TRANSACTIONS OF THE ROYAL SOCIETY OF TROPICAL MEDICINE AND HYGIENE (1999) 93,133-136
An outbreak of cutaneous leishmaniasis in an Afghan refugee settlement in notth-
Mark Rowland”*, Arif Muniti, Naeem Durrani’ , Harry Noyes and Hugh Reyburn’ ’ HealthNet International,
University Town, I? 0. Box 889, Peshawar, Pakistan; 2London School ofHygie ne and TropicalMedicine, London WClE 7HT,
UK; 3Natimal Institute of Health, Islamabad, Pakistan; 4Liverpool School of Tropical Medicine, Liverpool L3 5QA, UK
Cutaneous leishmaniasis (CL) due to Leishmania tropica appears to be an emerging disease in parts of north-
east Afghanistan and north-west Pakistan. Timargara, an Afghan refugee camp of 17 years’ standing, in the
district of Dir, North West Frontier Province of Pakistan, experienced a major outbreak of CL in 1997 for
the first time. As part of the investigation, each section of the camp was surveyed for CL. Around 38% of the
9200 inhabitants bore active lesions and a further 13% had scars from earlier attacks. According to interview
statements, 99% of earlier infections had healed within the previous 2 years. To confirm the diagnosis, a
sample of current CL lesions was examined parasitologically. Amastigotes were detectable by microscopy in
only 36% of lesions. However, 48% of slide-negative cases produced positive cultures and some cases
negative to both microscopy and culture were positive by PCR. Overall detection rate was about 80%. The
sandtly Phlebotomus ser.qenti, a known vector of L. tropica, was captured within the camn. indicating local
transmission. CL has not been reported from this area-of Pakistan before. Although the iijority of rekgees
left Afghanistan 2 decades ago, cross-border movement ofmen is common. The Afhanistan caoital.-I Kabul.->
is cur&ntly experiencing a major epidemic of CL; infected migrant carriers fi-okKabul are probably the
source of the outbreak in Timargara.
Keywords: cutaneous leishmaniasis, Leishmaniu rropica, epidemiology, refugees, Afghanistan, Pakistan, Phlebotomus
Introduction position of lesions, and any history of past CL were
Cutaneous leishmaniasis (CL) due to Letihmania collected. The appearance of lesions was examined
(Leishmania) tropica appears to be an emerging disease wherever possible.
in parts of north-east Afghanistan and north-west Paki- Second. a diaenostic confirmation of clinical CL was
stan. Recent epidemiological data indicate that the undertakin using 3 techniques: microscopy, parasite
Afghanistan capital, Kabul, is experiencing a large in- culture and polymerase chain reaction (PCR). Families
crease in CL (A. Brvceson, WHO/I-IN1 unpublished of the original survey were selected using a random
assignment rep&t: Trkatmen; of cutaneous leishmaniasis in number generator and the first 100 current cases of CL
Kabul, 4-15 December 1998; HEWITT et al., 1998). were invited to attend the camp’s health centre for
Refugee camps in north-west Pakistan are reporting diagnostic validation.
cases of CL for the first time (United Nations Hi& At the time of the diagnostic study, a number of light-
Commissioner for Refugees, *HCR, unpublished traps were deployed in houses in the camp to identify
report, 1998). likely vectors of CL.
The Timargara refugee camp in the district of Dir, Owing to logistical problems, there was a 7-week gap
north-west Pakistan, is a settled camp of 17 years’ between the initial survey and the diagnostic study. None
standing. There are over 9000 residents, mostly from of the cases had received treatment as no treatment was
eastern, central, and northern Afghanistan. A few cases available at the time.
of CL have been noticed over the past few years. How- Twenty cases were absent when called upon and 5
ever, in 1997-98 there was a large outbreak, causing refused to attend. These were replaced by the next 25
alarm within the refugee community and among the cases among the randomly selected families.
health authorities. CL also now occurs in neighbouring Cases were seen, their lesions counted and measured,
Pakistani villages. and the duration of disease was noted. The nodular
As part of the investigation of the outbreak, the camp margin of the most active-looking lesion was sliced open
was surveyed for CL and a sample of lesions was and a smear taken using a scalpel blade. Smears were
examined parasitologically to confirm the diagnosis. fixed in methanol, stained with Giemsa, and examined
Confirmation of this outbreak is important in the plan- during the evening. Those with negative slides were
ning of public health strategies both in the camp and in invited to re-attend the following day for a repeat smear
the local area. It also indicates the need for vigilance for and culture sample, taken from another slit-skin pre-
emergent diseases associated with population move- paration. Twenty samples were also taken sequentially
ment. for PCR from the slide-negative cases (6 of which were
found to be slide-positive on the second examination).
