Abstract
Introduction
Sub-Saharan Africa remains challenged by the highest burden of human immunodeficiency virus (HIV), an epidemic of tuberculosis (TB), and increasing number of people with HIV (PWH) on antiretroviral therapy (ART), all of which may result in kidney injury.
Methods
This observational cohort study describes the spectrum of kidney disease in PWH in South Africa, between 2005 and 2020. Kidney biopsies were analyzed in 4 time periods as follows: early ART rollout (2005–2009), tenofovir disoproxil (TDF) introduction (2010–2012), TDF-based fixed dose combination (2013–2015), and ART at HIV diagnosis (2016–2020). Logistic regression was used to identify factors associated with HIV-associated nephropathy or focal segmental glomerulosclerosis (HIVAN/FSGS) and tubulointerstitial disease (TID).
Results
We included 671 participants (median age 36, interquartile range, 21–44 years; 49% female; median CD4 cell count 162 [interquartile range, 63–345] cells/mm3). Over time, ART (31%–65%, P < 0.001), rate of HIV suppression (20%–43%, P < 0.001), nonelective biopsies (53%–72%, P < 0.001), and creatinine at biopsy (242–449 μmol/l, P < 0.001) increased. A decrease in HIVAN (45%–29% P < 0.001) was accompanied by an increase in TID (13%–33%, P < 0.001). Granulomatous interstitial nephritis accounted for 48% of TID, mostly because of TB. Exposure to TDF was strongly associated with TID (adjusted odds ratio 2.99, 95% confidence interval 1.89–4.73 P < 0.001).
Conclusion
As ART programs intensified and increasingly used TDF, the spectrum of kidney histology in PWH evolved from a predominance of HIVAN in the early ART era to TID in recent times. The increase in TID is likely due to multiple exposures that include TB, sepsis, and TDF as well as other insults.
Keywords: HIV, HIV-associated nephropathy, kidney failure, kidney histology, tuberculosis, tubulointerstitial disease
Graphical abstract
Kidney failure is a life-threatening complication of HIV infection in Sub-Saharan Africa. Because of the high cost of kidney replacement therapy (KRT), access to chronic dialysis is severely restricted in most health care systems in this region.1 In South Africa, which has severely restricted KRT programs in most provinces, suitability for kidney transplantation is the overarching criterion for eligibility for chronic dialysis.2,3 Before 2008, PWH were not accepted onto chronic dialysis programs in most state sector hospitals because of the perceived poor outcomes associated with transplantation. A pivotal program to utilize kidneys from HIV-positive donors for HIV-positive recipients was commenced in September 2008. Once this program was commenced, it promoted the acceptance of PWH onto state dialysis programs.4 By 2018, the South African renal registry reported that 981 (11%) of 8948 patients requiring KRT were HIV-positive.5
HIVAN and TID are major kidney diagnoses, especially in the setting of advanced immunodeficiency, uncontrolled HIV viremia, and opportunistic infections, such as TB.6 HIVAN particularly affects populations of West African ancestry in whom genetic variants of APOL1 are highly prevalent,7 whereas TID reflects the effects of infection, drugs, and systemic illness on the kidney.8,9 Immune complex mediated glomerulonephritis (ICGN), another important cause of kidney disease, also disproportionally affects Black populations with uncontrolled HIV.10 The scale up of ART in Sub-Saharan Africa in the past 15 years is therefore likely to have averted many cases of HIVAN, TID, ICGN, and acute kidney injury thereby preventing end-stage kidney disease.11
Several studies have reported a change in the epidemiology and spectrum of kidney disease in PWH following the widespread introduction of ART. An early study from the United States of America reported a marked decline in the proportion of kidney biopsies demonstrating HIVAN from 1998 onwards when combination ART became standard of care,12 and a similar decline in the proportion of kidney biopsies showing HIVAN from 2004 onwards was reported in a French study.13 In South Africa, a substantial reduction in the proportion of biopsies showing HIVAN has been reported in recent years in 2 centers, with increases in the median CD4 cell count and/or the proportion of people receiving ART at the time of biopsy.14,15 The aim of this study was to expand on these observations and describe changes in other kidney pathologies as well as kidney outcomes and mortality in a large cohort of PWH in South Africa who underwent a kidney biopsy from 2005 to 2020 in the Western Cape Province.
Methods
In this large observational cohort study, we identified PWH who were referred to the nephrology service at Groote Schuur Hospital, Cape Town, South Africa for clinical evaluation and required a kidney biopsy between January 2005 and December 2020. The departmental biopsy registry was used to establish the study cohort and biopsies were performed for clinical indications (e.g., unexplained kidney impairment, proteinuria, and/or hematuria) as per best clinical practice. Kidney biopsies in HIV-positive transplant recipients, those younger than 18 years, as well as inadequate biopsies were excluded. Approval was obtained from the University of Cape Town Human Research Ethics Committee (UCT HREC: R048/2019 & 109/2020).
