1.1 THE EPIDEMIOLOGY OF HUMAN IMMUNODEFICIENCY VIRUS (HIV)

Transcription

1 1.0 INTRODUCTION 1.1 THE EPIDEMIOLOGY OF HUMAN IMMUNODEFICIENCY VIRUS (HIV) INFECTION The clinical condition called acquired immune deficiency syndrome (AIDS) is caused by infection with human immunodeficiency virus (HIV). AIDS was defined in 1982 by the Centres for Disease Control (CDC) (Apetrei et al., 2004). Reports of AIDS in Africa started emerging in 1984, but the epidemic was not fully appreciated until 1986 (Apetrei et al., 2004). Human immunodeficiency virus (HIV) is a retrovirus with viral genetic material stored as ribonucleic acid (RNA). Two strains of HIV are known to infect humans, HIV-1 and HIV-2 (Apetrei et al., 2004, Krogstad, 2003). HIV-1 is by far the most common and pathogenic worldwide. Statistical data released by the Joint United Nations Programme on HIV/AIDS (UNAIDS) for 2007 (figure 1.1) revealed the following: The number of people living with HIV worldwide is estimated to be 33.2 million The majority of people (61%) living with HIV in sub-saharan Africa are women More than two-thirds (68%) of all people infected with HIV live in sub-saharan Africa where more than three quarters (76%) of all AIDS-related deaths in 2007 occurred Within the region, Southern Africa is the worst-affected with national adult HIV prevalence exceeding 15% in eight Southern African countries from 2005 {UNAIDS The Joint United Nations Programme on HIV/AIDS, 2007 #340} It is 27 years since the identification of HIV. The majority of those who require antiretroviral therapy (ART) in sub-saharan Africa continue to live without access to treatment. In South Africa only 33% of those requiring ART received it in 2006 (Pembrey, 2007). These sobering statistics reflect South Africa s position as the epicentre of the global HIV-pandemic. 1

2 Figure 1.1 Adults and children estimated to be living with HIV in AIDS epidemic update [Accessed 10 December 2007] HIV INFECTION AND THE KIDNEY HIV infection and its devastating consequences have placed an enormous strain on healthcare resources and those living both with/without HIV. In addition to the broad, multi-dimensional impact of this infection, when considering kidney disease, HIV has unleashed a new burden of both acute and chronic kidney disease in South Africa. There is a wide clinical spectrum of renal disease in the course of HIV infection which includes: Acute kidney injury (AKI) Electrolyte and acid-base disturbances Intrinsic renal disease unrelated to HIV itself (e.g. diabetes mellitus and hypertension) 2

3 HIV-associated glomerulonephropathies (table 1.1): these may present as either acute on chronic or chronic renal failure and it is this group that is primarily implicated in the burden of chronic kidney disease (CKD) Side effects related to treatment of HIV which include ART and drugs used to treat complications of HIV Long term metabolic side-effects of ART Table 1.1 Spectrum of glomerular disease with HIV Glomerular pattern HIV-FGS (Focal Glomerulosclerosis) or Classic HIVAN (HIV-associated Subtypes Some have described a mixed variant of HIV-FGS in combination with a proliferative glomerulonephritis nephropathy) HIV-ICD (HIV-immune complex disease) (this group may have co-infection with Hepatitis B or C) Mesangial proliferative Membranoproliferative (type I and III) Lupus-like Exudative-Proliferative Crescentic IgA Membranous Various glomerulonephropathies (this is a heterogeneous group with different aetiologies) HIV-TTP/HUS (HIV-associated thrombotic thrombocytopaenic purpura / Minimal change Immunotactoid Amyloidosis TTP: Thrombotic Thrombocytopenic Purpura HUS: Haemolytic Uremic Syndrome haemolytic uraemic syndrome) Co-morbid disease Diabetic nephropathy Hypertensive Nephrosclerosis Auto-immune disease (e.g. Lupus nephritis) 3

