Epidemiology and Social: Concise Communication
Impact of malnutrition and social determinants on survival of HIV-infected adults starting antiretroviral therapy in resource-limited settings
Argemi, Xaviera; Dara, Somb; You, Sengb; Mattei, Jean F.c; Courpotin, Christianc; Simon, Bernardc; Hansmann, Yvesa; Christmann, Daniela; Lefebvre, Nicolasa
aDepartment of Infectious Diseases, Nouvel Hopital Civil, Strasbourg, France
bHIV Care Center, Junka Jek Hospital, Sihanoukville, Cambodia
cDepartment of International Relations, French Red Cross, Paris, France.
Correspondence to Xavier Argemi, Infectious Diseases Department, Nouvel Hopital Civil, 1 Place de l’Hôpital, 67000 Strasbourg, France. Tel: +33 609220088; e-mail: email@example.com
Received 7 August, 2011
Revised 13 March, 2012
Accepted 21 March, 2012
Objectives: Determining the impact of malnutrition, anaemia and social determinants on survival once starting antiretroviral therapy (ART) in a cohort of HIV-infected adults in a rural HIV care centre in Sihanoukville, Cambodia.
Methods: Retrospective and descriptive cohort study of adults starting ART between December 2004 and July 2009. We used the Kaplan–Meier and Cox regression survival analyses to identify predictors of death.
Results: Out of 1002 patients, 49.7% were men; median age was 40; median time of follow-up was 2.4 years and 10.4% died during the follow-up. At baseline, median CD4 cell count was 83 cells/μl, 79.9% were at WHO stage III or IV. In multivariate analysis, malnutrition appeared to be a strong and independent risk factor of death; 11.2% had a BMI less than 16 kg/m2 and hazard ratio was 6.97 [95% confidence interval (CI), 3.51–13.89], 21.5% had a BMI between 16 and 18 kg/m2 and hazard ratio was 2.88 (95% CI, 1.42–5.82), 30.8% had a BMI between 18 and 20 kg/m2 and hazard ratio was 2.18 (95% CI, 1.09–4.36). Severe anaemia (haemoglobin ≤8.4 g/dl) and CD4 cell count below 100 cells/μl also predicted mortality, hazard ratio were 2.25 (95% CI, 1.02–4.34) and 2.29 (95% CI, 1.01–2.97), respectively. Social determinants were not significantly associated with death in univariate analysis.
Conclusion: Malnutrition and anaemia are strong and independent prognostic factors at the time of starting ART. Nutritional cares are essential for the clinical success of HIV programs started in developing countries.
The prognosis of HIV-infected patients has improved dramatically since 1996 with the use of HAART [1,2]. Nevertheless, mortality remains high in resource-poor settings, particularly within the first months of therapy . Among other factors, malnutrition has already been identified as a predictor of death before the widespread of antiretroviral therapy (ART), and further recent cohort studies conducted in developing countries in the era of HAART still identify malnutrition as a strong predictor of death [4,5]. Conversely, social determinants as education level, illiteracy, employment and marital status may impact prognosis but have been poorly studied as potential prognostic factors at ART initiation in large cohort.
Thanks to the efforts that have been done for 7 years by the Cambodian Ministry of Health and nongovernmental organizations to promote comprehensive multifacet HIV prevention and to provide ART; the estimated ART coverage among adults living with HIV has reached nearly 70% in 2007 when coverage was less than 20% in 2004 . Consequently, HIV prevalence in Cambodia has declined from 1.9% in 2000 to 0.9% in 2006 of the adult population (15–49 years old) . However, it remains one of the highest in Southeast Asia . In addition, malnutrition is a critical health issue in the HIV Cambodian population, and previous cohort studies from Médecins Sans Frontières in Cambodia reported that 41% of people living with HIV had a BMI under 18 kg/m2 at ART initiation . Both malnutrition and major social determinants could explain the impact on HIV-related death in the Cambodian population and could be easily managed in this country with poor resource.
Consequently, we conducted a retrospective analysis among adults starting HAART at the French Red Cross (FRC) HIV care centre in Sihanoukville, to identify predictors of mortality, focusing on malnutrition, anaemia and social determinants.
