This was a cross-sectional study evaluating assessment of anal sphincter defects after obstetric anal sphincter injury (OASI) with endoanal, introital and transperineal ultrasound, assessing the predictive value for diagnosing anal incontinence (AI) [1]. This is the largest study to date performing this type of comparison. The authors concluded that defects diagnosed by endoanal ultrasound (EAUS) had a stronger association with AI than either introital ultrasound (IUS) or transperineal ultrasound (TPUS). Additionally, they have demonstrated an association between anal sphincter defects, as demonstrated in all sonographic modalities, and lower anal pressures on anorectal manometry. Although the rationale for the study is highly appropriate, we beg to differ with the authors’ conclusions, in particular regarding the alleged inferiority of TPUS compared to EAUS. First, as mentioned by the authors themselves in the Discussion, TPUS was performed with a GE Voluson i ultrasound system, which is obsolete and significantly inferior to modern machines commonly used in recent studies of OASI imaging [2–4]. The poor image quality would be expected to lower sensitivity and specificity. The first author (I.V.) has in fact published a recent paper using a more modern ultrasound system [3], and the difference in the image quality of the figures between those studies is striking. Second, the criteria for diagnosing significant external and internal anal sphincter defects (EAS and IAS, respectively) on TPUS are flawed. The authors use cutoffs that are derived from previous comparisons to EAUS in which the latter was considered the gold standard. This assumption is quite inappropriate since suggested cutoffs should be validated against symptoms and tested in a normal control group rather than in women after OASI repair, as previously done for TPUS [2, 5]. Third, the authors tested for an association of AI with EAS and IAS defects combined. This may well be misleading since IAS defects on TPUS are sometimes seen in nulliparous women and are less predictive of AI [2]. Finally, IAS assessment was not conducted according to current validated methodology for IAS imaging, which differs from EAS with assessment criteria, using a more cranial slice placement and a wider inter-slice interval on tomographic imaging. In conclusion, it appears to us that the authors have performed an inappropriate comparison with suboptimal conditions, which prevents drawing any firm conclusions regarding superiority of any ultrasound modality for predicting anal incontinence after obstetric anal sphincter injury.

中文翻译：

---

评论：括约肌缺损的肛门内、阴道口或经会阴超声诊断是否与肛门失禁更密切相关？

这是一项横断面研究，通过肛门内、阴道口和经会阴超声评估产科肛门括约肌损伤 (OASI) 后肛门括约肌缺陷的评估，评估诊断肛门失禁 (AI) 的预测价值 [1]。这是迄今为止进行此类比较的最大研究。作者得出的结论是，通过肛内超声 (EAUS) 诊断出的缺陷与 AI 的相关性比插入超声 (IUS) 或经会阴超声 (TPUS) 更强。此外，他们还证明了肛门括约肌缺陷（如所有超声检查方式所示）与肛门直肠测压法的较低肛门压力之间存在关联。尽管这项研究的理由非常恰当，但我们不同意作者的结论，特别是关于 TPUS 与 EAUS 相比的所谓劣势。首先，正如作者自己在讨论中提到的，TPUS 是使用 GE Voluson i 超声系统进行的，该系统已经过时并且明显不如最近 OASI 成像研究中常用的现代机器 [2-4]。预期较差的图像质量会降低灵敏度和特异性。事实上，第一作者 (IV) 最近发表了一篇使用更现代的超声系统 [3] 的论文，这些研究之间图像质量的差异是惊人的。其次，在 TPUS 上诊断明显的外部和内部肛门括约肌缺陷（分别为 EAS 和 IAS）的标准是有缺陷的。作者使用从之前与 EAUS 的比较得出的临界值，后者被认为是金标准。这种假设是非常不恰当的，因为建议的临界值应该针对症状进行验证并在正常对照组中进行测试，而不是在 OASI 修复后的女性中进行测试，如之前对 TPUS 所做的那样 [2, 5]。第三，作者测试了 AI 与 EAS 和 IAS 缺陷的关联。这很可能具有误导性，因为 TPUS 上的 IAS 缺陷有时见于未产妇，并且对 AI 的预测性较低 [2]。最后，IAS 评估不是根据当前经过验证的 IAS 成像方法进行的，这与 EAS 的评估标准不同，使用更多的颅骨切片放置和更宽的断层扫描成像间隔。总之，在我们看来，作者与次优条件进行了不适当的比较，