4 1.2.1 ACUTE KIDNEY INJURY (AKI) IN HIV The causes of AKI in hospitalised HIV infected patients may be community or hospital acquired, the latter being 5-10 times more common than the former, with a worse outcome in hospital acquired AKI (Lamiere et al., 2006). The known causes of AKI are similar in the HIV and non-hiv group with the most common being acute tubular necrosis (ATN) secondary to sepsis, hypotension, dehydration and nephrotoxins (appendix A) (D'Agati and Appel, 1997, Lamiere et al., 2006, Roling et al., 2006, Weiner et al., 2003). Although potentially reversible with appropriate medical treatment and, if indicated, dialysis support, AKI carries a high mortality in this population (D'Agati and Appel, 1997). Biopsy studies have confirmed a wide range of aetiologies ( D'Agati and Appel, 1997, Gardenswartz et al., 1984, Sreepada Rao et al., 1987). In one study of hospitalised patients with HIV infection, AKI occurred in up to 20% of cases (Kimmel, 2000). In another study, the short-term prognosis in this group of patients showed 18% mortality at 2 months, with 80% of patients diagnosed with AIDS at the time of hospital admission (Kimmel et al., 1999). Since the advent of ART, a recent prospective study on AKI in ambulatory HIV infected outpatients with access to ART concluded that more severe immunosuppression (CD4 <200 cells/mm 3 and/or HIV RNA level >10000 copies/ml) is still the predominant risk factor for AKI (Franceschini et al., 2006) ELECTROLYTE AND ACID-BASE DISTURBANCES IN HIV Numerous electrolyte and acid-base abnormalities have been documented with HIV infection (appendix B). Abnormalities may arise either from HIV infection itself, opportunistic infections and malignancies associated with advanced immunosuppression or with medication used to treat HIV and its complications GLOMERULAR DISEASE IN HIV Initial autopsy reports of renal disease in HIV infected patients with advanced disease showed a broad spectrum of lesions that include ATN, interstitial nephritis, nephrocalcinosis, minimal change, mesangial proliferation/hyperplasia and FGS (focal glomerulosclerosis) (Bourgoignie, 1990, Hailemariam et al., 2001, Pardo et al., 1984, Pardo et al., 1987). Reports then began to appear in 4

5 the literature describing various glomerulopathies, the most common of which was FGS. The distinguishing feature about the classic glomerular lesion in HIV was collapse of the glomerular tuft, so-called collapsing glomerulopathy (Bourgoignie, 1990, D'Agati and Appel, 1997, Shahinian et al., 2000, Sreepada Rao et al., 1987). European and American-based biopsy studies reveal varying frequencies of the different histological patterns (Kimmel, 2000, Szczech, 2001, Weiner et al., 2003). HIVAN (HIV-associated nephropathy) is the most common (up to 60%), with the other lesions accounting for one quarter to one third of the total. There have been studies documenting patterns of renal involvement in South Africa. A study of inpatients biopsied (total of 99 biopsies) by the nephrology unit at Chris Baragwanath Hospital showed: HIVAN 27%: relatively less common than international data HIV-ICD (HIV- immune complex disease) 21% Membranous nephropathy 13% Other non-glomerulonephritic renal disease 10% Other glomerulonephritides 9% Post infectious glomerulonephritis 8% Mesangioproliferative glomerulonephritis 6% IgA nephropathy 5% Very little data exists on screening asymptomatic HIV infected patients for early renal disease, especially in Africa. A few studies revealed that on screening for proteinuria, additional abnormalities in particular, leucocyturia and microscopic haematuria were found the significance of which has not been investigated, nor is it understood. In a study comparing HIV infected patients in an industrialized setting (Paris) with a sub-saharan setting (Abidjan), the respective prevalence of leucocyturia and albuminuria was higher in the cohort from Abidjan (16 versus 1% for leucocyturia and 26 versus 5% for albuminuria, both achieving statistical significance). Erythrocyturia was also more prevalent in the Abidjan group, but did not achieve statistical 5

