Incidence, Risk Factors, and Outcome of Cardiac Arrhythmia Postcardiac Surgery in Children.

OBJECTIVE: To study the incidence of postoperative cardiac arrhythmias in children undergoing cardiac surgery and to evaluate the risk factors and outcome of these patients.
MATERIALS AND METHODS: This retrospective observational study was conducted in the cardiac pediatric intensive care unit and included children <18 years of age. Children were monitored in the early postoperative period (72 h) for any sustained rhythm abnormality and were classified using standard definition. Details of treatment and their response were assessed. Risk factors for arrhythmias were evaluated using multivariate logistic regression analysis.
RESULTS: Five hundred and thirty-six children were included and the prevalence of arrhythmia was 14.4% (n = 77). The most common arrhythmia was complete heart block (CHB) (n = 28; 5.2%), followed by junctional ectopic tachycardia (JET) (n = 25; 4.7%), junctional escape rhythm (n = 13; 2.4%), supraventricular tachycardia (SVT) (n = 8; 1.5%), and ventricular tachycardia (VT) (n = 3; 0.6%). Cardiac pacing was required in all CHB; 8 (28.6%) required a permanent pacemaker. Six (24%) patients with JET responded to conventional measures; 19 (76.0%) patients required amiodarone and 5 (20%) required cooling to 34°C or cardiac pacing. Temporary cardiac pacing was required in 9 (69.2%) cases of junctional escape rhythm. Seven (87.5%) events of SVT responded to adenosine and 1 (12.5%) required cardioversion. Two (66.7%) of VT responded to cardioversion while 1 (33.3%) was refractory. Five (6.5%) patients with arrhythmia died. In the multivariate logistic regression analysis, age <1 year, risk adjustment for congenital heart surgery category ≥3, and cross-clamp time >67 min were independent risk factors.
CONCLUSION: Early postoperative period following cardiac surgery is extremely vulnerable to cardiac arrhythmias. Although majority are self-limiting, some can be life-threatening.

INTRODUCTION

Arrhythmia is a common event in postoperative cardiac pediatric intensive care unit (PICU) often manifesting with significant hemodynamic deterioration, resulting in increased risk for mortality and morbidity.[12] The reported prevalence of arrhythmia is between 7.5% and 48%.[134567] Most events occur during the early postoperative period as a result of direct injury or manipulation to cardiac conduction system during the surgery or due to myocardial edema and inflammation adjacent to the conducting system. High catecholamine levels and electrolyte disturbances during the postoperative period are also a likely cause.[89] The long-standing myocardial dysfunction from pressure or volume overload also makes these patients, especially vulnerable to arrhythmias.[10] Reported risk factors include young age and low body weight at surgery, longer cardiopulmonary bypass time (CPB) and aortic cross-clamp (AXC) time, electrolyte imbalance, temperature disturbance, and use of deep hypothermia and circulatory arrest.[1112]

Postoperative arrhythmias can be life-threatening, and prompt management is required. Although the majority of the arrhythmias are transient or self-limiting and recover with proper management, some especially bradyarrhythmias may require a permanent pacemaker.

This study was conducted to identify the incidence rate of different postoperative cardiac arrhythmias in our cohort and evaluate the associated risk factors. We also evaluated the response of individual arrhythmias with treatment and outcome of these patients.

MATERIALS AND METHODS

This retrospective observational study approved by the institutional review board and ethics committee was conducted in a 65-bedded pediatric cardiac ICU of a cardiac institute in India. Patients ≤18 years of age undergoing cardiac surgery between April and July 2016 were included in the study. Patients with preexisting rhythm abnormality or transient rhythm abnormalities in the operation theater which did not recur postoperatively were not included in the study.

All the children were intensively monitored in the early postoperative period (72 h) for the detection of any rhythm abnormality. Any sustained rhythm abnormality (those lasting for ≥30 s duration, recurrent and/or causing hemodynamic disturbance) was assessed by standard 12-lead electrocardiogram (ECG).[3] Sinus tachycardia and ectopic rhythm which were hemodynamically insignificant were not included in the study as they are quite common in postoperative patients and the majority have noncardiac etiology. Their inclusion may falsely increase the number of events.

Individual arrhythmias were defined using standard definition, and specific abnormality was labeled after consultation with pediatric cardiologist. Relevant data were collected from the PICU charts and case records. Dosage of inotropes at the time of initiation of arrhythmia was recorded. Electrolyte (potassium, calcium, and magnesium) levels were assessed at the time of arrhythmia and temperature recorded whenever required. Treatment given for individual arrhythmias was decided by a pediatric cardiac intensivist in consultation with a pediatric cardiologist, and the response of individual arrhythmias to therapy was recorded. Patients were followed up till discharge from PICU and the outcome assessed.

