|
|
Back to 2016 Annual Meeting
Can Non-Home Discharge Following Major Hepatectomy Be Predicted Using Preoperatively Determined Factors?
Megan V. Beems*1, Christina W. Lee1, Adam Brinkman2,3, Michael E. Egger3, Malcolm H. Squires4, David Kooby4, Shishir Maithel4, Clifford S. Cho1, Emily R. Winslow1, Robert C. Martin5, Kelly McMasters5, Charles Scoggins5, Sharon M. Weber1
1Department of Surgery, University of Wisconsin, Madison, WI; 2Department of Pediatric Surgery, Vanderbilt University, Nashville, TN; 3MD Anderson Cancer Center, The University of Texas, Houston, TX; 4Department of Surgery, Emory University, Atlanta, GA; 5Department of Surgery, University of Louisville, Louisville, KY
Introduction:
Hepatic resection is the treatment of choice for a wide variety of benign and malignant lesions. Recent studies have suggested that an increasing proportion of hepatectomy patients will require medical assistance after discharge.
Methods:
A retrospective review involving 70 preoperative, intraoperative, and postoperative variables was performed on 1239 patients who had undergone major hepatectomy at three academic centers from 2000 - 2012. Chi-square and independent sample T-test analyses were performed. Significant preoperative variables with rare events, large proportion of missing data, and/or collinearity were excluded; all remaining variables were then entered into a bivariate logistic regression analysis using backward stepwise likelihood ratio estimation. A random half of the dataset was used to design a formula predicting the probability of non-home discharge, and the remaining half to validate it. The entire dataset was then used to create a final formula.
Results:
Of the 1239 patients undergoing hepatectomy, 4.3% were discharged to sub-acute rehabilitation centers (3.4%), nursing facilities (0.7%), or hospice (0.2%). On univariate analysis, 10 preoperative variables were significantly associated with non-home discharge: Medicare/Medicaid insurance, increasing age, lower or higher BMI category, poorer Eastern Cooperative Oncology Group performance status, higher American Society of Anesthesiologists (ASA) risk classification, cancer, cardiac disease, and specific abnormal laboratory values (albumin, INR, and BUN). After excluding variables due to missing data or significant collinearity, multivariate analysis was performed using age, albumin, ASA class, cardiac disease, and cancer. Independent predictors of non-home discharge included higher ASA class, increasing age (continuous variable), and lower albumin level (continuous variable). Ageā„70 was associated with a 14% risk of non-home discharge, albumin<3.0 with an 11.7% risk, and ASA class of 4 with a 17% risk. Patients with all of these criteria were at highest risk of non-home discharge (22.2%, p<0.001), while patients with none of these criteria were at lowest risk (0%, p<0.001) (Figure 1). A formula to predict non-home discharge was derived and had an AUC of 0.843 (Figure 2).
Conclusion:
Higher ASA class, increasing age, and lower preoperative albumin levels are associated with non-home discharge after major hepatic resection. Our risk calculator has strong predictive ability for identifying patients who are less likely to discharge home, which may impact disposition planning. Prospective evaluation of this tool is necessary to assess its role in decreasing length of stay and hospital costs.
 
Back to 2016 Annual Meeting
|
