[Estimation of creatinine clearance by the height/plasma creatinine formula in critically-ill children]

Critically-ill children frequently show impaired renal function, necessitating adjustment of drug dosages. Our objectives were to study estimated creatinine clearance through the correlation between the height/plasma creatinine formula (CrClest) and measured creatinine clearance (CrClms) and to examine whether CrClest over- or underestimates CrClms by analyzing the influence of diagnosis, severity, and the practical consequences.

Seventy-seven patients admitted to the pediatric intensive care unit were included. CrClms was calculated using serum creatinine and creatinine in urine collected over 24 hours. CrClest was estimated using serum creatinine, height, and a constant. The difference between CrClms and CrClest was expressed as a percentage: (CrClms CrClest) x 100/CrClms. Differences of greater than 15 % were considered poor estimates. ResultsThe mean percentage difference was 29.2 (standard error: 39.9). There were no differences among diagnoses in the distribution of significant bias, although the frequency of metabolic diagnoses was high. Incorrect evaluation of CrClest would result in a therapeutic error in 11.69 % of the cases, with overdosage in 10.39 %. The Pediatric Risk of Mortality (PRISM) score was higher (p < 0.05) in patients at risk for overdosage.

CrClest estimation using the height/plasma creatinine formula was not an accurate method in critically ill children. In 10.39 % of patients with more severe illness, the dosage of renally excreted drugs would be too high. The highest risk was found in patients with metabolic and neurological diagnoses.