The Use of Pulsed – Field Gel Electrophoresis(PFGE) to Characterize an Outbreak of Acinetobacter baumanii in Baltimore, Maryland

Background:

Acinetobacter baumanii (ACIN) is a non lactose-fermenting gram-negative coccobacillary organism that is often a component of normal respiratory and/or skin flora.  This organism has also emerged as an important MDRO (multiple drug-resistant organism) often found to colonize and/or infect critically ill patients.  We experienced an outbreak of ACIN beginning in July of 2007 and ending in March of 2009.

Objective:

To use PFGE analysis to determine the epidemiological relationship between isolates of ACIN collected during our outbreak period. 

Methods:

Clinical isolates were sent to a reference laboratory (ARUP) for PFGE analysis.  Reference isolates for select clones were identified and used for ongoing analyses.  DNA fragment analysis was performed with 0 bands difference indicating the same indistinguishable clone and part of an outbreak, 2-3 bands difference indicating that isolates were closely related and probably part of the same outbreak, 4-6 bands difference indicating that isolates may possibly be related and possibly part of the same outbreak. >=7 bands difference are clearly different clones and not part of the same outbreak.

Results:

63 isolates from 52 patients were PFGE tested.  These isolates were collected between March of 2007 (pre-outbreak) and March of 2009 (at the end of the outbreak).

Testing revealed a total of  10 different clones.  6 of the clones appeared to be unique PFGE patterns (Clone D-I).  One small clone (Clone J) contained 2 patients, both with identical PFGE patterns, both from the same nursing home.  Clone B had 3 patients, all with identical PFGE patterns, 2 from the same nursing home and the last patient from a chronic care hospital associated with a large Baltimore medical center.  Clone C, the largest, had a total of 36 patients and a total of 42 different isolates.  22 of these patients had identical PFGE patterns, 12 additional patients had no more than 3 bands difference, and 2 others 4 bands difference.  Of the 36 patients, 6 were from the community, i.e. from their home, 1 patient from a very large tertiary hospital and the remainder from a total of  6 different nursing homes in the vicinity of the hospital with 13 patients coming from ventilator-dependent units in 3 of those facilities.  The last Clone, A, contained 8 patients, 3 from the community and the rest from nursing homes in our local area.

Conclusions:

69% of our patient population had isolates collected during a two-year period and demonstrated probable epidemiological association so close that they are most likely part of the same large “outbreak”.  Furthermore, the span of time involved suggests to us that common strains of ACIN are most likely circulating in various acute and subacute settings throughout Baltimore and have become endemic among critically ill patients that transition through these various levels of healthcare settings in the Baltimore area.