Real-World Analysis of the Duration of Risk of Venous Thromboembolism In US Surgical Patients

Patients hospitalized for major surgery are at significant risk of developing venous thromboembolism (VTE). In the US, the length of acute hospitalization for surgery has been decreasing. Given the decreased length of hospital stays, the time during which surgical patients are at risk for VTE is not well-understood in real-world clinical practice. Such information could provide insights into whether current recommendations regarding VTE prophylaxis reflect the real world needs. This retrospective, observational study assessed the incidence and time course of VTE events following admission to the hospital for major surgery in a large, real-world patient population.

This study employs administrative claims data derived from the Thomson Reuters MarketScan^®Inpatient Drug Link File, which comprises longitudinal patient-level inpatient and outpatient medical and pharmaceutical claims data for individuals with employer-sponsored primary or Medicare supplemental health insurance. Patients at risk for VTE due to a hospital admission between January 1, 2005, and December 31, 2008, for either abdominal or orthopedic surgery were included in this study (obtained through charge codes for types of surgical procedures that reflect the American College of Chest Physicians guideline risk groups for VTE). Patients were included if they had been continuously enrolled for at least 12 months prior to admission and at least 180 days after admission. Additionally, patients included in the analysis were required to have received some form of pharmacological prophylaxis during their hospitalization. Appropriateness of prophylaxis was not determined. The cumulative risk and hazard of VTE were established across an evaluation period of 180 days. The distribution of cumulative risk was described by the Kaplan Meier product limit method of survival analysis. The distribution of the hazard of VTE was described by a locally weighted regression procedure. The hazard function is measured as the number of VTE events per 1,000 person-days by number of days after hospital admission date to the first occurrence of VTE.

The study cohort consisted of 6,445 surgical patients who received prophylaxis, of whom 3,726 (57.8%) were admitted for orthopedic surgery and 2,719 (42.2%) for abdominal surgery. Patients had a mean (standard deviation [SD]) age of 64.0 (13.6) years and 63.8% were female. The mean (SD) length of stay in hospital was 5.4 (6.1) days, during which all patients received prophylaxis for VTE (criteria for inclusion) with a mean (SD) duration of 4.7 (4.3) days. 30% of patients received anticoagulation therapy within the period extending from discharge to 35 days after discharge. A total of 243 VTE events (171 DVT and 96 PE) occurred during the 180-day evaluation period. 104 (43%) VTE events occurred during the index hospitalization, and 139 (57%) VTE events occurred post-discharge. The time course of VTE events showed that the highest number of events occurred during the first 9 days (66 events, 80% in-hospital; proportion of 180-day cumulative risk ∼22%) and during days 10–19 (55 events, 58% in-hospital; proportion of 180-day cumulative risk ∼50%) following admission for the index hospitalization. VTE hazard peaked at approximately 1.4 per 1,000 person-days on the 9^th day following admission for the index hospitalization. By the 21^st day following index admission 50% of the cumulative 180-day risk had been incurred. The frequency of VTE then further decreased during days 20–29 (21 events, 38% in-hospital; proportion of 180-day cumulative risk ∼58%) and gradually declined thereafter, fluctuating at a background level of 1–7 events during each 10-day interval from 70–79 days up to 170–180 days.

Over a 180-day period after admission to the hospital 243 (3.8%) out of 6,455 at-risk surgery patients receiving pharmacological prophylaxis experienced an incident VTE event. Of these VTE events, 43% occurred during the index hospitalization, while 57% occurred post-discharge. Of all VTE events that occurred within the 180-day period, a quarter occurred within 10 days and half within 3 weeks. Results from this study show that the risk of VTE was highest within the first 19 days after admission to the index hospitalization. During this high-risk period, 30% of VTE events occurred after discharge, suggesting that a non-trivial risk of VTE extends into the period after discharge.

Amin:sanofi-aventis US Inc.: Acknowledgment: This study was funded by sanofi-aventis U.S., Inc. The authors received editorial/writing support provided by Tessa Hartog, PhD of Excerpta Medica, in the preparation of this poster funded by sanofi-aventis U.S., Inc., Honoraria, Speakers Bureau. Princic:sanofi-aventis US Inc.: Research Funding. Lin:sanofi-aventis US Inc.: Employment, Research Funding. Thompson:sanofi-aventis US Inc.: Employment. Johnston:sanofi-aventis US Inc.: Research Funding.