Kidney biopsies were processed for light microscopy with hematoxylin and eosin, methenamine silver, and periodic acid-Schiff stains. Standard techniques were performed to demonstrate immunohistochemistry (IgA, IgM, IgG, C3, and C1q once available from 2016). Biopsies were also processed for electron microscopy. All kidney biopsy specimens were reviewed by a kidney histopathologist (BP) and a nephrologist (NW), in accordance with the HIV KDIGO classification.6 The dominant pathologic finding (primary diagnosis) and up to 2 additional diagnoses (secondary and tertiary) were recorded. A primary diagnosis was made when a single dominant process was identified, which was thought to contribute significantly to the clinical presentation. A secondary diagnosis was made when an additional pathology was identified that was contributing to the clinical presentation, particularly TID and hypertension. A tertiary diagnosis was made if a pathological process was present in the biopsy that was either focal (e.g., minimal acute tubular injury [ATI]) or if its presence was not dominating could still contribute to overall kidney survival. Biopsies showing HIVAN or primary FSGS were grouped together because both share pathological and clinical characteristics and are strongly associated with APOL1 genetic variants.7
Granulomatous interstitial nephritis was defined by the presence of epithelioid granulomas in the kidney interstitium with or without necrosis or multinucleated giant cells.16,17 ICGN was diagnosed as per KDIGO guidelines.18 We defined the ball-in-cup phenomenon in accordance with the initial description that is, large subepithelial deposits associated with a basement membrane reaction resulting in a “ball-in-cup” architectural pattern.19,20 Lupus-like glomerulonephritis was defined as a diffuse proliferative glomerulonephritis resembling class IV lupus nephritis in the absence of clinical or serological evidence of systemic lupus erythematosus. ATI associated with TDF nephrotoxicity was attributed as a diagnosis if there was temporal association with exposure to TDF, proximal tubular injury, prominent eosinophilic intracytoplasmic inclusions, and when performed, dysmorphic mitochondria on electron microscopy. The improvement of kidney function on TDF withdrawal also supported the diagnosis.21
The following data were collected within 2 weeks of the kidney biopsy: demographics (age, sex, and ethnicity) and comorbidities (diabetes, hypertension, and TB, based on the use of medications to treat these conditions at the time of biopsy or a diagnostic test confirming TB). HIV parameters (HIV RNA [copies/ml], CD4 count [cells/mm3], and ART history); kidney function data (creatinine [μmol/l], estimated glomerular filtration rate (eGFR) calculated using the CKD-EPI race neutral equation [ml/min per 1.73 m2],22 urinary protein-to-creatinine ratio [mg/mmol]); and results of urine microscopy, culture, kidney size on ultrasound, and requirement of dialysis were obtained. The following blood results were obtained: hemoglobin (g/l), albumin (g/l), hepatitis B surface antigen, antihepatitis C antibodies, antinuclear antibody, antidouble-stranded DNA, venereal disease research laboratory test, antistreptolysin titre, antiDNase B titre, antineutrophil cytoplasmic antibody, and complement. Exposure to medications with nephrotoxic potential, including TDF, rifampicin, co-trimoxazole, methamphetamines, and use of traditional medicines were recorded. ART status at the time of biopsy was categorized as receiving ART (with or without TDF), ART-naïve, or having defaulted from HIV care or ART. All participants were followed up in a dedicated HIV kidney clinic, with regular assessments of kidney function and ART status. Dates of commencement of chronic KRT were recorded. Mortality was based on information from clinical records and/or interviews with family members.
Data Analysis
We stratified participants into 4 periods, which were chosen to coincide with early ART rollout (2005–2009), introduction of TDF (2010–2012), preferred use of a TDF-based fixed dose combination (TDF/emtricitabine/efavirenz) (2013–2015), and when ART was commenced at HIV diagnosis regardless of CD4 count with a TDF-containing combination (2016–2020). We described and compared the clinical characteristics and kidney histological patterns during the 4 periods using Wilcoxon rank-sum test for continuous data and Pearson’s chi-squared test or Fisher’s exact test for categorical data.
We used multivariable logistic regression to identify factors including each of the 4 time periods that were associated with a histological diagnosis of HIVAN/FSGS or TID. In these analyses, factors significantly associated (P < 0.05) with either outcome in univariable analysis were retained in the multivariable models. We used a composite end point of mortality and end-stage kidney disease requiring chronic KRT to estimate survival probability using Kaplan-Meier survival analysis. Survival time was calculated from the time of kidney biopsy to the date of death or initiation of chronic KRT, whichever occurred first. Participants who did not experience the composite outcome were censored 5 years post-biopsy. Participants lost to follow-up were censored at their last clinic visit. We used log-rank tests to compare the survival probabilities among the 4 time periods. A Cox proportional hazards model was used to evaluate the factors associated with mortality and end-stage kidney disease requiring chronic KRT and restricted to participants with an eGFR <15 ml/min per 1.73 m2. We tested the proportional hazards assumption by analyzing the correlation between Schoenfeld residuals and survival time. The proportional hazards assumption was met for all covariates. All analyses were performed in Stata (Stata Statistical Software: Release 16, StataCorp LLC, College Station, TX).
Results
Between January 2005 and December 2020, 695 HIV-positive adults underwent a kidney biopsy. This was 19% of a total 3663 biopsies performed during the study period. Twenty-four (3%) were inadequate, thus 671 biopsies were included in the analyses. Electron microscopy was performed in 82% of biopsies. The median age of the participants was 36 (interquartile range, 31−44) years, 49% were women, and 32% were on dialysis at the time of biopsy. There were 275 of 652 (42%) who had a TB diagnosis at baseline; of these, 260 patients had disseminated TB. The clinical characteristics and the primary histological diagnoses are outlined in Table 1. The proportion of patients on ART and with suppressed HIV RNA increased over time, as did the proportion of nonelective biopsies. Participants who underwent biopsies in recent times had more severe kidney dysfunction. HIVAN and TID were the common primary diagnoses seen in 44% (293/671) and 25% (116/671) respectively. When the tubulointerstitial dominant compartment was further examined, granulomatous interstitial nephritis accounted for the greatest proportion 48% (79/166) with confirmed TB identified in 40%. ATI was the second most common primary pathology identified in this compartment. The full spectrum of primary diagnoses for the total cohort by compartment as per the KDIGO classification is further summarized in Table 2.
Table 1.
