Authors’ reply

Excerpt

Sir: We agree with Berghmans et al. that differences in hospital mortality among the two study populations might be attributed to a dissimilar case mix and, furthermore, to possibly different decision making on whom to admit to the ICU. The same reasons might apply to the contradictory results with respect to the performance of the scores. Furthermore, the fact that both studies were carried out in single center institutions including a relatively small number of patients might have influenced the results, as well as different methods of analysis and the amount of missing data entered into score calculation. In our opinion, the comparison of these two studies suggests that no scoring system is yet able reliably to predict outcome for cohorts of patients, not to speak of individual outcome. Importantly, Berghmans et al. could show once more that not the characteristics of the underlying type of cancer but rather the acute physiologic alteration has an impact on outcome. Therefore it is important not to withhold intensive care treatment from patients with underlying malignant diseases. The available scoring systems should not be used for individual clinical decision making. …