c-TTE and c-TEE with “Curtain effect” for detecting right-to-left shunt in patients with cryptogenic stroke

Abstract

This study compares right heart contrast transthoracic echocardiography (c-TTE) and contrast transesophageal echocardiography (c-TEE) at different states for detecting and grading the right-to-left shunt (RLS) in patients with cryptogenic stroke (CS). A total of 150 CS patients were enrolled. All patients underwent c-TTE and c-TEE at 3 different states: the Rest, the Valsalva Maneuver, and the state of rest shortly after the Valsalva Maneuver (referred to as “Curtain effect”). Right-to-left shunt due to patent foramen ovale (PFO-RLS) or pulmonary right-to-left shunt was identified by the microbubble characteristics in the left atrium. The detection rates and semiquantitative grades of RLS were compared between c-TTE and c-TEE under different states. c-TTE at Valsalva Maneuver detected more RLS than c-TEE (83.3% vs 65.3%, P < .05). c-TTE at “Curtain effect” revealed more PFO-RLS and higher grades of PFO-RLS than c-TTE at Rest (91.8% vs 72.7%, P < .05). c-TTE showed higher grades of PFO-RLS compared to c-TEE at any state (P < .05). c-TTE at Valsalva Maneuver or “Curtain effect” has superiority in detecting RLS and grading PFO-RLS compared to c-TEE; it can be a practical screening approach for suspected RLS in CS patients.

1. Introduction

Stroke is a common disease that threatens human life and health. Among them, strokes for which the etiology cannot be determined through existing diagnostic methods are called cryptogenic stroke (CS).^[1] Previous studies have found that right-to-left shunt due to patent foramen ovale (PFO-RLS) is one of the reasons for CS patients.^[2–4] Moreover, recent research has reported that pulmonary right-to-left shunt (P-RLS) is another cause for CS patients.^[5] Right heart contrast transthoracic echocardiography (c-TTE) and contrast transesophageal echocardiography (c-TEE) are widely used to detect right-to-left shunt (RLS); however, which is better for evaluating RLS is still controversial.

The patent foramen ovale (PFO) functions as a dynamic anatomical “curtain” between the atria. Repeated Valsalva maneuvers would transiently induce PFO to be easily opened (a phenomenon we define as the state of “Curtain effect”). This study aims to compare different approaches (c-TTE and c-TEE) at various states (Rest, Valsalva Maneuver, and “Curtain effect”) for detecting and grading RLS in CS patients, find the best combination of approach and state for initial assessment of RLS in CS patients, and provide a practical reference for clinical practice.

2. Materials and methods

2.1. Study protocol

The Ethics Committee of our hospital approved this study, and all patients signed informed consent forms ([2015]084-2). One hundred fifty patients diagnosed with CS in the Department of Neurology, comprising 72 males and 78 females, were enrolled between January 2013 and December 2023. Inclusion criteria: neurologists accurately diagnosed CS patients based on diagnostic criteria.^[6] All patients completed both c-TTE and c-TEE examinations with satisfactory image quality. Exclusion criteria: inability to cooperate with the Valsalva Maneuver or achieve a pressure of 35 mm Hg. Poor image quality. Other severe congenital heart diseases, atrial septal defects, heart failure, etc. The patient recruitment pathway is shown in Figure 1.

Figure 1.

The patient recruitment pathway.

2.2. c-TTE and c-TEE approaches

2.2.1. Instruments

Philips i.e. 33, EPIQ7C (Royal Philips N.V., Boston), and GE VIVID7 (GE HealthCare Technologies Inc., Milwaukee) with a transthoracic ultrasound probe S5-1 (1–5 MHz) and a transesophageal ultrasound probe X7-2t (2–7 MHz).

