HIV and renal impairment in the cART era
Renal impairment in general is an emerging focus area in the care of HIV positive people. It may lead to severe complications such as
This is a short introduction to the field, focusing on causes, risk factors, possible strategies for patient management and emphasizes that the role of specific antiretrovirals (ARVs) in relation to development of renal impairment warrants further evaluation.
Renal impairment in HIV positive people is either acute or chronic, and can be related to HIV-infection and opportunistic infections, treatment (antiretrovirals as well as other HIV/AIDS-related drugs) as well as factors not related to HIV-infection.
Acute renal impairment
Most common causes of acute renal impairment in HIV positive patients are opportunistic infections and their treatment, severe dehydration and usage of specific antiretrovirals. Indinavir and Atazanavir, both protease inhibitors, are known to cause acute renal impairment via crystaluria/ kidney stones, which is seen in up to 27% of patients on Indinavir (II).
A common clinical finding in patients receiving this treatment is proteinuria and hematuria; however, the significance of this is not completely known. Usage of tenofovir/adefovir/cidofovir all nucleotide reverse transcriptase inhibitors, is also known to cause acute renal impairment through an affection in the proximal tubules of the kidneys causing a Fanconi-like syndrome (III)
Chronic renal impairment
Chronic kidney disease (CKD) may arise from multiple causes. The classic type is caused directly by the HIV virus itself, causing a focal segmental glomerulosclerosis (HIVAN, HIV associated nephropathy) especially in patients of African ethnicity.
Recent research has focused on identifying the genes responsible for predisposition to HIVAN-like kidney affection, and to determine if genetic polymorphism in specific renal tubular transporters play a significant role in predisposing to a more fragile response to potential nephrotoxic treatment.
Several studies have assessed common risk factors for HIV-associated CKD in general and have identified the following factors to predispose; advanced age, diabetes, hypertension, co-infection with viral hepatitis, prior AIDS events, treatment of opportunistic infections (e.g. aminoglycosides, amphotericin B and acyclovir), late HIV diagnosis/treatment failure with low CD4 count/high viral load and exposure to certain ARVs (II, IX, XII, XV). Other risk factors (low and high BMI, dyslipidemia, statin use) are believed to play a role as well but need to be further investigated.
The overall effect of ARVs on the incidence of CKD is not completely clear; some experience improvement in kidney function after initiation of anti-HIV treatment (those with e.g. HIVAN) whereas others develop a decline in kidney function (IV, V, VI, VII).
The renal impairment associated with certain ARVs is believed to be caused by either direct nephrotoxic damage or indirect via other side effects to the treatment such as diabetes and hypertension. The definitive role of antiretroviral drugs in the CKD pathogenesis is still unclear partly due to issues regarding defining CKD, having sufficient follow up time on the suspected ARVs and an appropriate number of patients in the study due to low overall incidence.
A recent study performed based on almost 7000 patients followed in EuroSIDA (a large prospective European cohort study) documents that cumulative exposure to certain ARVs is associated with development of at least moderate CKD within a median follow-up time of approximately 4 years (VI). Patients had a 16% increased risk of CKD per year of exposure to tenofovir (VIII). Of note, these data are of observational nature and do not allow for firm conclusions on causality. Similar associations were found for indinavir (12% increased risk per year of exposure), and atazanavir (21%). For these three drugs, the results were consistent, whereas the evidence for lopinavir/ritonavir was less clear (8% per year of exposure). Most patients developing CKD in this study had other pre-existing traditional renal risk factors (IX). Multiple randomised studies have also investigated the cumulative effect of tenofovir on kidney function, however these were all fairly small studies with a relatively short follow up (X). Further the included patient population does not necessarily reflect the general HIV population e.g. age and co-morbidities. Many observational studies lack adequate power to analyse the potential associations due to low overall incidence, and cannot at the present address the role on newly introduced drugs due to limited follow-up time.
