Are any radiologic parameters independently associated with degenerative postero-superior rotator cuff tears?

Abstract

Several radiologic parameters representing scapular morphology have been proposed as risk factors for degenerative posterosuperior rotator cuff tear (PSRCT). The causes of PSRCT are multifactorial; many reported risk factors are compound entities. However, no determination, made after adjustment with previously reported risk factors, of which radiologic parameters are independent risk factors for degenerative PSRCT, has been reported. Therefore, this study's purpose was to determine which radiologic parameters representing scapular morphology are independent risk factors for degenerative PSRCT in a general population after adjustment with previously suggested risk factors. This study involved 629 subjects (629 shoulders) drawn from a cohort of rural residents. PSRCT diagnoses were based on magnetic resonance imaging findings. Logistic regression analysis was used to determine the odds ratios and 95% confidence intervals for the following variables: sex, age, height, weight, waist circumference, dominant-side involvement, smoking, alcohol consumption, manual labor, diabetes, hypertension, metabolic syndrome, thyroid dysfunction, serum lipid profile, high-sensitivity C-reactive protein, and radiologic parameters. The evaluated radiologic parameters were the superior displacement of the humeral head (SDHH) and those representing the scapular morphology. The radiologic parameters representing the scapular morphology were lateral acromial angle, acromial slope, acromial tilt, acromial index, and critical shoulder angle (CSA). Univariate logistic regression analyses for all variables, as well as multivariable logistic regression analyses for variables with significant associations and clinical effect sizes in the univariate logistic regression analyses, were performed. In order to determine the cutoff value for significant radiologic parameters, each pair of candidate cutoff values was obtained as follows: 1 candidate, as the lowest significant ordinal value in a univariate analysis, and the other candidate, according to a receiver operating characteristic curve analysis. Then, the cutoff values were determined by selecting the value in each candidate pair that had the larger area under the receiver operating characteristic curve, using the DeLong test. In the multivariable analyses, CSA was significantly associated with degenerative PSRCT after controlling for the following factors: age, waist circumference, dominant-side involvement, manual labor, diabetes, metabolic syndrome, hypo-high-density lipoproteinemia, and SDHH (P ≤ .033). CSA (odds ratio, 1.58 [95% confidence interval, 1.45-1.72]; P < .001) was the only significantly associated radiologic parameter among the studied radiologic parameters representing scapular morphology. CSA (≥35°) is an independent associated factor for degenerative PSRCT after adjustment for the following suggested risk factors: age, waist circumference, dominant-side involvement, manual labor, diabetes, metabolic syndrome, hypo-high-density lipoproteinemia, and SDHH.