Total serum and intraerythrocyte magnesium concentrations in hemodialysis patients using different dialysate solutions.

Beside routinely used 0.5 mmol/L dialysate-magnesium, higher dialysate-magnesium (1.0 mmol/L) was recently introduced. The aim of this study was to evaluate the impact of different dialysate-magnesium on serum and intraerythrocyte levels of magnesium (Mg) before and after dialysis. The study included 43 patients receiving chronic hemodialysis, divided into two groups based on dialysate-magnesium (0.5 or 1.0 mmol/L) used prior to study initiation and during 12 months of follow-up. Blood samples were taken at the mid-week dialysis; total serum Mg was measured colorimetrically and intraerythrocyte Mg by atomic absorption spectrophotometry. Hypermagnesiemia-associated complications were observed for 12 months. Total serum Mg was 1.14 ± 0.19 mmol/L before and 0.95 ± 0.16 mmol/L after dialysis in patients using lower dialysate-Mg (p < 0.001), whereas it was 1.47 ± 0.25 mmol/L before and 1.49 ± 0.18 mmol/L after dialysis in patients using higher dialysate-Mg (p = 0.926). Intraerythrocyte Mg was 1.98 ± 0.34 mmol/L before and 1.97 ± 0.28 mmol/L after dialysis in the lower dialysate-Mg group (p = 0.939), while it was 2.09 ± 0.37 mmol/L before and 2.19 ± 0.48 mmol/L after dialysis in the higher dialysate group (p = 0.067). After 12 months total serum Mg decreased in both the groups, remaining lower in 0.5 mmol/L dialysate-Mg group. No hypermagnesiemia-related symptoms occur during 12 months of follow-up in both the groups. In patients using lower dialysate-Mg total serum Mg remains within the reference range and shows postdialytic decline, while in higher dialysate-Mg group it exceeded reference range before and after dialysis without significant intradialytic change. The intraerythrocyte values remain within reference range with both dialysates used. No clinical signs and symptoms of hypermagnesiemia occur during longer administration of higher dialysate-magnesium despite high total serum Mg level.

Influence of SLCO1B1 polymorphisms on lopinavir Ctrough in Serbian HIV/AIDS patients.

AIMS: Lopinavir (LPV) is not a first-line regimen. According to recent WHO data, LPV usage in low- and middle-income countries accounted for approximately 52% of the adult and 23% of the paediatric protease inhibitor market in 2017. Since LPV is a substrate for the SLCO1B1 (OATP1B1) transporter, the aim of this study was to assess the impact of SLCO1B1 polymorphisms (rs11045819, rs4149032 and rs4149056) on LPV trough plasma concentrations (Ctrough ) in Serbian patients. METHODS: Plasma samples from 104 HIV/AIDS Caucasians were collected. LPV Ctrough was quantified using liquid-chromatography-mass spectrometry. Genotyping was carried out using real-time-PCR-based allelic discrimination. One-way analysis of variance, t test and linear regression were used for data analysis. RESULTS: The overall mean (SD) LPV Ctrough was 5885 ± 2755 ng/mL. Significant differences were between patients with different rs11045819 genotypes: CC (LPV median Ctrough = 6072 ng/mL, interquartile range (IQR) = 4318-7617 ng/mL), CA (LPV median Ctrough = 4987 ng/mL, IQR = 4300-6295 ng/mL) and AA (LPV median Ctrough = 3648 ng/mL, IQR = 1949-4072 ng/mL) (P = .005). Significant differences were also observed according to rs4149032 genotype: CC (LPV median Ctrough = 6027 ng/mL, IQR =4548-8250 ng/mL), CT (LPV median Ctrough = 5553 ng/mL, IQR = 4300-6888 ng/mL) and TT (LPV median Ctrough = 4408 ng/mL, IQR = 3361-5233 ng/mL) (P = .007). For rs4149056 a statistically significant difference between T-homozygotes (LPV median Ctrough = 5434 ng/mL, IQR = 3855-6830 ng/mL), heterozygotes (LPV median Ctrough = 6707 ng/mL, IQR = 5088-8063 ng/mL) and C-homozygotes (LPV median Ctrough = 13906 ng/mL, IQR = 12946-14866 ng/mL) was observed (P = .002). In multivariate regression analysis, only the SLCO1B1 rs4149056 polymorphism was independently associated with higher LPV Ctrough (ß = 2834.5 [1442-4226.9] ng/mL [P = .001]). CONCLUSIONS: Our results demonstrate a statistically significant influence of the SLCO1B1 rs4149056 polymorphism on higher LPV Ctrough in Caucasian HIV/AIDS patients.

