Evaluation of POSSUM and P-POSSUM scoring systems in assessing outcome after laparoscopic colectomy (Br J Surg 2003; 90: 1280–1284)

Sir Senagore et al suggested that P-POSSUM becomes accurate when laparoscopic colorectal surgery (LAC) is considered to involve a minimal stress response. Whether LAC is actually minimally invasive remains controversial and needs to be assessed by further studies1. In France a multicentre prospective study has been performed on mortality after colorectal surgery. P-POSSUM overpredicted mortality two-fold in elective colorectal surgery. For open surgery (n = 1143), expected mortality (95 per cent confidence interval) was higher than observed mortality: 5·4 per cent (4·9–5·9) versus 4·1 per cent (P = 0·14). For LAC (n = 261), expected mortality was also higher than observed mortality: 1·5 per cent (1·2–1·7) versus 0·4 per cent. When the correct P-POSSUM formula was applied as suggested by Senagore et al, in contrast to their results, P-POSSUM still overpredicted mortality two-fold: 0·9 per cent (0·7–1·1) versus 0·4 per cent (P = 0·37). This lack of statistical significance may have been due to lack of power, we calculated that 3000 patients needed to be included for reliability. The authors' hypothesis was thus not confirmed in France. On the other hand, due to patient selection, the mortality rate after LAC was so small, that developing a predictive score in this setting will be a very hard task.