HIV-related nephropathy: A South African perspective
2006; Elsevier BV; Volume: 69; Issue: 10 Linguagem: Inglês
10.1038/sj.ki.5000351
ISSN1523-1755
AutoresTrevor Gerntholtz, Stewart Goetsch, Ivor Katz,
Tópico(s)HIV-related health complications and treatments
ResumoHuman immunodeficiency virus (HIV)-associated nephropathy (HIVAN) is an important cause of renal failure in those of African origin. A number of other kidney diseases occur in HIV-positive patients. We conducted a retrospective review of renal biopsies in HIV-positive Black African patients to determine the prevalence of both 'classic HIVAN' and non-HIVAN pathologies in this group. Clinical and laboratory data from HIV-positive patients who underwent renal biopsy from 1st January 2003 to 31st December 2004 were collected. Similar information on HIV-negative patients biopsied during the same period was also recorded by way of comparison to try and assess the influence of the virus on renal histologic patterns. HIV-positive group – 99 biopsies were suitable for study. The main histologic categories were 'classic HIVAN' (27%) and HIV immune complex kidney disease ('HIVICK') (21%). The subepithelial immune deposits in 'HIVICK' induced a newly described 'ball-in-cup' basement membrane reaction. Other glomerulonephritides included membranous, post-infectious disease, mesangial hyperplasia, and immunoglobulin A nephropathy. Overlapping clinical presentations prevented pre-biopsy histologic predictions. HIV-negative group – There were no examples of collapsing focal segmental glomerulosclerosis or nonspecific immune complex disease, but increased numbers of minimal change and membranoproliferative disease. 'Classic HIVAN' accounted for less than a third of the nephropathies occurring in HIV-positive Black South Africans. 'HIVICK' is another important cause of chronic kidney disease in this group. Future research is needed into the earlier detection and treatment of these diseases, which have a high mortality in our context. Human immunodeficiency virus (HIV)-associated nephropathy (HIVAN) is an important cause of renal failure in those of African origin. A number of other kidney diseases occur in HIV-positive patients. We conducted a retrospective review of renal biopsies in HIV-positive Black African patients to determine the prevalence of both 'classic HIVAN' and non-HIVAN pathologies in this group. Clinical and laboratory data from HIV-positive patients who underwent renal biopsy from 1st January 2003 to 31st December 2004 were collected. Similar information on HIV-negative patients biopsied during the same period was also recorded by way of comparison to try and assess the influence of the virus on renal histologic patterns. HIV-positive group – 99 biopsies were suitable for study. The main histologic categories were 'classic HIVAN' (27%) and HIV immune complex kidney disease ('HIVICK') (21%). The subepithelial immune deposits in 'HIVICK' induced a newly described 'ball-in-cup' basement membrane reaction. Other glomerulonephritides included membranous, post-infectious disease, mesangial hyperplasia, and immunoglobulin A nephropathy. Overlapping clinical presentations prevented pre-biopsy histologic predictions. HIV-negative group – There were no examples of collapsing focal segmental glomerulosclerosis or nonspecific immune complex disease, but increased numbers of minimal change and membranoproliferative disease. 'Classic HIVAN' accounted for less than a third of the nephropathies occurring in HIV-positive Black South Africans. 'HIVICK' is another important cause of chronic kidney disease in this group. Future research is needed into the earlier detection and treatment of these diseases, which have a high mortality in our context. Since its first description over 20 years ago,1.Rao T.K. Filippone E.J. Nicastri A.D. et al.Associated focal and segmental glomerulosclerosis in the acquired immmunodeficiency syndrome.N Engl J Med. 1984; 310: 669-673Crossref PubMed Scopus (478) Google Scholar, 2.Pardo V. Aldana M. Colton R.M. et al.Glomerular lesions: the acquired immunodeficiency syndrome.Ann Intern Med. 1984; 101: 429-434Crossref PubMed Scopus (240) Google Scholar human immunodeficiency virus (HIV)-associated nephropathy (HIVAN) has established itself as an important cause of kidney failure. In fact, HIV-related kidney disease is now the third