The FRC started in 2004 an HIV program in the province of Kompong Som in cooperation with the Junka Jek hospital, Sihanoukville, Cambodia. Global Fund founded the program. The staff members were mainly Khmer: three doctors, two nurses, two pharmacists, five peer educators and two counselors. The FRC expatriate staff members were limited to the health coordinator of Phnom Penh and a resident physician in Sihanoukville. The FRC and the Reproductive Health Association of Cambodia, a Khmer nongovernmental organization, supported home-based care, including transportation to the hospital when needed, support adherence to ART, help with nutrition. HIV cares were all free of charge, HAART, treatment of opportunistic infections, nutritional support and paraclinical tests. HAART was proposed to patients with WHO stage IV or WHO stage I, II or III with CD4 cell count below 200 cells/μl. Three pre-ART consultations were systematically organized to prepare patients for the treatment, and to collect social, clinical and biological data. The first-line regimen was a triple association of lamivudine (3TC), stavudine (d4T) and nevirapine (NVP) or efavirenz (EFV). Alternative regimens were proposed in case of contraindication or intolerance to zidovudine or abacavir. Nutritional support associated one meal for all the patients the day of consultation, micronutrients supplementation and protein powder. Patients were examined 1 week and 1 month after ART initiation and at least every 2 months after. Clinical follow-up included adherence monitoring with counselors. CD4 cell count was monitored every 6 months. If the consultation was missed, social workers investigated by phone or home visit the reason why (death, treatment interruption, move to unknown place). One computer scientist was dedicated to the database collection.
Nutritional status was defined by BMI [weight (kg) divided by height (m)2]. Body weight and height were measured systematically for all the patients during the pre-ART consultation. BMI was stratified according to established criteria and previous report: less than 16 kg/m2 (severe malnutrition), 16–18 kg/m2 (moderate malnutrition), 18–20 kg/m2 (mild malnutrition) >20 kg/m2 (no malnutrition) [8,9].
Haemoglobin level cutoff points were determined using WHO recommendations  and data from previous studies [11,12]. Anaemia was defined as severe (haemoglobin <8.5 g/dl for both sex) or moderate (haemoglobin range 8.5–14.0 g/dl for men and 8.5–12.0 g/dl for women). Normal haemoglobin level was more than 14 g/dl for men and more than 12 g/dl for women.
CD4 cell counts were performed using flow cytometry (Facscount; Beckton Dickinson, Franklin Lakes, New Jersey, USA) at the Institut Pasteur du Cambodge.
A total of 1214 patients started HAART at the FRC Centre from 1 December 2004 to 31 July 2009. Pregnant women (n = 35), patients under 18 (n = 66) and non-naive patients for ART (103 patients) were excluded from the analysis. Body-weight measurement was not available for eight patients among the 1010 remaining patients (0.8%) and 124 were lost to follow-up. Patients transferred to other clinics during the study period were considered as lost to follow-up, as well as if more than two consultations were missed (after investigation by social worker to determine the reason why).
Relationships between factors hypothesized to influence outcomes were explored using scatter plots, univariate and multivariate tests of association. Main outcome in this study was survival following start of HAART. Baseline characteristics recorded in the dataset included sex, age, WHO stage, BMI, marital status, education level, literacy level, HIV spouse status, type of HAART, haemoglobin level, CD4 cell count. Haemoglobin and CD4 cell count were tested both as continuous and class variables.
Cox proportional hazards models were used to obtain unadjusted and adjusted hazard ratios with 95% confidence intervals (CI). Covariates with P values less than 0.20 in the unadjusted model were included in the adjusted multivariate model. Additionally, any variable specified as a potential confounder a priori based on previous biological plausibility was also included in the model. The proportional hazards assumption was tested by visually inspecting scaled Schoenfeld residuals versus ranked time. All data were censored as of 31 July 2009. For all statistical tests, two-sided P values of less than 0.05 were considered to be statistically significant. Statistical analyses were performed using Stata software, version 11.0 (Stata Corp, College Station, Texas, USA).
