Rapid Diagnostic Tests for the diagnosis of uncomplicated

Document Sample
Rapid Diagnostic Tests for the diagnosis of uncomplicated Powered By Docstoc
					                                                    Rapid Diagnostic Tests for the
                                                      diagnosis of uncomplicated
                                                    Plasmodium falciparum malaria

                                                  Validation and ease of use of three new
                                                              pan pLDH tests




                                                Final Report
                                                January 2006


                                                Report written by Carole Fogg (PI)




              Centre Collaborateur de l’OMS
          pour la Recherche en Epidémiologie
        et la Réponse aux Maladies Emergentes   This study is supported by the MSF
                                                Malaria Working Group and funded by MSF.

TELEPHONE :          00 33 (0)1 40 21 28 48
FAX :                00 33 (0)1 40 21 28 03
E-MAIL :       EPIMAIL@EPICENTRE.MSF.ORG
WEB :       HTTP://WWW.EPICENTRE.MSF.ORG



ASSOCIATION LOI 1901.
                                                       1
ACRONYMS


ACT        Artemisinin Combination Therapy
BS         Blood smear
CRF        Case Report Form
HCG        Human chorionic gonadotrophin
HRP2       Histidine Rich Protein II
IPD        In-Patient Department
IQR        Inter-quartile range
IV         Intra-venous
LNMP       Last normal menstrual period
NPV        Negative Predictive Value
OPD        Out-Patient Department
PCR        Polymerase chain reaction
pLDH       parasite lactate de-hydrogenase
PPV        Positive Predictive Value
MAbs       Monoclonal antibodies
MSF        Médecins Sans Frontières
MUST       Mbarara University of Science & Technology
MRRH       Mbarara Regional Referral Hospital
MWG        Malaria Working Group
RDT        Rapid Diagnostic Test
SD         Standard Deviation
Se         Sensitivity
SOP        Standard Operating Procedure
Sp         Specificity
UNCST      Uganda National Council of Science & Technology
WHO        World Health Organisation




                           2
  DEFINITIONS

Hyperparasitaemia           Infection with Plasmodium falciparum at a density of more
                            than 250,000 parasites per microlitre.
Inter-quartile range        The value of the first and third quartiles of a sample, or the
                            numerical difference between them.
Negative Predictive Value   The probability that the individual is not infected by
(NPV)                       plasmodium when the RDT is negative.
Pan-malarial antigen        An antigen expressed by all four of the plasmodium species
                            which are pathogenic in human beings, ie P. falciparum, P.
                            vivax, P. malariae, P. ovale.
Positive Predictive Value   The probability that an individual actually is infected by
(PPV)                       plasmodium when the RDT is positive.
Reliability                 The probability of 2 readers reading the same test and giving
                            the same result, as analysed using a Kapppa test.
Sensitivity (Se)            The proportion of true positives (blood smear +) that are
                            correctly identified by the test.
Severe anaemia              Haemoglobin ≤5g/dL
Specificity (Sp)            The proportion of true negatives (blood smear -) that are
                            correctly identified by the test.
Validity                    An overall term for 4 parameters used when assessing the test
                            – ie the NPV, PPV, Se and Sp.




                                            3
Table of contents

SUMMARY............................................................................................................................... 7
1 INTRODUCTION ................................................................................................................. 9
   1.1 HRP2 TESTS ................................................................................................................... 10
      1.1.1 Development of HRP2 tests and characteristics ..................................................... 10
      1.1.2 Constraints of HRP2 tests ....................................................................................... 12
   1.2 PLDH TESTS .................................................................................................................... 13
      1.2.1 Development of pLDH tests and characteristics..................................................... 13
      1.2.2 Possible advantages of pLDH RDTs ....................................................................... 15
   1.3 OTHER RAPID DIAGNOSTIC TESTS .................................................................................... 15
2 RATIONALE....................................................................................................................... 16
3 OBJECTIVES...................................................................................................................... 17
   3.1 PRIMARY OBJECTIVE ....................................................................................................... 17
   3.2 SECONDARY OBJECTIVES ................................................................................................. 17
4 STUDY DESIGN ................................................................................................................. 17
5 SAMPLE SIZE .................................................................................................................... 18
6 STUDY AREA ..................................................................................................................... 18
   6.1 STUDY SITE ..................................................................................................................... 18
   6.2 CATCHMENT POPULATION............................................................................................... 19
   6.3 MALARIA IN THE CATCHMENT POPULATION .................................................................... 19
7 STUDY PROCEDURES..................................................................................................... 20
   7.1 SCREENING ...................................................................................................................... 20
   7.2 INCLUSION PROCESS ........................................................................................................ 21
   7.3 DAY 0 POST-INCLUSION PROCEDURES ............................................................................. 21
   7.4 TREATMENT WITH COARTEM® ........................................................................................ 22
   7.5 FOLLOW-UP PROCEDURES................................................................................................ 23
8 TESTS & LAB TECHNIQUES ......................................................................................... 23
   8.1 BLOOD SMEAR ................................................................................................................. 23
      8.1.1 Method of preparation and reading ........................................................................ 23
      8.1.2 Quality control ........................................................................................................ 24
   8.2 RDTS .............................................................................................................................. 24
      8.2.1 Method..................................................................................................................... 24
      8.2.2 Interpretation........................................................................................................... 26
      8.2.3 Ease of use............................................................................................................... 27
   8.3 PREGNANCY TESTING ...................................................................................................... 27
   8.4 HAEMOGLOBIN ................................................................................................................ 28
9 DATA MANAGEMENT AND ANALYSIS...................................................................... 28
10 RESULTS........................................................................................................................... 29
   10.1 BASELINE CHARACTERISTICS OF SUBJECTS .................................................................... 30

                                                                     4
     10.1.1 Study profile .......................................................................................................... 30
     10.1.2 Demographic and clinical characteristics ............................................................ 30
     10.1.3 Parasitological characteristics ............................................................................. 31
   10.2 VALIDITY OF THE RAPID DIAGNOSTIC TESTS .................................................................. 32
     10.2.1 Overall results – Day 0 ......................................................................................... 32
     10.2.2 Validity according to age group – Day 0 .............................................................. 34
     10.2.3 Sensitivity according to level of parasitaemia – Day 0......................................... 36
     10.2.4 Sensitivity and PPV according to presence/absence of measured fever – Day 0 . 39
     10.2.5 Sensitivity according to duration of illness ........................................................... 41
     10.2.6 Sensitivity according to history of taking antimalarials ....................................... 41
   10.3 INTER-READER RELIABILITY .......................................................................................... 41
     10.3.1 Invalid and doubtful tests ...................................................................................... 41
     10.3.2 Inter-reader reliability on day of diagnosis (D0).................................................. 42
   10.4 PERCENTAGE OF POSITIVES ON POST-TREATMENT FOLLOW-UP ...................................... 42
   10.5 EASE OF USE .................................................................................................................. 43
   10.6 COSTS............................................................................................................................ 45
11 DISCUSSION..................................................................................................................... 45
   11.1 PLDH TESTS COMPARED TO PARACHECK ...................................................................... 45
   11.2 COMPARISON OF PLDH TESTS ....................................................................................... 46
   11.3 FACTORS AFFECTING VALIDITY OF PLDH TESTS ........................................................... 47
   11.4 POTENTIAL ROLE OF RDTS IN REDUCING THE COST OF DIAGNOSING AND TREATING
   MALARIA. .............................................................................................................................. 47
   11.5 STUDY LIMITATIONS ...................................................................................................... 48
   11.6 CONCLUSIONS ............................................................................................................... 48
   11.7 RECOMMENDATIONS ..................................................................................................... 49
12 ACKNOWLEDGEMENTS.............................................................................................. 50
13 REFERENCES ................................................................................................................. 50
14 APPENDICES ................................................................................................................... 54
   14.1 LIST OF STANDARD OPERATING PROCEDURES ................................................................ 54
     14.1 Clinical SOPs ........................................................................................................... 54
     14.2 Laboratory SOPs...................................................................................................... 54
   14.2 INFORMED CONSENT FORM ............................................................................................ 55
   14.3 CASE REPORT FORM...................................................................................................... 57
   14.4 SOPS FOR RAPID DIAGNOSTIC TESTS ............................................................................ 61




                                                                     5
                                                       Tables and figures


Tables
Table 1. Summary of HRP2 validity parameters in field tests................................................. 11
Table 2. Summary of pLDH validity parameters in field tests ................................................ 14
Table 3. Summary of HRP2/aldolase validity parameters in field tests................................... 16
Table 4. Coartem dosage....................................................................................................... 22
Table 5. Denominators for analysis.......................................................................................... 29
Table 6. Baseline characteristics of included population ......................................................... 30
Table 7. Parasitological characteristics of included population ............................................... 31
Table 8. Parasitaemia categories according to age group ........................................................ 31
Table 9. Species distribution on Day 0 by age group............................................................... 32
Table 10. Parasite density and IQR by species ........................................................................ 32
Table 11. 2x2 tables for the four RDTs as compared to microscopy on Day 0 (ALL SPECIES)
     .......................................................................................................................................... 33
Table 12. Sensitivity, specificity, PPV and NPV of the 4 RDTs (ALL SPECIES) ................. 33
Table 13. Sensitivity, specificity, PPV and NPV of the 4 RDTs (FALCIPARUM ONLY).... 34
Table 14. Validity according to age group. .............................................................................. 36
Table 15. Sensitivities of the RDTs at different levels of parasitaemia ................................... 37
Table 16. Sensitivity of RDTs according to level of parasitaemia, stratified by age group.... 38
1Table 17. Sensitivity and PPV of RDTs according to presence/absence of measured fever . 39
Table 18 . Validity according to fever...................................................................................... 40
Table 19. Percentage and number of invalid and doubtful tests according to day of follow-up
     .......................................................................................................................................... 41
Table 20. Inter-reader discordances on day 0........................................................................... 42
Table 21. Percentage of positive tests on each follow-up visit ................................................ 43
Table 22. Ease of use questionnaire results.............................................................................. 44
Table 23. Cost comparison for current diagnostic and treatment system versus RDT system48


Figures
Figure 1: Map of Uganda and study site location .................................................................... 19
Figure 2: Proportion of OPD consultations due to malaria in Mbarara district, 2000-3
     (excluding MRRH)........................................................................................................... 20
Figure 3 (i). Photograph of the four Rapid Diagnostic Tests – packaging and buffer vials..... 25
Figure 3 (ii). Photograph of the four Rapid Diagnostic Tests – cassettes, loops/pipette and
buffer vials................................................................................................................................ 26
Figure 4. Pictorial instructions for interpretation of rapid test results...................................... 27
Figure 5. Study profile, RDT validation study, Mbarara 2005 ................................................ 30
Figure 6. Diagrammatic representation of the sensitivity and specificity of rapid tests
     according to age group ..................................................................................................... 35




                                                                        6
                                         Summary

Background
The current gold standard for laboratory confirmation of diagnosis of malaria is a peripheral
blood smear, examined microscopically. However, trained staff and quality equipment and
supervision are scarce within populations requiring such diagnosis. Alternative methods of
diagnosis are available, the simplest of which are Rapid Diagnostic Tests (RDTs). These are
antigen detection tests which are simple to use and to interpret, and also use peripheral blood.
Currently, the RDTs which have been most developed detect an antigen called Histidine Rich
Protein II (HRP2) produced by Plasmodium falciparum trophozoites and young gametocytes.

Rationale
HRP2 tests have been shown to have constraints, namely that they remain positive for some
time after the blood smear has become negative, and that they only detect falciparum malaria.
Development of tests using another antigen – parasite lactate dehydrogenase (pLDH) – is
ongoing, and a few tests has been commercially available for several years. The advantages of
using pLDH-based tests are that pLDH is supposed to be cleared more quickly from the
bloodstream after parasites have been destroyed, and that it is a pan-malarial antigen produced
by all 4 plasmodium species. In a region such as sub-Saharan Africa where falciparum
malaria is predominant but other species are also present and levels of drug resistance are
high, patients with a negative HRP2 test may still be given antimalarials due to uncertainty as
to whether the patient has non-falciparum species. A test which detects all species would
reduce the chance of patients with a negative test being treated and assist to reduce
development of drug resistance. This study looked at the validity, reliability, percentage of
positives during post-treatment follow-up and ease of use of three new pLDH tests. Results
were compared to that of one currently widely used HRP2 test.

Methods
Out-patients with suspected malaria were screened using one HRP2 test (Paracheck), three
pLDH tests (Parabank, Carestart and Vistapan) and a blood smear. A total of 460 patients
(248 positive smears, 212 negative smears) were recruited into two age groups – under 5 years
and 5 and above. Each test was read by two independent readers (blinded to the results of each
other) and results were compared to that of the blood smear, considered as the gold standard.
Outcomes were sensitivity, specificity, positive and negative predictive values. Inter-reader
reliability was assessed by the kappa coefficient, with kappa>0.80 considered as very reliable.
Patients with a positive blood smear were treated with artemether+lumefantrine (Coartem)
and followed for 14 days to assess the percentage of tests remaining positive at days 3, 7 and
14 after treatment. Lab technicians completed a questionnaire at the end of the study to assess
the ease of use and other important characteristics of the tests.

Main results
Validity
Overall, the three panPLDH tests were more specific (VistaPan: 89.6%; Carestart: 91.5%;
Parabank: 94.3%) and had a higher PPV (VistaPan: 91.2%; Carestart: 92.9%; Parabank:
94.6%) than the Paracheck (values of 87.3% and 89.6%, respectively) although no differences
were significant. Two of these tests were less sensitive and had lower NPV (Vistapan: 91.9%
and 90.5%, respectively; Parabank: 84.7% and 84.0% - NPV significantly lower) than the
Paracheck (94% and 92.5%, respectively). The exception was Carestart which was more


                                               7
sensitive (95.6%) and had a higher NPV (94.6%) than Paracheck (although not significantly
so). The 4 validity parameters were higher in the under 5 compared to the 5+ age group for
each of the 4 tests evaluated, significantly so for the sensitivity and NPV of Vistapan and
Parabank. There was a significant increasing trend in sensitivity with increasing level of
parasitaemia for all 4 tests (p<0.0001) and an increased sensitivity (although not statistically
significant for Carestart) in patients presenting with fever.

Percentage of positives on post-treatment follow-up
After treatment with Coartem, the percentage of tests remaining positive on days 3, 7 and 14
of follow-up was as follows: Paracheck 86.2%, 80.8%, 69.7%; Vistapan 36.1%, 23.4%, 8.9%;
Carestart 42.5%, 27.6%, 9.5%; Parabank 17.8%, 8.9%, 4.6%.

Reliability
All tests were extremely reliable on day 0, with Kappa>0.90.

Ease of use
All tests were generally considered simple and quick to use. Differences between the tests in
terms of the score are mainly related to their different stages of product development, and
these scores will change over the coming year as further improvements and temperature
stability studies are made. There was no test which was felt to be vastly superior to others in
terms of use.

Conclusions
Overall, two of the pan-pLDH tests, Vistapan and Carestart, were considered alternatives to
Paracheck in sub-Saharan Africa where P. falciparum is the predominant species. The
Parabank is not yet a suitable replacement due to its lower sensitivity compared with
Paracheck (p<0.001) and the two other tests (p=0.019 and p<0.001, respectively). The
Carestart was considered the best of the three pLDH tests evaluated. The faster time to
become negative after treatment of the pLDH tests could be useful for monitoring recovery
from the episode of malaria, but with confidence only after 2 weeks. This characteristic also
makes it easier to decide whether a patient who returns within 2-3 weeks of treatment for
malaria does actually have parasites, either from a re-infection, or from treatment failure.
However, the higher cost of these tests needs to be addressed.




                                               8
1 Introduction
Malaria is a major health problem in most of the vulnerable populations in low income
countries. The recommended strategy to control malaria is based on prompt and accurate
diagnosis and the use of a combination therapy with artemisinin combination therapy (ACT)
(WHO 2003). Accurate confirmation of malaria diagnosis can reduce occurrence of
antimalarial treatment given under misdiagnosis, and therefore help to reduce the selection of
resistant strains of P. falciparum, as well as increasing patient confidence in health services by
providing accurate information for a differential diagnosis.

Currently the most commonly used gold standard diagnostic method is the microscopic
reading of a blood film. This requires a laboratory technician with training and experience,
quality reagents and equipment, including a microscope and lens which are costly. Ideally,
mechanisms should also be in place for maintenance and quality control to ensure the quality
of the ‘gold standard’. However, in many populations where malaria is prevalent, no
laboratory facilities or trained staff are present, making access to microscopically confirmed
diagnosis impossible.

An alternative to the classical thick/thin blood smear and microscopic examination is provided
by a technology known as the Rapid Diagnostic Test (RDT) (WHO, 1999). RDTs are based
on the detection of antigens derived from malaria parasites using immuno-chromatographic
methods. Fabricated in the form of a dipstick or a device bearing monoclonal antibodies, they
detect plasmodium-specific antigens in capillary blood taken from a finger-prick sample. The
WHO recommendation is that a RDT should have a sensitivity ≥95% at a level of ≥100
parasites/µl for Plasmodium falciparum, and should be at least as accurate as results derived
from microscopy performed by an average laboratory technician under routine field
conditions (WHO 1999). Other characteristics considered to be ‘essential’ include quantitative
or semiquantitative information on parasite densities in circulating blood, ability to distinguish
viable parasites from parasite products (eg antigens, nucleic acids), and an ability to indicate
the prediction of treatment outcomes or resistance to common antimalarial drugs. To date, no
test fulfils all of these criteria. The sensitivity of the RDTs is also affected by many factors
which will cause local variation, including the species of the parasite, the number of parasites
present, the condition of the RDT (affected by temperature/humidity etc), the correctness of
technique used to perform the test, the correctness of interpretation by the reader and the
parasite viability and variation in production of antigen by the parasite (WHO 2004). The
WHO further recommends that RDTs can be used when a clear benefit will occur in health
outcomes, and demonstration of parasitaemia will allow more rational use of antimalarial
drugs. This requires a clear plan of action to deal with the positive and negative results (for
example an algorithm), good health worker training and monitoring and quality control of
RDTs.

