Heart Rate is Not a Reliable Predictor of Adverse Cardiac Outcomes or Mortality after Major Elective Vascular Surgery
Salvatore T. Scali1, Daniel Neal2, Daniel J. Bertges3, Jens Eldrup-Jorgensen4, Jack L. Cronenwett5, Adam W. Beck1
1university of florida- gainesville, Gainesville, FL;2University of Florida- Gainesville, Gainesville, FL;3University of Vermont-Burlington, Burlington, VT;4Maine Medical Center, Portland, ME;5Dartmouth-Hitchcock Medical Center, Lebanon, NH
Introduction: Arrival heart rate(AHR) and heart rate control(HRC) are known indicators of cardiovascular complications after cardiac surgery, but there is little evidence of their role in predicting outcome after major vascular surgery. The purpose of this study was to determine whether AHR and HRC are predictive of mortality or major adverse cardiac events(MACE) after elective vascular surgery in the Vascular Quality Initiative(VQI).
Methods: Using the VQI dataset, a retrospective analysis was performed on patients undergoing elective infrainguinal bypass(IIB), suprainguinal bypass(SIB) and open AAA repair(oAAA). MACE was defined as any postoperative myocardial infarction(POMI), dysrhythmia or congestive heart failure(CHF). Controlled HR was defined as HR less than 75 upon OR arrival, while HRC was determined by taking the highest intraoperative HR-AHR. Procedure specific MACE models were applied to risk stratify patients. Beta blocker(BB) status[none vs. acute(0-30 days preoperatively) vs. chronic (>30 days preoperatively)] was evaluated, and a Bonferroni correction with a P<.002 was used to declare any result significant at a 5% error rate. Logistic regression generalized estimating equations and linear mixed models were used with MACE as the outcome and AHR/HRC, BB status and the interaction between AHR/HRC and BB status as fixed factors.
Results: A total of 13,141 elective patients were reviewed[IIB, N=8,155(62%); SIB, N=2,629(18%);oAAA, N= 2,629(20%)]. The raw non-risk adjusted rates of MACE and 30-day mortality were: IIB, MACE 9.8%(N=633), 30-day death 2.0%(N=126); SIB, 15.7%(N=263), 2.6%(N=43); oAAA, 21.7%(N=486), 3.3%(N=73). Rates of any preoperative BB exposure were: IIB, 66.5%(N=5412); SIB, 57%(N=1342); oAAA, 74.2%(N=1949). Association of AHR and HRC with outcomes was variable across patients and procedures. AHR less than 75 was associated with increased risk of POMI for IIB patients across all cardiac risk strata(OR 1.4, 95%CI 1.03-1.9;P=.03), while AHR greater than 75 was associated with increased risk for dysrhythmia(OR 2.4, 1.6-3.6,P=.0001) and 30-day death(OR 2.0,1.3-3;P=.001) in both moderate and high cardiac risk patients. These HR effects disappeared when controlling for BB status. SIB had no association between AHR and 30-day mortality; however, AHR greater than 75 was associated with increased risk of CHF across all cardiac risk strata(OR 2.2, 1.2-3.8;P=.007). These effects also disappeared when controlling for BB exposure. No association with HR and MACE outcome was noted among oAAA patients irrespective of BB status. However, 30-day mortality was increased (OR 1.9, 1.03-3.4, P=.04) for patients with AHR greater than 75. This association was sustained when controlling for acute but not chronic BB status. HRC was analyzed among all 3 operations and inconsistent associations with MACE or 30-day mortality were detected.
Conclusions: HR is highly confounded by patient presentation, operative variables and beta-blocker therapy. The discordance between cardiac risk and HR as well as the lack of consistent correlation to outcome makes this an unreliable predictor. The VQI has elected to discontinue collecting heart rate data nationally given the lack of sufficient data to suggest its importance as an outcome measure.
Back to 2015 Annual Meeting Abstracts