Renal AA-amyloidosis in intravenous drug users – a role for HIV-infection?

Renal disease related to intravenous drug use (IVDU) has been reported since the 1970s, mostly in the context of heroin-associated nephropathy, characterized by nephritic syndrome and rapid-progression [1, 2]. The incidence of heroin associated-nephropathy decreased in the last four decades. More recent reports starting from the 1990s related chronic kidney disease in IVDU more with concomitant chronic HIV-, HBV- and HCV-infection [3‐5]. Two recent studies from Europe observed changing patterns of renal disease in patients with IVDU, reporting an increased prevalence of renal AA-amyloidosis [6, 7].

Renal AA-amyloidosis is a complication of chronic and/or recurrent inflammatory disease [8]. Autoimmune diseases, mainly rheumatoid arthritis, are common causes in resource rich countries whereas untreated chronic infections are the predominant cause in countries with limited medical resources. Severe proteinuria, nephrotic syndrome as well as renal insufficiency are the typical clinical manifestations of renal AA-amyloidosis [8]. Successful treatment of the underlying inflammation, by immunosuppressant’s for autoimmune diseases or by antimicrobials for chronic infections, can lead to stabilization of or even improvements of renal function [9].

Although cases of AA-amyloidosis in patients with IVDU have already been described in the 1970s [10, 11], only sporadic cases have been reported in the last decades [5, 12‐14]. To date, it remains unclear to what extent IVDU and associated co-morbidities, especially chronic infections, contribute to the development of renal AA-amyloidosis. Our aim was to describe the histologic pattern of nephropathy and its associated risk factors in a contemporary cohort of patients with IVDU.

The Nephrology departments at Goethe-University (GU) and the Agaplesion Markus Krankenhaus (AMK) are two main nephrologic care providers and the only centres conducting renal biopsies in Frankfurt, Germany. All patients with IVDU were identified by using the Patient Data management System of the renal units of the GU and AMK that are based on primary and secondary diagnosis according to the international Classification of Diseases 10th Revision (ICD-10). We then retrospectively analyzed all patients with ongoing or a prior history of IVDU referred to us between 01.04.2002 and 31.03.2012 that underwent renal biopsy because of chronic kidney disease. Biopsies were performed because of progressive renal failure or overt proteinuria. Clinical data and the patient′s medical history were extracted from the medical records. Glomerular filtration rate was estimated (eGFR) using the Modification of Diet in Renal Disease (MDRD) equation [15]. Proteinuria was assessed by collection of 24-h urine at the time of renal biopsy. Proteinuria prior to renal biopsy was determined from spot urine. End-stage renal disease (ESRD) was defined as initiation of chronic renal replacement therapy (RRT) in an ambulatory setting. All data were collected retrospectively; thus no informed consent or approval from the ethics committee was obtained.

Chronic hepatitis B virus (HBV) infection was defined by a detectable HBs antigen persisting at least 6 months. Past HBV exposure was defined by detection of HBV core or surface antibodies in absence of previous vaccination. Chronic hepatitis C virus (HCV) infection was defined by detection of HCV antibodies and positive HCV RNA PCR persisting at least 6 months. HIV infection was defined by positive HIV antibodies or positive HIV RNA PCR.

All renal biopsy specimens were submitted for diagnostic purposes and proceeded using our standardized routine protocol. Briefly, formalin-fixed and paraffin-embedded kidney biopsy specimens were sectioned into 2 μm thick paraffin sections with at least 8 serial sections stained either by periodic acid-Schiff (PAS) or hematoxylin-eosin (HE) stainings. Additionally, a Congo red staining as well as immunohistochemical stainings with antisera specific for IgA (1:150000), IgG (1:100000), IgM (1:75000), C1q (1:75000) and C3c (1:75000, all polyclonal rabbit antisera, Dako Cytomation, Hamburg, Germany) were performed of all cases. In Congo red positive cases additional stainings with antibodies specific for the kappa (1:50000) and lambda (1:100000) light chains (both polyclonal rabbit antisera, Dako) and for amyloid A (1:500, monoclonal, mouse, clone mc-1, Dako) for differentiation of the amyloid deposits were initiated.

Continuous variables were expressed as median and interquartile-range (IQR) or as proportions as appropriate. Continuous and categorical variables were compared for univariate analysis between groups using the t-test or Mann–Whitney U-test and Fisher exact test, respectively. All p-values reported are two-sided. Statistical significance was assumed when the p-value was <0.05. A statistical trend was assumed when the p-value was >0.05 but <0.08. Kaplan-Meier (Mantel-Cox log-rank test).estimates were used to analyze time to death or progression to ESRD.

In this retrospective study we found renal AA-amyloidosis to be the predominant cause of progressive kidney disease in IVDU accounting for 50% of cases, followed by membranoproliferative glomerulonephritis as the second leading cause in 20.8%. This finding is well in accordance with two recent observations from other European Centres. These studies described an increasing rate of renal AA-amyloidosis in IVDU, while in older biopsy studies AA-amyloidosis was found only sporadically, indicating a changing pattern of nephropathy in this patient population [5‐7].

