Remember me
ORIGINAL ARTICLE
Year : 2022 | Volume
: 32
| Issue : 4 | Page : 193-199
Two-dimensional strain echocardiographic parameters and clinical outcomes associated with significant atrioventricular regurgitation in a single-center adult fontan population
Tse Ben Chen1, Gnalini Sathananthan2, Mikyla Janzen2, Jasmine Grewal2
1 Division of Cardiology, St. Paul's Hospital, University of British Columbia; Department of Medicine, University of British Columbia, Vancouver, Canada
2 Division of Cardiology, St. Paul's Hospital, University of British Columbia, Vancouver, Canada
Correspondence Address:
Tse Ben Chen
St. Paul's Hospital, RM 478-1081 Burrard Street, Vancouver, BC
Canada
Source of Support: None, Conflict of Interest: None
CheckDOI: 10.4103/jcecho.jcecho_24_22
Background: Significant atrioventricular valve regurgitation (AVVR) is prevalent in Fontan adults. Two-dimensional speckle-tracking echocardiography allows for evaluation of subclinical myocardial dysfunction and offers technical benefits. We aimed to evaluate the association of AVVR with echocardiographic parameters and adverse outcomes. Materials and Methods: Fontan adults (≥18 years) with lateral tunnel or extracardiac connection actively followed at our institution were retrospectively reviewed. Patients with AVVR on most recent transthoracic echocardiogram (≥grade 2 as per American Society of Echocardiography guidelines) were matched with Fontan controls. Echocardiographic parameters, including global longitudinal strain (GLS), were measured. The composite outcome of Fontan failure included Fontan conversion, protein losing enteropathy, plastic bronchitis, and New York Heart Association Class III/IV. Results: Sixteen patients (14%, mean age 28.4 ± 7.0 years) with predominantly moderate AVVR (81%) were identified. The mean duration of AVVR was 8.1 ± 5.8 months. There was no significant reduction in ejection fraction (EF) (51.2% ± 11.7% vs. 54.7% ± 10.9%, P = 0.39) or GLS (−16.0% ± 5.2% vs. −16.0% ± 3.5%, P = 0.98) associated with AVVR. Larger atrial volumes and longer deceleration time (DT) were observed in the AVVR group. Patients with AVVR and a worse GLS (≥−16%) had higher E velocity, DT, and medial E/E' ratio. The incidence of Fontan failure did not differ from controls (38% vs. 25%, P = 0.45). Patients with worse GLS (≥−16%) demonstrated a marked trend toward a higher incidence of Fontan failure (67% vs. 20%, P = 0.09). Conclusions: In Fontan adults, a short duration of AVVR did not influence EF or GLS but was associated with larger atrial volumes and those with worse GLS demonstrated some differences in diastolic parameters. Larger multicenter studies throughout its disease course are warranted.
Keywords: Atrioventricular valve, echocardiography, Fontan, global longitudinal strain, regurgitation, strain
The Fontan procedure has transformed the survival and quality of life of newborns with single ventricle physiology since its inception.[1],[2] Complex cardiac malformations characterized by a single functional ventricle are palliated through staged procedures to create a systemic single ventricle while maintaining passive flow from the systemic venous to pulmonary circulation.[3] However, these patients remain susceptible to various long-term complications and require meticulous monitoring into adulthood.[4],[5],[6],[7]
Significant atrioventricular valve regurgitation (AVVR) is a common observation, with a reported incidence of 16%–41% in the adult Fontan population and recent studies demonstrating its relationship with higher rates of Fontan failure.[8],[9],[10],[11],[12] The cumulative incidence of AVVR is expected to increase with age and is higher in patients with common atrioventricular (AV) valves and single tricuspid valves, compared to single mitral or two AV valves.[11] Given the high recurrence and mortality risk of surgical AV valve intervention performed after Fontan palliation,[11],[13] it is important to improve our understanding of the relationship between AVVR and other markers of poor prognosis.
Ventricular assessment is important in the context of AVVR. This becomes a challenge in the setting of a single ventricle given structural heterogeneity, lack of ventricular–ventricular interaction, and the loading conditions associated with Fontan palliation.[14] Assessment of myocardial strain via two-dimensional (2D) speckle-tracking echocardiography (STE) has gained traction in single ventricle cohorts, given its ability to measure myocardial deformation without geometric assumptions or a reliance on qualitative observation.[15] Global longitudinal strain (GLS) has been shown to provide additional prognostic value and early detection of subclinical myocardial dysfunction compared to conventional echocardiographic parameters.[15],[16] The utility of strain analysis and the impact of AVVR on myocardial deformation in the adult Fontan population have not been investigated in detail.