2.2.2. Right heart contrast agent

Patients were placed in the left lateral position, and the left elbow vein was selected for venipuncture (right elbow vein in 144 cases, right dorsal hand vein in 2 cases due to poor condition of the right elbow vein, left elbow vein in 3 cases, left dorsal hand vein in 1 case). The right heart contrast agent was prepared: 8 mL of saline + 1 mL of sterile air + 1 mL of blood.^[7]

2.2.3. Three different states

Microbubble characteristics in the left atrium (LA) are observed in 3 different states: Rest: the state of rest. The state of Valsalva Maneuver: instruct the patient to practice 1 to 2 times Valsalva Maneuvers until they perform it well (momentarily releasing the pressure gauge when the pressure reaches 35 to 40 mm Hg). The state of “Curtain effect”: after resting for 5 minutes, instruct the patient to perform Valsalva Maneuvers 5 times and then calm down quickly.

2.2.4. Contrast transthoracic echocardiography

Firstly, perform a routine TTE examination, observe the thin and soft part of the interatrial septum, check if the primary septum is separated from the secondary septum, observe the presence of shunt jets and their direction, comprehensively scan the heart, and exclude other lesions. After injecting the saline solution, observe and record the microbubble characteristics in the LA at 3 different states.

2.2.5. Contrast transesophageal echocardiography

Patients were fast and refrained from drinking water for over 6 hours. Oral Lidocaine gel was applied for local anesthesia of the throat surface (Lidocaine gel stays in the throat for exceeding 20 minutes to achieve a better anesthetic effect, instructing the patient to take it orally in divided doses, head-down when swallowing to reduce the incidence of aspiration pneumonia). Patients were placed in the left lateral position, and the probe was inserted into the mid to lower esophagus. Firstly, a routine TEE examination will be performed to observe the presence of PFO, atrial septal aneurysm (ASA), Eustachian valve, or Chiari network. The steps for performing right heart acoustic imaging were the same as those for c-TTE. Observe whether there were microbubbles in the LA at 3 different states. If necessary, check the LA’s left or right pulmonary vein section to distinguish between PFO-RLS and R-RLS.

2.3. RLS grading criteria^[8]

RLS was graded based on the number of microbubbles appearing in a single frame image: Grade 0: no microbubbles in the LA, indicating no RLS; Grade 1: 1 to 10 microbubbles per frame in the LA; Grade 2: 11 to 30 microbubbles per frame in the LA; Grade 3: >30 microbubbles per frame in the LA, or the LA almost filled with microbubbles. Differentiation between PFO-RLS and P-RLS was based on^[9]: time of left atrial microbubble imaging: considered PFO-RLS if appearing within 5 cardiac cycles, or it will be considered P-RLS. Duration of left atrial microbubble imaging: PFO-RLS is transient and related to right atrial microbubble concentration, while P-RLS persists until the right atrial microbubbles become faint/disappear. Origin of left atrial microbubbles: PFO-RLS originates from the gap between the primary and secondary septa, while P-RLS originates from the left/right pulmonary veins, and P-RLS is not significantly associated with the end of the Valsalva Maneuver (Fig. 2).

Figure 2.

The RLS grade shown by c-TTE and c-TEE in different states. (A) At Rest, c-TTE shows Grade 1 RLS. (B) At Valsalva maneuver, c-TTE shows Grade 3 RLS. (C) At “Curtain effect,” c-TTE shows Grade 2 RLS. (D) c-TEE in biatrial and bicavity views reveals fine separation between the primary and secondary septa with LRS. (E) At Rest, c-TEE shows Grade 1 RLS. (F and G) At Valsalva maneuver, c-TEE shows Grade 2 RLS, with microbubbles observed entering the left atrium from the atrial septum. (H) At “Curtain effect,” c-TEE shows Grade 1 RLS. c-TEE = contrast transesophageal echocardiography, c-TTE = contrast transthoracic echocardiography, LRS = left-to-right shunt, RLS = right-to-left shunt.

2.4. Statistical methods

Data were analyzed using SPSS 20.0 statistical software (IBM Corp., Armonk). Count data were expressed as percentages. McNemar’s test was used to compare the detection results of the 2 methods. The Wilcoxon rank-sum test was used to compare the grade data of RLS detection between the 2 methods. The Mann–Whitney U test was used to compare the grade data of RLS detection between different states. P < .05 was considered statistically significant.

