Prolonged open sternotomy and delayed sternal closure after cardiac operations

Abstract

Maintenance of an open sternotomy (OS) after a complicated cardiac operation is an adjunct in the treatment of the severely impaired heart. We hypothesized that predictors of the timing, morbidity, and prognosis of delayed sternal closure (DSC) could be determined by intensive case review. Prolonged OS was used in 107 of 6,030 adult open heart patients (1.8%) between 1987 and 1991. Indications for OS were hemodynamic instability (40), myocardial edema (18), intractable bleeding (23), relentless arrhythmias (9), and ventricular assist devices (17). Delayed sternal closure was carried out in 75 of 107 patients at a mean of 3.4 ± 0.3 days after OS. Fifty of these 75 (67%) survived and were discharged an average of 43 ± 6 days after closure. Fifty-seven patients died: 32 before DSC at 3.7 ± 0.8 days after OS and 25 after DSC at 27 ± 4 days after OS. Baseline cardiac index (1.7 ± 0.1 L · min −1 · m −2) improved an average of 1.0 ± 0.1 L · min −1 · m −2 after OS ( p ≤ 0.001) and remained stable through DSC (2.5 ± 0.3 L · min −1 · m −2) and late (9 ± 0.7 days) follow-up (2.8 ± 0.1 L · min −1 · m −2). Delayed sternal closure in patients without ventricular assist devices was significantly more likely to be successful ( 45 / 63 versus 9 / 27 ; p < 0.002) when carried out after the onset of a negative daily fluid balance. Sternal infection occurred in 4 of 75 (5%) patients after DSC and was associated with bleeding as an indication for OS ( 3 / 15 versus soll / 60 ; p < 0.02). Predictors of mortality after OS by multivariate analysis were renal insufficiency (serum creatinine level >265 μmol/L [3 mg/dL]) and serious ventricular arrhythmias ( p < 0.01 for both). This study shows that OS and DSC can now be performed with a more complete understanding of the benefits, timing, and risks of these potentially lifesaving maneuvers.