Assessment Of Fungal Infection Risk Associated With Specific Contaminated Lots Of Methylprednisolone Acetate, New Jersey 2012

BACKGROUND:   In September 2012, the New Jersey Department of Health (NJDOH) was notified that six NJ facilities received three potentially contaminated lots (05212012@68, 06292012@26, 08102012@51) of methylprednisolone acetate (MPA) from the New England Compounding Center (NECC), associated with a fungal infection outbreak. Two facilities received >80% of product shipped to NJ, and reported an estimated 598 patients were exposed to the product. Between September and December 2012, NJDOH identified 46 fungal infection cases associated with MPA, per CDC case definitions; of these cases, 44 (96%) were associated with the two facilities.  In conjunction with case data, NJDOH analyzed MPA invoice and patient injection information at these two facilities to characterize infection risk associated with particular MPA lots.

METHODS:   MPA shipment information by facility (ship date, number/dosage of MPA vials ordered, lot supplied) was obtained from NECC invoices. Since facilities had limited data on injections administered to patients (e.g., lack of lot number documentation for each injection administered), facilities’ product usage and record keeping practices (e.g., “first in, first out” product use; methods of reordering and replenishing products, documentation of vial usage and dosages administered) were evaluated to determine which MPA lot patients likely received.  Calculated cumulative frequencies of MPA vials used and dates of patient injections were compared with shipment information in order to assign lot numbers to injections administered by patient.  Attack rates were calculated based on injections administered to cases versus non-cases according to MPA lot received.

RESULTS:   The two facilities administered a total of 693 injections to 598 patients; lot numbers could not be assigned for 106 injections, and these injections were excluded from analysis. Of the remaining 587 injections, 48 injections were administered to case-patients and 539 injections where administered to non-case patients.  Those patients receiving lot 08102012@51 had the highest fungal infection attack rate (14.81%).  Patients receiving lot 05212012@68 had an attack rate of 2.63%, and lot 06292012@26 was associated with an attack rate of 9.14%.

CONCLUSIONS:   In these two NJ facilities, injections with MPA from lot 08212012@51 were associated with the greatest risk of becoming a case. Many assumptions, based on available facility and NECC invoice data, were made to determine which MPA lot patients received, and therefore the reliability of calculated attack rates is based on the validity of these assumptions. The ongoing nature of this multistate fungal infection investigation may reveal additional information that would facilitate affected states, including NJ, to assess lot-associated risk more accurately.