Chronic Kidney Disease from Polyvinylpyrrolidone Deposition in Persons with Intravenous Drug Use

Visual Abstract

Abstract

Background and objectives

Persons with intravenous drug use have a higher risk of developing CKD compared with the general population. In Norway, deposits of polyvinylpyrrolidone have been observed in kidney biopsies taken from persons with opioid addiction and intravenous drug use since 2009. Polyvinylpyrrolidone is an excipient commonly used in pharmaceuticals, and the polyvinylpyrrolidone deposits observed in these patients were caused by intravenous injection of a specific oral methadone syrup containing very high molecular weight polyvinylpyrrolidone. Here, we present the clinicopathologic findings from 28 patients with CKD associated with polyvinylpyrrolidone deposition in the kidney.

Design, setting, participants, & measurements

The 28 patients and their kidney biopsies were included when polyvinylpyrrolidone deposition was recognized, either retrospectively or at the time of diagnostic evaluation. Biopsies were taken between 2009 and 2016. We collected laboratory parameters and clinical data from digital patient charts. For each kidney biopsy, the glomerular volume, extent of polyvinylpyrrolidone deposition, and tubulointerstitial area with tubular atrophy were assessed quantitatively.

Results

All patients (mean age: 37 years) had CKD (mean eGFR: 33 ml/min per 1.73 m²) and normal urine protein or non-nephrotic–range proteinuria. Biopsies showed moderate to severe tubular atrophy (mean extent: 65%) and interstitial infiltrates of vacuolated macrophages containing polyvinylpyrrolidone (mean share of biopsy area: 1.5%). Underperfused and ischemic glomeruli were common findings. In 22 samples, ultrastructural investigation revealed polyvinylpyrrolidone-containing vacuoles in the mesangial or endothelial cells of glomeruli. At the last follow-up, most patients had stable or improved eGFR. Two patients had developed kidney failure and underwent hemodialysis.

Conclusions

Intravenous injection of a specific oral methadone syrup caused polyvinylpyrrolidone deposition in the kidney in persons with opioid addiction and intravenous drug use. Kidney biopsy findings suggested an association between polyvinylpyrrolidone deposition and tubular atrophy.

Introduction

Intravenous (IV) drug use is associated with a wide range of diseases, including CKD (1,2). The risk of CKD in this patient group is due in part to effects of the drugs themselves (3) and in part to comorbidities associated with IV drug use, such as soft tissue infections (4,5), chronic viral infections (6), and episodes of rhabdomyolysis (7). Common histopathologic diagnoses are glomerulopathies (e.g., membranoproliferative GN), amyloidosis, acute interstitial nephritis, and acute tubular necrosis (3,8).

Starting in 2009, pathology departments in Norway received biopsies from patients with opioid addiction and IV drug use showing deposits of polyvinylpyrrolidone (PVP) (9–12), a common excipient in pharmaceuticals (13). Moderate and high molecular weight PVP is exclusively utilized in oral and topical medications (13), as it may accumulate in the body if injected (14,15). In most countries, the use of moderate and high molecular weight PVP in parenteral medication ended in the late 1970s (16) when it was recognized that extensive PVP deposition could cause severe organ dysfunction (17–19). However, cases of CKD resulting from PVP deposition have rarely been reported (20,21).

In this case series, we present the clinicopathologic findings from 28 patients with PVP deposition in the kidney and CKD. The PVP deposited in these patients originated from IV use of a specific oral methadone syrup marketed as an opioid substitution drug in Norway and several other countries from 2007 to 2014 (22). IV use of oral opioid substitution drugs is widespread among persons with IV drug use (23), and the drugs are also shared and sold on to others (24). In this specific methadone syrup, very high molecular weight PVP was added as a thickener with the aim of preventing unintended IV use (11,22). Unfortunately, the higher viscosity did not have the intended effect, and injection of the methadone syrup caused PVP deposition in various organs, including the kidneys, in some cases, with detrimental clinical consequences (11,12).

