|Home | About | Journals | Submit | Contact Us | Français|
Highly Active Antiretroviral Therapy (HAART) has resulted in a marked decrease in AIDS-related conditions and death. With improved survival, cardiovascular disease (CVD), hepatic, renal disease and non-AIDS related cancers represent an increasing burden for HIV infected individuals.
HIV Associated Nephropathy (HIVAN), acute renal injury, HAART, and co-morbid conditions such as Hepatitis C, hypertension and diabetes are among the multiple causes of renal disease. In HIVAN there is incomplete understanding of the interaction of the virus with renal cellsand the host genetics leading to susceptibility to this form of renal dysfunction. There is agreement that a baseline estimate of glomerular filtration (eGFR) should be obtained and that renal function should be monitored during antiretroviral therapy. There is, however no agreement as to the most accurate method of estimating GFR. Renal transplantation has emerged as a feasible and successful modality of management of end stage renal disease (ESRD) in HIV infected individuals.
Kidney disease represents an increasing concern in the care of HIV infected persons although there are questions remaining regarding the pathophysiology of HIVAN. Transplantation, however, can be carried out safely in infected persons with ESRD.
Suppression of HIV replication and partial restoration ofimmune competence following administration of effective antiretroviral therapy has resulted in prolongation of survival of HIV infected individuals. In addition, HIVAN has diminished as a cause of renal dysfunction in areas of the world with adequate medical resources. Recipients of HAART, however, are subject to diseases associated with aging such as diabetes, hypertension, and chronic kidney disease (CKD). Furthermore, specific antiretroviral agents are associated with diminished renal function
Adih et al have examined causes of death listed on death certificates of HIV infected persons in the HAART era [1*]. Their analysis documented declines of death due to Acquired Immunodeficiency Syndrome (AIDS) in the United States. Non-AIDS deaths have in part replace AIDS as a cause of death and renal disease was listed as the primary cause on 1682 of 13,750 (12.2%) of death certificates. A similar pattern of an increasing proportion of deaths due to non-AIDSconditions has been reported from Latin America and by the D:A:D study , however, fewer deaths than reported by Adih were attributed to renal disease. A community based investigation in Rakai, Uganda, noted that 8% of infected persons initiating HAART had mildly reduced estimated glomerular filtration (eGFR) or moderately reduced eGFR. This frequency of moderate reduction in eGFR was lower than that seen in HiV infected African-Americans (AA) in Baltimore Md. The investigators speculated that modification of diet in renal disease (MDRD) equation which they used to estimate GFR does not incorporate weight and would overestimate eGFR in lighter individuals who eat a diet lower in proteinthan that eaten by persons in industrialized countries where the MDRD was developed [4*].
Multiple publications have linked kidney function to morbidity and mortality [5*]. Hemmelgarn et al reported that all cause mortality, myocardial infarction and progressive renal failure are increased in non-HIV infected in a community based cohort study in Alberta [6*]. Acute renal injury greatly increases the risk of CVD, ESRD and mortality in HIV infected individuals [7*]. Choi et al also reportdata from a study of HIV infected persons notinga 48 % mortality in persons with eGFR of less than 60 and albuminuria and 23% mortality in those eGFR <60 alone and 20% mortality in those with albuminuria alone [8**]. Estrella et al using data from the Women’s Interagency HIV Study reported that kidney function at HAART initiation remained a determinant of mortality after adjusting for age, race, Hepatitis C, AIDS and CD4+T cell count. Adjustment for hypertension and diabetes did not completely attenuate this risk [9**]. Specifically renal dysfunction as determined by eGFR and proteinuria were noted by George et al to be associated with cardiovascular events in 63 patients compared to 252 matched controls without CVD in the Johns Hopkins HIV Clinic [10*].
The critical issue of the method used to determine eGFR in the absence of a direct measure of GFR was discussed by Post et al. Serum creatinine-based estimates of GFR all have limitations [11**]. The Cockcroft-Gault equation adjusts for muscle mass and gender but not for tubular reabsorption of creatinine. The MDRD, also using serum creatinine, again does not account for tubular reabsorption of creatinine. Derived from both men and women the MDRD was compared to a direct measure of GFR, iothalamate clearance. The CKD Epidemiology Collaboration (CKD-EPI) also based upon serum creatinine combined data from more 8000 persons in 10 different studies and was validated in an external sample. The CKD-EPI is more accurate than the Cockcroft-Gault or MDRD equations in persons with normal renal function. However, onlysmall studies of renal function have been conducted in HIV infected persons. Cystatin C has been proposed as an alternate to creatinine as it is not influenced by muscle mass and tubular secretion. Post et al conclude that until creatinine and cystatin Cbased eGFR are compared to a gold standard the best estimate of renal function in HIV infected persons remains unclear. Post et al also noted that routine measurement of urinary bio-markers of tubular function in clinical practice is not indicated [11**].