RDT technology is attractive to health care settings lacking in human resources and with a
high percentage of outpatients with suspected malaria requiring a blood smear (BS). Such
tests can be used by trained personnel, not necessarily a laboratory technician, and do not
require heavy infrastructure. They can be used easily within clinics and also on outreach
programmes. They are also useful in emergency settings where the usual health care system
has broken down or is overloaded, for example those where Médecins sans Frontières (MSF)
are often present. Currently, the Paracheck RDT is used in many MSF programmes where it
simplifies diagnosis of cases in mobile clinics, population-based surveys, and in health
facilities where standard laboratory facilities are lacking. However, most RDTs are currently


                                                9
rather expensive to be considered for routine use in public health systems in the populations
that would benefit.

The two types of RDT most often used belong to two groups according to the detected antigen
– those detecting histidine rich protein II (HRP2) and those using parasite lactate
dehydrogenase (pLDH).


1.1 HRP2 Tests
1.1.1 Development of HRP2 tests and characteristics

HRP2 antigen is one of the three histidine-rich proteins produced by trophozoites and young
gametocytes of Plasmodium falciparum. It is a soluble histidine-alanine rich protein,
expressed on the erythrocyte membrane surface. HRP2 antigen was identified in the early
1990’s as an antigen which is detectable in the plasma of patients infected with P. falciparum
and therefore a potential antigen for diagnosis of malaria (Parra et al., 1991). Since then, a
wide variety of tests, comprising dipsticks and devices, have been developed by various
manufacturers. Many examples of field testing in a variety of countries and population types
are available and a summary of the major studies is given in Table 1.




                                             10
Author                Test         Country        Study        Validity parameters           Comments
                                                Population
Beadle et al     HRP2 *           Kenya         general      Se: 96.5-100% for Pf >60/µl
1994                                                             70-81% for Pf 11-60/µl
                                                                 11-67% Pf <10/µl
                                                             Sp: 88%
                                  US            US           Sp: 95% - naïve subjects
                                                volunteers       98% - exposed to Pf
Van den          ParaSight-F      Belgium       returned     Pf: Se: 95%, Sp: 90%
Ende et al                                      travellers   All species: Se: 71%, Sp:
1998                                                         87%
                 ICT Malaria Pf                              Pf: Se: 95%, Sp: 89%
                 test                                        All species: Se: 71%, Sp:
                                                             86%
Funk et al       MalaQuick /      Switzerland   returned                   -               difficulties of
1999             ParaSight F                    travellers                                 interpretation at
                                                                                           parasitaemia
                                                                                           <100/µl
Gaye et al       PATH Malaria     Senegal       OPD          Se: 96%, Sp: 92.7%,
1999             assay                                       PPV: 96.8%, NPV: 91%
Tjitra et al     ICT Pf/Pv        Indonesia     OPD          Pf: Se: 95.5%, Sp: 89.8%,
1999                                                         PPV: 88.1%, NPV: 96.2%
                                                             Pv: Se: 75%, Sp: 94.8%,
                                                             PPV: 50%, NPV: 98.2%
Wolday et        Quorum RTM       Ethiopia      OPD          Se: 97.2%, Sp: 96.3%,
al 2001          dipstick                                    PPV: 77.8%, NPV: 99.6%
Guthmann         Paracheck Pf     Uganda        OPD          Se: 97.4%, Sp: 88.1%
et al 2002       dipstick
                 Paracheck Pf                                Se: 97.2%, Sp: 87.7%
                 device
                 ParaHIT f                                   Se: 97.6%, Sp: 87.3%
                 BIO P.F.                                    Se: 89.5%, Sp: 93.1%
                 Malaria Rapid                               Se: 98.3%, Sp: 75.3%
Forney et al     ParaSight F+V    Thailand      OPD          Pf: Se: 98%, Sp: 93%          decrease in Se
2003             prototype        and Peru                   Pv: Se: 87%, Sp: 87%          with decreasing
                                                                                           density for both
                                                                                           species
Grobusch et      Parasight-F      Germany       returned     Se: 95.1%, Sp: 97.1%
al 2003                                         travellers
                 ICT                                         Se: 95.7%, Sp: 99.2%
Dev 2004         HRP2*            India         OPD          Se: 100%, Sp: 94-100%
*: commercial name not given


Table 1. Summary of HRP2 validity parameters in field tests


Although false negative HRP2 tests are rare, it has been shown that anti-HRP2 IgM can be
correlated with relatively low HRP2 antigenaemias in microscopy-positive RDT-negative
samples, (Biswas S et al., 2005). This could be because the antibodies react with HRP2 in the
plasma to form immune complexes that, unlike the free antigen, are not detectable in an RDT
based on the antigen.


                                                       11
HRP2 tests may also be able to detect parasitaemias below the microscopic detection level,
and preliminary studies using polymerase chain reaction (PCR) detection of plasmodium have
shown RDTs to detect low-level chronic infections in some patients with a negative BS (Bell
et al., 2005). However, such detection would depend on the epidemiological characteristics of
the population, as another small study amongst Cambodian refugees showed no positive
RDTs which had positive PCRs but negative blood smears (Causer LM et al., 2005).



1.1.2 Constraints of HRP2 tests

HRP2 is slowly eliminated from the bloodstream: the mean (standard deviation) (SD) plasma
elimination half-life (t½) of PfHRP2 in 13 Thai adult patients was 3.67 (1.47) days (Dondorp
et al., 2004). Most commercially available tests using HRP2 can remain positive for up to 4
weeks after the infection has been cleared (Humar et al., 1997), although the majority can
become negative by 7 days after the initiation of treatment, (Beadle et al., 1994). Such false
positives can also be caused by the presence of gametocytes in the blood, even after the
asexual parasitaemia is no longer present. This makes it difficult to interpret such tests at the
day of diagnosis, especially in settings where antimalarials are often bought from private
pharmacies before the patient attends the health facility for diagnosis, and also to assess
whether a patient has been cured or not. Studies which have used HRP2 tests for monitoring
treatment outcome have not found it possible to predict treatment failure from repeated RDT
testing during the efficacy study follow-up (Tjitra E et al., 2001 (1)). This has been shown in
part to be due to persistent gametocytaemia, especially after treatment with relatively non-
efficacious drugs such as chloroquine and sulphadoxine-pyrimethamine (SP) (Tjitra et al.,
2001 (2)).

Another problem with the HRP2 tests is that, although in most of sub-Saharan Africa
falciparum comprises more than 90% of malaria infections, the HRP2 test classifies all non-
falciparum infections as malaria negative, thus missing the small but significant number of
monoinfections of vivax, malariae or ovale. Since ACT treats both falciparum and non-
falciparum, there is no need in this region to differentiate between species. The possibility that
an HRP2 negative test could be a non-falciparum infection leads to continuing clinical
diagnosis of malaria without parasitological confirmation and/or the continuing use of
chloroquine for presumed non-falciparum infections.

A further constraint of HRP2 tests may be their variability in performance according to
conditions at the time of manufacturing, storage, transport and utilization, as demonstrated in
the Philippines using ParaHIT-f and Paracheck Pf which had much lower sensitivities than
expected (Belizario et al., 2005).

Recent evidence comparing the gene sequences of PfHRP2 from worldwide isolates also
suggests that diversity in the sequence can contribute to differing validity in the tests between
sites, and identifies a possible role for PfHRP3 in cross-reactivity in some isolates (Baker J et
al., 2005).




                                               12
1.2 pLDH tests
1.2.1 Development of pLDH tests and characteristics

Parasite Lactate Dehydrogenase (pLDH) is an intracellular metabolic enzyme present in all
four plasmodium species which infect humans, and is produced by sexual and asexual stages
of the parasite. For this reason, they are also called ‘pan-pLDH’ test. Quantitative
immunocapture pLDH assays (ICpLDH assay) showed that the level of pLDH activity closely
mirrored the levels of parasitaemia both in initial diagnosis and while following patient
therapy, thus being a potential assay for monitoring therapy and indicating drug-resistant
infections. The late 1990’s saw the advent of the pLDH RDT. The first marketable test to be
produced was the OptiMAL 48 assay, commercially available since 1998, which could detect
infections of >200 parasites/µl and could differentiate between P falciparum and non-
falciparum malaria (Piper et al. 1999). A later study showed a loss of sensitivity of the test
strips due to humidity around 70% with routine use in the original OptiMAL (also called
OptiMAL 48) as compared with an improved test – the OptiMAL-IT (Moody AH, Chiodini
PL, 2002). Examples of validity of the two tests in different settings are given in the below
table.




                                             13
Author(s)           Test(s)     Country        Study           Validity parameters           Comments
                                             Population
Palmer et al     OptiMAL        Honduras    OPD –             Pf: Se: 94%, Sp: 100%       false negatives
1997                                        outbreak          Pv: Se: 88%, Sp: 99%        normally with
                                            situation                                     <100/µl
Cooke et al      OptiMAL        The         OPD               Se: 91.3% Sp: 92%           decreased Se of test
1999                            Gambia                        PPV: 87.2% NPV 94.7%        at parasitaemias
                                                                                          <100/µl
Piper et al      OptiMAL        London      returned          Pf: Se:60% <5/µl
1999                                        travellers               81% 50-500/µl
                                                                     94% 500-1,500/µl
                                                                     100% >1,500/µl
Moody et al      OptiMAL        London      returned          Pf: Se: 95.3%, Sp: 100%
2000                                        travellers        Pv: Se: 96%, Sp: 100%
Fryauff et       OptiMAL        Indonesia   OPD               Pf : Se : 60.4%, Sp : 97%   false negatives and
al 2000                                                       Pv/Pm : Se : 70.2%, Sp :    misidentifications
                                                              89%                         related to
                                                                                          parasitaemia<500/µl
Ricci et al      OptiMAL        Italy       returned          Se: 83% overall
2000                                        travellers        96-100% Pf>100/µl
                                                              0-44% Pf<100/µl
Congpuong        OptiMAL        Thailand    OPD               Se: Pf 92%; Pv 97.6%        detects only 20% of
et al 2001                                  (hyperendemic)    Sp: 100%                    infection with
                                                                                          parasites <200/µl
Iqbal et al      OptiMAL        Kuwait      immigrants        Se: 89%, Sp: 99.5%
2001
Ferro et al      OptiMAL        Colombia    OPD – urban       ‘overall diagnostic         Pv 98.1%, Pf 94.9%
2002                                        referral centre   efficiency’ 96.9%
Iqbal et al      OptiMAL        Kuwait      immigrants        Se: Pf: 87%; Pv 79%         Se: 44% at
2002                                                          Sp:>98%                     parasitaemia
                                                                                          <500/µl

Grobusch et      OptiMAL        Germany     returned          Se: 76.2%, Sp: 99.7%
al 2003                                     travellers
Pattanasin       OptiMAL-IT     Thailand    OPD               Pf: Se: 88%, Sp: 92%        Sensitivity
et al 2003                                                    non-Pf: Se : 65%, Sp :      decreased to 70%
                                                              99%                         below 500/µl
                                Colombia    OPD               Pf: Se: 100%                Pf range 42-
                                                              Pv: Se: 100%                129,000/µl
                                                                                          Pv range 200-
                                                                                          39,500/µl
Dev 2004         pLDH*          India       OPD               Se : 81-89%, Sp : 100%
De               OptiMAL-IT          -      non-endemic       Se : 79.8%, Sp : 98.4%
Monbrison                                   area
et al 2004
* : commercial name not given

Table 2. Summary of pLDH validity parameters in field tests




                                                         14
As can been seen in comparison between HRP2 and pLDH, (Tables 1 and 2), sensitivity in
operational conditions for the OptiMAL test is much lower than for the variety of HRP2 tests.
The most likely cause for loss of sensitivity on the field is due to either physical deterioration
or the device or to a manufacturing problem. Sensitivities of the OptiMAL test are generally
acceptable at a level between 200-500 parasites/µl (Moody AH, Chiodini PL 2002).


1.2.2 Possible advantages of pLDH RDTs

PLDH is eliminated from the bloodstream more rapidly than HRP2 after starting an effective
treatment, as it is an enzyme only produced by viable parasites (Laferi et al., 1997, Piper et
al., 1999). Therefore, the pan pLDH test should quickly revert to negative when the infection
is cleared (although the clearance rates depend on the circulating amounts), thus reducing
false positive tests which would lead to prolonged treatment with antimalarials. The faster
clearance of pLDH will reduce the possibility of having false positive patients who have been
treated with antimalarials in the previous 2-3 weeks and present again with fever. It would
also make the antigen potentially of value in monitoring efficacy of drug therapy, as it has
been shown to clear within 5 days following effective drug therapy (Wu et al., 2002). The
pLDH tests also have an advantage over HRP2 in that they can detect all parasite species, and
therefore reducing situations in which a negative test may be attributed to another species of
malaria.


1.3 Other rapid diagnostic tests
Another antigen also used in RDTs is aldolase, which also occurs within the glycolytic
pathway. Aldolase has been used and evaluated most commonly in a combination
HRP2/aldolase test, ICT malaria Pf/Pv. The advantage of such tests should be that they can
maintain the high sensitivities of HRP2 tests whilst also being able to detect other species of
malaria, although at not such a high sensitivity. As non-falciparum species tend to have lower
parasite densities, it is expected that aldolase/pLDH tests would be less sensitive for non-
falciparum species. As the table below shows, sensitivity of the aldolase both for falciparum
and for non-falciparum species is generally poor and highly variable between sites.




                                               15
First          Test      Country       Study           Validity parameters             Comments
Author                               Population
Tjitra et    ICT Pf/Pv   Indonesia   OPD          Pf: Se: 95.5%, Sp: 89.8%,       Se 29% when parasites
al 1999                                           PPV: 88.1%, NPV: 96.2%          <500/µl
                                                  Pv: Se: 75%, Sp: 94.8%,
                                                  PPV: 50%, NPV: 98.2%
Cho et al    ICT Pf/Pv   Korea       OPD          Se: 44.6%, Sp: 100%             Se dropped to half at
2001                                                                              <5,000/µl. All vivax
                                                                                  infections.
Iqbal et     ICT Pf/Pv   Kuwait      immigrants   All species: 82%, Sp:99%        >50% Pv infections
al 2001
Huong et     ICT Pf/Pv   Vietnam     OPD          Pf: Se: 82.6%, Sp: 100%
al 2002                                           Pv: Se: 20%, Sp: 100%
Iqbal et     NOW         Kuwait      immigrants   Pv: Se: 58%, Sp:>98%
al 2002      Malaria
             ICT Pf/Pv
Mason et     ICT Pf/Pv   Burma       OPD          Pf: Se: 86.2%, Sp: 76.9%        non-Pf = Pm and Pv -
al 2002                                           non-Pf: Se: 2.9%, Sp: 100%      misdiagnosis as Pf in
                                                                                  38% of cases
Farcas et    NOW ICT     Canada      returned     Pf monoinfection: Se: 95.5%     PCR as gold standard.
al 2003      Pf/Pv                   travellers   Pf mono+mixed: Se: 94.3%        Pf : Se: 75% at <100/µl
                                                  Pv: Se: 87%                     Pv: Se: ≤55% at
                                                  Po and Pm: Se: 62%              <1,000/µl
                                                  Overall Sp: 99%
Richter et   Malaria-    Germany     returned     Pf: HRP2 band: Se: 100%         Co-reaction of bands at
al 2004      now                     travellers   Pf: Aldolase band: Se: 48.1%    Pf
                                                  Pv: (aldolase only):Se: 37.5%   parasitaemia≥40,000/µl
                                                  Po+Pm (5 cases): Se: 0%


Table 3. Summary of HRP2/aldolase validity parameters in field tests


Combination HRP2/aldolase tests have also been suggested for a role in monitoring response
to antimalarial therapy, where a small sample of returned travellers to Australia showed a
decreasing intensity of aldolase detection in line with declining parasite density (Eisen DE,
Saul A, 2000).

2 Rationale
The RDT currently used in the field by MSF is the Paracheck HRP2 test (Orchid/Zephyr,
India) which only detects P. falciparum. It is a very sensitive test, and performs well even
when exposed to high ambient temperatures. However, the drawbacks of Paracheck, chiefly
the prolonged positivity after treatment and the failure to detect other species of malaria, mean
that it is not always the best diagnostic tool on which to base an effective treatment.

The MSF Malaria Working Group (MWG) defined the “ideal” profile of a new RDT, which is
an RDT which can detect all species of malaria with a sensitivity and specifity of at least 90%
for P falciparum and P vivax. To date, available pLDH tests have had a poor sensitivity. The
new tests use newly developed monoclonal antibodies (MAbs) which have shown a much
higher sensitivity. However, there is need to measure the performance of these new pLDH


                                                  16
tests in field conditions before their wider use, particularly in MSF programs, as there is no
international and independent body responsible for the evaluation of such products. Moreover,
an overall consensus from recent publications is that pLDH disappears from the blood as the
viable parasites disappear, but this needs to be shown in the field by comparing the pLDH
tests directly with an HRP2 test to see the numbers of false positive tests after treatment.
Validation of these alternatives to the OptiMAL-IT, the only evaluated pLDH test currently
available on the market which has an insufficient sensitivity and costs approximately 1.5
dollars/test, is therefore needed.

The new pan pLDH monoclonal antibody which has been developed by Flow Inc. (USA) has
been used to manufacture test strips by 3 different companies, which are listed below:

   1. Vista-pan malaria test (Mitra, India)
   2. CareStart antigen test (AccessBio, US)
   3. Parabank device (Orchid/Zephyr, India)


3 Objectives


3.1 Primary objective

   •   To measure the validity of rapid tests when compared to microscopy (gold standard).


3.2 Secondary objectives

   •   To assess the percentage of positive tests after efficacious treatment of a blood smear
       positive patient (with a limit of 14 days).
   •   To measure the reliability of 2 readers reading the same rapid test.
   •   To assess the ease of use of each rapid test.
   •   To assess factors contributing to differences in validity, eg age group, parasite density.

4 Study design
A cross-sectional diagnostic evaluation study of malaria. Patients with suspected malaria
presenting to Mbarara Regional Referral Hospital (MRRH) OPD were diagnosed using
microscopy which was compared to the result of each of the 4 RDTs under investigation
(Carestart, Vistapan, Paracheck, Parabank), to assess validity. Patients with positive blood
smears were treated with an efficacious drug and followed-up on the 3rd, 7th and 14th days
after the first dose of treatment, when all RDTs and the blood smear were repeated to assess
the number of tests remaining positive after treatment.




                                               17
5 Sample size
With an estimated sensitivity of a test of 90% and an alpha error of 0.05, the number of blood
smear positive subjects needed to achieve a precision of 6% is 96. The analysis was stratified
according to 2 categories of age (under 5 years versus 5 years and above), therefore requiring
96 cases in each category of age, i.e. a total number of approximately 200 positive subjects.
Expected specificity was calculated with the same value, precision and alpha error, giving a
sample size of 96 blood smear negative patients in each age group.

