A dissertation submitted to the Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of MSc (Med) Johannesburg, 2011 / Peripheral sensory neuropathy is a common neurological complication of antiretroviral therapy, typically occurring within 6-months of starting Highly Active Antiretroviral Therapy (HAART) which includes stavudine. Therefore, the primary aim of the study was to determine the 6-month incidence of ATN in patients free of neuropathy and beginning stavudine-based HAART for the first time. Also, we examined whether initiating stavudine-based HAART altered the symptoms of patients who had a pre-existing, virus-mediated distal symmetrical polyneuropathy (HIV-DSP). Seventy-five HIV-positive patients were screened for neuropathy, at the Chris-Hani Baragwanath Hospital, using the AIDS Clinical Trials Group neuropathy screening tool. The bilateral presence of atleast one sign (decreased vibration sense in the great toe or absent ankle reflex) and one symptom (pain, paraesthesia or numbness) in the feet was indicative of neuropathy. On recruitment, 52 patients presented without neuropathy and 13 patients presented with HIV-DSP. After 3- months of follow-up (n=46), 23% (10/46) of patients had developed peripheral neuropathy, and by 6-months (n=44), 41% (18/44) of patients had developed neuropathy. Greater disease severity was the only risk factor significantly associated with the development of neuropathy. Eleven (61%) of the 18 patients that developed neuropathy, developed painful symptomatic neuropathy, and only 6 (55%) of these patients were receiving treatment for symptom relief. In patients with HIV-DSP, numbness was the most common symptom reported at baseline and was the only symptom to reduce in frequency across the 6-months. In conclusion, we found that the development of neuropathy is common in the first 6-months of patients initiating stavudine-based HAART.

The Human Immunodeficiency Virus (HIV) which causes Acquired Immunodeficiency Syndrome (AIDS) has caused a global scare with mainly poor African countries suffering the greatest burden. Treatment of HIV is more of palliation rather than cure such that there is no room for treatment interruption if treatment goals are to be met. Antiretroviral treatment is associated with short term and long term side effects which have the potential to negatively impact on the high levels of adherence to treatment that is required to maintain virological suppression and may eventually lead to development of drug resistance and treatment failure. This research aims to identify the extent to which these side effects, through possible poor adherence, impact on treatment successes by measuring the risk that side effects contribute towards treatment failure. Methods Secondary data analysis was conducted on a cohort of patients who initiated ART between 2004 and 2010 at a large tertiary facility in Johannesburg. Patients who were switched to second line ART due to treatment failure were identified. Assessment of side effects on adherence was done. The hazards of side effects among patients switching and not switching to second line were calculated using Cox proportional hazards regression adjusting for other socio-demographic and clinical predictors for treatment failure. Interaction between side effects, gender, age and that of side effects and adherence was investigated. Time dependent covariates were also investigated. Confounding was controlled using multivariate Cox regression analysis. Results There were 5285 patients in the baseline cohort with multiple entry points who contributed 16035 person-years of follow up. The cohort consisted of 63.2% females and 36.8% males. Of these 85.9% were initiated on stavudine (d4T)- based regimen, 7.1% on tenofovir (TDF), 6.3% on zidovudine (AZT)-based regimen and 0.7% on other regimens. The median and mean time at risk per subject was 2.2 and 2.3 years respectively. A total of 770 episodes of side effects due to first line ART were experienced with some patients recording multiple side effects at different time points. Adherence data were found to be missing and incoherent in some of the regimen dosages and could not be used to objectively compare patients. There were 430 patients who were switched to second line ART due to treatment failure. Relative to the group of no side effects, the adjusted hazard ratios for mild, moderate and severe side effects were 1.40 (95% CI=0.94-2.09) p=0.10; 1.72 (95% CI=1.35-2.20) p<0.01 and 1.24 (95% CI=0.65-2.35) p=0.52 respectively. Therefore, overally side effects did not seem to play a role in the time to switch to second line ART. Sex, baseline CD4 cell count, the period during which ART was initiated and the time between date of testing HIV positive and date of initiating were significantly associated with the time to switching to second line ART. Conclusion The study informs that side effects overally may not play a significant role in switching patients from first line to second line ART with the exception of moderate side effects. However, patients who experience side effects should be closely monitored and adequately counselled to help them cope with the side effects so that optimal adherence levels are maintained. Availability of adherence scores or additional information on pills that should have been taken on periods during which pills were reported to have been missed would have made the research more valuable by allowing objective comparison of adherence among patients.

