Unexpected prolapse of the anterior leaflet during saline testing in mitral valve repair

Abstract

Open in new tabDownload slide

OBJECTIVES

Saline testing is used in mitral valve (MV) surgery to evaluate the repair intra-operatively. Sometimes, saline testing shows a prolapse of the anterior leaflet, not seen on preoperative echocardiography. Our objective was to investigate the incidence, predisposing factors and consequences of this phenomenon.

METHODS

We retrospectively reviewed all consecutive patients undergoing surgery for posterior leaflet prolapse between 2013 and 2017. All data, including intraoperative video recordings of the repair and saline testing, were collected prospectively.

RESULTS

Isolated posterior leaflet repair was performed in 91 patients. In 17 patients (18.7%), saline testing showed an unexpected anterior leaflet prolapse. Patients with unexpected prolapse presented with higher body mass index (BMI) compared to the reference group (27.5 ± 2.3 vs 25.0 ± 4.2, P = 0.01). Binomial logistic regression analysis showed BMI, surgical approach, number of prolapsing segments, left ventricular ejection fraction, left ventricular end systolic diameter and left atrial diameter to be predictive for unexpected anterior leaflet prolapse. In patients with unexpected anterior leaflet prolapse, no adequate saline testing was possible and repair was accomplished based on correction of the prolapse as seen on a preoperative echocardiogram. In both groups, 100% repair rate was achieved. Predischarge mitral regurgitation grading showed mild or less mitral regurgitation in all the patients in the unexpected prolapse group in comparison with 98.6% of patients in the reference group.

CONCLUSIONS

When saline testing shows an unexpected prolapse of the anterior leaflet, not present on preoperative echocardiography, no additional surgical techniques should be performed in order to achieve an excellent postoperative result. Further research is warranted to predict unexpected anterior leaflet prolapse preoperatively.

INTRODUCTION

Mitral valve repair (MVr) has become the preferred method for surgical correction of mitral regurgitation (MR) in degenerative MV disease [1, 2]. Understanding the complex anatomy of the mitral valvular apparatus and its functional entity is essential for successful repair. Preoperative transthoracic echocardiography (TTE) and additional transoesophageal echocardiography (TOE) provide anatomical characteristics of the different lesions in relation to segmental and functional anatomy, according to the Carpentier classification, in order to assess the feasibility of repair [3]. Even with extensive preoperative planning with TTE and TOE, intraoperative final analysis is crucial for understanding the complex MV pathology and to determine the optimal repair technique. In 49–56% of the patients with degenerative MV disease, the predominant lesion is a prolapse of the posterior leaflet, mainly resulting from chordal rupture [4–7]. Posterior leaflet prolapse can successfully be repaired in most patients with repair rates above 95% and equally high freedom from reoperation [7, 8]. Repair rates for MV surgery are shown to be procedural volume related, and the results may vary among cardiac surgeons [9, 10]. Several preoperative strategies and intraoperative tests have been proposed to achieve higher repair rates. However, intraoperative assessment of the repair highly relies on visual inspection. Saline testing is the most common technique used to inspect the repaired valve intraoperatively and could simulate the postoperative functional anatomy (Video 1, Part I) [11]. Confirmation of valve competency before closure of the atrium is crucial and will prevent reinstitution of cardiopulmonary bypass (CPB).

Although the published data show a high repair rate and long-term durability for repair of degenerative MV regurgitation in experienced hands, the overall repair rate in registry data is much lower, implying that less-experienced MV surgeons fail in repairing the most common valve pathology [12]. In MV repair for isolated posterior leaflet prolapse, saline testing sometimes shows an unexpected full prolapse of the anterior leaflet, which was not present on preoperative echocardiography (Video 1, Part II). This finding may lead inexperienced surgeons to correct the anterior leaflet prolapse, potentially resulting in a suboptimal repair, prolonged operating times and potential residual MR due to overcorrection and could eventually lead to valve replacement. Experienced surgeons use several intraoperative manoeuvres to counter this phenomenon, such as (i) pulling down the diaphragm, (ii) repositioning the atrial retractor and (iii) forcing down the anterior part of the annuloplasty ring. Despite these manoeuvres, some patients still maintain a prolapse of the anterior leaflet during saline testing (Video 2). A possible explanation for the occurrence of this phenomenon could be the displacement of the ventricle and the subvalvular apparatus in the arrested heart [13].

Our objective was to evaluate the incidence of an unexpected anterior leaflet prolapse in saline testing for isolated posterior leaflet repair, investigate the predisposing factors underlying its occurrence and determine the consequences of surgical repair.

