Free
Correspondence  |   September 1998
Is Lack of Statistical Power Always Evidence of Lack of Effect?
Author Notes
  • Assistant Professor of Anesthesiology; University of Milano; Milan, Italy; tommasino.concezione@hsr.it
Article Information
Correspondence
Correspondence   |   September 1998
Is Lack of Statistical Power Always Evidence of Lack of Effect?
Anesthesiology 9 1998, Vol.89, 798-799.. doi:
Anesthesiology 9 1998, Vol.89, 798-799.. doi:
To the Editor:-I read with interest the paper by Pittman et al. [1] recently published in Anesthesiology. The study in the article shows that rats undergoing 75-min middle cerebral artery occlusion during pentobarbital or propofol anesthesia, in doses sufficient to maintain electroencephalogram burst suppression, have similar neurologic and histologic outcomes. The study is well designed, and I think that this research is extremely important.
As stated in the article by the authors, the neurologic scores between the two groups were not significantly different. However, when neurologic scores are analyzed in detail (their fig. 2), animals treated with pentobarbital seem to have a better neurologic outcome. In fact, seven pentobarbital-treated animals had a neurologic score of 1 (as compared to one propofol-treated animal), and fewer pentobarbital-treated rats had a score of 2 (four versus seven) or a score of 3 (six versus nine). If we pool the results (0 to 1 versus 2 to 3 neurologic score), 10 animals treated with pentobarbital versus 3 animals treated with propofol had a "good" neurologic score (0 to 1, no deficit or left forelimb flexion only), and 10 versus 16 had a more severe hemiparesis (2 to 3 neurologic score).
The authors did not attempt to "force" the results in any direction. I also agree that histologic results (infarct areas) provide a better end point when considering the protective cerebral effects of any drug or treatment, or both. However, although the aforementioned differences are not statistically significant, they should be considered. The authors did not provide any correlation between histology (infarct area) and neurologic deficit. I would speculate that such scores are congruent, with smaller cerebral infarct size correlating with better neurologic outcome. However, if this is the case, I wonder which drug provides a better correlation.
Concezione Tommasino, M.D.
Assistant Professor of Anesthesiology; University of Milano; Milan, Italy; tommasino.concezione@hsr.it
(Accepted for publication May 5, 1998.)
REFERENCES
Pittman JE, Sheng H, Pearlstein R, Brinkhous A, Dexter F, Warner DS: Comparison of the effects of propofol and pentobarbital on neurologic outcome and cerebral infarct size after temporary focal ischemia in the rat. Anesthesiology 1997, 87:1139-44