Characteristics of the study population
A total of 1002 patients were included in the study. Baseline characteristics are shown in Table 1. There were 498 men (49.7%) and 504 women. Median age was 40 years [interquartile range (IQR) = 35–45 years]. Median BMI was 19.2 kg/m2 (IQR = 17.5–21 kg/m2); 11.2% of the patients had a BMI under 16 kg/m2 and 21.5% within 16–18 kg/m2. At ART initiation, 50% of the patients were in WHO stage III and 30.3% in WHO stage IV. The median CD4 cell count was 83 cells/μl (IQR = 23–182 cells/μl). Haemoglobin level was available for 938 patients; the median level was 11.6 g/dl (IQR = 10.4–12.9 g/dl). Out of these, 19.5% had severe anaemia and 59.9% moderate anaemia. Only 20.6% had a normal haemoglobin level. HIV spouse status was available for 476 (47.5%) patients. Among them, 388 (81.5%) had an HIV-positive partner. Available social data were mainly marital status, education level, literacy level and occupation at ART initiation. Details are shown in Table 1. Antiretroviral regimen included D4T/3TC in 975 patients (97.3%), EFV in 326 patients (32.5%) and NVP in 669 patients (66.8%).
Patients were followed up for a total of 2504.8 person-years and median duration of follow-up was 2.4 years. Out of 1002 patients, 105 (10.4%) died during the follow-up. The mortality rate was 4.19 per 100 person-years (95% CI: 3.46–5.07). During the study period, 42 patients out of the 105 dead patients (40.0%) died within 3 months after HAART initiation and 56 (53.3%) within the first 6 months.
The univariate analysis identified age, baseline WHO stage III and IV, baseline CD4 cell count below 100 cells/μl and severe anaemia as significant factors associated with death. Survival did not differ according to sex, HIV spouse status or social determinants. Data are reported in Table 2.
Using a Cox proportional hazard model, we identified that age, baseline CD4 cell count below 100 cells/μl, BMI class and severe anaemia were independently associated with increased risk of death during the follow-up. Moreover, the strength of the link between BMI classes and the risk of death was revealed in the inverse relationship between hazard ratios and BMI classes. Likewise, patients who started ART with severe anaemia versus normal haemoglobin level were presented with an increased risk of death (hazard ratio, 2.25; 95%CI, 1.02–4.34). Multivariate analysis is reported in Table 2.
Our study found that mortality was significantly associated with low BMI and anaemia but not with social determinants. BMI was used as an indicator of nutritional state. Recent reports have described an association between weight gain and improved outcomes within 6 months after starting ART, for patients with baseline BMI less than 16 kg/m2 or BMI 18 kg/m2 or less . Besides, macronutrient supplementation has shown to be effective in reducing HIV disease progression and mortality rate [15,16] and is now commonly included in HIV care programs. However, effect of global supplementary feeding for malnourished HIV patients on mortality, HIV disease progression or adherence has not been well defined in areas characterized by food scarcity. Two randomized trials conducted in sub-Saharan Africa did not find any positive effect of supplementary feeding on outcome, adherence or disease progression .
Unlike in Africa, data are still missing in Cambodia to estimate properly the incidence of undernourishment, particularly in HIV-infected populations. A retrospective analysis of demographic surveys from 11 sub-Saharan African countries reported that 10.3% of HIV-infected women between 15 and 49 years old had BMIs less than 18.5 kg/m2. Baseline characteristics at the beginning of HAART showed that 11.2% of the patients had a BMI less than 16 kg/m2 and 32.7% had a BMI less than 18 kg/m2. Ferradini et al. also found in the Médecins Sans Frontières population of HIV-infected patient in Phnom Penh, Cambodia, that 41.6% had a BMI less than 18 kg/m2 at baseline. Considering those data and the prognostic value of BMI, food supplementation has to become a priority in HIV-infected patients in Cambodia and probably more widely in the Asian populations.
However, even if BMI seems to be a good indicator of a poor nutritional state, cautions are warranted. Asian populations have significant lower median BMIs in the healthy population, compared with western populations. Those anthropometric variations have already conducted WHO to reconsider cutoff points for overweight and cardiovascular risk [18,19]. Although imperfect, BMI remains an independent predictor of death, easy to obtain and follow after HAART initiation in resource-limited settings.
Correction of nutritional deficiencies includes the correction of iron deficiency and lack of vitamin B12 and B9 which lead to anaemia. As reported in previous studies [11,12,20], anaemia was a strong predictor of death. As a consequence, clinicians should target a normal level of haemoglobin for their patients.
