Latent Tuberculosis is Highly Prevalent in Sub-Saharan Africans in Dublin - a Study Intended to Establish Normal CD4 Reference Ranges in this Population

Objectives: To establish the normal reference range for CD4 lymphocyte cells in human immunodeficiency virus (HIV) negative sub-Saharan Africans attending the Genito-Urinary and Infectious Diseases (GUIDE) Outpatient Clinic at St. James’s Hospital in Dublin. To correlate CD4 Count with lymphocyte count. Design: This was a prospective observational study. Methods: Volunteers were recruited among new sub-Saharan African patients attending the GUIDE Outpatient Clinic at St. James’s Hospital, Dublin. Recruitment took place over an eight-week period between July and August 2003. The study objectives and methods were explained to volunteers, and informed consent for participation was obtained. History and relevant physical examination, together with measurement of haematological parameters, screening for sexually transmitted infections (STIs) and examination of stool and urine samples were performed to exclude confounding co-morbidities. A chest radiograph and Mantoux skin test was performed to exclude pulmonary tuberculosis (TB). Results: Seventeen participants were recruited. Two (12%) were excluded on the basis of HIV infection. Ten men and five women form the CD4 study group. Of the fifteen suitable patients recruited, the range of CD4 count is 532–1537 x 10/L. The reference range used by the laboratory at St. James’s, based on CD4 counts in largely Caucasian populations, is 380 – 1500 x 10/L. Despite a high level of coincidental findings in the cohort, the range of CD4 counts measured falls within the range currently used by the laboratory. Women had significantly higher CD4 cell counts than men (median = 1245 and 899 respectively. P < 0.01), as has been previously described. There was a strong linear relationship between CD4 cell count and absolute lymphocyte count (r=0.5). Twelve of thirteen patients (92%) screened had evidence of latent pulmonary tuberculosis (TB), on the basis of a positive Mantoux reaction without radiographic evidence of TB. Two of fifteen (13%) HIV-negative patients defaulted before the result of their tuberculin skin test could be read. Only one patient in the entire cohort had a negative reaction to tuberculin challenge. We have referred 92% of this cohort for tuberculosis chemotherapy. Conclusions: The range of CD4 lymphocyte counts measured in this cohort falls within that used by the central pathology laboratory at St. James’s Hospital, and by clinicians at GUIDE. This is the range that is used to guide clinical care among HIV-positive patients at GUIDE. A high rate of latent TB exists in this cohort. INTRODUCTION Tuberculosis (TB) The World Health Organization estimates that one third of the world’s population is infected with Mycobacterium tuberculosis, and that there are eight million new cases of active TB annually. Nearly 2 million persons die of TB worldwide each year, including persons infected with the human immunodeficiency virus (HIV). The global incidence rate of TB is growing by approximately 0.4% per year, with a much faster growth rate in sub-Saharan Africa. TB is currently responsible for approximately 11% of deaths occurring due to the acquired immunodeficiency syndrome (AIDS) worldwide. HIV is the single most important precipitant of the increased incidence of TB in Africa in the past 10 years. TB is now the most common AIDSdefining illness in Africans resident in the UK. The crude incidence rate of tuberculosis in Ireland fell in four consecutive years until 2000, where it was recorded at 10.9 cases per 100,000 population. In the year 2000, 11.4% of cases of tuberculosis notified nationally were known to affect persons born outside Ireland. Background and Original Aims of the Study International guidelines for the treatment of HIV infection have been drawn up using CD4 lymphocyte cell count reference ranges determined by studies conducted on HIV-negative North American and European subjects. Several studies have shown that there is a significant difference in reference ranges between ethnic groups. The primary aim of this study was to establish the normal reference range for CD4 cells in HIV-negative sub-Saharan Africans attending the Genito-Urinary and Infectious Diseases (GUIDE) Outpatient Clinic at St. James’s Hospital, Dublin. This was a prospective observational study. Latent Tuberculosis in Sub-Saharan Africans in Dublin 73 STUDY METHODS General Considerations Measurement of CD4 count at St. James’s Hospital is undertaken as part of measurement of lymphocyte subsets. This is carried out by the department of immunology at the hospital’s central pathology laboratory. Candidates for the study were drawn from new sub-Saharan African patients who registered with the GUIDE outpatient service over an eight week period between July and August 2003. Consent, History-taking and Examination Prior to participation, a detailed consent form was discussed with the patient, and informed consent was obtained for participation in the study. Demographic details and past medical and infection history were recorded, as were details pertaining to smoking preference. All patients underwent infectious disease screening in order to exclude ongoing infection. Participants had a general physical