3. Results

3.1. Study characteristics

In this study, 150 CS patients were enrolled from our hospital (78 females and 72 males). A total of 127 RLSs were detected by c-TTE and c-TEE, including PFO-RLS (110/150, 73.3%), P-RLS (8/150, 5.4%), combined PFO-RLS and P-RLS (7/150, 4.7%), and unknown cause (2/150, 1.3%). Among PFO-RLSs, 15 were associated with ASAs, and 11 were associated with the Eustachian valve or Chiari network.

3.2. c-TTE versus c-TEE for detecting RLS at Rest and Valsalva Maneuvers

At Rest, c-TTE and c-TEE had similar detection rates of RLS (64.7% vs 58.0%, P > .05) (Fig. 3). But at Valsalva Maneuver, c-TTE detected more RLSs than c-TEE (83.3% vs 65.3%, P < .05) (Fig. 4).

Figure 3.

Comparison of c-TTE and c-TEE for detecting RLS in CS patients at Rest (n = 150). CS = cryptogenic stroke, c-TEE = contrast transesophageal echocardiography, c-TTE = contrast transthoracic echocardiography, RLS = right-to-left shunt.

Figure 4.

Comparison of c-TTE and c-TEE for detecting RLS in CS patients at Valsalva Maneuver (n = 150). CS = cryptogenic stroke, c-TEE = contrast transesophageal echocardiography, c-TTE = contrast transthoracic echocardiography, RLS = right-to-left shunt. *P < .05.

3.3. c-TTE versus c-TEE for detecting pure PFO-RLS at Rest and “Curtain effect”

At “Curtain effect,” c-TTE detected more PFO-RLSs than Rest (91.8% vs 72.7%, P < .05) (Fig. 5). c-TEE at Rest and “Curtain effect” had similar detection rates of PFO-RLS (63.6% vs 67.38%, P > .05) (Fig. 6).

Figure 5.

Comparison of c-TTE for detecting pure PFO-RLS at Rest and “Curtain effect” (n = 110). c-TEE = contrast transesophageal echocardiography, c-TTE = contrast transthoracic echocardiography, PFO = patent foramen ovale, RLS = right-to-left shunt. *P < .05.

Figure 6.

Comparison of c-TEE at Rest and “Curtain effect” for detecting pure PFO-RLS. c-TEE = contrast transesophageal echocardiography, PFO = patent foramen ovale, RLS = right-to-left shunt.

3.4. c-TTE versus c-TEE for semiquantitative grading pure PFO-RLS at different states

c-TTE at “Curtain effect” had higher grades of PFO-RLS than at Rest (P < .05) (Fig. 7). c-TTE had higher PFO-RLS grades than c-TEE at Rest and “Curtain effect” (Figs. 8 and 9).

Figure 7.

Comparison of c-TTE at Rest and “Curtain effect” for semi-quantitative grading pure PFO-RLS (n = 110). c-TEE = contrast transesophageal echocardiography, c-TTE = contrast transthoracic echocardiography, PFO = patent foramen ovale, RLS = right-to-left shunt.

Figure 8.

Comparison of c-TTE and c-TEE for semi-quantitative grading pure PFO-RLS at Rest (n = 110). c-TEE = contrast transesophageal echocardiography, c-TTE = contrast transthoracic echocardiography, PFO = patent foramen ovale, RLS = right-to-left shunt.

Figure 9.

Comparison of c-TTE and c-TEE for semi-quantitative grading pure PFO-RLS at Valsalva Maneuver (n = 110). c-TEE = contrast transesophageal echocardiography, c-TTE = contrast transthoracic echocardiography, PFO = patent foramen ovale, RLS = right-to-left shunt.

4. Discussion

c-TTE and c-TEE, better for detecting and grading RLS, are still controversial.^[10–12] Our results support that c-TTE at the Valsalva Maneuver or “Curtain effect” may be an initial approach for detecting and grading the RLS in CS patients.

The structure of the PFO is like a door curtain, which is equivalent to the “oval valve” door curtain. After an effective Valsalva Maneuver, it has not completely closed for a short time. Therefore, we propose a new state of “Curtain effect.” We found that c-TTE at the “Curtain effect” detected more and had higher grades of PFO-RLS than c-TTE at the rest. In addition, c-TTE at the Valsalva Maneuver detected higher grades of PFO-RLS than c-TEE. The reason may be that c-TEE is an invasive procedure requiring fasting, esophageal intubation, and pharyngeal anesthesia, which may result in poor patient compliance with the Valsalva Maneuver.

