Page 58 - Risk quantification and modification in older patients with colorectal cancer
P. 58
Chapter 3
date of surgery to time of death. Follow-up of all patients was at least 183 days. The regional ethics committee and institutional review board of both hospitals approved this study.
Statistical analysis
Patients were classified as ‘screened’ if a G8 and/or ISAR-HP screening was performed prior to surgery. We performed descriptive analysis of patient’s characteristics for both screened and non-screened patients and for the best performing screening tool. Normally distributed variables are presented as a mean with standard deviation (SD) and for non-normal distributed as a median with the interquartile range (IQR, 25th-75th percentile). The chi-square test (χ2) was used to compare ordinal variables and the Mann-Whitney U test or unpaired t-test for continues variables. Odds ratio (OR) was used as a measure for the association between ISAR-HP and G8 screening tool and primary and secondary outcomes. An OR is expressed with a 95% confidence interval. A p-value ≤ 0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 17.0 (SPSS, Inc., Chicago, IL, USA).
Results
A total of 268 patients aged >70 years, with colorectal cancer were identified. After exclusion of patients with emergency surgery (n=37), Transanal Endoscopic Microsurgery (n=4), stage IV disease (n=7) and synchronous cancer at time of diagnosis (n=6), a total of 214 patients were included. Of the latter, 139 patients (65%) were screened prior to surgery. From two out of these 139 patients, only an ISAR-HP screening was available.
Seventy-nine patients (57%) had a partial or hemicolectomy, 55 (40%) a low- anterior resection, three patients (2%) had an abdominoperineal resection and two patients (1%) a subtotal colectomy.
Baseline characteristics all screened patients are depicted in Table 1. Median age of screened patients was 77.7 years (IQR 75.0-82.8), 29% of patients were classified as ASA III or IV and 35% had a CCI score ≥2. Analysis of the non-screened patients of the total cohort showed no significant differences between screened and non-
56