Unnecessary harm is avoided by reliable paediatric index of mortality2 scores without arterial gas sampling

Aim To investigate whether unnecessary harm could be avoided in children admitted to paediatric intensive care ( PICU ), we analysed the impact of arterial blood gas on the paediatric index of mortality score2 ( PIM 2) and the derived predicted death rate ( PDR ). Methods From January 1, 2008 to December 31, 2010, 1793 consecutive admissions, newborn infants to 16 years of age (median 0.71 years) from a single, tertiary PICU in Gothenburg Sweden, were collected. Admission information on arterial oxygen tension (PaO 2 ) and fraction of inspired oxygen (FiO 2 ) was extracted from 990 admissions. Results There was close agreement between PIM 2 score and PDR regardless of whether the PaO 2 /FiO 2 ratio was omitted or not. In the subgroup of admissions with a respiratory admission diagnosis, the inclusion of the PaO 2 /FiO 2 ratio increased the accuracy of the PIM 2 score as well as the PDR . The standard mortality ratio was slightly but not significantly overestimated by excluding the PaO 2 /FiO 2 ratio. Conclusion To avoid unnecessary harm to children admitted to PICU , an arterial blood gas analysis should only be performed if clinically indicated or if the child has a respiratory admission diagnosis. Estimation of the PIM 2 score and PDR will not be less accurate by this approach.