Discussion
Many factors affect the morbidity and mortality in surgical ventricular septal defect closure. In most patients, surgeon experience and the anatomical location of the ventricular septal defect are as important as the patient’s clinical state in deciding whether to perform an early surgical ventricular septal defect closure. Owing to the heterogeneous patient distribution that occurs with the combination of these factors that affect surgical outcomes, questions are raised in surgical planning. Despite the multifactorial features of patients and the improvements in the surgical and perfusion approaches, outcomes in terms of mortality and morbidity have been improving. Although total correction of ventricular septal defect in patients with a younger age and lower body weight sometimes seem to be ambiguous and risky for cardiac surgeons and cardiologists, it is the first-choice approach with the advances in congenital heart surgery. Palliative pulmonary artery banding has lost importance together with complete repair in children with symptomatic ventricular septal defect and pulmonary overcirculation. It is no longer the preferred palliative surgical approach because of complete repair can now be performed safely at a younger age and in patients with lower body weights. In the decision making on the surgical timing during the follow-up period, the first factor should be the clinical condition of the patient with ventricular septal defect. By contrast, given the improvements of the results of neonatal total corrective surgery, the pediatric cardiologist challenges the surgeon to proceed with surgical closure of symptomatic ventricular septal defects in ever smaller patients [4]. Although surgical ventricular septal defect closure in patients low body weights and ages (<6 months) can be safely performed with a low incidence of complications, no clear consensus has been reached on the timing of surgical ventricular septal defect closure.
Currently, surgical ventricular septal defect closure seems to be a reasonably safe procedure, with near-zero hospital mortality and low complication rates. However, mortality and prolonged PICU or hospital stay due to complications have still been reported. According to previous studies, the mortality rate ranges from 0% to 2.8%, reoperation rates due to residual defects range from 0% to 4.9%, pacemaker implantation rates range from 0% to 5.6%, and major adverse event rates range from 2.9% to 5.9% [2–4, 6–8]. In this study, none of the patients died, had a residual defect requiring reoperation, and needed a permanent pacemaker, and the incidence of major adverse event was 0.9%. In fact, on postoperative echocardiography, none of the patients had a residual ventricular septal defect of ≥2 mm in size.
Schipper et al. found that genetic syndrome was associated with MV and PICU hospitalization times [3]. Similarly, Anderson et al. reported a strong relationship between genetic syndromes and prolonged hospitalization in their study [2]. In this study, we found that patients with genetic syndromes had longer hospitalization (p = 0.002), PICU hospitalization (p < 0.001), and MV durations (p < 0.001).
In the study of Anderson et al., patients who received delayed postoperative nutrition had longer postoperative hospitalization and MV durations [9]. Sahu et al. found that early nutritional therapy was associated with decreased ventilation duration, infection rate, length of hospital stay, lenght of ICU stay, and mortality [10]. Similarly, in our study, prolonged PICU stay, hospital stay, and MV duration, as well as the occurrence of complications, were associated with the initiation time of nutritional therapy.
Despite the development of surgical techniques and perfusion technology, some studies have shown that morbidity was higher in infants with lower body weights and ages < 6 months. In their study, Anderson et al. found that those who were aged >6 months had less hospitalizations and lower risk of temporary or permanent heart block. In addition, higher body weight during operation was associated with a significant decrease in the incidence of postoperative bleeding. Again, in the group aged <6 months, the ICU hospitalization duration decreased by 2.3 days in every kilogram increase in body weight, and the complication rate increased 1.8 times in every kilogram loss of body weight [2]. In the study of Schipper et al., patients who weighed <6 kg had longer postoperative ICU hospitalization and MV durations [3]. In a study conducted by Kogon et al., body weight did not affect the operation, aortic cross-clamp, bypass, and MV times, and complication rate [4]. Aydemir et al. showed no significant difference between age groups in terms of morbidity [7]. Ergün et al. showed that the increase in body weight was associated with the decrease in the risk of major adverse events and shortened ICU and hospital stays. In the study, <4-kg body weight was a risk factor of prolonged MV time [6]. Vaidyanathan et al. found that younger age was associated with longer ICU and hospital stays. In this study, no effects of body weight and height Z scores were demonstrated [11].
In our study, the univariate analysis revealed lower age and weight in the patients with complications and prolonged ICU and hospitalization stays, and MV duration. In terms of the presence of complications, the cutoff age and weight were 5 months and 5.8 kg, respectively, similar to those in other studies.
In the study of Mildh et al., in a population-based group of patients who underwent pediatric open-heart surgery, PRISM proved to be a poor tool for risk stratification owing to its low discrimination and calibration values [12]. The PIM-2 measures demonstrated good performance regarding the capacity to discriminate between survivors and non-survivors in a population of patients with postoperative congenital heart disease in the study of Rezende et al [13]. These studies only provided information about the effects of the PIM and PRISM score systems on mortality. We evaluated the effects of these scoring systems on morbidity. In our study, prolonged hospitalization was associated with higher PIM-2 and PRISM scores. In addition, the PRISM score was found to be higher in the patients who needed prolonged MV.
In the multivariable analysis of our study, prolonged PICU stay was associated with MV time of patients (odds ratio [OR], 0.9; 95% confidence interval [CI], -1.22–3.03; p < 0.001), and nutritional therapy time (OR, 0.05; 95% CI, 0.01–0.1; p = 0.044). Prolonged hospital stay was associated with MV time of patients (OR, 0.6; 95% CI, 0.44–0.74; p < 0.001). Prolonged MV time was associated with nutritional therapy time (OR, 0.25; 95% CI, 0.13–0.35; p < 0.001) and, PRISM score (OR, 0.43; 95% CI, 0.16–0.7; p = 0.002). In this respect, multicenter studies may provide more-precise results.
The limitations of our study are that it had a retrospective design and was conducted in a single center. The other limitations of the study are the absence of long-term follow-up, small number of patients with major adverse events, and absence of mortality cases.