Surgical heart block is a well-described complication of cardiac surgery for CHD. Prior reports estimate that atrioventricular block occurs in 1% – 6% of patients undergoing cardiac surgery. Reference Romer, Tabbutt and Etheridge1,Reference van Geldorp, Vanagt and Vugts2 A recent registry study from Pediatric Cardiac Critical Care Consortium found an incidence of post-operative heart block to be nearly 3%. However, almost 94% of these patients had resolution of atrioventricular block by post-operative day 10, resulting in 1% of patients receiving a permanent pacemaker. Reference Romer, Tabbutt and Etheridge1
Post-surgical atrioventricular block is a complication with significant morbidity and mortality. Reference Romer, Tabbutt and Etheridge1,Reference Anderson, Czosek, Knilans, Meganathan and Heaton3,Reference Murray, Smith, Flack, Crum, Owen and Kannankeril4 Patients requiring permanent pacemaker for surgical heart block have an overall 2.1 times longer length of stay and increased mortality when adjusting for surgical complexity. Reference Anderson, Czosek, Knilans, Meganathan and Heaton3–Reference Liberman, Silver, Chai and Anderson5 Placement of pacemaker in children can be associated with complications including lead fracture, phrenic nerve capture, pacer-induced cardiomyopathy, infectious complications, and coronary artery compression. Reference Wilhelm, Thone, El-Scheich, Livert, Angelico and Osswald6,Reference Mah, Prakash, Porras, Fynn-Thompson, DeWitt and Banka7 Guidelines suggest that a permanent pacemaker be placed for post-surgical heart block after 7 – 10 days as a class-1 indication. Reference Silka, Shah and Silva8 There is evidence that up to 97% of recovery of atrioventricular nodal function occurs within 9 days of surgery with post-surgical heart block following CHD repair. Reference Romer, Tabbutt and Etheridge1,Reference van Geldorp, Vanagt and Vugts2,Reference Weindling, Saul, Gamble, Mayer, Wessel and Walsh9 The registry report by Romer et al suggested that longer cardiopulmonary bypass times and specific operations such as the Nikaidoh, double switch procedure, or ventricular septal defect creation/enlargement were associated with higher risk for needing a permanent pacemaker. Reference Romer, Tabbutt and Etheridge1 There are reports of late recovery of atrioventricular nodal function after surgical heart block, but factors predicting recovery are not well understood. Reference van Geldorp, Vanagt and Vugts2,Reference Batra, Wells, Hinoki, Stanton and Silka10–Reference Bruckheimer, Berul and Kopf12 We sought to describe our centre’s experience with surgical heart block following cardiac surgery for CHD and examine the factors associated with heart block and late recovery of atrioventricular nodal function.
Materials and methods
Cohort selection
This is a single-centre retrospective cohort study that included all patients 0 – 21 years of age who underwent cardiac surgery for palliation or repair of CHD and developed heart block post-operatively between 1 January, 2010 and 31 December, 2019 at Children’s Healthcare of Atlanta, a free-standing, university-affiliated quaternary care children’s hospital. An internal surgical database was queried, and cardiac surgical encounters were obtained, and patients who experienced heart block were identified by ICD-10 code diagnoses. The study was approved by the Children’s Healthcare of Atlanta Institutional Review Board (IRB# 00000398).
