Drug-induced acute tubulointerstitial nephritis: Serial C-reactive protein measurements might predict the course of acute kidney injury.

Background: Elevated C-reactive protein (CRP) was proposed as a biomarker instead of renal biopsy in drug-induced acute tubulointerstitial nephritis (D-ATIN) in adults. However, there is no study in which patients were followed up with serial CRP without kidney biopsy. We evaluated the significance of CRP levels in predicting the clinical course of D-ATIN in children.

Materials and methods: Children with a clinical diagnosis of D-ATIN were evaluated for sex, age, symptoms, offending drug(s), latent period after exposure, blood pressure, urine output, urinalysis, eosinophiluria, urine albumin/total protein, serum creatinine, and CRP at presentation. In addition, serial CRP and creatinine levels were recorded daily until discharge and thereafter. Correlations of CRP with creatinine and of peak CRP with creatinine normalization time were evaluated.

Results: There were 13 patients (8 female, median age 15 years). None had oliguria or hypertension. Median CRP and creatinine at presentation were 36 mg/L and 1.6 mg/dL, respectively. Median times to peak CRP and creatinine were 6 and 7 days after drug exposure, respectively. The decrease in CRP preceded the fall in creatinine. Median CRP and creatinine normalization times after their peaks were 5 and 14 days, respectively. None required renal biopsy or corticosteroid treatment. CRP was correlated with creatinine throughout the study period. However, peak CRP was not correlated with creatinine normalization time.

Conclusion: CRP was increased at presentation and decreasing CRP predicted favorable outcome. Renal biopsy and corticosteroid treatment could be postponed safely in these children.