A and B, Intra-aortic balloon pump use was adversely associated with patient outcome, regardless of situation, in crude (A) and multivariable (B) analyses. In the logistic regression model, adjustments were made using all variables exhibiting a bivariate association with the use of IABP with P < .001 in the Table, which included all variables except the following: diabetes mellitus, previous coronary artery bypass graft, chronic lung disease, stable angina or silent ischemia, and 1-vessel disease. C, For evaluating the baseline inequality index, we redefined a list of the following baseline characteristics that are recognized markers of mortality risk: age, cardiogenic shock, prior heart failure, peripheral vascular disease, chronic lung disease, renal dysfunction, NYHA functional classification of at least 3 at the time of percutaneous coronary intervention, and clinical presentation (STEMI or NSTEMI). LMT indicates left main trunk; NSTEMI, non–ST-segment elevation myocardial infarction; NYHA, New York Heart Association; and STEMI, ST-segment elevation myocardial infarction.
Inohara T, Miyata H, Ueda I, Maekawa Y, Fukuda K, Kohsaka S. Use of Intra-aortic Balloon Pump in a Japanese Multicenter Percutaneous Coronary Intervention Registry. JAMA Intern Med. 2015;175(12):1980-1982. doi:10.1001/jamainternmed.2015.5119
We read with interest the recent meta-analysis by Ahmad et al,1 demonstrating a negative association between intra-aortic balloon pump (IABP) therapy and mortality among patients experiencing acute myocardial infarction. We agree that efforts are needed to clarify the role of IABP therapy and to examine its effect on care in other regions and countries. In Japan, IABP therapy is frequently used in patients with guideline-based indications and in patients with less established indications, and the judicious use of invasive procedures has been highlighted.2,3 Our objective herein was to investigate the prognostic effect of IABP use in patients undergoing percutaneous coronary intervention (PCI) for nonacute and acute indications registered in a contemporary multicenter Japanese PCI registry (Japan Cardiovascular Database-Keio Interhospital Cardiovascular Studies4).
We analyzed data from 14 378 consecutive patients treated between September 2, 2008, and May 19, 2014. Of those, 1124 patients were excluded because of missing baseline information (n = 192), registration for staged PCI during the same hospitalization (n = 801), or PCI performed under percutaneous cardiopulmonary support (n = 132). The remaining 13 253 patients were included herein, and logistic regression models for in-hospital mortality were used to correct for differences in variables. We included in the logistic regression model all variables exhibiting a significant (P < .10) bivariate association with IABP use. Baseline inequality between patients with and without IABPs was evaluated with the baseline inequality index, the same method used by Ahmad et al.1 Because our study focused on the effect of IABP on in-hospital mortality for all PCIs, we redefined a list of baseline characteristics recognized as markers of mortality risk based on a previous study.5 Data analyses were performed using statistical software (SPSS, version 22.0; SPSS Inc). This study was approved by each participating hospital’s ethics review board (Keio University School of Medicine, Saiseikai Utsunomiya Hospital, Ashikaga Red Cross Hospital, Saitama City Hospital, Saitama National Hospital, Hino Municipal Hospital, Tokyo Dental College Ichikawa General Hospital, Tokyo Saiseikai Central Hospital, Tokyo Medical Center, St Luke's International Hospital, Kawasaki Municipal Hospital, and Yokohama Municipal Citizen's Hospital), and written informed consent was obtained from each patient.
Baseline demographics in patients with and without IABPs are summarized in the Table. Overall, PCIs after ST-segment elevation myocardial infarctions and PCIs after non–ST-segment elevation myocardial infarctions or unstable angina accounted for 23.9% and 24.2% of the procedures, respectively. Before PCI, 486 patients (3.7%) and 900 patients (6.8%) manifested complications of cardiogenic shock and serious heart failure (New York Heart Association functional classification ≥3), respectively. The proportions of interventions for left main trunk and 3-vessel disease were 3.7% and 0.9%, respectively. Intra-aortic balloon pumps were inserted in 885 patients (6.7%). There were 134 in-hospital deaths (15.1%) among the patients receiving an IABP and 111 in-hospital deaths (0.9%) among the patients not receiving an IABP. In the crude analysis, the use of IABP was associated with an increased risk of in-hospital mortality (Figure, A).
Intra-aortic balloon pump use remained an independent predictor of in-hospital mortality after adjusting for baseline differences (odds ratio, 3.87; 95% CI, 2.71-5.52; P < .001). Among several subgroups thought to potentially have indications for IABP use, the use of IABPs was consistently associated with risk of in-hospital death (Figure, B), and IABP recipients had a worse baseline risk profile than nonrecipients (Figure, C). Notably, the risk of death appeared to be higher (with higher odds ratios) as the indications for IABP use became less established.
Among a cohort of Japanese patients undergoing PCI in whom IABP use was frequent, we found that the use of IABP was associated with a higher risk of in-hospital death. This unfavorable association was consistent across clinical settings and was more pronounced as the indications for IABP use became less established.
Several limitations need to be acknowledged. Because of the observational design, we cannot assume a causal relationship between IABP use and mortality. Despite rigorous risk adjustment, the possibility of confounding by unmeasured covariates remains. However, the consistency of the association between IABP use and mortality in various subgroups is notable. Our registry does not capture reasons for IABP insertion. Some physicians or patients may have declined IABP based on institutional or personal preferences.
Using a contemporary multicenter Japanese PCI registry, we have shown a negative association between IABP use and mortality. Our findings are consistent with the meta-analysis by Ahmad et al1 and suggest that it is time to reconsider the appropriate use of IABP therapy, a potentially life-saving but extremely costly and high-risk intervention for patients.
Corresponding Author: Shun Kohsaka, MD, PhD, Department of Cardiology, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan (firstname.lastname@example.org).
Published Online: November 2, 2015. doi:10.1001/jamainternmed.2015.5119.
Author Contributions: Dr Inohara had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Inohara, Kohsaka.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Inohara, Kohsaka.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Inohara, Miyata.
Obtained funding: Inohara, Ueda, Fukuda, Kohsaka.
Administrative, technical, or material support: Miyata, Ueda, Maekawa, Fukuda, Kohsaka.
Study supervision: Miyata, Ueda, Maekawa, Fukuda, Kohsaka.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was funded by grants 25460630 and 80571398 from the Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research and by the Pfizer Health Research Foundation.
Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.