Demographics (age, weight, gender, and race), clinical characteristics (cardiac ICU length of stay, post-operative length of stay, hospital length of stay, duration of mechanical ventilation, genetic syndrome), and operative variables (cardiopulmonary bypass time and aortic cross-clamp time) were collected. (Table 2) Patients with post-operative heart block that required ventricular pacing were stratified into 3 main outcome groups: 1) those who had spontaneous recovery of atrioventricular nodal function, 2) those who had permanent pacemaker placed but had no recovery of atrioventricular nodal function, and 3) those who recovered atrioventricular nodal function after permanent pacemaker implantation. Recovery of atrioventricular nodal function was defined as return to sinus rhythm or first-degree heart block. This definition is similar to what has been used in prior reports. Reference Romer, Tabbutt and Etheridge1
Statistical analysis
Normality was tested for all variables of interest using the Shapiro–Wilk test. Descriptive statistics were performed including counts, medians, and interquartile ranges. We used non-parametric methods to determine the difference in medians between continuous variables of the three outcome groups. Exact Fisher and χ2 tests were performed for differences in categorical variables between groups. A p-value of less than 0.05 was considered the cut-off for significance. Statistical analyses were performed using SAS Enterprise Guide version 7.15 (SAS Institute, Inc., NC).
Results
Patient characteristics
A total of 6333 cardiac surgeries for CHD were performed between 1 January, 2010, and 31 December, 2019. Of this group, 2% (128/6333) patients developed post-surgical heart block, and 0.6% (38/6333) of them underwent permanent pacemaker placement. There were 6/38 patients (15.8%) who had recovery of atrioventricular nodal function following permanent pacemaker placement. There was no difference in demographics between the three groups with regard to age, race, weight at the time of index surgery, gender, preoperative need for antiarrhythmic medications, or presence of chromosomal abnormalities (Table 1). For patients who had transient atrioventricular block, 91.2% had recovery of atrioventricular nodal function 10 days after developing heart block and 100% at 19 days (Fig 1). Of the patients with single-ventricle physiology who experienced heart block and received a pacemaker, a greater proportion had atrioventricular nodal recovery than not (50% versus 12.5%, p = 0.016) (Table 1). Given the heterogeneity in cardiac diagnoses with few patients in each category, further statistical analysis was not feasible (Table 2).
Table 1. Patient characteristics
Results depicted in n (percent), median (interquartile range).
† significant difference between Groups B and A p < 0.0001.
Abbreviations: AV = atrioventricular, CICU = cardiac ICU; PPM = permanent pacemaker.
Table 2. Primary cardiac diagnosis
ASD = atrial septal defect; AV canal complete: atrioventricular canal complete; DORV = double outlet right ventricle; PA/VSD = pulmonary atresia/ventricular septal defect; TGA = transposition of the great arteries; TOF = tetralogy of Fallot; VSD = ventricular septal defect.
Figure 1. Kaplan–Meier plot showing time from onset of atrioventricular block to either spontaneous recovery of atrioventricular nodal function (red) or placement of permanent pacemaker (blue).
Surgical characteristics
Median cardiopulmonary bypass time was 121 minutes (interquartile range: 95 – 166) in the transient atrioventricular block group, 144 minutes (interquartile range 120 – 190) in the permanent pacemaker group without recovery, and 138 minutes (interquartile range 111 – 178) in the permanent pacemaker group with later recovery of atrioventricular nodal function (Table 1). The transient heart block group had shorter cardiopulmonary bypass times compared to those who had pacemaker placement (with or without recovery) (p = 0.011). There was no significant difference between the pacemaker patients who had and did not have spontaneous recovery of nodal function (p = 0.963). Median aortic cross-clamp time was 70 minutes (interquartile range 51 – 110) in the transient atrioventricular block group, 91 minutes (interquartile range 72 – 128) in the permanent pacemaker group without recovery, and 75 minutes (interquartile range 62 – 144) in the permanent pacemaker group with later recovery of atrioventricular nodal function (p = 0.059) (Table 1).
Medical characteristics
There was no difference in the rate of post-operative steroid administration between groups (p = 0.445) nor was there a difference in infectious or wound complications (p = 1). Furthermore, there was no difference in long-term mortality between groups (8%, 19%, and 0% respectively, p = 0.191). However, the median cardiac ICU length of stay in the transient atrioventricular block group was 5 days (interquartile range 3 – 10) compared to 15 days (interquartile range 6.0 – 36.5) in those who underwent permanent pacemaker placement with or without recovery of atrioventricular nodal function (p < 0.0001) (Table 1). Various antiarrhythmic medications were used preoperatively (Table 3).