leading cause of end-stage renal failure among African Americans aged 20–64 in the United States,3.US Renal Data System USRDS 1999 Annual Data Report. National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD1999Google Scholar occurring almost exclusively in this group and pointing towards a possible genetic influence. Histologically, the disease is characterized by a focal segmental sclerosing glomerulopathy, with or without a collapsing component and prominent tubulointerstitial changes, including cystically dilated tubular profiles. Similar histologic findings have been reported in association with intravenous heroin use4.D'Agati V. Sun J.I. Carbone L. et al.Pathology of HIV-associated nephropathy: a detailed morphologic and comparative study.Kidney Int. 1983; 35: 1358-1370Abstract Full Text PDF Scopus (329) Google Scholar and in patients treated with bisphosphante pamidronate. It is now felt that HIVAN is caused by direct viral infection of renal cells, particularly the visceral epithelial cells of the glomerulus and the tubular epithelial cells. Besides HIVAN, it is known that a number of other renal diseases occur in HIV-infected patients. A recent report showed 'non-HIVAN' renal pathology is occurring in over half (47) of the 89 biopsies.5.Szczech L.A. et al.The clinical epidemiology and course of the spectrum of renal diseases associated with HIV infection.Kidney Int. 2004; 66: 1145-1152Abstract Full Text Full Text PDF PubMed Scopus (271) Google Scholar The etiological role of the virus in these pathologies is not known. It has not been well documented to what extent 'non-HIVAN' pathologies affect Black patients. Indeed, some authors consider these pathologies to be more common in Caucasian patients.6.Nochy D. et al.Renal disease associated with HIV infection: a multicentric study of 60 patients from Paris hospitals.Nephrol Dial Transplant. 1993; 8: 11-19PubMed Google Scholar, 7.Connolly J.O. Weston C.E. Hendry B.M. HIV-associated renal disease in London hospitals.Q J Med. 1995; 88: 627-634Google Scholar Although the vast majority of infected patients reside in sub-Saharan Africa (roughly 75%), there is a paucity of published data from this continent. This is particularly true for HIV-related kidney disease. For this reason, we undertook a retrospective biopsy review of our HIV-positive patients to determine the prevalence of both HIVAN and 'non-HIVAN' renal pathologies in Africans and to study their clinical presentations. We also wanted to compare our HIV-positive patients with our HIV-negative patients to try to understand the influence of the virus on the prevalent kidney disease patterns in this group. During the 2-year period, we obtained 104 biopsies from 104 HIV-positive patients. Five of these were considered inadequate to make a diagnosis. The remaining 99 biopsies were from Black African patients. Their clinical characteristics and histologic patterns can be seen in Table 1. The majority were young males. The mortality rate was high with 44 of the patients dead at the end of the study period and 23 being lost to follow-up.Table 1Clinical characteristics of all HIV-positive patients who were biopsiedMean (s.d.)HIVANICDMembranousPIGNMes prolifIgAOther GN'sOtherNumber272113865910Sex (females)138713224Age (years)35 (9)35 (9)34 (8)36 (8)34 (11)32 (10)27 (6)37 (9)Bpsys131 (30)132 (30)143 (34)136 (33)113 (16)121 (29)133 (17)152 (50)Bpdias80 (17)80 (19)87 (23)82 (27)67 (7)81 (17)78 (8)88 (44)Serum creatinine (μmol/l)a770 (506)493 (387)713 (707)319 (478)273 (214)507 (529)343 (468)1317 (87)CD4 counta149 (182)118 (85)327 (269)296 (117)157 (156)216 (192)358 (193)221 (132)Proteinuria (g/day)b11.8 (10.34.3 (5.4)6.8 (5)3.5 (1.7)6.9 (6.6)6 (6)6.4 (5.5)3.0 (1.0)Serum albumin (g/dl)b21 (9)29 (8)27 (9)23 (6)19 (9)27 (8)24 (9)31 (4)Serum cholesterol (mmol/l)5.7 (2.8)4.9 (2)6.0 (2.4)3.8 (0.8)5.9 (2.4)3.27.1 (3)4.7 (1.9)Hepatitis B surface antigencPositive results out of those whose relevant information was available (i.e. some data missing in records).4 out of 231 out of 182 out of 121 out of 71 out of 40 out of 52 out of 70 out of 8Hepatitis C antibodiescPositive results out of those whose relevant information was available (i.e. some data missing in records).0 out of 230 out of 181 out of 120 out of 70 out of 40 out of 50 out of 70 out of 8Anti nuclear factorcPositive results out of those