Before the introduction of highly active antiretroviral therapy (HAART) in industrialized countries, malnutrition was frequently observed in HIV-infected patients , and few nutritional interventions were shown to prevent or correct it . Nutritional benefits of HAART have now been clearly observed . Although there are some concerns regarding metabolic dysfunction with protease inhibitor drugs , most treated HIV-positive patients from industrialized countries recover or maintain a more acceptable nutritional status, with indirect benefits on their quality of life.
In sub-Saharan Africa, where most HIV-positive persons live , HAART is unavailable for all but very few. Even when malnutrition is recognized as a major complication of HIV disease , the interventions that improve nutritional status in HIV-positive patients in developed countries cannot be implemented because of limited resources and technical constraints, with the exception of diet counselling, which is yet to be fully appreciated in the developing countries . Consequently, appropriate nutritional interventions have not been evaluated in developing countries and clinicians have little interest in measuring nutritional indicators. However, it is possible that any prophylactic or curative intervention other than HAART that could be available for African HIV-positive patients might have indirect effects on nutritional status. Easy-to-measure anthropometric indicators for use in routine consultations could help clinicians to estimate the nutritional consequences of their package of care and to assess any improvement in the quality of life of their patients.
Recently, two placebo-controlled trials in Abidjan, Côte d'Ivoire showed that chemoprophylaxis with co-trimoxazole (CMX) dramatically reduced the early severe morbidity in HIV-infected adults [8,9].
The present study examines whether this reduction in early morbidity was accompanied by improved nutritional status, using available anthropometric data from patients included in the COTRIMO-CI ANRS 059 trial .
This study was part of a double-blind placebo-controlled trial conducted in Abidjan (Côte d'Ivoire) from April 1996 to March 1998 to assess the efficacy of CMX prophylaxis in reducing severe clinical events in HIV-infected adults recruited at early stages of HIV-1 infection. Study population and medication have been previously described . In brief, patients whose HIV infection was diagnosed in routine consultations at five urban community health centres in Abidjan were included if they met the following criteria: 18 years of age or more, HIV-1 or HIV-1 plus HIV-2 dual infection, clinical stage 2 or 3 of the WHO staging system for HIV infection and disease  and written consent. Reasons for exclusion were pregnancy, breast-feeding, prior history of intolerance to any sulphonamide, haemoglobin level < 7 mg/dl, platelet count < 75 × 109 cells/l, absolute neutrophil count < 0.75 × 109 cells/l and renal or hepatic failure. Subjects were randomly assigned to receive daily either CMX (800 mg sulfamethoxazole and 160 mg trimethoprim) in a single tablet or a matching placebo. No patient received antiretroviral therapy during the trial.
Baseline demographic, clinical and biological data were recorded through standardized questionnaires and in physical examinations and CD4 cell count estimations. Anthropometric measurements included weight (measured to the nearest 0.2 kg on an electronic scale for patients who were dressed but without shoes), height (measured to the nearest 1 cm on a wall scale, patients without shoes and with joined feet), mid-upper arm circumference (measured to the nearest 5 mm, on the left half arm) and skinfold thickness in four sites (bicipital, tricipital, subscapular and suprailliac) measured to the nearest 0.2 mm with a caliper (Siber Hegner, F-01702 Miribel, France).
After randomization, patients were asked to attend monthly scheduled visits and, in addition, to visit the medical team at the trial centre as soon as they had medical problems. All medications, consultations, hospitalizations and transportation were free. All clinical events (including death) were referred for independent review by an event documentation committee blinded to the treatment group. The primary outcome of the trial was the occurrence of severe clinical events, defined as all events leading to hospitalization and/or death whatever the cause. Deaths were systematically investigated for patients who did not keep scheduled appointments through home visits and telephone calls, and by screening obituary sections of the national newspapers. Every 3 months, the anthropometric indicators for baseline evaluation were measured. The body mass index (BMI) was the ratio of weight (kg) to height (m2); the arm muscle circumference (AMC) was the mid-upper arm circumference − π(tricipital skinfold in mm); and the percentage of fat mass (FM) was estimated using the Durnin and Womersley equation and the four skinfold thicknesses, age and gender .