METHODS

In our tertiary referral centre, patients accepted for MV surgery are included in a prospective database. Preoperative variables, intraoperative data, including intraoperative video recordings, and postoperative outcomes are collected prospectively. We retrospectively reviewed all data of patients who underwent MV surgery by a single surgeon (P.S.N.) for degenerative MV disease at our institution from January 2013 to December 2017 (n = 182). Emergency surgery, reoperations and patients who underwent beating heart procedure were excluded (n = 24) [14]. As it is clinically important to evaluate the unexpected anterior leaflet prolapse in isolated posterior MV prolapse, this group forms the focus of the current study. The institutional review board of the Maastricht University Medical Centre approved the study and waived the need for informed consent due to the observational character of the study.

ECHOCARDIOGRAPHY

All patients eligible for MV surgery underwent preoperative TTE to determine the exact MV pathology. In most of the cases, preoperative TTE was performed at the referral centre in compliance with local agreements and international recommendations [15]. In patients where the mechanism of regurgitation was not clarified on TTE and in all the patients who were eligible for minimally invasive surgery, a TOE was performed. Intraoperative TOE was performed in all patients, prior to incision and post repair to assess the quality of the MV repair and left ventricular function after weaning of CPB, by an imaging cardiologist experienced in MV treatment. Furthermore, all the patients underwent predischarge TTE to assess the quality of the repair and to grade postoperative MR. The severity of MR was graded based on the current international guidelines [16].

SALINE TESTING

After MVr is completed, a 50-cc bulb syringe is filled with heparinized saline which is injected at low pressure through the MV to fill the left ventricle in the arrested heart to mimic the systolic phase. Then, a second injection of saline with higher pressure is given simultaneously with anterograde crystalloid cardioplegia (St. Thomas’ cardioplegic solution No. 2, Plegisol, Abbott Laboratories, North Chicago, IL). An optimal result is indicated by a posterior position of the closure line with a symmetrical aspect. In all the patients, saline testing was performed in the same manner by the same surgeon. Additionally, during saline testing, ink testing was used to estimate the length of coaptation [17]. In patients with unexpected prolapse of the anterior leaflet, no adequate saline testing could be performed during pressurization of the left ventricle, and anterior leaflet prolapse occurred, resulting in immediate deflation of the ventricle. In these patients, only the posterior leaflet prolapse was corrected, relying solely on the preoperative TOE, and correction of only the prolapsing segments was performed.

Statistical analysis

Statistical analysis was performed using SPSS 24.0 for Macintosh (SPSS Inc., Chicago, IL, USA). Normality of the continuous variables was tested with visual inspection of the histograms and the Shapiro–Wilk test. Continuous normally distributed data are presented as mean with standard deviation. Continuous non-parametric distributed data are presented as median with interquartile range. Frequencies are displayed as absolute numbers and relative percentages. Baseline characteristics and outcomes were compared using the χ² test for categorical data and the Fisher’s exact test when the minimum expected cell-size assumption did not apply. The Student’s t-test and Mann–Whitney U-test were performed for continuous parametric and non-parametric data, respectively. Additionally, a binomial logistic regression analysis was performed. Variables for binomial logistic regression analysis were selected based on the baseline difference and theoretical hypothesis. Linearity of the continuous variables with respect to the logit of the dependent variable was assessed via the Box–Tidwell procedure. On the basis of this assessment, all continuous independent variables added to the model were found to be linearly related to the logit of the dependent variable. A receiver-operating characteristic analysis was used to determine the diagnostic value of our model. All reported P-values are 2-sided, and P-values of <0.05 are considered statistically significant.

RESULTS

Baseline characteristics

After an analysis of our database, we identified 158 patients with degenerative MV disease. The underlying mechanism of MR was an isolated prolapse of the posterior leaflet in 91 (57.6%) patients, isolated prolapse of the anterior leaflet in 13 (8.2%) patients, commissural prolapse in 3 (1.9%) patients, bileaflet prolapse in 48 (30.4%) patients and calcifications in 3 (1.9%) patients.

Subsequently, a total of 91 patients with isolated posterior leaflet prolapse were included. In 17 (18.7%) patients, saline testing showed an unexpected prolapse of the anterior leaflet (unexpected prolapse group). Mean age was 65.2 ± 7.3 and 65.5 ± 9.6 (P = 0.92) in the unexpected prolapse group and the reference group, respectively. Gender was equally distributed between the groups. Biometrics [length, weight, body mass index (BMI), body surface area] showed BMI to be significantly increased in the unexpected prolapse group, with a mean BMI of 27.5 ± 2.3 compared to a mean BMI of 25.0 ± 4.2 in the reference group (P = 0.01). Median logistic EuroSCORE was 2.2 (3.1) and 3.4 (4.4) (P = 0.33) in the unexpected prolapse group and the reference group, respectively. Further characteristics are depicted in Table 1.