In this study, we aimed to evaluate the association of AVVR with echocardiographic parameters and adverse clinical outcomes.
Materials and MethodsStudy cohort
This is a retrospective study of adult (≥18 years of age) Fontan patients followed at our institution as of April 2019. Cases included patients with a lateral tunnel or extracardiac Fontan connection with AVVR. Controls were Fontan adults matched to cases by age and sex on a 1:1 basis.
Patients without an echocardiogram performed at our institution within 12 months of the most recent clinical assessment were excluded. Due to the prospective nature of our institutional database, deceased patients or those who received cardiac transplantation were excluded from this study as there were significant gaps in data quality and availability. This study was approved by the institution's ethics review board before commencement.
Clinical data collection
All patients underwent at least annual clinical review at our institution. The following data were obtained from the patient records: underlying cardiac diagnoses, age at and type of Fontan procedure, other cardiac interventions, cardiac medications, and history of cardiac events that had occurred up until the time of last clinical follow-up. Data from additional investigations were included only if they occurred within 2 years of the most recent echocardiogram.
Echocardiography
Transthoracic echocardiography was performed using either the Philips IE 33 system (Philips Medical Systems, Bothell, WA, USA) or GE Vivid 7 system (GE Healthcare, Chicago, IL, USA). The most recent echocardiogram (as of April 2019) for cases and controls was independently reviewed as described below by a primary reader who was blinded to the clinical data. The most recent echocardiogram was chosen to allow for higher image quality sufficient for strain analysis. It is our institution's policy to perform, at minimum, annual echocardiograms as part of routine monitoring of these patients.
The morphology of the ventricle was determined using a combination of echocardiography and multimodality imaging and was classified as either left ventricle (LV), right ventricle (RV), or indeterminate. A standard 2D parasternal long axis (PLAX) was used to determine the linear dimensions of the single ventricle. The systemic ventricular ejection fraction (EF) was calculated using the Simpsons biplane method of discs. A single plane EF was calculated when inadequate images precluded biplane calculation. GLS was measured using the TOMTEC Imaging Systems GmbH, Unterschleissheim, Bavaria, Germany [Figure 1] using the 2-, 3-, and 4-chamber views of the systemic ventricle when possible. For this study, a more negative GLS represents a greater degree of myocardial deformation and is communicated as having better GLS. Traditional parameters for the assessment of diastolic function were used according to the American Society of Echocardiography (ASE).[17] A single plane or biplane method was used to calculate left atrial volume, depending on the quality of images available. The right atrial volume was measured excluding the Fontan.
Figure 1: GLS measurement using the Tom Tec AutoStrain software on a 4-chamber view of the systemic ventricle, GLS = Global longitudinal strainQualitative and quantitative assessment of valvular regurgitation was performed according to the ASE guidelines.[18] The severity of regurgitation was graded into the following categories: 1 = none/trivial, 2 = moderate, 3 = moderate-to-severe, and 4 = severe. We defined AVVR as a grade of ≥2. Given the limitations of evaluating valvular regurgitation in the heterogeneous group of single ventricle patients, Categories 3 and 4 were combined and denoted as “moderate-to-severe–severe” AVVR. The duration of AVVR was quantified in cases as the time interval between the most recent echocardiogram to the earliest echocardiogram in which AVVR (of moderate or greater severity) was detected.
The Fontan circuit was interrogated from multiple views as is the practice in our echocardiographic laboratory. The peak velocity in the distal end of the Fontan and Glenn was measured from the suprasternal and high parasternal views. The peak velocity in the proximal Fontan tunnel was measured from the subcostal view.
Fontan adverse clinical outcomes
Adverse clinical outcomes were adjudicated from each patient's last clinical visit. Fontan failure was defined as a composite of requiring Fontan conversion, protein-losing enteropathy,[19] plastic bronchitis, or New York Heart Association (NYHA) class III or IV. Additional markers of Fontan circulatory dysfunction such as clinical heart failure, refractory arrhythmias, and thromboembolic complications were documented. Thromboembolic complications were defined as follows: (1) Fontan circuit thrombus and (2) Pulmonary artery thromboembolism. Clinical heart failure was defined as either a hospital admission for inpatient treatment of heart failure or progression of relevant symptoms and/or escalation of diuretic therapy in an outpatient setting. Refractory arrhythmia was defined as ≥2 inpatient hospital visits/admissions occurring in a given year for arrhythmias refractory to pharmacologic management or electrophysiologic ablation. Extracardiac dysfunction was noted and included the presence of liver disease/dysfunction and/or kidney dysfunction. When available, and if performed within 2 years of the most recent echocardiogram, cardiopulmonary exercise test results were reviewed. The mean GLS of the combined cohort (–16%) was used to stratify patients for comparison.
