O. Tabatabaie1, G. G. Kasumova1, T. S. Kent2, M. F. Eskander1, A. Fadayomi1, S. Ng1, J. F. Critchlow2, N. E. Tawa3, J. F. Tseng1 1Beth Israel Deconess Medical Center,Surgical Outcomes Analysis & Research (SOAR),Boston, MA, USA 2Beth Israel Deaconess Medical Center,Department Of Surgery,Boston, MA, USA 3Beth Israel Deaconess Medical Center,Division Of Surgical Oncology,Boston, MA, USA
Introduction:
Totally implantable venous access devices (ports) are widely used for long-term central venous access, especially for cancer chemotherapy. Upper extremity DVT (U-DVT) is a reported complication of ports, however, prophylaxis remains controversial due to low event rates in the general population. The aim of this study was to determine the risk factors of U-DVT to help identify patients at increased risk who could potentially benefit from prophylaxis.
Methods:
Healthcare Cost and Utilization Project’s Florida State Ambulatory Surgery and Services Database (SASD) was queried between 2007-2011 for patients who underwent outpatient port insertion by CPT code. Patients were followed in the SASD, State Inpatient Database (SID) and State Emergency Department Database (SEDD) for U-DVT occurrence. The cohort was divided into a test cohort and a validation cohort based on the port placement time (2009-2011 and 2007-2008 for test and validation cohorts; respectively). A multivariable logistic regression model was developed to identify risk factors for U-DVT in patients with a port. The model was then tested on the validation cohort.
Results:
Of the 51,049 patients identified in the test cohort, 926 (1.81%) had at least one U-DVT coded at a follow-up visit. The mean age of the test cohort was 62.3 (SD=13.2) and there was a slight female predominance (61.94%). The median time of U-DVT development after port placement was 133 days (IQR: 47-297). Patients who had a U-DVT were more likely to be younger (61.2 vs 62.6), black (vs white, OR=1.9), a smoker (OR=1.29) or have an Elixhauser score of 1-2 (vs. 0; OR=1.22). They also had increased odds of having a history of hypercoagubility (OR=7.68), catheter-related complication at the time of port placement (OR=3.96), autoimmune disease, (OR=1.72), or a non-cancer indication for port placement (vs. genitourinary cancers; OR=2.74). Other univariate predictors were Medicaid insurance (vs. private insurance; OR=1.42), all-cause 30-day readmission (OR=2.44), previous DVTs (OR= 1.95) and end-stage renal disease (OR=4.94). All of the univariate predictors had a P-values<0.05. On multivariate analysis, age>65, black race, 30-day readmission, hypercoagubility, ESRD, indication for port placement and catheter complication were independent predictors of U-DVT (see Figure for details). C-statistics of the model for the test and validation cohorts were 0.68 and 0.66; respectively.
Conclusion:
Our model can be used to identify patients at increased risk of U-DVT after port insertion. Utility of DVT prophylaxis should be investigated in this group of patients in future prospective trials.
