Short-course antibiotics are sufficient for most patients, concludes an open-label trial that compared a 3-day course of antibiotics with an extended course for patients with febrile chemotherapy-induced neutropenia of unknown origin.
Carbapenem antibiotics can safely be stopped for patients who become afebrile after 3 days, say the investigators, although they note that the shorter course was associated with more adverse events.
The trial was carried out in patients with unexplained febrile neutropenia that resulted from intensive chemotherapy or hematopoietic stem cell transplantation (HSCT). The patients were randomly assigned to receive a 3-day short course or an extended-treatment protocol of carbapenem antibiotic therapy with either imipenem-cilastatin or meropenem.
There was no significant difference in the rate of treatment failure in the two arms.
However, more patients in the short-course group experienced serious adverse events, and there were more deaths in this treatment arm in comparison with the extended-treatment group, particularly among patients who were not afebrile after 3 days of treatment.
“Short treatment was noninferior to extended treatment with regard to treatment failure,” say the researchers, led by Nick A. de Jonge, MD, Department of Hematology, Vrije Universiteit Amsterdam, the Netherlands.
“However, because the safety of early discontinuation is unclear in patients with persistent fever, we recommend vigilance for infections that are not sensitive to carbapenem and early reinstitution of empirical therapy in patients who are deteriorating,” they add.
The study was published in The Lancet Haematology on June 9.
In an accompanying editorial, Benjamin W. Teh, MD, PhD, Department of Infectious Diseases, Peter MacCallum Cancer Center, Melbourne, Australia, said that despite “an increasing amount of evidence” in support of shorter courses of antibiotics in these patients, adoption in clinical practice “remains poor.”
The new study has “several aspects” that further “raise challenges for its implementation into clinical practice,” the editorialist comments.
In particular, microbiological workup and diagnostic tools for detecting infection are either not rapid enough for the required time frame or are insufficiently sensitive.
He nevertheless suggests that “incorporation of a clinical threshold, such as resolution of fever, will probably improve acceptability of early de-escalation or antibiotic cessation for undifferentiated neutropenic fever to clinicians.”
Patients at High Risk of Infections
In their article, the authors emphasize that patients who receive intensive chemotherapy or are undergoing HSCT for hematologic malignancies “are at high risk of infections” if they experience chemotherapy-induced neutropenia.
Broad-spectrum antibiotics have been shown to be effective in reducing mortality, and “historically, clinicians have been reluctant to discontinue empirical antibiotic treatment in patients who are immunocompromised,” they comment.
However, continuing with antibiotics increases the risk of unnecessarily prolonged treatment, breakthrough infections, and antimicrobial resistance.
The result, the researchers note, is that clinical practice “varies greatly around the world,” and guidelines “differ considerably,” despite recent analyses indicating that short antibiotic courses of less than 4 days are not associated with worse outcomes.
However, the safety of short-course antibiotics for patients who experience fever of unknown origin during neutropenia has been less clear, and this prompted the team to focus on this specific patient population.
The 292 patients enrolled into the trial had undergone intensive chemotherapy or HSCT for a hematologic malignancy at one of six hospitals in the Netherlands. The median age was 59 years, and 29% were women. The majority (73%) of patients had undergone HSCT.
All patients were considered to be at high risk for neutropenia (defined as a neutrophil count <0.5 x109/L expected for ≥7 days) at the onset of unexplained fever.
Initially, all patients were treated with intravenous imipenem-cilastatin 500 mg four times a day or intravenous meropenem 1000 mg three times a day at the start of fever.
From 48 hours to 72 hours after starting treatment, the patients were randomly assigned in a 1:1 ratio to receive either a short course of treatment, in which the carbapenem antibiotic was discontinued 72 hours after starting therapy, or an extended course.
In the extended-treatment group, carbapenem was continued until neutrophil recovery or for a total of 9 days of treatment, whichever came first. If the patient had not been afebrile for 5 days before the end of the treatment period, the antibiotic was continued up to a maximum of 14 days.
“Overall, antibiotic treatment was two days shorter in the short treatment arm, which was less of a reduction than expected,” the researchers note. “The lower than expected reduction was caused by participants going off-protocol and receiving extended treatment according to local guidelines.”
In an intention-to-treat analysis that included 281 patients, there was no significant difference in rates of treatment failure (a composite endpoint included recurrent fever, septic shock, respiratory failure, and death). The rate was 19% in the short-course group vs 15% in the extended-treatment group (P = .25).
A similar picture was seen in a per-protocol analysis of 225 patients, in which 23% of patients who received short-course therapy experienced treatment failure, vs 16% of the extended-treatment group (P = .11).
The proportion of patients with serious adverse events was higher in the short-course group than among those given extended treatment, at 16% vs 10%. Notably, the rate of hospital readmission was higher in the short-course group, at 10% vs 7%.
Overall, the most common grade 3–5 adverse events were mucositis (20% of short-course patients vs 29% of extended-treatment patients), fever of unknown origin (18% vs 16%), and bacteremia (13% vs 13%).
The team also reports that there were five deaths within 30 days of neutrophil recovery in the short-course group. These deaths were attributed to progressive leukemia in two patients, candidemia in two patients, and Enterococcus faecium bacteremia and drug-induced pneumonitis in one patient.
One patient died in the extended-treatment group. The cause of death was candidemia.
None of the deaths were due to carbapenem-sensitive infections, the investigators note. They emphasize that the “deaths were unrelated to the shorter antibiotic treatment they received.”
Subgroup analysis indicated that treatment failure was more common among participants with persistent fever after day 3 in the short-course group than in the extended-treatment group, as were excess mortality and serious adverse events.
“The higher rate of serious adverse events and the higher overall and infection-related mortality after neutrophil recovery in the short treatment group is a concerning and not well understood finding that could prevent a safe introduction of short term treatment in general practice,” the authors write.
“Therefore, we do not recommend this strategy in patients with ongoing fever after three days of empirical antibiotic therapy,” they add.
The study was funded by the Netherlands Organisation for Health Research and Development and Fonds NutsOhra. The authors have disclosed no relevant financial relationships.