Risk factors for perioperative mortality and transfusion in sacrococcygeal teratoma resections

Sacrococcygeal teratomas are a common congenital tumor. Surgical resection can occur in utero, in the neonatal period, or in the postneonatal period.

We describe patient and tumor factors associated with mortality and transfusion in this population.

We did a retrospective chart review of patients who underwent sacrococcygeal teratoma resection between January 1998 and March 2016. Demographic data, transfusion data, and tumor characteristics were collected. Descriptive statistics were calculated, and univariate comparisons were performed with chi-square test and Fisher's exact test. Variables significant at univariate level were used in multivariate logistic regression and negative binomial regression.

Of the 112 cases, 6 were in utero repairs, 73 were neonatal repairs, and 33 were repairs at >30 days of life. There was 17%, 1%, and 0% intraoperative mortality and 33%, 5%, and 0% 30-day mortality in the in utero, neonatal, and >30 days of life repairs, respectively. All six patients who died within the first 30 days of life had a postmenstrual age of <32 weeks at time of surgery. All six patients who died had noncystic tumors. Patients with noncystic tumors were more likely to be born prior to 30-week gestation (23/65 vs 6/47; χ² = 7.3; P = 0.007). Gestational age >30 weeks was associated with decreased intraoperative death (0% vs 10%; modified maximum likelihood estimate of OR 0.05; 95% CI 0.002-0.96; P = 0.02). Gestational age >30 weeks (2.4% vs 13.8%; OR 0.15; 95% CI 0.03-0.89; P = 0.04) and cystic morphology (0% vs 9.2%; modified maximum likelihood estimate of OR 0.1; CI 0.01-1.75; P = 0.04) were associated with decreased 30-day mortality and emergent surgery (17.9% vs 1.2%; OR 18; 95% CI 2-162.2; P = 0.004) was associated with increased 30-day mortality. Gestational age >30 weeks (33.7% vs 62.1%; OR 0.27; 95% CI 0.09-0.79; P = 0.02) and Altman class 3-4 (12.1% vs 52.7%; OR 0.1; 95% CI 0.03-0.34; P = 0.0002) were associated with decreased need for transfusion and noncystic tumor was associated with increased transfusion volume (131.6 ml·kg^-1 [95% CI 94-184] vs 63 ml·kg^-1 [95% CI 40-100.1]; P = 0.01).

Prematurity is associated with increased intraoperative and 30-day mortality. Noncystic tumor morphology was the only significant factor associated with transfusion volume and all six patients who died had transfusion volumes of 240 ml·kg^-1 or greater. In these patients at high risk of mortality due to blood loss, the anesthesia team should be prepared to manage massive transfusion and coagulopathy with blood components and pharmacologic measures.