Prognostic impact of 1-year permanent pacemaker implantation after mitral valve surgery with the Cox-maze procedure

Abstract

OBJECTIVES

This study aimed to evaluate the prognostic impact of permanent pacemaker (PPM) implantation within the first year after mitral valve (MV) surgery combined with the Cox-maze procedure, focusing on long-term outcomes, including overall mortality, infective endocarditis (IE) and ischaemic stroke.

METHODS

We conducted a retrospective cohort study using data from the National Health Insurance Service (NHIS) in South Korea, identifying 10 127 patients who underwent MV surgery with the Cox-maze procedure between 2005 and 2020. Patients were classified into the PPM and non-PPM groups based on PPM implantation within 1 year postoperatively. The primary outcome was overall mortality, and secondary outcomes included risk factors for overall mortality, IE and ischaemic stroke. Multivariable Cox proportional hazards regression and Fine-Gray competing risk models were utilized for statistical analysis.

RESULTS

Of the total cohort, 178 patients (1.76%) underwent PPM implantation. The overall mortality during the follow-up period was 20.5%, with no significant difference between the PPM and non-PPM groups. PPM implantation was not a significant risk factor for overall mortality (hazard ratio [HR], 0.825; 95% confidence interval [CI] 0.598–1.140; P = 0.244) or ischaemic stroke. However, PPM implantation was associated with a significantly increased risk of IE (HR, 2.015; 95% CI 1.179–3.442; P = 0.010).

CONCLUSIONS

PPM implantation within the first year after MV surgery with the Cox-maze procedure does not significantly impact long-term mortality or ischaemic stroke risk but is associated with an increased risk of IE. The Cox-maze procedure remains advisable for patients with atrial fibrillation undergoing MV surgery.

Open in new tabDownload slide

INTRODUCTION

The Cox-maze procedure, an established surgical approach for treating atrial fibrillation (AF), is commonly performed in conjunction with mitral valve (MV) surgery for patients with both conditions [1]. The Cox-maze procedure aims to restore sinus rhythm by creating scar tissue in the atria, which blocks abnormal electrical signals [2]. Numerous studies have demonstrated that the Cox-maze procedure, when combined with valve surgery, results in significant reductions in AF recurrence and leads to better long-term clinical outcomes [1, 3–7].

However, despite these benefits, the Cox-maze procedure is associated with certain risks, particularly the increased likelihood of permanent pacemaker (PPM) implantation [8–10]. This risk arises from the possibility of sinus node dysfunction or atrioventricular (AV) block following the procedure, with postoperative PPM implantation rates ranging between 3% and 20% [8, 11–13]. The variability in these rates is influenced by factors such as patient comorbidities, surgical technique and the complexity of the underlying heart condition. The necessity of PPM implantation raises concerns about its potential impact on long-term prognosis, as some studies have suggested that PPM insertion after cardiac surgery, including the Cox-maze procedure and MV surgery, could be associated with adverse clinical outcomes.

Given the controversy surrounding the long-term outcomes associated with PPM insertion, it is crucial to further investigate whether PPM implantation significantly affects patient prognosis following the Cox-maze procedure. This study aims to evaluate the impact of PPM implantation within the first year after MV surgery with the Cox-maze procedure, focusing on long-term outcomes, including overall mortality, infective endocarditis (IE) and ischaemic stroke, using a nationwide population-based dataset.

MATERIALS AND METHODS

Ethical statement

The project received approval from the Institutional Review Board (IRB) of Korea University Anam Hospital (Seoul, Republic of Korea), which waived the need for informed consent to use de-identified patient medical data in this retrospective observational study (IRB No. 2019AN0505).

Data sources

The health claims data were obtained from the National Health Insurance Service (NHIS) in South Korea (NHIS-2023-1-529) and analysed. The NHIS comprises a nationwide health insurance programme covering 96–97% of the 50 million population in South Korea [14, 15]. The NHIS database contains information on all insurance claims, including demographic details, diagnoses coded by the Tenth Revision of the International Classification of Diseases (ICD-10), procedures and mortality records.

Study population

A total of 10 177 patients (aged ≥18 years) undergoing MV surgery with the Cox-maze procedure between January 2005 and December 2020 were identified from the NHIS database using country-specific procedure codes (see Supplementary Material, Table S1). We excluded 50 patients who had PPM implantation before the index date, resulting in a final study cohort of 10 127 participants (Fig. 1). The index date was defined as the date of MV surgery with the Cox-maze procedure within the study period. During this time, the standard or modified Cox-maze III procedure was performed at institutions in South Korea, with the surgical techniques detailed in the existing literature [16–18].

