cjasn epress article tubulointerstitial nephropathies...

Article

Tubulointerstitial Nephropathies in HIV-InfectedPatients over the Past 15 Years: A Clinico-PathologicalStudy

Mohamad Zaidan,* Francois-Xavier Lescure,†‡ Isabelle Brocheriou,§| Sarah Dettwiler,¶ Jean-Baptiste Guiard-Schmid,†

Jerome Pacanowski,** Eric Rondeau,|††‡‡ Gilles Pialoux,†| Pierre-Marie Girard,|** Pierre Ronco,*|‡‡ andEmmanuelle Plaisier*|‡‡

SummaryBackground and objectives The therapy and outcome of HIV infection have dramatically changed over the last 15years, resulting in a change in renal complications. This study analyzed the characteristics of HIV-infectedpatients and biopsy-proven tubulointerstitial nephropathies to define disease patterns and therapeuticimplications.

Design, setting, participants, & measurements A clinico-pathologic retrospective study of 59 consecutive renalbiopsies showing predominant tubular and/or interstitial lesions in HIV-infected patients referred to thenephrology department between 1995 and 2011 was performed. HIV-associated nephropathy and vasculardiseases were excluded from the study.

Results Tubulointerstitial nephropathies accounted for 26.6% of 222 native renal biopsies performed in HIV-infected patients. Two pathologic groups were analyzed, tubulopathy and interstitial nephritis, which represented49% and 51% of tubulointerstitial nephropathies, respectively. Most patients presented with AKI (76.3%) and high-grade proteinuria (57.7%). Drug-related nephrotoxicity was the leading cause (52.5%). Alternative etiologiesincluded infections (15.2%), dysimmune disorders (8.5%), malignancies (3.4%), and chronic (10.2%) and acute(10.2%) tubulointerstitial nephropathies of undetermined origin. Tubulopathy was strongly associated withantiretroviral drug toxicity (75.9%) and mostly caused by tenofovir (55.2%), which was associated with proximaltubular dysfunction (87.5%), overt Fanconi’s syndrome (37.5%), and nephrogenic diabetes insipidus (12.5%). In-terstitial nephritis was associated with a broader spectrum of pathologic lesions and etiologies.

Conclusions In this series, tubulointerstitial nephropathies accounted for 26.6% of renal diseases in HIV-infectedpatients. Considering the therapeutic implications of diagnoses of drug toxicity, infection, and dysimmunesyndromes, this study underscores the importance of monitoring renal parameters in HIV-infected patients andpoints to the relevance of kidney biopsy to allow an accurate diagnosis.

Clin J Am Soc Nephrol 8: ccc–ccc, 2013. doi: 10.2215/CJN.10051012

IntroductionInterstitial nephritis (IN) and tubulopathy are twomain causes of renal disease, representing between15% and 27% of causes of AKI in Western countries(1,2). Tubulointerstitial nephropathies (TINs) are re-lated to a wide range of causes, including drug tox-icity, infections, and autoimmune disorders (1–3).Only a few studies have investigated the character-istics of TIN in HIV-infected patients (4). Indeed,HIV-associated renal diseases have been dominatedby glomerular diseases, particularly HIV-1–associatednephropathy (HIVAN) (5–7). In the mid-1990s, com-binational use of antiretroviral (ARV) agents in highlyactive antiretroviral therapy regimen was associatedwith a dramatic decrease in the mortality of HIVpatients and incidence of HIVAN (8–10). In themeantime, nephrotoxicity of new ARVs and renal

complications of dysimmune disorders, including im-mune reconstitution inflammatory syndrome (IRIS)and diffuse infiltrative lymphocytosis syndrome(DILS), were reported (4,11–14). In a North Americanseries of 262 HIV-infected patients who underwent akidney biopsy, diagnosis of acute IN was establishedin 11% of renal biopsies between 1995 and 2008 (15).To determine if these findings were similar to Euro-pean practice, we conducted a retrospective study ofthe characteristics and clinical course of HIV-infectedpatients with biopsy-proven TIN.

Materials and MethodsWe included all consecutive HIV-infected adults

who underwent a renal biopsy at Tenon Hospital(Paris, France) from January of 1995 to April of 2011 in

*Assistance Publique-Hopitaux de Paris(APHP), Departmentof Nephrology andDialysis, TenonHospital, Paris,France; †APHP,Department ofInfectious andTropical Diseases,Tenon Hospital, Paris,France; ‡Centre deRecherche Saint-Antoine, InstitutNational de la Sante etde la RechercheMedicale (INSERM)Unite Mixte deRecherche (UMR)S 938, Paris, France;§APHP, Department ofPathology, TenonHospital, Paris,France; |UniversitePierre et Marie Curie-Paris 6, Paris, France;¶Department ofPathology, GenevaUniversity Hospital,Geneva, Switzerland;**APHP, Departmentof Infectious andTropical Diseases,Saint-AntoineHospital, Paris,France; ††APHP,Department of RenalIntensive Care Unitand Transplantation,Tenon Hospital, Paris,France; and ‡‡INSERMUMR S 702, Paris,France

Correspondence: Dr.Emmanuelle Plaisier,Department ofNephrology andDialysis, HopitalTenon, 4 rue de laChine, 75020 Paris,France. Email:[email protected]

www.cjasn.org Vol 8 June, 2013 Copyright © 2013 by the American Society of Nephrology 1

