Concordance of shoulder strength assessments using a spring balance and isometric dynamometer in patients before and after arthroscopic rotator cuff repair

BACKGROUND: Shoulder strength is an essential assessment to monitor the outcome of treatment interventions. Isometric strength assessment in the Constant Score (CS) was initially measured with a cable tensiometer or spring balance (SB). Some authors have questioned the validity of this strength assessment and the resulting CS. The purpose of this study was to investigate the concordance of strength measurements using an unsecured SB vs. isometric dynamometer and outline the impact of these methods on the CS. METHODS: In the context of routine clinical examination as well as participation in a Swiss national cohort study, shoulder strength was measured to calculate baseline (before surgery) and 6-month postoperative CS in adult rotator cuff tear patients who had undergone primary arthroscopic rotator cuff repair. Measurements of each of the operated and contralateral shoulders were made per patient routinely using an unsecured SB and study-specific using an isometric dynamometer in patients with the shoulder at 90° abduction in the scapular plane. Absolute and change values of strength and CS data were presented in scatter plots and assessed using concordance correlation coefficients (CCCs) and Bland-Altman plots. RESULTS: Between June 2020 and October 2021, baseline strength measurements from the operated shoulder of 78 patients ranged from 0.0 to 13.6 kg with a CCC of 0.64 (P < .001) and a mean difference of 0.81 kg between the SB and dynamometer methods. There were 89 measurements of the contralateral healthy shoulder that ranged from 3.6 to 15.6 kg; CCC and mean strength difference were 0.76 (P < .001) and 0.70 kg, respectively. At 6 months postsurgery, strength measurements of the operated shoulder ranged from 1.4 to 12.0 kg with a CCC of 0.66 (P < .001) and mean strength difference of 0.9 kg (n = 68). Respective 6-month measurements of the contralateral side (n = 52) ranged from 2.0 to 15.9 kg with a CCC of 0.73 (P < .001) and mean strength difference of 0.03 kg. CONCLUSION: Absolute and change values in shoulder strength assessments using an unsecured SB and isometric dynamometer are fairly concordant with mean differences of less than 1 kg between methods. With the variability of strength differences among patients, interpretation of these values for individual patients may be challenging. Nonetheless, unsecured SB and dynamometer methods share only slight and clinically unimportant differences that can provide similar group mean values for use in research along with the calculation of the CS.