Demographic and clinical characteristics at time of kidney biopsy
| Characteristics | Total |
2005–2009 |
2010–2012 |
2013−2015 |
2016−2020 |
P-value |
|---|---|---|---|---|---|---|
| N = 671 | n = 166 | n = 155 | n = 129 | n = 221 | ||
| Age (yrs) IQR | 36 (31–44) | 33 (29–40) | 35 (30–42) | 37 (31–45) | 37 (33–45) | <0.001 |
| Sex-female | 331/671 (49%) | 87/166 (52%) | 87/155 (56%) | 58/129 (45%) | 99/221 (45%) | 0.100 |
| Black ethnicitya | 573/671 (85%) | 147/166 (89%) | 140/155 (90%) | 112/129 (87%) | 174/221 (79%) | 0.008 |
| Dialysis at biopsy | 209 (32%) | 50 (30%) | 41 (27%) | 39 (31%) | 79 (36%) | 0.29 |
| Creatinine (μmol/l) IQR | 351 (153–795) | 242 (118–669) | 297 (146–707) | 417 (180–814) | 449 (175–883) | <0.001 |
| eGFR (ml/min per 1.73 m2) | 16 (6–46) | 27 (8–59) | 20 (8–50) | 14 (6–36) | 12 (6–37) | <0.001 |
| uPCR (mg/mmol) | 310 (160–750) | 410 (180–900) | 290 (140–630) | 250 (150–560) | 357 (160–790) | 0.039 |
| Nonelective biopsy | 430/671 (64%) | 88/166 (53%) | 94/155 (61%) | 89/129 (69%) | 159/221 (72%) | <0.001 |
| CD4 (cells/mm3) | 162 (63–345) | 178 (64–330) | 156 (72–286) | 156 (64–352) | 162 (63–376) | 0.67 |
| ART status | <0.001 | |||||
| On ART | 353/665 (53%) | 52/166 (31%) | 80/153 (52%) | 79/128 (62%) | 142/218 (65%) | <0.001 |
| TDF regimen | 166/353 (47%) | 3/52 (6%) | 20/80 (25%) | 47/79 (59%) | 96/142 (68%) | <0.001 |
| Time since starting | 14 (4–86) | 14 (4–55) | 8 (3–63) | 23 (4–52) | 16 (6–156) | 0.05 |
| ART (wks) | ||||||
| ART naïve | 253/665 (38%) | 101/166 (61%) | 62/153 (41%) | 37/128 (29%) | 53/218 (24%) | |
| ART defaulted | 59/665 (9%) | 13/166 (8%) | 11/153 (7%) | 12/129 (9%) | 23/221 (10%) | |
| HIV viral load <200 c/ml | 138/486 (28%) | 24/122 (20%) | 16/105 (15%) | 23/84 (27%) | 75/175 (43%) | <0.001 |
| Hepatitis B | 63/612 (10%) | 16/147 (11%) | 17/141 (12%) | 11/111 (10%) | 19/213 (9%) | 0.79 |
| Hepatitis C | 6/480 (1%) | 2/132 (2%) | 1/107 (1%) | 2/82 (2%) | 1/159 (1%) | 0.61 |
| Hypertension | 159/662 (24%) | 30/165 (18%) | 35/155 (23%) | 31/123 (25%) | 63/219 (29%) | 0.11 |
| Diabetes mellitus | 43/665 (6%) | 10/165 (6%) | 5/154 (3%) | 10/126 (8%) | 18/220 (8%) | 0.21 |
| Tuberculosis | 275/652 (42%) | 65/163 (40%) | 67/154 (44%) | 62/122 (51%) | 81/213 (38%) | 0.1 |
| Disseminated TB | 151/275 (55%) | 29/65 (45%) | 28/67(42%) | 33/63 (52%) | 61/81 (75%) | |
| Albumin (g/l) | 25 (19–31) | 28 (20–32) | 27 (21–33) | 26 (20–29) | 23 (19–28) | <0.001 |
| Primary diagnoses | ||||||
| HIVAN | 293 (44%) | 90 (54%) | 85 (55%) | 54 (42%) | 64 (29%) | <0.001 |
| FSGS | 14 (2%) | 4 (2%) | 2 (1%) | 2 (2%) | 6 (3%) | 0.82 |
| -Primary (5) | ||||||
| -Secondary (9) | ||||||
| Immune complex GN | 114 (17%) | 30 (18%) | 19 (12%) | 22(17%) | 43 (19%) | 0.30 |
| Diabetic nephropathy | 24 (4%) | 7 (4%) | 4(3%) | 5 (4%) | 8 (4%) | 0.89 |
| Hypertensive nephropathy | 27 (4%) | 7 (4%) | 8 (5%) | 4 (3%) | 8 (4%) | 0.83 |
| Tubulointerstitial disease (TID) | 166 (25%) | 21 (13%) | 32 (21%) | 40 (31%) | 73 (33%) | <0.001 |
| Major categories of TID | ||||||
| GIN | 79/166 (48%) | 5/21 (24%) | 17/32 (53%) | 26/40 (65%) | 31/73 (42%) | |
| ATI | 45/166 (27%) | 7/21 (33%) | 5/32 (16%) | 8/40 (20%) | 25/73 (34%) | |
| ATI-TDF | 24/45 (53%) | 0/7 (0%) | 0/5 (0%) | 8/8 (100%) | 16/25 (64%) | |
| AIN/CIN | 28/166 (17%) | 9/21 (43%) | 5/32 (16%) | 5/40 (12%) | 9/73 (12%) | |
| Other TIDb | 14/166 (8%) | 0/21 (0%) | 5/32 (16%) | 1/40 (2%) | 8/73 (11%) | |
| Other pathologiesc | 33/671 (5%) | 7/166 (4%) | 5/155 (3%) | 2/129 (2%) | 19/221 (9%) | 0.018 |
AIN, acute interstitial nephritis; ART, antiretroviral therapy; ATI, acute tubular injury; CIN, chronic interstitial nephritis; eGFR, estimated glomerular filtration rate; FSGS, focal segmental glomerulosclerosis; GIN, granulomatous interstitial nephritis; GN, glomerulonephritis; HIVAN, human immunodeficiency virus associated nephropathy; IQR, interquartile range; TB, tuberculosis; TDF, tenofovir disoproxil fumarate; uPCR, urine protein creatinine ratio.
Most non-Black individuals were mixed-black race
Other TID = (N = 14) diffuse infiltrative lymphocytosis syndrome (n = 2); Direct renal parenchymal infection by pathogens (pyelonephritis) (n = 12).
Other Pathologies (N = 33) amyloidosis (n = 3), end-stage (n = 1), lymphoma(n = 4), TMA (n = 2) myeloma/ monoclonal gammopathy of undetermined significance(n = 3), minimal change (n = 1), mesangial proliferative GN (n = 4), nodular sclerosis (n = 3), pauci-immune (n = 7); nonspecific mesangial proliferation with tubulointerstitial nephritis associated with abnormal kappa/lamba ratio-not fulfilling the definition of light chain deposition disease (n = 5).
Table 2.
