Plasma B-Type Natriuretic Peptide Levels Are Associated With Early Cardiac Dysfunction After Subarachnoid Hemorrhage
Background and Purpose— Serum B-type natriuretic peptide (BNP) is elevated after subarachnoid hemorrhage (SAH), as well as in the setting of congestive heart failure and myocardial infarction. The aim of this study was to prospectively quantify the relationship between BNP levels and cardiac outcomes after SAH.
Methods— Plasma was collected for BNP measurements as soon as possible after enrollment; a mean of 5±4 days after SAH symptom onset. On days 1, 3, and 6 after enrollment, troponin I (cTi) was measured and 2-dimensional echocardiography was performed. The following cardiac variables were collected and treated dichotomously: left ventricular ejection fraction (LVEF), regional wall motion abnormalities (RWMA), diastolic dysfunction, pulmonary edema, and cTi.
Results— There were 57 subjects. The median BNP level was 141 pg/mL (range, 0.8 to 3330 pg/mL). Higher mean BNP levels were present in those with RWMA (550 versus 261 pg/mL; P=0.012), diastolic dysfunction (360 versus 44; P=0.011), pulmonary edema (719 versus 204; P=0.016), elevated cTi (662 versus 240; P=0.004), and LVEF <50% (644 versus 281; P=0.015).
Conclusion— Early after SAH, elevated BNP levels are associated with myocardial necrosis, pulmonary edema, and both systolic and diastolic dysfunction of the left ventricle. These findings support the hypothesis that the heart releases BNP into the systemic circulation early after SAH.
B-type natriuretic peptide (BNP) is released from the heart in patients with myocardial infarction1 and congestive heart failure caused by systolic or diastolic dysfunction.2 Elevated BNP levels also occur after subarachnoid hemorrhage (SAH),3 although the source of BNP in this setting is controversial. Previous studies have shown that BNP levels increase soon after SAH and return to baseline in 1 to 2 weeks.4 Proposed mechanisms include catecholamine release that increases the load on the cardiac ventricles,4 hypoxia of the hypothalamus,5 and endothelin-1 release.6 In a small study of 18 patients,7 plasma but not cerebrospinal fluid levels of BNP increased by 2- to 3-fold, suggesting that the brain is not the source of BNP.
We hypothesized that SAH may result in cardiac dysfunction and release of BNP from the heart. Therefore, the aim of this study was to quantify the relationships between plasma BNP levels and cardiac dysfunction after SAH.
Materials and Methods
This is a substudy from a prospective SAH cohort. The inclusion criteria for the cohort were age older than 21 years and a diagnosis
See Editorial Comment, pg 1570
of SAH by computed tomography or lumbar puncture. Patients were included in the substudy if they consented for storage of plasma samples and excluded if they had history (obtained from the patient, family, or medical records) of myocardial infarction or congestive heart failure. The study protocol was approved by the UCSF Committee on Human Research and informed consent was obtained from each patient or an appropriate designee.
The patients were enrolled as soon as possible after admission. Clinical data were collected from patient and family interviews and the medical record. Blood samples were collected in EDTA tubes, centrifuged, and stored at −70°C. After study completion, BNP was measured using the Triage BNP assay (Biosite, San Diego, Calif).
On days 1, 3, and 6 after enrollment, serum specimens were collected, echocardiography was performed, and a chest x-ray was reviewed. The level of cardiac troponin I (cTi) was measured using a fluorescent enzyme immunoassay (Abbot Diagnostics, Abbott Park, Ill).
Transthoracic echocardiography was performed using an Acuson Sequoia 6.0 system (Mountain View, Calif). For each echo, standard images8 and Doppler recordings of mitral inflow and pulmonary venous flow were obtained. All echocardiographic analyses were performed off-line (ProSolv; Indianapolis, Ind) by a blinded observer. Left ventricular ejection fraction (LVEF) was measured using the biplane Simpson’s method of discs.8 Regional wall motion abnormalities (RWMA) were defined as hypokinesis, akinesis, or dyskinesis of any of the 16 LV segments.8 Using established criteria,9 diastolic function was categorized as normal or abnormal (impaired relaxation, pseudonormal, or restrictive).
The cardiac abnormalities were treated as dichotomous variables based on an abnormal result on any of the 3 study days. Abnormal results were defined as cTi >1.0 μg/L, pulmonary edema on chest x-ray, LVEF <50%, RWMA, and abnormal diastolic function. We performed Wilcoxon rank-sum tests to compare mean BNP levels among patients with and without each cardiac abnormality.
The relationship between BNP levels and hospital discharge disposition (home, acute hospital, rehabilitation hospital, death) was quantified using linear regression (and logBNP as the dependent variable to normalize the BNP distribution). All statistical analyses were performed using commercially available software (STATA, College Station, Tex) and P≤0.05 was considered significant.