This study aims to describe the clinical, light microscopy, and ultrastructural findings in a series of patients with PVP deposition in the kidney and to alert both clinicians and pathologists to this diagnostic possibility in the context of CKD, opioid addiction, and IV drug use.

Materials and Methods

Patients

Twenty-eight kidney biopsies taken between 2009 and 2016 showing PVP deposition were included continuously or retrospectively following primary or second opinion diagnostic evaluation. Biopsies were obtained from four pathology departments in Norway: Haukeland University Hospital, Oslo University Hospital, St. Olav’s University Hospital, and Førde Hospital. PVP deposition was defined as the presence of macrophages with intracytoplasmatic vacuoles displaying the following staining properties: light blue in the hematoxylin and eosin (H&E) stain, red in the Congo stain, and gray or black in the periodic acid silver–methenamine stain (Figure 1) (17). Vacuole content should not be birefringent in Congo red stain or any other stain (17).

Figure 1.

Staining characteristics for polyvinylpyrrolidone deposited in tissue. (A) Light blue in hematoxylin and eosin (H&E); (B) red in Congo red stain, although not birefringent in polarizing light; and (C) gray or black in periodic silver–methenamine stain. Scale bars: 20 μm.

Patient clinical and demographic data were collected from pathology referral forms, digital patient charts (medical records), and/or the Norwegian Renal Registry. Follow-up data were collected in May 2019. Proteinuria was assessed by albumin-creatinine ratio (ACR) or protein-creatinine ratio depending on which was available. For one patient, proteinuria was assessed by strip test evaluation. Proteinuria was graded as normal to mildly increased (<30 mg/g [ACR] or <150 mg/g [protein-creatinine ratio]), moderately increased (30–300 mg/g [ACR] or 150–500 mg/g [protein-creatinine ratio]), or severely increased (>300 mg/g [ACR] or >500 mg/g [protein-creatinine ratio]) (25). In four cases, there was a discrepancy between ACR and protein-creatinine ratio levels. In these cases, proteinuria was graded according to the parameter showing the highest level. The diagnosis of CKD was defined as >3 months of abnormalities in kidney function documented as GFR <60 ml/min per 1.73 m² or albuminuria (25). Progression of CKD was categorized as slow progression (decrease in eGFR of 1–3 ml/min per 1.73 m² per year) or fast progression (decrease in eGFR of >3 ml/min per 1.73 m² per year) (25,26).

The study was approved by the Regional Committee for Medical and Health Research Ethics (REK; no. 27687). REK approved exemption from the consent requirement for the use of information and biologic material gathered in health care services. The study was performed in adherence to the Declaration of Helsinki.

Methods of Measurement

As part of the original diagnostic evaluation, formalin-fixed, paraffin-embedded biopsies were routinely stained with H&E, Congo red, periodic acid–Schiff (PAS), trichrome, and periodic acid silver–methenamine stain. The following immunohistochemical panel was used: IgG, IgA, IgM, C1q, and C3c. Immunohistochemical evaluation of macrophages with CD68 was performed in eight cases. Tissue for electron microscopy (EM) from 27 of the 28 cases was fixed and processed according to the standard procedure at each hospital. Available slides were scanned with ScanScope XT (Aperio) at ×40 magnification (resolution of 0.25 μm per pixel). We viewed and annotated the virtual slides in ImageScope 12.

The total number of assessable glomeruli (defined as >25% of the capillary tuft visible) was counted, as well as the numbers of glomeruli with necrosis, segmental glomerulosclerosis, global glomerulosclerosis, crescents, underperfused glomeruli, and ischemic glomeruli. An underperfused glomerulus was defined as a glomerulus with a collapsed capillary tuft. An ischemic glomerulus was defined as a glomerulus with a collapsed capillary tuft and periglomerular fibrosis in at least 50% of the glomerular circumference with involvement of the tubular pole (27). We measured the tuft area of each nonsclerotic glomerulus using the annotation pen tool in ImageScope 12. The glomerular tuft volume was calculated using the Weibel–Gomez formula: $V_{\mathit{\text{glom}}}=\frac{{A_{\mathit{\text{glom}}}}^{1.5}\times 1.38}{1.01}$ (28).

