研究对比增强经颅多普勒（c-TCD）与经食管超声心动图声学造影（c-TEE）评估卵圆孔未闭（PFO）相关右向左分流（RLS）的临床意义。

回顾性连续纳入2020年1月至2022年6月在苏州大学附属第一医院因头晕、晕厥、偏头痛、缺血性卒中或短暂性脑缺血发作（TIA）等临床表现就诊，临床可疑RLS的患者243例，所有患者均完成c-TCD、TEE和c-TEE检查。分析c-TCD、c-TEE对PFO-RLS的检出结果。采用McNemar-Bowker检验比较c-TCD和c-TEE对PFO-RLS的半定量分级；并分析c-TCD评估下PFO-RLS分流类型、分流等级及与PFO直径和隧道长度的关系。

纳入的243例患者中，偏头痛及头晕、晕厥等占71.2%（173/243），缺血性卒中或TIA占28.8%（70/243）。以TEE结合c-TEE结果为参考标准，确诊232例PFO-RLS。c-TCD、c-TEE对PFO-RLS的检出率均为99.1%（230/232）。c-TCD检查：静息状态、传统Valsalva动作、定量Valsalva动作时，大量分流的检出率分别为20.7%（48/232）、41.4%（96/232）、67.7%（157/232），两两比较差异均有统计学意义（P均<0.001）。c-TEE检查：静息状态、腹部加压下，小量分流的检出率分别为18.1%（42/232）、32.8%（76/232），中量分流的检出率分别为0.9%（2/232）、25.0%（58/232），大量分流的检出率分别为3.0%（7/232）、41.4%（96/232）。c-TCD与c-TEE评估的PFO-RLS分流等级差异有统计学意义（Boweker检验值为51.524，P<0.001）。小量分流，c-TEE检出率高于c-TCD（χ²=5.222，P=0.022）；中量分流，c-TEE检出率高于c-TCD（χ²=20.885，P<0.001）；大量分流，c-TCD检出率高于c-TEE（χ²=53.923，P<0.001）。34.9%（81/232）的患者分流等级c-TCD高于c-TEE；55.6%（129/232）的患者c-TCD和c-TEE评估的分流等级相同；9.5%（22/232）的患者分流等级c-TCD低于c-TEE。c-TCD评估下固有型RLS的PFO直径大于潜在型RLS的PFO直径（P<0.05），隧道长度差异无统计学意义（P>0.05）。PFO-RLS分流等级与PFO直径呈正相关（r=0.307，P<0.05），与隧道长度无相关性（P>0.05）。

c-TCD是评估RLS的有效血流动力学依据，充分的Valsalva动作下c-TDC可为PFO-RLS提供可靠的半定量诊断结果，c-TEE可明确RLS的来源，两者共同联合可为个体化诊断提供科学依据。

To investigate the clinical significance of contrast-enhanced transcranial Doppler (c-TCD) and contrast-transesophageal echocardiography (c-TEE) in the assessment of right-to-left shunt (RLS) associated with patent foramen ovale (PFO).

A total of 243 patients who presented with symptoms such as dizziness, syncope, migraine, palpitations, transient ischemic attack (TIA), or ischemic stroke and were clinically suspected of having a RLS were retrospectively enrolled from January 2020 to June 2022 at the Department of Neurology or Cardiology of the first Affiliated Hospital of Soochow University. Each patient was subject to c-TCD, transesophageal echocardiography (TEE), and c-TEE. The detection results of PFO-RLS by c-TCD and c-TEE were analyzed. The McNemar-Bowker test was used to compare the semi-quantitative grading of c-TCD and c-TEE for PFO-RLS, and the relationship of PFO-RLS shunt types and grades with PFO diameter and tunnel length under c-TCD evaluation was analyzed.

Among the 243 enrolled patients, migraines, dizziness, and syncope accounted for 71.2% (173/243), while ischemic stroke or TIA accounted for 28.8% (70/243). Using TEE combined with c-TEE results as the reference standard, 232 cases of PFO-RLS were diagnosed. The detection rates of c-TCD and c-TEE for PFO-RLS were both 99.1% (230/232). For c-TCD examination, the detection rates for large shunts were 20.7% (48/232) at rest, 41.4% (96/232) during traditional Valsalva maneuver, and 67.7% (157/232) during quantitative Valsalva maneuver. There were statistically significant differences between each pair of detection rates (P<0.001). However, there were no statistically significant differences in the detection rates for small and moderate shunts (P>0.05). For c-TEE examination, the detection rates for small shunts were 18.1% (42/232) at rest and 32.8% (76/232) under abdominal pressure. The detection rates for moderate shunts were 0.9% (2/232) at rest and 25.0% (58/232) under abdominal pressure. The detection rates for large shunts were 3.0% (7/232) at rest and 41.4% (96/232) under abdominal pressure. There was a statistically significant difference in the shunt grades of PFO-RLS evaluated by c-TCD and c-TEE (Boweker test value=51.524, P<0.001). For small shunts, the detection rate of c-TEE was higher than that of c-TCD (χ²=5.222, P=0.022); for moderate shunts, the detection rate of c-TEE was higher than that of c-TCD (χ²=20.885, P<0.001); for large shunts, the detection rate of c-TCD was higher than that of c-TEE (χ²=53.923, P<0.001). Among the patients, 34.9% (81/232) had a higher shunt grade assessed by c-TCD than by c-TEE; 55.6% (129/232) had the same shunt grade assessed by both c-TCD and c-TEE; 9.5% (22/232) had a lower shunt grade assessed by c-TCD compared to c-TEE. The PFO diameter of the inherent type RLS was greater than that of the potential type RLS (P<0.05), while there was no significant difference in tunnel length (P>0.05) as assessed by c-TCD. Correlation analysis showed a positive correlation between the shunt grade of PFO-RLS and PFO diameter (r=0.307, P<0.05), but there was no correlation with tunnel length (P>0.05).

c-TCD provides an effective hemodynamic basis for evaluating PFO-RLS, and its sufficient Valsalva maneuver can provide reliable semi-quantitative diagnosis results for PFO-RLS. c-TEE can identify the source of RLS. The combination of these diagnostic techniques provides a scientific basis for individualized diagnosis of RLS.