TO THE EDITOR: We read with interest the study by Greve et al.1 in which they compared neurological outcomes of patients with intraoperative neuromonitoring (IONM: somatosensory evoked potentials [SSEPs] and motor evoked potentials [MEPs]) to a historical control without IONM (Greve T, Stoecklein VM, Dorn F, et al. Introduction of intraoperative neuromonitoring does not necessarily improve overall long-term outcome in elective aneurysm clipping. J Neurosurg. 2020;132[4]:1188–1196). No difference was observed for clipping of unruptured intracranial aneurysms. The authors state, “Hence, the ethical burden to perform randomized controlled trials with and without the use of [IONM] in [elective microsurgical clipping of unruptured intracranial aneurysms] might by overcome.”
Clinical equipoise, “genuine uncertainty within the expert medical community,” should be informed by our certainty in the results of studies.2 In their nonrandomized comparative study, the authors did not account for differences in prognostic variables between patients that could also have predicted the use of IONM. This typically requires multivariate methods (regression, propensity analysis) or stratification of effects.3 In fact, patients in the IONM group were older than those in the no-IONM group, and age was correlated with worse postoperative outcomes. Confounding may have resulted in no observed difference when there was a true association between IONM and outcomes. Accounting for confounding in nonrandomized studies is a basic requirement before results can be accepted with confidence.4
As the authors acknowledged, a limitation of their study is the small sample size. Assuming a 10% risk of new neurological deficit and that IONM would cut this in half, 435 procedures in the IONM and no-IONM groups (870 total) would be required for a conventional 80% power (Pearson’s chi-square test). The authors reported results on 138 procedures with IONM and 136 procedures without IONM. Their underpowered study (power = 27%) may have missed an IONM effect on outcomes if one was present.
In the absence of randomized studies, 2 other comparative studies of IONM for middle cerebral artery (MCA) aneurysms might be given more weight in informing medical opinion. Both Byoun et al.5 (SSEPs for unruptured MCA aneurysms) and Yue et al.6 (MEPs for MCA aneurysms, 72% ruptured) used multivariate regression to account for confounding. Contrary to the results reported by Greve et al.,1 both studies reported improved postoperative outcomes for IONM compared with historical cohorts without IONM.
Disclosures
The authors report no conflict of interest.
References
- 1↑
Greve T, Stoecklein VM, Dorn F, et al. Introduction of intraoperative neuromonitoring does not necessarily improve overall long-term outcome in elective aneurysm clipping. J Neurosurg. 2020;132(4):1188–1196.
- 3↑
Agoritsas T, Merglen A, Shah ND, et al. Adjusted analyses in studies addressing therapy and harm: users’ guides to the medical literature. JAMA. 2017;317(7):748–759.
- 4↑
Schünemann HJ, Cuello C, Akl EA, et al. GRADE guidelines: 18. How ROBINS-I and other tools to assess risk of bias in nonrandomized studies should be used to rate the certainty of a body of evidence. J Clin Epidemiol. 2019;111:105–114.
- 5↑
Byoun HS, Bang JS, Oh CW, et al. The incidence of and risk factors for ischemic complications after microsurgical clipping of unruptured middle cerebral artery aneurysms and the efficacy of intraoperative monitoring of somatosensory evoked potentials: a retrospective study. Clin Neurol Neurosurg. 2016;151:128–135.
- 6↑
Yue Q, Zhu W, Gu Y, et al. Motor evoked potential monitoring during surgery of middle cerebral artery aneurysms: a cohort study. World Neurosurg. 2014;82(6):1091–1099.