Spectrum of primary histological diagnoses in patients with HIV infection at Groote Schuur Hospital from 2005 to 2020 (N = 671)
| Total N = 671 | Diagnosis | Number (n) (%) |
|---|---|---|
| 1. Glomerular-dominant | ||
| a. Podocytopathies | n = 312 | |
| HIVAN | 293/671 (43.7%) | |
| FSGS (NOS) in the setting of HIV FSGS (secondary) (9) FSGS (primary) (5) |
14/671 (2.1%) | |
| Diffuse mesangial hypercellularity in the setting of HIV | 4/671 (0.6%) | |
| Minimal change | 1/671 (1.1%) | |
| b. Immune complex-mediated glomerular disease | 114 /671 (17%) | |
| Membranoproliferative GNa Hepatitis B surface antigen positive (2) Unidentified cause) (33) |
35/114 (30.7%) | |
| Post-Streptococcal GN | 28/114 (24.6%) | |
| Membranous nephropathy Hepatitis B surface antigen positive (1) Hepatitis C positive (1) |
15/114 (13.2%) | |
| Lupus nephritis | 15/114 (13.2%) | |
| Mesangial proliferative GN | 5/114 (4.4%) | |
| IgM nephropathy | 4/114 (3.5%) | |
| Lupus-like GN | 4/114 (3.5%) | |
| ICGN-NOS | 4/114 (3.5%) | |
| IgA nephropathy | 3/114 (2.6%) | |
| C1 Q nephropathy | 1/114 (0.9%) | |
| Other glomerular lesions | ||
| Pauci-immune GN | 7/671 (1.0%) | |
| Amyloidosisb All 3 were associated with post-TB bronchiectasis |
3/671 (0.4%) | |
| Nodular sclerosis | 3/671 (0.4%) | |
| Tubulointerstitial dominant | 166/671 (24.7%) | |
| Granulomatous interstitial nephritis | 79/166 (47.6%) | |
| TB | 56 | |
| Drugs (Bactrim = 4; NSAID = 3; Ceftriaxone = 1) | 8 | |
| Uncertain | 5 | |
| Yeast | 4 | |
| Pyelonephritis | 3 | |
| MOTT | 2 | |
| Vasculitis | 1 | |
| Acute tubular injury 24/45 (53%) were attributed to TDF |
45/166 (27.1%) | |
| Drug induced tubulointerstitial nephritis (other than antiretroviral therapy) Rifampicin (15), NSAIDs (3), Ceftriaxone (2), Cotrimoxazole (2),Rifabutin (1), Sodium valproate (2), Furosemide (1), Angiotensin receptor blocker (1), Uncertain (4) |
31/166 (18.7%) | |
| Direct renal parenchymal infection by pathogens (pyelonephritis) (ascending pyelonephritis) | 12/166 (7.2%) | |
| Chronic interstitial nephritis-no cause identified | 3/166 (1.8%) | |
| Diffuse infiltrative lymphocytosis syndrome | 2/166 (1.2%) | |
| Vascular-dominant | ||
| Hypertension | 19/671 (2.8%) | |
| Malignant hypertensionc | 8/671 (1.2%) | |
| Thrombotic microangiopathy Haemolytic uraemic syndrome (1) In setting of malignant hypertension (1) |
2/671 (0.3%) | |
| IV. Other pathologies in the setting of HIV | ||
| Diabetic nephropathy | 24/671 (4%) | |
| Myeloma cast nephropathyd (2)/MGUS (1) | 3/671 (0.4%) | |
| Nonspecific mesangial proliferation with tubulointerstitial nephritis associated with abnormal kappa/lamba ratio-not fulfilling the definition of light chain deposition disease | 5/671 (0.7%) | |
| Lymphoma | 4/671 (0.6%) | |
| End stage–no diagnosis determined | 1/671 (0.1%) | |
| HIVAN + ICGN | Ball-in-cup | 9/69 (13%) |
| Membranous Hepatitis B surface antigen positive 1 |
5/69 (7%) | |
| Membranoproliferative GN Hepatitis B surface antigen positive 1 |
5/69 (7%) | |
| IgM nephropathy | 3/69 (4%) | |
| IgA nephropathy | 1/69 (1%) | |
| Lupus nephritis | 1/69 (1%) | |
| Lupus-like GN | 1/69 (1%) |
AIN, acute interstitial nephritis; ART, antiretroviral therapy; ATI, acute tubular injury; CIN, chronic interstitial nephritis; eGFR, estimated glomerular filtration rate; FSGS, focal segmental glomerulosclerosis; GIN, granulomatous interstitial nephritis; GN, glomerulonephritis; HIVAN, human immunodeficiency virus associated nephropathy; IQR, interquartile range; TB, tuberculosis; TDF, tenofovir disoproxil fumarate; uPCR, urine protein creatinine ratio.
Lupus excluded.
Amyloidosis: pauci-cellular mesangial expansion with PAS positivity and a negative silver stain in the glomeruli, congo red positive staining, and yellow-green birefringence with polarized microscopy. Amyloid deposits were confirmed ultrastructurally with electron microscopy.
Malignant Hypertension: mucoid intimal hyperplasia of arterioles, concentric arteriolar laminar fibrosis (“onion-skinnning”), glomerular features of basement membrane thickening with ischemic wrinkling and glomerulosclerosis with or without arteriolar fragmented erythrocytes (schistocytes) within arterioles.
Myeloma cast nephropathy: eosinophilic dense luminal proteinaceous PAS-negative material visualized in tubules (“cracking artifact”). Light chain restriction was also performed. MGUS: positive kappa light chains and consistent bone marrow findings.
The proportion of biopsies with HIVAN/FSGS as a primary diagnosis progressively declined in the recent time periods and this was accompanied by an increase in the proportion of biopsies showing TID (Figure 1). There was little change over time in the other diagnostic categories. The decrease in HIVAN/FSGS was driven by a reduction in HIVAN; the proportion of biopsies showing FSGS remained stable throughout the study period (Table 1).
Figure 1.
Primary histological diagnoses by time-period. FSGS, focal segmental glomerulosclerosis; HIVAN, human immunodeficiency virus associated nephropathy; ICGN, Immune complex glomerulonephritis.
The distribution of primary, secondary, and tertiary histopathological diagnoses are described in Supplementary Table S1. The most common primary diagnoses were FSGS/HIVAN (93%), with the majority of these attributed to HIVAN (89%) and ICGN, which accounted for 60% of biopsies. TID was a primary diagnosis in 32%, secondary diagnosis in 49%, and tertiary diagnosis in 18% of biopsies. Kidney function and HIV parameters at the time of kidney biopsy for each of the primary diagnoses are provided in Table 3. Median CD4 counts were lowest in those with HIVAN/FSGS and TID (122 and 128 cells/mm3, respectively) and kidney function was most preserved in those with ICGN despite the presence of heavy proteinuria.
Table 3.
Kidney function, CD4 and HIV viral load at biopsy in most common histological diagnoses
| Primary Diagnosis | HIVAN/FSGS N = 307 |
ICGN N = 114 |
DM/HPT N = 51 |
TID N = 166 |
Other N = 33 |
|
|---|---|---|---|---|---|---|
| eGFR (ml/min per 1.73 m2) | Median (IQR) | 15 (6–42) | 45 (14–87) | 14 (5–34) | 11 (6–3) | 11 (7–32) |
| uPCR (mg/mmol) | Median (IQR) | 460 (230–980) n = 284 | 410 (170–860) n = 108 | 180 (90–420) n = 49 | 180 (110–320) n = 141 | 440 (150–1190) n = 29 |
| CD4 (cells/mm3) | Median (IQR) | 122 (52–258) n = 295 | 287 (128–487) n = 107 | 300 (154–458) n = 49 | 128 (52–292) n = 159 | 226 (113–448) n = 33 |
| HIV RNA (<200 copies/ml) | n (%) | 43 (20%) | 28 (34%) | 14 (35%) | 39 (33%) | 14 (52%) |
| On ART | n (%) | 138 (45%) | 63 (57%) | 22 (44%) | 106 (64%) | 24 (73%) |
ART, antiretroviral therapy; DM, diabetes mellitus; eGFR, estimated glomerular filtration rate; FSGS, focal segmental glomerulosclerosis; HIV RNA, human immunodeficiency virus RNA; HIVAN, human immunodeficiency virus associated nephropathy; HPT, hypertension; ICGN, immune complex glomerulonephritis; TID, tubulointerstitial disease; uPCR, urine protein creatinine ratio.