In previous research, PFO-RLS was considered the main reason for paradoxical embolism.^[5,13] However, recent studies, including ours, reported that P-RLS may also be the pathway for CS. Physiological pulmonary arteriovenous anastomoses (diameter 20–50 μm) may exist in some healthy individuals, which are rarely and nonfunctionally open under normal circumstances. Under certain conditions, such as stress, fatigue, inflammation, and hypoxia, they may open and lead to P-RLS.^[14–16] Subtle physiological pulmonary arteriovenous anastomoses or PFO may explain our study’s 2 cases with unknown causes.

Nevertheless, c-TEE provides direct visualization of PFO size, morphology, residual atrial septal length, and other related anatomical structures, facilitating the differentiation of RLS sources more intuitively. In our study, c-TEE revealed 10 cases of P-RLS, clearly showing that left atrial microbubbles originated from the left upper pulmonary vein. It may be because the left upper pulmonary vein is easier to observe than the other pulmonary veins.

In addition, our study revealed that ASA, the Eustachian valve, and the Chiari network are related to PFO-RLS, according to previous literature.^[17] ASA provides favorable conditions for paradoxical embolism, possibly due to the increased opening amplitude and frequency of PFO caused by increased atrial septal activity, thereby increasing RLS.^[7,18] Some scholars argued that thrombi are more likely to form at the location of ASA, which may play a significant role in the pathogenesis of CS.^[19] Concurrent ASA and PFO are considered high-risk factors, with a 3 to 5 times higher risk of recurrent embolic events compared to patients with isolated PFO.^[20]

This study still has some limitations. Firstly, this study was conducted at a single tertiary center, potentially limiting the generalizability of our findings. While adequate for overall RLS detection, the sample size was insufficient for robust subgroup analyses, particularly for P-RLS (n = 8) and combined RLS (n = 7), limiting definitive conclusions about these less prevalent shunt types. Secondly, our cohort consisted exclusively of patients diagnosed with CS who underwent both c-TTE and c-TEE. This inherent selection bias may overrepresent patients with a higher pretest probability of RLS than the general CS population or asymptomatic individuals.

5. Conclusion

In summary, c-TTE at Valsalva Maneuver or “Curtain effect” has superiority in detecting RLS and grading PFO-RLS compared to c-TEE; it can be a practical screening approach for suspected RLS in CS patients.

Acknowledgments

The authors thank First Affiliated Hospital of Fujian Medical University for managing our patient database.

Author contributions

Conceptualization: Zi-Ling You, Da-Jun Chai, Qin-Yun Ruan, Xiao-Yan Lin.

Data curation: Zi-Ling You, Sheng Cheng, Liang-Liang Yan, Hui-Xin Wei, Long-Fei Chen, Chao-Yang Qu, Qin-Yun Ruan, Xiao-Yan Lin.

Formal analysis: Zi-Ling You, Da-Jun Chai, Sheng Cheng, Qin-Yun Ruan, Xiao-Yan Lin.

Funding acquisition: Xiao-Yan Lin.

Investigation: Zi-Ling You, Sheng Cheng, Hui-Xin Wei, Chao-Yang Qu, Xiao-Yan Lin.

Methodology: Zi-Ling You, Sheng Cheng, Liang-Liang Yan, Xiao-Yan Lin.

Project administration: Qin-Yun Ruan, Xiao-Yan Lin.

Resources: Da-Jun Chai, Xiao-Yan Lin.

Software: Da-Jun Chai, Sheng Cheng, Liang-Liang Yan.

Supervision: Da-Jun Chai, Sheng Cheng.

Validation: Da-Jun Chai, Sheng Cheng.

Visualization: Sheng Cheng.

Writing – original draft: Zi-Ling You, Da-Jun Chai, Sheng Cheng.

Writing – review & editing: Zi-Ling You, Xiao-Yan Lin.