Patients were divided into risk category based on risk adjustment for congenital heart surgery (RACHS) system. Vasoactive inotropic score (IS) was calculated using the formulae: VIS = Dopamine dose (μg/kg/min) + dobutamine dose (μg/kg/min) + (100 × adrenaline dose [μg/kg/min]) + (10 × milrinone dose [μg/kg/min]) + (10,000 × vasopressin dose [unit/kg/min]) + (100 × norepinephrine dose [μg/kg/min]).[13]

The results were analyzed using SPSS 22 software (IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.). Incidence and nature of arrhythmia were analyzed in relation to the procedure performed. Pearson's Chi-square test was used to evaluate categorical data and Mann–Whitney U-test for continuous data.

Patients were divided into two groups: those who developed arrhythmia and those who did not. Univariate analysis was performed to identify the possible risk factors. For further analysis, continuous variables were dichotomized as upper (worst) 25th percentile versus lower (best) 75th percentile. Variables which achieved significance in univariate analysis were entered into a multivariate logistic regression analysis to analyze independent predictors for arrhythmia. P < 0.05 was considered statistically significant, and odds ratio (OR) was calculated.

RESULTS

Five hundred and thirty-six children with a median age of 16 months were included in the study. The most common procedure performed was ventricular septal defect closure (n = 131; 24.4%) and intracardiac repair of tetralogy of Fallot (n = 88; 16.4%). The majority of the procedures (55.2%) were classified in RACHS category 2; none in category 5 and 6. Seventy-seven (14.4%) children developed arrhythmia. Forty-six (59.7%) events occurred within 1 h and 71 (92.2%) within first 24 h. Most common arrhythmia was complete heart block (CHB) (n = 28, 5.2%), followed by junctional ectopic tachycardia (JET) (n = 25; 4.7%), junctional escape rhythm (n = 13; 2.4%), supraventricular tachycardia (SVT) (n = 8; 1.5%), and ventricular tachycardia (VT) (n = 3; 0.6%). The frequency of the types of arrhythmias is depicted in Table 1.

Table 1

Frequency of the types of arrhythmias in the study

Arrhythmia	Frequency	Incidence in relation to
		Arrhythmia cases (n=77)	Total cases (n=536)
Complete heart block	28	36.4	5.2
Junctional ectopic tachycardia	25	32.4	4.7
Junctional escape rhythm	13	16.9	2.4
Supraventricular tachycardia	8	10.4	1.5
Ventricular tachycardia	3	3.9	0.6

No child operated for acquired heart disease developed arrhythmia. Intraventricular tunnel repair (61.1%) was the most common procedure associated with arrhythmia. The frequency of arrhythmia with the type of procedure is depicted in Table 2.

Table 2

Frequency and relation of individual arrhythmia with specific procedure

Procedure (n)	Frequency (%)
Complete heart block (n=28)
TOF repair (88)	8 (9.09)
VSD closure±PDA ligation (131)	8 (6.10)
AVSD repair (26)	4 (15.4)
Fontan procedure (16)	2 (12.5)
IVTR (18)	2 (11.1)
ASO (18)	2 (11.1)
TAPVC/PAPVC (51)	2 (3.9)
Junctional ectopic tachycardia (n=25)
IVTR (18)	5 (27.7)
ASD + VSD closure±PDA ligation (18)	5 (27.7)
TOF repair (88)	4 (4.5)
AVSD repair (26)	4 (15.4)
VSD closure±PDA ligation (131)	3 (2.3)
ASO (18)	2 (11.1)
Fontan procedure (16)	2 (12.6)
Junctional escape rhythm (n=13)
ASD + VSD closure±PDA ligation (18)	4 (22.2)
TOF repair (18)	4 (22.2)
IVTR (18)	2 (11.1)
TAPVC/PAPVC (51)	2 (3.9)
ASD closure (35)	1 (2.9)
Supraventricular tachycardia (n=8)
BT shunt (48)	3 (6.3)
IVTR (18)	2 (11.1)
ASO (18)	1 (5.6)
TOF repair (88)	1 (1.1)
VSD closure ± PDA ligation (131)	1 (0.8)
Ventricular tachycardia (n=3)
ASO (18)	1 (5.6)
BT shunt (48)	1 (2.1)
TAPVC/PAPVC (51)	1 (1.9)

PDA: Patent ductus arteriosus, VSD: Ventricular Septal Defect, TOF: Tetralogy of Fallot, TAPVC/PAPVC: Total anomalous pulmonary venous connection/partial anomalous pulmonary venous connection, IVTR: Intraventricular tunnel repair, BT: Blalock-Taussig, ASO: Arterial switch operation, ASD: Atrial septal defect

All the events of CHB were recognized in the operation theater and or within first 24 h of shifting to PICU. None of them were hemodynamically significant. Temporary cardiac pacing was required in all the patients with CHB. Although 24 (85.7%) patients recovered within 10 days; in 3 patients (10.7%) the CHB persisted long enough to require permanent pacemaker. One (3.6%) patient died although the cause of death could not be attributed directly to CHB [Figure 1].