Definitions and study outcomes

Patients who underwent PPM implantation within 1 year after MV surgery with the Cox-maze procedure were classified into the PPM group.

Comorbidities and previous treatments were included as covariates (see Supplementary Material, Table S1). Comorbidities were defined as the presence of two or more identical diagnostic codes within 3 years before the index date, using ICD-10 codes. Previous treatments were defined as having undergone one or more identical procedure codes before the index date, based on country-specific codes.

The primary outcome of this study was overall mortality. The secondary outcomes included risk factors of overall mortality, IE and stroke. Follow-up was initiated on the index date and continued until the occurrence of the outcomes, including death, or until 31 December 2021 (last day of the database), whichever came first.

Statistical analysis

Continuous variables were presented as medians with interquartile ranges (IQRs) when assessed as skewed using the Kolmogorov–Smirnov test. Categorical variables were expressed as frequencies and percentages using the chi-square test or Fisher’s exact test. The cumulative incidence function was used to statistically describe the survival data, and 95% confidence intervals (CIs) were calculated without adjustment for multiple comparisons. Cox proportional hazards regression models were used to calculate hazard ratios (HRs) and 95% CIs to identify risk factors for clinical outcomes. The multivariable model was adjusted for all covariates, including PPM implantation, age, sex, income quartile, comorbidities (e.g. hypertension, diabetes mellitus, dyslipidaemia, persistent AF, etc.), pure mitral regurgitation, IE, CHA₂DS₂-VASc score, Charlson comorbidity index (CCI) score, type of MV procedure and any concomitant procedures. The Fine-Gray proportional subdistribution hazards model was utilized to account for competing risks when evaluating the association between death and other clinical outcomes, including IE and ischaemic stroke. Statistical significance was determined using two-sided tests, with a P-value of <0.05 indicating significance. All statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA).

RESULTS

Baseline characteristics of study population

A total of 10 127 patients were enrolled in the study, of which 178 patients (1.76%) were classified into the PPM group. The baseline characteristics of the study population are summarized in Table 1. The median age of the cohort was 61.0 years (IQR, 52.0–69.0), with the PPM group being significantly older than the non-PPM group (65.0 years [IQR, 58.0–70.0] vs 61.0 years [IQR, 52.0–68.0]; P < 0.001). The proportion of female was notably higher in the PPM groups (64.6% vs 56.8%; P = 0.039). Hypertension, congestive heart failure, dyslipidaemia and diabetes mellitus were the most common comorbidities in the PPM and non-PPM groups. Duration of AF before the index date was significantly longer in the PPM group than in the non-PPM group (4.5 years [IQR, 0.4–10.1] vs 1.4 years [IQR, 0.1–5.8]; P < 0.001).

Operative procedures

In the overall cohort, 3169 patients (31.3%) underwent MV repair (Table 1). Concomitant valve procedures performed alongside MV surgery and the Cox-maze procedure were significantly more common in the PPM group compared to the non-PPM group (79.8% vs 65.4%; P < 0.001). Furthermore, coronary artery bypass grafting was significantly more frequent in the PPM group (11.8% vs 5.2%; P = 0.001).

Overall clinical outcomes

The overall mortality rate in the study population during the follow-up period was 20.5% (Table 2). Cumulative incidence analysis for long-term mortality following MV surgery with the Cox-maze procedure revealed no statistically significant difference between the PPM and non-PPM groups (Gray’s test P = 0.330; Fig. 2). The median hospital stay was 17.0 (IQR, 12.0–24.0) days, and the median follow-up duration was 6.5 (IQR, 3.3–11.1) years (Table 2).

The overall incidence rate of postoperative IE in the entire cohort during the follow-up period was 4.7% (Table 2). The PPM group had a significantly higher incidence of IE compared to the non-PPM group (8.4% in the PPM group vs 4.6% in the non-PPM group; P = 0.029). The overall incidence rate of postoperative ischaemic stroke in the total cohort was 10.6% during the follow-up period, with no significant difference observed between the PPM and non-PPM groups (P = 0.084).

Risk factors for overall mortality

Multivariable analyses were performed to identify risk factors associated with overall mortality, as presented in Fig. 3A and Supplementary Material, Table S2. PPM implantation was not found to be a significant risk factor for overall mortality after MV surgery with the Cox-maze procedure (HR, 0.825; 95% CI 0.598–1.140; P = 0.244). Similarly, persistent AF was not a significant risk factor for overall mortality (HR, 0.959; 95% CI 0.757–1.214; P = 0.729). Age (HR, 1.060; 95% CI 1.052–1.069; P < 0.001), CHA₂DS₂-VASc score (HR, 1.207; 95% CI 1.088–1.340; P = 0.004) and CCI score (HR, 1.086; 95% CI 1.064–1.110; P < 0.001) were identified as independent predictors of overall mortality. The competing risk analysis for overall mortality yielded results consistent with those of the Cox proportional hazards regression models (see Supplementary Material, Table S3).