. Published on May 2, 2013 as doi: 10.2215/CJN.10051012CJASN ePress

mailto:[email protected]

mailto:[email protected]

whom the histopathological diagnosis of TIN was retained.The indications for renal biopsies were acute or chronicrenal failure and/or proteinuria and/or hematuria. Pa-tients with prominent glomerular disease or vascularnephropathy were excluded. Approval of the study wasobtained from the Comité de Protection des Personnes, Ilede France.Demographics data and comorbid conditions included

age, sex, ethnicity, hypertension, and hepatitis B andhepatitis C coinfections. The following clinical data onHIV infection were collected: time since HIV infectiondiagnosis, Centers for Disease Control and Preventionstaging, prior and current opportunistic infections andmalignancies, and prior and current ARV therapy. Labo-ratory measurements at the time of the biopsy includedbaseline and peak serum creatinine, electrolytes, proteinuria-to-creatininuria ratio (UP/C), hematuria, leukocyturia,glycosuria, CD4 cell count, and plasma HIV RNA load.Estimated GFR (eGFR) was calculated using the four-variable Modification of Diet in Renal Disease equation.CKDwas defined by eGFR less than 60 ml/min per 1.73 m2

for at least 3 months. Acute renal injury was classifiedaccording to the Risk, Assessment, Failure, Loss, andEnd Stage Renal Disease classification (16). Proximal tu-bular dysfunction was defined as two signs among gly-cosuria, low-molecular weight proteinuria, and lowphosphate serum level, one sign plus low serum potas-sium or uric acid level, or metabolic acidosis. Follow-updata were assessed in the 6-month period after the renalbiopsy and at patients’ consultations when available.Complete recovery was defined as full recovery of base-line renal function. Patients were considered to have par-tial recovery if renal function improved but did notrecover to its baseline value. Patients with renal functionthat remained stable or worsened were considered as hav-ing stable or worsening renal outcome, respectively.Renal biopsies were blindly reviewed by three pathol-

ogists specialized in kidney pathology according to stan-dard protocols. Causes of TIN were identified on the basisof the clinician’s diagnosis and the review of follow-updata. Causal relationship between drug and renal damagewas based on (1) drug exposure at the onset of renal symp-toms, (2) consistent pathologic findings, (3) absence of al-ternative causes, and (4) complete or partial recovery afterdrug withdrawal. The first three criteria had to be met toconsider a causal relationship, whereas the fourth criterionwas optional.Continuous variables were compared using the Mann–

Whitney test, and categorical variables were compared us-ing the chi-squared or Fisher exact test. A value of P,0.05was considered significant.

ResultsBaseline Characteristics and Renal Parameters of 59PatientsAmong 222 renal biopsies analyzed in HIV-infected

patients during the study period, 59 (26.6%) patientswho showed a TIN (tubulopathy, n=29; IN, n=30) wereincluded in the study. All patients had HIV-1 infection.Four patients were biopsied before 1997, and they allhad IN. Baseline characteristics are detailed in Table 1.

Median age was 45 years (range=23–85), with a men-to-women ratio of 3.2. Sixteen (27.1%) patients were black.Hepatitis B and hepatitis C coinfections were reported in12 (20.3%) and 8 (13.6%) patients, respectively. The me-dian time between HIV diagnosis and renal biopsy was10 years (range=0–22); 29 (58%) patients had a CD4 cellcount,200/mm3, and 27 (52.9%) patients had detectableviral replication (median=15,000 copies/ml, range=94–13107). Thirty-seven (62.7%) patients had reached theAIDS stage. About 90% of patients received a combinationof ARVs at the time of biopsy. A past history of kidneydisease was reported in 19 (32.2%) patients, includingstage III/IV CKD in 13 (22%) patients, pyelonephritis in3 (5.1%) patients, and urolithiasis in 4 (6.8%; 3 related toindinavir and 1 related to atazanavir) patients. At the timeof biopsy median, eGFR was 26 ml/min per 1.73 m2

(range=3–146); 40 (76.3%) patients had AKI or acute kid-ney failure (AKF), with 11 acute-on-chronic cases. MedianUP/C was 160 mg/g (range=0–10,696), with 12 (23.1%) pa-tients showing UP/C.2652 mg/g (Table 2). Analysis ofcomposition of proteinuria was available for 12 of 30 (40%)patients with UP/C.1326 mg/g. LWM proteins predom-inantly composed proteinuria, and median albuminuria-to-proteinuria ratio was 7.3% (range=0%–33%). Hematuriaand leukocyturia were detected in 16 (30.8%) and9 (20.9%) patients, respectively.

Renal Biopsy FindingsKidney biopsies were divided into two groups: 29 (49%)

patients showed tubulopathy, and 30 (51%) patients dis-played acute or chronic IN. Baseline characteristics did notdiffer between the two groups, except that tenofovir andritonavir exposures were more frequently observed in thetubulopathy group (P=0.005 and P=0.02, respectively).Drug-related nephrotoxicity accounted for 52.5% of overallcases of renal injury. For the remaining 47.5% cases, etiol-ogies included infections (15.2%), dysimmune disorders(8.5%), malignancies (3.4%), chronic TIN (10.2%), andacute TIN of undetermined origin (10.2%) (Table 3).

TubulopathyBaseline Characteristics. Summary and detailed char-

acteristics of 29 patients with tubulopathy are shown inTable 2 and Supplemental Table 1, respectively. Twenty-three (79.3%) patients had AKI or AKF, and nine (31%)patients required renal replacement therapy (RRT). Four-teen patients with acute tubular necrosis (ATN) lesionshad a lower median GFR (11.5 versus 44 ml/min per1.73 m2, P=0.001) at presentation and required more fre-quent RRT (50% versus 13.3%, P=0.05) compared with pa-tients without ATN (n=15). Although median eGFR wascomparable with patients with IN, a higher proportion ofpatients with tubulopathy showed AKF and required RRTat presentation (P=0.02 and P=0.09, respectively). Fifteen(62.5%) patients had a UP/C.1326 mg/g. Proximal tubu-lar dysfunction was recorded in 21 (72%) patients, with 7patients presenting with complete Fanconi’s syndrome.Three patients displayed nephrogenic diabetes insipidus.Biopsy Findings. Renal biopsy specimens showed

acute tubular injury of varying intensity. Fifteen patientsdisplayed mild to moderate tubular damage that

