502 SAMT DEEL 68 28 SEPTEMBER 1985
A simple clinical index of Crohn's
disease activity - the Cape Town index
J. P. WRIGHT, I. N. MARKS, A. PARFITT
The assessment of disease activity in Crohn's disease is liable Index correlated better with the HBI and CDAI but less well
to observer error because of the predominance of subjective with the van Hees index. The latter two indices did not
symptoms such as frequent diarrhoea, abdominal cramps and correlate well with each other. The addition of the erythrocyte
lassitude. While this difficulty in assessment occurs when sedimentation rate (ESR) to the Cape Town Index did not
comparing different consultations by the same physician, it improve its correlation with the CDAI (Table Ill).
becomes more serious when comparing disease activity in
different centres and particularly when performing therapeutic
drug studies. Discussion
In 1976 the Crohn's Disease Activity Index (CDAI) was
published.' This index required a 7-day documentation by the Although the correlation of the Cape Town Index and the
patient of his or her symptoms including the exact diarrhoea CDAI was not as high as we had expected, it was nevertheless
frequency. Four years later the no less complicated van Hees similar to that obtained by the HBI when compared to the
Index2 appeared. It usually required calculation by an electronic CDAI. The latter correlation was however somewhat less than
calculator but had the advantage of including more objective the 0,93 originally reported. J The reason for this is uncertain.
parameters such as serum albumin and haemoglobin values. It The patients entered into our study reflect a 'normal' clinic
was also less patient-dependent; for example, bowel function population assessed on 1 068 occasions over a IS-month period.
was simply assessed as being watery, semi-formed or formed. This is a larger number than that reported by Harvey and
At the same time the simplified CDAI, the Harvey-Bradshaw Bradshaw but should not account for the large difference.
Index (HBI), which quantified the patient's previous 24-hour As the addition of the ESR to the index did not materially
symptoms, was described. J It had the avantage of easy calcu- improve its correlation with the CDAI, it has not been included
lation but the disadvantage of being heavily biased by the in the proposed index.
frequency of diarrhoea. In the development of the CDAI and van Hees indices a
The need for a simple clinical index of disease activity selection of clinical and laboratory parameters were chosen on
without single-parameter bias was discussed at a meeting of the basis of clinical experience and subjected to analysis using
the International Organization for the study of Inflammatory the physician's global assessment of the patient as the standard.
Bowel Disease held in Oxford during Wovember 1980. A The resulting index therefore best reflects the physician's
number of clinical and laboratory parameters were considered opinion, using the parameters entered into the analysis. If the
but the overriding need for simplicity led to the formulation of purpose of the index is to quantify the attending physician's
an index based on 10 parameters (9 clinical and 1 haemato- opinion, then these complicated indices and their simplified
logical). It was suggested that the correlation between this variants are adequate. Our view, however, is that 'activity
index and the established, more complicated indicies be indices' should reflect more than this. In addition to the
examined in a prospective series of patients with Crohn's clinical symptoms they should also reflect our understanding
disease. The present study was therefore undertaken using a of 'active Crohn's disease'.
modification of the suggested index - the Cape Town Index A patient with Crohn's disease may have marked diarrhoea
(Table I). and even lassitude for many reasons including bacterial over-
growth and malabsorption as well as histologically active
Crohn's disease. In the former, diarrhoea and lassitude may be
Patients and methods prominent, while in the latter local and systemic features are
usually also present. The proposed index therefore ranks
Patients being seen by the authors as part of a prospective diarrhoea occurring for instance 6 times a day as equivalent to
therapeutic tria14 were asked to complete diary cards in order the presence of a local complication such as an anal fistula,
to calculate the CDAI. At each clinic visit they were assessed while the CDAI scores this amount of diarrhoea IS times
by the authors and the HBI, van Hees, CDAI and Cape Town higher than an anal fistula. The underlying problem is that it
indices were calculated. is not known which of the various parameters measured best
The correlation coefficient between each index and the P reflect the underlying disease process and the likely_clinical
value was calculated. course.
The advantages of the Cape Town Index include the broad
base of clinical features used, the heavier weighting given to
Results complications over diarrhoea, and the use of ranking rather
than absolute values which are liable to miscalculation by
Eighty-six patients were seen on 1068 occasions. Most patients patient and medical attendant alike.
had mild-to-moderate symptoms (Table ll). The Cape Town There is need for further study of the reliability and
reproducibility of the proposed Cape Town and other published
indices in quantifying symptoms and in assessing the underlying
Gastro-intestinal Clinic, Groote Schuur Hospital, and disease process and concurrent clinical course in patients with
Department of Medicine, University of Cape Town Crohn's disease.
J. P. WRIGHT, M.B. CH.B., M.R.C.P.
I. T. MARKS, M.B. CH. B., F.R.C.P.
A. PARFIIT, S.RS. We acknowledge the support of the South African Medical
Research Council and the Nellie Atkinson Trust.
SAMJ VOLUME 68 28 SEPTEMBER 1985 503
TABLE I. CAPE TOWN INDEX DEFINITION
0 2 3
Diarrhoea ';;;4 5 ;;'6
Abdominal pain Mild Moderate Severe
Well-being Below par Unwell Terrible
Local Skin tag Sinus Fistula
Systemic Stomatitis Arthralgia Arthritis
Temperature (0C) ';;;37,0 ';;;38,0 ';;;39,0 >39,0
Weight compared with
last weight <95% <90%
Mass Indefinite Certain
Tenderness Mild Moderate Severe
Haemoglobin (g/dl) ;;'12,0 <12,0 <11,0 <10,0
(ESR (mm/1st h) ';;;15 >15 >25 >40t)
-The index is calculated by adding the score for each period. thus obtaining a value between 0 and 30.
TAdded to test the effect of a commonly used indicator of inflammation but not included in proposed index.
TABLE 11. VALUES OF INDICES STUDIED (N= 1068)
Mean SO Minimum Maximum
Cape Town 2,8 2,8 o 17
Cape Town + ESR 3,65 3,2 o 19
HBI 2,2 2,9 o 23
COAl 85 68 14 433
Van Hees Index 168 32 77 329
TABLE Ill. CORRELATION MATRIX OF THE INDICES STUDIED (N= 1 068)*
Cape Town 0,374
Cape Town + ESR 0,481 0,934
COAl 0,418 0,761 0,728
HBI 0,262 0,810 0,733 0,745
Van Hees Cape Town Cape Town COAl
"'The probability value of all the correlation coefficients is < 0.001.
REFERENCES index of inflammatory activity in patients with Crohn's Disease. Gm 1980;
L Best WR, Bectel JM, Singleton JW, Kern F jun. Development of a Crohn's 3. Harvey RF, Bradshaw JM. A simple index of Crohn's disease activity.
Disease Activity Index - National Crohn's Disease Study. Gaslroenrerology La"cer 1980; i: 534.
1976; 70: 439-444. 4. Wright JP, Mee AS, Parfirr A er al. vitamin a therapy in patients with
2. Van Heese PAM, Van Elteren P T, Van Lier NJJ, Van Tongeren JHM. An Crohn's disease. Gasrroemerology 1985; 88: 512-514.