Do antiemetic drugs benefit adult emergency department patients with nausea? The literature says no, but is it right?

Abstract

Nausea is a common problem in ED patients. Antiemetic drugs have been used in the ED for decades, but a recent Cochrane review found no convincing evidence for the benefit of antiemetic drugs over placebo. This was largely based on three placebo-controlled trials, which found mean Visual Analog Scale (VAS) changes for various drugs and placebo, to be similar. However, reliance on mean VAS change as the primary outcome measure has probably been a mistake. It does not give information on the number of improved patients, so these cannot be compared between groups. Alternative primary outcome measures warrant further exploration. Use of a VAS cut-off level indicative of clinically significant symptom improvement would allow comparison of numbers of patients with improved nausea ratings. This is proposed as the best option currently available. Preliminary testing of this outcome measure suggests that the conclusions of past studies may be misleading, and that the question of antiemetic efficacy for ED patients is not yet answered.

Randomized Placebo‐controlled Trial of Droperidol and Ondansetron for Adult Emergency Department Patients With Nausea

“…The primary objective of antiemetic treatment for ED patients with nausea is clinically significant symptom improvement. Mean group VAS change does not provide direct evidence for this objective . Recent research has demonstrated that ED patients with symptom improvement (“a little less” or “a lot less”) reliably report VAS reductions in excess of –5 mm .…”

“…Doubts have been expressed about these seemingly counterintuitive findings . Possible limitations of the currently used outcome measures for the detection of real differences and the difficulty of interpreting the main study results have been highlighted …”

“…Difference in mean VAS change between treatment groups has been the primary outcome measure for all three ED‐based placebo‐controlled trials conducted to date . Findings of superiority or equivalence have been based on the statistical significance of the between‐group differences, but the clinical interpretation of these results is not straightforward . The “minimum clinically significant difference” (MCSD) for nausea on the VAS, defined as the mean VAS change reported by people who describe their symptoms as being “a little less,” was intended to address this difficulty .…”

“…As these changes are greater than the MCSD, it can only be inferred that most patients in all groups were improved to some degree. Even if a between‐group difference is statistically significant, the clinical significance is difficult to determine when the mean VAS change for both groups is in excess of the MCSD …”

“…From an analysis point of view, compared with treating VAS change as a continuous variable, its reduction to a binary outcome must lessen its sensitivity for detection of between‐group differences . This may seem undesirable, but the advantage in this setting is that it will allow the primary objective of symptom improvement to be directly compared between groups . In conjunction with more standardly used outcome measures, this should aid the clinical interpretability of results.…”