Article
Plasma Estrogen Levels Are Associated With Severity of Injury and Outcomes After Aneurysmal Subarachnoid Hemorrhage
Biological Research for Nursing 2015, Vol. 17(5) 558-566 ª The Author(s) 2014 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/1099800414561632 brn.sagepub.com
Elizabeth A. Crago, RN, PhD1, Paula R. Sherwood, RN, PhD1, Catherine Bender, RN, PhD1, Jeffrey Balzer, PhD1,2, Dianxu Ren, PhD1, and Samuel M. Poloyac, PharmD, PhD3
Abstract Background: Biochemical mediators alter cerebral perfusion and have been implicated in delayed cerebral ischemia (DCI) and poor outcomes after aneurysmal subarachnoid hemorrhage (aSAH). Estrogens (estrone [E1] and estradiol [E2]) are mediators with neuroprotective properties that could play a role in DCI. This study explored associations between plasma estrogen levels and outcomes following aSAH. Methods: Plasma samples from 1–4, 4–6, and 7–10 days after hemorrhage from 99 adult aSAH patients were analyzed for estrogen levels using liquid chromatography tandem mass spectrometry. DCI was operationalized as radiographic/ultrasonic evidence of impaired cerebral blood flow accompanied by neurological deterioration. Outcomes were assessed using the Modified Rankin Scale at 3 and 12 months after hemorrhage. Statistical analysis included correlation, regression, and group-based trajectory. Results: Higher E1 and E2 levels were associated with higher Hunt and Hess grade (E1, p ¼ .01; E2, p ¼ .03), the presence of DCI (E1, p ¼ .02; E2, p ¼ .02), and poor 3-month outcomes (E1, p ¼ .002; E2, p ¼ .002). Trajectory analysis identified distinct populations over time for E1 (61% E1 high) and E2 (68% E2 high). Patients in higher trajectory groups had higher Fisher grades (E1, p ¼ .008; E2, p ¼ .01), more frequent DCI (E1, p ¼ .04; E2, p ¼ .08), and worse 3-month outcomes (E1, p ¼ .01; E2, p ¼ .004) than low groups. Conclusions: These results provide the first clinical evidence that plasma E1 and E2 concentrations are associated with severity of injury and outcomes after aSAH. Keywords subarachnoid hemorrhage, delayed cerebral ischemia, estrogen, outcomes
Aneurysmal subarachnoid hemorrhage (aSAH) is a sudden and devastating neurological event affecting approximately 33,000 individuals per year in the United States at a mean age of 55 years (with a higher incidence in women, Blacks, and Hispanics; de Rooij, Rinkel, Dankbaar, & Frijns, 2013). Despite advances in medical and surgical management, aSAH continues to be associated with complications necessitating vigilant monitoring and emergent interventions during the acute recovery period (Sehba, Pluta, & Zhang, 2011; Zubkov & Rabinstein, 2009). In addition to the initial bleed and despite early aneurysm repair, a large population of patients develops delayed neurological deficits and delayed cerebral ischemia (DCI; Cahill & Zhang, 2009). The rate of mortality and disability is greater than for other stroke etiologies, leaving survivors with profound physical and neuropsychological disability (de Rooij et al., 2013). Historically, prevention and recognition of cerebral vasospasm has been the primary target for improving outcomes after aSAH; however, clinical trials in which vasospasm was reduced have not
resulted in improved outcomes (Ma et al., 2012; Vergouwen, Algra, & Rinkel, 2012). As a result, focus has shifted to early brain injury that may be precipitated by injury to the brain’s neurovascular unit as a consequence of the hemorrhage and elevation in intracranial pressure. In particular, biochemical changes that alter the balance of vasoactive metabolites may impact recovery and outcomes following acute neurological
1
School of Nursing, University of Pittsburgh, Pittsburgh, PA, USA Neurological Surgery, University of Pittsburgh Medical Center, Pittsburgh, PA, USA 3 School of Pharmaceutical Sciences, University of Pittsburgh, Pittsburgh, PA, USA 2
Corresponding Author: Elizabeth A. Crago, RN, PhD, Acute and Tertiary Care, School of Nursing, University of Pittsburgh, 320B Victoria Building, 3500 Victoria Street, Pittsburgh, PA 15261, USA. Email: [email protected]
Downloaded from brn.sagepub.com at UNIV OF PITTSBURGH on November 14, 2015
Crago et al.