Definition of kidney function
A simple creatinin measurement is too insensitive to determine kidney function, since a significant rise is only seen when the kidney function has waned to under half of normal function. Determining the glomerular filtration can more precisely assess kidney function. This is done by measuring the creatinin clearance, and can be done in different ways. One way is to analyse 24h urine collections, however this may be associated with practical difficulties.
The gold standard today is therefore measurement of the clearance of a tracer substance that is excreted only by filtration (without reabsorption or secretion in the tubule) e.g. Cr-EDTA.
An easier way is to use a surrogate measurement called estimated glomerular filtration rate (eGFR) based on creatinine levels. eGFR can be calculated using different equations (Cockcroft-Gault (CG), Modification of Diet in Renal Disease (MDRD), Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI), an eGFR calculator can be found on the CHIP website. It is not known which formula in general has the most advantages.
An eGFR>90 ml/min per 1,73 m2 is considered normal kidney function (I, VIII). CKD is often defined as eGFR < threshold measured at least twice and consecutively over at least a 3-month period (II, VIII).
Individual stages (2-5) of CKD are based on severity and are independent of cause:
Of note, this classification may overestimate the prevalence of CKD in patients with only mildly reduced kidney function, especially the elderly. Another issue is the lack of proteinuria findings in the definition.
Renal impairment may lead to a variety of complications. The risk of these complications increases with the severity of the impairment. Renal impairment is often asymptomatic in early stages, making regular screening of all HIV patients very important. Screening procedure guidelines have been available since 2005 and recommend that all patients at the time of HIV diagnosis should be assessed for existing kidney affection with urine screening for protein and an eGFR calculation. After diagnosis, an eGFR and dip stick analysis should be done at least once a year (XI).
Several strategies are recommended if a diagnosis of renal impairment is made. These include determination of severity, identification of the possible cause(s), correction of the(se) cause(s) if possible (e.g. removal of nephrotoxic treatment, rehydration, treatment of infection/diabetes/ hypertension), frequent control of renal function, treatment of complications (hypertension, anaemia, malnutrition, electrolyte derangements etc.) and possibly referral to a nephrology department (see guidelines for referral in screening guidelines above).
Prevalence and incidence
Estimating the prevalence of CKD in HIV positive people depends greatly on how one defines the condition (see section above). Using a definition based on eGFR being below a threshold of 60 ml/min - and thus not including the presence of proteinuria- the prevalence is between 2,5 and 11% in the US and Europe (I, XII, XIII, IVX), naturally depending on a variety of other factors including the level of co-morbidity and other risk factors for CKD in a given population. In EuroSIDA, a CKD incidence of 1,05 per 100 PYFU was demonstrated and a prevalence of 3,5%.
If CKD is defined by eGFR and the presence of significant proteinuria, the estimate is increased but the clinical relevance is not determined (I, XIII, IVX, IV, VIII). The prevalence/incidence of more advanced stages of CKD (eGFR <30 and <15 ml/min) has not yet been finally determined due to low incidence. However, a European cross-sectional multi-center survey has found an end stage renal disease prevalence of 0,5%, most frequently caused by HIVAN (VX).
With time, the overall incidence and especially the prevalence of CKD is believed to increase in the HIV positive population partly due to reduced mortality in an ageing cohort and may also be due to lifelong cumulative exposure to certain ARVs.
Possible mechanisms for drug-associated chronic nephrotoxicity
The underlying mechanism for a possible tenofovir-associated chronic nephrotoxicity is still unknown. More studies have identified a high drug concentration in the proximal kidney tubule due to high uptake in certain anionic transporters and reduced secretion. This is believed to cause cell damage via mtDNA depletion and dysfunctions of the oxidative procedures. This is confirmed by a smaller study that showed that rats given 8 weeks of tenofovir treatment developed loss of body- and kidney weight as well as a dilation of the proximal tubules and enlarged number and size of mitochondrias in the affected tubule cells (XVI, III, XVII). The mechanism of the possible nephrotoxic effect of atazanavir and possibly also lopinavir is unknown.
