Discussion

Bidirectional cavopulmonary anastomosis has been adopted as part of a staged strategy for total cavopulmonary connection procedures since the late 1980s.^{Reference Norwood and Jacobs5} Overall in-hospital mortality varies between reports and ranges from 0.7 to 20%.^{Reference Manuel, Morais and Turquetto6–Reference Nichay, Gorbatykh and Kornilov9} Although some authors advocate that this procedure be performed between 3 and 6 months of age,^{Reference Tanoue, Kado and Boku1,Reference Schidlow, Gauvreau and Cherian10} patients with hypoplastic left heart syndrome (HLHS) and those from high-income countries undergo procedures earlier,^{Reference Schidlow, Gauvreau and Cherian10} and patients in developing countries may be submitted to this procedure later.^{Reference Manuel, Morais and Turquetto6,Reference Silvilairat, Pongprot, Sittiwangkul, Woragidpoonpol, Chuaratanaphong and Nawarawong11} In our cohort, the median age was 12 months, and the few cases with HLHS and lack of operating rooms may partly explain these data. Developed countries show mortality rates from 0.7 to 11%,^{Reference Tanoue, Kado and Boku1,Reference Kogon, Plattner, Leong, Simsic, Kirshbom and Kanter7,Reference François, Vandekerckhove, De Groote, Panzer, De Wolf and De Wilde8} while low- and middle-income countries show mortality rates from 6 to 24%,^{Reference Manuel, Morais and Turquetto6,Reference Schidlow, Gauvreau and Cherian10,Reference Silvilairat, Pongprot, Sittiwangkul, Woragidpoonpol, Chuaratanaphong and Nawarawong11} depending on the casuistics. Therefore, our mortality rate (13.4%) would be somewhere in between. Our cohort presented some similarities to other developing countries, as older age at operation and mainly right single ventricles. Low weight at surgery and longer cardiopulmonary bypass time were associated with increased chance of death or longer time of mechanical ventilation in this cohort of patients. These outcomes were unaffected by ventricular morphology, previous cardiac interventions, and the presence of left superior caval vein.

The presence of bilateral superior caval vein was not a risk factor in our analysis, as shown by other authors.^{Reference Kogon, Plattner, Leong, Simsic, Kirshbom and Kanter7} As this cohort had high incidence of complications, we could not identify a single complication as a factor for mechanical ventilation for over 24 hours.^{Reference Kogon, Plattner, Leong, Simsic, Kirshbom and Kanter7} A longer cardiopulmonary bypass time may be related to technical difficulties or to the need for pulmonary arterioplasty, which may impact the outcome of the procedure.^{Reference Jonas, Ramalho, Vicente and Manso12,Reference Cleveland, Tran, Takao, Wells, Starnes and Kumar13}

Because positive inspiratory pressure is a burden to these patients, and given that other authors showed a median ventilation time ranging from 4 to 16 hours,^{Reference François, Vandekerckhove, De Groote, Panzer, De Wolf and De Wilde8,Reference Nichay, Gorbatykh and Kornilov9,Reference Cleveland, Tran, Takao, Wells, Starnes and Kumar13} we arbitrarily chose a mechanical ventilation time above 24 hours as a marker of unfavourable outcome. There was no association between prolonged hospital stay and death which is why it was not used as an outcome.

Our patients may present low weight due to several factors, such as long-term hypoxaemia, heart failure, over-circulation, or social problems. However, the registry data preclude us from exploring which conditions would prevail in this cohort.

Even though the initial lactate level was usually associated with death, as it may be a hypoxaemia or inadequate oxygen delivery marker, we could not find an association between lactate levels and mechanical ventilation time. In a series of Fontan patients, lactate levels were not related to the outcome.^{Reference Hamamoto, Uemura, Imanaka and Yagihara14}