Anesth Analg 2007;105:287
© 2007 International Anesthesia Research Society
doi: 10.1213/01.ane.0000263015.53783.5c
LETTER TO THE EDITOR
Early Termination Penalty
David Wax, MD
Mount Sinai School of Medicine; New York, NY; David.Wax{at}msnyuhealth.org
To the Editor:
Roy et al. (1) reported promising results from an investigation of the efficacy of single-shot intrathecal opioids for analgesia in addition to patient-controlled analgesia (PCA) after liver resection. I subsequently tried this technique in three patients, one of whom failed to meet extubation criteria and required many hours of postoperative mechanical ventilation to manage persistent hypoventilation, a problem most likely related to the spinal morphine.
In their discussion, the authors suggested that the lack of a significant difference in adverse effects between the groups might have been due to their small sample size. Their intent was to study 20 patients in each group, but they stopped enrollment halfway through because an interim analysis showed superior results in the study group when compared with the control group. Presumably they felt that it would be unethical to subject any more patients to the inferior modality of pain management of the control group—that is, the PCA-only modality which represents a standard of care for postoperative analgesia around the world.
Another study (2) compared lactated Ringer's (LR) and normal saline (NS) solutions in renal transplantation. It too was terminated after only a quarter of the planned 100 patients were enrolled in each group because interim analysis showed the LR group to have better intraoperative laboratory profiles than the NS (standard care) group. Apparently the investigators felt that it was unsafe to administer NS to any more patients despite the large number of transplant recipients around the world that have already received NS without adverse outcomes and without knowledge of the laboratory values that concerned investigators (because intraoperative blood testing is typically utilized less frequently than in this study protocol).
In both studies, the investigators (or their institutional review boards) terminated enrollment early. In doing so, they effectively rejected the current standards of care as being intolerable or unsafe. Yet neither of these reports thoroughly convinces us that we should immediately adopt new standards of care. (Based only on these small studies, I do not believe that it would be cruel to provide PCA-only analgesia to liver resection patients, and I think it would be premature to abandon NS in renal transplant patients.) When a novel therapy causes serious adverse effects, a decision to stop the study is warranted. However, when it is the time-tested standard of care that appears inferior or unsafe in a small sample despite collective clinical experience and higher-powered studies to the contrary, termination seems overcautious. The early terminations of these studies represent missed opportunities to better characterize the very benefits and risks the investigators set out to study and to offer us more convincing conclusions.
REFERENCE
- Roy JD, Massicotte L, Sassine MP, et al. A comparison of intrathecal morphine/ fentanyl and patient-controlled analgesia with patient-controlled analgesia alone for analgesia after liver resection. Anesth Analg 2006;103:990–4.[Abstract/Free Full Text]
- O'Malley CM, Frumento RJ, Hardy MA, et al. A randomized, double-blind comparison of lactated Ringer's solution and 0.9% NaCl during renal transplantation. Anesth Analg 2005;100:1518–24.[Abstract/Free Full Text]
|