Risk factors for postoperative IE and ischaemic stroke

Multivariable analyses were performed to identify independent predictors of postoperative IE (Fig. 3B and Supplementary Material, Table S4). PPM implantation was found to be a significant risk factor for postoperative IE (HR, 2.015; 95% CI 1.197–3.392; P = 0.008). Persistent AF was also a significant risk factor for IE (HR, 2.320; 95% CI 1.423–3.784; P = 0.007). Additionally, age (HR, 1.030; 95% CI 1.015–1.044; P < 0.001) and CHA₂DS₂-VASc score (HR, 1.266; 95% CI 1.024–1.567; P = 0.030) were identified as independent predictors of IE. The competing risk analysis for IE yielded results consistent with those of the Cox proportional hazards regression models (see Supplementary Material, Table S5).

Further multivariable analyses were conducted to identify risk factors of postoperative ischaemic stroke (Fig. 3C and Supplementary Material, Table S6). PPM implantation was not significantly associated with postoperative ischaemic stroke (HR, 1.418; 95% CI 0.957–2.102; P = 0.082). In contrast, persistent AF was a significant risk factor for ischaemic stroke (HR, 2.440; 95% CI 1.779–3.346; P < 0.001). Age (HR, 1.039; 95% CI 1.028–1.049; P < 0.001) and CCI score (HR, 1.035; 95% CI 1.002–1.068; P = 0.036) were also found to be risk factors for ischaemic stroke. The competing risk analysis for ischaemic stroke provided results consistent with those of the Cox proportional hazards regression models (see Supplementary Material, Table S7).

Risk factors for PPM implantation

The independent risk factors for postoperative newly PPM implantation in the overall cohort were analysed and are presented in Supplementary Material, Table S8. Female sex was identified as a significant risk factor for PPM implantation (HR, 1.048; 95% CI 1.019–1.077; P = 0.009). Persistent AF was also a significant risk factor for IE (HR, 2.382; 95% CI 1.299–4.368; P = 0.005).

DISCUSSION

Our analysis, based on a large administrative dataset with comprehensive follow-up, revealed several important findings regarding the long-term impact of PPM implantation within the first year after MV surgery combined with the Cox-maze procedure. First and foremost, we found that PPM implantation within the first year postoperatively was an uncommon event, with a relatively low incidence of implantation in the studied population. Importantly, we did not observe an association between the PPM implantation and increased long-term mortality or ischaemic stroke risk when compared to patients who did not require PPM. However, a significant finding was the increased risk of IE in patients who underwent PPM implantation. The impact of the PPM implantation was benign prognosis in our study.

The clinical significance of PPM insertion following the Cox-maze procedure remains a topic of ongoing debate, particularly regarding its impact on long-term mortality and morbidity. Some previous studies have reported that the concomitant Cox-maze procedure to MV surgery increases the likelihood of PPM implantation, and that PPM insertion may negatively impact survival [8–10, 19]. A notable randomized trial investigating the effect of the Cox-maze procedure in patients undergoing MV surgery found a higher risk of PPM implantation, and in some cases, an associated increase in mortality risk [8]. However, this trial had notable limitations, including a relatively small sample size and a short follow-up period of just 1 year, which could have skewed the findings towards more adverse outcomes. Short-term follow-up studies often capture transient complications such as early conduction disturbances, which may resolve over time and reduce the need for permanent pacing, thus potentially overestimating the negative impact of PPM implantation [20–22].

Conversely, other studies, including meta-analyses, have found no significant association between PPM implantation and long-term adverse outcomes [10, 21, 23]. This disparity in findings may be partially explained by the timing of PPM implantation. Early PPM implantation (within the first 30 days after surgery) is often driven by perioperative factors such as intraoperative trauma to the conduction system or sinus node dysfunction. In these cases, the need for a pacemaker may be transient, and patients may not remain pacemaker-dependent in the long term [22]. On the other hand, late PPM implantation (within the first year) is more likely to be associated with comorbid conditions or progressive conduction disease, which may carry different implications for prognosis [22]. Our study’s longer follow-up period allowed for a more nuanced understanding of these distinctions, and our findings suggest that, overall, PPM implantation within the first year does not significantly increase the risk of hard clinical outcomes such as mortality or ischaemic stroke.

