To our knowledge, this is the first study to investigate the effect of COPD on the clinical outcomes of patients with VAP in Taiwan. In our study, patients with COPD accounted for 44% of the overall study population. Our findings showed no significant differences in mortality between the COPD and non-COPD groups. The length of ICU and hospital stay and IMV duration were similar between groups. P. aeruginosa was the most common etiology in patients with VAP, but the COPD group showed a higher frequency of Acinetobacter spp.
Previous studies on the impact of COPD on the outcomes of patients with VAP had different definitions of COPD. Numerous studies either failed to confirm COPD using forced spirometry (16, 19, 25) or could have incorrectly classified patients with probable COPD in the non-COPD group by solely eliminating COPD clinical diagnoses and smokers without spirometry (15). Discrepancies in results can be found if PFTs are not mandatory for diagnosis. We used rigorous COPD diagnostic criteria in this investigation, and patients were only included if they underwent post-bronchodilator spirometry (1). This should have led to a well-defined group of patients with COPD, yielding more robust results.
In our study, the microbiological etiology of pneumonia was determined in only 47% of the patients, similar to that reported in the literature (15, 19). Varying epidemiologies have been documented in different locations and territories. E. coli, S. maltophilia, P. aeruginosa, and non-fermenting GNB have been reported to be significantly more prevalent in patients with COPD than those without COPD (19, 21). In contrast, Makris et al. found no differences in pathogens between the two groups among the VAP population (16). Herein, we discovered that Acinetobacter spp. had a higher prevalence in patients with COPD than in those without COPD. These inconsistencies can be attributed to the different COPD or VAP definitions used in previous studies. However, in line with former reports (16, 19, 26, 27), we also found that P. aeruginosa was the most common pathogen in COPD patients with VAP, representing 20% of patients with COPD.
Regarding mortality, the outcome varies in different studies. We found no differences in mortality between the COPD and non-COPD patients with VAP. A prospective cohort study conducted in 20 countries that enrolled 439 patients with VAP found as well that COPD was not associated with mortality in these patients (20). Another multi-national investigation conducted in Europe also found that COPD had no significant impact on mortality in 369 patients with VAP (19). In contrast, several studies demonstrated that COPD is associated with higher ICU mortality in patients with VAP (15, 16). A broad range has been reported on ICU mortality of 38–60% in VAP patients with COPD (16, 19, 20). In our study, VAP patients with COPD presented an ICU mortality of 31% and a 90-day mortality of 53%.
The effect of COPD on the duration of mechanical ventilation or hospitalization of patients with VAP has also been controversial, as different results have been reported. Rouze et al. conducted a large multi-national prospective study and concluded that COPD had no significant impact on the length of hospital stay, ICU stay, or IMV duration, even in patients without VAP (19). Similarly, a single-center study conducted by Makris et al. also found no differences in the length of ICU stay and IMV duration; however, when patients were stratified using COPD GOLD staging criteria, the average length of ICU stay and IMV duration were significantly longer in GOLD stage IV patients than in non-COPD patients (16). Our findings also showed no differences in the length of ICU stay, hospital stay, and IMV duration, and similar results were observed in the surviving patients.
We realize that it is challenging to reach a firm conclusion because there is no agreed definition of pneumonia, and some study settings did not mandate the PFTs inclusion in COPD identification. However, since we used a rigorous definition to examine the two groups, we were able to determine more accurately whether COPD is linked to worse outcomes.
Strengths and limitations
As aforementioned, this study is the first to report the effect of COPD on patients with VAP in Taiwan. We also strictly adhered to the diagnostic standards for COPD and only included individuals who had undergone post-bronchodilator spirometry (1). These factors should aid in producing a well-defined homogeneous population of patients with and without COPD and provide more precise results, even though we have inevitably excluded many patients during the screening process. Nevertheless, the single-center design of this study is its primary drawback; thus, extrapolating its findings to other institutions should be performed with caution, especially the lack of a connection between COPD and mortality, given the retrospective nature of the study and small sample size.