We examined whether the decline in HIVAN/FSGS and the increase in TID over time was explained by changes in the demographic or immune-virological characteristics of the participants using logistic regression analysis. In univariable analysis, a biopsy performed in the 2 most recent time periods was less likely to show HIVAN/FSGS (Table 4). Exposure to ART was associated with lower odds, whereas Black ethnicity and a CD4 cell count <200 cells/mm3 increased the odds of HIVAN/FSGS. The association between a biopsy performed in 2016 to 2020 and a primary diagnosis of HIVAN/FSGS persisted in the multivariable model with adjustments for age, gender, ethnicity, exposure to ART, and CD4 cell count. In univariable analysis for TID, a biopsy performed in the 2 most recent time periods and exposure to ART containing TDF was associated with increased odds, whereas higher eGFR was associated with lower odds of TID (Table 5). The association between a biopsy in recent time periods and a diagnosis of TID was attenuated after adjustment for age, gender, ART exposure, and eGFR, whereas a strong association with exposure to ART containing TDF persisted (adjusted odds ratio, 3.08 [95% confidence interval, 1.94–4.90]).
Table 4.
Factors associated with HIVAN/FSGS
| HIVAN/FSGS | Univariable |
Multivariable |
||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P-value | aOR | 95%CI | P-value | |
| Time period | ||||||
| 2005–2009 (ref) | 1 | 1 | ||||
| 2010–2012 | 0.98 | 0.63–1.52 | 0.93 | 1.08 | 0.67–1.74 | 0.75 |
| 2013–2015 | 0.59 | 0.37–0.93 | 0.03 | 0.66 | 0.40–1.09 | 0.107 |
| 2016–2020 | 0.36 | 0.23–0.54 | <0.001 | 0.45 | 0.28–0.70 | <0.001 |
| Age <40 yrs (ref) | 1 | 1 | ||||
| 40–50 yrs | 0.74 | 0.51–1.08 | 0.12 | 0.81 | 0.54–1.22 | 0.32 |
| >50 yrs | 0.49 | 0.29–0.82 | 0.01 | 0.52 | 0.30–0.93 | 0.03 |
| Gender [Female] | 1.35 | 1.00–1.83 | 0.05 | 1.43 | 1.02–2.00 | 0.04 |
| Ethnicity [African] | 2.35 | 1.49–3.78 | <0.001 | 2.03 | 1.21–3.39 | 0.01 |
| On ART | 0.56 | 0.41–0.76 | <0.001 | 0.62 | 0.44–0.87 | 0.01 |
| CD4 (<200 cells/mm3) | 2.18 | 1.58–3.00 | <0.001 | 2.20 | 1.57–3.09 | <0.001 |
| eGFR (ref: <15 ml/min per 1.73 m2) | 1 | |||||
| eGFR 15–29 | 0.90 | 0.57–1.40 | 0.63 | |||
| eGFR 30–59 | 1.16 | 0.77–1.75 | 0.47 | |||
| eGFR ≥60 | 0.64 | 0.41–0.99 | 0.05 | |||
ART, antiretroviral therapy; aOR, adjusted odds ratio; CI, confidence interval; eGFR, estimated glomerular filtration rate; FSGS, focal segmental glomerulosclerosis; HIVAN, human immunodeficiency virus associated nephropathy; OR, odds ratio.
Table 5.
Factors associated with tubulointerstitial disease
| Tubulointerstitial disease | Univariable |
Multivariable |
||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P-value | aOR | 95% CI | P-value | |
| Time frame | ||||||
| 2005–2009 (ref) | 1 | 1 | ||||
| 2010–2012 | 1.79 | 0.99–3.27 | 0.056 | 1.60 | 0.86–2.97 | 0.13 |
| 2013−2015 | 3.10 | 1.72–5.6 | <0.001 | 1.93 | 1.02–3.66 | 0.04 |
| 2016−2020 | 3.41 | 1.99–5.82 | <0.001 | 1.97 | 1.09–3.54 | 0.02 |
| Age <40 yr (ref) | ||||||
| 40−50 | 1.23 | 0.81–1.89 | 0.330 | 0.98 | 0.62–1.54 | 0.92 |
| >50 | 1.63 | 0.96–2.76 | 0.068 | 1.49 | 0.85–2.61 | 0.17 |
| Gender (ref: female) | 0.66 | 0.46–0.94 | 0.022 | 0.67 | 0.46–0.98 | 0.04 |
| Ethnicity (ref: African) | 0.95 | 0.58–1.56 | 0.848 | |||
| ART (ref: Not on ART) | 1 | 1 | ||||
| On ART not containing TDF | 0.74 | 0.45–1.21 | 0.228 | 0.72 | 0.43–1.18 | 0.19 |
| On ART containing TDF | 3.72 | 2.46–5.64 | <0.001 | 2.99 | 1.89–4.73 | <0.001 |
| CD4 (<200 cells/mm3) | 1.30 | 0.90–1.87 | 0.156 | |||
| eGFR (ml/min per 1.73 m2) (ref: <15) | 1 | |||||
| eGFR 15−29 | 0.98 | 0.60–1.60 | 0.928 | |||
| eGFR 30−59 | 0.47 | 0.27–0.80 | 0.005 | |||
| eGFR ≥60 | 0.62 | 0.37–1.05 | 0.073 | |||
ART, antiretroviral therapy; aOR, adjusted odds ratio; CI, confidence interval; eGFR, estimated glomerular filtration rate; OR, odds ratio; TDF, tenofovir disoproxil fumarate.
