Back to 2015 Annual Meeting Program
National Trends in Utilization of Endoscopic Ultrasound (EUS) for Gastric Cancer: a SEER-Medicare Study
Ciara R. Huntington*1, Kendall W. Carpenter2, Yimei Han2, Jonathan Salo2, Joshua Hill2
1Department of Surgery, Carolinas Medical Center, Charlotte, NC; 2Levine Cancer Institute, Carolinas Medical Center, Charlotte, NC
PURPOSE: EUS is now recommended by the NCCN guidelines as a preferred component of staging for gastric cancer. This study examines nationwide utilization of EUS and the patient or disease characteristics associated with its use. Additionally, the association of EUS with other NCCN evidence-based treatment recommendations is determined, including peri-operative chemotherapy (Level 1) and retrieval of ≥15 lymph nodes. METHODS: A retrospective cohort study was performed using the Surveillance, Epidemiology, and End Results (SEER) and Medicare linked database. The database was queried for patients with pathologic diagnosis of gastric adenocarcinoma (SEER Site 12), who underwent resection. The cohort was stratified by preoperative EUS. Demographics, tumor characteristics, cancer treatments including peri-operative chemotherapy (defined as within four months of surgery), and pathologic node retrieval were measured. Standard statistical analyses were performed with chi square and Cochran-Armitage trend test; a two tailed p-value of <0.05 was considered significant. RESULTS: During the study period (1996-2009), 5,826 patients were diagnosed with non-metastatic gastric cancer; mean age = 77 years, 56% male, and 64% Caucasian. At surgery, 61.5% had 1-15 nodes retrieved and 31.7% had >15 nodes; 52.2% had positive nodes. Peri-operative chemotherapy was administered to 37.7% of patients (23.4% postoperative, 2.9% preoperative, 1.4% both pre-and postoperative). No peri-operative chemotherapy was administered in 72.3%. EUS utilization increased each year during the study period, with 2.6% undergoing EUS in 1996 and 22% in 2009 (p<0.001). Patients who underwent EUS (n=667,11.5%) were more often male (p<0.001), Caucasian (p<0.001), had higher education levels (p<0.001), had higher income (p<0.001), lived in large metropolitan centers (p<0.001), and had fewer comorbidities (p=0.024) than those who did not undergo EUS (Table 1). Peri-operative chemotherapy administration was more common in those who received EUS (p<0.001). Retrieval of >15 lymph nodes at resection was also more frequent in those who received EUS (p<0.001). CONCLUSIONS: The use of EUS has increased over the past decade, though still it is utilized in a minority of patients. Patient demographic characteristics differed between those receiving and not receiving EUS (Table 1). Utilization of EUS was associated with the administration of other NCCN recommended care, including administration of perioperative chemotherapy and adequate nodal retrieval, which are accepted indicators of quality of cancer care. EUS availability is a surrogate marker of a cancer care processes that adhere to nationally recommended guidelines. Table 1. Characteristics of Patients with and without EUS for Gastric Cancer | Patients with EUS (n=667) | Patients without EUS (n=5159) | p value | Male gender | 65.8% | 54.9% | <0.0001 | Caucasian race | 76.9% | 62.3% | <0.0001 | \"Localized\" stage | 49.2% | 39.8% | <0.0001 | \"Regional\" stage | 50.8% | 60.2% | <0.0001 | Big Metro | 66.7% | 59.8% | <0.0001 | Metro/Urban | 26.4% | 33.3% | <0.0001 | Suburban | 5.4% | 5.8% | <0.0001 | Rural | 1.5% | 1.2% | <0.0001 | Charlson Comorbidity Index (CCI) | | | | 0 | 50.8% | 45.2% | 0.024 | 1 | 27.9% | 28.8% | 0.024 | 2 | 11.5% | 14.3% | 0.024 | ≥3 | 9.8% | 11.8% | 0.024 | Income Quartiles | | | | Lowest Quartile | 12.0% | 23.8% | <0.0001 | Highest Quartile | 38.0% | 26.2% | <0.0001 | Education Quartiles | | | | Lowest Quartile | 14.4% | 26.0% | <0.0001 | Highest Quartile | 39.0% | 25.5% | <0.0001 | Nodes examined | | | | 0 | 6.0% | 7.0% | <0.0001 | 1-15 | 52.4% | 62.6% | <0.0001 | >15 | 41.6% | 30.4% | <0.0001 | Positive nodes | 42.1% | 53.4% | <0.0001 | Chemotherapy | | | | None | 63.9% | 73.4% | <0.0001 | Postoperative only | 18.1% | 24.1% | <0.0001 | Preoperative only | 12.3% | 1.7% | <0.0001 | Pre and postoperative chemotherapy | 5.7% | 0.8% | <0.0001 |
Back to 2015 Annual Meeting Program
|