Liberal Use of Delayed Sternal Closure in Children Is Not Associated With Increased Morbidity

Delayed sternal closure (DSC) is often employed to optimize hemodynamics following pediatric cardiac surgery. Prior reports have suggested that DSC may be associated with increased morbidity. We sought to analyze the impact of a liberal policy of DSC on surgical outcomes at our center.

We retrospectively evaluated the clinical course of 1,000 consecutive patients between July 2005 and June 2015 whose sternum was electively left open following pediatric cardiac surgery. Data are presented as mean and standard error (parametric) or median and quartiles (nonparametric). Receiver-operating characteristic curve analysis was undertaken to identify significant points of inflection. A p less than 0.05 was considered significant.

An a priori decision to leave the sternum open is made when complex surgery, especially in neonates and usually involving circulatory arrest, is expected to result in postoperative hemodynamic instability. Age at index surgery for the 1,000 patients was 7 (interquartile range [IQR], 3 to 19) days and weight 3.3 (IQR, 2.8 to 3.7) kg. There were 816 (82%) neonates and 569 (57%) boys. Index operations included 332 (33%) Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery (STAT) category 5, 483 (48%) STAT category 4, and 185 (19%) STAT category 3 procedures. A total of 103 (10%) patients required postoperative extracorporeal support. Following hemodynamic recovery, DSC was undertaken 3 (IQR, 2 to 4) days postoperatively and in 98.3% patients was performed in the intensive care unit. Overall, mortality was 6.3% and major Society of Thoracic Surgeons morbidity was 21.6%. There were 42 (4.2%) positive mediastinal surveillance cultures at the time of DSC, with the most common organism being coagulase-negative staphylococcus. Fifty-nine (5.9%) clinical sternal and mediastinal wound infections and a total of 117 infectious complications were encountered in 94 patients. Using Society of Thoracic Surgeons database outcome as benchmark, mortality and length of stay in our patients were comparable when analyzed by STAT categories or for the 2 most common index procedures (eg, Norwood and arterial switch operations). Receiver-operating characteristic curve analysis showed that 5 days of open sternum had a weak, but statistically significant, correlation with incidence of infectious complications (area under the curve, 0.56; p = 0.002). The need for DSC 5 or more days after the index procedure was observed in 177 (18%) patients and was not associated with increased wound infection. It was, however, independently associated on multiple regression analysis with major morbidity (odds ratio, 1.7; 95% confidence interval, 1.2 to 2.5; p = 0.002) and, in the subset of 897 patients who did not require extracorporeal support, with increased mortality (odds ratio, 2.2; 95% confidence interval, 1.3 to 3.6; p = 0.003).

A liberal policy of DSC does not adversely affect surgical outcomes, including infectious complications and length of stay. We submit that need for DSC should not, by itself, be considered a source of morbidity.