2 Clinical Journal of the American Society of Nephrology

http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.10051012/-/DCSupplemental

predominated on proximal tubules without ATN. Theselesions included cytoplasmic vacuolization and focal loss ofbrush border but no tubular basement membrane denu-dation. Fourteen patients had ATN consisting of flattenedand ragged tubules with luminal ectasia, cellular casts,diffuse loss of brush border, and tubular basement mem-brane denudation. Nuclear atypia (dystrophic nuclei andprominent nucleoli) were seen, mostly in patients withATN. No tubular microcystic dilation, epithelial cell hy-pertrophy or hyperplasia, or mitotic figures, as usually seenin HIVAN, were observed. The interstitium was focallyedematous, with only rare inflammatory cells. Interstitialfibrosis was unremarkable or mild. Vessels and glomeruliwere normal, except for four patients with mild to mod-erate nephroangiosclerosis lesions and one patient withmild diabetic nephropathy. In three patients exposed totenofovir, eosinophilic cytoplasmic inclusions of proximaltubular cells were associated with ATN (Figure 1, A–C). Elec-tron microscopy was performed in three other tenofovir-treated patients but without tubular cytoplasmic inclusionsby light microscopy, and it disclosed dysmorphic, and

sometimes abnormally enlarged, mitochondria in proxi-mal tubular cells (Figure 1D).Etiologic Diagnosis. Drug-related nephrotoxicity ac-

counted for 79.3% of tubulopathy cases, whereas it repre-sented 26.7% of IN cases (P,0.001) (Table 3). A singleculprit agent was identified in 23 cases. Tenofovir accountedfor the main offending agent (16 patients, 69.5%). Theremaining seven cases were caused by other antiviral drugs(two cidofovir cases and one didanosine case), antimicro-bial agents (one vancomycin case, one rifampicin case,and one amphotericin B case), or metformin in additionto tenofovir (one case). ATN caused by septic shock wasidentified in four patients. Two of them concomitantly re-ceived tenofovir and showed proximal tubular dysfunc-tion. One patient presented with duodenal immunoblasticplasmacytoid lymphoma responsible for myeloma cast ne-phropathy.Outcome. Renal follow-up data were available in 21

(72.4%) patients, with a median follow-up of 8 months(range=1–132). Complete recovery of renal function wasobserved in 6 (28.6%) patients, and stage III–V CKD was

Table 1. Baseline characteristics of 59 HIV-infected patients and biopsy-proven tubulointerstitial nephropathies

All Patients(n=59)

Tubulopathy(n=29, 49%)

Interstitial Nephritis(n=30, 51%)

Median age, yr (range) 45 (23–85) 44 (23–85) 45.5 (29–67)Men, n (%) 45 (76.3) 22 (75.8) 23 (76.7)Black, n (%) 16 (27.1) 6 (20.7) 10 (33.3)AIDS stage, n (%) 37 (62.7) 20 (69) 17 (56.7)Median CD4 count, cell/mL (range) 178.5 (1–974) 169 (1–729) 205 (6–974)CD4 count,200/mm3, n (%) 29 (58) 18 (64.3) 11 (50)Detectable HIV viral load, n (%) 27 (52.9) 15 (55.6) 12 (50)Median, cp/ml (range) 15.103 (94–10.106) 8.103 (200–10.106) 21.103 (94–39.104)

Opportunistic infections, n (%) 37 (62.7) 20 (69) 17 (56.7)Tuberculosis 11 (29.7) 4 (20) 7 (41.2)Atypical mycobacterial infection 7 (18.9) 3 (15) 4 (23.5)CMV disease 17 (45.9) 11 (55) 6 (35.3)Pneumocystosis 8 (21.6) 6 (30) 2 (11.8)Toxoplasmosis 3 (8.1) 3 (15) —Esophageal candidiasis 3 (8.1) 2 (10) 1 (5.9)Other 7 (18.9) 6 (30) 1 (5.9)

Malignancies, n (%)Kaposi sarcoma 5 (8.5) 3 (10.3) 2 (6.7)Multicentric Castleman disease 1 (1.7) — 1 (3.3)Lymphoma 2 (3.4) 1 (3.4) 1 (3.3)

Coinfection, n (%)Hepatitis B 12 (20.3) 7 (24.1) 5 (16.7)Hepatitis C 8 (13.6) 4 (13.8) 4 (13.3)

Exposure to ARV therapy, n (%) 50 (89.3) 27 (93.1) 23 (85.2)Ritonavir 37 (75.5) 24 (88.9) 13 (59.1)Lamivudine 33 (67.3) 17 (63) 16 (72.7)Tenofovir 29 (59.2) 21 (77.8) 8 (36.4)Abacavir 18 (36.7) 9 (33.3) 9 (40.9)Lopinavir 15 (30.6) 7 (25.9) 8 (36.4)Zidovudine 11 (22.4) 6 (22.2) 5 (22.7)Atazanavir 9 (18.4) 7 (25.9) 2 (9.1)Indinavir 8 (16.3) 3 (11.1) 5 (22.7)Stavudine 5 (10.2) 3 (11.1) 2 (9.1)Didanosine 5 (10.2) 4 (14.8) 1 (4.5)

CMV, cytomegalovirus; ARV, antiretroviral.

Clin J Am Soc Nephrol 8: ccc–ccc, June, 2013 Tubulointerstitial Nephropathies in HIV-Infected Patients, Zaidan et al. 3

seen in 15 (71.4%) patients, with 3 (14.3%) patients requir-ing long-term dialysis. Four (16%) patients, all presentingwith ATN, died within the first 1 month after kidney bi-opsy (three patients from sepsis and one patient from he-matologic malignancy).

Interstitial NephritisBaseline Characteristics. Summary and detailed char-

acteristics of 30 patients with IN are shown in Table 2 andSupplemental Table 2, respectively. Six patients showedacute fever at presentation without skin rash or eosinophilia.