6 significance (Mortier et al., 2003). An explanation for this finding was not given, but the group from Abidjan had statistically significantly lower CD4 counts with fewer patients on ART (34% versus 66%). Leucocyturia and albuminuria may therefore reflect more advanced or untreated HIV disease, but may also reflect host genetic factors. A study on police officers in Tanzania found albuminuria, leucocyturia and erythrocyturia to be more frequent in an HIV infected cohort, when compared with a non HIV infected cohort (albuminuria 28.4 versus 16.8%; leucocyturia 14.2 versus 11.6%; erythrocyturia 1.6 versus 0.2% respectively) (Janabi et al., 2002). A urine dipstick screening study in Boston showed proteinuria in 16.6%, haematuria in 15.5% and leucocyturia in 11% of HIV infected outpatients, there was no HIV negative control group (Borkan et al., 1992).The HERS study (HIV Epidemiology Research Study) found dipstick leucocyturia in 10.1% of HIV infected and 9.9% of HIV negative women; bacteriuria (from urine culture) was seen in 6.3% of HIV infected and 6.5% of HIV negative women. Within the HIV infected group there was no difference in prevalence based on CD4 count or whether co-trimoxazole was taken (Smith et al., 1995). In a Kenyan study, 216 antiretroviral-naïve patients with an average CD4 count of 383 cells/mm 3 were screened for renal disease (Wools-Kaloustian et al., 2007). Those with known risk factors for renal disease were excluded (Wools-Kaloustian et al., 2007). 25% of the remaining patients had creatinine clearance (Cr Cl) < 90ml/min (normal 90ml/min), 2% had Cr Cl <60ml/min and 8% had proteinuria of >1gram/day. Although not statistically significant, there was a trend to more severe renal insufficiency and heavier proteinuria in those with a CD4 count <200 cells/mm 3 (Wools- Kaloustian et al., 2007). A Ugandan study involved 229 patients with World Health Organisation (WHO) clinical stage 3 disease. CD4 counts were not reported. They showed a high prevalence of renal disease, with a Cr Cl of <80ml/min in 48.5% of patients. Screening showed that 20% had proteinuria >100mg/dl and leucocyturia was present in 132/299 patients (44.1%) but confirmatory urine cultures were not done (Andia et al., 2005). In the above studies, no renal biopsies were performed, nor was there further investigation into the causes of leucocyturia and microscopic haematuria. 6

7 A screening study was conducted at King Edward Hospital in Durban on HIV infected patients looking for proteinuria, including microalbuminuria (Han et al., 2006). This is one of very few studies that have evaluated the prevalence and potential significance of microalbuminuria in the HIV infected population, as a marker of early renal disease. In a study by Luke et al. 72 ambulatory HIV infected outpatients were prospectively screened for microalbuminuria (3 asymptomatic, 32 AIDSrelated complex, 37 AIDS). 14 patients (19.4%) had microalbuminuria. Microalbumin levels were not correlated with race, sex, risk factors of AIDS, disease history, or concurrent drug therapy. In contrast, urinary microalbumin levels were correlated with CD 4 T-cell and white blood cell (WBC counts, tumour necrosis factor alpha (TNF-) and beta 2-microglobulin levels, suggesting an association between AIDS progression and microalbuminuria. It was speculated by the authors whether monitoring of urinary microalbumin levels may identify those patients susceptible to the development of nephrotic syndrome, but this was not explored with further follow up or renal biopsy in the microalbuminuric group (Luke et al., 1992). In the Durban study 615 patients were screened. Two percent were hospitalised patients; 98% were asymptomatic (from a renal perspective) outpatients attending an HIV clinic. Black patients accounted for 98% of those screened and 30 renal biopsies were done, of which 7 were for persistent microalbuminuria and 23 for overt proteinuria (Han et al., 2006). Histology showed the following: HIVAN 25/30 (83.3%); HIVAN + membranous glomerulonephritis (GN) 4/30 (13.3%) - included in initial HIVAN group; membranoproliferative GN 2/30 (6.7%); interstitial nephritis 3/30 (10%). The data from the Chris Hani Baragwanath and King Edward hospital cohorts are contradictory. Possible explanations may be population demographics - the studies were done in different provinces of South Africa which serve varying black ethnic groups. Another explanation may be patient selection. The Chris Hani cohort comprised of symptomatic in patients with advanced renal failure, many of whom required dialysis support. The King Edward cohort comprised mostly outpatients with relatively normal renal function, presumed to be at earlier stages of renal disease. The spectrum of HIV-related glomerular disease has evolved, with HIVAN being the most frequently described and to date, the lesion most researched regarding pathogenesis. 7

