Bringing doctors together

or

A free, private network for verified physicians

Leading voices

Dr. Brahmajee Nallamothu

Interventional Cardiology · UMHS

View

Dr. Mohamad Adnan Alkhouli

Interventional Cardiology · Mayo Clinic College of Medicine

View

Dr. Ajay Kirtane

Interventional Cardiology · CUIMC

View

Dr. Manesh Patel

Interventional Cardiology · Duke University

View

Dr. Deepak Bhatt

Interventional Cardiology · Mount Sinai

View

Dr. Sunil Rao

Interventional Cardiology · NYU Langone

View

Dr. Alexander Fanaroff

Interventional Cardiology · UPenn

View

Dr. James Tcheng

Interventional Cardiology · Duke

View

Dr. Pinak Shah

Interventional Cardiology · PHS

View

Dr. Dorairaj Prabhakaran

General Cardiology

View
  • Discover your community

  • Stay current

  • A profile that represents you

  • Find your people

About RoonPrivacyTerms

    Overview

    Incidence, Risk Factors, and Outcomes Associated With Permanent Pacemaker Implantation Following Tricuspid Valve Surgery.

    Jan 31, 2024DOI: 10.1161/jaha.123.032760Journal ArticleAHA

    Written by: Joseph KassabSerge C HarbMilind Y DesaiA Marc GillinovHabib LayounJoseph El DahdahMichel Chedid El HelouShady NakhlaHaytham ElgharablySamir R KapadiaPaul C CremerAmgad Mentias

    Find these authors on Roon

    JK

    Dr. Joseph Kassab

    Interventional Cardiology · UT Southwestern

    MD

    Dr. Milind Desai

    Advanced Cardiac Imaging · Cleveland Clinic

    Methods

    Data used for this study are covered under data use agreement between the Cleveland Clinic and the Centers for Medicare and Medicaid Services. The data are not available for distribution by the authors but may be obtained from the Centers for Medicare and Medicaid Services with an approved data use agreement.Medicare patients who underwent TVS (repair or replacement) from January 2013 to December 2020 were identified from Medicare Provider Analysis and Review 100% inpatient files using International Classification of Diseases (ICD) procedure codes (35.14, 35.27, 35.28, 02QJ0, 02RJ0). Surgeries included in the study were isolated tricuspid valve (TV) repair, isolated TV replacement, or concomitant TV repair or replacement with different combinations of surgical aortic valve replacement, mitral valve repair or replacement, and coronary artery bypass graft. The primary exposure of interest for the study was new pacemaker placement after TVS and before discharge, or within 30 days from discharge date. Patients who had Medicare Fee‐for‐Service coverage for <1 year before the TVS, patients who had pacemaker or implantable cardioverter‐defibrillator before TVS and patients who underwent TVS for endocarditis were excluded. Patients' demographics including age, sex, race, and dates of enrollment were extracted from Medicare Beneficiary Summary Files. A period of lookback of 1 year was used and all ICD codes submitted in claims during that period were used to ascertain patients' comorbidities using Elixhauser's method.6, 7 ICD diagnoses on the index TVS claim that were marked as present on admission were also used to determine patients' comorbidities.The primary study outcome was all‐cause mortality in follow‐up in patients who underwent pacemaker placement after TVS versus patients who did not require pacemaker placement. The study secondary outcomes included 1‐year mortality and the risk of hospital readmission with a primary diagnosis of endocarditis or HF in follow‐up. Data regarding mortality were available through August 2021 and data regarding hospital readmissions were available through December 2020. The Institutional Review Board at the Cleveland Clinic approved the study with waiver of informed consent due to the retrospective nature of the study.Categorical variables are reported as frequencies and percentages and compared using chi‐square test. Numeric variables are reported as mean and SD and compared using Student t test if normally distributed or reported as median and interquartile values and compared using Mann–Whitney U test if not normally distributed. Propensity score matching analysis was used to adjust for differences between the 2 groups. In this analysis, patients from both groups (PPM versus no PPM) were matched on exact age, sex, and type of surgery and matched on tight caliper on all other comorbidities. Importantly, thorough propensity matching was performed with special emphasis on congestive HF, because a diagnosis of HF in follow‐up was a primary outcome of the study. Each patient from the pacemaker group was matched to up to 3 controls from the no pacemaker group. Robustness of the matching algorithm was examined by calculating standardized differences between the 2 groups and matching was considered successful and well balanced if all differences were <0.1. For the primary outcome of the study, a Cox regression analysis was done with strata of match ID and Breslow method to break ties to account for the matching design,8 with PPM placement as a time‐dependent covariate. For the secondary outcomes of endocarditis and HF in follow‐up, sub‐distribution hazards Fine–Gray models were used, accounting for the competing risk of death.9 Kaplan–Meier curves were created for the primary and secondary outcomes as a function of events over time and were compared using log‐rank test. For the outcome of 1‐year mortality, a conditional logistic regression was used and asymptotic odds ratios (ORs) were estimated.10 Multivariable Cox regression analysis was performed as a sensitivity analysis, in which robust variance estimator was used to account for hospital clustering. To study variables associated with PPM placement after TVS, we analyzed the entire cohort, including all potential confounders, using 2 mixed‐level relative risk regression models with a random intercept for the hospital. The first model encompassed patient age, sex, and comorbidities, and the second model included type of surgery in addition to variables in the first model. Analysis was conducted using SAS 9.4 and R 4.0.2.