Table 2. Renal parameters at presentation and outcome at last follow-up

All Patients(n=59)

Tubulopathy(n=29)

Interstitial Nephritis(n=30)

Renal parameters at kidney biopsyHypertension, n (%) 12 (24.5) 6 (24) 6 (25)Median peak sCr, mg/dl (range) 2.56 (0.57–16.33) 2.71 (0.84–16.33) 2.49 (0.57–13.8)Median estimated GFR, ml/minper 1.73 m2 (range)

26 (3–146) 25 (3–94) 29 (4–146)

Acute renal dysfunction, n (%)Risk 7 (11.8) 3 (10.3) 4 (13.3)AKI 26 (44) 9 (31) 17 (56.7)AKF 19 (32.2) 14 (48.3) 5 (16.7)

Acute-on-chronic renal failure, n (%) 11 (18.6) 6 (20.7) 5 (16.7)Renal replacement therapy, n (%) 12 (20.3) 9 (31) 3 (10)Median Pu/Cu, mg/g (range) 1414 (0–10,696) 1630 (0–10,696) 1330 (0–6541)Pu/Cu ratio, mg/g, n (%),442 9 (17.3) 4 (16.7) 5 (17.9)442–1326 13 (25) 5 (20.8) 8 (28.5)1326–2652 18 (34.6) 8 (33.3) 10 (35.7)$2652 12 (23.1) 7 (29.2) 5 (17.9)

Hematuria, n (%) 16 (30.8) 5 (20) 11 (40.7)Leukocyturia, n (%) 9 (20.9) 2 (9.1) 7 (33.3)

Renal outcome at last follow-upAvailable renal outcome, n (%) 44 (74.6) 21 (72.4) 23 (76.7)Median follow-up time, mo (range) 9.5 (0.5–156) 8 (1–132) 12 (0.5–156)Complete recovery, n (%) 19 (43.2) 6 (28.6) 13 (56.5)Stage III–V CKD, n (%) 24 (54.5) 15 (71.4) 9 (39.1)ESRD 3 (6.8) 3 (14.3) 0 (0)

Death, n (%) 5 (8.5) 4 (13.7) 1 (3.3)

sCr, serum creatinine level; AKF, acute kidney failure; Pu/Cu, proteinuria-to-creatininuria ratio.

Table 3. Etiologies of tubular and interstitial nephropathies in 59 HIV-infected patients

All Patients(n=59)

Tubulopathy(n=29)

Interstitial Nephritis(n=30)

Drugs, n (%) 31 (52.5) 23 (79.3): tenofovir, 16; cidofovir, 2;didanosine, 1; vancomycin, 1;rifampicin/IoCM, 1;amphotericin b, 1;metformin/tenofovir, 1

8 (26.7): indinavir, 2; foscarnet, 2;abacavir, 1; NSAIDs, 2;cotrimoxazol, 1

Infections, n (%) 9 (15.2) 4 (13.8): sepsis shock (2 withtenofovir)

5 (16.7): MAC, 3; tuberculosis, 1;hantavirus, 1

Dysimmune syndromes,n (%)

5 (8.5) — 5 (16.7): DILS, 3; IRIS, 2

Malignancies, n (%) 2 (3.4) 1 (3.4) IPL 1 (3.3) MCDATIN of undeterminedorigin, n (%)

6 (10.2) 1 (3.4) 5 (16.7)

CTIN, n (%) 6 (10.2) — 6 (20)

IoCM, iodinated contrast media; NSAIDs, nonsteroidal anti-inflammatory drugs; MAC,Mycobacterium avium Complex; DILS, diffuseinfiltrative lymphocytosis syndrome; IRIS, immune reconstitution inflammatory syndrome; IPL, immunoblastic plasmacytoid lym-phoma; MCD, multicentric Castleman disease; ATIN, acute tubulointerstitial nephropathy; CTIN, chronic tubulointerstitial ne-phropathy.



Twenty-two (73.3%) patients presented with AKI or AKF.Three (10%) patients required RRT at presentation. Fifteen(53.6%) patients had a UP/C.1326 mg/g. Hematuria andleukocyturia were observed in 40.7% and 33.3% of patients,respectively.Biopsy Findings and Etiologic Diagnosis. Six (20%)

patients presented with granulomatous IN related to My-cobacterium avium complex (three patients), M. tuberculosis(one patient), and undetermined origin (two patients) (Fig-ure 2A). Twenty-four (80%) patients had nongranuloma-tous IN. Drug-related IN was associated with crystalnephropathy and intratubular precipitation in four pa-tients caused by indinavir or foscarnet (Figure 2B),whereas immunoallergic IN was observed in four patientsrelated to nonsteroidal anti-inflammatory drugs (NSAIDs),abacavir, or cotrimoxazol. DILS was diagnosed in threepatients. The kidney biopsy specimen showed marked in-filtration by mononuclear cells, essentially CD8+ lympho-cytes, with tubulitis (Figure 2C). Two patients with activetuberculosis infection displayed IRIS. They presented withsystemic inflammatory symptoms, AKI, bilateral kidneyenlargement, and dense interstitial infiltrate composedpredominantly of CD68+ macrophages (Figure 2D). Inboth cases, IRIS occurred a few weeks after antituber-culosis and ARV initiation. The patients were treatedwith steroids and antimicrobial therapy, leading to an im-provement of the renal function. Two patients presentedwith IN related to multicentric Castleman disease andHantavirus infection, respectively. Finally, three patientspresented with acute IN of undetermined origin, and sixpatients displayed chronic TIN. In these cases, no etiologicdiagnosis was established. Particularly, we did not ob-serve prominent plasma cell interstitial infiltrates as recog-nized in the interstitial disease related to HIVAN.However, causative factors could be suspected on retro-spective chart review, including a past history of sepsisshock-related ATN, recurrent indinavir-related renal colic,administration of foscarnet a few months before kidneybiopsy, and tenofovir-induced AKI.Outcome. Renal follow-up data were available in 23

(80%) patients, with a median follow-up of 12 months(range=0.5–156). Renal function returned to baseline in 13(56.5%) patients. Dialysis was discontinued in three pa-tients who initially required RRT. Nine (39.1%) patientsdeveloped stage III–V CKD. One additional patient diedshortly after kidney biopsy.