whose relevant information was available (i.e. some data missing in records).0 out of 170 out of 131 out of 100 out of 41 out of 30 out of 32 out of 70 out of 4Treatment: ACE inhibitors16171244276 Prednisone42402030 Simvastatin95612043 HAART34000001Follow-up (weeks)17 (23)14 (16)24 (29)28 (29)22 (30)17 (16.8)24 (16)23 (30)Lost to follow-up93223013Alive5 out of 189 out of 184 out of 113 out of 61 out of 31 out of 56 out of 83 out of 7ACE, angiotensin-converting enzyme inhibitor; HAART, highly active antiretroviral therapy; HIVAN, HIV-associated nephropathy; ICD, immune complex disease; PIGN, post-infectious glomerulonephritis; Mes prolif, immune complex-negative mesangial proliferation; IgA, IgA nephropathy; other GN'S, other glomerulonephritides; other, other renal diseases.P=0.006; bP=0.003; P-values only shown where <0.05.c Positive results out of those whose relevant information was available (i.e. some data missing in records). Open table in a new tab ACE, angiotensin-converting enzyme inhibitor; HAART, highly active antiretroviral therapy; HIVAN, HIV-associated nephropathy; ICD, immune complex disease; PIGN, post-infectious glomerulonephritis; Mes prolif, immune complex-negative mesangial proliferation; IgA, IgA nephropathy; other GN'S, other glomerulonephritides; other, other renal diseases. P=0.006; bP=0.003; P-values only shown where <0.05. A breakdown of the various histological categories can be seen in the Figure 1. We subdivided our patients according to their histological patterns using the following guidelines: Twenty-seven patients (27%) were placed in this category. Biopsies showed focal segmental sclerosing glomerulopathy, with or without an associated collapsing component. Mesangial hyperplasia, with a variable increase in the mesangial cellularity and/or mesangial matrix tissue deposition was usual. There was overlying visceral epithelial cell prominence, often associated with hyaline droplet change within Bowman's space. The interstitium showed microcystic change of the proximal convoluted tubular profiles, with varying degrees of interstitial fibrosis, tubular atrophy, and interstitial inflammation. The blood vessels were either normal or showed segmental arteriolar hyalinosis with fibro-elastic hyperplasia of the interlobular arteries. Immunofluorescence microscopy was either negative or showed nonspecific trapping of immunoglobulin (Ig)M with or without complement component C3. These patients tended to present with substantial proteinuria (mean of 11.8 g/day), advanced renal failure (mean creatinine 770 μmol/l), and low CD4 counts (mean 149). However, there was a wide variation among the above variables, with proteinuria ranging from 1.7 to 40 g/day, creatinines from 39 to 1971 μmol/l, and CD4 counts from 1 to 651 cells/cm3. In the absence of antiretrovirals or renal replacement therapy, the majority was dead by the end of the study (13 out of the 18 who were not lost to follow-up). Twenty-one patients (21%) were placed in this category. Glomerular changes were predominantly mesangial, ranging from segmental to global with mild to moderate hyperplasia. Varying numbers of immune deposits were discernible on light microscopy within the mesangial and paramesangial regions (see Figures 2 and 3). Numerous biopsies also showed varying numbers of large subepithelial deposits associated with a basement membrane reaction and resulting in a 'ball-in-cup' architectural pattern (see Figures 4 and 5). Scattered basement membrane holes were visible. In some, marked crescent formation was evident. We felt the above morphologic pattern did not fit with any well-described classic immune-complex-mediated glomerulonephritis, but had overlap features between incompletely expressed 'post-infectious' and membranous glomerulopathies. Immunofluorescence microscopy confirmed the immune deposits, with varying staining patterns for IgG, IgM, IgA, and complement component C3. Several of the biopsies showed a 'full-house' 'lupus-like' immunofluorescence pattern. The interstitium invariably had some degree of inflammation, fibrosis, and tubular atrophy. Furthermore, some of the biopsies showed microcystic change of the tubular profiles. The blood vessels were usually normal, but occasionally had combined features of a thrombotic microangiopathic process, with areas of transmural fibrinoid change, red