Acute Physiology Score prediction in patients with traumatic brain injury
Georgios Matis¹, Olga Chrysou², Theodossios Birbilis¹, Nikolaos Lyratzopoulos³, Georgios Minopoulos³, Konstantinos Manolas^3
1Department of Neurosurgery, Democritus University of Thrace
²First Pneumonology Department, General Hospital of Kavala
³First Department of Surgery, Democritus University of Thrace, Alexandroupolis, Greece

SUMMARY. Background: APACHE-II is one of the severity-of-illness scoring systems. It describes in an arithmetical manner the severity of injury by calculating 12 variables, the Acute Physiology Score (APS), age and the chronic health status. Objectives: The aim of this study was to find a regression equation to predict the APS from a limited number of variables in patients with traumatic brain injury (TBI). Population and Method: The characteristics were studied of 74 subjects admitted to the Intensive Care Unit (ICU) of the University General Hospital of Alexandroupolis, Greece, during the decade 1994-2003 with the diagnosis of "TBI". The subjects had a mean age of 45.19±2.55 years, 65 were male (87.8%) and 9 female (12.2%) and their mean Glasgow Coma Scale (GCS) score was 6.39±0.55. Statistical analysis was made using stepwise regression, with APS as the dependent variable and the 12 physiological variables and age being the independent variables. The following selection criteria were also taken into consideration: Akaike, Amemiya, Mallows' and Schwarz. Results: The data analysis provided the following equation: APS = 148.843 - (1.074*GCS) - (16.878*pH) - (0.190*HCT) + (4.082*CR), where HCT is the haematocrit and CR is serum creatinine. The statistics of the equation were: F=116.304 (p=0.000), adjusted coefficient of determination (R'²)=0.863 and standard error of the estimate=2.624. The corresponding values for Akaike, Amemiya, Mallows' and Schwarz selection criteria were: 147.575, 0.148, 3.053 and 159.095. Conclusions: These results suggest that it is possible to estimate the APS value based on four variables only, thus saving time in APACHE-II calculation. In this series, 86.3% of the APS change was due to the four aforementioned variables, while the remaining 13.7% was due to other factors which are not entered in the equation. Pneumon 2008; 21(4):374-379.
Key words: Traumatic Brain Injury (TBI), APACHE-II, Acute Physiology Score (APS), regression.
Correspondence: Georgios K. Matis, 1st Emmanouil Roidi Rd, GR-542 48, Thessaloniki, Tel.: +302310-307588, +30-6977990208, e-mail: gkmatis@yahoo.gr

FULL TEXT