DiscussionDuring the last 15 years, 59 of 222 (26.6%) HIV-infected

patients who underwent a kidney biopsy at our institution

Figure 1. | Light microscopic and ultrastructural findings in patientswith tubulopathy related to tenofovir nephrotoxicity. (A) Toxic acute

tubular necrosis affecting the proximal tubules that associated loss ofbrush border, cytoplasmic reduction, luminal enlargement of tubularsections, epithelial desquamation, and interstitial edema and fibrosis(trichrome stain, 3400). (B) A higher-power view showing swellingand vacuolization of proximal tubular cell cytosol (trichrome stain,3600). (C) Presence of red intracytoplasmic inclusions (arrows) inproximal tubular epithelial cells (trichrome stain, 31000). (D) Ul-trastructural examination showing abnormally enlarged and dys-morphic mitochondria within a proximal tubular epithelial cell.Original magnification, 38000.


showed either prominent tubular (13.1%) or interstitial(13.5%) lesions. These data are consistent with the pre-viously reported frequencies of IN (11%–20%) and biopsy-proven ATN (10%–16.7%) in this population, respectively(4,11,15,17–19). Compared with HIV-infected patients withhistopathological diagnosis of glomerular disease estab-lished between 1995 and 2007 (10), TIN patients wereless frequently of black origin (27% versus 63%), had abetter virological control (undetectable viral load: 47% ver-sus 28%), and had a higher percentage of received ARVtherapy (89.3% versus 73%). None of the patients withdrug-induced IN displayed the classic triad of fever, skinrash, and eosinophilia, which was previously underlinedin the series of Parkhie et al. (15). Strikingly, the high-gradeproteinuria observed in some patients is unusual for TIN.However, this finding was also pointed out in studies byParkhie et al. (15) and Herlitz et al. (20), which reported 29HIV-infected patients with acute IN and 13 HIV patientswith tenofovir-induced nephropathy, respectively (15,20).In the series by Parkhie et al. (15), median proteinuria was2.2 g/d (range=0.1–12.9), with 65.5% of patients present-ing with proteinuria.1.5 g/d and 31% of patients present-ing with proteinuria.3.5 g/d (15). Median level ofproteinuria was 1.5 g/d (range=1.2–2.1) in patients withtenofovir-induced nephropathy (20).Drug toxicity accounted for nearly 80% of tubulopathy

cases. ARV-related tubular toxicity was usually observedwith a nucleotide analog reverse transcription inhibitor,mainly tenofovir. Tenofovir may have contributed to ATNin three additional patients presenting with septic shock(two patients) and in conjunction with metformin (onepatient), because they all displayed proximal tubulardysfunction. Tenofovir has gained widespread use forthe treatment of HIV-1 and hepatitis B infections on thebasis of its efficacy and tolerability (21–24). Phase III stud-ies showed excellent renal safety, but cases of tenofovir-related nephrotoxicity were first reported in 2002 (12,25).Proximal tubular dysfunction was then further character-ized, and sometimes, overt Fanconi syndrome with renalfailure was characterized (19,25–32). In our study, almost90% and 40% of patients with tenofovir nephrotoxicitydisplayed proximal tubular dysfunction and overt Fanconi’ssyndrome, respectively. Similar findings were noted in pa-tients with cidofovir- and didanosine-related tubulopathy.Most of the patients had at least one risk factor for teno-fovir nephrotoxicity: 50% of patients were more than

Figure 2. | Light microscopic findings in patients with interstitialnephritis. (A) Granulomatous interstitial nephritis associated with

tuberculosis characterized by epithelioid granulomas with multinu-cleated giant cells (arrows) and necrosis (hematoxylin and eosin,3200). (B) Foscarnet-related crystal nephropathy. The glomerular tuftis partially obliterated by clear foscarnet crystals (trichrome stain,3400). Some crystals were also seen within tubular lumens, and therenal interstitium showed mild to moderate interstitial fibrosis andtubular atrophy. (C) Interstitial nephritis associated with diffuse in-filtrative lymphocytosis syndrome characterized by dense cellularinfiltrates composed of lymphocytes, monocytes, and plasma cells(hematoxylin and eosin,3200). Immunophenotyping analysis showsthat lymphocytes were mainly CD3+ and CD8+ T cells (Inset,3200).(D) Interstitial nephritis related to immune reconstitution in-flammatory syndrome (trichrome stain, 3200). The interstitial in-flammatory infiltrate was mainly composed by CD68+ macrophages(Inset, 3400).