Determination of urinary protein or albumin can contribute to estimates of renal dysfunction. A comparison of the urinary albumin-creatinine ratio (ACR), urinary total protein-creatinine ratio (PCR), 24 hour urinary albumin or total protein demonstrated that ACR and PCR were as effective as 24 hour urinary albumin or protein in predicting renal outcomes. Spot urine for either ACR or PCR moreover, are more convenient for patients, clinicians and laboratories. Tonelli et al used total proteinuria measured by dipstick or ACR plus the MDRD equation to classify the risk of all cause mortality and a composite renal outcome of failure or doubling of serum creatinine in a large sample of uninfected persons.[12**] An accompanying editorial in the Annals of Internal Medicine discussed revision of the definition of CKD to include ACR and clinical diagnoses as well as eGFR to determine risk categories for progression to CKD .
HIVAN is the most common cause of CKD in HIV infected persons of African descent who are not receiving HAART. HIV mediates dysregulation of glomerular podocytes, the epithelial cells which maintain the glomerular basement membrane, and apotosis of renal tubular cells. The resulting lesion of HIVAN is a focal glomerulosclerosis (FGS) and microytic dilation of tubules [14**].
It appears that HIVAN occurs in the setting of specific host genes and it t has been widely accepted that the gene encoding a non-myosin A heavy chain (MYH9) is associated with FGS, HIVAN and end stage kidney disease due to hypertensionin persons of African descent [16,17]. Recently, however, Genovese and co-workers have reported that genome wide association analysis in AA and European Americans (EA) with FGS identified a 60kb region on chromosome 22 which contains MYH9 and a gene which transcribes atrypanolytic APOL1 variant protein which lyses Trypanasoma brucei. This variant was found in AA but not in EA and was presumably selected because of exposure to Trypanasomein Africa. Genovese et al state that this 60kb region spans disequilibrium blocks which includes the gene encoding MYH9. This and a later publication from this group noted that APOL1 variants are absent in Europeans [18,19**]. Other investigators have confirmed the association of the APOL1 variant and FGS in AA [20*].
The interaction of HIV with renal cells has not been clarified. Renal cells lack the classic receptors for HIV, CD4 and the chemokine receptors CCR5 and CXCR4. However, HIV mRNA has been detected in renal epithelialcells and HAARTslows progression of HIVAN to renal failure . Non-conventional receptors on renal cells such C-type lectins including DC-Sign and lipid rafts have been suggested as portals of entry of HIV into renal cells. [22*]. Renal cell infection by HIV, however, is non-productive . Host factors such as activation of the renin-angiotension-aldosterone system have also been implicated in the pathogenesis of HIVAN) and blocking production of Angiotensin II slows progression of HIVAN . Finally, Sharma et al suggest that the NOTCH signaling pathway is activated by HIV proteins that leads to cellular dysregulation of podcytes and dedifferentiation of these cells . Of interest is a recent report that in AA men HIVAN with low PBMC proviral DNA. Conversely, AA men with undetectable plasma HIV RNA but high proviral DNA were at risk for HIVAN as documented by renal biopsy [25*].
The major co-infection associated with renal disease in the presence of HIV infection is Hepatitis C (HCV). As reviewed by Operskalski and Kovacs, patients with HIV/HCV demonstrate proteinuria, acute interstitial nephritis and acute tubular necrosis although the mechanism underlying the pathogenesis of these conditions in unknown. In contrast, membranoprolifrative glomerulonephritis secondary to B-cell stimulation, increased cryoglobulin production and deposition in the renal vasculature has been well documented [26**]. CKD in HIV/HCV infected persons is associated with reduced survival as reported from a large study from the Veterans Administration [27**] Smaller studies have documented that co-infection increases the risk death as compared to HCV or HIV mono infections . Finally, in Cape Town South Africa where tuberculosis is endemic, granulomatous tubercular interstitial nephritis has been recognized as a cause of proteinuric CKD in children .
Several large investigations have reported that renal impairment is associated with administration receipt of tenofovir or indinavir[30–33*]. In the EuroSida study the incidence rate ratio of developing CKDfor cumulative exposure to tenofovir was 1.16 (95% CI 1.06–1.25), indinavir 1.12 (95% CI 1.06–1.18), atazanavir 1.12 (95% CI 1.09–1.35) and lopinavir/ritonavir 1.08 (95% CI 1.01–1.16) all significant increases. [34*]. Rodriguez et al noted that although tenofovir has clinically unimportant effects upon eGFR, high plasma trough levels are associated with renal tubular dysfunction. Specific polymorphisms in genes coding transport proteins have been associated with tubular damage with tenofovir administration. Labarga et al have also reported tubular damage in patients with little or no loss of creatinine clearance . In multivariate analysis comparing 154 patients receiving tenofovir, 49 on other HAART regimens and 81 untreated HIV patients the odds ratios predicting tubular dysfunction was 21.6 (95% CU 4.1–113) . Post et al also noted that routine measurement of urinary bio-markers of tubular function in clinical practice is not indicated. Several reports in children have also emphasized the use of tenofovir in association with tubular dysfunction in the absence of changes in eGFR. Judd et al notes that tubular dysfunction is associated with prolonged use of the tenofovir and is generally reversible with withdrawal of the drug .
