Effect of Heart Rate Variabilities on Outcome After Acute Intracerebral Hemorrhage: A Post Hoc Analysis of ATACH-2

Abstract

Background To explore how the clinical impact of heart rate (HR) and heart rate variabilities (HRV) during the initial 24 hours after acute intracerebral hemorrhage (ICH) contribute to worse clinical outcomes. Methods and Results In the ATACH-2 (Antihypertensive Treatment in Intracerebral Hemorrhage 2) trial, the HR was recorded for every 15 minutes from baseline to 1 hour and hourly during the initial 24 hours post-randomization. We calculated the following: mean, standard deviation, coefficient of variation, successive variation, and average real variability (ARV). Outcomes were hematoma expansion at 24 hours and unfavorable functional outcome, defined as modified Rankin Scale score 4 to 6 at 90 days. Of the 1000 subjects in ATACH-2, 994 with available HR data were included in the analyses. Overall, 262 experienced hematoma expansion, and 362 had unfavorable outcomes. Increased mean HR was linearly associated with unfavorable outcome (per 10 bpm increase adjusted odds ratio [aOR], 1.31, 95% CI, 1.14-1.50) but not with hematoma expansion, while HR-ARV was associated with hematoma expansion (aOR, 1.06, 95% CI, 1.01-1.12) and unfavorable outcome (aOR, 1.07, 95% CI, 1.01-1.3). Every 10-bpm increase in mean HR increased the probability of unfavorable outcome by 4.3%, while every 1 increase in HR-ARV increased the probability of hematoma expansion by 1.1% and unfavorable outcome by 1.3%. Conclusions Increased mean HR and HR-ARV within the initial 24 hours were independently associated with unfavorable outcome in acute ICH. Moreover, HR-ARV was associated with hematoma expansion at 24 hours. This may have future therapeutic implications to accommodate HR and HRV in acute ICH. Registration URL: https://www.clinicaltrials.gov; Unique Identifier: NCT01176565.

Keywords: heart rate; heart rate variabilities; intracerebral hemorrhage; randomized controlled trial; stroke.

Conflict of interest statement

Toyoda reports honoraria from Daiichi‐Sankyo, Bayer Yakuhin, Brystol‐Meyers‐Squibb, and Nippon Behringer Ingerheim, outside the submitted work. Koga reports honoraria from Bayer Yakuhin, BMS/Pfizer, Otsuka, Daiichi‐Sankyo, Nippon Boehringer Ingelheim and Takeda, scientific advisory board from Ono, and research supports from Takeda, Daiichi‐Sankyo, Nippon Boehringer Ingelheim, Astellas, Pfizer and Shionogi, outside the submitted work. Ihara reports grants from Otsuka Pharmaceutical Ltd., grants from BMS, and grants from Shimadzu Corporation outside the submitted work. The remaining authors have no disclosures to report.

Figures

Figure 1. The odds ratio according to quartiles of heart rate (HR) measures.

A, for hematoma expansion at 24 hours; (B) for unfavorable functional outcome at 90 days. Lowest quintile is reference. ARV indicates average real variability; CV, coefficient of variation; and SV, successive variation.

Figure 2. Associations between heart rate‐average real variability (HR‐ARV) and hematoma expansion at 24 hours.

A, The adjusted cubic spline model. This curve (black line) displays that higher HR‐ARV appears to be associated with a rise in the risk of outcome. The dotted curves represent the upper and lower 95% confidence limits, respectively. The horizontal black line represents the odds ratio of 1. B, Probability of hematoma expansion with margins plot, including 95% CIs. Every 1 increase in HR‐ARV is associated with an absolute increase in the probability of hematoma expansion by 1.1%.

Figure 3. Associations between mean‐heart rate (HR) and unfavorable functional outcome.

A, The adjusted cubic spline model. This curve (black line) displays that higher mean HR appears to be associated with a rise in risk of outcome. The dotted curves represent the upper and lower 95% confidence limits, respectively. The horizontal black line represents the odds ratio of 1. B, Probability of unfavorable functional outcome with margins plot, including 95% CIs. Every 10‐bpm increase in the mean HR was associated with an absolute increase in the probability of unfavorable functional outcome by 4.3%.

Figure 4. Associations between heart rate‐ average real variability (HR‐ARV) and unfavorable functional outcome.

A, The adjusted cubic spline model. This curve (black line) displays that higher HR‐ARV appears to be associated with a rise in the risk of outcome. The dotted curves represent the upper and lower 95% confidence limits, respectively. The horizontal black line represents the odds ratio of 1. B, Probability of unfavorable functional outcome with margins plot, including 95% CIs. Every 1 increase in HR‐ARV was associated with an absolute increase in the probability of unfavorable functional outcome by 1.3%. C, Probability of unfavorable functional outcome, according to the treatment arm. In the intensive treatment arm, there was a 0.6% increased probability of worse functional outcome per 1 increase in HR‐ARV. Whereas in the standard treatment, there was a 2.2% increased probability of it.