In conclusion, the sample size was 100 positive and 100 negative slides in each age group,
thus a minimum total of 200 positive and 200 negative slides.

6 Study area

6.1 Study Site
The study was conducted in Mbarara Municipality, a town of 69,000 persons (2002
Population Census Provisional Results) located 300 km south-west of Kampala (Figure 1).
The town is located in Mbarara District, which has a population of approximately 650,000
thousand (2002 Population Census Provisional Results), and is mainly rural with a density of
98.6 inhabitants/km2. The main health facility is the Mbarara Regional Referral Hospital
which is the referral hospital for the western region of Uganda. It is also the teaching hospital
for the Medical School of the Mbarara University of Science and Technology (MUST). It has
a bed capacity of 240 beds, OPD and IPD services, a paediatric unit and various diagnostic
laboratories.




                                               18
Figure 1: Map of Uganda and study site location



6.2 Catchment Population

Patients were recruited from the OPD of the Mbarara main Hospital, (MRRH), which has a
general OPD and a Children’s Clinic for children under the age of 13 years, and from Kakoba
Health Centre III, also located within Mbarara Municipality.

6.3 Malaria in the catchment population
Malaria constitutes one of the major causes of illness in the district. In the years 2000 to 2003,
malaria was diagnosed in more than 60% of the total OPD consultations in children under 5
years in the district, and more than 50% in patients 5 years and above (excluding MRRH)
(Figure 2). In Mbarara referral hospital, 74.8% of the total OPD consultations in under 5’s and
21.2% patients above 5 years in 2003 were clinically attributed to malaria (not all slide
confirmed).




                                               19
                                            100
                                                                                                     under 5 years
                                             90
                                                                                                     5 years and above
    % of OPD consultations due to malaria



                                             80

                                             70

                                             60

                                             50

                                             40

                                             30

                                             20

                                             10

                                              0
                                                       2000              2001             2002              2003
                                                                                 Year


Figure 2: Proportion of OPD consultations due to malaria in Mbarara district, 2000-3
(excluding MRRH)


There is seasonal variation of malaria in Mbarara. There are two peaks of rainfall, the first
peak usually occurs between March and May and the second between September and
November. Malaria cases usually increase about two months after the beginning of the rainy
season and represent approximately one half to one third of malaria suspects depending on the
season and the age group. 95% of confirmed cases are Plasmodium falciparum
monoinfections.

7 Study procedures
All study procedures were carried out according to approved standard operating procedures
(SOPs), which are listed in Appendix 1.


7.1 Screening
A clinical officer in the OPD screened patients according to the following criteria:

•                                           Suspected clinical malaria (history of fever in the last 48 hours, no other obvious causes
                                            of fever, presence of other malaria-related symptoms eg headache, nausea, fatigue);
•                                           Weight ≥ 5 kilos;
•                                           Resident in Mbarara Municipality;
•                                           NO signs of severe illness;
•                                           NO visibly pregnant women.

Patients were given a screening number and a referral slip and escorted to the Epicentre clinic
within the hospital grounds by the patient escort.


                                                                                    20
7.2 Inclusion process
On arrival at the clinic, the patient was logged into the screening register and the residence of
the patient double-checked by the registration clerk. The patient was then referred to a
clinician, who completed a screening form containing a simple physical examination and the
below eligibility criteria. In case of suspicion of anaemia, a fingerprick blood sample was
taken and assessed by the laboratory using a haemoglobinometer (HemoCue). Patients with
severe anaemia or other signs of severe disease were referred to MRRH and were not included
in the study.

Inclusion criteria:
    • Clinical suspicion of malaria, ie history of fever in the last 24 hours and/or axillary
        temperature >37.5°C;
    • Weight > 5 kilos;
    • Resident in Mbarara Municipality;
    • Available for follow-up of two weeks;
    • Signed informed consent.

Exclusion criteria:
   • Signs of severe or complicated malaria (WHO 2000);
   • Signs of severe disease;
   • Women with visible pregnancy or suspicion of pregnancy based assessment of last
       normal menstrual period (LNMP).

Patients (or a parent/guardian for children under 18 years of age) who met all entry criteria
were then informed about the purpose of the study and asked to sign a written informed
consent in Runyankole or English (see Appendix 2 for the English version). Children under
18 years who could comprehend study content were also asked to sign. In case the
patient/guardian was not able to read the consent form, an independent witness was found to
witness the explanation of the clinician and to confirm that what was explained to the
patient/guardian was what was written on the consent form, and to sign the informed consent
in addition to the patient/guardian.

The clinician then opened a case report form (CRF) for the patient and gave the next available
study number for that age group which was used on all subsequent source data forms. The
patients name, age and study number was then recorded on the enrolment log, and the
screening and referral forms filed in a secure place. The CRF used for the study is given in
Appendix 3.


7.3 Day 0 post-inclusion procedures

A laboratory request form was written for the patient. A fingerprick was performed and a
thick and thin blood smear made. From the same fingerprick, samples were collected on the
respective pipettes/loops for the four RDTs (Paracheck, Vistapan, Carestart and Parabank).
RDTs were processed in the clinic, and blood smears were taken to the main lab for
processing. (Refer to Chapter 8 for more information on laboratory methods). The blood



                                               21
smear was systematically double-read, results were compared, and a third reading performed
in case of discordance. The RDTs were read by two independent health professionals/co-
investigators who had been trained before the beginning of the study. Each person read the
RDT without knowing the result of the other reader and the result of the blood smear. Results
for each RDT were compared by a lab technician, and a third reading performed by an
independent lab technician in case of discordance. An RDT considered invalid by either
reader was repeated. First readings of tests were done at the time specified by the
manufacturer, and all second readings were done within a further 20 minutes to collect data
for the inter-reader reliability of the test.

RDT and BS results were entered into the CRF by the clinician. Patients with a negative blood
smear and NO positive rapid test were given appropriate out-patient treatment and discharged
from the clinic. Patients with a negative blood smear and at least one positive rapid test were
also treated appropriately (for example completing an unfinished course of antimalarial
treatment) and asked to return to the clinic for the follow-up period.

Patients with a positive blood smear (regardless of the RDT results) were further assessed. A
full physical address was taken by the registration clerk. All patients were admitted onto the
observation ward and treated for 3 days with a 6-dose Coartem regimen EXCEPT for the
following categories of patients:

        1. Patients with hyperparasitaemia who were referred to the paediatric or medical ward
for treatment with quinine 10mg/kg, thrice daily, for 7 days, intravenous (IV) initially and
switching to oral treatment after improvement. Any patient developing signs of severity
before/after the inclusion period was also referred for quinine.

       2. All women aged 14 years and above were requested to consent to a pregnancy test
(HCG), as Coartem is not yet proven safe in the first trimester of pregnancy, and only to be
used in emergency. Women found to be pregnant were given the nationally recommended
treatment and discharged.

No further laboratory samples were taken during the treatment period unless the clinician felt
it was clinically necessary, for example in case of deterioration of the patient.


7.4 Treatment with Coartem®

Patients admitted onto the observation ward received artemether + lumefantrine (Coartem®)
treatment according to the schedule in Table 4. Each tablet contains 20mg artemether and
120mg lumefantrine. Each dose was supervised, and doses vomited or spat within 30 minutes
of administration were repeated.

                               Day 0                          Day 1                     Day 2
Weight range            0h               8h           8am             8pm       8am             8pm
5.0 – 14.9 Kg         1 tab            1 tab         1 tab            1 tab    1 tab            1 tab
15.0 – 24.9 Kg        2 tabs           2 tabs        2 tabs           2 tabs   2 tabs           2 tabs
25.0 – 34.9 Kg        3 tabs           3 tabs        3 tabs           3 tabs   3 tabs           3 tabs
≥ 35.0 Kg             4 tabs           4 tabs        4 tabs           4 tabs   4 tabs           4 tabs
Table 4. Coartem dosage


                                                22
7.5 Follow-up procedures

Patients with a positive blood smear (treated with either Coartem or quinine) and/or at least
one positive rapid test were requested to come for follow-up visits on days 3, 7 and 14.
Patients were given a follow-up card with details of the expected visits.

On each follow-up visit, a brief clinical examination was performed and fingerprick samples
were taken off to repeat the blood smear and all four rapid tests. If a patient who had a
negative blood smear but at least one positive RDT on day 0 became BS positive on follow-
up, they were given a new CRF (but not a new study number), admitted to the observation
ward, given Coartem and followed anew from Day 0 – ie their total follow-up period was
lengthened by a further 14 days.

If any patient did not attend a follow-up visit, a tracing team located him/her at home. If they
were not located, the tracers would attempt on the following day. If the patient was still not
located, s/he was considered lost to follow-up for that visit. If the patient did not come for the
subsequent visit, tracing procedures were again applied.




8 Tests & Lab Techniques

8.1 Blood smear
8.1.1 Method of preparation and reading
Two to three drops of capillary blood were drawn from each subject for preparation of thick
and thin blood films. Blood films were dried, thin films were fixed in methanol for a few
seconds, thick films were left unfixed and stained with 3% Giemsa for 45 minutes and read
under the microscope at 1,000 magnification for the presence of malaria parasites. 200 high
power fields were read before considering a blood slide to be negative for malaria parasites.
Once a slide was found to be positive for asexual parasites, the number of parasites present
per 200 or 500 white blood cells (WBCs) was recorded (or the number of WBCs reached
when the field was finished). In case the number of parasites counted per 200 WBCs were
<10, the count was continued until 500 WBCs were reached. The parasite density per
microlitre (µl) was calculated using the following formula (WHO 1991):

                               parasite count x 8000____
                              number of WBC counted
Parasite species were determined on the thick film and confirmed on the thin film.
Gametocyte presence/absence on all slides was recorded with a species identification where
possible, but density counts performed for falciparum only. Slides with mixed infections had
only falciparum quantified. For slides with a non-falciparum monoinfection, an estimate was
made of the asexual density per microlitre. Blood smears on follow-up visits were prepared
with the same method.




                                               23
After the first reading, all inclusion slides were blinded by the lab supervisor and read by a
second technician using the same methods.

8.1.2 Quality control

8.1.2.1 Day 0 slides
Results of the two readings were entered into an Internal Quality Control Excel sheet, which
informed the lab supervisor if the results were concordant or not. If only 2 readings were
necessary, the final parasitaemia was given as the average of the two readings.

Four types of discordance were assessed for each slide:
   • positive/negative discordance for asexual stages
   • species discordance for asexual stages
   • asexual density discordance – more than 50% difference between the parasitaemias of
       the two readers
   • positive/negative discordance for gametocytes

In case of discordance, a third technician read the slide and the definitive result was obtained
from the two results which concurred most closely.

8.1.2.2 Follow-up slides
A internal quality control for follow-up slides was performed on 20% of all follow-up slides
using the method described above. Any positive follow-up slide also had a second blinded
reading for confirmation.

8.1.2.3 External quality control
The sample size was calculated on a 3% precision, and an expected sensitivity/specificity of
>93%. A sample of 145 positive slides and 145 negative Day 0 slides were randomly selected
for external quality control at Shoklo Malaria Research Unit, Thailand. The Epicentre
Mbarara laboratory had a sensitivity of 95.5%, a specificity of 100%, a PPV of 100% and an
NPV of 94.9%. The 7 false negatives of the Mbarara lab were detected positive by SMRU
between 1-5 parasites per 500 WBCs, ie at very low densities which could have been missed
due to chance. There were two species discordances - one P falciparum recorded by Epicentre
as a mixed infection with falciparum and vivax, and one P ovale recorded by Epicentre as P
vivax. Gametocytes were unrecorded by Epicentre on 12 of 42 occassions, and wrongfully
reported on 2 slides.


8.2 RDTs
8.2.1 Method
All RDTs have similar methods of operation. They consist of a cellulose nitrate strip which is
fixed within a cassette or device. For the pLDH tests, the capture well contains an anti-pLDH
antibody conjugated with colloidal gold. The strip has two further reactive lines – one in the
‘test’ window which consists of anti-pLDH antibodies, and another in the ‘control’ window.
The cassette has two wells - one at the base known as the ‘buffer well’ and one further up
called the ‘sample well’. Blood is collected from the patient using either a loop or a
micropipette which is pressed into the sample well and the blood dispensed. A known
quantity of buffer solution is dropped into the the buffer well, and flows up the cellulose
nitrate strip by chromatographic action, past the sample well and carries the blood over the


                                              24
test line and the control line before reaching the end of the cassette. The tests are placed on a
flat surface and left for either 15 or 20 minutes, after which the results are read. Figures 3 and
4 give a photographic impression of the tests. The Vistapan test had individual buffer vials
produced later on plus the test cassette was modified, but these developments are not shown
here. Detailed SOPs for the tests are given in Appendix 14.4.




Figure 3 (i). Photograph of the four Rapid Diagnostic Tests – packaging and buffer vials
Left to right: Vistapan, Carestart, Parabank, Paracheck




                                               25
Figure 3 (ii). Photograph of the four Rapid Diagnostic Tests – cassettes, loops/pipette and
buffer vials
Left to right: Vistapan, Carestart, Parabank, Paracheck

Parabank and Paracheck RDTs were transported at ambient temperature from India, Vistapan
and Carestart were transported between 2-8°C from the US. RDTs were stored at an average
temperature of 24°C (range 12°C to 32°C) and average humidity of 58% (range from 27% to
83%) during the study period (data summarised from clinic records).


8.2.2 Interpretation
The RDTs were read by two independent health care providers (ie clinicians, nurses,
investigators or lab technicians) trained according to the manufacturers instructions and
reading according to a rota. A discordant result was read by the lab technician in charge in the
clinic. This method of reading enabled us to have a final accurate result for each RDT, and to
be able to compare inter-reader reliability between the two readers. All RDTs were interpreted
in the following manner:
    • one line at the control position: negative
    • one line at the control position plus one line at the ‘test’ position: positive
    • one line at the ‘test’ position and no line at the control position OR no lines present:
        invalid
A pictorial representation of the results is given below.




                                                26
                    CONTROL                                             CONTROL
                      LINE                                                LINE

                      TEST                                                 TEST
                      LINE                                                 LINE




                    SAMPLE                                               SAMPLE
                     WELL                                                 WELL

                    BUFFER                                              BUFFER
                     WELL                                                WELL


  POSITIVE                      NEGATIVE                              INVALID
Control line: YES             Control line: YES                    Control line: NO
 Test line: YES                Test line: NO                  Test line: either Yes or No


 Figure 4. Pictorial instructions for interpretation of rapid test results.


 Invalid tests were repeated.

 As can be seen from the above figure, the control line must be present for the result to be
 valid, but its presence does not prove that the RDT result is accurate.


 8.2.3 Ease of use
 A series of characteristics considered important for the use of a RDT in the field was recorded
 and a score given to each test according to these characteristics. This was done by two
 independent persons at the end of the study inclusions and any discordant opinions were
 resolved in a discussion with the PI. A score for each sub-section was totalled out of 10. A
 weighting was then given to each section according to its perceived importance in the field.

 The questionnaire was designed by first listing the parameters considered to be most
 important using information from a previous RDT study (Guthmann et al., 2002) and an
 informal discussion with the laboratory team. The parameters were then organised into
 common groups (or sub-sections) and each option for each parameter given a score. The total
 score for each sub-section was designed to total 10 for a perfect score. We then ranked the
 sub-sections in order of importance and chose a weight for each parameter based on its
 importance, as is sometimes done for other commercially available products. Sub-section
 totals were multiplied by the weight and then added to produce a final score out of 100. There
 is no standard method for this procedure, so the method we have followed is very much a
 locally devised one.


 8.3 Pregnancy testing

 Pregnancy testing was performed using HCG latex antigen agglutination test on a freshly
 collected urine specimen.




                                                   27
8.4 Haemoglobin

Blood was collected from a fingerprick into a HemoCue cuvette. The cuvette was then
inserted into a HemoCue haemoglobinometer (HemoCue B-Haemoglobin analyser,
Angelholm, Sweden) which gave a digital result in g/dL. The Hemocue machine was quality
controlled on each day when haemoglobin estimation was to be performed.


9 Data management and analysis
Data was double-entered using Epidata 3.1 (Epidata, Denmark). Cross-checking was
performed using Epidata and a Stata cleaning program used to ensure authenticity of the data.
The final database was transferred into a Stata dataset, and analysed using Stata 9.1 (College
Station, Texas).

Analysis proceeded in the following manner:

i)     Study profile: constructed using actual figures and percentages derived directly from the
       database.

ii)    Baseline characteristics (demographic, clinical, and parasitological): analysed using
       descriptive statistics (means, inter-quartile range and value, etc) and univariate
       comparative tests (Chi square or Mann-Whitney) for differences between age groups.

iii)   Validity: calculation of the sensitivity, specificity, PPV and NPV for each test overall
       and stratified by
             • age group;
             • category of parasitaemia (4 levels: 1-99 parasites/µl, 100 – 199 parasites/µl,
                 200-499 parasites/µl, and ≥500 parasites/µl; validity: level 1 with parasitaemias
                 between 1-99, level 2 with parasitaemia ≥100, level 3 with parasitaemia ≥200
                 and level 4 with parasitaemia ≥500);
             • fever;
             • duration of illness (categorised with 0 to 2 days of illness in one category vs 3
                 days and above in the other. 2 days of illness was considered to be a
                 ‘psychological threshold’ of illness which is not going to resolve without
                 visiting the health centre);
             • and history of taking antimalarials.

Values between groups were compared using comparison of proportions (Chi square).

iv)    Invalid and doubtful tests: actual figures and percentages derived directly from the
       database.

v)     Inter-reader reliability (day 0 and on follow-up): use of the Kappa test to compare
       reader 1 and reader 2 results for all RDTs.

vi)    Percentage of positives on post-treatment follow-up: Proportion of the number of
       positive tests over the total number of tests at each follow-up visit. Eligible patients are
       those with a positive blood smear on Day 0 who received Coartem treatment. Only



                                                28
     subjects with a positive relevant test on Day 0 are included in the denominator for each
     test.

vii) Ease of use questionnaire: Summary of data, weighting and calculation of the final
     score (as described in methods).



10 Results
Due to some violations of eligibility criteria and thus differences in patient management and
within the study, the following details guide the denominators for the analysis.