An atrophic tubule was defined as a tubule with thickened basement membranes and/or a reduction in tubule diameter (29). We quantified the tubulointerstitial area affected by tubular atrophy by point counting using a square test lattice with 100 μm between test lines applied to PAS-stained calibrated digital slides. The extent of tubular atrophy was expressed as the percentage of points hitting cortical tubulointerstitial areas with tubular atrophy of the total number of points hitting cortical tubulointerstitial tissue.

Arteriosclerosis was graded from zero to two by the extent of intimal thickening compared with the thickness of the media (30). Arteriolohyalinosis was graded from zero to three, representing increasing severity according to the Banff Hyaline Arteriolar Thickening score (29,31).

PVP-containing vacuoles were noted present or absent in the interstitium, glomeruli, and tubular epithelium. The extent of PVP deposition was expressed as the percentage of the total biopsy area occupied by PVP-containing vacuoles as measured using the pen tool in Aperio ImageScope. We used sections stained with Congo red stain, H&E stain, or PAS depending on which sections were available or which were most easily interpreted.

By EM, PVP-containing vacuoles were noted present in glomerular cells if they met all of the following criteria (17,20,32): (1) at least three round/oval vacuoles of similar size present in a single cell; (2) vacuole content has low to intermediate electron density and low granularity; (3) vacuole profile diameter 0.3–1.5 µm; and (4) vacuoles are not collocated with osmiophilic material, making the structure suspicious of lipofuscin.

Statistical Analyses

Statistical analyses were performed using SPSS for Windows, version 24. Central tendency for continuous variables was reported as mean (SD or range). We performed correlation analyses using the Spearman rank correlation coefficient for nonparametric data. Statistical significance was assumed at a P value of 0.05.

Results

Demographics, Comorbidities, and Clinical Kidney Parameters at the Time of Biopsy

Summarized clinical data are available in Table 1. This case series included 28 adults. All had a history of opioid addiction and IV drug use, and 27 (96%) of them received treatment with opioid substitution drugs. Hepatitis B and C virus positivity was common. None of the patients were HIV positive. Six patients were receiving treatment for hypertension, and two patients had diabetes mellitus type 2.

Table 1.

Demographics, comorbidities, and clinical kidney parameters at the time of biopsy

Characteristic	Value
Age at time of biopsy, yr	37 (SD 6; range, 23–52)
Men, n (%)	22 (79)
HBV positive, n (%)	16 (57)
HCV positive, n (%)	26 (93)
Proteinuria, n (%) a
No/mild	18 (64)
Moderate	9 (32)
Severe	1 (4)
Hematuria present, n (%)b	4 (14)
Serum creatinine, mg/dl	2.45 (SD 0.70; range, 1.25–4.32)
eGFR, ml/min per 1.73 m2	33 (SD 11; range, 14–58)

Values represent frequency (percentage) or mean (SD; range). eGFR was calculated by the Chronic Kidney Disease Epidemiology Collaboration formula. HBV, hepatitis B virus (seropositivity); HCV, hepatitis C virus (seropositivity).

^aOn the basis of measurements of albumin-creatinine ratio or protein-creatinine ratio. Graded according to the Kidney Disease Improving Global Outcomes guidelines (25).

^bOn the basis of strip test evaluation.

For 27 of 28 patients, indication for biopsy was kidney disease with reduced eGFR. For one patient with previously diagnosed SLE, indication for biopsy was CKD with newly discovered proteinuria. At the time of biopsy, 22 patients (79%) had CKD, two patients (7%) had AKI on CKD, and three patients (11%) had newly discovered kidney disease. The last patient had reduced eGFR, but prebiopsy duration is unknown to us.

