Travaglini Silvia, Bonvicini Maria, Bandinelli Stefania, Ferrucci Luigi, Antonelli Incalzi Raffaele, Pedone Claudio
Residency Program in Internal Medicine, Department of Medicine and Surgery, Università Campus Bio-Medico, Rome, Italy.
Residency Program in Geriatrics, Department of Medicine and Surgery, Università Campus Bio-Medico, Rome, Italy.
J Gerontol A Biol Sci Med Sci. 2025 Feb 10;80(3). doi: 10.1093/gerona/glaf003.
Poor muscle strength is a risk factor for disability; nonetheless, its discriminative capacity in identifying people who will become disabled is poor. We evaluated whether muscle power, which also is a risk factor for disability, has better discriminative capacity compared to muscle strength.
We used data from the population-based InCHIANTI study. Our outcome measure was the loss of at least 1 basic or instrumental activity of daily living between baseline and 3-year follow-up visit. Body weight standardized knee isometric strength and leg power (power rig) were used as exposure variables. Discriminative capacity was estimated using the area under the receiver-operating curves. Both strength and power were dichotomized as below versus equal of above sex-specific first quartile. Sensitivity, specificity, and positive/negative predictive values (PPV, NPV) were calculated.
We included 763 participants (415 women), with a mean age of 73.5 years (standard deviation: 6.4). In men, using muscle strength we obtained an area under the curve (AUC) of 0.70, with sensitivity = 0.45, specificity = 0.80, PPV = 0.27, and NPV = 0.90; using muscle power we obtained an AUC of 0.82, sensitivity = 0.73, specificity = 0.86, PPV = 0.46, and NPV = 0.95. In women, using muscle strength we obtained an AUC of 0.62, with sensitivity = 0.39, specificity = 0.81, PPV = 0.39, and NPV = 0.81; using muscle power we obtained an AUC = 0.69, sensitivity = 0.40, specificity = 0.83, PPV = 0.42, and NPV = 0.82.
We found that in men muscle power had better discriminative capacity, especially higher sensitivity, compared to muscle strength for prediction of worsening disability. No meaningful difference in overall discriminative capacity was found in women.
肌肉力量差是导致残疾的一个风险因素；然而，其在识别即将致残人群方面的区分能力较差。我们评估了同样作为残疾风险因素的肌肉功率相比于肌肉力量是否具有更好的区分能力。
我们使用了基于人群的基安蒂研究（InCHIANTI study）的数据。我们的结局指标是在基线和3年随访期间至少丧失一项基本或工具性日常生活活动能力。体重标准化的膝关节等长肌力和腿部功率（功率测试装置）被用作暴露变量。使用受试者工作特征曲线下面积来估计区分能力。力量和功率均被分为低于或等于与性别相关的第一四分位数。计算敏感性、特异性以及阳性/阴性预测值（PPV，NPV）。
我们纳入了763名参与者（415名女性），平均年龄为73.5岁（标准差：6.4）。在男性中，使用肌肉力量时我们得到的曲线下面积（AUC）为0.70，敏感性 = 0.45，特异性 = 0.80，PPV = 0.27，NPV = 0.90；使用肌肉功率时我们得到的AUC为0.82，敏感性 = 0.73，特异性 = 0.86，PPV = 0.46，NPV = 0.95。在女性中，使用肌肉力量时我们得到的AUC为0.62，敏感性 = 0.39，特异性 = 0.81，PPV = 0.39，NPV = 0.81；使用肌肉功率时我们得到的AUC = 0.69，敏感性 = 0.40，特异性 = 0.83，PPV = 0.42，NPV = 0.82。
我们发现，在男性中，相比于肌肉力量，肌肉功率在预测残疾恶化方面具有更好的区分能力，尤其是更高的敏感性。在女性中未发现总体区分能力有显著差异。