8 Evidence from small studies and case reports have shown that pharmacological interventions for HIVAN (less so for other glomerular lesions) delay the progression or may reverse even renal disease in isolated cases (Eustace et al., 2000, Han et al., 2006, Pope et al., 2005, Winston et al., 2000). Due to the paucity of data regarding the non-hivan lesions, the primary focus of the remaining introduction will be on HIVAN. 1.3 HIV- ASSOCIATED NEPHROPATHY (HIVAN) INTRODUCTION An association between HIV and renal disease was first reported in 1984 describing HIV infected individuals with nephrotic range proteinuria and progression to end stage renal disease (ESRD) within 8-16 weeks (Mazbar et al., 1990, Pardo et al., 1984, Sreepada Rao et al., 1984). Mortality in these patients approached 100% within six months of diagnosis. The existence of a specific HIVassociated nephropathy (HIVAN) has been confirmed as a distinct clinicopathological entity (Cohen and Nast, 1988, Klotman, 1999, Pardo et al., 1987, Strauss et al., 1989). HIVAN was initially thought to be associated with advanced immunosuppression but it was later recognised that the lesion can occur at any stage of HIV infection, even prior to antibody seroconversion (Levin et al., 2001). Most patients with HIVAN present late with advanced renal failure. This late detection of HIVAN could be due to a lack of screening for proteinuria and/ or renal dysfunction. The relative absence of overt symptoms and signs such as peripheral oedema and hypertension in those with HIVAN may also delay diagnosis. Outcome of patients with HIVAN has been correlated with the clinical stage of their disease, suggesting that survival improves with earlier detection (Winston and Klotman, 1996). There is an increased relative risk of for overall mortality with proteinuria, after correcting for other risk factors (Gardner et al., 2000, Gardner et al., 2003). In one study 77% of renal abnormalities developed with CD4 counts above 200 cells/mm 3 (Gardner et al., 2003). This was also seen in the study from Durban where the mean CD4 count of those with biopsy-proven HIVAN was 232 cells/mm 3 (Han et al., 2006). HIVAN was not previously an AIDS-defining illness, but criteria have recently been revised by the WHO and symptomatic HIVAN is considered clinical stage 4 disease (World Health Organization, 2007). 8

9 1.3.2 EPIDEMIOLOGY AND RACIAL PREDILECTION OF HIVAN There is a marked racial predilection for the development of HIVAN - over 90% of patients are black (Bourgoignie, 1990, Freedman et al., 1999, Kopp and Winkler, 2003, Laradi et al., 1998, Sreepada Rao, 2001, Winston and Klotman, 1996). This has been confirmed in paediatric studies (Strauss et al., 1989). Studies from African countries regarding the susceptibility of Africans to HIVAN are scanty. The reasons for the racial predilection of HIVAN are unexplained. Some have postulated a genetic predisposition, but candidate genes have not been identified (Bourgoignie, 1990, Kopp and Winkler, 2003). One study has shown that there appears to be a strong familial clustering of ESRD among blacks commencing renal replacement therapy due to HIVAN, independent of HIV status (Freedman et al., 1999). In most cases, the affected relatives had ESRD due to hypertension or diabetes. HIVAN also appears to follow a more severe clinical course in black patients (Cohen, 1990). Studies in black patients have shown a higher prevalence of both severe glomerular lesions (FGS) and nephrotic range proteinuria with renal dysfunction in the presence of normo/hypotension (Cohen, 1990, Laradi et al., 1998, Sreepada Rao, 2001). In the 1980s HIVAN was an uncommon cause of ESRD in the USA (Ross and Klotman, 2002). This was in part, due to the high mortality in this group and the absence of ART. With improved survival after the introduction of ART, HIVAN became the most rapidly increasing cause of ESRD, being the third leading cause of ESRD after diabetes and hypertension in African Americans aged years in the USA by 1990 (Ross and Klotman, 2002). While the rate of new cases of ESRD due to AIDS has fallen slightly since the beginning of the decade: reaching 2.7 per million population in the period, prevalence in the USA has grown steadily, reaching 8.9 in , indicating that people are living longer with the disease (USRDS United States Renal Data System, 2007). In 2004, HIVAN was the seventh leading cause of ESRD in the African American population (USRDS United States Renal Data System, 2007). Statistics in the USA estimate the incidence of HIVAN to be % (Ross and Klotman, 2002). If this were to be extrapolated to Sub-Saharan Africa, with an estimated 22.5 million infected with HIV, between million people would be predicted to have HIVAN. With the advent of unrestricted access to ART, the epidemiological 9