559
insult (Caner, Hou, Altay, Fuj, & Zhang, 2012; Crago et al., 2011; Sehba et al., 2011). Estrogen has been widely studied in neurological disorders including both experimental ischemic and hemorrhagic insults (Azcoitia, Arevalo, De Nicola, & Garcia-Segura, 2011; Herson, Koerner, & Hurn, 2009). E2 alters vascular reactivity, reduces tissue damage, improves functional recovery, and may stimulate repair processes in the central nervous system (Hurn & Brass, 2003; Liu, Dziennis, Hurn, & Alkayed, 2009). The absence of E2 has been linked to vasoconstriction, impaired cellular metabolism and cellular death in animal models (Azcoitia et al., 2011; Brown, Suzuki, Jelks, & Wise, 2009; Lin et al., 2009; Liu et al., 2009). In addition, researchers have attributed a lower incidence of ischemic neurologic insults and neurodegenerative diseases as well as improved outcomes after traumatic brain injury (TBI) in premenopausal females compared to men of equal age to higher estrogen levels (Liu et al., 2009; Miller & CroninGolomb, 2010; Sealy-Jefferson et al., 2012). E1 is a physiologic estrogen that is the predominant estrogen remaining after menopause. In experimental studies, E1 has been related to vascular reactivity and vasodilation through nitric oxide pathways (Rauschemberger, Sandoval, & Massheimer, 2011). It has also demonstrated neuroprotective properties in stroke, reperfusion injury, and TBI animal studies (Gatson et al., 2012). In studies of patients with critical illness, associations between estrogen levels and improved outcomes have been inconsistent (Gatson et al., 2012; Selles, Polini, Alvarez, & Massheimer, 2005). No study to date has explored whether estrogen levels are related to the incidence of DCI or outcomes in patients after aSAH. The purpose of this study was to determine the relationship between plasma E1 and E2 concentrations and outcomes (DCI and Modified Rankin Scale [MRS]) following aSAH.
Materials and Methods Patient Sample This study was a secondary analysis of prospectively collected data from aSAH patients admitted to the neurovascular intensive care unit who were enrolled in a National Institute of Health (NIH)-funded study (R01NR004339) exploring the relationships between biomarkers and complications of aSAH. For this analysis, we drew a convenience sample of eligible patients from the parent study between 2006 and 2011. Inclusion criteria for the parent study were (1) 21–75 years old, (2) verified aSAH, and (3) Fisher grade > 1 admitted within 3 days of aneurysm rupture. Exclusion criteria were (1) nonaneurysmal SAH, (2) mycotic aneurysm, or (3) history of preexisting debilitating neurologic disease. The sample was stratified to include women (both pre- and postmenopausal) and men. Patients taking any type of hormone or hormone replacement therapy were excluded. The parent study was approved by the University of Pittsburgh Institutional Review Board and informed consent was obtained from the patient or proxy prior to data collection.
