Department of Surgery, National Naval Medical Center, Bethesda, MD, USA.
Surgery. 2010 Jun;147(6):861-70. doi: 10.1016/j.surg.2009.11.017. Epub 2010 Jan 21.
The Dysphonia Severity Index (DSI) is an objective multiparametric acoustic calculation of vocal function; however, its changes after thyroidectomy have not yet been described.
Patient-reported symptoms, as well as auditory perceptual, acoustic, and videolaryngostroboscopic (VLS) data, were collected prospectively before and after thyroidectomy. Voice outcomes (normal versus negative voice outcome [NVO]) at 6 months after thyroidectomy were based on a combination of voice symptoms and objective findings. The DSI was assessed over the peri-operative course, and differences were determined with Wilcoxon signed rank tests. The DSI was compared between study groups (normal versus NVO) using t tests, analyses of variance (ANOVAs), or rank sum tests as appropriate. The predictive value of DSI for long-term voice dysfunction was assessed by an area under the receiver operating characteristics curve analysis. Correlations between DSI and Consensus Auditory Perceptual Ratings of Voice (CAPE-V) and the patient reported Voice Handicap Index (VHI) were determined with Pearson's correlation coefficients.
In all, 62 patients were evaluated before, 1-4 weeks after, and 6 months after thyroidectomy. Eight (13%) patients were diagnosed with NVO at 6 months. The DSI was different postoperatively between NVO and normal voice (P=.005, repeated measures [RM]-ANOVA), with the NVO group demonstrating a lesser DSI value and greater change from pre-operative assessment at the first postoperative visit when compared with the normal group (P<.006 each). The DSI differed significantly for pre-operative and 6-month assessments according to sex, smoking status, and age. Short-term postoperative DSI (area under the curve [AUC]=0.795) and DSI change from baseline to 1-4 weeks (AUC=0.835) were highly predictive of 6-month NVO. DSI measurements over the post-thyroidectomy course were correlated poorly to moderately (maximum r = -0.62) with CAPE-V and VHI assessments for the same time points.
The DSI is decreased in the early post-thyroidectomy period, mostly in persons who were ultimately found to have a long-term NVO. Early postoperative DSI and change of DSI from baseline at 1--4 weeks postoperation predict long-term post-thyroidectomy voice dysfunction. The modest correlations between the DSI and other vocal assessments point to the utility of DSI as an independent predictor of voice dysfunction after thyroidectomy, which can select patients who may benefit from voice therapy.
嗓音障碍严重指数（DSI）是一种客观的多参数声学嗓音功能评估方法；然而，其在甲状腺切除术后的变化尚未被描述。
前瞻性地收集甲状腺切除术前和术后患者报告的症状，以及听觉感知、声学和视频喉镜频闪（VLS）数据。甲状腺切除术后 6 个月的嗓音结果（正常与负嗓音结果[NVO]）基于嗓音症状和客观发现的综合评估。围手术期评估 DSI，使用 Wilcoxon 符号秩检验确定差异。使用 t 检验、方差分析（ANOVA）或秩和检验比较研究组（正常与 NVO）之间的 DSI，具体取决于研究设计。通过受试者工作特征曲线分析评估 DSI 对长期嗓音功能障碍的预测价值。通过 Pearson 相关系数确定 DSI 与共识性听觉感知嗓音评估（CAPE-V）和患者报告嗓音障碍指数（VHI）之间的相关性。
共有 62 例患者在甲状腺切除术前、术后 1-4 周和术后 6 个月接受评估。8 例（13%）患者在术后 6 个月被诊断为 NVO。NVO 组和正常嗓音组之间术后 DSI 存在差异（P=.005，重复测量[RM]-ANOVA），与正常组相比，NVO 组在术后首次就诊时的 DSI 值较低，且与术前评估相比变化更大（P<.006 各）。根据性别、吸烟状况和年龄，术前和 6 个月评估的 DSI 差异显著。短期术后 DSI（曲线下面积[AUC]=0.795）和从基线到 1-4 周的 DSI 变化（AUC=0.835）对 6 个月 NVO 具有高度预测价值。甲状腺切除术后 DSI 测量与同一时间点的 CAPE-V 和 VHI 评估相关性较差至中度（最大 r = -0.62）。
甲状腺切除术后早期 DSI 降低，主要发生在最终被诊断为长期 NVO 的患者中。术后早期 DSI 和术后 1-4 周时 DSI 从基线的变化预测甲状腺切除术后长期嗓音功能障碍。DSI 与其他嗓音评估之间的中度相关性表明，DSI 作为甲状腺切除术后嗓音障碍的独立预测因子具有一定的应用价值，它可以选择可能受益于嗓音治疗的患者。