Methods All slides were examined by 2 experienced microscopists.
There were 2 main comuonents to this studv. The microscopists each checked the findings of the other,
First, the camp was surveyed in May 1997 bia team of unaware of the first result.
resident communitv health workers (CHWsj trained in The samples for culture were inoculated into biphasic
the use of structures questionnaires ahd ide&ication of NNN media, transported on ice to the National Institute
CL lesions. Pairs of CHWs started in different sections of of Health in Islamabad, and incubated according to the
the camp and selected households at random, using a method of EVANS et al. (1989). The solid phase of the
spinning bottle as a direction pointer. Sixty households NNN media comnrised fresh. aseoticallv collected.
were visited and adults interviewed. Data on all residents, defibrinated rabbit blood, mixeh wi& agar; NaCl, glul
the number of current cases of CL, the number and case, and gentamicin. The liquid phase comprised RPM1
1640 medium with glutamine but without sodium
bicarbonate, filtered through 0.22-p disposable filters
Address for correspondence: Mark Rowland, Disease Control to avoid bacterial contamination.
and Vector Biology Unit, London School of Hygiene and For PCR, samples were placed in 500 ~.LLof 4-M
Tropical Medicine, Keppel Street, London WClE 7HT, UK; guanidine thiocyanate (GuSCN) and 0.25-M EDTA,
fax +44 (0) 171 580 9075, e-mail firstname.lastname@example.org and stored at 4°C until shipped by air to the Liverpool
134 MARK ROWLAND ETAL.
School of Tropical Medicine, UK. PCR was carried out was similar between boys (42%) and girls (43%) aged
as previously described by NOYES et al. (1998). < 15 years, but was significantly lower among adult males
(17%, n = 114) than among adult females (49%,
Results n = 140). Adult males seem to be less at risk (relative
Surveys risk: 0.35, x2 = 28, d.f. = 1, P <O.OOOl).
According to interviews, most families surveyed (57%) The prevalence of active lesions increased with age up
originated from the Afghan eastern province of Kunar, to 15 years (Figure); the correlation was significant
16% hailed from the northern provinces of Takhar and (r = O-77, P = 0.01). Older children seem to be more
Kundus, and others had come from other refugee camps at risk than any other age-group.
in the vicinity of Timargara. Sandfly light-traps were positioned indoors and out, in
Over one-third, 304 (38%), of the 799 refugees the upper, middle, and lower parts of the camp. Phlebo-
surveyed had active lesions, and 106 (17%) reported tomus (Paraphlebotomus) sergentiwere caught indoors and
scars and history characteristic of previous CL. All cases outdoors in each of the 6 traps (mean = 10 specimens
were recent: 87% of scars had healed within the past year per trap). No other species of phlebotomine was caught.
and 11% in the year before that. The mean number of
lesions per case was 2.2 (median 2, maximum 13). The Diagnostic study
mean duration of active lesions was 5.3 months, and no Microscopy showed that only 36 of the 100 cases
lesion was reported to be older than 12 months. sampled were positive for Leishmaniu amastigotes. Many
The population of Timargara was very stable: 82% of of the lesions appeared to be healing. However, of the 50
families had lived there for 3 15 years, and only 2% had culture specimens taken from the 64 slide-negative cases,
moved in during the past 6 years. The average age was 17 20 (48%) were positive for Leishmaniu and 8 were
years (median 10). Some two-thirds (61%) of the discarded owing to bacterial or fungal contamination.
population had lived their entire lives in Pakistan. Fourteen slide-negative patients refused to re-attend for
The Figure shows the prevalence of lesions and scars the second smear and culture (Table 1).
by age-group. Since an attack of CL gives life-long Nine (64%) of the 14 PCRs taken from slide-negative
immunity, this shows in effect the cumulative prevalence cases generated a PCR product of the same size as an L.