       1. Seven patients in the under 5 group had parasitaemia above the threshold for
          severe malaria (>250,000/µl) and were referred to Mbarara University Teaching
          Hospital (MRRH) for treatment with quinine.
       2. One subject was found not to be a resident of Mbarara after inclusion, but follow-
          up was continued as normal.
       3. Five women were found to be pregnant after inclusion and were therefore treated
          with the recommended antimalarials (not Coartem). Only one of the four
          participated in the follow-up.
       4. After inclusion and admission, two patients vomited a Coartem dose twice and
          were given rescue treatment with quinine. As this was supervised by Epicentre, it
          was considered to also be an efficacious treatment and subjects were still included
          in the time to negativity analysis.
       5. Five subjects with a positive RDT and negative BS on day 0 were followed-up and
          became BS positive during the follow-up period.

Subjects were included in different parts of the analysis as detailed in the below table.

                                       RDT         Inter-reader    Time to     Invalid/doubtful
                                     Validation     reliability   negativity         tests
                                       Day 0          Day 0       Follow-up
Parasitaemia >250,000/µl, quinine        X              X                             X
treatment (by MRRH) (n=7)
Non-resident of Mbarara (n=1)            X              X            X                X
Pregnancy (n=5)                          X              X                             X
Supervised rescue treatment after        X              X            X                X
Coartem (n=2)
RDT+/BS- on day 0 – details of                          X            X                X
second CRF. (n=5)
RDT+/BS- on day 0                        X              X
Missing RDT data points on               X              X                             X
follow-up visits (days 3, 7, 14)

Table 5. Denominators for analysis




                                                  29
10.1 Baseline characteristics of subjects
10.1.1 Study profile
The RDT study commenced in Mbarara on 26th April with a 2-day pilot, and then continued
into the study inclusions. The final inclusions were on the 27th of July, with follow-up
continuing for 2 more weeks. During the study period, 460 subjects were included (including
the pilot) – 239 under 5 years (group A) and 221 aged 5 years and above (group B). Sixteen
patients/ parents/guardians (10 in group A and 6 in group B) did not consent to enter the study
(3.4% of eligible patients). The only reason given by non-consenting patients was the lack of
other parents/guardians to look after other family members at home during the 3-day period of
observation on the ward. The study profile is illustrated below.

                                          Screened: 485


                                                                                                        Severe illness: 3
                                                                      Found ineligible at clinic: 9     Non-resident: 5
                                                                                                        Outside age group: 1


                                                                      Non-consent: 16 (ALL with other
                                           Included: 460              family members to care for)




               Group A (under 5): 239                                         Group B (5+): 221




   BS positive: 129            BS negative: 110            BS positive: 119                       BS negative: 102




Figure 5. Study profile, RDT validation study, Mbarara 2005

10.1.2 Demographic and clinical characteristics
Baseline characteristics of the included population are shown in the below table. The χ2 p-
value represents a comparison between the two age groups.
Characteristic                    Group A (<5)              Group B (5+)                Overall                χ2
                                     n=239                     n=221                    N=460
Gender ratio (M:F)                        0.98                    0.52                    0.73              p=0.001
                                        (118:121)              (76:145)               (194:266)
Mean age (SD)                         2 years                  22 years                12 years               N/A
                                    (14 months)               (12 years)              (13 years)
Median duration of illness           3 (1 – 14)               3 (1 – 30)              3 (1 – 30)             p=0.2*
in days, (range)
Previously taken                    81 (33.9%)                60 (27.3%)             141 (30.7%)             p=0.13
antimalarials (n, %)

Fever on presentation               99 (41.4%)                31 (14.0%)             130 (28.3%)           p<0.0001
(axillary temp. ≥37.5°C)
*Kruskal Wallis equality of populations test
Table 6. Baseline characteristics of included population


                                                             30
   Ages of study subjects ranged from 2 months to 62 years.

   The most frequently taken antimalarial drug before attending MRRH OPD was chloroquine
   16.7% (n=77), followed by quinine 10.9% (n=50) and fansidar 10.0% (n=46). Amodiaquine
   had been used by 2 subjects and artemether and mefloquine had been used by 1 subject each.
   33 subjects (7.2%) had used the drugs in combination either simultaneously or subsequently.

   10.1.3 Parasitological characteristics
   Parasitological characteristics of study subjects with positive blood slides on inclusion are
   given in the below table, using the geometric mean and the inter-quartile range (IQR).

Characteristic                  Group A (<5)         Group B (5+)       Overall          Comparison
                                   n=129                n=119           N=248            between age
                                                                                            groups
Asexual parasitaemia range      16 – 703,411         16 – 233,241     16 - 703,411         p=0.0001
                                                                                          (K-Wallis)

geometric mean of asexual           7,433                1,524            3,475            p=0.001
parasitaemia [95% CI]          [4,869 – 11,346]      [975 – 2,384]   [2,521 – 4,790]        (t-test)

Inter-quartile range (inter-   1,682 – 45,748        166 – 11,070     641 – 23,827             -
quartile value)                   (44,066)             (10,904)         (23,186)

Gametocyte carriage (n, %)       36 (27.9%)           22 (18.5%)       58 (23.4%)         p=0.11 (χ2)

   Table 7. Parasitological characteristics of included population


   A further 9 subjects with a negative blood smear for asexual stages had gametocytes on
   inclusion – 5 in group A and 4 in group B. This gives an overall gametocyte carriage rate of
   14.6% (n=67), with 17.2% (n=41) in group A and 11.8% (n=26) in group B.

   Parasitaemia stratified according to density categories are shown below according to age
   group.


                      Under 5 (n=129)            5+ (n=119)          Overall (N=248)
   Density             n          %             n          %          n           %
   1-99                10        7.8%           21       17.7%        31        12.5%
   100-199              3        2.3%           9         7.6%        12          4.8%
   200-499              4        3.1%           9         7.6%        13          5.2%
   500+                112      86.8%           80        67.2%       192         77.4%

   Table 8. Parasitaemia categories according to age group



   There was a clear significant difference between “low” (<100 parasites/µl) and “high” (≥100
   parasites/µl) parasitaemias between age groups. About 92.2% of the under 5’s had a high
   parasitaemia compared to 82.3% of 5 years and above (p=0.031).




                                                     31
The full species distribution on Day 0 is given in the below table.

Species                     Under 5’s (n=129)           ≥5 (n=119)             Overall (N=248)
                               n           %           n         %               n         %
falciparum monoinfection     124       96.1%          108      90.8%            232      93.6%
malariae monoinfection         2          1.6%        4            3.4%           6     2.4%
vivax monoinfection            2          1.6%        4            3.4%           6     2.4%
falciparum + malariae          1          0.8%         1          0.84%           2     0.81%
falciparum + vivax                    0               2            1.7%           2     0.81%

Table 9. Species distribution on Day 0 by age group


The geometric mean and interquartile range of each species (regardless of age group) is
shown below. (In mixed infections the density is calculated for P. falciparum only, and
therefore these densities are not included in the means summary. Only monoinfections are
shown).

Species                         Geometric mean             Inter quartile range
                                   [95% CI]                     (IQ value)
falciparum monoinfection      3822 [2,741 – 5,328]         689 – 26,548 (25,859)
malariae monoinfection         1472 [898 – 2415]            1132 – 1942 (810)
vivax monoinfection                156 [41 – 600]             79 – 376 (297)

Table 10. Parasite density and IQR by species



10.2 Validity of the rapid diagnostic tests
10.2.1 Overall results – Day 0

The 2 by 2 tables showing the concordances and discordances for each rapid diagnostic test
on the day of diagnosis (Day 0) as compared to the blood smear are shown below (N=460).
The total number of positive and negative blood smears is the same for each test evaluated
(248 and 212 respectively). Therefore column totals per test are not shown.




                                                 32
                                                       BLOOD SMEAR
                                       Positive            Negative         Total
                                       (n=248)             (n=212)         (N=460)
   Paracheck       Positive           233 (94%)*          27 (12.7%)          260
                   Negative            15 (6%)            185 (87.3%)         200


   Vistapan        Positive           228 (91.9%)         22 (10.4%)          250
                   Negative            20 (8.1%)          190 (89.6%)         210


   Carestart       Positive           237 (95.6%)          18 (8.5%)          255
                   Negative            11 (4.4%)          194 (91.5%)         205


   Parabank        Positive           210 (84.7%)          12 (5.7%)          222
                   Negative           38 (15.3%)          200 (94.3%)         238
*column percentages
Table 11. 2x2 tables for the four RDTs as compared to microscopy on Day 0 (ALL SPECIES)


The sensitivity/specificity/PPV and NPV of tests as compared to the blood smear as the gold
standard on the day of inclusion for 460 subjects are summarised for each test with 95%
confidence intervals in the following tables. Table 12 includes all parasites species, whereas
table 13 includes Plasmodium falciparum only. The highest value in each column is in bold.
Values above a level of 90% are starred*.


                   PARACHECK            VISTAPAN          CARESTART        PARABANK
Sensitivity              94%*             91.9%*             95.6%*           84.7 %
                      [90.2 - 96.6]      [87.8 – 95]       [92.2 – 97.8]    [79.6 – 88.9]
Specificity             87.3%             89.6%              91.5%*           94.3%*
                      [82.0 – 91.4]     [84.7 – 93.4]      [86.9 – 94.9]    [90.3 – 97.0]
PPV                     89.6%             91.2%*             92.9%*           94.6%*
                      [85.3 – 93]        [87 – 94.4]       [89.1 – 95.8]    [90.7 – 97.2]
NPV                     92.5%*            90.5%*             94.6%*            84.0%
                      [87.9 – 95.7]     [85.7 – 94.1]      [90.6 – 97.3]    [78.7 – 88.4]

Table 12. Sensitivity, specificity, PPV and NPV of the 4 RDTs (ALL SPECIES)




                                              33
                    PARACHECK           VISTAPAN          CARESTART          PARABANK
Sensitivity              97.9%*            93.2%*             96.2%*              86 %
                       [95.1 – 99.3]     [89.2 – 96.1]     [92.9 – 98.2]       [80.9 – 90.2]
Specificity              87.3%              89.6%             91.5%*             94.3%*
                       [82.0 – 91.4]     [84.7 – 93.4]     [86.9 – 94.9]       [90.3 – 97]
PPV                      89.5%             90.9%*             92.7%*             94.4%*
                       [85.1 – 93]       [86.6 – 94.2]     [88.6 – 95.6]       [90.5 – 97.1]
NPV                      97.4%*            92.2%*             95.6%*             85.8%
                       [94 – 99.1]       [87.7 – 95.5]      [91.8 – 98]        [80.7 – 90]

Table 13. Sensitivity, specificity, PPV and NPV of the 4 RDTs (FALCIPARUM ONLY)



Comparison of sensitivity and specificity between each of the pan-pLDH tests and Paracheck
shows no significant differences between Vistapan or Carestart vs Paracheck, but Parabank
has a significantly lower sensitivity (p=0.0013) and higher specificity (p=0.02).

Comparison between each of the pLDH tests for sensitivity and specificity showed no
significant difference between Vistapan and Carestart, but both had significantly higher
sensitivities than Parabank (p=0.019 and p<0.001 respectively).

10.2.2 Validity according to age group – Day 0
A summary profile of the blood smear results and the sensitivity/specificity of the tests by age
group is given in the figure below. ‘BS positive’ refers to the total number of positive tests (of
all species). Underneath the sensitivity, the species and frequency is displayed in italics of the
number of tests which were falsely negative.




                                               34
                                                            Study sample: 460




                                under 5: 239                                                    5+: 221



             BS positive: 129                  BS negative: 110            BS positive: 119               BS negative: 102



    % of positive tests (SENSITIVITY)                              % of positive tests (SENSITIVITY)
    and species in false negative tests                            and species in false negative tests




Paracheck     Vistapan      Carestart     Parabank                  Paracheck     Vistapan Carestart Parabank
96.9%         97.7%         97.7%         95.4%                     90.8%         85.7%    93.3%     73.1%
2 Pf, 2 Pv    2 Pf, 1 PV    1 Pf, 2 Pv    5 Pf, 1 Pv                3 Pf, 4 Pv,   14 Pf,   8 Pf      28 Pf,
                                                                    4 Pm          3 Pv               4 Pv

                                               % of negative tests                                         % of negative tests
                                                (SPECIFICITY)                                               (SPECIFICITY)


                           Paracheck Vistapan          Carestart     Parabank            Paracheck Vistapan Carestart        Parabank
                           90.0%     91.8%             92.7%         95.5%               84.3%     87.3%    90.2%            93.1%




Figure 6. Diagrammatic representation of the sensitivity and specificity of rapid tests according
to age group




The validity parameters according to age group are given in the below table. Parameter values
with significant comparative results (ie parameters which differed significantly between age
groups) are shown in bold.




                                                                   35
      Test         Age Group       Sensitivity         Specificity       PPV            NPV
                        <5           96.9%                90%           91.9%           96.1%
                                   (125/129)            (99/110)      (125/136)        (99/103)
                     95% CI       [92.3 – 99.1]    [82.8 – 94.9]     [86 – 95.9]     [90.4 – 98.9]
   Paracheck
                        5+           90.8%               84.3%          87.1%           88.7%
                                   (108/119)            (86/102)      (108/124)        (86/97)
                     95% CI       [84.1 – 95.3]    [75.8 – 90.8]     [79.9 – 92.4]   [80.6 – 94.2]
         Paracheck                  p=0.080             p=0.30         p=0.29          p=0.086
  (comparison of proportions)
                        <5           97.7%               91.8%          93.3%           97.1%
                                   (126/129)           (101/110)      (126/135)       (101/104)
    Vistapan         95% CI       [93.4 – 99.5]        [85 – 96.2]   [87.7 – 96.9]   [91.8 – 99.4]
                        5+           85.7%               87.3%          88.7%            84%
                                   (102/119)            (89/102)      (102/115)        (89/106)
                     95% CI       [78.1 – 91.5]        [79.2 – 93]   [81.4 – 93.8]   [75.6 – 90.4]
          Vistapan                 p=0.0012             p=0.40         p=0.29          p=0.003
  (comparison of proportions)
                        <5           97.7%               92.7%          94%             97.1%
                                   (126/129)           (102/110)      (126/134)       (102/105)
                     95% CI       [93.4 – 99.5]    [86.2 – 96.8]     [88.6 – 97.4]   [91.9 – 99.4]
   Carestart
                        5+           93.3%               90.2%          91.7%            92%
                                   (111/119)            (92/102)      (111/121)        (92/100)
                     95% CI       [87.2 – 97.1]    [82.7 – 95.2]     [85.3 – 96]     [84.8 – 96.5]
          Carestart                  p=0.17              p=0.69        p=0.64          p=0.19
  (comparison of proportions)
                        <5           95.3%               95.5%          96.1%           94.6%
                                   (123/129)           (105/110)      (123/128)       (105/111)
                     95% CI       [90.2 – 98.3]    [89.7 – 98.5]     [91.1 – 98.7]   [88.6 – 98]
   Parabank
                        5+           73.1%                93.1          92.6%           74.8%
                                    (87/119)            (95/102)       (87/94)         (95/127)
                     95% CI       [64.2 – 80.8]    [86.4 – 97.2]     [85.3 – 97]     [66.3 – 82.1]
          Parabank                  p<0.001             p=0.65          p=0.4          p<0.001
  (comparison of proportions)

Table 14. Validity according to age group.


A t-test of parasitaemias according to the Parabank result for the 5+ age group showed a
significant difference in the parasitaemia for a negative and positive test (p=0.008). This could
explain the sharp decrease in sensitivity between the two age groups.


10.2.3 Sensitivity according to level of parasitaemia – Day 0
Chi2 for trend was performed with the result of each RDT according to the categories of
density for day 0 results. Each Chi2 showed a significant reduction in the number of false


                                                  36
negative tests as the parasitaemia increased. Sensitivities at each level for each test are shown
in the below table.


                                     Parasitaemia level (parasites/µl)
                       1 – 99       100 – 199    200 – 499        ≥500        χ2
Paracheck               77.4%         83.3%         92.3%        97.4%     p<0.0001
                       (24/31)       (10/12)       (12/13)     (184/192)
Vistapan                58.1%         91.7%         76.9%        98.4%     p<0.0001
                       (18/31)       (11/12)       (10/13)     (189/192)
Carestart               67.7%         100%          92.3%        100%      p<0.0001
                       (21/31)       (12/12)       (12/13)     (192/192)
Parabank                41.9%         58.3%         46.2%        95.8%     p<0.0001
                       (13/31)        (7/12)        (6/13)     (184/192)

Table 15. Sensitivities of the RDTs at different levels of parasitaemia



Chi2 for trend stratified by age group gave significance levels of p<0.01 except for Paracheck
in patients aged 5 and above, where p=0.48, ie that there was no significant difference in the
sensitivity of Paracheck adjusting for parasite level in the older age group.

The below table shows an analysis of the sensitivity, stratified by age group. Sensitivity is
explored at different levels of parasites using different minimum levels of parasitaemia - ie
sensitivity at parasitaemias of at least 100 parasites/µl, at least 200 parasites/µl etc. The
sensitivity was compared between the age groups using a comparison of proportions test, and
the p-value is displayed below, with significant results highlighted in bold.