    Results

    The final study cohort included 13 294 patients who underwent any TVS, and 2518 (18.9%) patients required PPM implantation after surgery compared with 10 776 (81.1%) patients who did not require a PPM. Figure 1 shows the flow chart for the study cohort. Table 1 shows baseline characteristics of the 2 study groups before propensity score matching. There was no difference in age or race between the 2 groups; however, patients who underwent PPM placement were more likely to be women (64.4% versus 57.6%, P<0.001). Patients who underwent PPM placement also had a higher prevalence of prior sternotomy, chronic kidney disease, HF, preoperative atrial fibrillation, preoperative second‐degree heart block, right bundle‐branch block, and bifascicular block. In addition, patients who underwent PPM had a significantly higher prevalence of isolated TV replacement (P<0.001) or TV replacement with concomitant surgeries, compared with patients who had TV repair. After propensity score matching, there were 2267 patients in the PPM group and 5486 patients in the no PPM group. By study design, the 2 groups had identical age, sex, and types of surgeries and were well balanced in all other variables.Flow chart for study cohort derivation. Propensity score matching analysis was used to adjust for differences between the 2 groups. Patients who underwent TV surgery for endocarditis were excluded. In this analysis, patients from both groups (patients who received a pacemaker after TV surgery vs patients who did not receive a pacemaker) were matched on exact age, sex, and type of surgery and matched on tight caliper on all other comorbidities. Each patient from the pacemaker group was matched to up to 3 controls from the no‐pacemaker group. FFS indicates fee‐for‐service; PPM, permanent pacemaker; and TV, tricuspid valve.Baseline Characteristics of the 2 Study Groups Before and After Propensity Score MatchingCABG indicates coronary artery bypass graft; MV, mitral valve; PPM, permanent pacemaker; SAVR, surgical aortic valve replacement; and TV, tricuspid valve.P values for categorical variables were calculated using chi‐square test, and P values for numeric variables were calculated using Student t test if normally distributed or Mann–Whitney U test if not normally distributed.After a median follow‐up of 3.1 years (Q1, 1 year; Q3, 5.5 years), there was no difference in the risk of mortality in patients who received PPM compared with patients who did not receive PPM after TVS (HR, 1.02 [95% CI, 0.93–1.12], P=0.7; Figure 2). PPM placement was not associated with 1‐year mortality (18.9% versus 20.7%, OR, 0.90 [95% CI, 0.79–1.03], P=0.1) and was not associated with increased risk of endocarditis during follow‐up (HR, 1.03 [95% CI, 0.74–1.45], P=0.8) (Figure 3A). However, PPM placement was associated with longer hospital stay (median, 12 days [interquartile range, 8–19 days] versus 10 [7–17 days], P<0.001) and higher risk of HF readmission in follow‐up (HR, 1.28 [95% CI, 1.14–1.43], P<0.001) (Figure 3B). Moreover, when analysis was restricted to patients without history of HF before surgery, new PPM was still associated with higher risk of HF in follow‐up (HR, 1.24 [95% CI, 1.00–1.55], P=0.05).Kaplan–Meier curves for all‐cause mortality in patients who require permanent pacemaker placement after tricuspid valve surgery vs patients who do not require permanent