of CL with age. In endemic conditions cumulative tropica reference strain and distinct from L. infuntum and
prevalence tends to increase with age; in epidemic L. major strains. Restriction digests of the PCR product
conditions it is more constant across age-groups, as with Hue111 and HpaII produced fingerprint patterns
indicated here. Also shown in the Figure is the prevalence similar to L. tropica strains from Iran and central Pakistan
of active CL among the unscarred. Since a lesion lasts < 1 (Balochistan province). All 6 of the slides that were
year this approximates to the current annual incidence parasite-positive on the second examination were also
risk. In fact, the prevalence of active lesions in the positive to PCR. Five samples were negative by micro-
unscarred was significantly greater in the age-poups scopy, PCR and culture (Table 2).
younger than 15 years (relative risk: 1.4, x = 15, Ten specimens were sent for both PCR and culture; 3
d.f. = 1, P = O.OOOl), aswasthecumulativeprevalence. of the 5 culture-negative samples were positive to PCR
Further analysis reveals that prevalence of active lesions (Table 3).
0 10 20 30 40 60 60 70 so 90
+cumu!ative prevalence Oactii pfevalenw in unscarred
Figure. Prevalence of cutaneous leishmaniasis at different ages (years).
Table 1. Comparison of L. tropica cases identified by microscopy and by cell
Positive Negative Not done Total
Positive 3 2 31 36
Negative 20 22 22 64
Total 23 24 53 100
Table 2. Comparison of L. tropica cases identified by microscopy and
polymerase chain reaction (PCR)
Positive Negative Not done Total
Negative 8 z z: 2
Total 15 80 100
Table 3. Comparison of L. tro@ica cases identified tion (A. Bryceson, unpublished report, 1998; H. Rey-
by cell culture and polymerase chain reaction bum, 1998, unpublished surveys). As a result of
@‘CR) instability and civil war there is a regular flow of people,
some of them infected, between Kabul and refugee
PCR camps of Pakistan (I’. Marsden, unpublished report of
the European Commission: Living in exile: coping strute-
Positive Negative Total gihs of Afghan refugees, 1997). Adult male refugees are
mobile in the search for work (I’. Marsden, unpublished
Culture report, 1997), and those from Timargara would visit
Positive 3 Kabul or Parwan when travelling to the central or
Negative t 2 z northern Afghan provinces. Their frequent absence from
Total 5 5 10 the camp may explain why the prevalence of CL was
lower in this group and why they appear less at risk.
The diagnostic exercise described here would seem to
The mean duration of disease was significantly longer validate the findings of the initial survey. Although the
among cases negative to any diagnostic test (8 months, proportion of slide-positive cases was very low (36%), 20
SD = 2.9) than for cases positive to any test (6.5 months, of the 42 slide-negative cases sent for culture were
SD = 3.0) (Mann-Whitney U-test, .a = -2.5, P = positive (48%). By extrapolation, it would be expected
0.014). Similarly, the diameter of the largest lesion was that 48% of the 64 slide-negative cases would have been
greater for cases negative to any test (mean 3.1 cm, positive to culture, raising the rate ofpositive diagnosis to
SD = 2.3) than for cases positive to any test (2.4 cm, 76%. This rate might have been higher had there not
SD = 1.5) (a = 2.06, I’ = 0.039). been a 7-week delay between the initial survey and the
diagnostic study, during which perhaps 15% of cases
Discussion may have become negative to parasitological diagnosis.
The survey of Timargara camp indicates a large out- In addition 3 of the PCRs taken on 5 culture-negative
break of CL, with 38% of the population affected in 1997 cases were positive. If this is similarly extrapolated, the
(and 13% in the year or so before that). The recent case rate of positive diagnosis would have exceeded 80%.
histories, the population stability of the camp, and the Cases negative to any of the parasitological tests were
presence of the vector sandfly I? sergenri (KILLICK- likely to be of longer duration, as successful parasitolo-
KFNDRICK et al., 1994,1995) indicate that the outbreak gical diagnosis of CL is known to be inversely related to
emanated from within the camp. The clinical appearance the duration of the disease (RIPLEY & RIDLEY, 1983;
of the lesions was consistent with L. tropica infection, and AZADEH et al., 1985).