Table 3. Preoperative antiarrhythmic therapies
Recovery of atrioventricular nodal function
In those who had transient heart block, median time to recovery of atrioventricular nodal function was 2.0 days (interquartile range 1 – 6) with a median time to pacemaker placement of 8.0 days (interquartile range 7 – 12). There were 6 patients who had recovery of atrioventricular nodal function after placement of permanent pacemaker. Following the index surgery resulting in heart block, the median duration to pacemaker placement was 5.5 days (interquartile range 4 – 8) in patients who had late atrioventricular nodal function recovery versus 9.0 days (interquartile range 8 – 14) in those who did not have recovery of atrioventricular nodal function (p = 0.006) (Fig 2). Median time from index operation to recovery was 13 days (interquartile range 5 – 117).
Figure 2. Kaplan–Meier plot showing time from onset of AV block to recovery of AV nodal function without pacemaker (red) or recovery after PPM was placed (blue).
In a subgroup analysis of patients stratified by age, those in the children group (aged 1 year to 18 years) who had permanent pacemaker placed without atrioventricular nodal recovery had lower weight (median 13.5 kg, interquartile range 11 – 18.2) compared to those who had permanent pacemaker placed but had recovery of atrioventricular nodal function (median 43.15 kg, interquartile range 28.1 – 58.5) (Table 4). In a subgroup analysis of the infant and adult cohorts (<1 year old and > 18 years old), weight at the time of surgery was not different between the three groups. Complication rates between these groups were similar (p = 1).
Table 4. Patients with permanent pacemaker with and without recovery
Results depicted in n (percent), median (interquartile range).
AV = atrioventricular; PPM = permanent pacemaker.
Discussion
In this study, we demonstrate an incidence of early post-surgical heart block of 2% at our centre, with permanent heart block rate of 0.5%. This aligns with prior reports. Reference Romer, Tabbutt and Etheridge1,Reference Liberman, Silver, Chai and Anderson5 Romer et al in a large registry study evaluating 15,901 surgical hospitalisations found that 2.7% of patients experienced post-surgical heart block. Reference Romer, Tabbutt and Etheridge1 In their study cohort, approximately 1% of patients who underwent cardiac surgery required permanent pacemaker placement compared to our study, where 0.6% of patients underwent permanent pacemaker placement. Of note, our study is more specific to paediatric cardiac surgery as our cohort contained only three adult patients 18 – 21 years of age, compared to 713 in the Romer study. They identified that longer cardiopulmonary bypass time was associated with increased risk for needing permanent pacemaker placement. Our study confirms this finding as well, with the median cardiopulmonary bypass time of 121 minutes in those with transient heart block compared to 144 and 138 minutes in those who had permanent pacemaker placed with recovery of atrioventricular nodal function and those who had permanent pacemaker placed but had no resolution of heart block, respectively. Longer aortic cross-clamp time trended towards but was not significantly associated with increased rate of permanent heart block. The aetiology for this is not well known but may be related to longer atrioventricular node ischaemic time, or cardiopulmonary bypass time may be a surrogate for more difficult operations or ones with more challenging intracardiac visualisation.