cell extravasation, intimal basophilic mucoid degeneration, and endothelial cell swelling. The clinical profile of patients with 'HIV immune complex kidney (HIVICK)' was similar to those with 'classic HIVAN', although the former tended to have less proteinuria (4.3 vs 11.8 g/day) together with better creatinines (493 vs 770 μmol/l) and better albumins (29 vs 21 g/dl) on presentation. This was in spite of slightly worse CD4 counts (118 vs 149 cells/cm3). Of the 18 patients not lost to follow-up, nine survived. Thirteen patients (13%) were grouped in this category. Glomeruli showed thickening and rigidity of the peripheral capillary walls and glomerular basement membranes. Numerous subepithelial deposits were evident, inducing a typical 'spike and chain' basement membrane reaction. In addition to pure membranous-type changes, some biopsies had features of 'classic HIVAN' with associated microcystic change of the tubular profiles. These patients were classified as having membranous nephropathy. Occasional biopsies showed additional segmental sclerosing glomerular lesions suggestive of 'classic HIVAN'. Clinically, patients tended to have heavy proteinuria with marked renal impairment, although there was a wide range of variation. Two patients were Hepatitis B positive. Seven out of the 11 patients who were not lost to follow-up died. Eight patients (8%) belonged to this group. Biopsies displayed a diffuse proliferative glomerulonephritis, with endocapillary features and subepithelial 'hump'-type deposits. These patients had marked renal impairment with marked proteinuria, but less severe than those with 'classic HIVAN' or 'HIVICK'. Six patients (6%) had pure mesangial changes consisting of an increase in cellularity and matrix tissue deposition. The immunofluorescence findings were negative. The interstitium showed varying degrees of inflammation and fibrosis, with occasional cystic dilatation of tubules. This group had low CD4 counts (average 157 cells/cm3) together with high creatinines (273 μmol/l) and heavy proteinuria (6.9 g/day). These patients were considered to represent the forme-fruste of an evolving HIVAN, caught before the development of segmental sclerosing lesions. The five patients (5%) in this group all had mesangial hyperplasia with immunofluorescence microscopy showing dominant IgA deposition. These patients had low CD4 counts together with marked renal dysfunction (creatinine 507 μmol/l) and heavy proteinuria (6 g/day). 1.Three patients had 'non-HIVAN' focal segmental glomerulosclerosis with hyalinosis (FSH). Although these patients had focal segmental sclerosing glomerular lesions, they were non-collapsing and there were no typical tubulointerstitial changes of 'classic HIVAN'.2.Two patients had membranoproliferative glomerulonephritis with 'tram tracking' of glomerular basement membranes on silver staining and subendothelial immune deposits. There was diffuse and global glomerular hypercellularity, often accompanied by lobular accentuation of the glomerulus. Both were Hepatitis C antibody negative3.Two patients with minimal change disease had normal histology on light microscopy and immunofluorescence.4.One patient with lupus nephritis, WHO Type IV (diffuse nephritis), was known to have lupus before biopsy.5.One patient with chronic glomerulonephritis could not be typified further because of diffuse global glomerulosclerosis. 1.Three patients had malignant hypertension with ischemic glomeruli, collapsed peripheral capillary loops, and wrinkled basement membranes. Arterioles showed hyperplasia with fibrinoid change. The arteries showed a hyperplastic arteritis, with 'onion-skin' vessel wall thickening. Accompanying high blood pressures led us to favor a diagnosis of malignant hypertension over an HIV-associated thrombotic microangiopathy. The interstitium showed extensive fibrosis, with tubular atrophy.2.Three patients had normal glomeruli but marked tubulointerstitial nephritis, with features of both activity and chronicity. No cystically dilated tubular profiles were discernible.3.Two patients had features of acute renal tubular necrosis with normal glomeruli4.Two patients showed the features of a diabetic nephropathy. The diagnoses of diabetes were known before biopsy. The histologic diagnoses in this group can be seen in the Figure 5. The HIV-positive and HIV-negative groups