L. tropica kDNA was detected in the lesions by PCR The occurrence of this large outbreak of CL in the
(NOYES et al., 1998) indicating that the CL is due to L. Timargara refugee camp has important public health
tropica which is usually anthroponotic and urban. implications. CL may be carried by refugees into areas
Older children seem to be more at risk than any other previously unaffected by the disease (RAB et al., 1997).
age-group and men seem least at risk. Perhaps older Epidemics may flare up with little warning if vector
children are less disciplined in covering up during the hot sandflies are present, and l? sergenti, the vector of L.
summer nights. Women are required by their culture and tropica, is widespread in Pakistan (NASIR, 1964; LEWIS,
religion to retain their modesty at all times, and use of 1967). Cases due to L. tropica may be persistent and
chadors (the clothing for covering the head) is common disfiguring, and difficult and expensive to treat (A.
even at night. Bryceson, unpublished report, 1998). There is now a
CL has not been reported from this part of Pakistan trickle of cases reported from the local Pakistani popula-
before (RAB et al.. 1997). The outbreak mav have tion (UNHCR, unpublished report, 1998). Prompt and
resulted’from a single introduction, as the PCR evidence effective measures will be required to prevent CL from
is of a single, homogeneous schizodeme clone in the 15 becoming endemic in the area.
cases examined (NOYES et al., 1998). CL in Pakistan due
to L. tropica commonly occurs in Multan in the southern Acknowledgements
Punjab (BURNEY & LARI, 1986), in Balochistan province We are grateful to Musafa Khan and Fazle Rahim for the
where there was a major epidemic in the capital Quetta microscopy, Hameed ur Rahman for the sandfly detection, and
after the earthquake of 1935 (MASSOOM & MARRI, Dr Fayez of the Timargara basic health unit for alerting us to the
1993), and most recently in Azad Kashmir (RAB et al., outbreak. Rita Reybum assisted with the diagnostic survey. Dr
1997). Anthroponotic CL is more common in Afghani- Clive Davies and Professor Anthony Bryceson kindly comrnen-
stan; it is endemic to the southern city of Kandahar, the ted on the manuscript. HealthNet International’s Malaria and
Leishmaniasis Control Programme is supported by the Eur-
western city of Herat, and the central provinces of Kabul opean Commission (DGl and ECHO), the United Nations
and Parwan (OMR et al., ~~~~;NADIM & ROSTAMI, High Commissioner for Refugees, and Norwegian Church Aid.
1974; NADIM et al., 1979), but has not been reported
from the northern and eastern Afghan provinces from
which the majority of Timargara families originate. The References
Ashford, R. W., Kohestany, K. A. & Karimzad, M. A. (1992).
source of the outbreak is probably the capital, Kabul. Cutaneous leishmaniasis in Kabul: observations on a ‘pro-
The city is experiencing what has been described as a longed epidemic’. Annals of Tropical Medicine and Parasitol-
prolonged epidemic of CL (ASHFORD et al., 1992) ogy, 86,361-371.
currently affecting about 5% of the 1.2 million popula- Azadeh, B., Samad, A. & Ardehali, S. (1985). Histological
136 MARK ROWLAND ETAL.
spectrum of cutaneous leishmaniasis due to Leishmaniu S. (editors). India: Council of Scientific and Industrial
tropica. Transactions of the Royal Society of Tropical Medicine Research, pp. 231-236.
and Hygiene, 79,631-636. Nadim, A. & Rostami, G. S. (1974). Epidemiology of cutaneous
Bumey, M. I. & Lari? F. A. (1986). Status of cutaneous leishmaniasis in Kabul, Afghanistan. Bulletin of the World
leishmaniasis in Palustan. Pakistan Journal of Medical Re- Health Organization, 51, 45-49.
search, 25, 101-108. Nadim, A., Javadian, E., Noushin, M. K. & Nayh, A. K. (1979).
Evans, D. A., Godfrey, D. G. & Lanham, S. (1989). Letihmania. Epidemiology of cutaneous leishmaniasis in Afghanistan. II.
In: Handbook on Isolation, Characrerization and Cryopreserva- Anthroponotic cutaneous leishmaniasis. Bulletin de la Soci&
rion of Leishmania, Evans, D. A. (editor). Switzerland: de Pathologic Exotique, 5,46 l-466.
UNDP/World Bank/WHO (TDR), pp. l-28. Nasir, A. S. (1964). Sandflies as a vector of human disease in
Hewitt, S., Reybum, H., Ashford, R. & Rowland, M. (1998). Pakistan. Pakistan Journal of Health, 14, 26-30.