Higher Levels of Cystatin C in HIV/AIDS Patients with Metabolic Syndrome.

Data about Cystatin-C levels in HIV-infected patients with metabolic syndrome (MetS) are still limited. Therefore, the aim of this study was to evaluate the possible correlations of serum levels of Cystatin-C in HIV/AIDS patients treated with combined antiretroviral therapy (cART) with or without MetS. This cross-sectional study included 89 HIV/AIDS Caucasian patients receiving cART at the HIV/AIDS Centre Belgrade, Serbia, divided into two groups according to the presence of MetS. Cystatin-C and other biochemical parameters were measured using Cytokine-Array-I, Metabolic-Array-I and Metabolic-Array-II, at the Department of Clinical Biochemistry, Royal Free Hospital and University College London, UK. A linear regression model was performed to evaluate which clinical and laboratory variables had an independent effect on Cystatin-C levels in HIV/AIDS patients. There were 33 (37%) patients with MetS and 56 (63%) without MetS. Patients with and without MetS were homogenous for age, duration of cART, number of cART combinations and CD4+ T cell count. Statistically increased Cystatin-C levels were observed in HIV/AIDS patients with MetS (p = 0.017), when compared to patients without MetS. Data showed a positive correlation of Cystatin-C and C-reactive protein (r = 0.349, p = 0.001). Using linear regression modelling, significant correlations were obtained between Cystatin-C and MetS in univariate analysis (p < 0.001). Cystatin-C levels were significantly higher in HIV/AIDS patients with MetS versus without MetS. Early assessment of MetS using Cystatin-C as a marker may ultimately help increase the lifespan of HIV/AIDS patients, as these patients appear to be at high risk of cardiovascular diseases.

Treatment outcomes of first-line antiretroviral therapy in HIV-1-positive patients in Serbia.

INTRODUCTION: Access to combination antiretroviral treatment (cART) and toxicity profiles of antiretroviral medications have significantly improved during the last three decades. In order to optimise treatment outcomes, achieve favourable virological suppression and immunological status, balanced with potential adverse effects of cART, it is considered beneficial to maintain first-line antiretroviral treatment for as long as possible. However, the Republic of Serbia, as a resource-limited setting, often experiences interruptions to drug supplies. Data are very limited in Serbia concerning the initial antiretroviral regimens prescribed and the reasons for treatment changes. AIMS: The aim of this study was to determine the most frequently prescribed antiretroviral drugs within first-line cART regimens in drug-naïve patients in Serbia and the reasons for switching drugs. METHODS: All HIV-infected individuals who started cART at the HIV/AIDS Center of Infectious and Tropical Diseases, Clinical Centre of Serbia, from 1 January 2004 until 1 July 2014 were included. A cohort of 339 patients were retrospectively analysed to review their initial treatment regimens. All analyses were performed using the SPSS statistical package version 11.0. Descriptive measurements and Kaplan-Meier survival curves were used. RESULTS: The most frequently prescribed nucleoside reverse transcriptase inhibitor (NRTI) backbones in the cART regiment were fixed combinations of abacavir and lamivudine (n=181, 53.3%) and of zidovudine and lamivudine (n=103, 30.5%). Efavirenz was the most commonly prescribed 'third' drug (n=254, 75%). Where given, reasons for switching initial cART were shortage of antiretroviral drugs (e.g. out of stock, n=53, 37.6%), toxicity (n=49, 34.3%), physician choice (n=21, 14.6%), resistance (n=15, 10.6%), and patient choice (n=4, 2.9%). Mean duration of first-line cART was 20±17 months. CONCLUSION: The most frequently prescribed initial cART regimen in Serbia is not the preferred first choice, but an alternative option according to the international antiretroviral treatment guidelines. Duration of first-line cART is short and a switch to second-line cART is often made due to a shortage of antiretroviral medications and the more severe side effects resulting from the use of older drugs.

Influence of manual thrombus aspiration on left ventricular diastolic function in patients with ST-segment elevation myocardial infarction treated with primary percutaneous coronary intervention.