50 years old, 50% of patients had baseline eGFR under 90ml/min per 1.73 m2, 25% of patients had stage III/IVCKD, and all but one patients also received ritonavir-boosted ARV regimen (31). Recently, the Columbia groupreported pathologic findings in 13 HIV-infected patientswith tenofovir nephrotoxicity (20,30). Histologic patternassociated toxic ATN and eosinophilic intracytoplasmicinclusions that corresponded with giant mitochondriawithin proximal cells (20). Dysmorphic mitochondriawere also observed by electron microscopy (20). Similarintracytoplasmic inclusions were observed in 18.8% ofour cases of tenofovir-related tubulopathy (n=3), and ul-trastructural analysis performed in three biopsy specimensdisclosed dysmorphic and giant mitochondria. Proximaltubular dysfunction usually improved after tenofovirwithdrawal. However, only 40% of patients achieved com-plete renal recovery as previously reported (20,31,33). TheIN group was associated with a broader spectrum of path-ologic lesions and causes. Compared with the tubulopathygroup, IN patients tended to have a better renal and over-all outcome, maybe because of a less severe renal dysfunc-tion at presentation, although differences in eGFR were notstatistically significant. Drug toxicity was a significantcause of IN. Antiviral therapy, including indinavir andfoscarnet, was associated with crystal nephropathy (34–37). No case of atazanavir-related IN was identified in con-trast to recent reports (38,39). Classic forms of acuteIN were also observed with NSAIDs and cotrimoxazol.Nevertheless, drug-related IN only accounted for 26.7%of cases compared with 72% of acute IN cases in the JohnsHopkins’ series (15). Antiviral agents were involved in62.5% of drug-induced IN in our series, whereas NSAIDswere the main culprits in the Johns Hopkins’ series (15).Such discrepancy may be explained by over-the-counteravailability of NSAIDs in the United States and specificdemographic characteristics, with predominance of blackrace (79%), high prevalence of hepatitis C coinfection(62%), and low rate of ARV therapy (approximately50%) in North American patients (15). In our study, only27.1% of patients were of black origin, the rate of hepatitisC virus infection was 13.6%, and about 90% of patientsreceived ARV.We found that mycobacterial infections remained an

important cause of granulomatous IN. We also identifiedIRIS and DILS as specific causes of acute IN. IRIS is relatedto rapid immune restoration after ARV initiation in patientswith a previously treated or subclinical opportunisticinfection (39–43). It manifests as systemic inflammationand various organ involvement, including kidneys (40–43). Tissue infiltrates are predominantly composed ofCD68+ macrophages. DILS usually associates CD8+ lym-phocytosis and CD8+ lymphocytic visceral infiltration,predominantly affecting the salivary glands, lungs, andless frequently, other organs, including the kidney in lessthan 10% of cases (44–48). HIV patients with IRIS or DILSrespond to corticosteroid therapy (49,50). In about onethird of cases of acute or chronic IN, no definitive etiologycould be established, although some of the patients pre-sented past exposition to cumulative risk factors for de-veloping chronic interstitial lesions. Of note, recent studiessuggested that HIVAN might present with subtle glomer-ular changes (i.e., podocyte hypertrophy and hyperplasia)

and prominent tubulointerstitial lesions with tubular mi-crocysts and plasma cell interstitial infiltration (51–53).Such findings were not observed in our series.Our study has some limitations. First, the prevalence of

TIN in the whole HIV population is underestimated,because the study was only based on biopsy-proven cases.Although the indications for kidney biopsy included acuteor chronic renal failure and/or persistent proteinuria and/or hematuria, patients with mild deterioration of renalfunction and/or mild proteinuria were unlikely to bebiopsied. Additionally, only patients with a main patho-logic diagnosis of TIN were included, whereas patientswith TIN lesions secondary to glomerular or vasculardisease were not analyzed. The high percentage of prox-imal tubular dysfunction among patients with tenofovirtoxicity may also represent a referral bias. Finally, patientswere referred from two Infectious Disease Units that arenot representative of the European HIV cohorts becauseof a higher proportion of black patients.In summary, we report a series of French HIV-infected

patients, focusing on biopsy-proven tubulopathy and INover the past 15 years. Both pathologic entities are closelyassociated with specific etiologies. Drug-related nephrotox-icity, mostly caused by ARV therapy, is the leading causeof TIN, underscoring the importance of a careful renalmonitoring in HIV-infected patients under ARV therapy.Tenofovir accounts for the main cause of tubulopathy. Be-cause proximal tubular dysfunction is usually observedin cases of tenofovir nephrotoxicity, renal biopsy may bevaluable in patients with atypical presentation, particularclinical characteristics (i.e., hepatitis B coinfection and lim-ited ARV alternatives), or lack of recovery after tenofovirwithdrawal. In IN, renal pathologic analysis and immuno-phenotyping of inflammatory infiltrates are useful to rec-ognize granulomatous forms and kidney involvementrelated to DILS or IRIS. Given the wide range of renal lesionsand atypical presentation, kidney biopsy remains a key stepin the accurate diagnosis of TIN in HIV-infected patients.

DisclosuresNone.

References1. Baker RJ, Pusey CD: The changing profile of acute tubulointer-

stitial nephritis. Nephrol Dial Transplant 19: 8–11, 20042. Praga M, Gonzalez E: Acute interstitial nephritis. Kidney Int 77:

956–961, 20103. Davison AM, Jones CH: Acute interstitial nephritis in the elderly:

A report from the UK MRC Glomerulonephritis Register and areview of the literature. Nephrol Dial Transplant 13[Suppl 7]:12–16, 1998

4. Wyatt CM, Morgello S, Katz-Malamed R, Wei C, Klotman ME,Klotman PE, D’Agati VD: The spectrum of kidney disease in pa-tients with AIDS in the era of antiretroviral therapy. Kidney Int 75:428–434, 2009

5. Rao TK, Filippone EJ, Nicastri AD, Landesman SH, Frank E, ChenCK, Friedman EA: Associated focal and segmental glomerulo-sclerosis in the acquired immunodeficiency syndrome. N Engl JMed 310: 669–673, 1984

6. Gardenswartz MH, Lerner CW, Seligson GR, Zabetakis PM,Rotterdam H, Tapper ML, Michelis MF, Bruno MS: Renal diseasein patients with AIDS: A clinicopathologic study. Clin Nephrol21: 197–204, 1984

7. Pardo V, Aldana M, Colton RM, Fischl MA, Jaffe D, Moskowitz L,Hensley GT, Bourgoignie JJ: Glomerular lesions in the acquired


immunodeficiency syndrome. Ann Intern Med 101: 429–434,1984

8. Rachakonda AK, Kimmel PL: CKD in HIV-infected patients otherthan HIV-associated nephropathy. Adv Chronic Kidney Dis 17:83–93, 2010