Anthroponotic cutaneous leishmaniasis in Kabul, Afghani- Noyes, H. A., Reybum, H., Bailey, J. W. & Smith, D. (1998). A
stan: vertical distribution of cases in apartment blocks. nested PCR based schizodeme method for identifying Leish-
Transactions of the Royal Society of Tropical Medicine and mania kinetoplast minicircle classes directly from clinical
Hygiene, 92,273-274. samples and its application to the study of the epidemiology
Killick-Kendrick, R., Killick-Kendrick, M. & Tang, Y. (1994). of Leishmania tropica in Pakistan. Journal of Clinical Micro-
Anthroponotic cutaneous leishmaniasis in Kabul, Afghani- biology, 36, 2877-2881.
stan: the low susceptibility of Phlebotomus papatasi to Leish- Omar, A., Saboor, A,, Saidi, S. & Sery, V. (1968). Cutaneous
mania tropica. Transactions of the Royal Society of Tropical leishmaniasis in Afghanistan. Afghanistan Journal of Public
Medicine and Hygiene, S&252-253. Health, 1, l-6.
Killick-Kendrick, R., Killick-Kendrick, M. & Tang, Y. (1995). Rab, M. A., Rustamani, L. A., Bhutta, R. A., Mahmood, M. T.
Anthroponotic cutaneous leishmamasis in Kabul, Afghani- &Evans, D. A. (1997). Cutaneous leishmaniasis: iso-enzyme
stan: the high susceptibility of l’hlebotomzts sergenti to Leish- characterisation of Leishmania tropica. Journal of the Pakistan
mania tropica. Transactions of the Royal Society of Tropical MedicalAssociation, 47,270-273.
Medicine and Hygiene, 89,477. Ridley, D. S. & Ridley, M. J. (1983). The evolution of the lesion
Lewis, D. J. (1967). The phlebotomine sandflies of West in cutaneous leishmaniasis. Journal of Pathology, 141, 83-96.
Pakistan (Diptera: Psychodidae). Bulletin of the British Mu-
seum (Natural Hisroy) Entomology, 19, l-75.
Massoom, M. & Mani, S. M. (1993). Current status of
leishmaniasis in Pakistan. In: Current Trends in Leishmania Received 16 October 1998; revised 18 December 1998;
Research, Bhaduri, A. N., Basu, M. K., Sen, A. K. & Kumar, accepted for publication I8 December 1998
ROYAL SOCIETY OF TROPICAL MEDICINE AND HYGIENE
Denis Burkitt Fellowships
The Denis Burkitt Fund was set up by his family in memory of Denis Burkitt, FRS, who died in 1993; it is
administered by the Royal Society of Tropical Medicine and Hygiene.
One Fellowship (maximum value E7000) or two separate Fellowships (of E3500 each) are awarded annually for
practical training, travel, or direct assistance with a specific project (preferably clinico-pathological, geographical
or epidemiological studies of non-communicable diseases in Africa).
Applications must be made at least six months before the commencement of the proposed study (by 15 March or
15 September in each year). A short report on the study should be submitted, within 3 months of the recipient’s
return. Application forms are available from the Administrator, Royal Society ofTropical Medicine and Hygiene,
Manson House, 26 Portland Place, London, W 1N 4EY, UK; fax +44 (0) 17 1436 1389, e-mail email@example.com
ROYAL SOCIETY OF TROPICAL MEDICINE AND HYGIENE
Robert Cochrane Fund for Leprosy
The fund, in memory of the great leprologist Robert Cochrane, is administered by the Royal Society of Tropical
Medicine and Hygiene. It is used to finance up to three travel fellowships each year to a maximum value of El000
The fund will support travel for
l Leprosy workers who need to obtain practical training in field work or in research
l Experienced leprologists to provide practical clincal training in a developing country
There is no restriction on the country of origin or destination providing the above requirements are met.
Applications must be made at least six months ahead of the proposed trip, sponsored by a suitable representative
of the applicant’s employer or study centre and agreed by the host organization. A short report on the travel/study
should be submitted, within one month of the recipient’s return. Application forms are available from the
Administrator, Royal Society of Tropical Medicine and Hygiene, Manson House, 26 Portland Place, London,
WIN 4EY, UK; fax +44 (0)171 436 1389, e-mail firstname.lastname@example.org