                                                37
                 Age                                 Parasitaemia category
                group
                             0 vs 1 - 99         ≥100*               ≥200*              ≥500*
                  <5         70% (7/10)     99.2% (118/119)     100% (116/116)     100% (112/112)
                95% CI      [34.8 – 93.3]      [95.4 – 100]       [96.9 – 100]       [96.8 – 100]
                  5+        81% (17/21)      92.9% (91/98)       93.3% (83/89)      93.8% (75/80)
  Paracheck     95% CI      [58.1 – 94.6]     [85.4 – 97.1]       [85.9 – 97.5]       [86 – 97.9]
                p-value         0.82              0.037              0.016               0.026
                Overall    77.4% (24/31)    96.3% (209/217)     97.1% (199/205)    97.4% (187/192)
                95% CI      [58.9 – 90.4]     [92.9 – 98.4]       [93.7 – 98.9]       [94 – 99.1]
                  <5         80% (8/10)     99.2% (118/119)     99.1% (115/116)    99.1% (111/112)
                95% CI      [44.4 – 97.5]      [95.4 – 100]       [95.3 – 100]       [95.1 – 100]
                  5+       47.6% (10/21)     93.9% (92/98)       94.4% (84/89)      97.5% (78/80)
   Vistapan     95% CI      [25.7 – 70.2]     [87.1 – 97.7]       [87.4 – 98.2]      [91.3 – 99.7]
                p-value         0.19              0.071               0.11               0.77
                Overall    58.1% (18/31)    96.8% (210/217)     97.1% (199/205)    98.4% (189/192)
                95% CI      [39.1 – 75.5]     [93.5 – 98.7]       [93.7 – 98.9]      [95.5 – 99.7]
  Carestart       <5        70% (7/10)       100% (119/119)     100% (116/116)     100% (112/112)
                95% CI      [34.8 – 93.3]      [96.9 – 100]       [96.9 – 100]       [96.8 – 100]
                  5+       66.7% (14/21)      99% (97/98)        98.9% (88/89)       100% (80/80)
                95% CI       [43 – 85.4]       [94.4 – 100]       [93.9 – 100]       [95.5 – 100]
                p-value         0.82                0.92              0.89            - (equal Se)
                Overall    67.7% (21/31)    99.5% (216/217)     99.5% (204/205)    100% (192/192)
                95% CI      [48.6 – 83.3]      [97.5 – 100]       [97.3 – 100]       [98.1 – 100]
  Parabank        <5         60% (6/10)     98.3% (117/119)     98.3% (114/116)    98.2% (110/112)
                95% CI      [26.2 – 87.8]     [94.1 – 99.8]       [93.9 – 99.8]      [93.7 – 99.8]
                  5+        33.3% (7/21)     81.6% (80/98)       85.4% (76/89)      92.5% (74/80)
                95% CI       [14.6 – 57]      [72.5 – 88.7]       [76.3 – 92]        [84.4 – 97.2]
                p-value         0.31             0.00007             0.0012              0.11
                Overall    41.9% (13/31)    90.8% (197/217)     92.7% (190/205)    95.8% (184/192)
                95% CI      [24.5 – 60.9]     [86.1 – 94.3]       [88.2 – 95.8]       [92 – 98.2]
*100/200/500 parasites/µl up to the maximum parasitaemia
Table 16. Sensitivity of RDTs according to level of parasitaemia, stratified by age group.




                                               38
10.2.4 Sensitivity and PPV according to presence/absence of measured fever –
Day 0

Measured fever has been shown in previous studies to be the best predictor (although not a
good one) of a positive blood smear. The validity of the tests according to the presence of
fever on Day 0 is overall and for each age group is shown in the below tables. For the table
with age groups, a comparison of proportions test was performed to look for significant
differences in sensitivity and PPV between the age groups. Both tables display a comparison
of proportions p-value comparing the differences in sensitivity and PPV between a subject
with or without fever.


                                          Fever presence or absence
             Axillary temperature ≥37.5°C          Axillary temperature <37.5°C       p-value
                Se (%)         PPV (%)              Se (%)            PPV (%)        Se     PPV


Paracheck   98.8% (84/85)     94.4 (84/89)    91.4% (149/163)     87.1% (149/171)   0.041   0.11
             [93.6 – 100]     [87.4 – 98.2]        [86 – 95.2]      [81.2 – 91.8]
Vistapan    97.6% (83/85)    94.3% (83/88)     89% (145/163)      89.5% (145/162)   0.032   0.29
             [91.8 – 99.7]    [87.2 – 98.1]     [83.1 – 93.3]       [83.7 – 93.8]
Carestart   98.8% (84/85)    95.5% (84/88)    93.9% (153/163)     91.6% (153/167)   0.14    0.38
             [93.6 – 100]     [88.8 – 98.7]         [89 – 97]       [86.3 – 95.3]
Parabank    91.8% (78/85)    96.3% (78/81)     81% (132/163)      93.6% (132/141)   0.040   0.59
             [83.8 – 96.6]    [89.6 – 99.2]     [74.1 – 86.7]         [88.2 – 97]

1Table 17. Sensitivity and PPV of RDTs according to presence/absence of measured fever




                                              39
       Age group                                     Fever presence or absence

                      Axillary temperature ≥37.5°C          Axillary temperature <37.5°C       p-value
                         Se (%)             PPV (%)            Se (%)            PPV (%)      Se     PPV


               <5     98.4% (60/61)    93.8% (60/64)       95.6% (65/68)    90.3% (65/72)    0.69    0.67


                      [91.2 – 100]     [84.8 – 98.3]        [87.6 – 99.1]        [81 – 96]
Paracheck
               5+     100% (24/24)         96% (24/25)     88.4% (84/95)    84.8% (84/99)    0.18    0.25


                      [85.8 – 100]     [79.6 – 99.9]        [80.2 – 94.1]    [76.2 – 91.3]
            p-value       0.63                0.92              0.18               0.41
               <5     96.7% (59/61)    95.2% (59/62)       98.5% (67/68)    91.8% (67/73)    0.92    0.66


Vistapan              [88.7 – 99.6]        [86.5 – 99]      [92.1 – 100]     [83 – 96.9]
               5+     100% (24/24)     92.3% (24/26)       82.1% (78/95)    87.6% (78/89)    0.056   0.76
                      [85.8 – 100]     [74.9 – 99.1]        [72.9 – 89.2]    [79 – 93.7]
            p-value       0.92                0.98             0.0023              0.55
               <5     98.4% (60/61)    95.2% (60/63)       97.1% (66/68)     93% (66/71)     0.92    0.85


                       [91.2 - 100]        [86.7 – 99]      [89.8 – 99.6]    [84.3 – 97.7]
Carestart
               5+     100% (24/24)         96% (24/25)     91.6% (87/95)    90.6% (87/96)    0.31    0.64


                      [85.8 – 100]     [79.6 – 99.9]        [84.1 – 96.3]    [82.9 – 95.6]
            p-value       0.63                0.68              0.27               0.80
               <5     98.4% (60/61)    96.8% (60/62)       92.6% (63/68)    95.5% (63/66)    0.26    0.94


Parabank               [91.2 - 100]    [88.8 – 99.6]        [83.7 – 97.6]    [87.3 – 99.1]
               5+     75% (18/24)      94.7% (18/19)       72.6% (69/95)     92% (69/75)     0.98    0.93
                      [53.3 – 90.2]        [74 – 99.9]      [62.5 – 81.3]    [83.4 – 97]
            p-value       0.020               0.78             0.0026              0.62

  Table 18 . Validity according to fever


  To ascertain whether the differences could be confounded by a relation between presentation
  with fever and higher parasitaemias (leading to higher sensitivity), t-tests were performed to
  check for differences in the means of the parasite density between subjects presenting with a


                                                     40
  without fever. There was a significant increase in the mean parasitaemia of subjects
  presenting with fever overall (p=0.0014) and in the under 5 age group (p=0.028). However,
  the 5 and over age group showed no such difference (p=0.70).


  10.2.5 Sensitivity according to duration of illness
  A t-test to investigate the variation in means of parasitaemia according to the duration of
  illness category showed a significantly higher mean parasitaemia in subjects presenting within
  0 to 2 days (p=0.0012). T-tests repeated for each age group showed that this difference
  persists in the under 5’s (p=0.0004) but not the 5 and above group (p=0.51).
  There were no significant difference in sensitivity between patients who attended the health
  centre before 2 days (inclusive) and those who came after 2 days (Paracheck p=0.28, Vistapan
  p=0.18, Carestart p=0.33, Parabank p=0.051).


  10.2.6 Sensitivity according to history of taking antimalarials
  T-tests comparing the mean densities overall and by age group of subjects who had/had not
  taken antimalarials before attending the hospital were performed. The densities did not differ
  significantly either overall (p=0.25), in the under 5’s (p=0.24) or in the 5+ (p=0.36).
  Sensitivity according to a history of taking antimalarials in the previous 14 days before
  attending the clinic was analysed, and did not differ significantly between any of the tests.


  10.3 Inter-reader reliability
  10.3.1 Invalid and doubtful tests
  The number of invalid and doubtful tests on each visit day of the study are shown below.
  Denominators differ between days due to either missed follow-up visits or the 5 subjects who
  became positive on follow-up and began again from day 0 (therefore missing visits from day
  7 or 14 on the initial CRF). One invalid Carestart was not repeated on day 14. However, it is
  clear that the overall numbers of invalid or doubtful tests for each RDT are small, all less than
  1.5%.

                                                            Follow-up day
                   Day 0                Day 3                     Day 7                Day 14               TOTAL
                   n=460                n=285                     n=286                n=279                N=1310
             Invalid   Doubtful   Invalid   Doubtful        Invalid   Doubtful   Invalid   Doubtful   Invalid   Doubtful
              n (%)     n (%)      n (%)     n (%)           n (%)     n (%)      n (%)     n (%)      n (%)     n (%)

Paracheck      0           0        0           0             0       2 (0.7)      0            0       0       2 (0.2)
                                                                                                       15
Vistapan     6 (1.3)   3 (0.7)    2 (0.7)   3 (1.1)     3 (1.0)           0      4 (1.4)        0               6 (0.5)
                                                                                                      (1.1)
                                                                                                       10
Carestart    2 (0.4)   1 (0.2)    3 (1.1)   1 (0.4)     2 (0.7)       4 (1.4)    3 (1.1)        0               6 (0.5)
                                                                                                      (0.8)
Parabank     1 (0.2)       0      1 (0.4)       0             0       2 (0.7)      0            0     2 (0.2)   2 (0.2)

  Table 19. Percentage and number of invalid and doubtful tests according to day of follow-up




                                                       41
10.3.2 Inter-reader reliability on day of diagnosis (D0)
Results for inter-reader reliability for each RDT on the day of diagnosis (day 0) are given in
the below table, with Kappa presented for each test. A Kappa of 0.8 or more is considered to
be acceptable.


                                                     READER 1             Kappa
                                          Positive    Negative   Total
                               Positive        256       5        261
               Paracheck       Negative         1       198       199       0.97
                               Total           257      203       460


                               Positive        237       9        246
   READER 2




                Vistapan       Negative         5       209       214       0.94
                               Total           242      218       460


                               Positive        249       7        256
                Carestart      Negative         2       202       204       0.96
                               Total           251      209       460


                               Positive        215       7        222
                Parabank       Negative         2       236       238       0.96
                               Total           217      243       460


Table 20. Inter-reader discordances on day 0




10.4 Percentage of positives on post-treatment follow-up

The total number of subjects eligible for this analysis (ie who had a positive blood smear and
took Coartem) are 241 – 123 in the under 5’s and 118 in the 5 and above group. There were
no positive BS results for any of the above patients on any of the 3 days of follow-up (D3, D7
and D14).

The raw percentage of positive tests on each follow-up visit are given in the below table.
Baseline positives are given in the column ‘Day 0’.




                                               42
                               Day 0       Day 3           Day 7             Day 14
             Paracheck          226         86.2%           80.8%               69.7%
                                        [81.7 – 90.7]   [75.6 – 86.0]      [63.1 – 75.7]
                                          (193/224)       (181/224)          (152/218)
             Vistapan           221         36.1%           23.4%                8.9%
                                        [29.7 – 42.5]   [17.8 – 29.0]       [5.1 – 12.7]
                                           (79/219)        (51/218)           (19/213)
             Carestart          230         42.5%           27.6%                9.5%
                                        [36.1 – 48.9]   [21.8 – 33.4]       [5.6 – 13.4]
                                           (97/228)        (63/228)           (21/221)
             Parabank           204         17.8%            8.9%                4.6%
                                        [12.5 – 23.1]    [5.0 – 12.8]        [1.7 – 7.5]
                                           (36/202)        (18/202)            (9/196)
          Table 21. Percentage of positive tests on each follow-up visit


Comparison between all pLDH tests at each day of follow-up and Paracheck shows that all
pLDH tests have significantly less positives than Paracheck on follow-up (p<0.001 for all,
χ2). Between the pLDH tests, Parabank has a significantly lower rate of positives than
Vistapan or Carestart at days 3 and 7 (p<0.001 for all, χ2 comparison of proportions) but by
day 14 there was no significant difference between the proportion of positive pLDH tests. The
lower rate of positives for Parabank on day 3 could be related to its lower sensitivity - ie a test
which does not detect pLDH at low levels will also become negative faster as pLDH levels
fall after treatment.



10.5 Ease of use

A questionnaire to assess the ease of use of each of the tests and compare their costs and other
characteristics considered to be important in operations. Results for each category were
weighted according to perceived importance to make an overall percentage score for each test.




                                                43
CHARACTERISTIC                                               Paracheck   Vistapan   Carestart    Parabank

I EASE OF PERFORMANCE
i. global ease of performance 1=simple, 0=difficult             1           1           1            1
ii. quality of the instruction     2=excellent,                 1           2           1            1
sheet                              1=average, 0=poor
iii. blood application                                        Loop        Loop       Pipette       Loop
a) ease of filling                 2=easy, 1=moderate,         2           2            1           2
b) ease of emptying onto device    0=difficult                 1           2            2           1
iv. buffer application
a) time to wait after blood        1= immediate, 0=1            1           1           0            1
application                        min delay
b) buffer format                   2=individual vials,          1           2           1            1
                                   1=1 vial per box
II SAFETY                          10=no concerns,              5*         5*           10          5*
                                   5=small concerns, 0=
                                   unsafe
III STABILITY OF THE               10=>24 h, 6=12 to 24         10         10           10          10
RESULT                             h, 4=6 – 12 h, 2=1 to 6
                                   h, 0= <1 hour
IV INTERPRETATION
i. ease of interpretation
a) positive line                     2=always clearly           2           1           1            2
b) control line                      visible, 1=not             2           2           2            2
                                     always clearly
                                     visible, 0=usually
                                     faint
ii. invalid                          3=<0.5%, 2=0.5 –           3           2           2            3
iii. doubtful                        2%, 0=≥2%                  3           2           2            3
V STORAGE
i. shelf-life                        3=>12 months,              3           3           3            3
                                     1=6-12 months,
                                     0=<6 months
ii. temperature stability            6=above 37°C,              6           6           6          4**
                                     4=30-37°C, 2=8-         ≤40°C 24     >37° 4      -20°C -    ≤30°C 24
                                     29°C, 0= Some             mths        mths      30°C >15      mths
                                     items at 4-8°C                       =37° 6       mths
                                                                           mths      =37°C 7
                                                                                       mths
                                                                                     =45°C 2
                                                                                       mths
iii. volume for 1000 tests           1=<1m3, 0=≥1m3             1           1           1            1
                                        Weighting            original score in brackets (before weighting)
I performance                                  3              21 (7)      30 (10)      18 (6)      21 (7)
II safety                                     1.5             7.5 (5)     7.5 (5)     15 (10)      7.5 (5)
III stability of the result                   0.5              5 (10)      5 (10)      5 (10)      5 (10)
IV interpretation                              3             30 (10)      21 (7)       21 (7)     30 (10)
V storage                                      2             20 (10)      20 (10)     20 (10)      12 (6)
GRAND TOTAL                                                    83.5          64          79         75.5
*given score 5 due to concern over potential splashing of blood from the loop into mucous
membranes
**real time studies above 37°C are in progress
Table 22. Ease of use questionnaire results




                                                44
Globally, there are no large differences between the tests in terms of ease of use. They were
all fairly simple to use and interpret, had stable results, and were simple to store with no cold
chain requirement for any test.

There were some small differences between the tests which were felt to be either favourable
or a disadvantage by the study team. The instruction sheet for Vistapan was classed as
excellent due to it’s very clear explanation and the use of photographs (rather than diagrams)
to illustrate the exact position and nature of the lines expected on the test strip. The pipette of
Carestart was slighty harder to use than the loops as it was hard to fill the pipette exactly to
the mark indicated (to collect the correct volume of blood). However, it was found to be easier
to empty the pipette onto the test than the loops, which sometimes had to have slightly more
blood added to them to enable the gap in the loop circle to be full enough to permit the blood
to be dispensed onto the test at this point of contact. The positive line of Carestart was not
always clearly visible and often faint, especially on follow-up. Due to the shiny surface of the
test strip, it was sometimes hard to ascertain whether a line was present or not due to other
reflections on the strip. The Vistapan test strip was also faint on follow-up, and sometimes
would be visible at either side of the strip, but not in the center, leading to doubts whether it
was a positive test.


10.6 Costs

Paracheck: $0.45 per test.
Parabank: $0.80 per test. Large volumes: $0.6 per test.
Carestart:
              number of tests                  cost (US$)
              <5,000                               1.00
              5,001 – 20,000                       0.85
              20,001 – 50,000                      0.75
              50,001 – 100,000                     0.65
              >100,000                             0.60

.Vistapan: $0.70 per test.

Prices provided by the manufacturer to Martine Guillerm (MSF Access Campaign) on
request.


11 Discussion

11.1 pLDH tests compared to Paracheck

The pLDH tests were able to detect species other than Plasmodium falciparum, and thus could
detect more malarial infections. Two of the pLDH tests, Vistapan and Carestart, showed no
significant differences in sensitivity or specificity for Plasmodium falciparum as compared to
Paracheck. Carestart had the highest sensitivity of all tests, and was the only test to have all
four validity parameters above 90% in our sample. This demonstrates that, in terms of
validity, Vistapan or Carestart could be used instead of Paracheck for diagnosis, with a slight


                                                45
preference towards Carestart. Parabank had a significantly lower sensitivity than Paracheck,
thus not being a suitable candidate at present to replace the HRP2 test.

As there were no positive blood smears on any day of follow-up, all rapid tests positive on
day 3 or after were false positives. All of the pLDH tests showed a clear advantage in their
significantly lower rate of false positives after successful treatment than Paracheck on each
follow-up day (p<0.0001 for each pLDH vs Paracheck comparison).

Although the pLDH tests detected all cases of P malariae, Vistapan only detected 2 of the 6 P
vivax infections, Parabank detected 1 of 6, and Carestart 4 of 6. This lower rate of detection
for P vivax may be related to the averagely lower densities of this species (geometric mean of
156 parasites/µl for vivax versus 1472 parasites/µl for malariae). However, as one of the
major advantages of the pLDH tests is supposed to be that they can detect all species of
parasite, the lower rate of detection of vivax will still present a problem for diagnosis. In
settings where vivax is more common, doubt could remain about the presence of vivax with a
negative pLDH test, and antimalarials could still be prescribed. Tests would have to be
evaluated locally as they may perform better in areas where vivax infections are predominant.