pacemaker placement after propensity matching. Propensity score matching analysis was used to adjust for differences between the 2 groups. In this analysis, patients from both groups (patients who received a pacemaker after TV surgery vs patients who did not receive a pacemaker) were matched on exact age, sex, and type of surgery and matched on tight caliper on all other comorbidities. Each patient from the pacemaker group was matched to up to 3 controls from the no‐pacemaker group. After a median follow‐up of 3.1 years, the risk of all‐cause death was similar between the 2 groups (hazard ratio, 1.02 [95% CI, 0.93–1.12], P=0.7) TV indicates tricuspid valve.Kaplan–Meier curves for endocarditis (A) and heart failure readmission (B) in patients who require permanent pacemaker placement after tricuspid valve surgery vs patients who do not require permanent pacemaker placement after propensity matching. Propensity score matching analysis was used to adjust for differences between the 2 groups. In this analysis, patients from both groups (patients who received a pacemaker after TV surgery vs patients who did not receive a pacemaker) were matched on exact age, sex, and type of surgery and matched on tight caliper on all other comorbidities. Each patient from the pacemaker group was matched to up to 3 controls from the no‐pacemaker group. After a median follow‐up of 3.1 years, the risk of endocarditis was not increased in the pacemaker group (HR, 1.03 [95% CI, 0.74–1.45], P=0.8). The risk of HF rehospitalization was increased in the pacemaker group (HR, 1.28 [95% CI, 1.14–1.43], P<0.001). HF indicates heart failure; HR, hazard ratio; and TV, tricuspid valve.Following sensitivity analysis, the results were consistent with the results from the primary analysis. PPM was not associated with higher mortality (adjusted HR, 0.97 [95% CI, 0.91–1.04], P=0.4) and was not associated with higher risk of endocarditis (adjusted HR, 1.12 [95% CI, 0.87–1.44]). However, PPM was associated with higher risk of HF readmission in follow‐up (adjusted HR, 1.13 [95% CI, 1.04–1.23], P=0.006).In multivariable relative risk regression, female sex, prior sternotomy, history of malignancy, preoperative second‐degree heart block, right bundle‐branch block, and bifascicular block were all independently associated with a higher risk of PPM placement after TVS (Table 2). After accounting for the type of surgery, patients' variables associated with a higher risk of PPM were female sex and second‐degree heart block. In assessing the type of surgery, the highest risk for PPM was associated with TV replacement with concomitant surgeries and isolated TV replacement, compared with TV repair (relative risk [RR], 3.20 [95% CI, 2.16–4.75], P<0.0001 and RR, 2.20 [95% CI, 1.87–2.58], P<0.0001, respectively) (Table 3).Mixed‐Level Relative Risk Regression, With Patient Variables Associated With the Risk of PPM Placement After TV Surgery, With Random Intercept for the HospitalPPM indicates permanent pacemaker; RR, relative risk; and TV, tricuspid valve.Mixed‐Level Relative Risk Regression for Patient and Surgical Risk Factors Associated With the Risk of PPM Placement After TV Surgery With a Random Intercept for the HospitalCABG indicates coronary artery bypass graft; MV, mitral valve, PPM, permanent pacemaker; RR, relative risk; SAVR, surgical aortic valve replacement; and TV, tricuspid valve.

