DOI: 10.4103/lungindia.lungindia_293_26 ISSN: 0970-2113

Seven versus 14 days of polymyxin-based therapy for ventilator-associated pneumonia due to carbapenem-resistant Acinetobacter baumannii: A randomised feasibility trial

Gurkirat Kaur, Shrikant K. Malegaonkar, Sahajal Dhooria, Kuruswamy T. Prasad, Valliappan Muthu, Ritesh Agarwal, Manisha Biswal, Kajal Jain, Ashutosh N. Aggarwal, Inderpaul S. Sehgal
Show PDF Cite

ABSTRACT

Background:

Ventilator-associated pneumonia (VAP) due to carbapenem-resistant Acinetobacter baumannii ( CRAB ) is associated with substantial morbidity and mortality in intensive care units (ICUs), yet the optimal antibiotic duration of therapy remains uncertain. We compared outcomes of 7-day versus 14-day polymyxin-based therapy for CRAB-VAP.

Methods:

We conducted a single-centre, open-label, randomised feasibility trial in a tertiary-care respiratory ICU from March 2017 to October 2024. Patients with microbiologically confirmed CRAB-VAP who met predefined clinical stability criteria on day 7 were randomised 1:1 to either discontinue antibiotics on day 7 or continue treatment for 14 days. Treatment comprised intravenous colistin or polymyxin B with adjunctive nebulised colistin. The primary outcome was relapse of CRAB-VAP. Secondary outcomes included antibiotic-free days through day 28, duration of mechanical ventilation, ICU and hospital length of stay, and 28-day all-cause mortality.

Results:

We randomised 41 patients (7-day n = 22; 14-day n = 19). Relapse occurred in 1 (4.5%) versus 2 (10.5%) patients (risk ratio [RR] 0.43; 95% CI: 0.04, 4.31) in the 7-day versus 14-day groups, respectively. The 7-day group had more antibiotic-free days (mean difference: 6.9 days; 95% CI: 5.1, 8.7 days). Acute kidney injury occurred in 0% versus 31.6% (RR 0.07; 95% CI: 0.004, 1.15). Other outcomes, including 28-day mortality (45.5 vs 63.2%; RR 0.72; 95% CI: 0.40, 1.30), were similar.

Conclusion:

In clinically improving CRAB-VAP, a 7-day discontinuation strategy was feasible. Clinical outcome estimates, including relapse and mortality, were imprecise and should be interpreted as hypothesis-generating. These findings support evaluation in larger, adequately powered multicentre trials.

More from our Archive