Pathology Features

Pathologic features are summarized in Table 2.

Table 2.

Pathology findings

Characteristic	Value
Extent of tubular atrophy, %	65 (SD 20; range, 34–92)
Extent of PVP deposition, %	1.5 (SD 2.0; range, 0.05–9.1)
No. of glomeruli, %	24 (SD 12; range, 4–53)
Underperfused	51 (SD 20; range, 0–86)
Ischemic	20 (SD 16; range, 0–61)
Globally sclerosed	8 (SD 11; range, 0–50)
Arteriosclerosis, n (%) a
0	8 (29)
1	2 (7)
2	18 (64)
Arteriolohyalinosis, n (%) b
0	9 (32)
1	7 (25)
2	4 (14)
3	8 (29)

Values represent frequency (percentage) or mean (SD; range). PVP, polyvinylpyrrolidone.

^aGraded as suggested by Roberts et al. (30) (zero, no intimal thickening; one, intima thickened to less than the thickness of the media; two, intima thickened to more than the thickness of the media).

^bGraded according to the Banff Hyaline Arteriolar Thickening score (29,31) (zero, no hyaline arteriolar thickening; one, hyaline deposits in only one arteriole; two, hyaline deposits in more than one arteriole; three, hyaline deposits with circumferential involvement independent of the number of arterioles involved).

Tubulointerstitial Findings

The most prominent histologic finding was the widespread tubular atrophy (Figure 2A). Areas with atrophic tubules showed extended interstitium with varying degrees of fibrosis. PVP-containing macrophages, singular or clustered, were dispersed in the interstitium, mainly in areas affected by tubular atrophy (Figure 2, A–C). In some biopsies, extensive aggregates of PVP-containing macrophages displaced tubules in larger areas. In rare incidences, we also found PVP-containing vacuoles within the tubular epithelium. The extent of PVP deposition compared with the extent of tubular atrophy varied between biopsies (Figure 2, B and C), and we could not find a significant correlation between these two parameters.

Figure 2.

Tubulointerstitial and glomerular findings in biopsies with polyvinylpyrrolidone (PVP) deposits. (A) Abrupt transition from an area with tubular atrophy to more normal-looking tubules. There is a dense infiltrate of PVP-containing macrophages in the extended interstitium (H&E with saffron). Scale bar: 50 μm. (B and C) Whole-biopsy micrographs (H&E). To illustrate the distribution, PVP-containing vacuoles are encircled with turquoise, glomeruli are encircled with yellow, and normal tubulointerstitial areas are encircled with blue. (B) The infiltrates of PVP-containing macrophages and areas with tubular atrophy are largely overlapping. (C) As in (B), the infiltrates of PVP-containing macrophage are localized in areas with tubular atrophy; however, tubular atrophy is more extensive, and the PVP deposition is markedly less extensive than in (B). Scale bars: 500 μm. (D–F) Glomerular findings. (D) Underperfused glomerulus (periodic acid–Schiff [PAS]). The capillary tuft is collapsed with wrinkling of the basement membranes. (D, Inset) Normal glomerulus for comparison. (E) Glomerulus with PVP deposits (PAS). PVP-containing vacuoles appear to be located in the extended mesangium. (F) Glomerulus (same as in [E]) with a higher number of CD68-positive macrophages corresponding to PVP deposits in (E). Scale bars: 50 μm.

No case showed amyloid deposition, and except for the PVP-containing macrophages, there was little or no tubulointerstitial inflammation.

Glomerular Findings

The most striking finding in the glomerular compartment was the many underperfused (Figure 2D) and ischemic glomeruli. Glomerular volume varied greatly both within and between biopsies. Mean glomerular volume was 2.3 × 10⁶ μm³ (SD=1.7; 95% confidence interval, 2.2 to 2.4), indicating a low mean glomerular volume compared with the normal value of 2.6 × 10⁶ μm³ (SD=1.00) reported by Denic et al. (33). In 22 cases, we observed PVP-containing vacuoles in glomeruli by light microscopy, which in some cases, corresponded to a higher number of macrophages (Figure 2, E and F).