10 pattern of HIVAN that evolved in the USA over the last 14 years may predict what will happen in sub-saharan Africa. This presents a potentially unprecedented burden of chronic kidney disease, given the current resources available for managing renal disease in this region HISTOPATHOLOGY OF HIVAN HIVAN is characterised by a specific constellation of pathological findings on biopsy that involve glomerular, tubular and interstitial compartments of the kidney (figure 1.2) (Cohen and Nast, 1988, Ross et al., 2001, Sreepada Rao, 2001): FGS with collapse of the glomerular tuft Tubular dilatation, flattening of tubular epithelial cells, microcysts Lymphocytic infiltrates of interstitium with patchy oedema and /or interstitial fibrosis Endothelial tubuloreticular inclusions (on electron microscopy) Increased mesangial matrix, hyperplasia of mesangial cells may be seen but whether this is part of the spectrum of HIVAN is debatable Figure 1.2 Typical histopathologic findings of HIVAN (Ross et al., 2001) Periodic acid-schiff stain: focal segmental glomerulosclerosis (arrowhead); collapse of glomerular tuft (arrow), tubular microcystic disease (*), interstitial lymphocytic infiltration and interstitial fibrosis. Magnification x200 10

11 1.3.4 AETIOPATHOGENESIS OF HIVAN HIVAN is just one of many pathological manifestations of HIV-related renal disease. The protean manifestations of kidney disease may be the result of a complex interplay between the particular pheno- and/or genotypic variants of HIV, the genetic make-up of the host and environmental factors. The evidence in support of potential mechanisms for the development of HIVAN has been summarised as follows (Husain et al., 2002): Cytopathic effects of HIV gene products Apoptosis mediated by HIV infection Elaboration of chemokines and cytokines as a result of viral or host protein synthesis in patients with a genetic susceptibility to nephropathy Subversion of the host metabolic and synthetic machinery by the virus may allow for rapidly progressive disease 1.4 DOES HIV INFECT THE KIDNEY? The aetiological link between HIV and HIVAN has been explored in animal models, in-vitro studies and human biopsy studies. HIV transgenic mouse models showed that mice develop proteinuria, progressive renal failure and histological disease identical to HIVAN (Bruggeman et al., 1997 Kopp et al., 1992). Reciprocal transplantation studies confirm that the HIVAN phenotype in these mice is dependent on HIV gene expression in the renal epithelium (Bruggeman et al., 1997). In 2000 it was proven on human biopsy studies that HIV gains entry into renal epithelial cells and generates full length mrna transcripts (Bruggeman et al., 2000). HIV infection was demonstrated in several cell groups including visceral and parietal epithelial cells, and tubules (Bruggeman et al., 2000, Dickie et al., 1991). This pattern of infection correlates with the histological pattern of disease in HIVAN (Bruggeman et al., 2000, Dickie et al., 1991). It is now accepted that an aetiological link between direct HIV infection of renal tissue compartments and the histological lesion of HIVAN has been proven. Antiretroviral therapy has been demonstrated to decrease viral replication, improve immune function, delay progression of disease and prolong survival (Furtado et al., 1999). 11

12 In spite of ART a long-lived reservoir of infectious, replication-competent HIV persists in resting memory CD4+ T cells and possibly macrophages (Furtado et al., 1999). This makes it impossible, on current therapy, to eradicate HIV completely from the host (Furtado et al., 1999). It has been proposed that other anatomical sites act as reservoirs for HIV, including the kidney (Ross and Klotman, 2004). The implications of this with respect to adequacy of penetration of ART into the kidney in the treatment of HIVAN, and/or the generation or harbouring of mutant strains of HIV within the kidney are unknown. 1.5 HOW DOES HIV INFECT RENAL TISSUE? In addition to HIV and its CD4 lymphocyte co-receptor, various lymphocyte chemokine receptors have been implicated in the pathogenesis of HIV infection. Studies in peripheral blood lymphocytes showed that chemokine receptors CXCR4 and CCR5 are essential co-receptors for the penetration of HIV into susceptible lymphoid cells (Eitner et al., 2000). With regard to penetration of renal tissue, early studies confirmed that renal epithelium does not express CCR5 in human and subhuman primates infected with HIV (Eitner et al., 1998). A more recent study showed that cultured human podocytes express CCR4, CCR8, CCR9 CCR10, CXCR1, CXCR3, CXCR4 and CXCR5 co-receptors, which potentially explains how HIV may gain access to the podocyte (Huber et al., 2002). Expression of chemokine receptors may be altered in various disease states for example, in membranous nephropathy, the expression of CXCR1, CXCR3 and CXCR5 was upregulated (Huber et al., 2002). These studies have not been performed in HIV infection. 1.6 THE EFFECT OF HIV INFECTION ON THE KIDNEY GLOMERULAR TISSUE (mesangial cells, podocytes, endothelial cells) Mesangial cells Increased mesangial cell apoptosis has been demonstrated in HIV transgenic mice when compared with controls, an effect mediated by increased levels of tumour necrosis factor - alpha (TNF-α). HIV glycoprotein (gp)-120 has been shown to stimulate mesangial cell proliferation (Singhal et al., 1997, Singhal et al., 1998). HIV infection of mesangial cell cultures stimulates mesangial cells to 12