Nineteen participants out of the total cohort commenced chronic KRT, whereas there were 275 deaths over the study period. We evaluated the probability of survival for participants during each of the 4 time periods with participants stratified by baseline eGFR (< or ≥ 15 ml/min per 1.73 m2). For those with eGFR <15 ml/min per 1.73 m2 (Figure 2a), survival probability was higher in the more recent time periods as a result of lower mortality rather than commencement of KRT during this time. In all 4 time periods, there was a steep decline in survival probability in the first 3 months. Nonsignificant differences in survival probability were observed for those with eGFR ≥15 ml/min per 1.73 m2 (Figure 2b). However, survival in this group was higher for all time periods compared with those with an eGFR <15 ml/min per 1.73 m2 (Figure 2a).
Figure 2.
Kaplan-Meier curves for survival by time periods stratified by (a) eGFR <15 ml/min per 1.73 m2) and (b) eGFR >15 ml/min per 1.73 m2). eGFR, estimated glomerular filtration rate.
Survival probability was substantially better for those diagnosed with TID than those with HIVAN/FSGS (Supplementary Figure S1). In the Cox proportional hazards regression, we restricted the analysis to patients with an eGFR <15 ml/min per 1.73 m2 to further describe the association among kidney histological diagnosis, time-period, exposure to ART, and the combined end point of mortality and those with end-stage kidney disease accepted onto chronic KRT (Table 6). In univariable analysis, a diagnosis of TID, a biopsy in recent time periods, and exposure to ART were associated with a lower mortality and/or commencement of KRT. In the multivariable analysis, these associations persisted for TID, ART, and both the ([2013–2015] and [2016–2020]) time periods.
Table 6.
Factors associated with mortality and commencement of KRT in patients with an eGFR <15 ml/min per 1.73 m2
| Variables | Univariable |
Multivariable |
||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P-value | aHR | 95% CI | P-value | |
| HIVAN/FSGS (ref) | ||||||
| TID | 0.30 | 0.19–0.45 | < 0.001 | 0.43 | 0.27–0.69 | 0.001 |
| Other | 0.86 | 0.61–1.21 | 0.40 | 0.99 | 0.69–1.41 | 0.95 |
| Time period | ||||||
| 2005–2009 (ref) | ||||||
| 2010–2012 | 0.78 | 0.53–1.16 | 0.23 | 0.91 | 0.60–1.38 | 0.65 |
| 2013–2015 | 0.34 | 0.21–0.55 | <0.001 | 0.46 | 0.27–0.78 | 0.004 |
| 2016–2020 | 0.55 | 0.37–0.80 | 0.002 | 0.45 | 0.30–0.69 | < 0.001 |
| Age <40 yr (ref) | ||||||
| 40–50 yr | 0.81 | 0.57–1.15 | 0.23 | |||
| >50 yr | 0.72 | 0.44–1.16 | 0.18 | |||
| Gender (female) | 0.93 | 0.69–1.23 | 0.63 | |||
| Ethnicity (African) | 1.05 | 0.67–1.66 | 0.83 | |||
| On ART | 0.48 | 0.36–0.65 | < 0.001 | 0.61 | 0.44–0.86 | 0.005 |
| CD4 (<200 cells/mm3) | 0.96 | 0.71–1.32 | 0.82 | |||
AHR, adjusted hazard ratio; ART, antiretroviral therapy; CI, confidence interval; HR, hazard ratio; HIVAN, human immunodeficiency virus associated nephropathy; FSGS, focal segmental glomerulosclerosis.
Discussion
This South African study describes the evolution of kidney histology in PWH over a 15-year period, during which ART was progressively rolled out in the Western Cape Province. It describes the influence of ART usage and changes in ART regimens on trends in kidney histology, particularly HIVAN and TID, and associated outcomes in PWH. Although many patients remained severely immunodeficient with only two-thirds being on ART by the final time period, we observed a decrease in the proportion of biopsies showing HIVAN/FSGS with a concomitant increase in TID. We also observed higher survival in later time periods in patients who had severe kidney failure (eGFR <15 ml/min per 1.73 m2). There are no clear clinical or laboratory characteristics that adequately distinguish different kidney pathologies other than a kidney biopsy. Furthermore, the prognostic value afforded by biopsy provides support for its continued use in the management of PWH who present with kidney disease.
There are several publications that have reported results consistent with our study findings and attributed their findings similarly to increased uptake of combination ART12, 13, 14, 15 and its beneficial effect on reducing the incidence of HIVAN.23 Unfortunately, despite efforts to reach and provide ART to all PWH in South Africa, a substantial proportion remain undiagnosed and a further proportion have disengaged from HIV care or are otherwise not on ART.11 Consequently, we observe large numbers of PWH with HIV viremia, which increases risk of progressive immunodeficiency and kidney disease, particularly HIVAN/FSGS. The lower reported rates of ART coverage and viral suppression in West Africa are of particular concern given the high prevalence of APOL1 variants in this region, which strongly predispose to HIVAN/FSGS.7,24 Persistent challenges in early HIV diagnosis and universal long-term ART coverage suggest PWH in Sub-Saharan Africa will remain at risk of developing HIVAN/FSGS.
The decrease in the proportion of biopsies showing HIVAN has been accompanied by a relative increase in other kidney pathologies. Both Lescure et al. 13 and Diana et al. 15 reported a 3-fold to 5-fold increase in FSGS, which may represent an attenuated form of HIVAN in ART-treated patients.13,15 The study by Diana et al.15 from South Africa reported a 4-fold increase (from 2%–7.8%) in tubulointerstitial dominant pathologies, including both tubulointerstitial nephritis and ATI. We did not observe an increase in FSGS. The substantially higher proportion of cases with TID in our series may reflect the exceptionally high prevalence of TB (42%), which can result in granulomatous interstitial nephritis in the kidney, TB drug-induced interstitial nephritis, or ATI from kidney hypoperfusion in those with severe systemic illness.25,26 Furthermore, the greater degree of immunodeficiency in our patients may have resulted in increased susceptibility to other infections and increased exposure to nephrotoxic agents. Of note, about half of the patients with HIV/TB had initiated ART within 3 months of kidney biopsy, raising the possibility that some kidney pathology may have been the result of immune reconstitution inflammatory syndrome in the kidney.25
Several antiretrovirals have the propensity to cause TID. Atazanavir is a recognized cause of tubulointerstitial nephritis27 and TDF may cause proximal tubulopathy, which is characterized by low molecular weight proteinuria, phosphate wasting, normoglycemic glycosuria, and ATI with prominent eosinophilic intracytoplasmic inclusions and dysmorphic mitochondria on kidney biopsy.21,28 We observed an increase in biopsies showing ATI from 2010 to 2020 when TDF was increasingly used in first and/or second line ART regimens in our region. Although it is important to acknowledge that the etiology of ATI is often multifactorial, particularly in acutely ill hospitalized patients, TDF (particularly when administered in full dose) could be a significant contributor to ATI. Although Diana et al.15 also reported an increase in the proportion of biopsies showing ATI, the study did not provide data on ART exposures, specifically TDF. Alternatively, a biopsy series from the United States of America29 showed a decrease in the proportion of biopsies showing TDF nephrotoxicity over time, which may firstly reflect familiarity with TDF nephrotoxicity, allowing its management without kidney biopsies; and secondly the replacement of TDF with tenofovir alafenamide would have lowered the nephrotoxic potential of ART in their cohort.30
Finally, we report greater survival probability in the later time periods (2013–2015 and 2016–2020) in those who had severe kidney failure (eGFR <15 ml/min per 1.73 m2) at the time of biopsy. This was largely explained by increased ART usage with subsequent reduction in early mortality and histological evolution from HIVAN/FSGS to TID, which led to better prognostic outcomes in this time frame. Together with the long-term kidney recovery rates exceeding 50%, transient KRT should be provided to people with HIV who present with severe kidney failure in the setting of severe illness and/or exposure to TDF.