Clinically significant immunopathologic findings with corresponding electron-dense deposits were present in only two biopsies. One of these was from a patient with previously diagnosed SLE and showed a mesangioproliferative lupus nephritis with secondary segmental sclerosis (class 3). The second biopsy showed a mesangioproliferative immune-complex GN. No case showed evidence of glomerular disease associated with IV drug use, such as subepithelial electron-dense deposits or membranous or membranoproliferative patterns.

Vascular Findings

Arterial vessels did not show PVP deposits. Arteriosclerosis and/or arteriolohyalinosis were present in 27 of 28 cases, and moderate to severe vascular sclerosis was common. We could not find a correlation between severity of vascular sclerosis and the extent of tubular atrophy or the share of underperfused or ischemic glomeruli.

Electron Microscopy Findings

In 22 of 27 biopsies examined by EM, we observed cells in glomeruli with vacuoles that met our criteria for PVP-containing vacuoles (Figure 3, A–C). All 22 biopsies revealed vacuolated mesangial cells, and 11 of them showed vacuolated endothelial cells. There were often numerous vacuoles in a single cell, occupying much of the cytoplasm and frequently making indentations in the nucleus. Some vacuoles contained granular electron-dense material along the vacuoles’ limiting membrane. The ultrastructural appearance of vacuoles in interstitial macrophages was similar, but considerably larger vacuoles were common (Figure 3D). In podocytes, vacuoles were typically fewer in each cell, more variable in size, and collocated with osmiophilic material suspicious of lipofuscin (Figure 3E). Hence, we could not separate these from lipid-filled vacuoles with certainty.

Figure 3.

Electron microscopy findings in biopsies with PVP deposits. (A) Overview showing extended mesangial areas with vacuolated cells. (B) Cells in the mesangium containing multiple vacuoles. Vacuole content has low to intermediate electron density and low-grade granularity. The vacuole profile diameter is typically 0.3–1.5 μm. Some vacuoles have granular electron-dense material along the limiting membrane. (C) Vacuolated endothelial cells. Vacuole appearance is the same as described in (B). (D) The macrophage in the interstitium is distended by multiple vacuoles. Vacuole profile diameter is up to 5 μm. (E) Vacuolated podocyte. The vacuoles’ appearance is similar to those in the mesangial and endothelial cells. However, the vacuoles are fewer, with larger variation in size and collocated with osmiophilic material suspicious of lipofuscin.

Further Development and Patient Outcome

Although the first biopsy in this study was performed in 2009, the deposits observed were not recognized as PVP until 2013 (Figure 4). The methadone product that was the likely source of PVP in these patients was retracted from the market in Norway and the European Union in 2014 (22). After 2016, we know of no new cases of kidney PVP deposition.

Figure 4.

Timeline from the start of the marketing license in Norway for the methadone product containing high molecular weight PVP. The number of observed kidney biopsies showing PVP deposits for each year is shown to the left, with each point representing one kidney biopsy. EU, European Union. *Start of marketing license for five other European countries in 2010.

Most patients in this study received follow-up at a nephrology outpatient clinic, and we have follow-up data for 26 patients (mean observation time: 4.8 years; SD=2.4; range, 1.3–9.1). In 23 of 26 patients (88%), eGFR had slowly declined (8%), stabilized (31%), or improved (50%), but normalization was not observed. Only three patients (12%) had a rapid decline in eGFR, one of whom had lupus nephritis. The other two patients had developed kidney failure requiring hemodialysis. Both had concomitant chronic soft tissue infections and osteomyelitis. Six patients were deceased by the time of the last follow-up. One patient, who had biopsy-proven PVP deposition in several organs, died with multiple organ failure and kidney failure with ongoing hemodialysis. One patient died from injuries caused by violence, and two likely died from overdoses. The cause of death for the two remaining patients is unknown to us.