Patients with complete anthropometric data at baseline and at least one quarterly visit were included in the present analysis. To analyse indicators in both men and women, the variation was calculated as the difference between the quarterly value and the baseline value, divided by the baseline value, and expressed as a percentage. Patients included in the present analysis compared with other patients in the trial, and changes between treatment groups, were analysed using Student's t-test or the t-test for unequal variances if the significance of Bartlett's test was < 0.05. Comparisons for qualitative variables were performed by the chi square test. Differences between the quarterly and the baseline visits in each treatment group were analysed using Student's t-test for matched data. Analyses of anthropometric changes and comparisons between treatment groups were first performed using measurements from inclusion to the end of the trial (17 March 1998) in both treatment groups. After the trial was stopped, all patients received CMX and were followed according to the same scheduled procedures. Therefore, the nutritional measures for up to 1 year after the end of the trial for patients in the CMX group were subsequently included in the analyses to increase the statistical power of the analysis
Among the 545 patients randomized in the trial from 23 April 1996 to 17 March 1998, 128 were excluded from the present study: four because they were subsequently confirmed to be HIV-2 positive only; 16 because they had incomplete data at inclusion; and 118 because they did not complete 3 months of follow up or, if they did, they did not have complete anthropometric data. Of the 417 patients analysed, 58% were women, 48% had a regular professional activity (including activity in the informal sector) and 56% were illiterate or had primary school level education. The median age was 31 years [interquartile range (IQR), 26–37]. At entry in the trial, 36% of patients were classified at WHO clinical stage 2, 58% at stage 3, and 6% at stage 4. The median CD4 lymphocyte count was 282 × 106 cells/l (IQR, 148.5–481), with 36% of patients with ≤ 200 × 106 cells/l, 41% with 200–500 × 106 cells/l and 23% with > 500 × 106 cells/l. Median BMI was 20.1 (IQR, 18.6–21.8), median AMC 223 mm (IQR, 208–241) and median FM 16.8% (IQR, 10.5–21.9). None of these indicators was different between treatment groups at inclusion (data not shown). Baseline characteristics of these 417 patients did not differ from those of the 128 patients excluded from this analysis, except for median age (31 versus 33 years; P = 0.03) and median CD4 lymphocyte count (282 × 106 versus 231 × 106 cells/l; P = 0.0001). Table 1 shows the measures available at each scheduled visit during the trial follow up (CMX and placebo groups) and after the trial was stopped (CMX group only). The number of attended visits was comparable in both treatment groups (Table 1).
Figure 1 shows the evolution of mean BMI (Fig. 1a), mean AMC (Fig. 1b) and mean FM (Fig. 1c), according to the treatment group up to the end of the trial. In the CMX group, a rapid (first 3 months) and significant improvement of nutritional status was observed for all anthropometric indicators (BMI P < 10−−4; AMC P = 0.0006; FM P < 10−−4). This improvement was maintained up to the ninth month of follow up (BMI P < 10−−4; AMC P = 0.006; FM P = 0.008), but was not significant after this point (BMI P = 0.20; AMC P = 0.99; FM P = 0.60 for the relative variation at 12 months). A slight improvement was observed in the placebo group at 3 months for BMI (Fig. 1a; P = 0.07) and FM (Fig. 1c; P = 0.06), but not for AMC (Fig. 1b; P = 0.56). The three anthropometric indicators did not vary significantly beyond the third month, except that there was a significant decrease of AMC at 9 months (P = 0.03). The difference between treatment groups was statistically significant at 3, 6 and 9 months (P < 0.01) for all anthropometric indicators but not at any time after this point.
