Blood Pressure Variability on Antihypertensive Therapy in Acute Intracerebral Hemorrhage
The Stroke Acute Management With Urgent Risk-Factor Assessment and Improvement-Intracerebral Hemorrhage Study
Background and Purpose—The associations between early blood pressure (BP) variability and clinical outcomes in patients with intracerebral hemorrhage after antihypertensive therapy, recently clarified by a post hoc analysis of Intensive Blood Pressure Reduction in Acute Cerebral Hemorrhage Trial 2 (INTERACT2), were confirmed using the Stroke Acute Management with Urgent Risk-factor Assessment and Improvement (SAMURAI)-intracerebral hemorrhage study cohort.
Methods—Patients with hyperacute (<3 hours from onset) intracerebral hemorrhage with initial systolic BP (SBP) >180 mm Hg were registered in a prospective, multicenter, observational study. All patients received antihypertensive therapy based on a predefined standardized protocol to lower and maintain SBP between 120 and 160 mm Hg using intravenous nicardipine. BPs were measured hourly during the initial 24 hours. BP variability was determined as SD and successive variation. The associations between BP variability and hematoma expansion (>33%), neurological deterioration within 72 hours, and unfavorable outcome (modified Rankin Scale, 4–6) at 3 months were assessed.
Results—Of the 205 patients, 33 (16%) showed hematoma expansion, 14 (7%) showed neurological deterioration, and 81 (39%) had unfavorable outcomes. The SD and successive variation of SBP were 13.8 (interquartile range, 11.5–16.8) and 14.9 (11.7–17.7) mm Hg, respectively, and those of diastolic BP were 9.4 (7.5–11.2) and 13.1 (11.2–15.9) mm Hg, respectively. On multivariate regression analyses, neurological deterioration was associated with the SD of SBP (odds ratio, 2.75; 95% confidence interval, 1.45–6.12 per quartile) and the successive variation of SBP (2.37; 1.32–4.83), and unfavorable outcome was associated with successive variation of SBP (1.42; 1.04–1.97). Hematoma expansion was not associated with any BP variability.
Conclusions—SBP variability during the initial 24 hours of acute intracerebral hemorrhage was independently associated with neurological deterioration and unfavorable outcomes. Stability of antihypertensive therapy may improve clinical outcomes.
Elevated blood pressure (BP) is significantly associated with hematoma expansion,1 neurological deterioration,2 and unfavorable outcomes.3 BP lowering with antihypertensive treatment (AHT) has been reported to be tolerated and feasible for acute patients with intracerebral hemorrhage (ICH).4 The recent result of the Intensive Blood Pressure Reduction in Acute Cerebral Hemorrhage Trial 2 (INTERACT2) demonstrated that BP lowering therapy with a target systolic level of <140 mm Hg tended to be beneficial for acute ICH.5
Additionally, the importance of BP stability in management of acute ICH is recently noted as a potential treatment target. Rodriguez-Luna et al6 revealed that early SD of systolic BP (SBP) independently predicted early neurological deterioration in 117 consecutive patients with ICH but was not associated with hematoma expansion and mortality at 3 months. Recently, the INTERACT2 investigators clarified that SD of SBP in the initial 24 hours have a significant linear association with poor outcome at 3 months in 2645 patients with ICH.7 They also showed the similar linear association using SD of SBP over days 2 and 7. The aim of the present study was to confirm the associations between early BP variability and clinical outcomes including functional outcome at 3 months after onset in patients with hyperacute ICH after standardized AHT using The Stroke Acute Management with Urgent Risk-factor Assessment and Improvement (SAMURAI)-ICH data set.
The SAMURAI-ICH study was a prospective, multicenter, observational study. The details were documented previously.4,8,9 Briefly, the criteria for inclusion are as follows: ≥20 years of age; total Glasgow Coma Scale score ≥5; initial SBP >180 mm Hg; computed tomography <2.5 hours of onset demonstrating a supratentorial intraparenchymal hematoma with manual volume measurement <60 mL; and absence of extensive intraventricular hemorrhage. The study was approved by each institutional ethics and hospital management committee.