Mortality rate in our cohort was similar [21–23] or lower [24–26] to those previously reported in other developing countries. However, it remained high compared with high-income countries, wherein mortality rates are now approaching the mortality rate of the general population . Excess of mortality was particularly higher within the first months of the therapy [25,26]. Lower CD4 cell count and more advanced clinical stage by the time of starting ART have been identified as pejorative factors.  Likewise, in our study, patients were in more advanced clinical stages (80% with WHO stage III or IV) and had lower CD4 cell count (median CD4 cell count at baseline, 83 cells/μl) as usually seen in western cohorts.
The main limitation of our study is the lack of follow-up data to identify the prognostic value of changes in CD4 cell count, haemoglobin level or BMI stage, as was demonstrated in one previous study in Cambodia. Unfortunately, these data were not available by the time of the analysis in our dataset. Also a potential bias could have been due to patients lost of follow-up as they might have been significantly sicker than those included in the analysis. However, baseline characteristics were available for more than 93.6% of the patients (except for HIV spouse status, 47.5%) and no significant differences were found between the characteristics of the patients who were lost to follow-up and those who were included in the analysis.
Finally, we investigated relationships between survival and several social determinants. No significant associations were found in our analysis, probably due to missing data and a lack of statistical power. Conversely, we didn’t find in the literature any trial focusing on the prognostic value of social factors in similar settings, with the exception of the income level . Unfortunately, this data was missing in our dataset. Nevertheless, employment, social status, and education are major issues in HIV populations, exposed to discrimination, poverty and isolation. They remain central issues in resource-limited settings and should be considered as priorities, as should undernourishment.
The authors want to thank the FRC for developing HIV care program in the province of Kompong Som and Sophie for her technical support. We also thank all the people involved in the Center, the patients and the personnel of the Junka Jek hospital.
Conflicts of interest
There are no conflicts of interest.
1. Hogg RS, Yip B, Kully C, Craib KJ, O'Shaughnessy MV, Schechter MT, et al. Improved survival among HIV-infected patients after initiation of triple-drug antiretroviral regimens. CMAJ 1999; 160:659–665.
2. Mocroft A, Vella S, Benfield TL, Chiesi A, Miller V, Gargalianos P, et al. Changing patterns of mortality across Europe in patients infected with HIV-1. EuroSIDA Study Group. Lancet 1998; 352:1725–1730.
3. Braitstein P, Brinkhof MW, Dabis F, Schechter M, Boulle A, Miotti P, et al. Mortality of HIV-1-infected patients in the first year of antiretroviral therapy: comparison between low-income and high-income countries. Lancet 2006; 367:817–824.
4. Süttmann U, Ockenga J, Selberg O, Hoogestraat L, Deicher H, Müller MJ. Incidence and prognostic value of malnutrition and wasting in human immunodeficiency virus-infected outpatients. J Acquir Immune Defic Syndr Hum Retrovirol 1995; 8:239–246.
5. Tang AM, Forrester J, Spiegelman D, Knox TA, Tchetgen E, Gorbach SL. Weight loss and survival in HIV-positive patients in the era of highly active antiretroviral therapy. J Acquir Immune Defic Syndr 2002; 31:230–236.
7. Ferradini L, Laureillard D, Prak N, Ngeth C, Fernandez M, Pinoges L, et al. Positive outcomes of HAART at 24 months in HIV-infected patients in Cambodia. AIDS 2007; 21:2293–2301.
8. Moh R, Danel C, Messou E, Ouassa T, Gabillard D, Anzian A, et al. Incidence and determinants of mortality and morbidity following early antiretroviral therapy initiation in HIV-infected adults in West Africa. AIDS 2007; 21:2483–2491.
9. Ferro-Luzzi A, Sette S, Franklin M, James WPT. A simplified approach of assessing adult chronic energy deficiency. Eur J Clin Nutr 1992; 46:173–186.
11. Harris RJ, Sterne JA, Abgrall S, Dabis F, Reiss P, Saag M, et al. Antiretroviral Therapy Cohort CollaborationPrognostic importance of anaemia in HIV type-1-infected patients starting antiretroviral therapy: collaborative analysis of prospective cohort studies. Antivir Ther 2008; 13:959–967.
12. O’Brien ME, Kupka R, Msamanga GI, Saathoff E, Hunter DJ, Fawzi WW. Anemia is an independent predictor of mortality and immunologic progression of disease among women with HIV in Tanzania. J Acquir Immune Defic Syndr 2005; 40:219–225.
