Antimicrobials for Wound Infections after Earthquakes: Closing the Gap between Guidelines and Microbes

Background: During the past 12 years five catastrophic earthquakes have occurred. Most caused widespread destruction, injury and loss of life and overwhelmed local medical capacity. Medical teams were dispatched both from within the country and from abroad to treat the excess patient load and to bridge the resource gap until local health services recovered. Many of the injured presented with infected wounds, necessitating empirical antimicrobial treatment. The CDC and WHO have proposed guidelines for this scenario, targeting primarily Gram-positive pathogens.

Objective: To examine the appropriateness of the empirical antibiotic treatment outlined by these guidelines in earthquake settings.

Methods: In response to the January 12, 2010 Haiti earthquake, the Israel Defense Forces dispatched a rapid response team, based on a 72 bed mobile hospital, to Port au Prince, Haiti. This hospital was operational less than four days after the quake and was active for 10 days during which time 737 patients were admitted, the majority due to limb trauma requiring orthopedic procedures and for treatment of infected wounds. An on-site field microbiology laboratory was used to identify wound pathogens, with identifications confirmed later at a hospital laboratory in Israel. Organisms isolated were compared with those reported from victims of previous natural disasters.

Results: 77% of wound infections sampled were polymicrobial, involving primarily Gram-negative pathogens (41/46). 65% of the isolates were enterobacteria including E. coli, Proteus spp., K. pneumoniae and Pantoea spp.; 23% were Gram-negative non-fermenters primarily A. baumannii and P. aeruginosa. These isolates were generally resistant to antimicrobials suggested in the current guidelines. In only 4 patients were Gram-positive bacteria identified. Previous reports from such settings were from referral hospitals and did not distinguish between nosocomial pathogens and those acquired in the field. Our observations were based on cultures taken upon arrival of patients who had no previous exposure to a healthcare environment. We found a remarkable similarity in the range of organisms isolated from wounds of Haiti victims and those of previous incidents, suggesting that the isolates reported in earlier studies indeed reflected the wound pathogens that might be anticipated in earthquake settings.

Conclusions: Emergency relief medical teams should be equipped with antimicrobial drugs appropriate for Gram-negative bacillary infections as well as those included in current guidelines. In addition, we recommend the inclusion of field microbiology laboratories in future disaster medical relief operations to enable the early adaptation of empiric antibiotic therapy to the local pathogen spectrum and antibiotic resistance patterns.