Surgical procedure

Patients in the reference group were more frequently operated on through a sternotomy, n = 37 (50.0%), in comparison with the unexpected prolapse group, n = 2 (11.8%). All other patients were operated on through a minimally invasive approach (endoscopically port-access incision). More concomitant surgical procedures were performed in the reference group in comparison with the unexpected prolapse group (20 vs 2 patients). Aortic valve replacement, rhythm surgery and aortic surgery were performed only in the reference group. The median number of prolapsing segments and additionally the median number of artificial neochordae were higher in the unexpected prolapse group in comparison with the reference group 2 (2) vs 1 (0) (P < 0.01) and 6 (1) vs 4 (0) (P < 0.01). Nearly all patients (96.7%) received a semi-rigid annuloplasty ring with no difference in ring size between the groups. The repair rate was 100% for both groups. The surgical procedural details are depicted in Table 2.

Pre- and postoperative echocardiography

Mean preoperative left ventricular ejection fraction was 55.9 ± 9.1 in the unexpected prolapse group and 57.9 ± 8.6 (P = 0.40) in the reference group. No significant baseline differences were found in left ventricular end-diastolic diameter, left ventricular end-systolic diameter and left atrial diameter. Pre- and postoperative echocardiographic parameters are summarized in Table 3. Echocardiography with MR grading at discharge was performed in all patients. Predischarge MR grading showed mild or less MR in all the patients in the unexpected prolapse group (n = 17) in comparison with 98.6% (n = 54) of patients in the reference group, with no significant difference between the groups. Only one patient in the reference group had mild-to-moderate residual MR. At a mean follow-up of 6.6 months, all patients in the unexpected anterior leaflet group showed mild (n = 1) or less MR (n = 16).

Multivariable logistic regression analysis

A binomial logistic regression analysis was performed to ascertain the effects of BMI, surgical approach, number of prolapsing segments, left ventricular ejection fraction, left ventricular end-systolic diameter, left ventricular end-diastolic diameter and left atrial diameter on the likelihood that patients show an unexpected prolapse during intraoperative saline testing. The logistic regression model was statistically significant, χ²(4) = 37.078, P < 0.001. The model explained 54.1% (Nagelkerke R2) of the variance in unexpected anterior leaflet prolapse. Sensitivity was 47.1%, specificity was 95.9%, positive predictive value was 72.7%, and negative predictive value was 88.8%. Of the 7 predictor variables, 6 were statistically significant (Table 4).

DISCUSSION

The phenomenon of unexpected anterior leaflet prolapse in saline testing was found in 18.7% of the isolated posterior leaflet cases. Baseline characteristics demonstrated an increased BMI in the unexpected prolapse group. Binomial logistic regression analysis showed BMI, surgical approach, number of prolapsing segments, left ventricular ejection fraction, left ventricular end-systolic diameter and left atrial diameter to be predictive for unexpected anterior leaflet prolapse. In all the patients, a posterior leaflet repair was performed without correction of the anterior leaflet, with no differences in postoperative residual MR and with excellent surgical results.

A possible explanation for the occurrence of anterior leaflet prolapse could be displacement of the arrested heart with a shift of the ventricle wall and its papillary muscles towards the annulus, resulting in a prolapse of the anterior leaflet during saline testing. External factors contributing to the displacement of the heart could originate from anatomical characteristics. Habitus, diaphragm elevation, aortic elongation and thorax deformities may all dislocate the heart in an arrested state. We found an increased BMI in patients with unexpected anterior leaflet prolapse. Increased abdominal pressure, similar to that in overweight patients, could contribute to an elevation of the diaphragm in sedated patients with subsequent displacement of the arrested heart. Furthermore, the extensiveness of the disease and the severity of the regurgitating volume, combined with relatively preserved left ventricular dimensions and a slightly reduced ejection fraction, could cause geometric remodelling of the subvalvular apparatus with a shift of the papillary muscles towards the annulus in the arrested heart, resulting in an unexpected anterior leaflet prolapse during saline testing. Additionally, in patients with a prolapse of multiple segments of the posterior leaflet, papillary muscle dysfunction will play a certain role. As papillary muscle functions, in general, seem to follow left ventricular function, patients with a reduced ejection fraction will be more prone to unexpected anterior leaflet prolapse during saline testing [18].