examination. Chest radiograph and tuberculin skin test (Mantoux method) were performed to identify tuberculosis. Stool samples were collected for culture and examination for ova, cysts and parasites. Optimal monoclonal antibody kits were used to screen for malaria. Urinalysis was performed to exclude pathological proteinuria, haematuria, glycosuria, and bacteruria (evidenced by the presence of nitrite in the urine), and of leucocyte esterase (indicative of the presence of white cells in the sample). Female patients underwent a pregnancy test based on urinary qualitative human chorionic gonadotrophin (βHCG) measurement. A screen for sexually transmitted infection was performed. This included genital swabs for chlamydia, gonorrhoea, trichomoniasis and non-specific urethritis, and blood tests for HIV, hepatitis A, B and C, and syphilis. Absolute lymphocyte count was measured in each volunteer, together with lymphocyte subsets (including CD4 count). Statistical Analysis Statistical analysis of CD4 lymphocyte counts was undertaken using the Statistical Program for the Social Sciences (SPSS). The range of a group of CD4 cell counts is positively skewed. "Skew" or "bias" is used to describe a distribution that is not normal. This reflects the fact that although the upper limit of a range of CD4 cell counts may be very high, the lower limit cannot be less than zero. Because of this positive skew, standard statistical tests based on normal distributions such as the Student’s T-test are weakened. Nonparametric analysis does not require that variables under examination must be distributed normally about the mean of the sample. This is the reason that non-parametric analysis is more appropriate to a sample consisting of CD4 cell counts than a simple parametric test such as the t-test. The Kruskal-Wallis rank sum test used in this instance is a non-parametric statistical test. The coefficient of determination r was calculated as a test of the strength of the correlation between CD4 Count and absolute lymphocyte count. For two given variables x and y measured in a sample, r represents the proportion of the variability of y that can be attributed to its linear relationship with x, where r is the Pearson product moment correlation coefficient "correlation coefficient". In this study, r is quoted as a measure of the strength of the relationship between CD4 count and absolute lymphocyte count. RESULTS Seventeen new patients of sub-Saharan origin were recruited over an eight week period between July and August, 2003. Eleven men and six women were enrolled during this time. The original intention of the study was to establish a range for CD4 lymphocytes in sub-Saharan African patients. Recruitment was hampered by a sharp decline in the number of new patients from subSaharan Africa attending GUIDE outpatient clinics during the study period. Most new patients from sub-Saharan Africa at GUIDE are asylum seekers who are in the process of applying for refugee status in this country. This curtailment of numbers may reflect a national trend in the rate at which people sought asylum in Ireland during 2003 (figure 1). Figure 1: Applications for asylum received by the Refugee Applications Commissioner’s Office between December 2000 and October 2003. Seventeen volunteers from eight countries in sub-Saharan Africa were recruited. The majority of our volunteers were born in TSMJ Volume 5: Original Research 74 Nigeria (N=9). Two were born in Angola, and one in each of the other countries listed in figure 2. Two of our seventeen original volunteers (12%) were diagnosed HIV-positive, a Nigerian woman and a man from the Ivory Coast. These patients were excluded from the CD4 study calculations. The range of CD4 lymphocyte counts in the fifteen suitable patients recruited is 532 – 1537 x 10/L. The reference range used by our laboratory, based on CD4 counts in largely Caucasian populations, is 380 – 1500 x 10/L. The range of CD4 lymphocyte cell counts measured in our cohort falls within the range Figure 2: Country of origin of study recruits (N=17). Figure 3: Sex variation in CD4 count. currently used by our laboratory. Women had significantly higher CD4 cell counts than men (median = 1245 and 899 respectively. P < 0.01), as has been previously described (figure 3). This is based on a Kruskal-Wallis rank sum test. Smoking is known to affect CD4 count. Smokers are known to have significantly higher CD4 counts than non smokers. Three (20%) of the patients in this cohort were smokers. No difference was found between the CD4 counts of smokers and non-smokers in this small group. There was a strong linear relationship between CD4 cell count and absolute lymphocyte count (r=0.5) (See figure 4). This value indicates that at least 50% of the variation in CD4 is explained by the variation in lymphocyte count. In order to establish that the range of CD4 lymphocyte counts in our cohort was valid as a reference range in the wider population, it was important for us to exclude infective causes of altered lymphocyte count in the study cohort. Table 1 details potentially confounding pathologies identified in members of the study group. Despite the wide range of incidental pathologies identified in the study group, measurement of CD4 count lies within the range currently used by the hospital. Figure 4: CD4 and lymphocyte counts in men (●) and