Discussion

This series of 28 patients is a systematic description of the clinicopathologic findings in patients with kidney PVP deposition after injection of a specific oral methadone syrup containing high molecular weight PVP as an excipient. All patients had CKD with absent or low-grade proteinuria, and all had a history of opioid addiction and IV drug use. The main biopsy findings were moderate to severe tubular atrophy associated with infiltrates of PVP-containing macrophages, highlighting a likely causal role of PVP deposits in the development of CKD in these patients. This case series is unique in describing PVP deposition in the kidneys in the context of CKD, opioid addiction, and IV drug use, and it may serve to raise awareness among both clinicians and pathologists regarding this diagnostic possibility.

PVP is considered an inert substance; it is not absorbed from the gastrointestinal tract, and it is not enzymatically degraded (34). If administered parenterally, the only mechanism of elimination is through glomerular filtration, which is progressively ineffective with increasing molecular weight of the PVP molecules (14). In the kidney, circulating PVP molecules enter the interstitium via the postglomerular capillaries (35). We hypothesize the following pathogenesis of CKD in these patients; the PVP deposits increase the distance between the postglomerular capillaries and the tubules, leading to an inadequate supply of nutrients and oxygen. This causes initial acute and, later, chronic tubular epithelial cell damage, and eventually, it leads to a decreased GFR via altered tubuloglomerular feedback (36). The frequent observation of underperfused glomeruli in our cases and the reduced mean glomerular volume would be consistent with such a mechanism.

It should be noted that biopsy samples with a similar extent of tubular atrophy showed a markedly different extent of PVP deposition and that no significant correlation between the extent of PVP deposition and the extent of tubular atrophy was found. One possible explanation for lack of correlation may be that there is a potential redistribution of PVP from the kidney to other tissues (14). Such a mechanism may also explain the improvement in kidney function observed in several patients after they ceased injecting the methadone syrup.

The pronounced arteriosclerosis could be an alternative explanation or contributing factor for tubular atrophy, glomerular underperfusion, and ischemia. However, arteriosclerosis is a common finding in persons with illicit drug use, as shown in an autopsy study by Buettner et al. (37) in a cohort with similar age and sex distribution. Although arteriosclerosis was pronounced, Buettner et al. (37) reported interstitial fibrosis and tubular atrophy >10% in only 16% of cases, whereas all of our cases showed a significantly higher proportion of tubular atrophy. Likewise, we could not find a correlation between vascular sclerosis and tubular atrophy or glomerular underperfusion. Glomerulonephritides with significant nephron loss as an alternative explanation for the observed tubular atrophy could be excluded. These observations taken together support the notion that PVP deposition was central in the development of tubular atrophy in our patients.

CKD in concurrence with PVP deposition in the kidney has so far only been reported in two case reports from 1968 and 1972 (20,21). The PVP deposits in these cases originated from repeated injections of parenteral medication containing moderate molecular weight PVP. Similar to our cases, the biopsies in these reports showed tubular atrophy and large interstitial accumulations of macrophages with PVP-containing vacuoles as well as glomerular deposits. However, in contrast to our findings, the case report by Grünfeld et al. (20) described vacuolated tubular epithelial cells as a main finding, and Grünfeld et al. (20) found most of the glomerular PVP deposits residing in the podocytes. Although our biopsies did show vacuolated podocytes, we were not able to distinguish these from lipid-filled vacuoles, which is a common finding in kidney biopsies (38). The differences in findings between the case reported by Grünfeld et al. (20) and the cases in this study could be due to the much lower molecular weight of the PVP previously used in parenteral medication (16,20), which allows for partial glomerular filtration and, possibly, reabsorption by the tubular epithelium (14). The lower molecular weight might also explain why CKD from PVP deposition has rarely been reported.