We also describe the rate of resolution of post-surgical heart block in the largest group of children following cardiac surgery, to our knowledge. Of 6333 cardiac surgeries, 128 (2%) were complicated by atrioventricular block requiring ventricular pacing, and 38 (0.6%) had a permanent pacemaker placed. Of the patients that underwent pacemaker placement (n = 38), 6/38 (15.8%) had resolution of heart block. van Geldorp et al describe a group of patients with the resolution of atrioventricular block following initial pacemaker placement. They examined 2850 patients who underwent cardiac surgical procedures of whom 59 (2.1%) experienced chronic (>14 days) post-operative heart block. Of this group, 7 patients (7/59, 11.9%) demonstrated resolution of atrioventricular block following initial pacemaker placement. Reference van Geldorp, Vanagt and Vugts2 Notably, that study did not analyse variables that are associated with recovery of atrioventricular nodal function. Our findings demonstrate a slightly higher (6/38, 15.8%) rate of recovery of atrioventricular nodal function following post-operative heart block. While the patients in our study who had recovery of atrioventricular nodal function after permanent pacemaker placement had median placement of pacemaker on post-operative day 5, all but one had recovery 7 days after the development of post-surgical atrioventricular block. Thus, these patients would have still met the current Pediatric and Congenital Electrophysiology Society/Heart Rhythm Society recommendation for permanent pacemaker placement at post-operative day 7 – 10. Reference Silka, Shah and Silva8 Interestingly, had we waited for post-operative day 10 to determine the need for pacemaker implantation, 3 of our 38 patients would not have received pacemakers. Romer et al. suggested that in their large multi-centre cohort, that if atrioventricular nodal function recovers after post-operative heart block, this occurs by post-operative day 10 in 94% of patients. Reference Romer, Tabbutt and Etheridge1 Ours is also the first study to evaluate for factors that might predict recovery of atrioventricular nodal function. A greater proportion of patients with single-ventricle physiology had recovery of atrioventricular nodal function after pacemaker implantation compared to those who had permanent post-operative heart block. The reasons for this are not entirely clear, but could be due to the fact that Glenn and Fontan procedures do not generally require intracardiac surgery.
Of those in the single-ventricle physiology late recovery group, one patient had ventricular septal defect enlargement, but the remaining two patients did not have intracardiac surgery (Kawashima and extracardiac Fontan procedures). This is compared to those single-ventricle physiology patients who required permanent pacing, all of whom underwent intracardiac surgery (Ebstein’s anomaly of tricuspid valve repair, Fontan with atrioventricular valvuloplasty, Glenn with ventricular septal defect enlargement, and Starnes procedures). Oster et al. studied arrhythmias in single-ventricle patients who underwent Norwood operation. They found the only significant factor contributing to heart block in this cohort was a concomitant procedure in addition to the Norwood procedure, with the strongest risk factor being tricuspid valve repair. Reference Oster, Chen and Dagincourt13 Thus, our data suggest that in single-ventricle patients, the need for a concomitant procedure in addition to Glenn or Fontan procedure is associated with developing irrecoverable heart block (in addition to the other previously identified high-risk operations like Ebstein repair and Starnes procedure). Reference Romer, Tabbutt and Etheridge1
This is a single-centre retrospective study with its attendant limitations. We did not study operator as a variable. Pacemaker implantation timing was based on surgeon and cardiac ICU team preference and not based on a set protocol. For this reason, pacemaker placement could have been performed earlier than guidelines suggest for other reasons (i.e., early unplanned reoperation). The group in whom late atrioventricular node recovery was present was small and warrants a larger, multi-centre study to better characterise this population.
In conclusion, post-surgical heart block occurs in 2% of patients undergoing cardiac surgery for CHD. In our study, 91% of resolution occurred by post-operative day 10. Permanent pacemaker placement occurred in 0.6% (38/6333) of patients, and permanent post-operative heart block occurred in 0.5% (32/6333) of patients. Higher weight, shorter cardiopulmonary bypass time, and presence of single-ventricle physiology are associated with the resolution of post-surgical heart block. Late recovery of atrioventricular node conduction following permanent pacemaker placement occurred in 15.8% (6/38) patients and is associated with greater age and weight, but is not associated with gender, post-operative steroids, infectious, or wound complications.
Acknowledgements
None.
Financial support
This research received no specific grant from any funding agency, commercial, or not-for-profit sectors.
Conflicts of interest
None.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national guidelines on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008, and have been approved by the institutional committees (Institutional Review Board at the Children’s Healthcare of Atlanta).