The lowest survival was observed in the earliest time period (2005−2009). At this time, ART rollout was in its infancy and not freely available to all PWH. Furthermore, hospital policy at the time did not actively support patients with KRT particularly those with HIVAN who were not on ART. This likely explains the higher mortality in this first period. In the final time period (2016−2020), despite decreases in HIVAN, and increased access to KRT and ART, patients were frequently admitted with severe systemic illness and underlying sepsis culminating in acute kidney injury. This shift in disease severity may have contributed to the somewhat lower survival in the most recent period than that in the preceding period (2013−2015). Furthermore, our survival data highlights that having severe kidney failure is associated with poor outcomes, particularly in the first 3 months post biopsy.
There remained a low rate of chronic KRT acceptance in our biopsy cohort, with only 19 PWH meeting criteria for our program from 2005 to 2020, which reflects the severe resource constraints of KRT in the South African health care system.3 However, this finding should be interpreted in context with the high proportion of acutely ill patients who died soon after admission to hospital, which precluded consideration for the program.
The strengths of this study are the large cohort, the prospective kidney histology registry on which the cohort was based, and post-biopsy follow-up in a dedicated HIV/kidney clinic. Limitations include that our findings reflect the experience of a single tertiary referral center in a region with a high rate of HIV/TB coinfection. The cohort increasingly comprised of hospitalized patients, who are likely to have experienced multiple kidney insults resulting in multiple pathologies on biopsy, making an independent histological diagnosis challenging. The low CD4 cell counts and low rates of HIV suppression in this group indicate that many of the paticipants are individuals who have not received the full benefits of ART, most commonly because of suboptimal adherence or complete cessation of ART medications. Fewer outpatient referrals from HIV clinics suggest clinicians may have become more comfortable managing mild to moderate kidney disease in this setting. If this is the case, the shift from HIVAN/FSGS to TDF-associated TID may have been even more pronounced. Although the association between exposure to TDF and TID is intriguing, the common occurrence of TID in hospitalized, systemically unwell patients with multiple kidney insults precludes inference of causality.
Conclusion
As ART programs have intensified in the Western Cape Province, the spectrum of kidney histology in PWH has evolved. An increase in ART has contributed to a decline in HIVAN/FSGS and substantial improvements in kidney survival justify maximal kidney support during acute hospitalization. Kidney biopsy remains a valuable tool to establish an accurate histological diagnosis, identify opportunities to preserve kidney function through additional treatments, and to provide prognostic information. The relationship between increases in TID and exposure to TDF warrants further investigation.
Disclosure
All the authors declared no competing interests.
Acknowledgments
The authors thank Dr. Maureen Duffield for mentorship provided on kidney histology. NW is supported by the United States National Institutes of Health/Fogarty International Center grant No. 1D43TW010937-01A1(University of Pittsburgh HIV Comorbidities Research Training Program in South Africa–Pitt-HRTPSA).
Translational Statement
This paper provides insight into the effect of phasic ART rollout on the spectrum of kidney disease observed in a tertiary referral center in PWH in South Africa over a 15-year period (2005–2020). We also showed that kidney histology corresponded with changes to degree of ART coverage over time. The spectrum of kidney disease in our region differs from developed countries and is likely due to availability and coverage of ART as well as differences in the population’s communicable and noncommunicable disease profiles. Our results contribute to the knowledge in HIV-associated kidney disease and will be specifically useful for those working in high HIV endemic areas within low resource settings.
Footnotes
Figure S1. Kaplan-Meier of survival comparing patients with a primary diagnosis of HIVAN/FSGS versus TID
Table S1. Distribution of dominant pathologies for primary, secondary, and tertiary histopathological diagnoses.
STROBE Statement.
Supplementary Material
Supplementary Figure 1.
Figure S1. Kaplan-Meier of survival comparing patients with a primary diagnosis of HIVAN/FSGS versus TID.
Table S1. Distribution of dominant pathologies for primary, secondary, and tertiary histopathological diagnoses.
STROBE Statement.