The predefined standardized protocol of AHT was used to lower and maintain the SBP level below 160 mm Hg and above 120 mm Hg. Titrating of intravenous nicardipine was initiated at a rate of 5 mg/h within 3 hours of symptom onset and continued for 24 hours. Levels of BP and pulse rate were measured every 15 minutes during the first 2 hours after initiation of AHT and every 60 minutes during the next 22 hours, as well as at 48 and 72 hours. Oral antihypertensive agents were started after the first 24 hours. To test our hypothesis, variability profiles of BP—SD and successive variation (SV)10 —were calculated based on 24 successive BP values measured every 60 minutes within 24 hours after initiation of ATH.
The clinical outcomes included hematoma expansion (>33% in volume from baseline to 24 hours), neurological deterioration (a decrease of ≥2 in Glasgow Coma Scale or an increase of ≥4 in the National Institutes of Health Stroke Scale score from baseline to 72 hours), and unfavorable outcome (modified Rankin Scale score, 4–6 at 3 months). Hematoma expansion was also evaluated by other indexes such as an absolute volume difference and a relative volume change.5
Data including BP variability are presented as median values (interquartile range) or number (%). Multivariate analysis of variance was performed to determine the associations between BP variability and clinical outcomes adjusted by the known predictors of clinical outcomes, including sex, age, previous antithrombotic medication, initial SBP, initial pulse rate, initial National Institutes of Health Stroke Scale score, onset to treatment time, initial hematoma volume, and serum glucose level at baseline. BP variability was categorized into quartiles and analyzed as a continuous variable. A P value of <0.05 was considered significant.
Of the 211 patients in the registry, 205 (81 women, median age 65 [interquartile range, 59–75] years, median initial National Institutes of Health Stroke Scale score 13 [8–17]) whose BP data were available throughout the 24-hour observation period were studied. The Table in the online-only Data Supplement shows the baseline clinical characteristics of the included patients. The SD and SV of SBP were 13.8 (11.5–16.8) and 14.9 (11.7–17.7) mm Hg, respectively, and those of diastolic BP were 9.4 (7.5–15.9) and 13.1 (11.2–15.9) mm Hg, respectively.
Hematoma expansion was observed in 33 patients (16%), neurological deterioration was seen in 14 (7%), and unfavorable outcomes occurred in 81 (40%). As shown in Table 1, the SD and SV of SBP were larger in patients with neurological deterioration than in those without neurological deterioration (SD of SBP, P=0.001; SV of SBP, P=0.003). The SV of SBP was larger in patients with an unfavorable outcome than in those with a favorable outcome (P=0.015). Neither SD nor SV of SBP was associated with hematoma expansion (>33% in volume). Figure 1 shows the frequencies of clinical outcomes according to BP variability in quartiles. Neurological deterioration gradually and significantly increased across quartiles of SD and SV of SBP (P=0.005, P=0.008), whereas unfavorable outcomes increased across quartiles of SV of SBP (P=0.031).
Multivariate regression analyses demonstrated that the SD of SBP (odds ratio, 2.75; 95% confidence interval, 1.45–6.12 per quartile) and the SV of SBP (odds ratio, 2.37; 95% confidence interval, 1.32–4.83 per quartile) were associated with neurological deterioration, and the SV of SBP (odds ratio, 1.42; 95% confidence interval, 1.04–1.97 per quartile) was associated with unfavorable outcome (Table 2). Neither SD nor SV of diastolic BP demonstrated associations with any clinical outcomes (Tables 1 and 2; Figure 2).
Finally, hematoma expansion was evaluated by absolute and relative volume changes. They were significant across quartiles of both the SD and SV of SBP after adjustment for prognostic variables (absolute difference: SD-SBP [P=0.009], SV-SBP [P=0.025]; relative change: SD-SBP [P=0.033], SV-SBP [P=0.048], respectively), and both were highest in the upper quartiles of SD and SV.
In the present subanalysis of the SAMURAI-ICH study, the associations of early BP variability with early and long-term outcomes after ICH were examined. The major findings were that the variability of SBP assessed using both SD and SV during the initial 24 hours of ICH was independently associated with both early neurological deterioration and unfavorable outcome at 3 months.
