Supraventricular tachycardia (SVT) is the most common sustained arrhythmia in the neonatal age group, with an estimated incidence of infants as 0.25 per 1000 patient and < 1 month as 0.06 per 1000 patient [16]. AHF occurred more commonly in 35% of patients under 4 months of age [17], that is close to the incidence of 39.5% in our study. As well toleration in the first 12–24 hours in SVT of neonate, symptoms may be mild and tachycardia may not be recognized for a long time, thus leading to AHF commonly [18], so that explicit risk factors influence the occurrence of AHF would do me a favor to early detection for preventing decompensation.
In this study, we found that longer duration of SVT— time to initial control of tachycardia > 24h could increase the risk of AHF, similarly to the finding reported by Nadas et al that a 19% incidence of AHF if tachycardia continued until 30 h, and a 50% incidence of AHF if tachycardia last 48 h [19]. Comorbid conditions such as an inflammatory state, hypoxia, acidosis and electrolyte imbalance etc may be capable of triggering arrhythmia of SVT and devoted to hemodynamic instability [20, 21]. In our study, hyperkalemia and anemia were risk factors for AHF in SVT. B-type natriuretic peptide (BNP), a cardiac peptide released by the heart ventricles in response to changes in the ventricular pressure and/or volume, reported associating with AHF in children from other causes such as CHD [22]. Gisela L. Salas has found that an increase in BNP levels measured in critically-ill neonates requiring assisted mechanical ventilation may predict hemodynamic changes and poor prognosis [23]. BNP in our study was identified to be a risk factor for AHF in neonates secondary to SVT.
Structural heart disease contributes to cardiovascular collapse during a tachycardia episode [24], but in our study not identify it as a risk factor, may be that major subjects were simple congenital heart diseases with small shunt volumes, while others’ study involved large shunt volumes or complex congenital heart disease. As cardiac pump reserve function is limited especially in immature, and fast heart rate would lead to a declined cardiac output, prenatal history, prematurity, intrauterine tachycardia and urgent caesarian section regarded as an indicator of decreased fetal circulation might be associated with unfavorable clinical outcome in neonates with SVT [25, 26]. But, we didn’t find these results, perhaps condition is not that serious. Lower body weight and younger age reported associating with fatal or near-fatal outcome in infant with SVT [8], that seems not with AHF in our study, maybe because that subject of similar age and weight in each group.
Indeed, BNP is rarely used as a biomarker in newborns, because it can be affected by extra-cardiac conditions like anemia, severe infections, even by some prenatal and postnatal factors like mothers with type 1diabetes, prematurity, cesarean section [27]. Shawn Reeves etal observed a extremely high level approximately 20,000 pg/mL of N-terminal pro-brain natriuretic peptide (NT-proBNP), which originating from breaking down of BNP, in 3 neonates with decompensated SVT [28], indicating the potential of plasma BNP in predicting AHF secondary to SVT in neonates. In our study, the value of BNP for predicting AHF was identified to be 2460.5pg/ml. That is significantly higher than 758.7 pg/mL ~ 741.4 pg/mL at 97.5th percentile in normal infants aged from 0-30d reported by Cantinotti M [27]. Further more, among the associated risk factors, BNP > 2460.5pg/ml was identified to be an independent predictor, indicating that the incidence of AHF secondary to SVT is not only affected by the tachycardia itself, but also by many other factors associated the overall cardiovascular status, precisely that BNP may appropriately reflect the picture.
In addition to early intervention according to type of tachycardia, an efficient option of treatment to hasten terminating SVT is also critical to prevent decompensation. Digitalis was the most used first line drug in our study, it seems not superior to other first-line drugs in preventing HF, though the combination of positive inotropic activity with negative chronotropic effects has been shown to reduce hospital admissions in heart failure [29]. Inverse, esmolol seems more beneficial for preventing AHF, as we found that esmolol terminated SVT without developing HF in 4 neonates including who did not respond to digitalis, indicating the positive inotropic effect may be not fully advantageous when tachycardia with preserved ejection. In this case, esmolol alone or in combination with digitalis may be more effective in the control of elevated haemodynamic parameters in patients with SVT as reported [30].
Till now, it's still a dilemma for cardiologists to balance efficiency and safety in terminating acute recurrent and persistent SVT in neonates. Compared to first-line treatment, the second line therapy has an attractive efficacy in case of refractory SVT, but remains a reserved option owing to reported relatively high incidence of systemic adverse effects, such as cardiac arrest inducing by propafenone, hypothyroidism and pulmonary fibrosis by amiodarone, cardiac depression by DC cardioversion, potential pro-arrhythmia in themselves [2, 15]. However, some authors advocate amiodarone and propafenone are equally safe and effective used with monitoring as a first choice drug especially in infancy [31, 32, 33]. In this study, 8 newborns who developed AHF terminated the prolonged SVT by the second line therapy ultimately without adverse effects, and we found that neonates with a BNP > 2460.5pg/ml had a higher demand for second-line therapy to control SVT. The equal value of BNP in predicting AHF and anti-arrhythmic treatment may be not only a coincidentally. That's exactly Bjeloševič M reported that heart failure is a possible predictor of arrhythmia persistence, the need for ablation and mortality rate are reduced by commonly use amiodarone and propafenone in arrhythmia termination [34]. Though the diagnostic accuracy is not significant, the value of BNP in monitoring and predicting response to treatment deserves further study.
This study has some limitations. First, the sample size of this study is small. Second, the diagnosis of heart failure is made by ROSS score, which is partly subjective. Third, selection of anti-arrhythmic drugs and doses were at the preference of the physician on duty. This circumstance made it impossible to study and characterize a well-defined therapy protocol. Forth, retrospective study design impeded assessment of adverse effect of anti-arrhythmic treatment, none of these neonates showed significant adverse reactions. Therefore, the definition of relatively higher risk anti-arrhythmic treatment is based on previous literature reports. Despite these limitations, we believe that our data about neonatal SVT can be useful for neonatologists in the neonatal emergency room. Further multicenter prospective studies are needed to confirm our findings.