Antidote triggers find unreported ADRs

Research at the University of Pittsburgh has shown that tracking adult ICU use of antidotes such as protamine and phytonadione can signal adverse drug reactions (ADRs). A report on the study was given at the recent American College of Clinical Pharmacists meeting.

Sandra Kane-Gill, PharmD, MSc, tells Drug Formulary Review trigger research generally has not included the adult ICU. The primary study objective, she says, was to evaluate positive predictive values of triggers such as protamine, phytonadione, and methylprednisolone for detecting ADRs in the adult ICU. A secondary objective was to compare the number of ADRs detected using triggers to those voluntarily reported.

Kane-Gill says that while the University of Pittsburgh has a voluntary reporting system that produces a reasonable number of ADRs, many people often feel they are too busy to make a report or there are other reasons why reports are not made.

"We wanted to find a way to identify ADRs in a more assertive way," she says.

As a study at King's College Hospital, London, UK, says, triggers can be abnormal laboratory values, medication stops, and prescriptions for certain drugs, all of which can prompt a more detailed search of patient records for information on an ADR. In that study, almost half of prescriptions for trigger drugs were written in response to an adverse drug event, with vitamin K, Beriplex®, naloxone, calcium resonium, and hydroxyzine found to be the most specific.

In Kane-Gill's study, a retrospective chart review was applied to all patients admitted to the medical ICU from July 1, 2005, to June 30, 2006, who received protamine, phytonadione, and methylprednisolone.

So many patients received phytonadione and methylprednisolone that a random sample of 50 patients each from those two groups was evaluated for related ADRs. Only 11 patients received protamine during the study period and all were evaluated.

Objective assessment used for ADRs

A clinical pharmacist doing the chart reviews used an objective assessment instrument to determine that an adverse drug reaction had occurred. The study defined ADRs as undesirable clinical manifestations consequent to and caused by drug administration.

Positive predictive values were calculated as the proportion of triggers that occurred divided by the number of times that triggers occurred and ADRs were confirmed.

The study found positive predictive values of 0.36 for protamine, 0.18 for phytonadione, and 0.0 for methylprednisolone, respectively. Kane-Gill says none of the ADRs detected using the triggers had been reported on the voluntary reporting system.

Protamine and phytonadione make good triggers, she says, because they typically are given in response to an ADR. "If you see that one of these antidotes has been administered," she tells DFR, "you can look at the record and are likely to see that an ADR has occurred."

While this research was done with a retrospective chart review, Kane-Gill says it could be done prospectively and the findings could be used to prevent ADRs.

"If clinical pharmacists could get a real-time alert that an antidote has been administered, we could intervene quickly," she says.

[Editor's note: Contact Dr. Kane-Gill at (412) 624-5150 or e-mail her at slk54@pitt.edu.]