Among 71 patients of TOF who underwent ICR in the Department of Cardiovascular and Thoracic Surgery from 1^st January, 2016 to 30^th April, 2017, six had previous modified BT shunts, three required deep hypothermic circulatory arrest, two got self-extubated post-operatively. These 11 patients were excluded and the remaining 60 patients were finally included for retrospective analysis of results. The total number of cases included in the study was based on the number of patients available who had undergone the above mentioned surgical procedure at our institute over the period of time mentioned. Informed consents were taken from the parents to allow the results of analysis being published.

All procedures performed in the cases were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Approval was taken from the Institutional Ethics Committee (Letter number: CTVS/15-331).

Patients below 18 years of age, who underwent ICR for TOF and whose parents had given informed consent, were included.

Patients with prior cardiac/thoracic surgery, TOF with other major intracardiac anomalies, those requiring deep hypothermic circulatory arrest, patients with significant comorbidities and syndromic patients, emergency cases and patients unwilling to participate were excluded.

A retrospective analysis of outcome of CPB temperature was done. Thirty patients who had undergone normothermic CPB and 30 patients who had undergone hypothermic CPB were divided into two groups and results were analysed between them.

Objective data were obtained from the patients’ medical records.

In this study, the co-primary endpoints were requirement of inotropic support, total duration of post-operative mechanical ventilation, total duration of post-operative hospital stay.

The secondary outcomes were in-hospital mortality and morbidity rates in each group.

Morbidity was defined as prolonged intubation (greater than six hours post-operatively), need for re-exploration, post-operative wound infection, need for re-intubation during the first 24 hours from extubation (failure of extubation), post-operative fever, prolonged post-operative ICU stay (more than seven days), post-operative renal dysfunction (>30% decrease in serum creatinine clearance levels in the post-operative period as compared to preoperative baseline values).

Other secondary outcomes were intraoperative need for vasodilators/vasoconstrictors, after coming off CPB need for DC (direct current) shock to revert to normal sinus rhythm, after coming of CPB following adequate rewarming presence of cardiac arrhythmias, post-operative blood loss, transfusion requirements, biventricular function (from echocardiography), blood gas parameters, blood sugar levels, clotting parameters, haemolysis (recorded in terms of presence or absence of it in terms of urine colour) and total post-operative duration of ICU stay.

Normally distributed variables were expressed by their mean and standard deviation; not normally distributed variables were expressed by their medians and interquartile ranges; categorical variables were expressed as n(%). In test groups of continuous normally distributed variables, Student’s t-test was used. Likewise if continuous data were not normally distributed the Mann-Whitney U test was used. Categorical were compared with the Chi-square test or Fisher’s-exact tests (if expected cell count <5) between two study groups or when appropriate as relative risks. Statistical uncertainty was expressed by 95% confidence levels. Baseline balance was statistically tested.

Time to event variables were analysed using univariate nonparametric Kaplan Meier survival analysis and a Cox proportional hazard model adjusted for possible imbalances of patients’ baseline characteristics. Time course variables were analysed using repeated measure ANOVA (analysis of variance) and a linear mixed model. Area under curve for various blood parameters were also estimated and compared between two trial arms. Statistical softwares Stata IC version 14, R version 3.2.1 and MedCal were used for analysis and all tests were two-tailed with level of significance as <0.05.

In the normothermia group out of the 30 patients, 13 (43.3%) were females and 17 (56.7%) were males. In the hypothermia group 14 (46.7%) were females and 16 (53.3%) were males.

The two groups were comparable in terms of the age, height, weight and Body Surface Area (BSA), Myocardial Performance Index (MPI) Right Ventricle and Left Ventricle (RV and LV), as well as presence or absence of Major Aortopulmonary Collaterals MAPCAs) [Table/Fig-1,2].

Mean duration of surgery in the normothermic group was 4.22±1.006 hours. In the hypothermic group the mean duration of surgery was 5.13±1.050 hours (p-0.001) and this difference was statistically significant [Table/Fig-3].

The need for vasopressors, presence of arrhythmias, need for DC shock, presence of haemolysis, intra and post-operative blood sugar levels, mean duration of ventilation, incidence of delayed extubation and inotropic scores for the two groups have been provided in the [Table/Fig-4,5,6,7,8 and 9]. The incidence of delayed extubation was more common in the hypothermia group which was statistically significant. Rest of the parameters were comparable.

There was no effect on the CPB temperature on the intra-operative and post-operative ABG reports.

The coagulation parameters and the post-operative transfusion requirements have been summarised in [Table/Fig-10]. Apart from the transfusion of FFP, the two groups were similar in their results.

The post-operative blood parameters were also not statistically significant when compared between the groups [Table/Fig-11].

The mean post-operative MPI for LV and RV along with difference from the preoperative values, mean number of post-operative ICU stays, incidence of prolonged ICU stay, need for re-exploration, hourly drain output, fever, mean number of days of post-operative hospital stay as well as incidence of prolonged hospital stay have been documented in the [Table/Fig-12,13,14,15,16,17 and 18]. None of the patients had surgical site infection in either group. There was significantly better post-operative RV function in the normothermia group. Besides the normothermia group had less duration of post-operative ICU stay.

The baseline characteristics between the two groups were comparable. When the two groups were compared in this regard the mean VIS (vasoactive inotrope score) was less in the normothermic than in the hypothermic group. However, this difference was not high enough to be statistically significant. In a study involving normothermic CPB in paediatric patients, there was statistically significant difference in inotropic requirements [11]. The greater incidence of RV dysfunction associated with ICR for TOF patients perhaps was the predominant reason behind this difference in outcome.