were similar in terms of age, gender, and racial make-up using the Mann–Whitney U-test. There were no examples of a collapsing 'HIVAN'-like nephropathy in the HIV-negative group. There was a high prevalence (17%) of patients with features of an FSH/focal glomerulosclerosis. This was not associated with the collapsing form of FSH, nor was there microcystic dilatation of the tubular profiles, as can be seen with HIVAN. There were statistically significant increased numbers of minimal change disease (P=0.0006) and membranoproliferative glomerulonephritides (P=0.0016) in the HIV-negative group. There were no examples of immune-complex-mediated glomerulopathies in that they did not fit a classic morphologic reaction pattern. With regards the other histologic subtypes (membranous, 'post-infectious', IgA nephropathy, and 'other' non-glomerulonephritic lesions), there was no statistical difference between the HIV-positive and HIV-negative groups. Although HIVAN is thought to be the most common form of chronic kidney disease in those of African descent who are HIV positive, this entity accounts for only about half of all renal pathology. HIVAN is most likely owing to direct viral infiltration of renal cells,8.Bruggeman L.A. et al.Nephropathy in human immunodeficiency virus-1 transgenic mice is due to renal transgene expression.J Clin Invest. 1997; 100: 84-92Crossref PubMed Scopus (217) Google Scholar, 9.Marras D. et al.Replication and compartmentalization of HIV-1 in kidney epithelium of patients with HIV-associated nephropathy.Nat Med. 2002; 8: 522-526Crossref PubMed Scopus (246) Google Scholar, 10.Kimmel P.L. et al.HIV-associated immune-mediated renal disease.Kidney Int. 1993; 44: 1327-1340Abstract Full Text PDF PubMed Scopus (134) Google Scholar but the role of the virus in non-HIVAN HIV-related nephropathy remains one of conjecture. In our biopsy series of 99 in-hospital HIV-positive patients, we have described a broad spectrum of histologic patterns. Comparing our HIV-positive and HIV-negative patients, we found a similar prevalence of many histologic variants (excluding non-HIVAN focal glomerulosclerosis, minimal change, and membranoproliferative glomerulonephritis). The predominant pathologies in the HIV-positive group, however, were 'classic HIVAN' and 'HIVICK'. This would tend to support a causative role for the virus in these two pathologies, in our view. 'Classic HIVAN' accounted for less than a third of all patients. They presented at any stage of HIV infection and with varying degrees of renal disease in terms of glomerular filtration rate and proteinuria, in agreement with other authors.11.Glassock R.J. et al.Human immunodeficiency virus (HIV) infection and the kidney.Ann Intern Med. 1990; 112: 35-49Crossref PubMed Scopus (136) Google Scholar, 12.Seney Jr, F.D. Burns D.K. Silva F.G. Acquired immunodeficiency syndrome and the kidney.Am J Kidney Dis. 1990; 16: 1-13Abstract Full Text PDF PubMed Scopus (107) Google Scholar, 13.Winston J.A. et al.Nephropathy and establishment of a renal reservoir of HIV type 1 during primary infection.N Engl J Med. 2001; 344: 1979-1984Crossref PubMed Scopus (259) Google Scholar All the patients in our series with HIVAN acquired their HIV infection in ways other than intravenous drug use. This re-emphasizes that the route of transmission is irrelevant to the pathogenesis of the condition. Although there are limitations to drawing conclusions from our retrospective review (single center study, single pathologist not blinded to clinical data, small numbers), it appears difficult predicting renal histology on the basis of clinical criteria. Patients with HIVAN and HIVICK made up the biggest proportion, but accounted for less than half the group (48%). Although patients with HIVAN tended to have heavier proteinuria, higher creatinines, and lower CD4 counts, there was marked overlap with the other groups. There was no difference between the groups in terms of any of the other clinical parameters. Thus, renal biopsy remains the gold standard for diagnosing HIV-related nephropathies. Following diagnosis, those with HIVAN tend to progress to end-stage renal failure within months.14.Weiner N.J. Goodman J.W. Kimmel P.L. The HIV-associated renal diseases: current insight into pathogenesis and treatment.Kidney Int. 2003; 63: 1618-1631Abstract Full Text Full Text PDF PubMed Scopus (104) Google Scholar This was certainly our experience. Approximately 30 out of the 38 million HIV-infected people worldwide reside in sub-Saharan Africa.15Executive Summary Report on the global AIDS epidemic Joint United Nations Programme on HIV/AIDS, 2004, pp 1–18Google Scholar If other authors are correct16.Shahinian V. Rajaraman S. Borucki M. et al.Prevalence of HIV-associated nephropathy in autopsies of HIV-infected patients.Am J Kidney Dis. 2000; 35: 884-888Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar, 17.Ahuja T.S. et al.Is the prevalence of HIV-associated nephropathy decreasing?.Am J Nephrol. 