Statistical analysis
SPSS version 27 software (SPSS, Inc., Chicago, Illinois, USA) was used for data analysis. Descriptive statistics were used with continuous variables reported as mean ± standard deviation and categorical variables reported as frequencies. Between-group comparisons were analyzed using Fisher's exact test or Chi-square test where appropriate for categorical variables. The unpaired t-test was used to analyze continuous variables. Results were deemed statistically significant if P < 0.05.
ResultsCohort characteristics
A total of 117 adult Fontan patients with lateral tunnel or extracardiac conduit followed in the PACH clinic were identified. 16 patients had AVVR (14%) and were analyzed along with 16 matched Fontan controls without AVVR. The baseline characteristics are detailed in [Table 1]. The most common anatomic diagnoses in the AVVR group were double inlet LV (n = 3), pulmonary atresia (n = 3), and AV septal defect (n = 3). Only one patient in the AVVR group had a diagnosis of hypoplastic left heart syndrome, given the exclusion of transplanted patients. One patient in the AVVR group had prior AV valve intervention, consisting of intervention 2 years before Fontan completion, with recurrence of significant AVVR 15 years following intervention. One control had AV valve intervention at the time of Fontan operation. Patients in the AVVR group tended to have a higher NYHA functional class and were more likely to be on an angiotensin-converting enzyme (ACE) inhibitor.
Table 1: Baseline characteristics of Fontan patients with significant atrioventricular valve regurgitation and Fontan controlsEchocardiographic parameters
Of the 16 patients with AVVR, 13 (81%) had moderate severity regurgitation and 3 (19%) had moderate-to-severe–severe AVVR. The mean duration of AVVR in this study was 8.1 ± 5.8 months [range, 1–17 months; [Table 2]]. None of the 16 patients with AVVR had additional valvular disease of moderate or greater severity.
Table 2: Characteristics of patients with significant atrioventricular valve regurgitationParameters of systolic function of the systemic ventricle were comparable between the AVVR and controls group [Table 3]. EF and GLS measurements were similar between groups. GLS was similar between the AVVR and control groups regardless of ventricular morphology: LV (–14.6% ± 6.2% vs. –15.7% ± 3.8%, P = 0.65) and RV (–17.5% ± 3.9% vs. –16.4% ± 3.8%, P = 0.63). Within the AVVR group, GLS did not differ significantly between the patients with RV morphology (n = 8) versus LV morphology (n = 8) (–17.5% ± 3.9% vs. –14.6% ± 6.2%, P = 0.28). The majority of patients had an EF >40% (94% of AVVR group; 88% of controls) and an EF >50% (63% of AVVR group; 75% of controls). 63% of the patients in the AVVR group and 43% in the control group had a GLS better than –16%. Patients with AVVR and a GLS better than −16% had a trend toward higher EF compared to those with a GLS of –16% or worse, although this did not reach statistical significance (55% ± 9% vs. 45% ± 13%, respectively; P = 0.09). ACE inhibitor or beta blocker use in AVVR patients was not associated with a difference in GLS or EF.
Table 3: Echocardiographic parameters in patients with significant atrioventricular valve regurgitation and controlsThe majority of diastolic parameters did not differ significantly between the two groups, as reflected by similar E/A ratio, medial and lateral E/E', and pulmonary vein S/D ratio. However, the deceleration time (DT) was higher in the AVVR group, at 208 ± 32 ms, compared to 170 ± 35.8 ms in the controls (P = 0.03). Furthermore, larger atrial volumes were observed in patients with AVVR. The mean left atrial dimension (PLAX view) was increased in the AVVR group (43.7 ± 11.3 mm vs. 35.8 ± 6.4 mm; P = 0.04), and the biplane left atrial volume index trended to be higher. With stratification by GLS, patients with AVVR and a worse GLS (≥–16%) had increased E velocity (92.3 ± 21 mm/s vs. 65 ± 13 mm/s, P = 0.03), increased DT (236 ± 1.4 ms vs. 199 ± 32 ms, P = 0.04), and increased medial E/E' ratio (15.3 ± 3.2 vs. 8.3 ± 1.5, P = 0.009) compared to patients with a better GLS (<–16%).
