The Science Journal of the American Association for Respiratory Care

2008 OPEN FORUM Abstracts


Griselda G. Garcia2,1, Brendon G. Hopgood1, Nicole Lahey2, Roger W. Yurt1, Octavio Lafuentes2, Palmer Q. Bessey1

Background: Short-term and laboratory studies have demonstrated improved measures of gas exchange with airway pressure release ventilation (APRV), since it was first described 20 years ago. To date, however, there has been no reported sizeable clinical series of critically ill patients supported with that mode. At this burn center, APRV was used for patients with severe acute lung injury (ALI) beginning in May 2004. The purpose of this study was to compare the clinical experience with APRV and that with conventional ventilation (CV) in burn patients with ALI.

Methods: The study groups consisted of 50 patients with ALI supported with CV who required PEEP of 15 or more between May 2002 and May 2004 and 54 severe ALI patients supported with APRV from May 2004 through May 2006. Data are expressed as Mean (95 % confidence interval) and CV vs APRV. Differences in means and proportions were tested by ANOVA and Chi squared analysis respectively.

Results: Mean age (50.7 years (44.6-56.8) vs 48.4 (42.4-54.3)), burn size (24 % Total BSA (16-31) vs 26 (19-33)), and incidence of inhalation injury (62% (48-76) vs 65 (52-75)) were similar in the two groups, as were statistical estimates of case fatality (44% (35-54) vs 46 (38-54)), indicating similarly severe illness.

The highest PEEP used for the CV patients averaged 22 cm H2O (20-24) and ranged from 15 to 35. The P/F ratio improved significantly in both groups after 12 hours of PEEP ≥ 15 or APRV. CV patients with were more than twice as likely to die in the first two weeks after PEEP ≥ 15 as APRV (20 patients compared with 11 (Odds ratio 2.61 (1.09 - 6.23) P=0.031). For those who remained ventilated at two weeks, CV patients had lower P/F ratios, higher peak pressures, and higher minute volumes than APRV (all p<0.05). Case fatality overall was strongly related to age and burn size by logistic regression (P<0.001), but seemed also to be reduced by APRV (Odds ratio 0.41 (0.16 - 1.04) P=0.062).

Conclusion: This is the first sizeable clinical series of patients supported with APRV during a substantial portion of their critical illness. APRV was at least as effective in recruiting airways as PEEP ≥ 15 and CV in patients with severe ALI and burns. APRV also had a short and possibly long term survival benefit. Further clinical experience with APRV in critically ill patients is warranted.