Abstract
Purpose: To describe the clinical features and outcomes of pts. with HIT over a 6-year period from a tertiary care medical center.
Design: Retrospective case series of 545 pts. enrolled in an IRB-approved HIT Registry from 1/1/99 to 12/31/05.
Measurements: demographics, co-morbid conditions, HIT presentations, platelet ct./HIT antibody results, treatment, hospital length of stay (LOS), composite outcome, and all-cause mortality.
Results: Median age-68 yrs.; 47%-females; 97%-caucasians. Co-morbid conditions: HTN (78%), coronary artery disease (73%), diabetes mellitus (42%), CHF (16%), ESRD on dialysis (5.7%) and active malignancy (4.2%). Clinical HIT settings included: adult open heart surgery (OHS)-69%, medical-23% and non-cardiac surgery-8%. Cardiac medical pts. comprised 50% of the medical cohort. Among the OHS, medical and non-cardiac surgery HIT cohorts, HIT occurred during the hospitalization with UFH exposure in 322 (59%), 125 (23%), and 39 (7.2%) pts., respectively. In these respective clinical HIT settings, delayed-onset HIT (D-HIT) developed after discharge from a hospitalization with UFH exposure in 54 (10%), 2 (0.36%), and 3 (0.55%) pts. During the study period, HIT was diagnosed in 2.5% of 15,152 OHS pts. The median time from UFH initiation to the time when HIT was first clinically suspected and diagnosed was 7 days and 10 days, respectively. The median time from UFH discontinuation to the time HIT was first clinically suspected and diagnosed was 1 and 4 days, respectively. Of 537 pts. in whom an H-PF4 ELISA (GTI) or HIPA assay or both were performed, 85% had at least one positive result. The median platelet ct. at baseline (last platelet ct. at hospital D/C for D-HIT), at the time HIT was first clinically suspected and at in-hospital HIT nadir was 209,500/mm3, 77,000/mm3 and 62,000/mm3, respectively. At HIT presentation, 53% had thrombocytopenia alone, 33% had both thrombocytopenia and thromboembolic complications (TEC) and 9.7% had TEC alone. At least one anticoagulant therapy was administered in 89.7% and varied by agent availability, clinical setting or physician selection: Argatroban (51.7%), lepirudin (36.5%), bivalirudin (4.4%), and danaparoid sodium (2.9%). The mean treatment durations on a DTI alone (n=479 pts.) or with subsequent warfarin co-therapy (n=368 pts.) were 9.7 days and 5.4 days, respectively. A new TEC after HIT diagnosis, major bleeding event or amputation occurred in 13%, 7.5%, and 1.8% (composite outcome: 22.6%). Of 446 (82%) pts. surviving to hospital D/C, 84.8% were on warfarin therapy alone. The mean HIT hospital LOS was 21.3 days. The all-cause mortality (ACM) was 18.2% with a quarterly-calculated linear trend in reduction over time (p=0.07). A 56% reduction in the annual ACM was observed from a peak of 28.8% in 2000 to 12.8% in 2005. Significant multivariate predictors of ACM included: TEC presentation (OR 2.09; 95% CI 1.33–3.28; p=0.001) and DTI use (OR 0.48; 95% CI 0.26–0.89; p=0.021).
Conclusions: In this HIT Registry 6-year review, HIT developed most frequently after OHS with a frequency of 2.5%. Fifty percent of medical HIT cases occurred in cardiac pts. Ninety-six percent presented with thrombocytopenia, TEC, or both. D-HIT developed following hospital discharge in 16% of OHS HIT pts. HIT survivorship was observed in 82%. The composite outcome was 22.6%. A strong trend for ACM reduction was observed. In a multivariate analysis, ACM was significantly influenced by a TEC presentation and DTI use.
Burden of herpes zoster among Brazilian adults – a hospital-based study