11.2 Comparison of pLDH tests

Vistapan and Carestart showed significantly better sensitivity than Parabank, and should be
the pLDH tests of choice. Parabank had a significantly lower rate of false positives than the
other two pLDH tests on days 3 and 7, (probably due to its lower sensitivity), and there was
no significant difference between any of the tests by day 14. The time to negativity shown by
Vistapan and Carestart is very similar to that of OptiMAL-IT, which had 61.6% of tests
becoming negative at day 3, 85.2% at day 7 and approximately 95% negative at day 14 (Iqbal
et al., 2004). Another study in Vietnam showed that, during a classical 28 day follow-up, no
subjects remained positive beyond 14 days after treatment (Huong et al., 2002). Reasons for
persistent pLDH antigenaemia after therapy could include persistent viable asexual-stage
parasitaemia below the detection limit of microscopy, delayed clearance of circulating antigen
(free or in antigen-antibody complex) (Tjitra et al., 2001), rheumatoid factor binding to either
IgM or IgG coating antibody (Grobusch et al., 1999, Singh et al., 2000) and detection of
circulating sexual stages in convalescence (Farcas et al., 2003, Tjitra et al., 2001). The pLDH
tests are therefore an option to monitor drug efficacy, certainly a better option than Paracheck.
However, due to a percentage remaining positive for several days, they should only be used
for follow-up when reliable microscopy is not available.

All pLDH tests were very reliable. The inter-reader reliability on day 0 was excellent,
indicating that wherever these tests are used, with a minimum training, the staff should be
able to give a reliable interpretation of the test.

All pLDH tests were similar in terms of ease of use. Differences between the tests in terms of
the score are mainly related to their different stages of product development, and these scores
will change over the coming year as further improvements and temperature stability studies
are made. There would be no reason to prefer one test over another based on these
characteristics.




                                               46
11.3 Factors affecting validity of pLDH tests

Validity changed with several variables for all the tests. These variables are confounded by
each other, for example lower immunity in the under 5’s leads to higher parasite densities and
more common occurrence of fever on presentation.

Sensitivities and specificities are higher for all pLDH tests in the under 5 age group than over
5, with the NPV and sensitivity being significantly higher for Vistapan and Parabank.
However, as for all RDTs for malaria, the level of sensitivity drops dramatically for parasite
densities below 100/microL, which ca be of concern with certain spcial groups of patients
infected with malaria (eg. pregnant women and individuals infected with HIV).

There are significant increases in the sensitivity with increasing parasitaemia for all tests. The
best overall sensitivity (although not significantly so) in the pLDH tests below 100
parasites/µl was shown by Carestart (67.7% vs 58.1% and 41.9%). All tests reached ≥90%
sensitivity at ≥100 parasites/µl. Parabank was the only test with a sensitivity of <95% at this
parasitaemia. All tests had a sensitivity of ≥95% at a parasitaemia of ≥500 parasites/µl.
Stratifying by age group, the only significant difference in sensitivity was shown by Parabank
at ≥100 and ≥200 parasites/µl.

Significant increases in sensitivity for Vistapan and Parabank with fever did not persist after
stratifying for age group, suggesting confounding by parasitaemia and a lower fever threshold
in under 5’s.


11.4 Potential role of RDTs in reducing the cost of diagnosing and
treating malaria.

As previously mentioned, the majority of malaria cases are currently clinically diagnosed. As
the commonly used antimalarials, namely chloroquine and fansidar, are cheap, the cost of
treating all malaria suspects without confirmation has been manageable. However, with the
change in first line treatment to ACTs, in particular the new first line treatment in Uganda,
artmether-lumefantrine (Coartem), such practices are going to have huge financial costs, as
the ACTs are at least 5 times more expensive than the current regimens.

A simple cost comparison of what it would cost to use an RDT to confirm diagnosis before
giving Coartem with the current system of mainly clinical diagnosis was performed to
illustrate the potential cost savings.

Figures for OPD consultations and the proportion of cases diagnosed as malaria per age group
were taken from the MRRH statistics of 2003. We assumed a cost of Coartem of $1.2 for
under 5’s and $2 for 5 years and above based on the price reductions under the WHO/Global
Fund agreement with Novartis (Snow et al., 2003). Diagnosic costs for the current system
were based on a blood smear for approximately 10% of all out-patients, with an estimated cost
of $0.1 per blood smear, assuming a technician working time of 20 minutes per smear with a
salary of $350 per month (based on 24 working days). The diagnostic costs for RDTs are
based on a cost per test (Carestart) of $0.6, with an estimated 50% positive RDTs in the under
5’s and 30% positive RDTs in the 5+ (based on Epicentre observations of the OPD population
- unpublished data). Technician costs are based on the same salary as above, with a working
time of 5 minutes per rapid test. The resulting costs of Coartem are given in the below table.


                                               47
                                            CURRENT SYSTEM                 RDT SYSTEM
                   OPD         % clinical    Cost of  Cost of     % RDT       Cost of  Cost of
               consultations    malaria     diagnosis Coartem     positive   diagnosis Coartem
                  (2003)                      (US$)    (US$)                   (US$)    (US$)
Under 5’s        11,200          75%          793      10,080      50%        6,316       5,040
5 and above      59,000          20%         4,175     23,600      30%        8,873       7,080
Total            70,200           -          4,968     33,680                 15,189      12,120
Total cost                                        38,648                      27,309

Table 23. Cost comparison for current diagnostic and treatment system versus RDT system


This represents a monetary saving, for this hospital alone in 1 year, of $11,339, ie of 29% of
the current budget for malaria diagnosis and treatment. The public health benefits of using
ACTs in a rational way are the reduction of an environment in which drug resistance can
develop and the improvement in diagnosis for non-malaria patients due to information
required for a differential diagnosis being available. However, clinicians would need to be
educated beforehand about the advantages and characteristics of the rapid tests and should be
confident in the test result to avoid continued overtreatment with antimalarials.


11.5 Study limitations

Validity of the tests very much depends on the quality of the laboratory analysis, and whether
an RDT can be useful in a situation can also depend on the sensitivity and specificity of the
lab technicians ability – ie an RDT may have better sensitivity and specificity than the
standard of microscopy available, and therefore be more useful. A difficulty in comparing the
sensitivity and specificity of these tests with that of other studies lies in method of reading the
blood films. A variety of methods have been used in previous studies, for example reading
between 100 to 300 thick blood film fields before declaring a slide negative, and performing
the count against between 200-4,000 WBCs (Hanschied T, Valada E, 1999). We have used
the recommended WHO method and we believe that the microscopy team, consisting of lab
technicians who have read slides in a research setting for more than 3 years, plus the double
reading of each slide, was enough to ensure accurate results, as was confirmed by the results
of the external quality control.


11.6 Conclusions

   1. Overall, the Vistapan and Carestart pLDH tests are as valid, reliable and easy to use as
      Paracheck. Parabank has slightly lower sensitivity, particularly in over 5’s, which
      should be rectified before considering it to replace Paracheck.
   2. The pLDH tests have a much reduced proportion of false positives after treatment, and
      could be considered for a role in monitoring treatment efficacy in a situation where
      reliable microscopy is not available.
   3. The Carestart pLDH test appears to be better than the other two in our setting, with
      overall validity parameters > 90% for each.



                                                48
  4. Further data needs to be collected concerning the sensitivity of the pLDH tests to other
     species, in particular P vivax, as the sample size of non-falciparum monoinfections
     was too small in this study.
  5. The pLDH tests are more costly than Paracheck, but if costs are reduced the WHO
     recommendation for use of HRP2 tests in Zone 1 (P. falciparum infections only, or
     with non-falciparum species occurring almost always as co-infections with P.
     falciparum as in most areas of sub-Saharan Africa, WHO 2005) could be challenged.



11.7 Recommendations

  •   That pLDH tests (in particular Carestart or Vistapan at this time) could replace
      Paracheck in African diagnostic settings.

  •   That the price of pLDH tests should be reduced by lobbying.

  •   That in epidemiological contexts where Plasmodium vivax is more present, the tests
      should be re-evaluated as their relative sensitivities and specificities may differ.




                                             49
12 Acknowledgements
The work of all Epicentre/MUST staff members is highly appreciated. Thanks to the
clinical team: Rogers Twesigye, Vincent Batwala, Samuel Tweheyo, Carol Nabasumba,
Stephen Mugisha, Job Baguma, Naome Natakunda, Gaudioza Mugabirwe, Beatrice
Ariho, Scovia Sanyu and Allen Asiimwe. Thanks to the lab team: James Kiguli,
Frederick Mutebi, Benon Tumwebaze, Juliet Kyomuhendo, Sulaiman Muwanga, Jeninah
Atwebembeire, Bernice Anyango. Thanks to the clerical/data and logistic/administrative
staff: Paul Orechoff, Alex Muhumuza, Doreen Mpeirwe, Tinka Kyoshabiire, Peronie
Uwera, Dunstan Bangirana, Faith Kabanda and Grace Natukunda.

Many thanks to those who gave support and helpful comments on the protocol and report,
namely Christa Hook (Malaria Working Group), Martine Guillerm and Anthony Moody
(Access to essential drugs campaign) and Michael Makler. Thanks to the representatives
of Orchid, Vista Diagnostics and AccessBio, Dr Sriram, Ian Buchanan and Young Ho
Choi, who always responded promptly to queries and were very co-operative throughout
the study.

Thanks to Francois Nosten for comments on the protocol, and to Stephane Proux
(SMRU) for performing the external quality control.

Finally, thanks to Jean-Paul Guthmann, Patrice Piola and Loretxu Pinoges for friendly
supervision and wise advice.

There are no conflicts of interest in this study. The test manufacturers only provided the
tests and did not participate in the study at any stage.


13 References

Baker J, McCarthy J, Gatton M, Kyle DE, Belizario V, Luchavez J, Bell D, Cheng Q. Genetic
diversity of Plasmodium falciparum Histidine-Rich Protein 2 (PfHRP2) and its effect on the
performance of PfHRP2-based rapid diagnostic tests. J Infec Dis 2005; 192(5):870-7

Beadle C, Long GW, Weiss WR, McElroy PD, Maret SM, Oloo AJ, Hoffman SL. Diagnosis of
malaria by detection of Plasmodium falciparum HRP-2 antigen with a rapid dipstick antigen-
capture assay. Lancet 1994; 343 (8897): 564-8
Belizario VY, Pasay CJ, Bersabe MJ, de Leon WU, Guerrero DM, Bugaoisan VM. Field
evaluation of malaria rapid diagnostic tests for the diagnosis of P. falciparum and non-P.
falciparum infections. Southeast Asian J Trop Med Pub Health 2005; 36(3):552-61

Bell DR, Wilson DW, Martin LB. False-positive results of a Plasmodium falciparum histidine-
rich protein 2-detecting malaria rapid diagnostic test due to high sensitivity in a community with
fluctuating low parasite density. Am J Trop Med Hyg 2005; 73(1): 199-203

Causer LM, Bishop HS, Sharp DJ, Flagg EW, Calderon JF, Keane V, Shah JJ, Macarthur JR,
Maloney SA, Cetron MS, Bloland PB. Rapid malaria screening and targeted treatment of United
States-bound Montagnard refugees from Cambodia in 2002. Am J Trop Med Hyg 2005;
72(6):688-93

                                               50
Cho D, Kim KH, Park SC, Kim YK, Lee KN, Lim CS. Evaluation of rapid immunocapture
assays for diagnosis of Plasmodium vivax in Korea. Parasitol. Res. 2001; 87:445-448
Congpuong K, Bualombai P, Jitchmroen S, Konchom S. Comparison of the OptiMAL rapid test
with routine microscopic examination of Giemsa-stained Thick Blood Film for diagnosis of
malaria. J Med Assoc Thai 2001; 84(3): 357-63
Cooke AH, Chiodini PL, Doherty T, Moody AH, Ries J and Pinder M. Comparison of a lactate
dehydrogenase-based immunochromatographic antigen detection assay (OptiMAL) with
microscopy for the detection of malaria parasites in human blood samples. Am J Trop Med Hyg
1999; 60(2): 173-6.
De Monbrison F, Gerome P, Chaulet JF, Wallon M, Picot S, Peyron F. Comparitive test
performance of two commercial rapid tests for malaria in a non-endemic area. Eur J Clin
Microbiol Infect Dis 2004; 23(10):784-6
Dev V. Relative utility of dipsticks for diagnosis of malaria in mesoendemic areas for
Plasmodium falciparum and P. vivax in northeastern India Vector Borne Zoonotic Dis 2004;
4(2): 123-30
Dondorp A , Desakorn V, Pongtavornpinyo W, Sahassananda D, Silamut K, Chotivanich K,
Newton P, Pitisuttithum P, Smithyman A, White, NJ, Day NP. Estimation of the total parasite
biomass in acute falciparum malaria from plasma PfHRP2. PloS Med. 2005 Aug 2(8):e204. Epub
2005 Aug 23.

Eisen DP, Saul A. Disappearance of pan-malarial antigen reactivity using the ICT Malaria Pf/Pv
kit parallels decline of patent parasitaemia as shown by microscopy. Trans Roy Soc Trop Med
Hyg 2000; 94:169-170

Farcas GA, Zhong KJY, Lovegrove FE, Graham CM, Kain KC. Evaluation of the Binax NOW
ICT test versus polymerase chain reaction and microscopy for the detection of malaria in returned
travellers. Am J Trop Med Hyg 2003; 69(6): 589-592

Ferro BE, Gonzalez IJ, Carvajal F, Palma GI, Saravia NG. Performance of OptiMAL® in the
diagnosis of Plasmodium vivax and Plasmodium falciparum infections in a malaria referral center
in Colombia. Mem Inst Oswaldo CruzI 2002; 97(5): 731-5

Forney JR, Wongsrichanalai C, Magill AJ, Craig LG, Sirichaisinthop J et al. Devices for rapid
diagnosis of Malaria: evaluation of prototype assays that detect Plasmodium falciparum histidine-
rich protein 2 and a Plasmodium vivax-specific antigen. J Clin Microbiol 2003; 41(6): 2358-66

Fryauff DJ, Purnomo, Sutamiharja MA, Elyazar IR, Krisin, Subianto B, Marwoto H. Performance
of the OptiMAL assay for detection and identification of malaria infections in asymptomatic
residents of Irian Jaya, Indonesia. Am J Trop Med Hyg 2000; 63(3-4):139-145
Funk M, Schlagenhauf P, Tschopp A, Steffen R. MalaQuick versus ParaSight F as a diagnostic
aid in travellers’ malaria. Trans R Soc Trop Med Hyg 1999; 93(3):268-272
Gaye O, Diouf M, Diallo S. A comparison of thick smears, QBC malaria, PCR and PATH
falciparum malaria test strip in Plasmodium falciparum diagnosis. Parasite 1999; 6(3):273-275
Grobusch MP, Jelinek T, Hanscheid T. False positivity of rapid antigen detection tests for
diagnosis of Plasmodium falciparum malaria: issue appears to be more complicated than
presented. J Clin Microbiol 1999; 37(11):3781-2
Grobusch MP, Hanscheid T, Gobels K, Slevog H, Zoller T, Rogler G, Teichmann D. Comparison
of three antigen detection tests for diagnosis and follow-up of falciparum malaria in travellers
returning to Berlin, Germany. Parasitol Res. 2003; 89(5):354-357



                                               51
Guthmann JP, Ruiz A, Priotto G, Kiguli J, Bonte L, Legros D. Validity, reliability and ease of use
in the field of five rapid tests for the diagnosis of Plasmodium falciparum malaria in Uganda.
Trans R Soc Trop Med Hyg 2002; 96: 254-257
Hanscheid T, Valadas E. Letter to the editor. Am J Trop Med Hyg 1999; 61(2): 179
Humar A, Ohrt C, Harrington MA, Pillai D, Kain KC. ParaSight® F test compared with the
polymerase chain reaction and microscopy for the diagnosis of Plasmodium falciparum malaria in
travellers. Am J Trop Med Hyg 1997; 56:44–48.
Huong NM, Davis TME, Hewitt S, Huong NV, Uyen TT, Nhan DH, Cong LD. Comparison of
three antigen detection methods for diagnosis and therapeutic monitoring of malaria: a field study
from southern Vietnam. Trop Med Int Health 2002; 7(4):304-308

Iqbal J, Hira PR, Sher A, Al-Enezi AA. Diagnosis of imported malaria by Plasmodium lactate
dehydrogenase (pLDH) and histidine-rich protein 2 (PfHRP2)-based immunocapture assays. Am
J Trop Med Hyg 2001; 64(1-2):20-3

Iqbal J, Khalid N, Hira PR. Comparison of two commerical assays with expert microscopy for
confirmation of symptomatically diagnosed malaria. J Clin Microbiol 2002; 40(12):4675-8

Iqbal J, Siddique A, Jameel M, Hira PR. Persistent Histidine-Rich Protein 2, Parasite Lactate
Dehydrogenase, and Panmalarial antigen reactivity after clearance of Plasmodium falciparum
monoinfection. J Clin Microbiol 2004; 42(9): 4237-4241

Laferi H, Kandel K, Pichler H. False positive dipstick test for malaria. N Engl J Med 1997;
337:1635–1636.
Mason DP, Kawamoto F, Kin K, Laoboonchai A, Wongsrichanalai C. A comparison of two rapid
field immunochromatographic tests to expert microscopy in the diagnosis of malaria. Acta
Tropica 2002;82:51-59
Moody A, Hunt-Cooke A, Gabbett E, Chiodini P. Performance of the OptiMAL® malaria antigen
capture dipstick for malaria diagnosis and treatment monitoring at the Hospital for Tropical
Diseases, London. Br J Haematol 2000; 109:891–894.
Moody AH, Chiodini PL. Non-microscopic method for malaria diagnosis using OptiMAL IT, a
second-generation dipstick for malaria pLDH antigen detection. Br J Biomed Sci 2002;
59(4):228-31

Palmer CJ, Lindo JF, Klaskala WI, Quesada JA, Kaminsky R, Baum MK, Ager AL. Evaluation
of the OptiMAL test for rapid diagnosis of Plasmodium vivax and Plasmodium falciparum
malaria. J Clin Microbiol 1997; 36(1): 203-6

Parra ME, Evans CB, Taylor DW. Identification of Plasmodium falciparum histidine-rich protein
2 in the plasma of humans with malaria. J Clin Microbiol 1991; 29(8): 1629-1634

Pattanasin S, Proux S, Chompasuk D, Luwiradaj K, Jacquier P, Looareesuwan S, Nosten F.
Evaluation of a new Plasmodium lactate dehydrogenase assay (OptiMAL-IT®) for the detection
of malaria. Trans R. Soc Trop Med Hyg 2003; 97:672-674

Piola P, Fogg C, Bajunirwe F, Biraro S, Grandesso F, Ruzagira E., et al., Supervised versus
unsupervised intake of six-dose artemether-lumefantrine for treatment of acute, uncomplicated
Plasmodium falciparum malaria in Mbarara, Uganda: a randomised trial. Lancet
2005;365(9469):1467-73

Piper R, Lebras J, Wentworth L, Hunt-Cooke A, Houze S, Chiodini P, Makler M. Immunocapture
diagnostic assays for malaria using Plasmodium lactate dehydrogenase (pLDH). Am J Trop Med
Hyg 1999; 60(1):109-118.