The roll-out of antiretroviral drugs in South Africa started in March 2004. In Mopani district, a rural district of Limpopo Province, the roll-out programme commenced in October 2004. While many resources were invested in this program, no study has assessed the clinical outcomes in this rural district. In addition, most studies conducted in South Africa were conducted in urban and tertiary settings. Assessing clinical outcomes is important in determining whether the program is making the desired clinical difference in the lives of the patients and may serve as feedback into the program for quality improvement purposes. Methodology The study was a retrospective record review of patients who were initiated on antiretroviral (ARV) treatment between December 2007 and November 2008. A structured questionnaire was used to collect data from 124 patient’s files and data was collected up to November 2011. The data collected included patients’ socio-demographic characteristics, clinical outcomes (CD4 count, viral load, presence of opportunistic infections, adverse effects and hospital admissions recorded at 6, 12, 24 and 36 months), the number of patients who were still attending the ARV clinic at 36 months and the reasons why patients are no longer attending the clinic. Data was analysed with Epi-Info and STATA. Results Of the 124 patients, 69% were females, 28% males and 3% did not have their sex specified. The majority of the patients were between 30 and 49 years. There was a significant improvement in CD4 count and viral load between baseline and all timeperiods after the initiation of ARV treatment. The mean CD4 count at baseline was 128The roll-out of antiretroviral drugs in South Africa started in March 2004. In Mopani district, a rural district of Limpopo Province, the roll-out programme commenced in October 2004. While many resources were invested in this program, no study has assessed the clinical outcomes in this rural district. In addition, most studies conducted in South Africa were conducted in urban and tertiary settings. Assessing clinical outcomes is important in determining whether the program is making the desired clinical difference in the lives of the patients and may serve as feedback into the program for quality improvement purposes. Methodology The study was a retrospective record review of patients who were initiated on antiretroviral (ARV) treatment between December 2007 and November 2008. A structured questionnaire was used to collect data from 124 patient’s files and data was collected up to November 2011. The data collected included patients’ socio-demographic characteristics, clinical outcomes (CD4 count, viral load, presence of opportunistic infections, adverse effects and hospital admissions recorded at 6, 12, 24 and 36 months), the number of patients who were still attending the ARV clinic at 36 months and the reasons why patients are no longer attending the clinic. Data was analysed with Epi-Info and STATA. Results Of the 124 patients, 69% were females, 28% males and 3% did not have their sex specified. The majority of the patients were between 30 and 49 years. There was a significant improvement in CD4 count and viral load between baseline and all timeperiods after the initiation of ARV treatment. The mean CD4 count at baseline was 128 The roll-out of antiretroviral drugs in South Africa started in March 2004. In Mopani district, a rural district of Limpopo Province, the roll-out programme commenced in October 2004. While many resources were invested in this program, no study has assessed the clinical outcomes in this rural district. In addition, most studies conducted in South Africa were conducted in urban and tertiary settings. Assessing clinical outcomes is important in determining whether the program is making the desired clinical difference in the lives of the patients and may serve as feedback into the program for quality improvement purposes. Methodology The study was a retrospective record review of patients who were initiated on antiretroviral (ARV) treatment between December 2007 and November 2008. A structured questionnaire was used to collect data from 124 patient’s files and data was collected up to November 2011. The data collected included patients’ socio-demographic characteristics, clinical outcomes (CD4 count, viral load, presence of opportunistic infections, adverse effects and hospital admissions recorded at 6, 12, 24 and 36 months), the number of patients who were still attending the ARV clinic at 36 months and the reasons why patients are no longer attending the clinic. Data was analysed with Epi-Info and STATA. Results Of the 124 patients, 69% were females, 28% males and 3% did not have their sex specified. The majority of the patients were between 30 and 49 years. There was a significant improvement in CD4 count and viral load between baseline and all timeperiods after the initiation of ARV treatment. The mean CD4 count at baseline was 128 cells/mm3; it increased to 310 cells/mm3 at 6 months, 380 cells/mm3 at 12 months and 470 cells/mm3 at 24 months. By 6 months, 67% of the patients had achieved viral suppression, but at 24 months, patients started having viral rebound. During the study, 20 patients fell pregnant and four patients fell pregnant twice. Overall, pregnant patients had a significantly higher viral load compared to non-pregnant patients (p-values = 0.015 at 6 months, 0.002 at 12 months and 0.027 at 24 months). Seventy two percent of patients were retained in the program at 36 months. Of the 28% that were no longer attending the clinic, 11.3% were transferred to other institutions, 6.5% were down referred to clinics, 3.2% died, 3.2% defaulted and 3.2% were lost to follow-up. Conclusion This study shows that good clinical outcomes can be achieved within an antiretroviral rollout program in a rural hospital. The biggest magnitude of clinical benefits was observed in the first six months after the initiation of ARV treatment with threats of viral rebound thereafter. There was good patient retention at 36 months after initiation of ARV treatment and a significant difference in viral load between pregnant and non-pregnant patients. The high rate of unplanned pregnancy signifies the need to place closer attention to family planning among female patients on antiretroviral treatment.

No description available.

