Infusion reactions are common after high-dose carmustine in BEAM chemotherapy and are not reduced by lengthening the time of administration

Abstract

Purpose

Carmustine (BCNU) is used in the conditioning regimens BEAM and CBV for autologous stem cell transplantation. Carmustine-related infusion reactions, while not described in the BEAM literature, occurred in 95 % of patients who received CBV. The most common symptoms include flushing, facial pain, headache, and hypotension. These reactions have been attributed to the absolute ethanol that is used in the reconstitution process or alternatively by a direct effect of carmustine. It is currently recommended that carmustine 300 mg/m² be infused over at least 100 min (3–5 mg/m²/min). Prior to October 2014, carmustine infusions were given over 90 min but were changed to 120 min based on the above recommendation. We compared the two infusion rates in patients receiving BEAM to see if lengthening the infusion decreased the frequency of reactions.

Methods

Overall, 100 patients received BCNU as part of BEAM or Zevalin BEAM and were equally divided between 90 and 120 min infusion times. The primary outcome was the incidence of infusion-related reactions which were graded based on CTCAE 4.03 descriptions of flushing and infusion-related reactions. We also evaluated the impact of premedication as well as the efficacy of medications used to treat infusion reactions.

Results

Between the years 2013–2016, there were 50 patients who received BCNU over 90 min and 50 patients over 120 min. There were no significant differences observed for diagnosis, age and gender between the two groups. Twenty-eight (56 %) in the 90-min and 26 (52 %) in the 120-min infusion intervals developed a reaction (p = 0.6882). Of the patients that developed a reaction, 19 patients (67 %) in the 90-min and all 26 patients (100 %) in the 120-min infusion were given premedications predominately acetaminophen, in addition to dexamethasone. Among reacting patients, 57 % of the 90-min and 65 % of the 120-min groups received additional intervention (p = 0.53).

Conclusion

Infusion reactions during high-dose BCNU are common and are not clearly reduced by modestly extending the duration of infusion or giving premedications.