INTRODUCTION: Data on effects of thrombus aspiration on left ventricular diastolic function in ST-elevation myocardial infarction (STEMI) population are scarce. OBJECTIVE: We sought to compare echocardiographic indices of the diastolic function and outcomes in STEMI patients treated with and without manual thrombus aspiration, in an academic, high-volume percutaneous coronary intervention (PCI) center. METHODS: A total of 433 consecutive patients who underwent primary PCI in 2011-2012 were enrolled in the study. Patients were not eligible for the study if they already suffered a myocardial infarction, had been previously revascularized, received thrombolytics, presented with cardiogenic shock, had significant valvular disease, atrial fibrillation or had previously implanted pacemaker. Comprehensive echocardiogram was performed within 48 hours. During follow-up patients'status was assessed by an office visit or telephone interview. RESULTS: Patients treated with thrombus aspiration (TA+, n=216) had similar baseline characteristics as those without thrombus aspiration (TA-, n = 217). Groups had similar total ischemic time (319 ± 276 vs. 333 ± 372 min; p = 0.665), but TA+ group had higher maximum values of troponin I (39.5 ± 30.5 vs. 27.6 ± 26.9 ng/ml; p < 0.001). The echocardiography revealed similar left ventricular volumes and systolic function, but TA+ group had significantly higher incidence of E/e' > 15, as a marker of severe diastolic dysfunction' (TA+ 23.1% vs. TA- 15.2%; p = 0.050). During average follow-up of 14 ± 5 months, major adverse cardiac/cerebral events occurred at the similar rate (log rank p = 0.867). CONCLUSION: Thrombus aspiration is associated with a greater incidence of severe diastolic dysfunction in unselected STEMI patients treated with primary PCI, but it doesn't influence the incidence of major adverse cardiovascular events.

Choice of first-line antiretroviral therapy regimen and treatment outcomes for HIV in a middle income compared to a high income country: a cohort study.

BACKGROUND: The range of combination antiretroviral therapy (cART) regimens available in many middle-income countries differs from those suggested in international HIV treatment guidelines. We compared first-line cART regimens, timing of initiation and treatment outcomes in a middle income setting (HIV Centre, Belgrade, Serbia - HCB) with a high-income country (Royal Free London Hospital, UK - RFH). METHODS: All antiretroviral-naïve HIV-positive individuals from HCB and RFH starting cART between 2003 and 2012 were included. 12-month viral load and CD4 count responses were compared, considering the first available measurement 12-24 months post-cART. The percentage that had made an antiretroviral switch for any reason, or for toxicity and the percentage that had died by 36 months (the latest time at which sufficient numbers remained under follow-up) were investigated using standard survival methods. RESULTS: 361/597 (61 %) of individuals initiating cART at HCB had a prior AIDS diagnosis, compared to 337/1763 (19 %) at RFH. Median pre-ART CD4 counts were 177 and 238 cells/mm(3) respectively (p < 0.0001). The most frequently prescribed antiretrovirals were zidovudine with lamivudine (149; 25 %) and efavirenz [329, 55 %] at HCB and emtricitabine with tenofovir (899; 51 %) and efavirenz [681, 39 %] at RFH. At HCB, a median of 2 CD4 count measurements in the first year of cART were taken, compared to 5 at RFH (p < 0.0001). Median (IQR) CD4 cell increase after 12 months was +211 (+86, +359) and +212 (+105, +318) respectively. 287 (48 %) individuals from HCB and 1452 (82 %) from RFH had an available viral load measurement, of which 271 (94 %) and 1280 (88 %) were <400 copies/mL (p < 0.0001). After 36 months, comparable percentages had made at least one antiretroviral switch (77 % HCB vs. 78 % RFH; p = 0.23). However, switches for toxicity/patient choice were more common at RFH. After 12 and 36 months of cART 3 % and 8 % of individuals died at HCB, versus 2 % and 4 % at RFH (p < 0.0001). CONCLUSION: In middle-income countries, cART is usually started at an advanced stage of HIV disease, resulting in higher mortality rates than in high income countries, supporting improved testing campaigns for early detection of HIV infection and early introduction of newer cART regimens.

Fondaparinux monitoring in a patient with heparin-induced thrombocytopenia on hemodialysis.

