Patient Factors Affecting Inpatient Mortality Following Surgery for Colorectal Cancer
Swati Sonal1, Chloe Boudreau1, Hiroko Kunitake1, Robert Goldstone1, Liliana G. Bordeianou1, Christy Cauley1, Rocco Ricciardi1, Todd Francone2, David L. Berger1
1General and GI Surgery, Massachusetts General Hospital, Boston, Massachusetts, United States, 2Colon and Rectal Surgery, Newton-Wellesley Hospital, Boston, Massachusetts, United States
Objective
Perioperative risk factors leading to inpatient mortality in colorectal cancer surgery are understudied. Our objective is to identify these risk factors for inpatient deaths in patients undergoing resection for colorectal cancer.
Design
Unmatched 1:3 case-control study from an IRB approved database of all surgically resected colorectal cancers at a tertiary care institution.
Setting
Patients operated on for colorectal cancer at a tertiary institution between 2004 and 2021 with an average follow-up duration of 31.6 months.
Patients
A total of 160 patients were included in this analysis (51.9% females). All patients who had inpatient deaths during their primary admission or readmission within 30-days of discharge of colorectal cancer surgeries were included (N=40). Simple random sampling was then used to select 120 case controls from the database containing Stage I-IV colorectal cancer patients who underwent surgical resection without mortality. Patients who underwent only transanal excisions were excluded.
Interventions
NA
Outcome measure
Inpatient mortality
Results
The patient cohort who died were older, had lower body mass index (BMI), higher Charlson Comorbidity Index, higher rates of Stage IV disease, preoperative anemia, preoperative leukocytosis, preoperative hypoalbuminemia, emergency surgeries, and open operative approach (p<0.05); as compared to the patient cohort who underwent surgical resection without inpatient mortality. Multivariate logistic regression model (Concordance index 0.92) demonstrated that non-metastatic disease presentation (OR 0.19; 0.04-0.79), emergency surgery (OR 5.07;1.37-20.34), preoperative anemia (OR 8.5, 1.56-73.89), perioperative blood transfusion (OR 4.39, 1.27-16.22) and postoperative ICU admission (OR 8.46, 1.98-42.54) significantly predicted inpatient mortality when controlled for age, gender, BMI, comorbidities, neoadjuvant therapy, operative approach and postoperative delirium. The backward elimination technique retained only emergency surgery (OR 5.01, 1.49-16.83) and perioperative blood transfusion (OR 5.66, 1.87-17.15) in the final model.
Conclusions
Surprisingly, it appears that perioperative factors are more important in predicting inpatient mortality of patients undergoing colorectal cancer surgery than baseline co-morbidity. Emergency surgery and perioperative blood transfusion seem to be the most critical factors determining inpatient mortality.
Back to 2022 Posters