13. Koethe JR, Lukusa A, Giganti MJ, Chi BH, Nyirenda CK, Stringer JS, et al. Association between weight gain and clinical outcomes among malnourished adults initiating antiretroviral therapy in Lusaka, Zambia. J Acquir Immune Defic Syndr 2010; 53:507–513.
14. Madec Y, Szumilin E, Genevier C, Ferradini L, Balkan S, Pujades M, et al. Weight gain at 3 months of antiretroviral therapy is strongly associated with survival: evidence from two developing countries. AIDS 2009; 23:853–861.
15. Schroecksnadel K, Zangerle R, Fuchs D. Multivitamin supplements and HIV disease progression. N Engl J Med 2004; 351:1353–1354.
16. Fawzi WW, Msamanga GI, Spiegelman D, Wei R, Kapiga S, Villamor E, et al. A randomized trial of multivitamin supplements and HIV disease progression and mortality. N Engl J Med 2004; 351:23–32.
17. Koethe JR, Chi BH, Megazzini KM, Heimburger DC, Stringer JS. Macronutrient supplementation for malnourished HIV-infected adults: a review of the evidence in resource-adequate and resource-constrained settings. Clin Infect Dis 2009; 49:787–798.
18. Choo V. WHO reassesses appropriate body-mass index for Asian populations. Lancet 2002; 360:235.
19. WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies.Lancet 2004; 363:157–163.
20. Mocroft A, Kirk O, Barton SE, Dietrich M, Proenca R, Colebunders R, et al. Anaemia is an independent predictive marker for clinical prognosis in HIV-infected patients from across Europe. EuroSIDA study group. AIDS 1999; 13:943–950.
21. Etard JF, Ndiaye I, Thierry-Mieg M, Guèye NF, Guèye PM, Lanièce I, et al. Mortality and causes of death in adults receiving highly active antiretroviral therapy in Senegal: a 7-year cohort study. AIDS 2006; 20:1181–1189.
22. De Beaudrap P, Etard JF, Ecochard R, Diouf A, Dieng AB, Cilote V, et al. Change over time of mortality predictors after HAART initiation in a Senegalese cohort. Eur J Epidemiol 2008; 23:227–234.
23. Lutalo T, Gray RH, Wawer M, Sewankambo N, Serwadda D, Laeyendecker O, et al. Survival of HIV-infected treatment-naive individuals with documented dates of seroconversion in Rakai, Uganda. AIDS 2007; 21 (Suppl 6):S15–S19.
24. Zachariah R, Harries K, Moses M, Manzi M, Line A, Mwagomba B, et al. Very early mortality in patients starting antiretroviral treatment at primary health centres in rural Malawi. Trop Med Int Health 2009; 14:713–721.
25. Stringer JS, Zulu I, Levy J, Stringer EM, Mwango A, Chi BH, et al. Rapid scale-up of antiretroviral therapy at primary care sites in Zambia: feasibility and early outcomes. JAMA 2006; 296:782–793.
26. Zachariah R, Fitzgerald M, Massaquoi M, Pasulani O, Arnould L, Makombe S, et al. Risk factors for high early mortality in patients on antiretroviral treatment in a rural district of Malawi. AIDS 2006; 20:2355–2360.
27. Antiretroviral Therapy Cohort Collaboration. Mortality of HIV-infected patients starting potent antiretroviral therapy: comparison with the general population in nine industrialized countries.Int J Epidemiol 2009; 38:1624–1633.
28. Cornell M, Myer L, Kaplan R, Bekker LG, Wood R. The impact of gender and income on survival and retention in a South African antiretroviral therapy programme. Trop Med Int Health 2009; 14:722–731.
This article has been cited 3 time(s).
Plos OneRelationship between Food Insecurity and Mortality among HIV-Positive Injection Drug Users Receiving Antiretroviral Therapy in British Columbia, CanadaPlos One
Nutrition JournalImpact of nutritional supplementation on immune response, body mass index and bioelectrical impedance in HIV-positive patients starting antiretroviral therapyNutrition Journal
Plos OnePrevalence and Characteristics Associated with Malnutrition at Hospitalization among Patients with Acquired Immunodeficiency Syndrome in BrazilPlos One
anaemia; HIV; malnutrition; mortality; social determinants
© 2012 Lippincott Williams & Wilkins, Inc.
Highlight selected keywords in the article text.