Introduction: Heparin-induced thrombocytopenia associated to hemodialysis is rare. In case when citrate dialysis and/or non-heparin anticoagulants are not available, only possible medication to use for anticoagulation during hemodialysis is fondaparinux. However, laboratory monitoring of fondaparinux based on anti-Xa activity in dialysis patients has not been sufficiently documented yet. Case Outline: We created a local anti-factor Xa assay for measuring fondaparinux plasma concentration and efficacy in a patient with heparin-induced thrombocytopenia during hemodialysis. Fondaparinux given subcutaneously increases risk of adverse events due to its extended release and prolonged maintenance of toxic levels. When used intravenously fondaparinux remains safe, with reached steady-state level within dialysis and low risk of toxicity afterwards. Conclusion: Fondaparinux may be used as an alternative anticoagulant medication during hemodialysis in patients who develop heparin-induced thrombocytopenia. Adequate dose must be adjusted to patients' dry weight (0.03 mg/kg intravenously) and fondaparinux anti-coagulation monitoring must be provided.

Comparison of four international cardiovascular disease prediction models and the prevalence of eligibility for lipid lowering therapy in HIV infected patients on antiretroviral therapy.

AIM: To compare four cardiovascular disease (CVD) risk models and to assess the prevalence of eligibility for lipid lowering therapy according to the 2013 American College of Cardiology/American Heart Association (ACC/AHA) guidelines, European AIDS Clinical Society Guidelines (EACS), and European Society of Cardiology and the European Atherosclerosis Society (ESC/EAS) guidelines for CVD prevention in HIV infected patients on antiretroviral therapy. METHODS: We performed a cross-sectional analysis of 254 consecutive HIV infected patients aged 40 to 79 years who received antiretroviral therapy for at least 12 months. The patients were examined at the HIV-treatment centers in Belgrade and Zagreb in the period February-April 2011. We compared the following four CVD risk models: the Framingham risk score (FRS), European Systematic Coronary Risk Evaluation Score (SCORE), the Data Collection on Adverse Effects of Anti-HIV Drugs study (DAD), and the Pooled Cohort Atherosclerotic CVD risk (ASCVD) equations. RESULTS: The prevalence of current smoking was 42.9%, hypertension 31.5%, and hypercholesterolemia (>6.2 mmol/L) 35.4%; 33.1% persons were overweight, 11.8% were obese, and 30.3% had metabolic syndrome. A high 5-year DAD CVD risk score (>5%) had substantial agreement with the elevated (≥7.5%) 10-year ASCVD risk equation score (kappa=0.63). 21.3% persons were eligible for statin therapy according to EACS (95% confidence intervals [CI], 16.3% to 27.4%), 25.6% according to ESC/EAS (95% CI, 20.2% to 31.9%), and 37.9% according to ACC/AHA guidelines (95% CI, 31.6 to 44.6%). CONCLUSION: In our sample, agreement between the high DAD CVD risk score and other CVD high risk scores was not very good. The ACC/AHA guidelines would recommend statins more often than ESC/EAS and EACS guidelines. Current recommendations on treatment of dyslipidemia should be applied with caution in the HIV infected population.

Prevalence and concordance of high cardiovascular disease scores in HIV/AIDS patients from Croatia and Serbia with four international algorithms.

INTRODUCTION: We evaluated cardiovascular risks in HIV-infected patients from Croatia and Serbia and the eligibility for statin therapy as recommended by the 2013 American College of Cardiology/American Heart Association (ACC/AHA) guidelines, European AIDS Clinical Society (EACS) Guidelines and European Society of Cardiology and the European Atherosclerosis Society (ESC/EAS) guidelines for cardiovascular disease (CVD) prevention [1-3]. MATERIALS AND METHODS: A cross-sectional analysis of consecutive patients between 40 and 79 years old who had received antiretroviral therapy for at least 12 months was performed. RESULTS: Of 254 (132 from Croatia and 122 from Serbia) persons included in the study, 76% were male; median age was 49 years. Up to 51.6% of persons had a high CVD risk. The prevalence of current smoking was 42.9%, hypertension 31.5% and hypercholesterolaemia (>6.2 mmol/L) 35.4%. Statins would be recommended to 21.3% (95% CI, 16.3% to 27.4%) of persons by the EACS, 25.6% (95% CI, 20.2% to 31.9%) by ESC/EAS and 37.9% (95% CI, 31.6 to 44.6%) by the ACC/AHA guidelines. A high 5-year data collection on adverse effects of anti-HIV drugs study risk score (>5%) had a moderate agreement with the high (≥20%) 10-year CVD Framingham risk score (kappa=0.47) and high (≥5%) 10-year European systematic coronary risk evaluation score algorithm (kappa=0.47), and substantial agreement with the elevated (≥7.5%) 10-year Pooled Cohort Atherosclerotic CVD risk equation score (kappa=0.63). CONCLUSION: We found a high prevalence of CVD risks in patients from Croatia and Serbia. The ACC/AHA guideline would recommend statins more often than ESC/EAS and EACS guidelines.