Figure 2 compares baseline data with values at the end of the trial for patients in the placebo group and with values 1 year after the end of the trial for the CMX group. The improvement of the three anthropometric indicators remained measurable at 3 months of follow up for the CMX group (BMI P < 10−−4; AMC P = 0.0009; FM P < 10−−4); it persisted to 24 months of follow up for BMI (Fig. 2a; P = 0.007), to 21 months of follow up for AMC (Fig. 2b; P = 0.02 at 21 months and P = 0.16 at 24 months), and to 12 months of follow up for FM (Fig. 2c; P = 0.04 at 12 months and P = 0.99 at 15 months). The difference between the CMX and placebo groups remained significant up to 15 months for BMI (P = 0.04) and up to 12 months for AMC and FM (P = 0.01 and 0.007, respectively).
The population studied was part of a group of HIV-1-infected adults included in the ANRS 059 trial in which early CMX prophylaxis was shown to be highly efficacious in preventing severe clinical events in Côte d'Ivoire . Available anthropometric measurements were analysed from patients enrolled in the trial and followed up at least 3 months. Overall, there was a significant and persistent improvement of nutritional indicators in patients receiving CMX; patients from the placebo group experienced an initial slight, transient and non-significant improvement before worsening anthropometric indicators, resulting in significantly lower values to those in the CMX group. This nutritional benefit under CMX prophylaxis was detectable at follow up for up to 9 months if available data between inclusion and the end of the trial were considering, and for up to approximately 1 year for FM and 2 years for BMI and AMC if data from inclusion to 1 year after the end of the trial was considered.
At enrolment in the trial, any clinical events were diagnosed and treated, which may explain why anthropometric indicators improved initially in both groups. However, the difference seen in the CMX treatment group during the overall follow up supports the hypothesis that prevention of severe clinical events with CMX may permit an improvement of nutritional status and then prevent malnutrition.
The relationship between nutritional status and severe clinical events has previously been described at advanced stages of HIV disease in patients from industrialized countries . In Africa, most of the nutritional studies performed in HIV-positive adults focused on the relationship between malnutrition and tuberculosis  or on other factors involved in malnutrition, such as dietary intakes . None have followed resting energy expenditure as carried out in developed countries . To our knowledge, this is the first time that a relationship has been observed between the prevention of severe clinical events at early stages of HIV disease, including non-tuberculous causes of morbidity, and improving nutritional status.
CMX may have improved nutritional status by indirect mechanisms rather than by prevention of morbidity-related malnutrition. The observed increase of FM at the beginning of follow up could be related to a noticeable increase in dietary intake, in parallel with an increase in fat-free mass, which is partly estimated by AMC measurement. Dietary intake values are not available in this report. AMC was maintained at higher than baseline level during follow up; given the pronostic value of fat-free mass on survival , this benefit could be very important. However, AMC is not an exact measure of fat-free mass, and a prognostic value of AMC on survival was not observed in our population (unpublished data). Analysis of data on bioelectrical impedance measurements, which were collected during this trial but not described here, will help to clarify this issue . To our knowledge, no CMX-related metabolic side-effect, such as the peripheral lipoatrophy and truncal adiposity observed with HAART , that could explain the anthropometric changes observed in our study has ever been reported.
Data for the placebo group provide information on the natural history of nutritional status in African HIV-positive patients. In a group of patients with a mean follow up of 10 months, there was no significant decrease in anthropometric values in the placebo group, although a tendency was noticed for AMC and FM to decrease. It is possible that the interval between two measurements (3 months) was too long to detect a rapid fall in the value of these anthropometric indicators just before death, which occurred in 16% of the included patients. Another limitation is that the sample analysed was slightly selected: patients with available data were younger and had a better immunological status than patients included in the trial but not included in the present analysis. However, the main cause of non-inclusion in this study was a follow-up time shorter than 3 months, and non-analysed patients were equally distributed between treatment groups.
