Telavancin hospital-acquired pneumonia trials: impact of Gram-negative infections and inadequate Gram-negative coverage on clinical efficacy and all-cause mortality.

BACKGROUND When hospital-acquired or ventilator-associated bacterial pneumonia (HABP/VABP) is caused by gram-positive and gram-negative pathogens or both (mixed infections), the adequacy of gram-negative coverage (GNC) can confound the assessment of a gram-positive agent under study. This analysis examines the influence of gram-negative infections and the adequacy of GNC on clinical efficacy and all-cause mortality in the telavancin HABP/VABP phase 3 ATTAIN trials (Assessment of Telavancin for Treatment of Hospital-Acquired Pneumonia). METHODS This post hoc analysis evaluated 3 patient groups from ATTAIN: (1) gram-positive-only infections, (2) gram-positive-only and mixed infections-adequate GNC, and (3) gram-negative-only infections and mixed infections with inadequate GNC. For each, clinical efficacy at test of cure and all-cause mortality at day 28 were compared for telavancin and vancomycin. RESULTS/CONCLUSIONS In the ATTAIN safety population there were 16 more deaths in the telavancin arms than in the vancomycin arms. Of these, 13 were in patients with gram-negative-only infections (n = 9) or with mixed infections and inadequate GNC (n = 4) and all had estimated baseline creatinine clearances of <30ml/min. Based on this analysis, clinical response and all-cause mortality could be confounded because there were more patients with gram-negative pathogens at baseline and more patients received inadequate treatment of these gram-negative infections in the telavancin groups.