                                               52
Ricci L, Viani I, Piccolo G, Fabio A, Calderaro A et al. Evaluation of OptiMAL assay test to
detect imported malaria in Italy. New Microbiol 2000; 23(4): 391-398
Richter J, Gobels K, Muller-Stover I, Hoppenheit B, Hausinger D. Co-reactivity of plasmodial
histidine-rich protein 2 and aldolase on a combined immuno-chromatographic-malaria dipstick
(ICT) as a potential semi-quantitative marker of high Plasmodium falciparum parasitaemia.
Parasitol. Res. 2004; 94(5):384-5
Singh N, Saxena A, Valecha N. Field evaluation of the ICT malaria Pf/Pv
immunochromatographic test for diagnosis of Plasmodium falciparum and P. Vivax infection in
forest village of Chindwara, central India. Trop Med Int Health 2000; 5:765-770
Tjitra E, Suprianto S, Dyer ME, Currie BJ, Anstey NM. Field evaluationof the ICT malaria
P.f/P.v immunochromatographic test for detection of Plasmodium falciparum and Plasmodium
vivax in patients with a presumptive clinical diagnosis of malaria in eastern Indonesia. J Clin
Microbiol 1999; 37(8): 2412-2417
Tjitra E, Suprianto S, Dyer ME, Currie BJ, Anstey NM. Detection of histidine rich protein 2 and
panmalarial ICT Malaria Pf/Pv test antigens after chloroquine treatment of uncomplicated
falciparum malaria does not reliably predict treatment outcome in eastern Indonesia. Am J Trop
Med Hyg 2001 (1); 65(5): 593-598
Tjitra E, Suprianto S, McBroom J, Currie BJ, Anstey NM. Persistent ICT malaria P.f/P.v
panmalarial and HRP2 antigen reactivity after treatment of Plasmodium falciparum malaria is
associated with gametocytaemia and results in false-positive diagnosis of Plasmodium vivax in
convalescence. J Clin Microbiol 2001; (2); 39(3): 1025-1031
van dan Ende J, Vervoort T, van Gompel A, Lynen L. Evaluation of two tests based on the
detection of histidine rich protein 2 for the diagnosis of imported Pplasmodium falciparum
malaria. Trans R Soc Trop Med Hyg 1998; 92(3): 285-288
WHO New perspectives for malaria diagnosis. Report of a joint WHO/USAID informal
consultation 25-27 October 1999 (WHO/CDC/RBM/2000.14).
WHO Management of severe malaria. 2000.
WHO (1991). Basic malaria microscopy, Part I & II. WHO, Geneva.
WHO Position of WHO's Roll Back Malaria Department on malaria treatment policy. WHO,
Geneva. http://www.who.int/malaria/treatment.html 2003
WHO The use of malaria rapid diagnostic tests. 2004.
WHO Interim notes on the selection of type of malaria rapid diagnostic test in relation to the
occurrence of different parasite species. RBM August 2005.
Wolday D, Balcha F, Fessehaye G, Birku Y, Shepherd A. Field trial of the RTM dipstick method
for the rapid diagnosis of malaria based on the detection of Plasmodium falciparum HRP-2
antigen in whole blood. Tropical Doctor 2001; 31(1): 19-21
Wu YS, Li M, Lu JH, Bi HX, Li YJ. Plasmodium lactate dehydrogenase activity assay and its
preliminary application in therapeutic efficacy monitoring of drugs. Di Yi Jun Yi Da Xue Xue Bao
2001; 22(1): 25-27




                                              53
14 Appendices

14.1 List of standard operating procedures
14.1 Clinical SOPs
   1. Screening
   2. Physical examination
   3. Informed consent
   4. Administering Coartem
   5. Supervision form completion
   6. CRF completion procedure
   7. Use of thermohygrometer


14.2 Laboratory SOPs
   1. RDT stability assessment
   2. Fingerprick
   3. Making blood smears
   4. Giemsa staining
   5. Reading a blood smear
   6. Haemocue
   7. Pregnancy test
   8. Vistapan
   9. Parabank
   10. Carestart
   11. Paracheck
   12. Completion of ease of use questionnaire




                                          54
14.2 Informed consent form

    INFORMED CONSENT STATEMENT FOR THE VALIDATION STUDY
Study title: Validation and ease of use of pLDH Rapid Diagnostic Tests for the diagnosis
of uncomplicated P. falciparum malaria, Mbarara (Uganda), 2005
Introduction: We are investigators from Epicentre and Mbarara University and we are
carrying out a study on rapid diagnostic tests for malaria. These are tests which are
supposed to make accurate malaria diagnosis easier in certain situations, as compared
with the standard method of using the blood slide and a microscope. A new type of rapid
test has been developed which is thought to be more accurate and also cheaper than the
existing test which is currently on the market. It can also detect the presence of any
species of malaria.
Study purpose: We would like to identify a suitable rapid diagnostic test for the
diagnosis of malaria. A test will be considered as suitable if it gives an accurate result, is
reliable, easy to use and cheap.

Study process: You/your child are suspected to have malaria and we would like to
confirm this by laboratory diagnosis. We will perform a finger prick and then will make a
blood smear, which is the usual method of malaria diagnosis. We will then use an
additional 4 drops of blood for comparison of the blood smear with 4 new rapid
diagnostic tests for malaria. No other test will be done on you/your child’s blood. If the
blood smear is positive, women over the age of 14 years will then provide urine for a
pregnancy test, as the antimalarial drug we will use is not yet registered for pregnant
women. Women who are pregnant will be treated with another drug to take at home, and
will not be followed up after the treatment, unless their condition deteriorates. For all
other participants, if the blood smear shows that you have malaria, you/your child will be
admitted on the supervision ward and given antimalarial treatment for 3 days. You will
be seen in the clinic on day 3, and also be to return to the clinic on day 7 and day 14 (i.e.
a period of 2 weeks). On these visits you/your child will be asked again to have a finger
prick for a blood smear plus 4 drops of blood for the rapid diagnostic tests. This will help
us know if the malaria parasites have cleared from the blood. If you do not have malaria
on day 0, you will not be asked to return to the clinic.

Confidentiality: The information obtained from the participants and the results of the
tests done will be kept under lock and key and will only be accessible to the researchers
and the ethics committees or the regulatory authorities. Your medical records will be kept
confidential. No records used for the results of the study will have your name on them.

Risks and benefits from participating in the study: By participating in this study you
will be fully investigated for malaria and you will be treated according to the results of
the blood smear and the assessment of the clinician. If you have malaria, you will be
treated with an antimalarial called Coartem, which is a fully registered drug on the
market. There are no obvious risks involved in participating in this study.

Withdrawal from the study and compensation: You can ask the doctor any questions
about the study. Your participation in this study is entirely voluntary and you are free to

                                             55
withdraw at any time. Withdrawal from this study will not prejudice your care or your
rights to receive treatment at the health unit. Your participation in the study will not
necessitate any direct monetary cost to you or your family. On days that you have been
asked to return to the clinic, Epicentre will refund your transport fare to and fro. If
you/your child have other health conditions that don’t improve you/your child will be
referred to MRRH for futher manangement.

Epicentre is based in Mbarara Hospital. If you want to contact us, please find Ms Carole
Fogg or Dr Rogers Twesigye in the Epicentre office within Mbarara Hospital or ring
telephone number 077 721748.

I have read and understood the above information and my questions have been answered
to my satisfaction. I give voluntary consent to my participation in this study.

I understand that I am free to withdraw from the study at any time and that I will continue
to receive the care I am entitled to.

Patient’s Name ……….………………………………………………
Patient’s signature/fingerprint ………………………………….                         Date …………….
Parent/guardian’s Name ……….………………………………………………
Parent/guardian’signature/finger print …………………………                       Date ……………
(if the participant is below 18 years)

Investigator’s Name ……….…………………………………
Investigator’s Signature . ………………………………….                               Date ……………

If oral consent given (if the participant or the parent/guardian cannot read) :
Witness Name ……….………………………………………………….
Witness signature ……………………………………………                                     Date ……………




                                             56
14.3 Case Report Form
               Validation study of pLDH Rapid Diagnostic Tests in uncomplicated P. falciparum malaria
                                          CASE REPORT FORM - PAGE 1
Country: Uganda                                                                                        Study Centre: Mbarara
1. SUBJECT IDENTIFICATION


Screening number: |__| / |__|__|__|            Subject Initials: |__|__|
                                                                                       Affix validation inclusion

                                                                                            number label here


2. DEMOGRAPHIC INFORMATION

Gender:        1=Male 2=Female |__|

Date of Birth: (dd/mm/yyyy)              |__|__|/|__|__|/|__|__|__|__|          Age: (years/months) |__|__|/|__|__|

Weight:       |__|__|.|__| kg

3. ELIGIBILITY (circle Y or N as appropriate)


Inclusion Criteria:                                                            Exclusion Criteria:
a) History of fever in last 24 hours and/or temperature ≥37.5oC     Y      N a) Signs of severe malaria (WHO 2000)    Y   N
b) Weight ≥ 5 kg                                                    Y      N b) Pregnancy                             Y   N
c) Resident in Mbarara municipality                                 Y      N c) Severe illness                        Y   N
d) Available for follow-up for 2 weeks                              Y      N
e) Signed informed consent                                          Y      N


To be included all Inclusion criteria have to be YES and all Exclusion criteria have to be NO


4. REGISTRATION
NB Make sure you have put the signed informed consent form in the ‘consent form file’.

Clinician name:                                                  Signature: _______________________________


                                                                        Date: |__|__|/|__|__|/2005




                                                            57
5. CLINICAL EXAMINATION D0
Axillary Temperature (°C)                |__|__| . |__|
Pulse (beats per minute)                 |__|__|__|
BP (mm/Hg)                               |__|__|__| / |__|__|__|
History of taking antimalarials
in the previous 2 weeks (Y/N)
                                         |__| if Y, specify:                         Duration of current illness :      |__|__|
Drug 1                            Drug :                                             Days of treatment prescribed:     |__|__|
                                  Date of first                                      Date of last
                                  dose:
                                                   |__|__|/|__|__|/|__|__|__|__| dose:                |__|__|/|__|__|/|__|__|__|__|
Drug 2                            Drug :                                             Days of treatment prescribed:     |__|__|
                                  Date of first                                      Date of last
                                  dose:
                                                   |__|__|/|__|__|/|__|__|__|__| dose:                |__|__|/|__|__|/|__|__|__|__|
Drug 3                            Drug :                                             Days of treatment prescribed:     |__|__|
                       Date of first                                             Date of last
                       dose:
                                                   |__|__|/|__|__|/|__|__|__|__| dose:        |__|__|/|__|__|/|__|__|__|__|
6. LABORATORY RESULTS D0
RDT:                                              Reader 1 Doubtful? y/n Reader 2 Doubtful? y/n Reader 3 Doubtful? y/n
1: Paracheck                                        |__|            |__|          |__|         |__|             |__|       |__|
                    Codes : 0=negative
2: Vista-pan                1=positive              |__|            |__|          |__|         |__|             |__|       |__|
                            9=not done
3: CareStart                                        |__|            |__|          |__|         |__|             |__|       |__|
4: Parabank                                         |__|            |__|          |__|         |__|             |__|       |__|
Any invalid result? (Y/N)         |__|        if Y, which test? [1-4]      |__|
Blood Smear:
Asexual forms                            0= Negative 1=Positive|__|      if positive:
     Species (use initials F, V, O, M)                          |__|__|
     Parasitaemia                                              |__|__|__|,|__|__|__| /µl
Gametocytes                             0= Negative 1=Positive |__|        if yes:
     Gametocyte density                                         |__|__|,|__|__|__| /µl
Third reading necessary? (Y/N)                                 |__|     if Yes, specify nature of discordance:
                                       A=pos/neg B=species C=density D=gametocyte E=other                |__|__|__|
For women aged 14+ years with a          HCG result: 0=Negative 1=Positive 9=Not Done                 |__|
positive BS/RDT:
NOTE: Follow-up to continue ONLY if the study subject has at least one positive RDT result and/or a positive BS
result, and has a negative pregnancy test (women aged 14+ years).
Follow-up to continue? (Y/N)                |__|           Note: complete pages 3 and 4 of CRF only if follow-up IS to continue.
7. TREATMENT
Coartem tabs/dose (1-4) |__|
Rescue treatment given? (Y/N)              |__|      If Yes, reason:              1=vomited twice        |__|
                                                                                  2=severe malaria
                                                                                  3=other (specify)
Other treatment given:




                                                               58
8. CLINICAL EXAMINATION D3
Date: ___/___/2005                       Clinician Initials:                                Subject seen? (Y/N) |__|
Axillary Temperature (°C)                     |__|__| . |__|
Pulse (beats per minute)                      |__|__|__|
BP (mm/Hg)                                    |__|__|__| / |__|__|__|
9. LABORATORY RESULTS D3
RDT:                                              Reader 1 Doubtful? y/n Reader 2 Doubtful? y/n       Reader 3    Doubtful? y/n
1: Paracheck                                        |__|            |__|    |__|            |__|         |__|            |__|
                    Codes : 0=negative
2: Vista-pan                1=positive              |__|            |__|    |__|            |__|         |__|            |__|
                            9=not done
3: CareStart                                        |__|            |__|    |__|            |__|         |__|            |__|
4: Parabank                                         |__|            |__|    |__|            |__|         |__|            |__|
Any invalid result? (Y/N) |__|      if Y, which test? [1-4] |__|
Blood Smear: Asexual forms     0= Negative 1=Positive |__|                   if positive:
     Species (use initials F, V, O, M)                            |__|__|
     Parasitaemia                                                |__|__|__|,|__|__|__| /µl
Gametocytes                               0= Negative 1=Positive |__|        if yes:
     Gametocyte density                                           |__|__|,|__|__|__| /µl
Third reading necessary? (Y/N)                                   |__|     if Yes, specify nature of discordance:
                                         A=pos/neg B=species C=density D=gametocyte E=other                |__|__|__|
10. CLINICAL EXAMINATION D7
Date: ___/___/2005                   Clinician Initials:                                    Subject seen? (Y/N)   |__|
Axillary Temperature (°C)                     |__|__| . |__|
Pulse (beats per minute)                      |__|__|__|
BP (mm/Hg)                                    |__|__|__| / |__|__|__|
11. LABORATORY RESULTS D7
RDT:                                              Reader 1 Doubtful? y/n Reader 2 Doubtful? y/n       Reader 3    Doubtful? y/n
1: Paracheck                                         |__|           |__|    |__|            |__|         |__|            |__|
                    Codes : 0=negative
2: Vista-pan                1=positive               |__|           |__|    |__|            |__|         |__|            |__|
                            9=not done
3: CareStart                                         |__|           |__|    |__|            |__|         |__|            |__|
4: Parabank                                          |__|           |__|    |__|            |__|         |__|            |__|
Any invalid result? (Y/N)        |__|   if Y, which test? [1-4] |__|
Blood Smear:         Asexual forms 0= Negative 1=Positive |__|               if positive:
     Species (use initials F, V, O, M)                            |__|__|
     Parasitaemia                                                |__|__|__|,|__|__|__| /µl
Gametocytes                               0= Negative 1=Positive |__|        if yes:
     Gametocyte density                                           |__|__|,|__|__|__| /µl
Third reading necessary? (Y/N)                                   |__|     if Yes, specify nature of discordance:
                                         A=pos/neg B=species C=density D=gametocyte E=other                |__|__|__|




                                                               59
                Validation study of pLDH Rapid Diagnostic Tests in uncomplicated P. falciparum malaria
                                           CASE REPORT FORM - PAGE 4
Screening number: |__| / |__|__|__|              Inclusion number: |__|__|__|         Subject Initials: |__|__|
12. CLINICAL EXAMINATION D14
Date: ___/___/2005                    Clinican Initials:                             Subject seen? (Y/N) |__|
Axillary Temperature (°C)                   |__|__| . |__|
Pulse (beats per minute)                    |__|__|__|
BP (mm/Hg)                                  |__|__|__| / |__|__|__|
13. LABORATORY RESULTS D14
RDT:                                           Reader 1 Doubtful? y/n Reader 2 Doubtful? y/n       Reader 3       Doubtful? y/n
1: Paracheck                                      |__|          |__|          |__|    |__|           |__|             |__|
                   Codes : 0=negative
2: Vista-pan               1=positive             |__|          |__|          |__|    |__|           |__|             |__|
                           9=not done
3: CareStart                                      |__|          |__|          |__|    |__|           |__|             |__|
4: Parabank                                       |__|          |__|          |__|    |__|           |__|             |__|
Any invalid result? (Y/N)       |__|         if Y, which test? [1-4]   |__|
Blood Smear:
Asexual forms                          0= Negative 1=Positive  |__|      if positive:
     Species (use initials F, V, O, M)                          |__|__|
     Parasitaemia                                              |__|__|__|,|__|__|__| /µl
Gametocytes                             0= Negative 1=Positive |__|        if yes:
     Gametocyte density                                         |__|__|,|__|__|__| /µl
Third reading necessary? (Y/N)                                 |__|     if Yes, specify nature of discordance:
                                       A=pos/neg B=species C=density D=gametocyte E=other                |__|__|__|


Notes:




                                                           60
14.4 SOPs for Rapid Diagnostic Tests


                                 SOP PARACHECK



 Identification number :


1. FUNCTION OF THE TEST AND LIMITATIONS
Paracheck is a rapid self-performing, qualitative, two site sandwich immunoassay for the
determination of P. falciparum specific histidine rich protein II (Pf HRP2) in whole blood
samples.

The test detects P. falciparum only and can remain positive for up to two weeks after a
successful treatment.
As with all diagnostic tests, the results must be correlated with other clinical findings, and
should be interpreted within an epidemiological, clinical and therapeutic context.
A parasitological technique (microscopic examination of a stained blood smear) should
be considered for comparision as a ‘gold standard’.