Variations in BMI and AMC were slight (about 3%) compared with the higher FM variation (10%), and their significance as a measure of clinical evolution and survival remains to be explored. The wide variation in FM could be partly explained by the fact that both sexes were analysed together and yet FM is known to differ in men and women . Such variation was observed in the men and women participating (data not shown). This variability could also be due to the low accuracy of FM measurements, and to the natural interindividual variability of this indicator . BMI offers an interesting alternative to estimate FM , but in our study its evolution was more comparable to AMC than to FM. This could be explained by the low sensitivity of BMI in predicting FM in patients with a low percentage of body fat , as observed in our population.
In the ANRS 059 trial where all patients had free access to medical care, all severe events were diagnosed early and managed, and CMX prophylaxis prevented severe morbidity but not mortality . Some authors have suggested that, given the lack of effect on survival, prompt outpatient treatment of illness episodes might be preferable to CMX prophylaxis  . The results described here indicate that the efficacy of CMX in reducing severe morbidity and the use of health care services was associated with a dramatic improvement in nutritional status in the medium term, thus supporting the hypothesis that a prophylactic strategy might be preferable to a curative one, at least in term of quality of life. This was part of the discussion in a recent UNAIDS/WHO consultation on the prophylactic use of CMX among HIV-positive patients in Africa, which finally recommended that CMX prophylaxis should be used in Africa as part of a minimum package of care for all symptomatic patients, regardless of their clinical stage . Use of anthropometric indicators when implementing such recommendations could validate our findings in various settings.
Since CMX prophylaxis should now be widely used in Africa, it will be important to monitor its clinical effectiveness over time. It is difficult to identify outcomes of effectiveness that can be easily monitored in field conditions where laboratory facilities are not available. Anthropometric indicators could provide such a set of indicators. Short-term improvement of these indicators in adults who start CMX prophylaxis should be considered as an indication of the clinical effectiveness of this intervention.
This study was part of the COTRIMO-CI/ANRS 059 trial. The authors wish to thank all the members of the COTRIMO-CI Study Group in Abidjan and Bordeaux, and all the patients who participated in the trial.
1. Grunfeld C. The wasting syndrome in AIDS: mechanisms and implications. Nutrition 1998, 14: 867 –868.
2. Macallan DC. Wasting in HIV infection and AIDS. J Nutr 1999, 129: 238S –242S.
3. Pernerstorfer-Schoen H, Schindler K, Parschalk B. et al
. Beneficial effects of protease inhibitors on body composition and energy expenditure: a comparison between HIV-infected and AIDS patients. AIDS 1999, 13: 2389 –2396.
4. Carr A, Samaras K, Thorisdottir A, Kaufmann GR, Chisholm DJ, Cooper DA. Diagnosis, prediction, and natural course of HIV-1 protease-inhibitor-associated lipodystrophy, hyperlipidaemia, and diabetes mellitus: a cohort study. Lancet 1999, 353: 2093 –2099.
5. World Health Organization. Global AIDS surveillance – Part II. Wkly Epidemiol Rec 1999, 74: 409 –414.
6. World Health Organization. WHO case definitions for AIDS surveillance in adults and adolescents. Wkly Epidemiol Rec
7. Schwenk A, Bürger B, Ollenschläger G. et al
. Evaluation of nutritional counselling in HIV-associated malnutrition. Clin Nutr 1994, 13: 212 –220.
8. Anglaret X, Chêne G, Attia A. et al
. Early chemoprophylaxis with trimethoprim-sulphamethoxazole for HIV-1-infected adults in Abidjan, Côte d'Ivoire: a randomised trial. Lancet 1999, 353: 1463 –1468.
9. Wiktor SZ, Sassan-Morokro M, Grant AD. et al
. Efficacy of trimethoprim-sulphamethoxazole prophylaxis to decrease morbidity and mortality in HIV-1-infected patients with tuberculosis in Abidjan, Côte d'Ivoire: a randomised controlled trial. Lancet 1999, 353: 1469 –1475.