Data Collection We collected sociodemographic data, including gender, age, race, menopausal status, severity of injury (Fisher grade and Hunt and Hess [HH] score), and clinical data from the patient/proxy and medical record. Estrogen. As a part of the parent study, plasma samples were collected each morning after initial hemorrhage, centrifuged and aliquoted into cryovials within 60 minutes of collection, and stored at 80 C until analysis. To capture changes in circulating estrogen levels (E1 and E2) during the time period of highest risk for ischemic complications (Days 3–10 after hemorrhage), we pulled and analyzed three plasma samples per subject: the first available sample after consent (generally acquired Days 1–3 after hemorrhage), a Day 5 sample (or middle sample, Days 5–7), and a Day 10 sample (or last sample, Days 9–10). E1 and E2 were assayed at the University of Pittsburgh (UP) Small Molecule Biomarker Core using liquid chromatographytandem mass spectrometry (LC-MS/MS). The UP LC-MS/MS method employs a validated liquid–liquid extraction, derivatization, and detection with a triple quad mass spectrometer as previously described by Nelson et al. with modifications (Nelson, Grebe, O’Kane, & Singh, 2004). The calibration curves, obtained from extracting known concentrations of E1 and E2 from 2-hydroxypropyl-b-cyclodextrin (0.2% in water), ranged from 0.5 pg/ml (lower limit of quantitation) to 200 pg/ml. All back calculations of calibrators, interday and intraday precision and accuracy, and stability were within acceptable limits (1 point] in score on the National Institute of Health Stroke Scale [NIHSS]; decline [>1 point] in the Glasgow Coma Scale score; deterioration in pupil reaction; or documentation of a focal deficit and/or decline in level of consciousness [lethargy, agitation, and confusion] not associated with seizure, hydrocephalus, fever, or medication administration [e.g., sedatives or paralytics]). Impaired CBF was defined using surrogate measurements including cerebral angiography, transcranial Doppler (TCD), and computerized tomography (CT) or magnetic resonance (MR) perfusion scans. Angiographic vasospasm was determined from cerebral angiograms read and coded by neurosurgeons blinded to participant identity and dichotomized as either ‘‘negative’’ or ‘‘positive’’ (25% narrowing of cerebral blood vessels). Daily TCDs were coded as abnormal when a systolic middle cerebral artery velocity >200 ml/sec and/or a Lindegaard ratio > 3.0 was identified. Finally, head CT/MR and head CT/MR perfusion scans were reviewed by study personnel for the presence of cerebral ischemia, infarction, or low blood flow in the absence of hydrocephalus or rebleed. All patients received the study institution’s standard therapy for
Downloaded from brn.sagepub.com at UNIV OF PITTSBURGH on November 14, 2015
560
Biological Research for Nursing 17(5)
SAH patients including strict blood pressure and central venous pressure parameters, nimodipine, hypertensive and hypervolemic therapy.
Table 1. Demographic Data and Plasma Estrogen Analysis.
Outcomes. Outcomes were measured using the MRS at 3 and 12 months after aSAH. Assessments were completed as a part of the parent study during a face-to-face or telephone interview with the patient or primary caregiver. Mortality was determined from follow-up phone calls or hospital records.