1999; 19: 655-659Crossref PubMed Scopus (77) Google Scholar and 3.5–12% of Black Africans with HIV are at risk of developing HIVAN, it is clear that Africa is facing an epidemic of renal failure (1–3.5 million people) for which it is ill-equipped. 'HIVICK', thought by some to occur predominantly in Caucasians, was commonplace in our series. Twenty-one patients had this histologic pattern, second only to those with 'classic HIVAN' (27 patients). This group included patients with a 'lupus-like' picture, in spite of being antinuclear factor negative. Although it is not known whether HIV is the cause of this nephropathy, it is known that immune dysregulation can occur. In HIV-infected patients, there are often large amounts of circulating antigen associated with a polyclonal antibody response. This could lead to the formation of immune complexes, either in the circulation or in situ in the kidney itself. The concomitant activation of inflammatory mediators would then result in secondary renal damage in a manner analogous to lupus nephritis. In the HIVICK group, the predominant histologic pattern included variable mesangial alterations, with identifiable immune deposits within the mesangial and paramesangial regions. In addition, some of the biopsies showed a peculiar and to our knowledge not previously described pattern, in which a varying number of subepithelial immune deposits were seen inducing a peculiar localized 'ball-in-cup' reaction pattern from the basement membrane. These deposits are reminiscent of the subepithelial 'humps' seen in a 'post-infectious' diffuse proliferative glomerulonephritis while the basement membrane reaction was similar to a membranous glomerulonephritis. However, diffuse spike formations were not evident, nor were there the numerous subepithelial deposits of a membranous nephropathy. Thus, this picture has overlap features between incompletely expressed 'post-infectious' diffuse proliferative and membranous glomerulonephritides. We propose this may be as a consequence of HIV immune dysregulation. The deposits themselves may well contain viral particles and stains for this may be informative in future. In addition, several of the 'HIVICK' patients had the tubulointerstitial changes of 'classic HIVAN', in the absence of focal segmental sclerosing glomerular lesions. This may indicate that the microcystic tubulointerstitial changes occur independently of the glomerular lesions. The typical microcystic tubulointerstitial change may be a result of direct viral infection of tubular epithelial cells, whereas the focal segmental sclerosing glomerular lesions may be a result of a visceral epithelial cell infection. The next most common histological pattern was membranous nephropathy. It is not clear that HIV is etiologically implicated, although it is known that other infective agents (including viruses) can cause the membranous pattern of disease. However, there was a similar prevalence of membranous nephropathy in our HIV-positive and HIV-negative groups, so that the virus may be co-incidental. There were a similar number of patients with 'post-infectious' nephropathy in the HIV-positive and HIV-negative groups so that it is difficult to draw any conclusions as to the etiology of the virus in this histologic variant. The group of six patients with immunofluorescence-negative mesangial hyperplasia, as mentioned above, may well represent classic HIVAN in an early stage of its evolution. Repeat biopsies may well be informative in this group. Although IgA nephropathy has been thought to be rare in people of Black African origin, it certainly occurred in both our HIV-negative and HIV-positive patients with equal frequencies. This nephropathy has been described by several authors in the context of HIV infection.18.Kimmel P.L. et al.Brief report: idiotypic IgA nephropathy in patients with human immunodeficiency virus infection.N Engl J Med. 1992; 327: 702-706Crossref PubMed Scopus (125) Google Scholar, 19.Katz A. et al.IgA nephritis in HIV-positive patients: a new HIV-associated nephropathy?.Clin Nephrol. 