A research report submitted to the Faculty of Health Sciences, the University of the Witwatersrand, Johannesburg, in partial fulfilment of the requirements for the degree of Master of Science in Medicine in Child Health Neurodevelopment Johannesburg, 2013 / Human Immunodeficiency Virus (HIV) infection in infancy may influence the developing brain and lead to adverse neurodevelopmental consequences. We aim to describe the neurodevelopmental characteristics of a cohort of young children infected with HIV prior to antiretroviral therapy (ART) initiation and after achieving viral suppression. A retrospective analysis of data collected as part of a randomised equivalence trial between April 2005 and May 2009, at a hospital in Johannesburg, South Africa. 195 HIV-infected children under 2 years of age were assessed. A simple, inexpensive screening questionnaire (Ages and Stages Questionnaire - ASQ) was used to identify neurodevelopmental delays. The ASQ was administered prior to ART initiation, and again after viral suppression on a protease inhibitor-based regimen had been achieved. Median age pre-ART was 8.8 months (range 2.2 - 24.9), 53.9% were male. Mean time to viral suppression was 9.4 months (range 5.9 - 14.5) and the ASQ was administered to 108 caregivers at this time. Compared to pre-ART, at viral suppression, there was significant reduction in the proportion of children failing the gross motor (31.5% vs. 13%, p<0.01), fine motor (21.3% vs. 10.2%, p=0.02), problem solving (26.9% vs. 9.3%, p<0.001) and personal social (17.6% vs. 7.4%, p=0.02) domains. The proportion of children failing the communication domain was similar at each time point (14.8% vs. 12%, p=0.61). At time of viral suppression 10.2% failed at least one of the five domains. Achieving viral suppression on ART resulted in significant improvements in the neurodevelopmental function of young HIV-infected children, however, neurodevelopmental problems still persisted in a large proportion. Appropriate screening for neurodevelopmental delay and timely referral could help improve outcomes.

HIV-associated peripheral neuropathy (PN), and related functional limitations that affect the quality of life (QoL), may now be one of the most formidable challenges in the health care of people living with HIV (PLHIV). The most common PN is distal sensory polyneuropathy (DSP). It is likely that there is a high prevalence of PN among PLHIV in Rwanda. The available data on the prevalence of PN are poor and there are none on how PN is associated with functional abilities and the QoL of PLHIV, which can guide management. In addition, current management of PN is mostly related to symptomatic management and is mainly pharmacological which may not rehabilitate the neuromuscular function that has been affected by PN. This thesis planned to re-validate and adapt the lower extremity functional scale (LEFS) and the brief peripheral neuropathy screen (BPNS), establish the prevalence of PN, and determine the effects of physiotherapeutic exercises on PN, lower extremity functional limitations and QoL, among Rwandan PLHIV receiving antiretroviral therapy (ART). Methods Study 1 translated LEFS from English to Kinyarwanda, modified it accordingly, and tested its reliability among 50 adult PLHIV on ART. The study also piloted

A research report submitted to the faculty of Health Sciences, University of the Witwatersrand, Johannesburg, in partial fulfillment of the requirements for the degree of Master of Science in Medicine (Pharmacotherapy) / Background: The decision of when to start treatment in an HIV-infected individual has always been problematic as far as CD4 count is concerned. Aims: To determine the clinical outcome of patients who commence HAART at different CD4 cell count levels. Method: Retrospective analysis of records of a cohort of patients who are received ART at workplace wellness clinics in three mines in Limpopo province from January 2003 to December 2009. Patients were divided into three groups based on their baseline, group A (CD4 <100), group B (CD4 101-200) and group C (CD4 201-250) Each patient’s data was analyzed one year after his/her commencement. Results: The percentage of patients who died in group A (16%) differs significantly from the percentage of patients who died in group B (4%) (Fisher exact test p= 0.038) and also differ significantly from the percentage of patients who died in group C (0%) (Fisher exact test p= 0.011). The percentages of patients who developed TB in the three groups are 8%, 8% and 2.9% respectively. When compared statistically, these percentages do not differ significantly (Fisher exact test p=0.059).The percentages of patients with severe bacterial pneumonia in the three groups (2%, 2% and 0% respectively) do not differ significantly (Fisher exact test p=0,276).The percentage of hospital admissions for patients in group A (18%) differ significantly from the percentage in group B (6%) and the percentage in group C (6%) (Fisher exact test p= 0.05). The percentage of patients with weight loss of more than 10% of baseline value in group A (24%) differ significantly from the percentage in group B (4%) (Fisher exact test p= 0.003) and also differ significantly to from the percentage in group C (0%) (Fisher exact test p= 0.001). The percentage of patients with undetectable viral load in group B (89%) is significantly different from the percentage in group A (69%) (Fisher exact test p= 0.03) and is also significantly different from the percentage in group C (61%) (Fisher exact test p= 0.008).The change in mean CD4 cell count was found to be statistically significant within each group (paired t test, p<0.0001), but the mean changes between the three groups (132,141 and 172) respectively, do not differ significantly (ANOVA test). Conclusion: Patients with baseline CD4 cell count of less than 100 have a poor clinical outcome when compared to patients with baseline CD4 cell count of more than 100. Efforts must be made to identify patients early before CD4 cell count fall to below 100 and preferably initiate HAART when CD4 cell count is above 200.

