Shah Anuja H, McCray Lauren R, Nguyen Shaun A, Camposeo Elizabeth, Dixon Peter R
Department of Otolaryngology-Head and Neck Surgery, Medical University of South Carolina, Charleston, South Carolina, USA.
Georgetown University School of Medicine, Washington, DC, USA.
Laryngoscope. 2025 Jun 20. doi: 10.1002/lary.32355.
Remote cochlear implant (CI) programming can address geographic disparities in CI care and improve patient convenience. While in-person programming is necessary in certain circumstances, it is important to determine the potential advantages in speech recognition outcomes. This study reviews the literature comparing speech recognition outcomes between remote versus in-person CI programming.
PubMed, Scopus, CINAHL, and Cochrane databases were queried from inception to May 7, 2025.
Eligible studies included CI recipients ≥ 12 years of age who underwent remote programming, with a comparison group who underwent in-person programming, and reported speech recognition outcomes. Primary outcome was standardized mean differences (SMD) of speech recognition outcomes, stratified by sentence- or word-recognition measures, pooled with random effects models.
Of 1597 abstracts identified, seven studies were included with a total of 123 CI users (mean age 53.8 years [range: 12-88, 95% CI: 42.4-65.3]; 50.6% male [95% CI: 26.9-74.2]). Study designs included crossover within-subjects designs (n = 4), prospective cohorts (n = 2), and cross-sectional studies (n = 1). Speech recognition outcomes tested in quiet were similar between in-person programmed and remotely programmed CI users for word lists (SMD: -0.06 [95% CI: -0.37-0.35], p = 0.71) and for sentences (SMD: -0.28 [95% CI: -0.79-0.24], p = 0.29). No studies were designed as non-inferiority analyses.
Remote CI programming is not associated with detectable differences in speech recognition outcomes compared to in-person programming. Future studies comparing remote and in-person programming outcomes should evaluate non-inferiority of remote programming to continue to justify its expansion to improve convenience and mitigate geographic care disparities.
远程人工耳蜗（CI）编程可解决CI护理中的地理差异问题，并提高患者的便利性。虽然在某些情况下需要亲自进行编程，但确定其在语音识别结果方面的潜在优势很重要。本研究回顾了比较远程与亲自进行CI编程的语音识别结果的文献。
从数据库创建至2025年5月7日查询了PubMed、Scopus、CINAHL和Cochrane数据库。
符合条件的研究包括年龄≥12岁且接受远程编程的CI接受者，以及作为对照组的接受亲自编程的CI接受者，并报告了语音识别结果。主要结果是语音识别结果的标准化平均差异（SMD），按句子或单词识别测量分层，采用随机效应模型进行汇总。
在1597篇摘要中，纳入了7项研究，共123名CI使用者（平均年龄53.8岁[范围：12 - 88岁，95%置信区间：42.4 - 65.3]；50.6%为男性[95%置信区间：26.9 - 74.2]）。研究设计包括交叉受试者内设计（n = 4）、前瞻性队列研究（n = 2）和横断面研究（n = 1）。在安静环境中测试的语音识别结果，对于单词列表，亲自编程的CI使用者和远程编程的CI使用者相似（SMD：-0.06[95%置信区间：-0.37 - 0.35]，p = 0.71）；对于句子，两者也相似（SMD：-0.28[95%置信区间：-0.79 - 0.24]，p = 0.29）。没有研究设计为非劣效性分析。
与亲自编程相比，远程CI编程在语音识别结果方面未发现可检测到的差异。未来比较远程和亲自编程结果的研究应评估远程编程的非劣效性，以继续证明其扩展的合理性，从而提高便利性并缓解地理护理差异。