1992; 38: 61-68PubMed Google Scholar, 20.Hsieh W.S. et al.Crescentic IgA nephropathy and acute renal failure in an HIV-positive patient with enteric salmonella infection.Nephrol Dial Transplant. 1996; 11: 2320-2323Crossref PubMed Scopus (11) Google Scholar, 21.Jindal K.K. et al.Crescentic IgA nephropathy as a manifestation of human immune deficiency virus infection.Am J Nephrol. 1991; 11: 147-150Crossref PubMed Scopus (41) Google Scholar Based on the fact that the other glomerulonephritides occurred with equal frequencies in both the HIV-positive and HIV-negative groups, we would argue that these diseases were co-incidental to the infection, but this remains pure conjecture. Three patients had morphologic features of FSH, non-collapsing subtype, with no typical tubulointerstitial changes of classic HIVAN. These were interpreted as non-HIV-related FSH, owing to the absence of microcystic tubular dilatation and the absence of a 'collapsing' form of FSH. The high prevalence of the non-collapsing form of FSH in the HIV-negative group (17%) may indicate that this group of patients displays susceptibility towards this form of renal scarring regardless of the initiating insult. These patients were nephrotic on clinical presentation and the HIV infection was associated with near normal CD4 counts. Of interest was the paucity of minimal change disease and membranoproliferative glomerulonephritis in our HIV-positive patients. The reasons for this are not clear, but would seem to argue against an etiological role for the virus in these nephropathies. Our remaining patients with non-glomerulonephritic-type renal disease had various pathologies ranging from diabetic nephropathy to malignant hypertension, acute tubular necrosis, and tubulointerstitial nephritis. As the majority of our patients with glomerular pathology had moderate to severe interstitial disease, it is very possible to imagine an etiological role for the HIV virus in the pathology of tubulointerstitial nephritis. In conclusion, it is obvious that there is a wide range of kidney diseases that can occur in the context of HIV-infected individuals of African origin. Although 'classic HIVAN' may well be the most common histologic pattern, a number of other pathologies may be found. These may or may not be as a result of viral infection. The immune complex disease pattern of HIVICK was almost as common in our experience and may well be a consequence of HIV infection. It is apparent that many patterns of renal pathology occur in HIV-infected patients. Thus, we would propose an international, multi-center collaboration to try and establish a classification system for HIV-related kidney disease. Even though much is to be learnt about the etiology, natural history, and treatment of the different types of HIV-related nephropathy, the renal biopsy plays a pivotal role in identifying the disease pattern present. In future, we would hope to establish screening programs for those at risk of developing HIV-related chronic kidney disease and then evaluate therapies aimed at slowing the often rapid progression of these pathologies. Screening strategies would probably include routine serum creatinine and urinary protein or even microabluminuric estimations in Black HIV-infected patients. One of the tragedies of HIV-related nephropathy is that the bulk of these diseases occur in a region of the world that is least able to cope at present. Optimal treatment regimens remain largely unknown. Effective therapeutic measures will need to be coupled with carefully thought out screening strategies. It is imperative that properly conducted randomized prospective trials be set up to evaluate these to try and combat this important cause of kidney failure in the developing world. Chris Hani Baragwanath Hospital is a 2888-bed institution located in Soweto on the outskirts of Johannesburg. It serves approximately five million people, the majority of whom are Black Africans. Based on antenatal clinic attendees, 29.4% of those living in Gauteng