The present findings are partly consistent with 2 recent studies on the same theme.6,7 The similarities of the findings between the study by Rodriguez-Luna et al6 and the present one were that SBP variability was predictive of early neurological deterioration but not predictive of hematoma expansion. The similarities between the INTERACT27 and the present one were that SBP variability was predictive of unfavorable outcome at 3 months, but again not predictive of hematoma expansion. However, ours showed the positive association between SBP variability and absolute and relative changes in hematoma volume. Differences between the study by Rodriguez-Luna et al6 and the present one were the baseline SBP level (mean, 172 versus 200 mm Hg), the proportion of patients receiving emergent AHT (43.6% versus 100%), and that the present one assessed the association of SBP variability with 3-month clinical outcomes, which was found to be significant. In the present study, not only vital outcome but also functional outcomes were assessed. A difference between the INTERACT2 and the present study was the timing of points of BP measurement within the initial 24 hours (6 versus 24 points). In addition, SV, which reflects the serial BP variation in a time sequence, was used as indicator of variability, as well as SD, in the present study.
Increased BP variability is associated with several factors, including female sex, advanced age, and hypertension,11 which are also major prognostic predictors of unfavorable outcome in patients with ICH. Increased BP variability was associated with clinical outcomes after adjusting for such established predictors in the present study. Thus, other additional mechanisms to connect BP variability and clinical outcomes seem to exist. Autonomic dysfunction, including sympathetic overactivity and diminished baroreflex sensitivity, may be one of the mechanisms. In patients with ICH, the decreased baroreflex sensitivity measured by a hemodynamic monitoring device and increased beat-to-beat BP variability were demonstrated.12 The impaired baroreflex sensitivity increased BP variability and accordingly altered cerebral perfusion, leading to secondary brain injury.13,14 Additionally, sympathetic overactivity affects the factors that boost the secondary brain injury and enhance brain edema, such as leukocytosis, proinflammatory cytokine production, hyperglycemia, hyperthermia, and increased blood–brain barrier permeability.15 In patients with ICH, the association of autonomic disorders with edematous change seems to be important, although data about perihematoma edema volume were not available in the present registry.
Increased BP variability was not associated with standard categorical index of hematoma expansion (>33% in volume) in the present cohort. A relatively low percentage of hematoma expansion (16%) in our cohort, probably attributable to the exclusion criteria of large hemorrhage (≥60 mL) and strict AHT in the hyperacute stage, might weaken the statistical power of the analysis of hematoma expansion.
The type of antihypertensive agent can affect BP variability. An intravenous calcium-channel blocker (nicardipine) was used for our patients.4 Because baroreflex sensitivity can be positively influenced by β-blockers,16 intravenous β-blocker administration may inhibit sympathetic overactivity and decrease BP variability in acute ICH. However, visit-to-visit variability in SBP was reported to be greater with β-blocker than with the calcium-channel blocker in treated hypertensive patients.17
In conclusion, SBP variability during the initial 24 hours of hyperacute ICH was independently associated with neurological deterioration and 3-month unfavorable outcome. Stability with appropriate AHT may ameliorate clinical outcomes in patients with hyperacute ICH. BP variability seems to be an important therapeutic target in acute ICH.
Sources of Funding
This study was supported in part by Grants-in-Aid (H20-Junkanki-Ippan-019 and H23-Junkanki-Ippan-010) from the Ministry of Health, Labour and Welfare, Japan.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.114.005420/-/DC1.
- Received March 11, 2014.
- Revision received May 19, 2014.
- Accepted May 20, 2014.
- © 2014 American Heart Association, Inc.
- Ohwaki K,
- Yano E,
- Nagashima H,
- Hirata M,
- Nakagomi T,
- Tamura A
- Koga M,
- Toyoda K,
- Yamagami H,
- Okuda S,
- Okada Y,
- Kimura K,
- et al
- Sakamoto Y,
- Koga M,
- Yamagami H,
- Okuda S,
- Okada Y,
- Kimura K,
- et al
- Kobayashi J,
- Koga M,
- Tanaka E,
- Okada Y,
- Kimura K,
- Yamagami H,
- et al
- Kato T,
- Kikuya M,
- Ohkubo T,
- Satoh M,
- Hara A,
- Obara T,
- et al
- Immink RV,
- van Montfrans GA,
- Stam J,
- Karemaker JM,
- Diamant M,
- van Lieshout JJ
- Ay H,
- Arsava EM,
- Vangel M,
- Oner B,
- Zhu M,
- Wu O,
- et al