HIV/AIDS mortality in a south east European country versus a west European country.

INTRODUCTION: Antiretroviral (ARV) treatment available in low-middle income countries differs as suggested in international HIV-treatment guidelines. Thus, we compared ARV regimens introduced as a first-line therapy, time of initiation, frequency of making combination antiretroviral therapy (cART) switches, frequency of viral and immunological monitoring and treatment outcome in south east European (SEE) country (i.e. HIV Centre in Belgrade, Serbia, (HCB)) and west European country (i.e. Royal Free Centre for HIV Medicine at the Royal Free Hospital London, UK (RFH)). MATERIALS AND METHODS: ARV naïve patients starting cART from 2003 to 2012 were included. Comparisons of the two cohorts were made using a chi-square test or Fisher's exact test for categorical variables and a Mann-Witney U test for continuous variables. Kaplan Meier survival curves were compared using the log rank test. RESULTS: Of 597 patients from HCB, 361 (61%) initiated cART with prior AIDS diagnosed, while 337 (19%) of 1763 patients from RFH. Average baseline CD4+ T cell counts were significantly lower in Serbia than in UK (177 cells/mm(3) vs 238 cells/mm(3)). The total (mediana, IQR) CD4+ T cell count measurements in the first year of cART was 2 (1, 2) at the HCB, while it was statistically significant higher at the RFH 5 (3, 7), respectively (p<0.0001). At the RFH, it appeared that the cART switching is due to patient's preference or toxicity (46%), while the lack of supply and toxicity (37%) were the most important reasons for treatment change in HCB, within the same period of time (p<0.05). Mortality rates were higher at the HCB versus RFH (p<0.0001). After 12, 24 and 36 months of cART, 3%, 5% and 8% of patients died in HCB, whereas 2%, 3% and 4% of patients died in RFH, respectively (Figure 1). CONCLUSIONS: In south European countries, as a consequence of low testing rate, ARV treatment is introduced at an advanced stage of disease, having a high mortality rate as a consequence. Switching within ARV drugs appears often due to lack of drug supplies and frequently drug-related toxicity in south east Europe, while in the east European country due to patient's preferences and rarely due to drug-related toxicity.

CYP2B6 516G>T (rs3745274) and smoking status are associated with efavirenz plasma concentration in a Serbian cohort of HIV patients.

BACKGROUND: Genetic factors have been associated with efavirenz (EFV) plasma concentrations in different populations. In this study, we investigated the effects of CYP2B6 516G>T (rs3745274), CYP2B6 c.485-18C>T (rs4803419), CAR c.540C>T (rs2307424), CAR c.152-1089T>C (rs3003596), and smoking status in a cohort of Serbian patients with HIV. METHODS: Patients with HIV positive, all whites, were recruited from the HIV/AIDS Center at the Infectious and Tropical Diseases Hospital, University of Belgrade Teaching Hospital, Belgrade, Serbia. Mid dose (10-14 hours after dose) EFV plasma concentration was determined using a validated liquid chromatography/tandem mass spectrometry method. Genotyping for CYP2B6 516G>T (rs3745274), CYP2B6 c.485-18C>T (rs4803419), CAR c.540C>T (rs2307424), and CAR c.152-1089T>C (rs3003596) was conducted using allelic discrimination real-time polymerase chain reaction assay. One-way analysis of variance, Mann-Whitney test, Pearson or Spearman correlation, and multiple linear regression were used for data analysis. RESULTS: Minor allele frequencies were similar to frequencies reported in other European populations. The overall mean (95% confidence interval) plasma EFV concentration was 2800 ng/mL (2460-3140). Significant differences between patients based on CYP2B6 516G>T (rs3745274) genotypes were observed: GG (n = 60), 2320 (range, 2160-2480) ng/mL; GT (n = 30), 3230 (range, 2790-3670) ng/mL; and TT (n = 2), 10,700 (range, 6170-15,300) ng/mL (P = 2.0 × 10). In multivariate linear regression analysis, CYP2B6 516G>T (rs3745274) [ß = 1770 (1230 to 2310) ng/mL, P < 0.0001] and smoking status [ß = -464 (-1250 to -43.3) ng/mL, P = 0.038] were independently associated with plasma EFV concentrations. CONCLUSIONS: The effects of CYP2B6 516G>T (rs3745274) and smoking status on EFV plasma concentration in the Serbian population have been established for the first time.