Lacking knowledge on CKD and ARVs
Little is known about the risk factors for more advanced stages of CKD, since these are still infrequent. The role of individual ARVs on kidney function with long-term exposure and the contribution (if any) relative to other renal risk factors has yet to be finally determined. A possible interplay between multiple risk factors and ARVs needs to be evaluated. The possibility of full reversibility of drug-induced CKD is unknown at the present.
It is very important with early diagnosis of renal impairment to prevent complications and reduce mortality. Increased attention on renal affection on both short and long term should be paid to all HIV patients, but especially to patients with pre-existing renal risk factors (e.g. high age, diabetes, hypertension or co infections with hepatitis) on ARVs until the role of individual ARVs are disentangled and guidelines for prescription and control are updated.
(I) Early recognition and prevention of CKD. MT James, BR Hemmelgarn, M Tonelli.
Lancet 2010; 375: 1296-309.
(II) Chronic renal failure among HIV-1-infected patients. A Mocroft, O Kirk, J Gatell et al.
(III) Acute renal failure and Fanconi syndrome in an AIDS patient on tenofovir treatment – case report and review of literature. A Malik, P Abraham, N Malik.
J Infect. 2005; 51:e61-e65.
(IV) Recent developments in HIV and the kidney. FA Post, SG Holt.
Curr. opinion in infectious disease 2009,22:43-48.
(V) HIV-infected persons continue to lose kidney function despite successful antiretroviral therapy. AI Choi, MG Shluipak, PW Hunt et al.
Aids 2009, 23:000-000.
(VI) Long term evolution and determinant of renal function in HIV-infected patients who began receiving CART in 1997-1998 ANRS CO8 APROCO-COPILOPE. C Leport, V Bouteloup, J Rossert et al.
(VII) Kidney disease in HIV infection: Introduction Semin Nephrol. CM Wyatt, PE Klotman.
Nov 2008; 28(&): 511-512.
(VIII) Definition and classification of CKD: a position statement from KDIGO. AS Levey, KU Eckardt, Y Tsukamoto et al.
Kidney Internatational, vol 67 (2005): 2089-2100
CKD as a global public health problem: approached and initiatives- a position from KDIGO. AS Levey, R Atkins, J Coresh.
Kidney International (2007) vol 72:247-259.
(IX) Estimated glomerular filtration rate, chronic kidney disease and antiretroviral drug use in HIV-positive patients. A Mocroft, O Kirk, P Reiss, S De Wit, D Sedlacek, M Beniowski, J Gatell, AN Phillips, B Ledergerber, JD Lundgren, for the EuroSIDA study group.
(X) Renal toxicity associated with tenofovir use. Rodriguez-Nóvoa S, Alvarez E, Labarga P, Soriano V.
Expert opin drug safety. 2010,9(4).
(XI) Guidelines for the management of CKD in HIV-infected patients: recommendations of the HIV medicine association of the infectious disease society of America. SK Gupta, JA Eustace, JA Winston et al.
CID 2005:40, (1.jun) 1559-85.
(XII) Spectrum of CKD in HIV infected patients. LJ Campbell, F Ibrahim, M Fisher et al. HIV medicine 2009, 10, 329-336.
(XIII) HIV and the kidney: a status report after 20 years. EM Cho, JB Kopp.
Curr HIV/AIDS reports 2004, I: 209-115.
(XIV) Prevalence of proteinuria and the development of CKD in HIV-infected patients. SK Gupta, BW Mamlim, CS Johnsin et al.
Clin nephrol. 2004jan; 61(1): 1-6.
(XV) Dialysis and renal transplantation in HIV-infected patients: a European survey. JC Trullas, A Mocroft, F Cofan et al.
J acquir immune defic syndr vol 00, nb 0, 2010.
(XVI) Impact of Tenofovir on renal function in HIV-infected, Antiretroviral –naïve patients. M Hornberg, B Tang, W Towner et al.
J. acquir Immune defic syndr vol53 1 Jan 2010.
(XVII) Mitochondrial tubulopathy in TNF disoproxil dumarate-treated rats. D Lebrecht, AC Venhoff, J Kirschner et al.
J. accuir. immune defi.c synd.r 2009 jul1; 51(3):258-63.