Severe head injury in children: intensive care unit activity and mortality in England and Wales

Abstract

Objective: To explore the relationship between volume of paediatric intensive care unit (PICU) head injury (HI) admissions, specialist paediatric neurosurgical PICU practice, and mortality in England and Wales.

Methods: Analysis of HI cases (age <16 years) from the Paediatric Intensive Care Audit Network national cohort of sequential PICU admissions in 27 units in England and Wales, in the 5 years 2004-2008. Risk-adjusted mortality using the Paediatric Index of Mortality (PIM) model was compared between PICUs aggregated into quartile groups, first to fourth based on descending number of HI admissions/year: highest volume, medium-higher volume, medium-lower volume, and lowest volume. The effect of category of PICU interventions - observation only, mechanical ventilation (MV) only, and intracranial pressure (ICP) monitoring - on outcome was also examined. Observations were reported in relation to specialist paediatric neurosurgical PICU practice.

Results: There were 2575 admissions following acute HI (4.4% of non-cardiac surgery PICU admissions in England and Wales). PICU mortality was 9.3%. Units in the fourth-quartile (lowest volume) group did not have significant specialist paediatric neurosurgical activity on the PICU; the other groups did. Overall, there was no effect of HI admissions by individual PICU on risk-adjusted mortality. However, there were significant effects for both intensive care intervention category (p<0.001) and HI admissions by grouping (p<0.005). Funnel plots and control charts using the PIM model showed a hierarchy in increasing performance from lowest volume (group IV), to medium-higher volume (group II), to highest volume (group I), to medium-lower volume (group III) sectors of the health care system.

Conclusions: The health care system in England and Wales for critically ill HI children requiring PICU admission performs as expected in relation to the PIM model. However, the lowest-volume sector, comprising 14 PICUs with little or no paediatric neurosurgical activity on the unit, exhibits worse than expected outcome, particularly in those undergoing ICP monitoring. The best outcomes are seen in units in the mid-volume sector. These data do not support the hypothesis that there is a simple relationship between PICU volume and performance.

Figure 1

Calibration plots for PIM. The two plots show the observed risk by expected risk in deciles of expected risk using a log scale for the two versions of PIM.

Figure 2

England and Wales PICU head injury practice by unit. Upper panel: Practice ordered according to size from smallest to largest. Lower panel: Cumulative practice by unit presented as percentage. The dotted lines in the lower panel show the borders of each quartile. The dotted lines in the upper panel show the size of practice defined at these quartile borders.

Figure 3

Specialist paediatric neurosurgical practice by PICU. Grouping of PICUs into quartiles defined in Fig. 2 from upperfirst to lowest-fourth, groups I to IV. PICU numbering is the same as in Fig. 2.

Figure 4

Funnel plots showing risk-adjusted mortality rate displayed as a scatter plot. The horizontal line shows the predicted mean 8.6%. Dotted lines show the 99.8% and 95% (2σ) limits. Points I–IV in the upper panel use the same notation as described in Fig. 3. Lower panel uses the same scheme of symbols, with each point representing PICUs within respective quartile-sectors of the health care system.

Figure 5

Half-cumulative-sum-risk-adjusted charts in the four quartiles of PICUs over their respective case series sequence, where: A, units 1–14, lower quartile and Group IV; B, units 15–20, third quartile and Group III; C, units 21–24, second quartile and Group II; D, units 25–25, upper quartile and Group I. Upper red line in each control chart tests for doubling of odds of death (h=4.6). Lower blue line in each control chart tests for halving of the odds of death (h= −4.6).