Contrasting Shamblin classification with tumor volume in predicting morbidity and mortality in carotid body tumor...

The Shamblin classification is the standard for predicting carotid body tumor (CBT) surgical outcomes; however, its subjective nature limits prognostic accuracy. This study compared the predictive value of preoperative tumor volume vs Shamblin classification for perioperative morbidity and mortality in CBT resection at a tertiary center.

A single-center retrospective cohort study analyzed data from 24 consecutive CBT resections performed at an Academic Hospital (January 2015-2025). Tumor volumes were calculated from preoperative contrast-enhanced computed tomography using three-dimensional volumetric reconstruction. Spearman's rank correlation, Mann-Whitney U test, and Fisher exact test were used; each correlation and proportion is reported with its 95% confidence interval (CI). Discriminative performance was assessed using sensitivity, specificity, positive predictive value, negative predictive value, the area under the receiver-operating-characteristic curve (AUC) with 95% CIs, and a DeLong paired test comparing the volume and Shamblin AUCs, and the F-statistic.

The cohort comprised 19 women (79%) and 5 men, with a mean age of 52 years. Ten tumors (42%) were Shamblin II, and 14 (58%) were Shamblin III; no Shamblin I tumors were identified. Median tumor volume was significantly larger in Shamblin III (63 cm³; interquartile range, 46-120) than Shamblin II (12 cm³; interquartile range, 9-18; P < .001). Tumor volume correlated with estimated blood loss (r = 0.637; 95% CI, 0.30-0.83) and operative time (r = 0.475; 95% CI, 0.09-0.74); the correlation with intensive care unit stay was weak and nonsignificant (r = 0.29; 95% CI, -0.13 to 0.62; P = .17). For predicting blood loss, tumor volume showed a numerically higher AUC (0.90; 95% CI, 0.76-1.00) than the Shamblin classification (0.83; 95% CI, 0.67-1.00); the two AUCs did not differ significantly (DeLong P = .26). The negative predictive value for blood loss was high for both (Shamblin 0.90; volume 0.75). All 14 patients with Shamblin III developed complications, compared with 60% of Shamblin II (P = .020). The iatrogenic injury rate was 20.8% (95% CI, 9.2-40.5). One 30-day death (4%) occurred in a patient with the cohort's largest tumor (224 cm³, Shamblin III) who died from a major intraoperative stroke.

Tumor volume correlated more strongly and more reproducibly with surgical outcomes than the Shamblin classification and may serve as an objective, continuous adjunct for risk stratification; the two systems are best regarded as complementary. A volume threshold of >73 cm³ warrants further validation as a marker of increased mortality risk, but with only one death this threshold is hypothesis-generating.