13 express platelet-derived growth factor (PDGF) and transforming growth factor-beta (TGF-β), two crucial mediators in the pathogenesis of glomerulosclerosis (Conaldi et al., 2000). HIV infected mesangial cells produce increased levels of cytokines, namely interleukin-6 (IL-6), interleukin-8 (IL- 8) and TNF-α (Conaldi et al., 2000) Podocytes and endothelial cells Numerous abnormalities have been shown in podocytes in HIV infection such as loss of contact inhibition and proliferation (Husain et al., 2002, Schwartz et al., 2001). Podocytes, once terminally differentiated, express markers like synaptopodin, P-cadherin, podocalyxin and Wilm s Tumour-1 (WT-1) (Barisoni et al., 1999). In human biopsy specimens from those with collapsing glomerulopathy (HIVAN), podocytes lose markers of terminal differentiation and re-enter the cell cycle causing proliferation, possibly due to loss of expression of WT-1, a cell cycle repressor (Barisoni et al., 1999). Collapse of the glomerular tuft is thought to be due to loss of vascular endothelial growth factor (VEGF), which is primarily secreted by the podocytes (Barisoni et al., 1999) TUBULOINTERSTITIUM Tubulointerstitial pathology in HIVAN is common. Tubules undergo characteristic morphological changes that include flattening and simplification of tubular epithelium, cystic dilatation and proteinaceous cast formation (D'Agati et al., 1989). Interstitial inflammation is characterised by the presence of infiltrating lymphocytes and plasma cells, interstitial oedema and fibrosis (D'Agati et al., 1989). Infection of proximal tubular epithelial cells (PTEC) with HIV induces apoptosis via activation of caspase proteins and up-regulation of expression of fas-ligand (Conaldi et al., 1998). HIV-gp 120 has been shown to induce apoptosis of tubular epithelial cells through activation of the stress and apoptosis-related pathway mitogen-activated protein kinases. Fibroblast growth factor-2 (FGF-2) has been shown to increase renal recruitment and attachment of HIV infected peripheral blood mononuclear cells to renal tubular epithelial cells (Tang et al., 2005).These changes induced tubulointerstitial injury which reversed when levels of FGF-2 were normalised (Tang et al., 2005). 13

14 1.7 PHARMACOLOGICAL TREATMENT OF HIV-ASSOCIATED RENAL DISEASE There have been no prospective randomised controlled studies with any form of therapy for HIVAN to date, let alone other forms of HIV-associated related renal disease. Therapies used in the treatment of HIVAN include corticosteroids, angiotensin converting enzyme inhibitors (ACE-I), cyclosporine and ART. Treatment options for patients who have reached ESRD include haemodialysis, peritoneal dialysis and renal transplantation. Corticosteroids Patients with HIVAN treated with prednisone experienced an improvement in renal function and reduction of proteinuria but complications such as relapse after steroid withdrawal, opportunistic infections, psychosis and gastrointestinal bleeding were relatively common ( Eustace et al., 2000, Smith et al., 1996). In addition, there were no controls in the analysis for the beneficial effects of ACE-I and ART (Eustace et al., 2000). An isolated case report showed response to corticosteroids and relapse after steroid withdrawal in a patient with HIVAN on zidovudine (AZT) monotherapy (Watterson et al., 1997). Because these studies have no long term follow up, are small in size and not randomised, no clear conclusions can be made regarding their use in HIVAN. Angiotensin Converting Enzyme Inhibitors (ACE-I) Captopril has been shown to delay progression of renal failure in a retrospective cohort of nine patients with HIVAN, when compared with controls (Kimmel et al., 1996). Fosinopril was shown, in patients with biopsy-proven HIVAN who had both nephrotic and sub-nephrotic range proteinuria, to stabilise renal function and proteinuria (Burns et al., 1997). This was compared to a control group who had declined the option of ACE-I and within 24 weeks of follow up had significant deterioration in renal function and proteinuria (Burns et al., 1997). Both studies were conducted prior to availability of ART. In a larger prospective study to determine the long-term effects of ACE-I on renal survival in HIVAN patients, those on ACE-I showed a median renal survival of days, with one patient progressing to ESRD, compared to a median renal survival of days in 14

