探讨肌少症对中老年急性胰腺炎患者的临床预测价值。

收集2013年8月至2021年8月南昌大学第一附属医院收治的160例45岁及以上急性胰腺炎患者的临床资料，根据CT第3腰椎层面骨骼肌指数确定其是否合并肌少症。以急性胰腺炎严重程度床旁指数（BISAP）为对照，采用Spearman秩相关分析和受试者工作特征曲线（ROC）评价肌少症评估对中老年急性胰腺炎患者疾病严重程度、住院时间、住院费用和疾病转归的预测价值。

91例（56.9%）患者诊断为肌少症。合并肌少症者BISAP评分、住院时间、住院费用均高于未合并肌少症者（H=9.800、12.560、10.170，P<0.05）。肌少症、BISAP与急性胰腺炎患者住院时间均呈正相关（r=0.310、0.190，P<0.05或0.01）。肌少症、BISAP预测中老年急性胰腺炎患者病情严重程度（中度重症和重症）的AUC分别为0.540、0.680，差异有统计学意义（Z=2.630，P<0.05）；预测住院费用（≥16 738元）的AUC分别为0.620、0.630，预测疾病转归（死亡）的AUC分别为0.702、0.960，差异均无统计学意义（Z=0.200、10.670，P>0.05）。

肌少症评估可以有效预测中老年急性胰腺炎患者的住院时间，对住院费用和疾病转归也有一定预测价值。

To explore the clinical predictive value of sarcopenia assessment for the middle-aged and elderly patients with acute pancreatitis.

The clinical data of 160 acute pancreatitis patients over 45 years old (including 45 years old) treated in the First Affiliated Hospital of Nanchang University from August 2013 to August 2021 were collected. According to L3-SMI on CT, it was determined whether it was complicated with sarcopenia. Taking the bedside index for severity in acute pancreatitis (BISAP) as the control, Spearman rank correlation analysis and receiver operating characteristic curve (ROC) were used to evaluate the predictive value of sarcopenia on disease severity, length of hospital stay, hospitalization expenses and the prognosis, and Delong test was used to evaluate the statistical significance of area under curve (AUC) values of each ROC curve.

91 patients were diagnosed as sarcopenia, accounting for 56.9%. There were significant differences in BISAP score, length of hospital stay and hospitalization expenses between the sarcopenia group and the non-sarcopenia group (H=9.800, 12.560, 10.170, P < 0.05) . Sarcopenia assessment and BISAP score had a relatively strong correlation with the length of hospital stay (r=0.310, 0.190, P < 0.05 or P < 0.01). The AUC of sarcopenia and BISAP in predicting the severity of acute pancreatitis in middle-aged and elderly patients (moderate severe and severe) were 0.540 and 0.680, and the difference was statistically significant (Z=2.630, P < 0.05). The AUC of predicting hospitalization cost (≥16 738 yuan) was 0.620 and 0.630, and the AUC of predicting disease outcome (death) was 0.702 and 0.960, and the difference was not statistically significant (Z=0.200, 10.670, P > 0.05).

Sarcopenia assessment can effectively predict the length of hospital stay of middle-aged and elderly patients with acute pancreatitis, and it also has a certain predictive value for hospitalization costs and disease outcomes.