10. World Health Organization. Acquired immunodeficiency syndrome (AIDS). Interim proposal for a WHO staging system for HIV infection and disease. Wkly Epidemiol Rec
11. Durnin JVGA, Womersley J. Body fat
assessment from total body density and its estimation from skinfold thickness: Measurements on 481 men and women aged from 16 to 72 years. Br J Nutr 1974, 32: 77 –97.
12. Nahlen BL, Chu SY, Nwanyanwu OC, Berkelman RL, Martinez SA, Rullan JV. HIV wasting syndrome in the United States. AIDS 1993, 7: 183 –188.
13. Niyongabo T, Henzel D, Idi M. et al
. Tuberculosis, human immunodeficiency virus infection, and malnutrition in Burundi. Nutrition 1999, 15: 289 –293.
14. Castetbon K, Anglaret X, Attia A. et al
. Dietary problems in African HIV-infected adults, Abidjan, Côte d'Ivoire. J Acquir Immune Defic Syndr 2000, 23: 357 –358.
15. Melchior JC, Raguin G, Boulier A. et al
. Resting energy expenditure in human immunodeficiency virus-infected patients: comparison between patients with and without secondary infections. Am J Clin Nutr 1993, 57: 614 –619.
16. Kotler DP, Tierney AR, Wang J, Pierson RN. Magnitude of body-cell-mass depletion and the timing of death from wasting in AIDS. Am J Clin Nutr 1989, 50: 444 –447.
17. Ott M, Lembcke B, Fischer H. et al
. Early changes of body composition in human immunodeficiency virus-infected patients: tetrapolar body impedance analysis indicates significant malnutrition. Am J Clin Nutr 1993, 57: 15 –19.
18. Saint-Marc T, Partisani M, Poizot-Martin I. et al
. A syndrome of peripheral fat
wasting (lipodystrophy) in patients receiving long-term nucleoside analogue therapy. AIDS 1999, 13: 1659 –1667.
19. Report of a WHO Expert Committee. Physical Status: the Use and Interpretation of Anthropometry.
Geneva: World Health Organization, 1995.
20. Baxter JP. Problems of nutritional assessment in the acute setting. Proc Nutr Soc 1999, 58: 39 –46.
21. Norgan NG. Population differences in body composition in relation to the body mass index
. Eur J Clin Nutr 1994, 48: S10 –S27.
22. Gartner A, Maire B, Traissac P. et al
. Sensitivity and specificity of the body mass index
to assess low percent body fat
in African women. Am J Hum Biol 2000, 12: 25 –31.
23. Brindle R. Co-trimoxazole in HIV-1 infection. Lancet 1999, 354: 333 –334.
24. UNAIDS, WHO. UNAIDS/WHO Hail Consensus on use of Cotrimoxazole for Prevention of HIV-related Infections in Africa.
[Press release] Access: 05/04/2000. Available from: http://www
Members of the COTRIMO-CI Study Group. Centre de Diagnostic et de Recherches sur le SIDA (CeDReS), CHU de Treichville, Abidjan, Côte d'Ivoire: X. Anglaret, D. Bonard, P. Combe, N. Dogbo, G. Gourvellec, E. Messou, H. Menan, T. Ouassa, F. Sylla-Koko, S. Touré, L. Zio-Diakité; Unité INSERM 330, Université Victor Segalen Bordeaux 2, Bordeaux, France: K. Castetbon, G. Chêne, F. Dabis, S. Lafont, R. Lassalle, V. Leroy, R. Salamon; Formations Sanitaires Urbaines de Yopougon Attié, Yopougon Toit-Rouge, Abobo Doumé, Abobo Sagbé, et CNPS de Yopougon: S. Coulibaly-Koné, J. Gnangbomon, D. Hamien, K. Koné, D. Koumi-Meledje; Service de Médecine, CHU de Yopougon, Abidjan, Côte d'Ivoire: A. Attia, A. Mahassadi, K. Manlan, T. N'Dri-Yoman; Service des Maladies Infectieuses et Tropicales, CHU de Treichville, Abidjan, Côte D'Ivoire: E. Aoussi, M. Coulibaly.