9. Lucas GM, Eustace JA, Sozio S, Mentari EK, Appiah KA, MooreRD: Highly active antiretroviral therapy and the incidence ofHIV-1-associated nephropathy: A 12-year cohort study. AIDS 18:541–546, 2004

10. Lescure FX, Flateau C, Pacanowski J, Brocheriou I, Rondeau E,Girard PM, Ronco P, Pialoux G, Plaisier E: HIV-associated kidneyglomerular diseases: Changes with time and HAART. NephrolDial Transplant 27: 2349–2355, 2012

11. Berliner AR, Fine DM, Lucas GM, Rahman MH, Racusen LC,Scheel PJ, Atta MG: Observations on a cohort of HIV-infectedpatients undergoing native renal biopsy. Am J Nephrol 28: 478–486, 2008

12. Coca S, PerazellaMA: Rapid communication: Acute renal failureassociated with tenofovir: Evidence of drug-induced nephro-toxicity. Am J Med Sci 324: 342–344, 2002

13. Izzedine H, Harris M, Perazella MA: The nephrotoxic effects ofHAART. Nat Rev Nephrol 5: 563–573, 2009

14. Atta MG, Deray G, Lucas GM: Antiretroviral nephrotoxicities.Semin Nephrol 28: 563–575, 2008

15. Parkhie SM, Fine DM, Lucas GM, Atta MG: Characteristics ofpatients with HIV and biopsy-proven acute interstitial nephritis.Clin J Am Soc Nephrol 5: 798–804, 2010

16. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P: AcuteDialysis Quality Initiative Workgroup: Acute renal failure—definition, outcome measures, animal models, fluid therapy andinformation technology needs. Presented at The Second In-ternational Consensus Conference of the Acute Dialysis QualityInitiative (ADQI) Group. Crit Care 8: R204–R212, 2004

17. Nochy D, Glotz D, Dosquet P, Pruna A, Guettier C, Weiss L,Hinglais N, Idatte JM, Mery JP, Kazatchkine M, Druet P, Bariety J:Renal disease associatedwith HIV infection: Amulticentric studyof 60 patients from Paris hospitals. Nephrol Dial Transplant 8:11–19, 1993

18. Peraldi MN, Maslo C, Akposso K, Mougenot B, Rondeau E, SraerJD: Acute renal failure in the course of HIV infection: A single-institution retrospective study of ninety-two patients anad sixtyrenal biopsies. Nephrol Dial Transplant 14: 1578–1585, 1999

19. Szczech LA, Gupta SK, Habash R, Guasch A, Kalayjian R, AppelR, Fields TA, Svetkey LP, Flanagan KH, Klotman PE, Winston JA:The clinical epidemiology and course of the spectrum of renaldiseases associated with HIV infection. Kidney Int 66: 1145–1152, 2004

20. Herlitz LC, Mohan S, Stokes MB, Radhakrishnan J, D’Agati VD,Markowitz GS: Tenofovir nephrotoxicity: Acute tubular necrosiswith distinctive clinical, pathological, and mitochondrial ab-normalities. Kidney Int 78: 1171–1177, 2010

21. Jimenez-Nacher I, Garcıa B, Barreiro P, Rodriguez-Novoa S,Morello J, Gonzalez-Lahoz J, deMendoza C, SorianoV: Trends inthe prescription of antiretroviral drugs and impact on plasmaHIV-RNAmeasurements. J Antimicrob Chemother 62: 816–822,2008

22. Nelson MR, Katlama C, Montaner JS, Cooper DA, Gazzard B,Clotet B, Lazzarin A, Schewe K, Lange J, Wyatt C, Curtis S, ChenSS, Smith S, Bischofberger N, Rooney JF: The safety of tenofovirdisoproxil fumarate for the treatment of HIV infection in adults:The first 4 years. AIDS 21: 1273–1281, 2007

23. Cooper RD, Wiebe N, Smith N, Keiser P, Naicker S, Tonelli M:Systematic review and meta-analysis: Renal safety of tenofovirdisoproxil fumarate in HIV-infected patients. Clin Infect Dis 51:496–505, 2010

24. Izzedine H, Hulot JS, Vittecoq D, Gallant JE, Staszewski S,Launay-Vacher V, ChengA,DerayG; Study 903 Team: Long-termrenal safety of tenofovir disoproxil fumarate in antiretroviral-naive HIV-1-infected patients. Data from a double-blind ran-domized active-controlled multicentre study. Nephrol DialTransplant 20: 743–746, 2005

25. Verhelst D, Monge M, Meynard JL, Fouqueray B, Mougenot B,Girard PM, Ronco P, Rossert J: Fanconi syndrome and renalfailure induced by tenofovir: A first case report. Am J Kidney Dis40: 1331–1333, 2002

26. Rifkin BS, Perazella MA: Tenofovir-associated nephrotoxicity:Fanconi syndrome and renal failure. Am J Med 117: 282–284,2004

27. Izzedine H, Isnard-Bagnis C, Hulot JS, Vittecoq D, Cheng A,Jais CK, Launay-Vacher V, Deray G: Renal safety of tenofovir inHIV treatment-experienced patients. AIDS 18: 1074–1076,2004

28. Creput C, Gonzalez-Canali G, Hill G, Piketty C, Kazatchkine M,Nochy D: Renal lesions in HIV-1-positive patient treated withtenofovir. AIDS 17: 935–937, 2003

29. Karras A, Lafaurie M, Furco A, Bourgarit A, Droz D, Sereni D,Legendre C, Martinez F, Molina JM: Tenofovir-related nephro-toxicity in human immunodeficiency virus-infected patients:Three cases of renal failure, Fanconi syndrome, and nephrogenicdiabetes insipidus. Clin Infect Dis 36: 1070–1073, 2003

30. Perazella MA: Tenofovir-induced kidney disease: An acquiredrenal tubular mitochondriopathy. Kidney Int 78: 1060–1063,2010

