The effect of concurrent cervical spine degenerative disease on the outcome of rotator cuff repair: a national database study

Abstract

Cervical spine degenerative disease (CSD) can cause shoulder pain, potentially confounding the management of patients with rotator cuff tears. This study aimed to investigate the relationships between CSD and rotator cuff repair (RCR). A national administrative database (PearlDiver) was used to study 4 patient cohorts: (1) RCR only (RCRo), (2) RCR with concurrent CSD (RCRC), (3) RCR after a cervical spine procedure (RCRA), and (4) RCR before a cervical spine procedure (RCRB). The outcomes of RCR were compared using multivariable logistic regression, controlling for age, sex, and Elixhauser Comorbidity Index, as well as preoperative opioid utilization in the analysis of opioid use. Between 2010 and 2021, a total of 889,977 patients underwent RCR. Of these patients, 784,230 (88%) underwent RCRo whereas 105,747 (12%) underwent RCRC, of whom 21,585 (2.4%) underwent cervical spine procedures (RCRA in 9670 [1.1%] and RCRB in 11,915 [1.3%]). At 2 years after RCR, compared with RCRo patients, RCRC patients had an increased risk of surgical-site infection (adjusted odds ratio [aOR]=1.25, P=.0004), deep vein thrombosis (aOR=1.17, P=.0002), respiratory complications (aOR=1.19, P=.0164), and ipsilateral shoulder reoperations (débridement [aOR=1.66, P<.0001], manipulation under anesthesia or arthroscopic lysis of adhesions [aOR=1.23, P<.0001], distal clavicle excision [aOR=1.78, P<.0001], subacromial decompression [aOR=1.72, P<.0001], biceps tenodesis [aOR=1.76, P<.0001], incision and drainage [aOR=1.34, P=.0020], synovectomy [aOR=1.48, P=.0136], conversion to shoulder arthroplasty [aOR=1.62, P<.0001], revision RCR [aOR=1.77, P<.0001], and subsequent contralateral RCR [aOR=1.71, P<.0001]). At 2 years, compared with RCRC patients who did not undergo cervical spine procedures, RCRC patients who underwent cervical spine procedures had an increased risk of incision and drainage (aOR=1.50, P=.0255), conversion to arthroplasty (aOR=1.40, P<.0001), and revision RCR (aOR=1.11, P=.0374), as well as a lower risk of contralateral RCR (aOR=0.89, P=.0469). The sequence of cervical spine procedures did not affect the risk of shoulder reoperations. At 1 year, the risk of opioid use after RCR was less for RCRA patients compared with RCRB patients (aOR=1.71 [95% confidence interval, 1.61-1.80; P<.0001] vs. aOR=2.01 [95% confidence interval, 1.92-2.12; P<.0001]). Concurrent CSD has significant detrimental effects on RCR outcomes. Patients with concurrent CSD undergoing cervical spine procedures have a greater risk of ipsilateral shoulder reoperations but a decreased risk of contralateral RCR. The risk of prolonged opioid use was lower if RCR followed a cervical spine procedure. Concurrent CSD must be considered and possibly treated to optimize the outcomes of RCR.