Raltegravir has been reported to be associated with rhabdomyolysis and renal dysfunction in three cases . The fact that raltegrivir has a low molecular weight and is highly soluble in water suggests that it would be removed by hemodialysis. In two patients with ESRD, however, hemodialysis minimally removed raltegravir . The pharmacokinetics of the as yet unreleased CCR5 antagonist, viciriviroc, are not altered in patients with ESKD and the plasma concentration post dialysis was similar to predialysis levels.
Reducing the prevalence of renal disease in HIV infected persons will have an impact upon non-AIDS morbidity and mortality. Szezech has pointed out that although HAART slows progression of HIVAN it is likely that the prevalence of CKD will increase with prolongation of survival and the probable increase of hypertension and diabetes in HAART treated HIV persons. Furthermore, biopsy data is rare which prevents developing rationale treatment of other forms of renal disease [21*]. Arendes et al have noted in Cape Town that acute tubular necrosis, nephrotoxic drug exposure or hypovolemeic shock were diagnosed in 58% of hospitalized HIV patients requiring renal support. Renal biopsies demonstrated HIVAN in 20% and 10% had other renal pathology. HIVAN was most commonly seen in late stages of HIV infection and it has not been established when institution of HAART is no longer effective in slowing or reversing HIVAN . The experience in Cape Town illustrates also that immune complex nephritis alone or in combination with HIVAN was found in a large number by biopsy and that the treatment of immune complex nephritis has not been established.
Szczech et al noted that microalbuminuria predicts proteinuria in HIV infected individuals and studies have linked proteinuria to CVD and mortality. She suggests that treatment strategies such as angiotensin receptor blockers and HAART need to be evaluated to determine if management of microalbuminuria can prevent these outcomes [42*].
Stock et al [43**]have reported the experience of a multicenter investigation of renal transplantation in HIV infected persons with ESRD receiving HAART. Eligibility depended upon control of HIV infection and a CD4+ T cell count greater than 200/mm3. In persons with evidence of viral hepatitis a liver biopsy was obtained to rule out cirrhosis. Prophylaxis for opportunistic pathogens was utilized as clinically indicated. The recipients were followed for a median of 1.7 years. Both living and deceased donors were utilized. Survival at one and three years was 94.6 and 88.2% and graft survival was 90.4% and 73.7%respectively. These results were equivalent to those seen in non-HIV infected graft recipients older than 65 years of age. Few recipients showed progression of HIV infection; however, the cumulative incidence of rejection was 31% (95% CI 24–40) at one year and 41% (95% CI 32–52) at three years, which is higher than seen in non-HIV infected graft recipients. The authors, in summary, state that renal transplantation is feasible in appropriately selected HIV infected patients. Strategies to lower graft rejection are needed.
Other investigators have evaluated whether or not HIV infected persons with ESRD would accept a kidney from an HIV infected donor. A majority, under specific circumstances would consider acceptance of a graft from an infected donor . A companion paper raised the issue of whether it was ethical to inform a living donor that the recipient was HIV infected. The authors concluded that the privacy of the recipient was paramount and that the donor should be told only that the potential recipient required a renal transplant .
Renal disease represents an increasing cause of morbidity and mortality in HIV infected individuals who are surviving longer due to effective antiretroviral therapy. Longitudinal studies of renal disease in HIV infected persons receiving or not receiving HAART, however, are not yet available. Effective treatment when initiated early. GFR in HIV infected persons has not been clearly established. Clinically management of HIV infection includes monitoring of serum creatinine and urinary albumin or protein as well as treatment of hypertension and diabetes. Successful renal transplantation of selected HIV infected persons represents a major breakthrough in the management of this condition.
Enhanced survival of treated HIV infected persons will result in an increased number with renal disease. The pathogenesis of HIVAN in persons of African descent is related to host genetics.
Both glomerular and tubular dysfunction can be related to exposure to specific antiretroviral agents. Renal transplantation is feasible for HIV infected patients with end stage renal disease.
Dr. Phair is supported by NIH grants UO1AIO35039, RO1AG0348522 and UO1 AI034041
Dr. Palella is supported by NIH grants 1RO1HL095129, 1520//RO1AG034853 and CDC 200-2006-18797
John Phair, Emeritus Professor of Medicine, Feinberg School of Medicine, Northwestern University.
Frank Palella, Professor of Medicine, Feinberg School of Medicine, Northwestern University.