Clinical and immunologic outcomes of HAART-treated HIV-infected women in resource constrain settings: the Belgrade Study.

We performed a study to identify factors related to favorable response to highly active-antiretroviral therapy (HAART) in HIV-infected women. A retrospective study was performed on 216 women who had initiated HAART from January 1, 1998 to December 31, 2012, at the HIV/AIDS Center, Belgrade, Serbia. Participants were followed-up for 8.2 ± 3.4 years. The mean age was 37 ± 9.7 years. During follow-up, it was found that 26 patients had died. Clinical AIDS at initiation of HAART was observed in 43.9% patients, while 64.8% had a CD4+ T-cell count below 200 cells/µL. Multivariate analyses revealed that the single factor independently related to a favorable response to HAART was good compliance (odds [OR] ratio for survival = 2.9, 95% confidence intervals [CI] = 1.0-8.6, p = 0.03), while a baseline CD4+ T-cell count below 100 cells/µL, hepatitis C virus coinfection, and aged 40 years and older were all associated with an unfavorable response to HAART (OR = 0.28, 95% CI = 0.15-0.52, p < 0.001; OR = 0.49, 95% CI = 0.22-0.8, p = 0.008; OR = 0.41, 95% CI = 0.21-0.79, p = 0.008, respectively). The estimated 14-year-survival was 100% in patients with sustained viral suppression, regardless of the CD4+ counts achieved (p = 0.6, log-rank). If women with advanced HIV-related immunodeficiency reach and maintain optimal viral suppression during HAART, regardless of the level of immune recovery, and if they continue to maintain this suppression for up to a mean 8 years of treatment, their prognosis may be fairly good, even in resource-limited settings.

Treatment outcome of HAART-treated patients in a resource-limited setting: the Belgrade Cohort Study.

INTRODUCTION: We evaluated the effects of highly-active-antiretroviral-therapy (HAART) in a resource-limited settings. METHODS: A cross-sectional study was performed in patients who had initiated HAART at the HIV/AIDS-Center, Belgrade, Serbia. Treatment response was considered favorable in case of the achievement of undetectable HIVRNA plasma-viral-load (pVL<50 copies/µL), and with the CD4+ T-cell counts increased above 350cells/µL. The treatment failure was defined as pVL over 1.7 log10 copies/mL, regardless of immunological improvement. RESULTS: Eight hundred and forty HIV infected patients were followed-up for 8.2±3.4years. Out of 697 patients available for follow-up, 113 (16.2%) patients died, 44 (6.3%) experienced treatment failure, while 540 (77.5%) had sustained undetectable viremia. In 419 (60.1%) favorable treatment response was achieved, while the dissociation between immunological and virological responses to HAART occurred in 121 (14.4%). A baseline CD4+ T-cell counts above 200 cells/µL was the single independent predictor of a favorable treatment response (HR=2, 95%CI=1.69-2.61, P=0.001), while pre-treatment with ART, HCV co-infection and AIDS at the time of treatment initiation, were all factors preventing a favorable response (HR=0.27, 95%CI=0.19-0.36, P=0.001; HR=0.75, 95%CI=0.56-0.95, P=0.02; HR=0.73, 95%CI=0.17-0.95, P=0.018, respectively). A sustained viral suppression was an independent predictor of survival (HR=0.2, 95% CI 0.07-0.61, P=0.004). HAART treated HIV-infected patients who reach and maintain undetectable viremia, have an 80% probability of a 14-years survival (P=0.08, log-rank). CONCLUSION: If patient with advanced HIV-related immunodeficiency reach and maintain optimal viral suppression during HAART, regardless of the level of immune recovery, and if they continue to maintain this, their prognosis may be fairly good even in the resource-limited settings.