A research report submitted to the Faculty of Health Sciences, University of Witwatersrand, Johannesburg, in partial fulfilment of the requirements for the degree of Master of Medicine in the branch of Diagnostic Radiology Johannesburg, 2013 / INTRODUCTION: There is limited information on the radiographic presentation of children eligible to start HAART in resource-limited settings. OBJECTIVES: Determine radiographic patterns on pre-HAART chest X-rays (CXRs) in children, compare findings in immune-suppressed vs. non immune-suppressed children, compare the percentage of children with radiographic features of pulmonary TB to the percentage of children on TB treatment and assess inter-observer agreement between 3 radiologists. METHODS: Children (0-8 years) participating in a cohort study of TB and BCG-IRIS who had an acceptable routine pre-HAART CXR were included. CXRs were independently assessed by 3 radiologists, blinded from clinical data, using a standardised assessment form. All 3 readings were used to create a majority consensus finding during the data analysis phase. RESULTS: Amongst 161 children, the median age at enrolment was 2.3 years (25% (41/161) were <1year), 54% (87/161) were on TB treatment and 65% (100/154) were immune-suppressed. The majority (71%) had an abnormal CXR finding, predominantly air space disease (42%) and parenchymal interstitial disease (21%). Of the sub-group of 112 (70%) CXRs that could be assessed for lymphadenopathy, 75(67%) had one or more features suggestive of TB (74 lymphadenopathy, 2 cavities, 18 miliary infiltration) and 65% (70/107) were immune-suppressed. Statistically significant differences between immune-suppressed and non-immune-suppressed children were noted for features of lymphadenopathy and radiographic pulmonary TB. Amongst the sub-group of 112 CXRs a high percentage 49/75 (65%) were on TB treatment, with 26/75 (35%) not on TB treatment. Inter-observer agreement between all 3 readers was fair for overall abnormal CXR findings (K=0.23), airspace disease (K=0.22), moderate for parenchymal interstitial disease (K=0.54) and slight for lymphadenopathy (K=0.05). CONCLUSION: Among children eligible to start HAART, most (71%) presented with abnormal CXR findings and the majority (67%) had one or more CXR signs suggestive of TB. Of concern was the high proportion of CXRs (30%) that were of insufficient quality to be assessed for lymphadenopathy and the poor inter-observer agreement for lymphadenopathy.

A research report submitted to the Faculty of Health Sciences, University of the Witwatersrand, in partial fulfillment of the requirements for the degree of Master of Science in Medicine in Pharmacotherapy, Johannesburg, 2011 / Long term antiretroviral therapy (ART) use is known to cause various toxic adverse effects in patients. Hepato-toxicity is one of the most significant adverse effects which have been associated with all antiretroviral therapy drugs in South Africa and worldwide.

Introduction: Stavudine is a commonly used drug in paediatric antiretroviral treatment (ART) regimens. Due to toxicity concerns, however, the drug abacavir has replaced stavudine in first-line paediatric regimens inmany countries.Wedescribe the frequency of stavudine toxicity in children receiving ART at a treatment clinic in Soweto, South Africa. Methods: Data on patient characteristics and outcomes of ART were collected from a cohort of 2222 HIV-infected children initiating ART between 2004 and 2008 when stavudine-containing regimenswere routinely recommended. At several time-points after treatment initiation, we estimate the proportion of children where an attending clinician discontinued stavudine due to lipodystrophy, pancreatitis, lactic acidosis or peripheral neuropathy. Factors associated with stavudine-related toxicities were identified. Results: At ART initiation, most children had advanced disease. The majority initiated an efavirenz/lamivudine/stavudine regimen (n¼1422), and 76% of children remained on their initial ART regimen after a median 19.9 months of ART. Replacement of stavudine due to drug toxicity occurred at a rate of 28.8 per 1000 child years on treatment (95% confidence interval¼23.6–35.2). Rates of toxicity increased with treatment duration (in their first year of ART stavudine was replaced in 0.5% of children, but after 3 years stavudine had been changed to abacavir in 12.6% of children). Toxicity was more common in older children and in girls. Lipodystrophy accounted for 87 of 96 toxic events. Conclusion: Stavudine-associated toxicity resulting in single-drug substitution was uncommon in this cohort, though its frequency increased steadily with ART duration, especially with lipodystrophy. Where drug options are limited, stavudine remains a relatively well tolerated and effective option for children.