Topical bovine thrombin: a 21-year review of topical bovine thrombin spontaneous case safety reports submitted to FDA's Adverse Event Reporting System

Abstract

Purpose: To review topical bovine thrombin spontaneous adverse event (AE) reports that were forwarded to the US Food and Drug Administration's (FDA) Adverse Event Reporting System (AERS) between January 1986 and December 2006.

Methods: Forty-one spontaneous AE reports were summarized for reported AE profile and chronological reporting patterns. Each AE report was adjudicated by a hematologist for the topical bovine thrombin product that was given and the AE(s) that were reported. AEs were grouped as allergic, coagulopathy/bleeding, and all other AEs combined. Grouped AE serial analyses were carried out using successive 3-year time increments between 1986 (the year an AE report was first noted for a bovine thrombin product) and 2006 (the first full year that was available at the time of initiation of the data summary).

Main outcome measures: The primary outcome measures were every 3-year trend lines for all-AE reports, all reporters, and topical bovine thrombin brand mentions for 2 AE groups of interest (allergic events and coagulopathy/bleeding events).

Results: The all-AE spontaneous reporter trend showed a downward appearance for AE reporting activity that started in 1995-1998 and continued through 2004-2006. The all-AE reports trend showed two potential safety signals that could be identified serially: (1) a prominent 1989-1991 peak that was attributable to allergic events (in particular, anaphylaxis), and (2) a small 1995-2000 broad peak that was attributable in part to coagulopathy/bleeding events. Allergic events were predominantly reported with products approved prior to 1995, were not temporally associated with prior medical literature case reports, and continued to be forwarded to the FDA at low levels up to the end of this study in 2006. Coagulopathy/bleeding events were reported only with products approved prior to 1995, were temporally associated with medical literature case reports, and were not forwarded to the FDA after 2000.

Conclusions: Overall, spontaneous AE reporting for topical bovine thrombin occurs at very low levels, and appears to have been decreasing since 1995. The serial reporting patterns for topical bovine thrombin are best explained as a strong safety signal for allergic events with ongoing, low level reporting, and a weak safety signal for coagulopathy/bleeding events that ceased on or before 2000. Although this descriptive trend analysis cannot measure associations or causation, the coagulopathy/bleeding signal may have been prompted by multiple, antecedent published case reports. The subsequent diminishment of signal attributed to thrombin likewise may coincide with lack of such reporting in larger follow-up clinical trials or, alternatively, in the introduction and growing market share of thrombin brands of greater purity. Currently marketed topical bovine thrombin formulations are rarely volunteered as possible causes of adverse events.