Figure 1

The frequency, treatment, and outcome of arrhythmia in the study

The majority of the patients with JET (n = 24; 96.0%) were hemodynamic unstable and required treatment. The majority (n = 23; 92%) of the events occurred within first 24 h of surgery. Hypomagnesemia was present in only 4 (16%) patients with JET, which was managed accordingly. No other significant electrolyte or temperature abnormality was present. Six (24%) patients responded to conventional measures which included reducing the dosage of inotropes, optimization of sedation and analgesia, temperature maintenance, and correction of electrolyte abnormality.

Dexmedetomidine infusion was the preferred sedative used. Fourteen (56.0%) patients responded to amiodarone infusion in addition to conventional therapy. Five (20%) patients who did not respond to the above therapy required cooling to 34°C and cardiac pacing. Three (12%) patients died due to refractory low cardiac output syndrome as a direct result of JET. The rest 22 (88%) patients reverted to normal sinus rhythm within 7 days [Figure 1].

None of the patients with junctional escape rhythm (n = 13; 2.4%) had hemodynamic instability and two presented with accelerated junctional rhythm. Eleven (84.6%) events occurred within 24 h of surgery and 1 within 48 h. Temporary cardiac pacing was required in 9 (69.2%) cases and normal sinus rhythm was achieved in all cases within 10 days. None of the patients required permanent pacemaker. None of the patients with junctional rhythm died. Six (75%) events of SVT presented within 24 h of surgery. Seven (87.5%) were hemodynamically stable and responded to adenosine. Five (62.5%) of them had recurrent events and required amiodarone. One (12.5%) was associated hemodynamic instability and required synchronized cardioversion followed by amiodarone. None of the patients with SVT died [Figure 1].

No significant electrolyte abnormality was present in any of the cases. VT was present in 3 (0.6%) patients, all being monomorphic VT and majority 2 (66.7%) presented with first 24 h. Normal sinus rhythm was achieved in 2 (66.7%) patients after synchronized cardioversion followed by which amiodarone infusion was started and no recurrent events were documented. One (33.3%) patient was refractory to cardioversion and required amiodarone and lidocaine. No possible correctable cause could be determined, and patient was put on extracorporal membrane oxygenator who later died [Figure 1].

In univariate analysis, age <1 year, higher RACHS category, procedure requiring CPB, higher CPB and AXC time, dobutamine >5 μg/kg/min, adrenaline >0.05 μg/kg/min, and higher VIS were found to be significantly associated with postoperative arrhythmia. For multivariate analysis, cutoff for CPB and AXC time was taken as 110 and 67 min whereas VIS of more than 10 was taken as high inotropic requirement. In the multivariate logistic regression analysis, age <1 year (OR - 1.96; P = 0.009), RACHS category ≥3 (OR - 2.21; P = 0.032), and AXC time >67 min (OR - 2.32; P = 0.024) were found to be independent risk factors [Table 3].

Table 3

Multivariate logistic regression analysis of the risk factors for arrhythmia in the study

Variable	Odds ratio	95% CI	P
Age <1 year	1.96	1.18-3.24	0.009
RACHS category ≥3	2.21	1.72-4.04	0.032
CPB time >110 min	1.87	0.89-3.95	0.056
AXC time >60 min	2.32	1.18-4.81	0.024
Adrenaline >0.05 µg/kg/min	1.34	0.39-4.64	0.643
AXC 24 h >10	1.91	0.67-5.42	0.224

CI: Confidence interval, RACHS: Risk adjustment for congenital heart surgery, CPB: Cardiopulmonary bypass time, AXC: Aortic cross-clamp time

DISCUSSION

This study cohort was assessed for the occurrence of arrhythmias following cardiac surgery and considering all possible risk factors.