2. PRINCIPLE OF THE TEST
Paracheck P.f is a rapid test for the detection of P.falciparum malaria that utilizes the
principle of immunochromatography. As the test sample flows through the membrane
assembly of the device after addition of the clearing buffer, the coloured anti-Pf HRP2
colloidal gold conjugate (monoclonal) anti-sera complexes the Pf HRP2 in the lysed
sample. This complex moves futher on the membrane to the test region where it is
immobilised by the anti-Pf HRP2 (monoclonal) antisera coated on the membrane, leading
to the formation of a pink coloured band which indicates a positive test result. Absence
of this coloured band in the test region indicates a negative test result. The unreacted
conjugate and unbound complex, if any, move further on the membrane and are
subsquentily immobilised by anti-rabbit antibodies coated on the membrane at the control
region, forming a pink band. This control band serves to validate the test perfomance.

3. SPECIMEN COLLECTION
Materials
       - Cotton wool
       - Methylated spirit
       - Blood lancets
       - Sharps container
       - gloves
       - 5µl sample loop (inside test device pouch)
       - Disposal bucket for soiled materials
       - Marker/pencils

                                             61
Procedure
      1. Seat the patient in a comfortable position and explain to him or her what you
          are about to do.
       2. Put on gloves, open the pouch and remove the test device plus the sample
          loop.
       3. Label the test device with the patient initials, inclusion number and date.
       4. Prepare the lancet on the table with the non-sharp end open.
       5. With an alcohol swab, clean the site of the finger to be pricked.
       6. Squeeze the end of the finger tip and make a prick with a sterile lancet and
          wipe off the first drop of blood with a clean, dry cotton swab. (Refer to the
          SOP Fingerprick).
       7. Using the sample loop provided, collect a loopful of blood by touching the
          drop of blood with a loop.
       8. Give the patient a clean cotton swab to apply to the puncture site.

4. EQUIPMENT
     - Paracheck test kit
     - digital timer

5. REAGENTS/STAINS
      - clearing buffer in a dropper bottle

6. METHOD OF CONTROL
The test device contains an internal control in the form of a control band. If the control
band does not appear after the test has been run, discard the test and repeat with a new
one.
Microscopic confirmation, especially for patients who have recently taken antimalarial
treatment, is important for cross-checking the result.

7. TEST METHOD
      1. Place one loop of blood into the sample well by pressing the loop gently
         against the bottom of the sample well and allowing the blood to flow out of
         the loop. Avoid touching the loop on other areas of the test or the sides of the
         well, as blood will be lost.
       2. Dispense six drops (300µl) of the clearing buffer into the buffer well whilst
          holding the plastic dropper bottle vertically. Avoid allowing any buffer to
          drop into the sample well.
       3. Set the timer for 15 minutes.
       4. At 15 minutes, read the test immediately. Then pass the test to the second
          reader.




                                            62
8. REPORTING RESULTS / INTERPRETATION

The test may be classified as positive, negative or invalid (as shown below). If the reader
cannot give an immediate result for the test, it should be indicated that the test is
doubtful. Results should be recorded on the RDT result form and in the Reader’s register.

Key
C = Control window
T= Test window
A= Sample well
B= Buffer well

NEGATIVE for P. falciparum malaria:


                                                                 Only one pink coloured
          C                       T               B              band in the control
                                                                 window ‘C’.

          Figure 1 Paracheck negative result


POSITIVE for P. falciparum malaria:


                                                                 In addition to the control band,
          C                       T               B              a distinct pink band also appears
                                                                 in the test window ‘T’.

         Figure 2 Paracheck positive result


INVALID test results:



          C                       T               B
                                                                 If the test band appears and the
                                                                 control band does not appear as
                                                                 in figure 3 above, or if no band
          Figure 3 Invalid result (i)                            appears on the test device as in
                                                                 figure 4, the test should be
                                                                 considered invalid. INVALID
                                                                 TESTS MUST ALWAYS BE
          C                       T               B              REPEATED.



                                            63
          Figure 4 Invalid result (ii)

If the result of the test has to be thought about for more than a few seconds, the reader
should also record ‘doubtful’ = Yes on the RDT reader form.

9. SOURCES OF ERROR
    - Defect of the test device as a result of manufacturing error or due to improper
      storage conditions.
    - Use of expired test devices or clearing buffer.
    - Storage out of +2°C to 30°C temperature range.


10. REFERENCES
Orchid Biomedical System, India information leaflet. (Manufacturer of Paracheck
device.)



                                   SOP VISTA-PAN


 Identification number :


1. FUNCTION OF THE TEST AND LIMITATIONS
The Vista-pan rapid diagnostic test (Vista diagnostics) is a visual, qualitative
immunoassay for the in vitro detection of plasmodium Lactate Dehydrogenase (pLDH), a
protein specific to parasites of the genus Plasmodium.

Limitations:
The test is intended as an aid in the diagnosis of malaria infection in humans, however it
is only a screening test and does not distinguish between falciparum and non-falciparum
malaria.The results are interpreted within the epidemiological, clinical context. When it is
indicated, the parasitological techniques of reference should be considered for
confirmation (examination of thin and thick blood smears).

Once a pouch is opened, the device should be used within one hour.
For in vitro diagnosis, use only with whole blood. Not for use with serum samples.

2. PRINCIPLE OF THE TEST
Vista pan-Malaria Test detects pLDH, an enzyme produced by the four species of
plasmodium known to cause malaria. A sample of blood is added to the sample window
and lysis buffer is added to a second window. The buffer contains a detergent that lyses
the red blood cells and releases the cell contents. If pLDH is present, this forms a
complex with colloidal gold anti-pLDH conjugate that is dried onto the test strip. The
liquid migrates through the nitrocellulose, and if colloidal gold–antibody- pLDH complex
is present, this binds to a second anti-pLDH antibody immobilized on the nitrocellulose,

                                             64
forming a visible purple line. To ensure assay validity, a second control line is
incorporated into the assay device.
3. SPECIMEN COLLECTION
 Materials
    - 5µl sample loop (inside test pouch)
    - sterile blood lancet
    - cotton wool
    - sharps container
    - gloves
    - methylated spirit
    - marker/pencil
    - disposal bucket

 Procedure
   1. Seat the patient in a comfortable position and inform him or her what you are
      about to             do.
   2. Put on gloves, open the pouch and remove the test device plus the sample loop.
   3. Label the test device with the patient initials, inclusion number and date.
   4. Open the lancet packet from the non-sharp end and place securely on the table.
   5. Using an alcohol swab, clean the site of the finger to be pricked.
   6. Squeeze the end of the fingertip and prick with a sterile lancet. (Refer to SOP
      Fingerprick).
   7. Wipe the first drop of blood with a clean, dry cotton swab
   8. Using the sample loop provided, collect a loopful of blood by touching the drop
      of blood with a loop.
   9. Give the patient a clean swab to press to the puncture site.

4. EQUIPMENT
    - Vista-pan test device
    - timer

5. REAGENTS AND STAINS
    - Clearing buffer (individual vial)

6. METHOD OF CONTROL
The test device has an internal control – ie the control band, which should appear to
confirm validity of the test. Any test in which the control band does not appear should be
discarded and a new test performed.

7. TEST METHOD
    1. If the test has been kept in a cold place (below room temperature), only open the
       pouch when the test has attained room temperature to prevent formation of
       condensation.


                                           65
   2. Place the loop containing blood (5 microlitres) in the sample well of the test
      device (shown below). Press gently against the bottom of the sample window and
      allow blood to flow out of the loop.




Figure 1      Photograph of Vista-Pan test device


   3. Twist off the seal of the clearing buffer and transfer the entire contents (150µl) to
      the buffer window.
   4. Set the timer for 20 minutes. The test should be read as soon as red colour of
      blood has cleared from the nitrocellulose (but not before 20 minutes). If the blood
      has not yet cleared by the end of the 20 minute period, set the timer for a further
      20 minutes.
   5. After the first reader has recorded the result, pass the test to the second reader
      immediately.

8. REPORTING RESULTS/INTERPRETATION




      Figure 2        Positive and negative Vista-Pan results

   1. A positive test (reaction) shows two coloured lines, one in the test area , one in
      the control area as shown in Figure 2.
   2. The test should be considered positive if any visible line is evident in the test line,
      even if it is very faint.
   3. A negative test shows only one coloured line in the control area.

                                            66
   4. If no lines are visible, or only the test line apears, then the test is invalid. ALL
      INVALID TESTS MUST BE REPEATED.
   5. If the result of the test has to be thought about for more than a few seconds, the
      reader should also record ‘doubtful’ = Yes on the RDT reader form.
   6. The results should be reported on the RDT results form and in the Reader’s
      register.

9. SOURCES OF ERROR
The test device should not be frozen, it should be stored in the coolest, driest area
available, preferably at 4-30°C.
Reagents from different batch numbers should not be combined.

10. REFERENCES
Vista Diagnostics information leaflet. (Manufacturers of Vista-Pan).



                               SOP CARESTART


 Identification number :


1. FUNCTION OF THE TEST AND LIMITATIONS
This test kit detects Plasmodium lactate dehydrogenase in patient’s whole blood and is
useful as a screening procedure for malaria diagnosis.
The test is limited to the detection of pan-pLDH enzyme present in Plasmodium species,
and false results can occur. As with all diagnostic tests, the results must be correlated
with other clinical findings, and should be interpreted within an epidemiological, clinical
and therapeutic context. A parasitological technique (microscopic examination of a
stained blood smear) should be considered for comparison with the test result as a “Gold
standard”.

2. PRINCIPLE OF THE TEST
The CarestartTM malaria antigen test contains a membrane strip, which is pre-coated with
a monoclonal antibody as a line across a test strip. The test can detect pan-pLDH antigen
which is found in all Plasmodium species (falciparum, vivax, ovale, and malariae). The
monoclonal antibody in the test strip reacts with the malaria antigen released by the
parasitized red blood cell and forms an antibody–antigen complex which causes the test
strip to become visible.

3. SPECIMEN COLLECTION
Materials
    - cotton swabs
    - methylated spirit
    - blood lancets
    - sharps container

                                             67
   -   gloves
   -   sample pipette (inside CareStart test pouch)
   -   disposal bucket
   -   marker/pencil

Procedure
   1. Seat the patient in a comfortable position and inform him or her what you are
      about to do.
   2. Put on gloves, open the test pouch and remove the test device and the sample
      pipette.
   3. Label the test device with the patient initials, inclusion number and date.
   4. Open the lancet packet from the non-sharp end and leave in a secure place on the
      table.
   5. Using an alcohol swab, clean the site of the finger to be pricked.
   6. Squeeze the end of the fingertip, and make a prick with a sterile lancet and wipe
      off the first drop of blood with a clean, dry cotton swab. (Refer to SOP
      Fingerprick).
   7. Gently squeeze the top of sample pipette, immerse the open end in the blood drop
      and then gently release the pressure to draw blood into the sample pipette up to
      black line.
   8. Place a small piece of cotton wool at the site and ask the patient to press on the
      puncture wound until the bleeding stops.

4. EQUIPMENT
      - Carestart test device
      - digital timer

5. REAGENTS AND STAINS
      - clearing buffer (assay buffer)

6. METHOD OF CONTROL
The Carestart test device has an internal control. A control band should appear in the
control window to confirm that the test is valid.
As recommended by the manufacturer, the comparison with parastological examination
constitutes a second level of control.

7. TEST METHOD
    1. Add 5µl of whole blood in the sample well (small well) by squeezing the sample
       pipette. Take care not to loose blood on the sides of the well.
   2. Set the timer for 1 minute.
   3. After 1 minute, add 2 drops (60µl) of assay buffer into the buffer well (large
      well).
   4. Set the timer for 20 minutes

                                           68
    5. At 20 minutes, read the result.
    6. Immediately pass the test to the second reader for the second reading.




                                          Result Window

                                           Sample Well

                                           Buffer Well


                Figure 1      Diagrammatic view of Carestart test

8. INTERPRETATION OF TEST AND REPORTING OF RESULTS
Results can either be positive, negative or invalid, as shown in the diagrams below. In
case of an invalid result, a new test should be performed. For each option, the result may
also be considered ‘doubtful’ if the result is not immediately obvious to the reader. After
reading, the result should be recorded on the lab source data form (‘RDT results’) and in
the reader’s register.

Key
C = Control window
T = Test window

 NEGATIVE                     POSITIVE                                  INVALID




            C                             C                                C
C

            T                              T                               T
T




                                            69
NB INVALID TESTS MUST ALWAYS BE REPEATED.

9. SOURCES OF ERROR
    - Use of an expired test device or clearing buffer may give wrong results
    - Defect in the test device as a result of manufacturing error or due to improper
      storage
    - Using a reagent of a different lot to the lot of the test device.

10. REFERENCES
Access Bio.Inc./Mitra information sheet (Manufacturers of Carestart).
E-mail correspondence with manufacturers.




                                 SOP PARABANK



 Identification number:



1. FUNCTION OF THE TEST AND LIMITATIONS
The test detects the presence of Plasmodium species parasite lactate dehydrogenase
(pLDH) antigen, released from parasitised red blood cells, thus permitting detection of
malaria parasites.

Limitations:
As with all diagnostic tests, the test results must always be correlated with other clinical
findings, and should be interpreted within an epidemiological, clinical and therapeutic
context. A parasitological technique (microscopic examination of a stained blood smear)
should be considered for comparison as a ‘gold standard’.
The test device can only be used for testing whole blood.
In a few fresh samples and especially in stored samples, the background clearance may
be delayed for 15-20 minutes more after the original reading time of 15 minutes. In such
cases it is strongly recommended to extend the reading time by another 15 minutes so
that the results can be interpreted against a clear background.

2. PRINCIPLE OF THE TEST
Parabank is an immunochromatographic test. As the test sample flows through the
membrane assembly of the device after the addition of the clearing buffer, the coloured
monoclonal anti-pan specific-colloidal gold conjugate antibodies form a complex with
the specific malaria antigen in the lysed sample. This complex moves further on the
membrane to the test region where it is immobilised by the monoclonal anti-pan specific
antibody coated on the membrane, leading to the formation of a pink-purple coloured
                                            70
band, which confirms a positive test result. Absence of the coloured band in the test
region indicates a negative test result.
The un-reacted conjugate together with the rabbit globulin-colloidal gold conjugate (and
un-bound complex, if any), move further on the membrane and are subsequently
immobilised by the anti-rabbit antibodies coated on the membrane at the control region,
forming a pink purple band. This control band serves to validate the test performance.

3. SPECIMEN COLLECTION
   Materials
       - 5µl sample loop
       - test device
       - blood lancets
       - methylated spirit
       - cotton swabs
       - gloves
       - sharps container
       - disposal bucket for soiled materials
       - marker/pencil


Procedure
   1. Open the pouch and remove the test device plus sample loop. Check the colour of
      the desiccant, it should be blue. If it has turned colourless or pink, discard the
      device and use another device. Once opened, the device must be used
      immediately. Label test device with the patient’s initials, inclusion number and
      date.
   2. Seat the patient in a comfortable position and explain to the patient what you are
      about to do as you arrange the materials needed to perform a finger prick. Perform
      a finger prick according to the “SOP for making a finger prick”.
   3. Using the sample loop provided in the test kit, touch the sample loop to the blood
      on the finger to collect a loop full of blood.
   4. Immediately blot the blood on to the sample pad in the sample port ‘A’. Never
      leave the blood to clot before you blot the sample onto the sample pad.
   5. Give the patient a clean cotton swab to apply to the puncture site.

4. EQUIPMENT
     - Timer

5. REAGENTS/STAINS
     - Clearing buffer in a dropper bottle

6. METHOD OF CONTROL
The test device has an internal control consisting of a control line. Absence of the control
line indicates an invalid test. The test should be discarded and repeated afresh.



                                            71
7. TEST METHOD

   1. Make sure that the device and the buffer are of the same lot.
   2. Tighten the vial cup of the clearing buffer provided with a kit in a clockwise
      direction to pierce the dropper nozzle (if the buffer has not previously been used).
   3. Dispense four drops of the clearing buffer into buffer well marked ‘B’, whilst
      holding the plastic dropper vertically. Avoid spillage of the buffer into the sample
      port.
   4. Set the timer for 15 minutes, and start the timer.
   5. Read the results at the end of 15 minutes. However, if the background of the
      result window has not cleared within this time, re-set the timer for another 15
      minutes before reading the results.
   6. As soon as the results have been read, pass the test to a second reader.




8. REPORTING RESULTS/INTERPRETATION

The test may be classified as positive, negative or invalid (as shown below). If the reader
cannot give an immediate result for the test, it should be indicated that the test is
doubtful. Results should be recorded on the RDT result form and in the Reader’s register.

Key
C= Control window
T=Test window
A= Sample well
B= Buffer well

NEGATIVE for Plasmodium species


                                                                       Only one pink-purple
                                                                       band appears in the
                                                                       control window ‘C’

        C           T                       A       B
Figure 1 Negative Parabank result




                                            72
POSITIVE for Plasmodium species

                                                                    In addition to the
                                                                    control band, another
                                                                    pink-purple     band
                                                                    appears in the test
                                                                    window ‘T’.
        C          T                       A       B
Figure 2 Positive Parabank result


INVALID RESULT
                                                                    If the test band appears
                                                                    and the control band
                                                                    does not as in figure.3
                                                                    above, or if no band
                                                                    appears on the test
                                                                    device as in figure 4
        C          T                       A       B
                                                                    even after extension of
Figure 3 Invalid Parabank result (i)                                the reading time for
                                                                    another 15 minutes,
                                                                    repeat the test with a
                                                                    new device to ensure
                                                                    that the test procedures
                                                                    have been followed
                                                                    accurately.        ALL
                                                                    INVALID          TESTS
        C          T                       A       B                MUST BE REPEATED.
Figure 4 Invalid Parabank result (ii)

If the result of the test has to be thought about for more than a few seconds, the reader
should also record ‘doubtful’ = Yes on the RDT reader form.

9. SOURCES OF ERROR
      - Intermixing reagents from different lots.
      - Defect in the test device as a result of manufacturing error or due to improper
        storage.


10. REFERENCES
Zephyr Biomedicals manufacturer’s manual. (Manufacturer of Parabank device.)




                                           73

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:27
posted:8/4/2011
language:English
pages:73