31. Hall AM,HendryBM,NitschD, Connolly JO: Tenofovir-associatedkidney toxicity in HIV-infected patients: A review of the evidence.Am J Kidney Dis 57: 773–780, 2011

32. WoodwardCL,Hall AM,Williams IG,Madge S, Copas A,NairD,Edwards SG, Johnson MA, Connolly JO: Tenofovir-associatedrenal and bone toxicity. HIV Med 10: 482–487, 2009

33. Zimmermann AE, Pizzoferrato T, Bedford J, Morris A, Hoffman R,Braden G: Tenofovir-associated acute and chronic kidney dis-ease: A case of multiple drug interactions. Clin Infect Dis 42:283–290, 2006

34. Said SM, Nasr SH, Samsa R, Markowitz GS, D’Agati VD:Nephrotoxicity of antiretroviral therapy in an HIV-infected pa-tient. Kidney Int 71: 1071–1075, 2007

35. Beaufils H,DerayG, KatlamaC,Dohin E, HeninD, SazdovitchV,Jouanneau C: Foscarnet and crystals in glomerular capillary lu-mens. Lancet 336: 755, 1990

36. Maurice-Estepa L, Daudon M, Katlama C, Jouanneau C,Sazdovitch V, Lacour B, Beaufils H: Identification of crystals inkidneys of AIDS patients treated with foscarnet. Am J Kidney Dis32: 392–400, 1998

37. DaudonM, Estepa L, Viard JP, Joly D, Jungers P: Urinary stones inHIV-1-positive patients treatedwith indinavir. Lancet 349: 1294–1295, 1997

38. Brewster UC, Perazella MA: Acute interstitial nephritis associ-ated with atazanavir, a new protease inhibitor. Am J Kidney Dis44: e81–e84, 2004

39. Izzedine H, M’rad MB, Bardier A, Daudon M, Salmon D:Atazanavir crystal nephropathy. AIDS 21: 2357–2358, 2007

40. Shelburne SA 3rd, Hamill RJ, Rodriguez-Barradas MC,Greenberg SB, Atmar RL, Musher DW, Gathe JC Jr, VisnegarwalaF, Trautner BW: Immune reconstitution inflammatory syndrome:Emergence of a unique syndrome during highly active anti-retroviral therapy. Medicine (Baltimore) 81: 213–227, 2002

41. French MA, Price P, Stone SF: Immune restoration disease afterantiretroviral therapy. AIDS 18: 1615–1627, 2004

42. French MA: HIV/AIDS: Immune reconstitution inflammatorysyndrome: A reappraisal. Clin Infect Dis 48: 101–107, 2009

43. Muller M, Wandel S, Colebunders R, Attia S, Furrer H, Egger M;IeDEA Southern and Central Africa: Immune reconstitution in-flammatory syndrome in patients starting antiretroviral therapyfor HIV infection: A systematic review and meta-analysis. LancetInfect Dis 10: 251–261, 2010

44. Solal-Celigny P, Couderc LJ, Herman D, Herve P, Schaffar-Deshayes L, Brun-Vezinet F, Tricot G, Clauvel JP: Lymphoid in-terstitial pneumonitis in acquired immunodeficiency syndrome-related complex. Am Rev Respir Dis 131: 956–960, 1985

45. Itescu S, Brancato LJ, Winchester R: A sicca syndrome in HIVinfection: Association with HLA-DR5 and CD8 lymphocytosis.Lancet 2: 466–468, 1989

46. Itescu S, Winchester R: Diffuse infiltrative lymphocytosis syn-drome: A disorder occurring in human immunodeficiency virus-1 infection thatmay present as a sicca syndrome. RheumDis ClinNorth Am 18: 683–697, 1992

47. Williams FM, Cohen PR, Jumshyd J, Reveille JD: Prevalence ofthe diffuse infiltrative lymphocytosis syndrome among humanimmunodeficiency virus type 1-positive outpatients. ArthritisRheum 41: 863–868, 1998


48. Basu D, Williams FM, Ahn CW, Reveille JD: Changing spectrumof the diffuse infiltrative lymphocytosis syndrome. ArthritisRheum 55: 466–472, 2006

49. Abdool Karim SS, Naidoo K, Grobler A, Padayatchi N, Baxter C,GrayA,Gengiah T,NairG, Bamber S, SinghA, KhanM, Pienaar J,El-Sadr W, Friedland G, Abdool Karim Q: Timing of initiation ofantiretroviral drugs during tuberculosis therapy. N Engl J Med362: 697–706, 2010

50. AbdoolKarimSS,NaidooK,GroblerA, PadayatchiN, BaxterC,GrayAL, Gengiah T, Gengiah S, Naidoo A, JithooN, Nair G, El-SadrWM,Friedland G, Abdool Karim Q: Integration of antiretroviral therapywith tuberculosis treatment. N Engl J Med 365: 1492–1501, 2011

51. WearneN, Swanepoel CR, Boulle A,DuffieldMS, Rayner BL: Thespectrum of renal histologies seen in HIV with outcomes, prog-nostic indicators and clinical correlations. Nephrol Dial Trans-plant 27: 4109–4118, 2012

52. Arendse C, Okpechi I, Swanepoel C: Acute dialysis in HIV-positive patients in Cape Town, South Africa. Nephrology(Carlton) 16: 39–44, 2011

53. Swanepoel CR, Wearne N, Duffield MS, Okpechi IG: The evo-lution of our knowledge of HIV-associated kidney disease inAfrica. Am J Kidney Dis 60: 668–678, 2012

Received: October 1, 2012 Accepted: January 14, 2013

Published online ahead of print. Publication date available at www.cjasn.org.

This article contains supplemental material online at http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.10051012/-/DCSupplemental.


http://www.cjasn.org

http://www.cjasn.org




cjasn epress article tubulointerstitial nephropathies...

Documents