province (where Soweto is located) are HIV positive.22.Provincial Profile 1999 Gauteng/Statistics South Africa. Statistics South Africa, Pretoria2003: 73Google Scholar It is estimated that 40–50% of all admissions to our Internal Medicine Department, from where we get most of our referrals, are infected. There are few outpatient referrals to our Nephrology practice because of the rudimentary primary health-care infrastructure in the area. Our retrospective cohort included all patients with documented HIV infection who underwent renal biopsy from 1st January 2003 to 31st December 2004. The indications for biopsy were left up to the attending Nephrologist in each case. Permission for the review was gained from the University of the Witwatersrand Ethics Committee. All patients who were biopsied signed consent and had routine blood tests performed before the procedure, including: serum creatinine and urea, bleeding profile, viral serology (Hepatitis B surface antigen, Hepatitis C antibodies, and HIV antibodies), syphilis serology, streptococcal serology, antinuclear antibody, and complement. Spot urine protein creatinine ratios were also performed as routine. Results from all investigations were documented where available. All biopsies were submitted for light microscopy and immunofluorescence, but only for electron microscopy where considered necessary by our pathologist. Biopsies were studied by the same pathologist who was provided with a basic clinical background, including the HIV status of the patient concerned. Biopsies insufficient to make a diagnosis were not included in the study. Further data were gathered from patient records and included: demographic details, blood results (as listed above), CD4 counts where possible, presenting blood pressure, and use of medications (angiotensin-converting enzyme inhibitors (perindopril), statins (simvastatin), steroids, and antiretrovirals). Antiretrovirals were supplied by our hospital from June 2004 onwards, so that the minority of patients was on this therapy. Owing to financial constraints, chronic dialysis was not made available to HIV-positive patients in our hospital. Thus, end-stage renal failure inevitably resulted in death for the patient concerned. Follow-up duration was measured from the date of the biopsy until the recorded date of death or, in the case of those who were lost to follow-up, the date of the last documented clinic visit. Patients were considered lost to follow-up if they did not arrive for scheduled clinic appointments and could not be tracked using either phone numbers or home visits, or their deaths were not recorded at the Department of Home Affairs in Johannesburg, where all death certificates are registered. For those who were known to still be alive, duration was from the date of biopsy until the study closure date of 31st December 2004. All patients were considered to be Black by the Nephrologists involved. Although mode of HIV transmission was not recorded, infection in Soweto is thought to be largely owing to heterosexual sexual contact. Intravenous recreational drug abuse is considered to be virtually non-existent. To try make inferences about the etiological role of HIV infection in certain forms of renal pathology, we compared the above-mentioned HIV-positive group with an HIV-negative 'control' group. The latter was made up of all HIV-negative patients attending our outpatient department who were biopsied during the same time (2003–2004). In the HIV-positive group, patients were categorized according to their histological patterns. In the case of a histological pattern that contained fewer than five examples, the patients were grouped together as either 'other glomerulonephritides', or else, in the case of non-glomerulonephritic disease, 'other kidney disease'. Statistical analysis was performed using the Statistica 6.0 software package from Statsoft. Differences between the various groups were analyzed using the non-parametric Kruskal–Wallis analysis of variance by Ranks test. The demographic make up of the HIV-positive and HIV-negative groups were compared using the Mann–Whitney U-test. Once the prevalence of varying histological diagnoses was known in the two groups, they were compared using the Fisher exact two-tailed test to try and asses any differences in the prevalence of renal pathology.
Referência(s)