    Discussion

    Several findings from this observational study should be emphasized. First, among a cohort of 13 294 Medicare beneficiaries, 18.9% of patients undergoing TVS required a PPM. To the best of our knowledge, this is the largest descriptive retrospective analysis evaluating the incidence of post‐TVS PPM implantation. The incidence of PPM following isolated TVS appears to be higher than the incidence of PPM following isolated left‐sided heart surgeries. In fact, 27.3% and 12.6% of patients undergoing isolated TV replacement or repair required a PPM respectively (Table 1). However, upon review of the literature, in patients undergoing isolated mitral valve surgery, coronary artery bypass grafting, surgical aortic valve replacement, or transcatheter aortic valve replacement, the rates were 7.75%,11 0.8% to 4%,12, 13 2.7%,14 and 9.98%,15 respectively. Our findings are consistent with a 10‐year (2004–2013) analysis of trends and outcomes in isolated TVS that described an incidence of PPM implantation of 26%.3 Our results are also concordant with a recently published randomized control trial comparing outcomes at 2 years in 401 patients undergoing mitral valve surgery for degenerative mitral regurgitation, with or without tricuspid annuloplasty, in which the incidence of PPM implantation was significantly higher in the group with surgery plus tricuspid annuloplasty than in the surgery‐alone group (14.1% versus 2.5%; rate ratio, 5.75 [95% CI, 2.27–14.60]).16 The reason behind the high risk of PPM implantation as a result of heart block following TVS is primarily anatomic. The atrioventricular node resides in the muscular atrioventricular septum at the apex of the triangle of Koch, which borders the TV. Thus, the former is highly susceptible to injuries during intervention, increasing the risk of postoperative high‐degree atrioventricular block, the most common indication of pacing following TVS.17, 18Second, among the Medicare population, mixed‐level relative risk regression of demographic and risk factors pointed out that female sex, prior sternotomy, preoperative conduction abnormalities (notably second‐degree heart block), and history of tumor were independent predictors of PPM implantation following TVS. Interestingly, a retrospective study of patients who underwent TVS at 2 academic medical centers in a 5‐year period found no significant differences in sex between those who needed a PPM and those who did not (P=0.359). Our study adds novel information in being a much bigger cohort, having more than 50 times the number of patients (13 294 versus 237). It is worth noting that other studies have also identified female sex as an independent risk factor for PPM after cardiac surgeries, especially TVS19, 20, 21 as well as conduction abnormalities.4 Moreover, Mar et al found that cross‐clamp time >60 minutes is also an independent risk factor for PPM implantation in TVS.17 This duration directly reflects cardiac ischemic time, which, if increased, could lead to ischemic injuries of the conduction system and bradyarrhythmias.22Third, the need for PPM differs among the types of TVS. In general, although comparative data are limited, TV repair is preferred over TV replacement, with the latter usually performed when repair is not feasible.23, 24 In addition to being an easier and faster procedure, TV repair is associated with significantly reduced in‐hospital and operative mortalities.23, 24 Nonetheless, similar survival rates at 1, 5, and 10 years were reported for repair and replacement (80%, 72%, and 66% versus 85%, 79%, and 49%, respectively, P=0.66).25 In this setting, we reported that isolated TV replacement was associated with a much higher risk of postprocedural PPM implantation in comparison to TV repair (RR, 2.2 [95% CI, 1.87–2.58], P<0.0001). The aforementioned close relationship between the tricuspid apparatus and the conduction system is an essential clue. The most common TV repair interventions include bicuspidization, suture annuloplasty, and prosthetic annuloplasty band.26 Interestingly, surgical tricuspid annuloplasty with dedicated rings is often incomplete in order to avoid placing stitches in the septal area, which lies next to the atrioventricular node. Valve replacement is not able to avoid stretching that area, explaining higher rates of postprocedural PPM placement.27 There are a few techniques that can be applied during TV replacement to avoid postoperative conduction disturbances, such as passing the valve sutures superficially in the annulus and through the septal leaflet in the area close to the conduction fibers or using a beating heart technique while closely monitoring for any changes in rhythm on the monitor.Fourth, after a median follow‐up of 3.1 years, we found no difference in mortality rates in patients with a postprocedural PPM and patients without one (P=0.7). Jokinen et al followed up 136 patients who underwent TVS for a mean duration of 7.9 years. They found that PPM implantation was significantly associated with increased survival especially early after surgery (P=0.01).20 The reasons for increased survival, especially in the first postoperative month, are unclear. One may hypothesize that PPM prevented fatal bradyarrhythmias, which usually occur early after TVS. In fact, when analyzing a cohort of 505 patients who underwent TVS, Herrmann et al found that although 5.2% of patients needed a PPM for atrioventricular block after 30 days, this number increased to only 5.8% at 1 year.18 We found that PPM