Outcomes From Root Cause Analysis of Hospital-Associated Thrombosis Over Twelve Months: A Single Centre Experience

Abstract 675

The national Venous Thromboembolism (VTE) Prevention Programme in England incorporates standardised guidance on risk assessment (RA) and thromboprophylaxis (TP) with a requirement for root cause analysis of all episodes of hospital associated thrombosis (HAT), defined as any VTE occurring whilst an inpatient or within 90 days of discharge. We report findings from audit of root cause analysis of HAT over 12 months at King's, a major London tertiary centre with 900 beds and an estimated 53 000 admissions per annum. 239 episodes of HAT were identified associated with surgical, medical and obstetric admission accounting for 101 (42.3%), 133 (55.6%) and 5 (2.1%) cases respectively. The estimated incidence of HAT is 4.5 per 1000 admissions. The mean age of patients with HAT was 62.7 (+/− 16.8) years, with males accounting for 53.6% of the cohort. HAT manifested as deep vein thrombosis in 121 (50.6%) and pulmonary embolism in 128 (49.6%). The median time to diagnosis of HAT following admission was 16 days (IQR 7–30). 171 (67.4%) of HAT occurred prior to discharge. Of the 78 (31.7%) events occurring post discharge, 60 (76.9%) required readmission for management of HAT and 2 (2.6%) represented with fatal events. HAT was associated with mortality in 51 cases (21.3%), with death directly attributable to PE in 16 (6.7%). The estimated incidence of HAT associated with fatal PE is 0.3 per 1000 admissions. Of note, autopsy was undertaken in 12/51 with PE identified as the primary cause of death in 11/12 (2/12 had known/suspected VTE prior to death). The five remaining patients with PE as the primary cause of death had the diagnosis made on clinical suspicion alone in four cases and radiological imaging in one case. The remaining deaths were attributed to other causes in 25, with 15 having unknown cause of death as certification occurred in the community (10 with known advanced malignancy at discharge). Root cause analysis has been completed for 149 (62.3%) of HAT episodes. Of these, 43.9% had RA undertaken on admission to hospital. Retrospective RA revealed 91.0% of patients were at high risk for VTE with 33.1% also at high risk of bleeding. 72% were prescribed anticoagulant TP. Anticoagulant prophylaxis was prescribed for 30/49 (61.2%) medical, 33/36 (91.7%) surgical and 3/4 (75%) obstetric HAT cases with a high VTE risk and low bleeding risk. Of those with a high bleeding risk, 8/23 (34.8%) and 15/27 (55.6%) medical and surgical patients respectively received anticoagulant TP for part of their admission. Mechanical TP was prescribed for 41/63 (65.1%) surgical, all obstetric (4) HAT cases and 5/15 (33.3%) medical patients in whom mechanical TP was indicated and appropriate. HAT was attributed to inadequate TP in 51 (32.5%), contraindication to chemical TP in 23 (14.6%), contraindication to all TP in 11 (7.0), TP failure in 43 (27.4%), line associated in 20 (12.7%), and was considered unexpected in 9 (5.7%) patients without any risk factors for VTE. Inadequate TP resulted from failure to prescribe in 17 (33.3%), unexplained delay in initiation in 8 (15.7%), unexplained missed doses in 7 (13.7%), inadequate duration of TP in 5 (9.8%) or inferior agent or dose in 9 (17.6%) cases with a combination of the above in 5 (9.8%) cases. Mortality associated with inadequate TP was 21.6% with death directly attributable to PE of 5.9%. TP remains underused in cases of HAT, with lowest rates associated with medical admission. At our centre, improved RA and TP could reduce the annual incidence of HAT by an estimated 21%. Further research is required to improve risk assessment and thromboprophylactic strategies to address unexpected events and those arising despite optimal TP.