15 untreated patients who all developed ESRD (Wei et al., 2003). The potentially beneficial effects of ACE-I may be related to improved renal haemodynamic, reduced proteinuria or cytokine modulation (Gupta et al., 2005). The effects of angiotensin-2 receptor blockers (ARB) in the treatment of HIVAN is unknown, just as the effects of ACE-I or ARB in the treatment of HIV-associated renal disease other than HIVAN are unknown. Cyclosporine The effectiveness of cyclosporine in inducing remission of proteinuria was reported in children with HIVAN (Ingulli et al., 1991). There are no studies evaluating the role of cyclosporine in adults. Highly active anti-retroviral therapy Because HIV itself appears to be the cause of HIVAN (and possibly other manifestations of HIVassociated renal disease), ART appears to be a logical choice in disease management. In one study with AZT monotherapy for HIVAN, at 20 months follow up, those on treatment had stable renal function and all those who were non-compliant progressed to ESRD (Ifudu et al., 1995). Another study with AZT monotherapy in 11 patients showed a beneficial effect on renal function only if the AZT was initiated before advanced renal failure was present (Michel et al., 1992). There appears to be a more beneficial effect of ART over AZT monotherapy in the management of HIVAN (Betjes and Verhagen, 2002, Chemlal et al., 2000, Kirchner, 2002, Wali et al., 1998, Winston et al., 2001). Regimens containing protease inhibitors were associated with significant slowing of the decline in creatinine clearance (Szczech et al., 2002). In addition to being effective in treating established HIVAN, ART may decrease the actual incidence of de-novo HIVAN (Ahuja et al., 1999, Lucas et al., 2004). 15

16 1.8 POSSIBLE IMPLICATIONS OF UNOPPOSED HIV REPLICATION ON KIDNEY DISEASE The SMART study (Strategies for Management of Antiretroviral Therapy) randomized HIV infected patients on stable ART with CD4 counts >350 cells/mm 3 to two treatment arms, either virological suppression (VS) or regularly interrupted, CD4-guided ART. The second arm, the drug conservation (DC) group, interrupted treatment when CD4 counts were >350 cells/mm 3 and restarted ART when CD4 counts dropped below 250 cells/mm 3. The study was prematurely terminated because the DC group showed a higher risk of severe AIDS-related illness and death, but also a three to four times higher risk of severe non-aids-related disease, in particular - hepatic, renal and cardiovascular events. It is thought that part of the increased risk of non-aids-related disease may be related to ongoing viral replication, with consequent inflammatory effects on endothelial function. Data from SMART has provided a strong scientific rationale for studying the effect of early ART on mortality and serious morbidity from AIDS and non-aids conditions, of which HIV-related renal disease is one (Neaton, 2007). 1.9 HYPOTHESIS FOR THIS RESEARCH REPORT Proteinuria (and persistent microalbuminuria) may be an early manifestation of renal disease in HIV infected individuals, requiring histological diagnosis by renal biopsy. Therapy of confirmed HIVassociated renal disease with ART and ACE-I (where indicated) may result in regression of renal disease Aims of the research report: Assess the prevalence of proteinuria, in particular microalbuminuria, in an ambulant outpatient HIV positive population in an attempt to detect early renal disease Assess those with persistent proteinuria, and where indicated, to determine renal histology In those with HIV-associated renal disease on renal biopsy, to commence ART and ACE-I (where indicated) and to monitor their response to treatment over 12 months using clinical and laboratory parameters in combination with renal biopsy. 16