The parent cohort study included 174 subjects. Institutional review board approval was obtained for stored blood samples starting with patient 101. The substudy included all 57 patients who consented for blood storage (of 74 eligible patients). The patients’ characteristics are shown in Table 1. There were no significant differences between the substudy patients and the other 117 patients in the cohort, except that more substudy patients had an anterior aneurysm (67% versus 49%; P=0.043). However, there were no significant associations between aneurysm location and the cardiac outcomes.
The mean time from SAH symptom onset to enrollment into the study was 3.3±3.2 days. BNP was measured at a mean of 5.1±3.5 days after SAH. The median BNP level was 141 pg/mL (interquartile range, 51 to 396 pg/mL). There was a trend for higher BNP levels in patients with an admission Hunt-Hess grade of 3 to 5 (449±668 pg/mL versus 189±241 for grade 1 to 2 patients, Wilcoxon rank-sum P=0.086). Patients with a history of hypertension had higher mean BNP levels than patients without hypertension (481±690 versus 188±251, Wilcoxon rank-sum P=0.044) and a higher frequency of diastolic dysfunction.
A total of 12 subjects (21%) had RWMA on any study day, 89% had evidence of diastolic dysfunction, 23% had pulmonary edema, 19% had a cTi level >1 μg/L, and 13% had a LVEF <50%. The mean of the subjects’ peak cTi levels was 2.7±8.2 μg/L (median, 0.3 μg/L; interquartile range, 0.3 to 0.5). The mean of the subjects’ lowest LVEF was 64% ± 15% (median 65%). Among the 50 subjects with diastolic dysfunction, 31 (55% of study cohort) had mild dysfunction (impaired relaxation) and 19 (34%) had high-grade dysfunction. The average time from SAH to each of 3 study days was 3.5±3.1, 5.3±2.6, and 8.3±2.7 days. The proportion of patients with a cTi level >1 μg/L was higher on study day 1 (13%) than study day 3 (6%). However, the other cardiac outcomes had similar rates across the 3 study days.
The Figure shows the relationships between plasma BNP levels and the measured cardiac end points. If the diastolic dysfunction analysis excluded patients older than age 65, for whom impaired relaxation is typical, the relationship between diastolic dysfunction and BNP levels was not markedly affected (P=0.054).
There was an increase in BNP levels in relation to worsening short-term outcomes (Table 2). By linear regression, patients who died during the hospitalization had higher log BNP levels (P=0.035).
This is the first study to our knowledge to demonstrate that cardiac injury and dysfunction occurring early after SAH are associated with elevated plasma BNP levels. Our findings are consistent with the hypothesis that BNP is released from the heart after SAH.
Cardiac injury and dysfunction are known to occur after SAH and the most likely mechanism is excessive myocardial catecholamine release.10 The results of the present study provide indirect evidence that the heart is the primary source of elevated BNP levels after SAH. The strong associations observed between cardiac dysfunction and BNP are consistent with what is known to occur in patients with congestive heart failure. The findings are novel in comparison to 2 previous SAH studies, which showed no correlation between levels of natriuretic peptides and either electrocardiographic abnormalities or troponin release.6,7⇓
Previous SAH studies have described an association between BNP levels and the development of cerebral vasospasm.11 In the present study, we observed that high BNP levels were significantly associated with inpatient mortality, which has not previously been reported.
In conclusion, this study provides novel evidence that cardiac injury and dysfunction occurring after SAH are associated with elevated plasma levels of BNP. These findings support the hypothesis that the heart releases BNP into the systemic circulation early after SAH. Furthermore, elevated BNP levels may have prognostic value, supporting the hypothesis that cardiac dysfunction contributes to poor neurological outcome after SAH.
Support for this study was provided by NIH/NHLBI 1 K23 HL04054-01 (PI: J.G.Z.), The Charles A. Dana Foundation (PI: J.G.Z.), and a gift from The Pritzker Cousins Foundation, John A. Pritzker, Director. These agencies had no direct role in the study design, data collection, analysis, interpretation, manuscript preparation, or the decision to submit the manuscript for publication.
- Received March 30, 2005.
- Revision received April 20, 2005.
- Accepted April 21, 2005.
Tomida M, Muraki M, Uemura K, Yamasaki K. Plasma concentrations of brain natriuretic peptide in patients with subarachnoid hemorrhage. Stroke. 1998; 29: 1584–1587.
Schiller N, Shah P, Crawford M, DeMaria A, Devereux R, Feigenbaum H, Gutgesell H, Reichek N, Sahn D, Schnittger I, al. e. Recommendations for quantitation of the left ventricle by two-dimensional echocardiography. Am Society of Echocardiography Committee on Standards, Subcommittee on Quantitation of Two-Dimensional Echocardiograms. J Am Soc Echocardiogr. 1989; 2: 358–367.