ABSTRACT
Title
Risk of potentially drug-induced hepatic injury in paediatric population: data mining on electronic healthcare databases in Europe
Authors
C. Ferrajolo1,2, K.M.C. Verhamme2, A. Capuano1, G. Trifirò2,3, A. Oteri2, G. Picelli4, G. Mazzaglia5, C. Cricelli5, F. Rossi1 and M.C.J.M. Sturkenboom2.
- Pharmacovigilance and Pharmacoepidemiology Centre of Campania Region, Dept. of Experimental Medicine, Pharmacology Section, Second University of Napoli, Napoli, Italy.
- Pharmacoepidemiology Unit, Dept. of Medical Informatics, Erasmus University Medical Center, Rotterdam, Netherlands.
- Dept. of Clinical and Experimental Medicine and Pharmacology, University of Messina, Messina, Italy.
- International Pharmacoepidemiology and Pharmacoeconomics Research Centre, Desio, Italy.
- Società Italiana di Medicina Generale, Firenze, Italy.
Abstract
Drugs are one of most important causes of hepatic injury in paediatrics but so far, the incidence rate of drug-related hepatic injury has only been estimated in the general population. A retrospective population-based cohort study was performed to estimate the incidence rates of hepatic injury in children and to identify the drugs associated with hepatic injury.
Between June 2000 and December 2008 data on population aged 1-18 years were collected using three general practice databases in two European countries: in the Netherlands, the Integrated Primary Care Information (IPCI) database, and in Italy, Pedianet (covering children up to 14), and Health Search/Thales (HSD) (covering children from 14 years). Potential cases of hepatic injury were identified through codes, free text and lab values search. All potential cases were manually validated using specific algorithms by two assessors, blinded to drug exposure. Crude incidence rates (IR; 95% CI) of hepatic injury by country and age and gender-adjusted relative risks (RR; 95% CI) were estimated using Jerboa (http://www.euadr-project.org).
In the study cohort of 346,490 subjects aged 1-18 years, contributing 1,260,506.4 person-years (PY) of follow-up, we identified 777 new definite cases of hepatic injury (46 cases in IPCI, 731 cases in HSD and Pedianet). The overall incidence rate for potentially drug-induced hepatic injury was 61.7 (95% 57.5-66.2) per 100,000 PY. Stratifying by each country, the incidence rate was higher in Italy (IR 73.0; 67.8-78.4) per 100,000 PY than in the Netherlands (17.9; 13.3-23.7) per 100,000 PY. Pooled data showed a statistically significant (p <<0.05) age and gender adjusted increased risk of hepatic injury for antimycobacterials (ATC code J04: RR 598.0; 95% CI 267.6-1336.2), followed by psycholeptics (N05: 15.7; 6.5-37.9), agents for acid related disorders (A02: 14.2; 5.9-34.2), antibacterials (J01: 13.7; 10.1-18.5), antiepileptics (N03: 9.3; 4.6-18.8), drugs for obstructive airway diseases (R03: 7.1;4.5-11.0), and NSAIDs (M01: 6.4; 2.4-17.1).
The incidence of potentially drug-induced hepatic injury is rare in the paediatric population and heterogeneous between the two European countries. Antimycobacterials, psycholeptics, antibacterials, antiepileptics, drugs for obstructive airway diseases, and NSAIDs are associated with an increased risk of hepatic injury.
Between June 2000 and December 2008 data on population aged 1-18 years were collected using three general practice databases in two European countries: in the Netherlands, the Integrated Primary Care Information (IPCI) database, and in Italy, Pedianet (covering children up to 14), and Health Search/Thales (HSD) (covering children from 14 years). Potential cases of hepatic injury were identified through codes, free text and lab values search. All potential cases were manually validated using specific algorithms by two assessors, blinded to drug exposure. Crude incidence rates (IR; 95% CI) of hepatic injury by country and age and gender-adjusted relative risks (RR; 95% CI) were estimated using Jerboa (http://www.euadr-project.org).
In the study cohort of 346,490 subjects aged 1-18 years, contributing 1,260,506.4 person-years (PY) of follow-up, we identified 777 new definite cases of hepatic injury (46 cases in IPCI, 731 cases in HSD and Pedianet). The overall incidence rate for potentially drug-induced hepatic injury was 61.7 (95% 57.5-66.2) per 100,000 PY. Stratifying by each country, the incidence rate was higher in Italy (IR 73.0; 67.8-78.4) per 100,000 PY than in the Netherlands (17.9; 13.3-23.7) per 100,000 PY. Pooled data showed a statistically significant (p <<0.05) age and gender adjusted increased risk of hepatic injury for antimycobacterials (ATC code J04: RR 598.0; 95% CI 267.6-1336.2), followed by psycholeptics (N05: 15.7; 6.5-37.9), agents for acid related disorders (A02: 14.2; 5.9-34.2), antibacterials (J01: 13.7; 10.1-18.5), antiepileptics (N03: 9.3; 4.6-18.8), drugs for obstructive airway diseases (R03: 7.1;4.5-11.0), and NSAIDs (M01: 6.4; 2.4-17.1).
The incidence of potentially drug-induced hepatic injury is rare in the paediatric population and heterogeneous between the two European countries. Antimycobacterials, psycholeptics, antibacterials, antiepileptics, drugs for obstructive airway diseases, and NSAIDs are associated with an increased risk of hepatic injury.