Study Design and Cohort
The Institute of Legal Medicine of Goethe University is responsible for autopsies in the Frankfurt/Main metropolitan area (3.5 million inhabitants), Germany. All individuals who underwent forensic autopsy by order of the judicial authorities between January 1, 2009, and April 30, 2011, because of suspected connection with illicit drug abuse were identified using the databases of the Institute of Legal Medicine. All cases were analyzed retrospectively using investigational reports provided by the judicial authorities. Investigation at the scenes was done by the criminal investigation department of the police, including securing evidence and interviewing witnesses, relatives, and general practitioners regarding various parameters, for example, age, sex, and information about abuse behavior.
Additionally, previous medical records from outpatient clinics and hospitalizations were retrieved from patient databases of the Goethe University hospital and affiliated teaching hospitals. Medical diagnoses of pre-existing disease conditions and drug dependency are based on International Classification of Diseases, 9th and 10th Revisions, respectively. Cause of death was rated in synopses of police investigations, autopsy findings, and toxicologic analyses in the final medico-legal report. Approval for this study was obtained from the institutional ethics committee of Goethe University.
All autopsies were performed by 2 forensic medical physicians according to Section 87 subs. 2 of the German Code of Criminal Procedure in accordance with the rules of the European Council in Legal Medicine regarding medico-legal autopsies and the DIN EN ISO 17025 standard, including the collection of hair, gastric contents, and urine and blood samples (obtained from the cardiac chambers and femoral veins) for further toxicologic analysis (Item S1, available as online supplementary material).
Remnant blood samples were analyzed further for HBV surface antigen, HCV antibodies, and HIV 1 and 2 antibodies by chemiluminescent microparticle immunoassays and defined as positive when confirmed by western blot (HCV antibodies and HIV antibodies) or neutralization test (HBV surface antigen). Antigen detection by polymerase chain reaction could not be performed because of the high viscosity of postmortem blood samples. Tissue samples were obtained from all organs, including the kidneys, and formalin fixed for further evaluation.
Histologic Evaluation of the Kidneys
Histologic specimens of 129 drug abusers were evaluated. For every paraffin block, hematoxylin and eosin, periodic acid–Schiff, Sirius, Congo Red, and, when calcifications were seen, von Kossa staining were performed. Additionally, immunoglobulin A (IgA) immunohistochemistry was performed for every case using an IgA-specific polyclonal rabbit anti-human antibody (1:150,000 dilution; Dako Cytomation) on a Ventana BenchMark ULTRA stainer immunohistochemistry device (Roche). In selected cases, additional immunohistochemical staining was performed, with polyclonal antibodies specific for IgG (1:100,000 dilution; Dako) and complement component C3c (1:75,000 dilution; Dako) were performed when glomerulonephritis was suspected. A monoclonal antibody specific for amyloid A (1:500 dilution, clone mc1; Dako) was applied in amyloidosis cases.
The following histologic parameters were evaluated: number of obliterated glomeruli per 100 counted glomeruli; presence of segmental sclerosis/scarring/segmentally accentuated increase in cellularity and matrix (yes/no), increase in mesangial matrix (yes/no), and signs of glomerular ischemia (wrinkling of basement membranes and widened multilayered Bowman capsules; yes/no); degree of interstitial fibrosis and tubular atrophy estimated as a percentage and arteriosclerosis and arteriolosclerosis semiquantitatively separated into 6 groups (0, no sclerosis; 1, mild; 2. mild to moderate; 3, moderate; 4, moderate to severe; and 5, severe); presence of interstitial inflammation (yes/no) with all degrees, even very mild, of inflammation included, except for minimal infiltration in the proximity of larger vessels; presence of parenchymal calcifications (yes/no) and whether they were located in the renal papillae/deep medulla or along tubular basement membranes; and, as signs of analgesic nephropathy, capillary sclerosis of the medulla/papillae and suburothelial soft tissue of the pelvis when present (n = 50) and signs of papillary necrosis tips of the papillae when present (n = 46).
Because of autolysis, the degree of tubular damage (acute tubular necrosis) was not evaluated. Additional electron microscopic investigations were performed for 7 cases, if the light microscopic diagnosis was ambiguous or in order to confirm amyloidosis.
Demographic, clinical, and laboratory variables were expressed as median with interquartile range (IQR) or as proportion, as appropriate. Continuous and categorical variables were compared for univariable analysis between groups using t test or Mann-Whitney U test and Fisher exact test, respectively. Associations between the presence of renal damage and demographic, clinical, and laboratory variables were estimated by stepwise multivariable logistic regression with an elimination criterion >0.1. All P values reported are 2 sided. Statistical significance was assumed for P < 0.05. All statistical analyses were performed using BiAS, version 10.04 (Epsilon-Verlag).
Am J Kidney Dis. 2014;63(6):945-953. © 2014 The National Kidney Foundation
The National Kidney Foundation
Published by Elsevier