© 2007 American Thoracic Society
Against Misleading Predictions for Severe Community-acquired PneumoniaTo the Editor:The definition of severe community-acquired pneumonia (SCAP) has been hampered by the lack of a generally applicable reference for the disease. In 1998, we advocated an approach that derived a prediction of SCAP from the identification of factors behind the clinical decisions to admit a patient with CAP to a respiratory ICU (1). This approach was adopted by the American Thoracic Society as a clinical guide for ICU admission (the modified American Thoracic Society [m-ATS] rule) (2). In fact, it remains a questionable one mainly because of the high variations in use of ICU resources across regions and nations. España and coworkers in their recent article propose a new prediction rule for SCAP (3). Unfortunately, there is concern that this rule will cause additional confusion. First, España and coworkers have selected mortality and/or mechanical ventilation and/or septic shock as reference for SCAP. This means by definition that pneumonias presenting with either acute respiratory failure not requiring mechanical ventilation or severe sepsis are not SCAP. While these patients probably represent the population at highest risk of early deterioration as well as with the highest chance of gaining benefit from intensive care, the authors focus on the strongest endpoints just for achieving high sensitivity. Accordingly, of the 414 "false positives" within 526 patients classified as SCAP in the entire cohort, 82% in fact had severe sepsis! Overall, the selection of patients seems to result in predictions of definitions of acute respiratory failure and severe sepsis/septic shock we already use: it is a case of circular reasoning. The endpoints of mechanical ventilation and septic shock are deconstructed into several predictors, and many calculations are needed to reconstruct them. Second, the claim of the authors to have derived a rule that provides better predictions than those defined so far is not correct. Their Table 4 clearly shows that in the external validation group, the most critical one in terms of validation, the proposed rule did not provide significantly better predictions. As regards the m-ATS rule, prediction wasn't better even in the internal validation group. Third, one of the most challenging issues in the care of patients with CAP is to identify those who present with mild-to-moderate CAP but in fact are at risk of early deterioration. The detection of these patients by the rule is only made at the cost of a positive predictive value of around 22%, which means that overtreatment is a potential issue in a large majority of patients classified as SCAP by the rule. In conclusion, in clinical practice, we should continue to rely mainly on the assessment (and repeated assessment!) of acute respiratory failure and severe sepsis/septic shock.
Thoraxzentrum Ruhrgebiet, Bochum, Germany FOOTNOTES Conflict of Interest Statement: S.E. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
