Anatomic versus reverse shoulder arthroplasty, are nerve injury rates different?

Abstract

BackgroundThe true incidence of iatrogenic neurological injury following total shoulder arthroplasty (TSA) surgery remains controversial at a cited rate of 1%-18.7%. Additionally, there is a reported incidence of new onset distal peripheral neuropathy (DPN) of 7.1% and 12.3%, for anatomic (aTSA) and reverse (rTSA) shoulder replacements, respectively. The purpose of this present study is to report on nerve injury rates of a large, consecutive series of primary TSA (ie, rTSA and aTSA) performed by one surgeon, utilizing intraoperative neuromonitoring and consistent implant design. MethodsA retrospective cohort review of consecutive patients who underwent primary TSA was performed from January 2016 to March 2020. Intraoperative neuromonitoring data included transcranial electrical motor evoked potentials (MEPs), somatosensory evoked potentials, and free-run electromyography. Motor alert threshold was set at ≥ 80% signal attenuation in any one or more peripheral nerves. Patients were screened for neurological deficits immediately following surgery and in the first 2 postoperative visits. Additional data collection included surgical indication, gender, laterality, age at surgery, procedure performed, body mass index, history of tobacco use, Charlson comorbidity index, past medical history and preoperative range of motion. ResultsA total of 290 consecutive shoulders in 270 patients were included. Ninety-one (31.4%) shoulders underwent aTSA, while 199 (68.6%) underwent rTSA. Patients in the rTSA cohort were significantly older (70.1 vs. 61.3 years; P < .001), had higher Charlson comorbidity index scores (4.3 vs. 3.0; P < .001), and shorter operative time (81.8 vs. 99.7; P < .001). A total of 96 MEP alerts occurred in 65 (22.4%) operations, with a significantly higher proportion of these occurring in patients undergoing aTSA. MEP alerts occurred in 28 (39.8%) aTSA versus 37 (18.5%) rTSA (P = .024). Logistical regression found operative time (P = .003, odds ratio = 1.020 [95%: 1.007-1.033]) was predictive of a MEP alert. The axillary nerve was the most commonly alerted nerve accounting for 61.5% of all alerts. Zero patients had a major or minor nerve injury in this large cohort. Nine patients (3.1%) had DPN in the postoperative period. ConclusionIn 290 surgeries, no (0%) patients had a postoperative major or minor nerve injury, and 9 (3.1%) patients complained of postoperative distal peripheral neuropathy. Patients that had MEP alerts had significantly longer operative times and were more likely to be in the aTSA group compared to the rTSA group. This may be attributed to longer exposure and implantation times on the glenoid side for an anatomic as opposed to a reverse arthroplasty. Level of evidenceLevel III; Retrospective Cohort study.