Shamsuddin AM et al., showed that with use of normothermic high flow CPB in the paediatric age group more than 90% patients were extubated within 3 hours [11]. However, the surgeries in their study were relatively simpler than ICR. Although, our study also showed a favourable outcome with respect to duration of mechanical ventilation like their study, the greater mean duration of ventilation in either group could be attributed to the greater complexity of the surgery involved. The results of our study were however comparable to other studies published in Europe [12].

Similar studies showed that the mean duration of post-operative hospital stay was about 7.2 days with normothermic CPB [13]. Although the duration of hospital stay was less for the patients with normothermic CPB, our study population had about 1.5 days of longer hospital stay. This could possibly be attributed to poor general condition as well as greater ICU stay in our patients which was again possibly due to the nature of the surgery as well as the greater requirements of mechanical ventilation.

We saw that although the duration of ICU stay was significantly lowered by application of normothermic CPB, hypothermic CPB does not necessarily result in a prolonged ICU stay. However, even a decreased stay in the ICU by one day would mean less of a financial as well as psychological burden for the patients and their guardians and hence application of normothermic CPB will be welcome in this regard. When we compared the results with other similar studies, although our study showed similar trends in the outcomes, the mean duration of ICU stay in the normothermic group was higher in our group (3.24 vs 2.7) [11]. The patient population in our study were operated for TOF and showed varied degrees of RV dysfunction post-operatively which increased the dose and duration of inotropic support post-operatively along with increased mechanical ventilation. Both these factors could possibly have resulted in the greater duration of ICU stay in our study population.

In terms of creatinine clearance the results of our study were comparable to the study of Xiong Y et al., [14]. Provenchere S et al., also had not found any correlation between CPB temperature and renal dysfunction in their study [15].

Bleeding is an inherent risk factor having a major impact on patient prognosis in cardiac surgery. Cyanotic patients exhibit a greater degree of platelet dysfunction which leads to increased blood loss. We hence tried to analyse the effect of CPB temperature on this vital parameter. Since the patients had wide range of body weights we decided to compute the hourly drain output per kg body weight. Bleeding was not significantly different. Thus, we concluded that as long as re-warming post CPB was complete, the CPB temperature did not have a significant effect on the volume of post-operative blood loss. When the two groups were thus compared requirement of FFP was significantly more with the hypothermic group. However, no differences were found in terms of PC and PRBC requirements. We could not find any similar study to have assessed this parameter in their groups.

Varying degrees of RV dysfunction was noted in almost all the patients while LV function was more or less preserved. When these data were compared in between the two groups, hypothermic CPB was found to be associated with a greater degree of RV dysfunction. Especially in a surgery like ICR for TOF where the patients are at a greater risk of RV dysfunction, this difference was of immense importance from the clinical perspective. However, in the patients requiring transannular patch for their repair which were more in the hypothermia group, the degree of RV dysfunction is usually greater, and this could have acted as a potential source of bias in this assessment. This parameter again had not been assessed in prior studies.

ABGs had been performed on the patients at several points during the course of surgery and post-operatively. We had taken up the parameters of pH, bicarbonate and base excess to assess the degrees of, metabolic acidosis and their normalisation. However when the two groups were compared in this regard, the differences were not statistically significant. Xiong Y et al., had used lactate levels to compare the levels of metabolic acidosis [14]. But all in all, our results were similar to theirs.

Lehot JJ et al., in their study showed that blood sugar rise was evident in patients with both normothermic and hypothermic CPB [16]. We in our study also noted similar derangements in blood sugar levels in the entire study population irrespective of the CPB temperature. However in our study, we did not measure serum insulin level, which was also seen to rise proportionately with the rise in blood sugar levels in the normothermic group in their study [16].

We compared the mean values of ACT and platelet function in the two groups. Similar to transfusion requirements, no gross difference was found between the two groups in this regard. Although several patients in both the groups did have a deranged platelet function, the almost uniform distribution of them in the two groups suggested that it could be a result of the primary background of cyanotic heart disease. Similar studies had not compared this parameter.

Haemolysis is one of the important side effects of CPB. This not only affects the renal function of the patients but can also lead to increased morbidity. Although the individual numbers were different, this difference was not statistically significant.

The increased duration of surgery in the hypothermia group could be possibly attributed to the time required to completely re-warm the patients in the hypothermia group. None the less, a decreased duration of surgery with decreased anaesthesia time is obviously significant especially in the background of patients falling in the paediatric age group with increased hazards of general anaesthesia. In this regard the outcome of our study was in a way contrary to what was seen in the study performed by Xiong Y et al., [14].

It was a small study with a study population of 60 patients. Patients were operated by different surgeons. Besides, this was a retrospective, observational study.

From the results of our study we concluded that ICR for TOF under normothermic CPB is feasible. Normothermic CPB resulted in better patient outcomes in terms of reduced post-operative ICU and hospital stay, lesser incidence of delayed extubation and post-operative RV dysfunction and lesser transfusion requirements of FFP. There was no impact of the CPB temperature on renal function, mean blood sugar levels, coagulation parameter and post-operative drain output, post-operative fever, wound infection, arrhythmia and need for re-exploration. For us, the outcomes have implied that if the intracardiac return can be manageable by proper venting, performing ICR for TOF using normothermic CPB can be attempted and this might result in better post-operative results in certain aspects. Besides, since our study is limited by small number of subjects as well retrospective nature, further studies can be conducted in the form of randomised control trials on a larger study population which would definitely give us more convincing results and more definitive conclusions can be reached on the topic.