One of the most clinically relevant findings of our study was the increased risk of IE in patients who received a PPM. This is consistent with prior research indicating that PPM implantation, particularly in patients who have undergone MV surgery or other valve-related procedures, increases the risk of device-related infections [24]. The presence of foreign material, such as the pacemaker leads and generator, provides a potential nidus for bacterial colonization, particularly in patients with underlying valve disease [25]. IE in the context of pacemaker devices is a serious complication associated with high morbidity and mortality [26]. Given this elevated risk, clinicians should remain vigilant in monitoring for signs of infection in patients with pacemakers, and prophylactic strategies may need to be considered in high-risk populations.

Strengths and limitations of this study

Our study has several strengths, including the use of a nationwide administrative dataset from South Korea’s national health insurance system, which covers the entire population. This allowed for comprehensive follow-up of all medical conditions, including death, ischaemic stroke and IE, providing a real-world view of clinical outcomes after the Cox-maze procedure and PPM implantation. The large sample size and complete follow-up make our findings highly generalizable to similar populations and clinical settings. Additionally, we were able to adjust for a wide range of baseline characteristics and perform rigorous statistical analyses to minimize confounding and bias.

However, our study also has limitations. One major limitation is the lack of detailed information about the surgical procedure itself, including the specific Maze lesion set used, the type of energy source (cryoprobe or radiofrequency), whether left atrial appendage exclusion was performed, and other operative details such as cardiopulmonary bypass time and aortic cross-clamp time. These procedural factors are known to influence postoperative outcomes and the risk of complications such as PPM implantation. Despite adjusting for baseline characteristics, the absence of this detailed procedural data may have introduced some residual confounding. Furthermore, due to the nature of the NHIS database, which is based on diagnosis and procedure codes, it was not possible to obtain information on anticoagulation therapy or the types of implanted PPMs. Consequently, these aspects could not be included in our analysis.

CONCLUSION

PPM implantation within the first year after the Cox-maze procedure combined with MV surgery does not appear to increase the risk of long-term mortality or ischaemic stroke, but it is associated with a higher incidence of IE. These findings suggest that while PPM implantation has a relatively benign impact on long-term survival and stroke risk, careful attention should be paid to the risk of device-related infections, particularly in high-risk patients. The Cox-maze procedure remains advisable for patients with AF undergoing MV surgery.

SUPPLEMENTARY MATERIAL

Supplementary material is available at EJCTS online.

FUNDING

This work was supported by the Basic Science Research Program through the National Research Foundation of Korea funded by the Ministry of Education (No. 2021R1I1A1A01048695). This work was also supported by the National Research Foundation graft funded by the Korea Government (MSIT) (No. RS-2023-00247757).

Conflict of interest: None declared.

DATA AVAILABILITY

All relevant data are within the manuscript and its Supporting Information files.

ACKNOWLEDGEMENTS

This study utilized the National Health Insurance Service (NHIS) database (NHIS-2023-1-529). The interpretations and conclusions reported here do not represent those of the NHIS.

Author contributions

Jun Ho Lee: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Validation; Visualization. Yun Jin Kim: Data curation; Formal analysis; Investigation; Methodology; Project administration; Validation. Ji Eon Kim: Data curation; Formal analysis; Investigation; Writing—review & editing. Kyungsub Song: Data curation; Formal analysis; Investigation; Writing—review & editing. Yonghoon Shin: Data curation; Formal analysis; Investigation; Writing—review & editing. Jae Seung Jung: Data curation; Formal analysis; Investigation; Writing—review & editing. Ho Sung Son: Data curation; Formal analysis; Investigation; Writing—review & editing. Seung Hyun Lee: Data curation; Formal analysis; Investigation; Writing—review & editing. Hee Jung Kim: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Supervision; Validation; Visualization; Writing—original draft; Writing—review & editing

Reviewer information

European Journal of Cardio-Thoracic Surgery thanks Manuel J. Antunes, Miha Antonic, Philipp Angleitner and the other anonymous reviewers for their contribution to the peer review process of this article.

Presented at the 38th EACTS Annual Meeting, Lisbon, Portugal, 10 October 2024.

REFERENCES

ABBREVIATIONS

ABBREVIATIONS
 
• AF

  Atrial fibrillation

 
• AV

  Atrioventricular

 
• CCI

  Charlson comorbidity index

 
• CI

  Confidence interval

 
• HR

  Hazard ratio

 
• ICD-10

  Tenth Revision of the International Classification of Diseases

 
• IE

  Infective endocarditis

 
• IQR

  Interquartile range

 
• IRB

  Institutional Review Board

 
• MV

  Mitral valve

 
• NHIS

  National Health Insurance Service

 
• PPM

  Permanent pacemaker

Author notes