Article Text

Download PDFPDF
PS-073 Patient records analysis for potentially preventable adverse drug events leading to acute kidney injury following a propensity matched cohort study
  1. S Amelung1–3,
  2. D Czock2,
  3. M Thalheimer4,
  4. T Hoppe-Tichy1,3,
  5. WE Haefeli2,3,
  6. H Seidling2,3
  1. 1Heidelberg University Hospital, Pharmacy Department, Heidelberg, Germany
  2. 2Heidelberg University Hospital, Department of Clinical Pharmacology and Pharmacoepidemiology, Heidelberg, Germany
  3. 3Heidelberg University Hospital, Cooperation Unit Clinical Pharmacy, Heidelberg, Germany
  4. 4Heidelberg University Hospital, Department of Quality Management and Medical Controlling, Heidelberg, Germany


Background Relevant inhospital adverse drug events (ADE) are often documented in clinical administrative data (CAD) using ICD-10 codes (International Classification of Diagnosis Related Diseases, 10th revision). In a previous propensity matched cohort study, we analysed the CAD of 48 072 inpatients of a university hospital for potentially preventable inpatient ADE affecting length of stay. From a hospital’s perspective, particularly ICD-10 codes coding for drug induced renal failure, appeared to be preventable.

Purpose We aimed to evaluate causes and conditions leading to renal failure in hospital and develop prevention strategies.

Material and methods We assessed the validity of such codes in patient records and evaluated whether acute renal failure occurred during the hospital stay. Using the updated causality score by Bégaud et al, we currently assess, using 2 independent reviewers, whether acute renal failure was drug related.1 Based on root cause analyses, preventive strategies will be developed.

Results The records of 69 patients were analysed (mean age 62 years (range 23–94), 33% women). 26 cases (38%) had a known history of renal failure or were hospitalised because of acute renal failure. In 43 cases (62%), the adverse event occurred in hospital. Nearly half of these cases (n=20) had a known history of renal failure. The pilot results of four randomly selected cases of this ongoing assessment revealed 6 suspect drugs with a high imputability to the ADE (1 drug with level I4 and 5 drugs with level I6 out of 7 possible scores from I0 to I6). As conceivable prevention strategies, we identified a priori dose adjustment and/or longer intravenous application duration for nephrotoxic drugs (eg, aciclovir) in patients with a known history of renal failure.

Conclusion Unless CAD do not explicitly flag inpatient ICD-10 codes, CAD based ADE identification is laborious, and adequate risk management by the hospital is challenging. Screening for (pre-existing) renal (dys)function at the stage of hospitalisation for appropriate dose adjustment could prove a promising preventive strategy for our hospital.

References and/or acknowledgements 1. Arimone Y, et al. Updating the French method for the causality assessment of adverse drug reactions. Therapie2013;68:69–76.

No conflict of interest

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.