The prevalence of arrhythmia (within 3 days of surgery) in the current study which included children <18 years irrespective of the procedure performed was 14.4%. The reported prevalence of arrhythmia range between 7.5% and 48%.[134567] The variation is probably because of the sensitivity of the method and the definition of arrhythmias used. Most studies have used overhead bedside monitor for detection of arrhythmias like the current study. Valsangiacomo et al. found the prevalence of arrhythmias as 48% within 1 day of cardiac surgery whereas Pfammatter et al. reported the prevalence of 27%.[14]

In another study by Grosse-Wortmann et al. who used Holter which is a more sensitive method than bedside monitoting. The peak prevalence of arrhythmias was 73.4% and 79.1% in neonates and older children, respectively, which when disregarding the usual benign rhythm abnormalities dropped to 29.6% and 38.9%.[6]

Definition of arrhythmia has a significant effect on prevalence. In the current study, benign arrhythmias such as sinus tachycardia and ectopic which was hemodynamic stable were excluded. Delaney et al. reported the prevalence of 15% with arrhythmia which they defined as events that necessitated intervention.[5] Another study by Yildirim et al. reported the prevalence of 8.8% using sustained arrhythmia as definition.[3]

CHB accounted for 36% of the arrhythmias in the current study with a prevalence of 5.2%. AV block is common in surgeries involving maximum manipulation of conducting system and occurs in complex procedure requiring major dissection. Fryda et al. in 1971 reported the incidence of postoperative CHB to be 25%; however, the improvement of surgical techniques and better understanding of the anatomy of the conduction tissue have reduced this risk to 1%–4% in the current era.[1415] CHB following surgical repair may be transient or permanent. In the current study, 71.4% of the CHB recovered with 10 days and the remaining required permanent pacemaker. Transient AV block generally reverts to sinus rhythm within the first 7–10 days after surgery though late recovery even months later is reported. In the current study, patients were followed up only till PICU discharge and any late recovery could not be commented.[1617] Weindling et al. have shown that 63% of patients with AV block after surgery recovered in the first postoperative month, with majority (97%) occurring in the first 9 days after surgery.[16] Recent reports have suggested the value to be around 43%–92%.[161718192021] This variation could be attributable to a variety of factors, such as era, inclusion criteria, and duration of follow-up.

JET is the next common arrhythmia in the current study with prevalence of 4.7%. It is one of the most common arrhythmias encountered in postoperative cardiac surgery unit with prevalence ranging between 2% and 11%.[1222] The precise mechanism of JET is not known, but it is believed to be a result of direct trauma to the AV node and bundle of His, although JET occurs in patients in whom there is no operation near the AV node.[232425262728] In the current study, 92% of cases were hemodynamically unstable. Six (24%) responded to conventional measures and 76% required additional therapy. JET resolved successfully in 88% of cases and 3 (12%) died. In the study by Abdelaziz and Deraz, JET successfully resolved in 86.5% of patients, with 57% responding to conventional measures and the remaining 43% were controlled by amiodarone infusion in addition to conventional measures.[29]

Junctional escape rhythm was noted in 13 patients, which compromised 32.4% of total events and incidence rate 4.7%. Cardiac pacing was required in 69.2% of cases and normal sinus rhythm was achieved in all patients within 10 days. Valsangiacomo et al. reported junctional escape rhythm to compromise 46% of total arrhythmia cases. Sinus bradycardia was the most common indication reported in their study.[1] Although SVT has been reported to be one of the most common arrhythmias reported with prevalence between 1% and 30% in postoperative cardiac patients, the current studies reported an incidence of 1.5%.[13512]

The prevalence of VT has been reported to be >3% in majority of the studies except that of Hoffman and associates, who reported a much higher incidence of nonsustained VT (15.2%), with 2% incidence of sustained VT.[25] Patients with preserved ventricular function and those with nonsustained VT have a favorable outcome. Patients with sustained VT and those associated with ventricular dysfunction may have mortality as high as 50%.[3031]

The current study found age <1 year, RACHS category ≥3, and AXC time >67 min to be independent risk factors for postoperative arrhythmia. Younger age group and lower body weight along with longer AXC time have been reported as independent risk factors for postoperative arrhythmia in previous studies.[1512] Longer CPB time has also been associated with arrhythmias in earlier reports which did not hold true in the current study although frequency of arrhythmia was higher in procedure with higher CPB time. Higher complexity of the surgical procedure has been associated with arrhythmias in earlier reports as in the current study.[112]

Alp et al. found an association of VSD closure with ventricular extrasystole, ASD closure, VSD closure, and TOF repair with SVT whereas Delaney et al. found an association of arrhythmia with atrioventricular septal defect.[511] In the current study, although the frequency of specific arrhythmias in different procedure was studied, no relation with any procedure was established.

The limitation of the study was the possibility of missing data due to the retrospective nature of the study. This was a small study from a single center. Furthermore, patients were monitored on real-time monitors and confirmation was done by 12-lead ECG, which is a less sensitive method and could have missed few abnormalities. The study also could not establish the association of arrhythmia and specific procedure. Association of mortality and morbidity outcome with different arrhythmias was also not studied.

CONCLUSION

The early postoperative period following cardiac surgery is extremely vulnerable to cardiac arrhythmias. Although the majority are self-limiting and recovers with appropriate treatment, some can be life-threatening. Younger age group, longer cross-clamp time, and complexity of procedure are independent risk factors.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.