Fontan adverse clinical outcomes
In this study with a mean AVVR duration of 8.1 months, the composite outcome of Fontan failure did not differ between patients with AVVR and controls [Table 4]. Indices of liver and kidney disease and exercise capacity were comparable between the two groups. In the AVVR group, patients with worse GLS (≥–16%) had a trend toward a higher incidence of Fontan failure compared to patients with better GLS (<–16%), although this did not achieve statistical significance (67% vs. 20%, P = 0.09). In the AVVR group, patients with worse GLS had similar incidence of clinical heart failure (33% vs. 30%, P = 0.65), refractory arrhythmia (17% vs. 0%, P = 0.38), and thromboembolic complications (0% vs. 20%, P = 0.38) compared to those with better GLS. Extracardiac organ dysfunction, exercise capacity, and routine blood work did not otherwise differ with stratification by GLS or EF within the AVVR group.
Table 4: Fontan adverse clinical outcomes in patients with significant atrioventricular valve regurgitation and controlsIn the overall group, patients with worse GLS (≥–16%) had a marked trend toward a higher incidence of Fontan failure compared to patients with better GLS (47% and 18%, respectively; P = 0.08). Patients with worse GLS had lower platelet levels (145 × 109/L vs. 193 × 109/L, P = 0.04) as well as a trend toward higher N-terminal pro B type natriuretic peptide (352 pg/mL vs. 179 pg/mL, P = 0.17) and lactate dehydrogenase levels (225 U/L vs. 162 U/L, P = 0.12). The incidence of adverse clinical outcomes, including Fontan circulatory dysfunction, extracardiac organ dysfunction, and exercise capacity, were otherwise similar between patients with worse GLS (≥–16%) and those with better GLS (<–16%) in the combined cohort.
DiscussionIn this contemporary cohort of adult Fontan patients with either lateral tunnel or extracardiac conduits, 14% had AVVR of moderate and greater severity, which is comparable to other studies reporting incidences of 16%–41%.[8],[9],[10],[11],[12] The echocardiographic parameters, specifically markers of ventricular function, did not differ significantly between the AVVR and control groups after a short duration of AVVR (mean, 8.1 months). Within the AVVR group, stratification by GLS revealed differences in diastolic parameters in patients with reduced myocardial deformation. Based on these initial data, although parameters of ventricular function likely carry some prognostic value in the Fontan population, they may not detect substantial changes in the short term and are likely, not the most robust tools, to help guide timing of intervention in the earlier course of AVVR.
Strain echocardiography has several advantages including the ability to detect subclinical myocardial dysfunction and the lack of geometric limitations. Interestingly, however, GLS was similar between Fontan patients with AVVR and controls in the short term. With a short duration of AVVR (8.1 months) and a predominantly moderate severity AVVR cohort (81%), this may represent adequate compensation of myocardial function in the setting of a lower burden of regurgitation or adaptation of the single ventricle over time. This may be applicable to our cohort with a mean age of 28 years and a mean duration of 20 years post-Fontan. Alternatively, the sequelae of regurgitation evidenced by increased atrial volumes in this study point toward the complicated nature of GLS interpretation in the Fontan population as multiple competing factors are expected to influence measurements. Certainly, although less preload-dependent than conventional measures,[20] the increased output associated with the regurgitant volume may result in increased strain so that it is similar compared to the preload-deficient state of Fontan controls.[15],[21] This is juxtaposed by progressive dilatation of the single ventricle which would cause a decrease in strain measurements. A recent study demonstrated this preservation of GLS despite AVVR on cardiac magnetic resonance strain analysis in the Fontan population but also identified a worse global circumferential strain with AVVR. It was suggested that these findings may represent adaptive fiber arrangements or preferred directions of deformation, which may not be captured via GLS and may in part also explain our findings.[22],[23] At last, the AVVR group in our cohort was more likely to be on an ACE inhibitor, which may improve strain measurements, although this benefit cannot be distinguished from treatment bias. Clearly, there are a number of confounding factors that make it difficult to conclude that the absence of a difference in GLS truly reflects comparable myocardial function between Fontan patients with and without AVVR. These findings in the context of unique physiology and short burden of AVVR likely highlight the shortcomings of traditional parameters of ventricular dysfunction in detecting subclinical signs of deterioration in adult Fontan patients with AVVR in its early disease course, although these changes are expected to be more apparent in the long term.