implantation was significantly associated with a higher risk of readmission for HF (P<0.001). The increased risk of HF readmission could be linked to the increased rates of TR, which is commonly described after PPM implantation with a frequency ranging from 7% to 45%, according to different studies.28, 29, 30, 31 Delling et al screened 93 592 echocardiograms (1245 with PPM) in 58 556 individual patients (634 with PPM) and found that the prevalence of significant TR (defined as moderate or severe TR) was higher in patients after PPM placement compared with those without a PPM (adjusted OR, 2.32 [95% CI, 1.54–3.49], P<0.0001).32 Mechanisms proposed to explain PPM‐induced TR include lead‐related interference on TV leaflet motion, lead impingement causing reduced leaflet coaptation, lead entanglement in the subvalvular apparatus, leaflet perforation, and leaflet laceration, with an increased risk of scarring and fibrosis.33, 34 Leadless pacemakers theoretically cause less mechanical interference with the TV apparatus.35 However, data on TR after leadless pacemaker implantation are sparse and conflicting. Nonetheless, Vaidya et al compared 90 patients who received leadless pacemakers (Micra and Nanostim implants) 1:1 to an age‐ and sex‐matched control group who received conventional pacemakers. They observed a higher increase in TR severity in patients who received conventional pacemakers compared with patients with leadless pacemakers (P=0.017). The median follow‐up duration of this study was only 62 days, limiting conclusions about the effects of the devices on valve function.36 Further data from randomized controlled trials are needed to support the widespread adoption of these devices in clinical practice and to better understand their impact on TR. Another explanation for increased HF would be an increased risk of left ventricular and right ventricular dyssynchrony following right ventricular apical pacing, which could have detrimental effects on left ventricular function.37, 38 Proposed mechanisms include increased left ventricular remodeling, increased left ventricular strain, and changes in the mechanical activation pattern of the left ventricular, which becomes comparable to a left bundle‐branch block, especially in patients with high right ventricular pacing burden (>20%).38, 39 Auger et al reported that in 169 patients with PPM, ventricular dyssynchrony induced by right ventricular apical pacing was associated with increased hospitalization rates during follow‐up (24% versus 4% at 5 years of follow‐up, log‐rank P<0.001).40 Impaired quality of life and higher rate of thromboembolic events were also described after PPM.20, 41Therefore, careful consideration should be given to the risk of PPM implantations after TVS. Accurate risk stratification, adequate patient counseling, and the development of standardized, externally validated score predictors specific to TVS could be helpful. Koplan et al proposed a validated simple risk score to predict the need for PPM after cardiac valve surgery. The presence of TVS or a right bundle‐branch block are the 2 strongest predictors for PPM implantation.4 However, a risk score specific to tricuspid interventions has yet to be developed. Patients who are at high risk for postoperative PPM after TV intervention, especially TV replacement, should undergo a direct discussion between the surgeon and the electrophysiologist about considerations for cardiac resynchronization therapy using epicardial lead placement during surgery. Finally, it would be interesting to study the outcome of transcatheter TV interventions in terms of conduction abnormalities and PPM need, as studies are lacking.Our study is not without limitations. First, data for our study were derived from an administrative database and ICD coding errors and subsequent misclassifications are possible. Second, we lacked echocardiographic data including ejection fraction, and right ventricular function and size, which could have affected outcomes. Third, because our study cohort is exclusively from Medicare patients, the described results might not be reproducible to the rest of the population. Fourth, despite the matching process on a wide range of confounders, residual unmeasured confounding is possible. However, despite the retrospective design of this study, the very large sample size makes the results relevant and provides crucial data to add to the current body of the literature.

    Conclusions

    In conclusion, postprocedural PPM implantation is a very frequent complication of TVS, especially following TV replacement with or without concomitant left‐sided heart surgeries. Numerous demographic and clinical factors are associated with an increased risk of PPM placement. PPM implantation is associated with higher HF rehospitalization rates and worsening TR. Adequate risk stratification and counseling of patients before surgery decision is essential.

    Sources of Funding

    The current research was partly funded by philanthropic gifts by the Haslam family, Bailey family, and Khouri family to the Cleveland Clinic for Dr Desai's research. The funders had no role in the design or conduct of the study; the collection, management, analyses, or interpretation of the data; the preparation, review, or approval of the manuscript; or the decision to submit the manuscript for publication.

    Disclosures

    None.
    JK

    Dr. Joseph Kassab

    Cardiology · Interventional Cardiology
    The University of Texas Southwestern Medical Center
    josephk