Furthermore, mechanical displacement of the heart could contribute to a shift of the ventricle wall and its papillary muscles. We observed an association between the surgical approach and the incidence of unexpected anterior leaflet prolapse during saline testing in posterior leaflet prolapse. A potential underlying mechanism could be the use of a dedicated atrial retractor in minimally invasive surgery, whereby the atrium is lifted vertically, whereas in sternotomy patients, the atrium is lifted vertically and to the left with rotation of the heart.

In ischaemic MR, ‘pseudoprolapse’ is described as a factor contributing to early recurrence resulting from the presence of an untreated prolapse of the anterior leaflet [19]. Pseudoprolapse is characterized as a functional prolapse of the anterior leaflet with a prolapse limited to the level of the MV annulus, without deterioration of the leaflets. Pre- and postoperative echocardiography revealed no pseudoprolapse in our cases. In contrast to pseudoprolapse, a full prolapse of the anterior leaflet above the line of the MV annulus was observed during saline testing. On the basis of the literature, we additionally hypothesize that the pressure in saline testing could be higher than in the physiological situation and could overstretch the papillary muscles and chordae tendineae, contributing to a prolapse of the anterior leaflet [20].

An unexpected prolapse of the anterior leaflet could lead to technically more challenging repairs and could eventually result in an inferior surgical outcome with prolonged surgical times. An unexpected prolapse forces the surgeon to perform time-consuming additional manoeuvres. Repositioning of the atrial retractor, pulling down the diaphragm or forcing down the anterior part of the annuloplasty ring will sometimes resolve the anterior leaflet prolapse. In case the prolapse is persistent, surgeons have to determine the optimal length and position of the neochordae based on caliper measurements and their personal experience. However, for less-experienced surgeons, the intraoperative finding of a prolapse of the anterior leaflet could change their operative strategy and eventually lead to repair of the anterior leaflet. This, in turn, could result in a suboptimal repair, prolonged operating times and potentially severe postoperative MR due to overcorrection, which could eventually lead to valve replacement.

Our postoperative echocardiographic observations at discharge showed no significant difference in MR grading between the unexpected prolapse group and the reference group, which therefore, support the strategy of ignoring a prolapse of the anterior leaflet during saline testing that was not present on preoperative TTE or TOE. Only one patient in the unexpected prolapse group showed mild (grade 1) MV regurgitation with a small eccentric jet, of which the severity decreased at last follow-up. At last follow-up, all the patients are in good clinical condition without the recurrence of MR grade 2+.

Limitations

We recognize the relatively small sample size and limited follow-up of the current study. Although a larger sample size would enable sufficient statistical power to detect preoperative predictors for unexpected anterior leaflet prolapse, our postoperative results are less likely to be influenced by the potentially underpowered number of patients as no residual MR grade 2+ was found at discharge. Our limited follow-up showed no recurrence of MR, and none of the patients underwent reoperation for failure of MVr. Nevertheless, the retrospective design with its accompanying disadvantages limits the generalizability of our cohort.

When surgeons encounter an unexpected anterior leaflet prolapse during intraoperative saline testing, preoperative echocardiography provides a definite answer regarding anterior leaflet involvement. As operative strategy is dependent on the quality of preoperative echocardiography and subsequently determined by the experience and expertise of the echocardiographist, it plays an important role in decision making and the postoperative outcome. Nowadays, MV surgery is performed in experienced centres with dedicated teams. In our centre, in most of the cases, preoperative TOE is performed by the same imaging cardiologist as the intraoperative echocardiogram directly after weaning from bypass, thereby reducing interobserver variability.

Furthermore, the experience of the surgeon is crucial to recognize an unexpected anterior leaflet prolapse and thus perform a repair of only the posterior leaflet. In the current study, a single experienced surgeon performed the MVrs, which may limit the generalizability of our results.

CONCLUSION

In MVr for posterior leaflet prolapse, intraoperative saline testing for valve competency sometimes shows an unexpected full prolapse of the anterior leaflet, which was not present on the preoperative echocardiogram. The underlying mechanism could be displacement of the arrested heart, resulting in a prolapse of the anterior leaflet during saline testing. When saline testing shows a prolapse of the anterior leaflet, although it was absent on the preoperative echocardiogram, no additional surgical techniques should be performed for the anterior leaflet in order to achieve excellent surgical results. Further research is warranted to predict unexpected anterior leaflet prolapse preoperatively.

Conflict of interest: none declared.

REFERENCES