Is development of postoperative venous thromboembolism related to thromboprophylaxis use? A case-control study in the Veterans Health Administration.

BACKGROUND Observational studies continue to report thromboprophylaxis underuse for postoperative pulmonary embolism/deep vein thrombosis (pPE/DVT) despite the long-standing existence of prevention guidelines. However, data are limited on whether thromboprophylaxis use differs between patients developing pPE/DVT versus those who do not or on why prophylaxis is withheld. METHODS Administrative data (2002-2007) from 28 Veterans Health Administration hospitals were screened for discharges with (1) pPE/DVT as flagged by the Agency for Healthcare Research and Quality (AHRQ) Patient Safety Indicator software and (2) pharmacoprophylaxis-recommended procedures, and the medical records were reviewed to ascertain true pPE/DVT cases. Controls were selected by matching cases by hospital, age, sex, diagnosis-related group, and predicted probability for developing pPE/DVT, and who underwent a pharmacoprophylaxis-recommended procedure. Records were assessed for "appropriate pharmacoprophylaxis use," defined primarily per American College of Chest Physicians (ACCP) guidelines, and reasons for anticoagulant nonuse. RESULTS The 116 case-control pairs were similar in terms of demographics, surgery type, ACCP risk category, and appropriate pharmacoprophylaxis rates overall. Of the highest-risk patients, respective pharmacoprophylaxis rates among cases and controls were 88% versus 92% among hip/knee replacements and 31% versus 48% among cancer patients. Of the cases and controls who did not receive appropriate pharmacoprophylaxis, only about 25% had documented contraindications. Reviewers identified contraindications in 14% of cases and 9% of controls. CONCLUSIONS Similarities in preventive pPE/DVT practice between cases and controls suggest that pPE/DVTs occur despite implementation of guideline-adherent practices.