Anzeige
Erschienen in:
17.03.2017 | Gastrointestinal Oncology
External Validation of the Simplified Preoperative Assessment for Low-Grade Mucinous Adenocarcinoma of the Appendix
Erschienen in: Annals of Surgical Oncology | Ausgabe 7/2017
Einloggen, um Zugang zu erhaltenAbstract
Background
MD Anderson Cancer Center developed a computed tomography (CT)-based preoperative assessment tool simplified preoperative assessment for appendix tumor (SPAAT) for predicting incomplete cytoreduction (IC) in low-grade mucinous adenocarcinoma (LGMA) of the appendix, based on preoperative CT scans. This study independently evaluates the tool’s performance.
Methods
Seventy-six preoperative CT scans of LGMA patients were evaluated by two surgeons unfamiliar with the patients’ medical history. Scores were assigned based on SPAAT criteria, with a SPAAT ≥3 predictive of IC. Binary regression analyses and area under the receiver operating characteristic (AUROC) curve analyses were performed. Patients with splenic resection were excluded due to the structure of the SPAAT assessment tool.
Results
Seventy-six LGMA patients underwent attempted cytoreductive surgery with hyperthermic intraperitoneal chemotherapy (CRS/HIPEC). Of 68 patients, 58 had complete cytoreduction and 10 had IC; 8 patients were ineligible due to prior splenectomy. The mean SPAAT score was 0.8, with six patients having SPAAT scores ≥3. SPAAT scores ≥3 were predictive of IC, with a hazard ratio (HR) of 19 (95% confidence interval 2.8–124.1) (p = 0.002). Sensitivity, specificity, positive predictive value (PPV), and negative predictive value were 40, 97, 50, and 90%, respectively. A SPAAT score ≥3 was not associated with worse survival prognosis. Median follow-up was 2.4 years and AUROC curve was 71%. SPAAT components with respective HR and p-values were foreshortening of the bowel mesentery (29.5; p = 0.004), and scalloping of the pancreas (9; p = 0.008), spleen (4.3; p = 0.04), portal vein (3.1; p = 0.4), and liver (2.1; p = 0.3).
Conclusion
A SPAAT score ≥3 predicted IC based on a binary regression model. The clinical value of this score is controversial due to low sensitivity and PPV.