References
- 1.Swanepoel C.R., Wearne N., Okpechi I.G. Nephrology in Africa--not yet uhuru. Nat Rev Nephrol. 2013;9:610–622. doi: 10.1038/nrneph.2013.168. [DOI] [PubMed] [Google Scholar]
- 2.Moosa M.R., Maree J.D., Chirehwa M.T., Benatar S.R. Use of the “accountability for reasonableness” approach to improve fairness in accessing dialysis in a middle-income country. PLoS One. 2016;11 doi: 10.1371/journal.pone.0164201. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Moosa M.R., Wearne N. Invited commentary Should We Be Rationing Dialysis in South Africa in the 21st Century? Perit Dial Int. 2018;38:84–88. doi: 10.3747/pdi.2017.00179. [DOI] [PubMed] [Google Scholar]
- 4.Muller E., Kahn D., Mendelson M. Renal transplantation between HIV-positive donors and recipients. N Engl J Med. 2010;362:2336–2337. doi: 10.1056/NEJMc0900837. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Davids M.R., Jardine T., Marais N., Jacobs J.C., Sebastian S. South African renal registry annual report 2018. Afr J Nephrol. 2020;23:185–196. doi: 10.21804/23-1-4458. [DOI] [Google Scholar]
- 6.Swanepoel C.R., Atta M.G., D’Agati V.D., et al. Kidney disease in the setting of HIV infection: conclusions from a Kidney Disease: improving Global Outcomes (KDIGO) Controversies Conference. Kidney Int. 2018;93:545–559. doi: 10.1016/j.kint.2017.11.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hung R.K., Binns-Roemer E., Booth J.W., et al. Genetic variants of APOL1 are major determinants of kidney failure in people of African ancestry with HIV. Kidney Int Rep. 2022;7:786–796. doi: 10.1016/j.ekir.2022.01.1054. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Wearne N., Davidson B. Springer; 2022. Bacterial and parasitic-related tubulointerstitial nephritis. Tubulointerstitial Nephritis; pp. 69–80. [Google Scholar]
- 9.Wearne N., Davidson B., Blockman M., Swart A., Jones E.S. HIV, drugs and the kidney. Drugs Context. 2020;9 doi: 10.7573/dic.2019-11-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Booth J.W., Hamzah L., Jose S., et al. Clinical characteristics and outcomes of HIV-associated immune complex kidney disease. Nephrol Dial Transplant. 2016;31:2099–2107. doi: 10.1093/ndt/gfv436. [DOI] [PubMed] [Google Scholar]
- 11.UNAIDS data 2021. UNAIDS. https://www.unaids.org/en/resources/documents/2021/2021_unaids_data
- 12.Berliner A.R., Fine D.M., Lucas G.M., et al. Observations on a cohort of HIV-infected patients undergoing native renal biopsy. Am J Nephrol. 2008;28:478–486. doi: 10.1159/000112851. [DOI] [PubMed] [Google Scholar]
- 13.Lescure F.-X., Flateau C., Pacanowski J., et al. HIV-associated kidney glomerular diseases: changes with time and HAART. Nephrol Dial Transplant. 2011;27:2349–2355. doi: 10.1093/ndt/gfr676. [DOI] [PubMed] [Google Scholar]
- 14.Bookholane H., Wearne N., Surapaneni A., et al. Predictors and prognosis of HIV-associated nephropathy on kidney biopsy in South Africa. Kidney Int Rep. 2020;5:1799–1804. doi: 10.1016/j.ekir.2020.06.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Diana N.E., Davies M., Mosiane P., Vermeulen A., Naicker S. Clinicopathological correlation of kidney disease in HIV infection pre- and post-ART roll out. PLoS One. 2022;17 doi: 10.1371/journal.pone.0269260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Joss N., Morris S., Young B., Geddes C. Granulomatous interstitial nephritis. Clin J Am Soc Nephrol. 2007;2:222–230. doi: 10.2215/CJN.01790506. [DOI] [PubMed] [Google Scholar]
- 17.Shah S., Carter-Monroe N., Atta M.G. Granulomatous interstitial nephritis. Clin Kidney J. 2015;8:516–523. doi: 10.1093/ckj/sfv053. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Rovin B.H., Adler S.G., Barratt J., et al. Executive summary of the KDIGO 2021 Guideline for the Management of Glomerular Diseases. Kidney Int. 2021;100:753–779. doi: 10.1016/j.kint.2021.05.015. [DOI] [PubMed] [Google Scholar]
- 19.Gerntholtz T., Goetsch S., Katz I. HIV-related nephropathy: a South African perspective. Kidney Int. 2006;69:1885–1891. doi: 10.1038/sj.ki.5000351. [DOI] [PubMed] [Google Scholar]
- 20.Wearne N., Swanepoel C.R., Boulle A., Duffield M.S., Rayner B.L. The spectrum of renal histologies seen in HIV with outcomes, prognostic indicators and clinical correlations. Nephrol Dial Transplant. 2012;27:4109–4118. doi: 10.1093/ndt/gfr702. [DOI] [PubMed] [Google Scholar]
- 21.Herlitz L.C., Mohan S., Stokes M.B., Radhakrishnan J., D’Agati V.D., Markowitz G.S. Tenofovir nephrotoxicity: acute tubular necrosis with distinctive clinical, pathological, and mitochondrial abnormalities. Kidney Int. 2010;78:1171–1177. doi: 10.1038/ki.2010.318. [DOI] [PubMed] [Google Scholar]
- 22.Inker L.A., Eneanya N.D., Coresh J., et al. New creatinine-and cystatin C–based equations to estimate GFR without race. N Engl J Med. 2021;385:1737–1749. doi: 10.1056/NEJMoa2102953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lucas G.M., Eustace J.A., Sozio S., Mentari E.K., Appiah K.A., Moore R.D. Highly active antiretroviral therapy and the incidence of HIV-1-associated nephropathy: a 12-year cohort study. AIDS (Lond Engl) 2004;18:541–546. doi: 10.1097/00002030-200402200-00022. [DOI] [PubMed] [Google Scholar]
- 24.South Africa 2021. UNAIDS. https://www.unaids.org
- 25.Wearne N., Hung R., Bohmer R., et al. Kidney disease in Africans with HIV and tuberculosis. AIDS (Lond Engl) 2019;33:1207–1213. doi: 10.1097/QAD.0000000000002204. [DOI] [PubMed] [Google Scholar]
- 26.Davidson B., Nel D., Jones E., et al. Granulomatous interstitial nephritis on renal biopsy in HIV positive patients: prevalence and causes in Cape Town, South Africa. Nephrology. 2019;24:681–688. doi: 10.1111/nep.13564. [DOI] [PubMed] [Google Scholar]
- 27.Brewster U.C., Perazella M.A. Acute interstitial nephritis associated with atazanavir, a new protease inhibitor. Am J Kidney Dis. 2004;44:e81–e84. doi: 10.1016/S0272-6386(04)01093-5. [DOI] [PubMed] [Google Scholar]
- 28.Hamzah L., Jose S., Booth J., et al. Treatment-limiting renal tubulopathy in patients treated with tenofovir disoproxil fumarate. J Infect. 2017;74:492–500. doi: 10.1016/j.jinf.2017.01.010. [DOI] [PubMed] [Google Scholar]
- 29.Kudose S., Santoriello D., Bomback A.S., et al. The spectrum of kidney biopsy findings in HIV-infected patients in the modern era. Kidney Int. 2020;97:1006–1016. doi: 10.1016/j.kint.2020.01.018. [DOI] [PubMed] [Google Scholar]
- 30.Gupta S.K., Post F.A., Arribas J.R., et al. Renal safety of tenofovir alafenamide vs. tenofovir disoproxil fumarate: a pooled analysis of 26 clinical trials. AIDS (Lond Engl) 2019;33:1455–1465. doi: 10.1097/QAD.0000000000002223. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.