胶原基止血贴片使用与双侧甲状腺切除术后患者报告恢复之间的关联：一项回顾性队列研究

文献信息

中文摘要

背景/目的： 局部止血剂在甲状腺手术中被广泛使用，但其临床价值仍不确定。我们评估了双侧甲状腺切除术中术中使用胶原基止血贴片是否与患者报告的恢复及术后并发症相关。
方法： 这项回顾性队列研究纳入193例接受双侧甲状腺切除术的患者（98例接受Hemopatch，95例接受不使用贴片的常规止血）。疼痛和吞咽评分以中位数及四分位距汇总，并采用双侧Mann-Whitney U检验进行比较。报告了未调整的平均差及其基于10,000次独立组重抽样的95%百分位自助法置信区间作为效应估计。作为敏感性分析，进行了多变量比例优势序数逻辑回归，并调整年龄、性别、手术时间、术中失血量和病理。
结果： 主要术后并发症不常见，组间差异无统计学意义。贴片使用与术后第7天（平均差 -0.34，95% CI -0.59至-0.09；p = 0.0008）、术后第30天（-0.11，95% CI -0.22至0.01；p = 0.0223）和术后第90天（-0.12，95% CI -0.18至-0.05；p = 0.0005）较低的疼痛评分相关，但与术后第1天无关。吞咽障碍也与术后第7天（-0.35，95% CI -0.67至-0.03；p = 0.0068）、术后第30天（-0.17，95% CI -0.38至0.05；p = 0.0272）和术后第90天（-0.23，95% CI -0.38至-0.09；p = 0.0005）较低评分相关，但与术后第1天无关。瘢痕满意度与6个月时较高的平均评分相关（+0.46，95% CI +0.21至+0.71；p = 0.0018）。在多变量调整和Holm校正后，若干关联仍具有统计学意义，但并非所有比较对多重性调整都稳健。
结论： 胶原基止血贴片使用与部分患者报告恢复指标的适度差异相关。这些差异的幅度和临床重要性仍不确定。未观察到主要并发症方面有统计学显著差异，但该研究不足以确立对不常见事件的比较安全性。这些探索性发现需要在前瞻性多中心研究中得到验证。

英文摘要

Background/Objectives: Topical hemostatic agents are widely used in thyroid surgery, but their clinical value remains uncertain. We evaluated whether intraoperative use of a collagen-based hemostatic patch during bilateral thyroidectomy was associated with patient-reported recovery and postoperative complications. Methods: This retrospective cohort study included 193 patients undergoing bilateral thyroidectomy (98 received Hemopatch and 95 underwent conventional hemostasis without a patch). Pain and swallowing scores were summarized as medians with interquartile ranges and compared using two-sided Mann-Whitney U tests. Unadjusted mean differences with 95% percentile-bootstrap confidence intervals based on 10,000 independent-group resamples were reported as effect estimates. Multivariable proportional-odds ordinal logistic regression adjusting for age, sex, operative time, intraoperative blood loss, and pathology was performed as a sensitivity analysis. Results: Major postoperative complications were infrequent, with no statistically significant between-group differences. Patch use was associated with lower pain scores on POD7 (mean difference -0.34, 95% CI -0.59 to -0.09; p = 0.0008), POD30 (-0.11, 95% CI -0.22 to 0.01; p = 0.0223), and POD90 (-0.12, 95% CI -0.18 to -0.05; p = 0.0005), but not POD1. Swallowing disturbance was also associated with lower scores on POD7 (-0.35, 95% CI -0.67 to -0.03; p = 0.0068), POD30 (-0.17, 95% CI -0.38 to 0.05; p = 0.0272), and POD90 (-0.23, 95% CI -0.38 to -0.09; p = 0.0005), but not POD1. Scar satisfaction was associated with a higher mean score at 6 months (+0.46, 95% CI +0.21 to +0.71; p = 0.0018). Several associations remained significant after multivariable adjustment and Holm correction, although not all comparisons were robust to multiplicity adjustment. Conclusions: Collagen-based hemostatic patch use was associated with modest differences in selected patient-reported recovery measures. The magnitude and clinical importance of these differences remain uncertain. No statistically significant differences in major complications were observed, but the study was not powered to establish comparative safety for uncommon events. These exploratory findings require confirmation in prospective multicenter studies.