The occurrence of renal AA-amyloidosis in IVDU has been linked to repeated bacterial infections, especially to skin and soft tissue infections, endocarditis as well as longer survival time after the onset of IVDU (as a consequence of successful opoid maintenance therapy) [6, 7, 10, 11]. When comparing patients with renal AA-amyloidosis to those without, we found a trend towards a longer self-reported time of IVDU. Repeated infections tended also to be more frequent in patients with AA-amyloidosis. Yet, the low number of patients in our series makes it difficult to establish a link between isolated sites of infection and occurrence of AA-amyloidosis. Alternatively, it might be the case that not one specific focus but the mere total number of bacterial infections predisposes to the development of renal amyloidosis.

The proportion of patients with HIV infection was significantly higher in patients with renal AA-amyloidosis. This observation has not been made in two published studies on AA-amyloidosis in patients with IVDU. One study did not provide data concerning HIV serostatus, while the other found only one out of 20 patients to be HIV infected [6, 7]. However, this does not rule out a higher rate of HIV infections in these studies, as in some of our patients the HIV infection was just diagnosed before the renal biopsy was conducted. Several cases of amyloidosis in HIV infected IVDU have been described in the literature [5, 6, 13, 14, 16, 17]. Besides IVDU, renal amyloidosis in HIV infected patients has been described in association with visceral leishmaniosis or multicentric Castleman`s disease [18‐21], which were both not found in our cohort. During the same time period of this retrospective study, 25 HIV-infected patients without a history of IVDU underwent renal biopsy at our institutions and none of them had evidence of renal AA-amyloidosis [data not shown]. Although these patients cannot be properly matched, because of a higher median age and 25% being of African-descent, this could indicate that the HIV infection by itself does not predispose for AA-amyloidosis, but coeval IVDU might increase the risk. Indeed, in a recent study, we found IVDU to be an independent risk factor for ESRD in an HIV-infected population [22]. Moreover, another study demonstrated that the combination of HIV infection and IVDU increases the risk for chronic kidney disease (CKD) in IVDU, with a higher prevalence of proteinuria in HIV-infected compared to uninfected IVDU [23]. The majority of HIV infected patients with renal AA-amyloidosis had a poorly or uncontrolled HIV infection prompting a severe immunosuppression and a chronic inflammation. Thus the uncontrolled HIV infection might be an important direct and indirect factor for the development of AA-amyloidosis predisposing to more frequent and severe bacterial infections.

Although there are several case reports of AA-amyloidosis in HIV infected IVDU, such a high rate has not been previously described. What might explain this finding? First, medical management of the IVDU population is challenging, as patients often do not return for follow-up and seek medical advice only when acutely ill. Hence, renal biopsy is infrequently performed in IVDU. Second, the time course of renal amyloidosis, its clinical presentation with lack of peripheral edema or systemic hypertension, and enlarged echogenic kidneys on ultrasound makes it indistinguishable from certain other types of renal disease, especially HIV-associated nephropathy (HIVAN) [3, 17, 24]. The diagnosis of HIVAN is often established solely clinically and is often not confirmed by renal biopsy. Thus the true prevalence of renal AA-amyloidosis might even be underestimated [25, 26]. Third, development of AA-amyloidosis also depends on predisposing genetic factors and is more frequent in Caucasian compared to Black patients. Vice versa HIVAN occurs mostly in Black patients and is a rarity in Caucasians [27]. Thus different ethnic compositions of IVDU cohorts may also contribute to the higher incidence observed by us and other European centres when compared to studies from the US with far higher rates of black IVDU [6, 7].

IVDU patients with AA-amyloidosis had a poor prognosis, when compared to patients with other underlying causes of AA-amyloidosis [6‐8]. We and others found that patients, who remained abstinent from IVDU after the initial diagnosis, achieved stable or even improving renal function during follow-up. This supports the idea that removal of the recurrent inflammatory stimuli can lead to at least some reversal of disease [6]. These observations, underscore the importance of renal biopsies, especially in the light of the oligosymptomatic clinical presentation of these patients. We found that overt proteinuria preceded the GFR decline by approximately 1–2 years. But unfortunately urine analysis is often missed by IVDU because unannounced illicit drug testing with possible consequences for opioid maintenance therapy is feared.

Besides the already discussed, our study has several limitations. First, the population studied is small and from a single metropolitan area. Thus, our results cannot be generalized to other HIV, often ethnically highly divers, cohorts. Second, IVDU often consume more than one substance of varying, uncontrolled purity. Thus our retrospective study cannot determine possible differences between substances used.

In conclusion we found AA-amyloidosis was the predominant cause of progressive renal disease in the last 10 years in IVDU patients. The high incidence of AA-amyloidosis observed was related to coeval HIV infection. We speculate that the HIV mediated immunosuppression promotes development of AA-amyloidosis by increasing frequency and severity of infectious complications acquired by ongoing IVDU. Overt proteinuria preceded the decline of GFR for approximately 1–2 years, providing an opportunity for early intervention in an otherwise fatal disease.