In this third EMC Journal Club (where we take the “boring” out of journal clubs and deliver clear, concise, practical critical appraisal knowledge based on an Emergency Medicine journal article that may have flown by your radar – not too detailed and not too brief), Dr. Rohit Mohindra, an Emergency Physician at North York General in Toronto and SREMI researcher works his critical appraisal magic on the article “Dexamethasone and ketorolac compare with ketorolac alone in acute renal colic: A randomized clinical trial” by Razi et al. Plus, for the EBM keeners, we have Dr. Shelley McLeod, clinical epidemiologist at SREMI give us a research methodology hot take on what needs to be considered when a statistical testing is done to compare the experimental and the control group in randomized clinical trials.
To sign up for an email blast with the EMC Journal Club of the month delivered to your inbox go to our subscribe page
Dexamethasone and ketorolac compared with ketorolac alone in acute renal colic: A randomized clinical trial
Case
You are assessing an otherwise healthy 47M with acute onset flank pain. Imaging confirms urolithiasis, and you start to order medications to help with his pain. You wonder if adjunct medications can reduce the need for opioids or additional analgesia.
Summary
Patients: Adults with a diagnosis of renal colic (although no standard definition was given)
Intervention: 8mg of dexamethasone plus 30mg ketorolac IV
Comparison: Placebo plus 30mg IV ketorolac
Outcomes: Visual analog pain scale (VAS) 30 minutes after administration
Results: There was a statistically significant improvement in VAS at 30 minutes: The control group had VAS 5 (CI 3-7), and the intervention group had 3.5 (0.25-6). The reported statistical significance of p = 0.009 was below their pre-specified cutoff. However, the improvement was not sustained at 60 minutes. Although pain scores decreased to 1 in the intervention group, 35% still required narcotic rescue therapy after 60 minutes.
Study author’s conclusions: According to the study authors, adding dexamethasone provided improved pain control after 30 min of therapy and decreased opioid and antiemetic need at 60 minutes
Critical appraisal
A few things warrant attention. First, the inclusion criteria is patients with flank pain, but it’s not clear how the presence of urolithiasis was confirmed. The authors stated they used a combination of CBC, UA, diagnostic imaging and specialist opinion, but which criteria for each of these tests was not specified (and was not listed on their RCT study protocol). It would be important to know how many cases were confirmed by CT, which is the gold standard for diagnosis.
Additionally, there is the issue that although patients had low VAS pain scores, ⅓ still required rescue opioids.
Finally, from what I could find, the therapeutic effect of dexamethasone starts around 1 hour and doesn’t peak until about 18 hours [2,3], so I really wonder if the effect they report is biologically plausible. The presence of analgesia at 30 minutes but disappears at 60 minutes also suggests that the dexamethasone was not directly responsible for the observed effect.
Take home message on adding dexamethasone to ketorolac when treating acute renal colic
Despite the author’s positive conclusion, you don’t think this study will change practice. You continue with your usual combination of NSAIDs and rescue opioids.
Research Methodology Hot Take by Dr. Shelley McLeod
The biggest issue I have with this trial was the multiple significance tests of between group differences. When designing analgesia trials, pain intensity is usually the primary outcome, measured on a 0 to 10 numerical rating or visual analog scale. This outcome can be defined at a single time point (e.g., change from baseline to the end of follow-up) or across several time points. However, if outcomes at multiple time points are all deemed primary as it was in this trial, then appropriate adjustment for multiple testing is needed using an approach (e.g., Bonferroni correction) that controls the overall type I error probability, or the probability of rejecting the null hypothesis of no treatment effect for at least one of the endpoints when in fact the treatment has no effect. The authors conducted MANY between group statistical tests (see Table 2), including all the variables listed in Table 1 which should not be formally compared in a randomized controlled trial.
They also reported significance tests for the secondary outcomes of grade of vomiting and the need for narcotics or antiemetic drugs at the end of the study. That’s a lot of multiple comparisons without adjustment! Sadly, it’s typical in clinical trials of pain treatments for the results of analyses of secondary endpoints and of other secondary analyses to be presented without any attention to the risk of type I error that results from multiple testing.
The form of the primary outcome variable also needs to be specified. For example, change from baseline in pain intensity measured using a 0 to 10 VAS can be specified as a continuous variable or as a dichotomous variable (e.g., ≥ 30% pain reduction). The authors of this study reported the sample size was based on a Cohen’s d of 0.56 (medium effect size) of pain severity, but the primary disadvantage of using Cohen’s d is that it’s harder to interpret because it has no units. What does it mean if the mean difference is 0.56 the size of the standard deviation? In general, the mean difference should be used as the simple effect size whenever possible rather than Cohen’s d as the natural data units are much easier to interpret. Also, Cohen’s d is not a reliable effect size for non-parametric data, which was likely an issue with this trial from the data reported in Table 2.
To sign up for an email blast with the EMC Journal Club of the month delivered to your inbox go to our subscribe page
Drs. Mohindra, McLeod and Helman are supported by the Schwartz-Reisman Emergency Medicine Institute, a nonprofit organization dedicated to excellence in EM research and education.
References
- Razi A, Farrokhi E, Lotfabadi P, Hosseini SS, Saadati H, Haghighi R, Rameshrad M. Dexamethasone and ketorolac compare with ketorolac alone in acute renal colic: A randomized clinical trial. Am J Emerg Med. 2022 Aug;58:245-250.
- Newton R. Molecular mechanisms of glucocorticoid action: what is important? Thorax . 2000 Jul;55(7):603-13. doi: 10.1136/thorax.55.7.603.
- Czock D, Keller F, Rasche FM, Häussler U. Pharmacokinetics and pharmacodynamics of systemically administered glucocorticoids. Clin Pharmacokinet. 2005;44(1):61-98.
Leave A Comment