Despite the lack of significant differences in systolic function early on, the evident atrial enlargement and differences in diastolic parameters, especially with lower GLS, reaffirm the need for routine surveillance of adult Fontan patients with AVVR and may precede the deterioration of ventricular function over time.[24] Atrial enlargement has been shown to be associated with AVVR, higher rates of atrial tachyarrhythmias, and an increase in the combined end point of mortality and heart transplantation in Fontan survivors.[25],[26] Along with the general lack of consensus in the optimal timing and indication for surgical intervention of the AV valve,[27] these findings further support the need for additional research into the utility of Speckle-Tracking Echocardiography (STE) and additional imaging modalities in the ongoing assessment of AVVR in the adult Fontan population as well as how these parameters may vary over its disease course. Many potential GLS cutoffs for intervention have been proposed in adults with acquired mitral regurgitation to identify patients at risk of worse clinical outcomes.[20] In this study, patients with AVVR and a GLS of ≥–16% demonstrated some differences in diastolic parameters and a trend toward a higher rate of Fontan failure although the latter was not statistically significant. Additional research is necessary to identify the role of STE in stratifying patients in a similar manner. Provided that GLS may be preserved in AVVR early on, higher GLS thresholds than the control population have been proposed in adults with acquired mitral regurgitation.[28],[29] A similar consideration for the use of STE in the Fontan population is certainly of value.
AVVR is associated with Fontan circulatory dysfunction likely secondary to an increase in systemic venous pressures due to elevated atrial and Fontan pressures as well as decreased effective cardiac output.[30] One recent study by Moon et al. identified AVVR of at least moderate severity after Fontan completion to be an independent risk factor of death, takedown or heart transplant after Fontan (hazard ratio: 2.4, 95% confidence interval: 1.5–3.8).[31] Although our study did not include patients who were later deceased or received heart transplantation, we noted a trend toward higher baseline NYHA class and higher rates of clinical heart failure, liver cirrhosis, and Fontan failure, but this was not statistically significant and deviates from the findings established by recent studies in large cohorts.[11],[31] This is most likely due to our exclusion of transplanted and deceased patients as well as the small sample size and short duration of AVVR in our cohort (mean, 8.1 months), representing an early stage of the disease trajectory that is not consistent with longer-term outcomes. This was suggested in the Moon et al.'s study where the severity of AVVR as well as its associated ventricular dysfunction was observed to increase with time. Another recent study by King et al. similarly demonstrated a twofold increase in Fontan failure in patients with AVVR, with death and transplantation representing a significant portion of the composite outcome.[11] The majority of data in clinical outcomes associated with AVVR have been collected from pediatric cohorts and often assess AVVR present before Fontan completion, which would encompass operative risks in the reported mortality and morbidity.[4],[7],[32],[33] Our study is unique in its focus on the adult population and contemporary Fontan circuits early in the AVVR disease course, although further research is necessary to replicate these contradictory findings in the short term, compared to established long-term findings. This may prove to inform surveillance protocols and the optimal timing of intervention.
Limitations
As the severity of AVVR in our cohort was predominantly moderate with few cases of worse severity, the findings are only applicable to the population as described. In our opinion, this distribution of AVVR severity is applicable to many modern cohorts and may represent an earlier stage of disease at which consideration for intervention might be most prudent and effective. Despite being the only referral center in our province, the small cohort and short duration of AVVR are limitations to this study and preclude decisive extrapolation of these findings to other cohorts. However, we believe that these findings establish the value of larger multicenter studies on this topic. Similarly, the small sample size of patients with AVVR in our cohort precluded the grouping of patients by cardiac diagnoses. This may contribute to strain measurements given the heterogeneity of loading conditions experienced throughout the stages of Fontan palliation. The cross-sectional analysis of the clinical outcomes similarly precludes any establishment of temporal relationships between echocardiographic changes, AVVR, and outcomes.
ConclusionsCompared to Fontan controls, there was no significant reduction in EF or GLS seen in patients with AVVR in the short term. Larger atrial volumes were observed in patients with AVVR early on, with differences in diastolic parameters seen in those with significantly worse GLS (≥–16%) and may suggest further deterioration over time. Strain assessment in adult Fontan patients with AVVR at various timeframes throughout its disease course may inform surveillance and intervention decisions and warrants further investigation through multicenter studies.
Ethical clearance
This study was approved by the University of British Columbia Research Ethics Board prior to commencement.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References
Comments (0)