Failure of APACHE II alone as a predictor of mortality in patients receiving total parenteral nutrition

We followed prospectively over 5 months all medical and surgical ICU patients placed on total parenteral nutrition (TPN) and recorded their Acute Physiology and Chronic Health Evaluation II (APACHE II) scores on the day of admission, on the day TPN was started, length of time in ICU before TPN was started, and the number of days TPN was administered. Sixty-one patients (15 to 82 yr) had an inhospital mortality of 47%. The APACHE II score was significantly higher for nonsurvivors vs. survivors both on the day of admission (24.4 +/- 9.6 vs. 18.4 +/- 6.5; p less than .003) and also on the day TPN was started (21 +/- 8.6 vs. 16.4 +/- 5.6; p less than .002). However, at a 60% risk of dying, specificity was 96.9% and sensitivity 27.6%. The mean number of days before TPN was started was 3.2 and mean number of days on TPN was 9.2 (p = NS). We conclude that calculation of APACHE II score either on ICU admission or on the day TPN is considered does not seem useful in identifying patients who will not benefit from TPN.