The methadone syrup linked to our cases was marketed in six European countries between 2007 and 2014 (22). However, moderate molecular weight PVP is still used as an excipient in several opioid substitution drugs and other oral medications with addictive properties both in Europe and in the United States (11,22,39). Frequent injection of such medications may lead to some level of accumulation of PVP (34). The question remains whether our cases are isolated incidents or if PVP deposition is under-reported due to milder disease in cases exposed to PVP with lower molecular weight. Deposition of a lipid-like material in kidney biopsies from patients with IV drug use has been described in four case reports (40–43). Three of these described a total of seven patients with pathologic findings that differ from our cases in that the deposits were primarily located in the glomeruli and that most patients presented with nephrotic syndrome (40–42). We therefore suspect that an excipient other than PVP caused the glomerular changes in these cases. Finally, a case report from Germany described two patients in a methadone substitution program who were biopsied in 2012 (43). These cases were very similar to ours in both their clinical presentation and the localization of the deposits in both glomeruli and the interstitium. The deposits could therefore very well correspond to PVP deposits. However, the authors of this report did not state whether they had investigated this possibility. Nevertheless, all four of these reports linked the deposits observed to injection of opioid substitution drugs and other oral medications with addictive properties and their excipients (40–43).

Our study has some limitations. Biopsies and patients were included in the study only when PVP deposits were recognized, which means that individual biopsies might have escaped detection. Furthermore, detailed data, such as the amount of medication injected and the time between PVP exposure and kidney biopsy, were not available to us. Our study benefits from the relatively large number of cases, but it could still be underpowered to detect significant correlations. Furthermore, we were not able to investigate any of the biopsies with lipid stains, as fresh frozen tissue was not available to us.

In conclusion, IV use of a specific oral methadone syrup has caused PVP deposition in the kidney in persons with opioid addiction and IV drug use. Clinicopathologic findings suggest an association between PVP deposition, tubular atrophy, and CKD. This specific methadone syrup containing high molecular weight PVP is no longer marketed, but moderate molecular weight PVP is an excipient in several opioid substitution drugs and other medications with addictive properties. As injection of oral medication is widespread among persons with IV drug use (39), it is likely that cases with deposits of excipients, such as PVP, are underdiagnosed and under-reported, as are kidney diseases in general in this patient group (44).

Disclosures

S. Leh reports speaker fees from Genzyme and other interests or relationships with the executive committee for the European Society of Digital and Integrative Pathology. H.-P. Marti reports research funding from Alexion, Amicus, Sanofi-Genzyme, and Takeda-Shire and other interests or relationships with the American Society of Nephrology, European Renal Association, the Norwegian Medical Association, the Norwegian Society of Hypertension, the Norwegian Society of Nephrology, and the Swiss Society of Nephrology. F.P. Reinholt reports other interests or relationships with the Norwegian Medical Association, the Norwegian Society for Pathology, and the Renal Pathology Society, United States. E. Svarstad reports honoraria from Sanofi Genzyme. All remaining authors have nothing to disclose.

Funding

Helse Vest (Western Norway Health Authority) grant 912001 funded this study.

Author Contributions

S. Leh conceptualized the study; I.V. Stalund was responsible for data curation; H. Grønseth, F.P. Reinholt, I.V. Stalund, and E. Svarstad were responsible for investigation; I.V. Stalund was responsible for formal analysis; H. Grønseth, S. Leh, F.P. Reinholt, and I.V. Stalund were responsible for methodology; S. Leh was responsible for project administration; H. Grønseth was responsible for validation; I.V. Stalund was responsible for visualization; S. Leh was responsible for funding acquisition; S. Leh and H.-P. Marti provided supervision; I.V. Stalund wrote the original draft; and H. Grønseth, S. Leh, H.-P. Marti, F.P. Reinholt, and E. Svarstad reviewed and edited the manuscript.