Presented by Dr Natalie YW LEUNG at the daily Journal Club, ICU, PYNEH, Hong Kong, on 15 May 2009
In the critical appraisal of an open-label randomized trial of oropharyngeal cleansing with 0.2% chlorhexidine (n = 224) vs 0.01% potassium permanganate (n = 247) as control for the prevention of nosocomial pneumonia in critically ill patients (see CHEST May 2009 vol. 135 no. 5 1150-1156), which of the following point is false?
A. The intended statistical power was 75%, lower than the usual of at least 80 – 90%
B. A Hawthorne effect was present
C. The most frequent diagnoses in the study patients were COPD, pneumonia and heart failure
D. No statistical difference was found between the chlorhexidine and the potassium permanganate group
Answer: C
Editor’s and the discussants’ impression
The intended statistical power to be detected was less than the conventional of at least 90 – 90%, and the difference to be detected is a 50% (relative) reduction. An incidence of nosocomial pneumonia rate of 16% in the control group was assumed, lower than the the actual incidence of 7.7%. Even with such conservation assumption, the sample size required has to be 506 subjects (i.e. 253 in each treatment group). One has to take into account these factors in the critical appraisal of this paper.
A big Hawthorne effect was likely present. In the non-study period of using potassium permanganate cleansing, the baseline nosocomial pneumonia rate was as high as 21.7%, but during the study, in the control group (i.e. potassium permanganate group), the rate was only 7.7%.
The most frequent diagnoses in this Indian study were malaria, Guillain-Barré syndrome, ARDS and Leptospirosis.
Reference
Tanmay S. Panchabhai, MBBS, Neha S. Dangayach, MBBS, Anand Krishnan, MD, Vatsal M. Kothari, MD and Dilip R. Karnad, MD*. Oropharyngeal Cleansing With 0.2% Chlorhexidine for Prevention of Nosocomial Pneumonia in Critically Ill Patients. An Open-Label Randomized Trial With 0.01% Potassium Permanganate as Control. CHEST May 2009 vol. 135 no. 5 1150-1156