Variable
Data Analysis Neither E1 nor E2 data were normally distributed; therefore, we applied data transformation (natural and square root logarithmic). Since few metabolite measurements were not quantifiable or below limits of detection (E1: 1 sample [
Plasma Estrogen Levels Are Associated With Severity of Injury and Outcomes After Aneurysmal Subarachnoid Hemorrhage
Biological Research for Nursing 2015, Vol. 17(5) 558-566 ª The Author(s) 2014 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/1099800414561632 brn.sagepub.com
Elizabeth A. Crago, RN, PhD1, Paula R. Sherwood, RN, PhD1, Catherine Bender, RN, PhD1, Jeffrey Balzer, PhD1,2, Dianxu Ren, PhD1, and Samuel M. Poloyac, PharmD, PhD3
Abstract Background: Biochemical mediators alter cerebral perfusion and have been implicated in delayed cerebral ischemia (DCI) and poor outcomes after aneurysmal subarachnoid hemorrhage (aSAH). Estrogens (estrone [E1] and estradiol [E2]) are mediators with neuroprotective properties that could play a role in DCI. This study explored associations between plasma estrogen levels and outcomes following aSAH. Methods: Plasma samples from 1–4, 4–6, and 7–10 days after hemorrhage from 99 adult aSAH patients were analyzed for estrogen levels using liquid chromatography tandem mass spectrometry. DCI was operationalized as radiographic/ultrasonic evidence of impaired cerebral blood flow accompanied by neurological deterioration. Outcomes were assessed using the Modified Rankin Scale at 3 and 12 months after hemorrhage. Statistical analysis included correlation, regression, and group-based trajectory. Results: Higher E1 and E2 levels were associated with higher Hunt and Hess grade (E1, p ¼ .01; E2, p ¼ .03), the presence of DCI (E1, p ¼ .02; E2, p ¼ .02), and poor 3-month outcomes (E1, p ¼ .002; E2, p ¼ .002). Trajectory analysis identified distinct populations over time for E1 (61% E1 high) and E2 (68% E2 high). Patients in higher trajectory groups had higher Fisher grades (E1, p ¼ .008; E2, p ¼ .01), more frequent DCI (E1, p ¼ .04; E2, p ¼ .08), and worse 3-month outcomes (E1, p ¼ .01; E2, p ¼ .004) than low groups. Conclusions: These results provide the first clinical evidence that plasma E1 and E2 concentrations are associated with severity of injury and outcomes after aSAH. Keywords subarachnoid hemorrhage, delayed cerebral ischemia, estrogen, outcomes
Aneurysmal subarachnoid hemorrhage (aSAH) is a sudden and devastating neurological event affecting approximately 33,000 individuals per year in the United States at a mean age of 55 years (with a higher incidence in women, Blacks, and Hispanics; de Rooij, Rinkel, Dankbaar, & Frijns, 2013). Despite advances in medical and surgical management, aSAH continues to be associated with complications necessitating vigilant monitoring and emergent interventions during the acute recovery period (Sehba, Pluta, & Zhang, 2011; Zubkov & Rabinstein, 2009). In addition to the initial bleed and despite early aneurysm repair, a large population of patients develops delayed neurological deficits and delayed cerebral ischemia (DCI; Cahill & Zhang, 2009). The rate of mortality and disability is greater than for other stroke etiologies, leaving survivors with profound physical and neuropsychological disability (de Rooij et al., 2013). Historically, prevention and recognition of cerebral vasospasm has been the primary target for improving outcomes after aSAH; however, clinical trials in which vasospasm was reduced have not
resulted in improved outcomes (Ma et al., 2012; Vergouwen, Algra, & Rinkel, 2012). As a result, focus has shifted to early brain injury that may be precipitated by injury to the brain’s neurovascular unit as a consequence of the hemorrhage and elevation in intracranial pressure. In particular, biochemical changes that alter the balance of vasoactive metabolites may impact recovery and outcomes following acute neurological
1
School of Nursing, University of Pittsburgh, Pittsburgh, PA, USA Neurological Surgery, University of Pittsburgh Medical Center, Pittsburgh, PA, USA 3 School of Pharmaceutical Sciences, University of Pittsburgh, Pittsburgh, PA, USA 2
Corresponding Author: Elizabeth A. Crago, RN, PhD, Acute and Tertiary Care, School of Nursing, University of Pittsburgh, 320B Victoria Building, 3500 Victoria Street, Pittsburgh, PA 15261, USA. Email: [email protected]
Downloaded from brn.sagepub.com at UNIV OF PITTSBURGH on November 14, 2015
Crago et al.
559
insult (Caner, Hou, Altay, Fuj, & Zhang, 2012; Crago et al., 2011; Sehba et al., 2011). Estrogen has been widely studied in neurological disorders including both experimental ischemic and hemorrhagic insults (Azcoitia, Arevalo, De Nicola, & Garcia-Segura, 2011; Herson, Koerner, & Hurn, 2009). E2 alters vascular reactivity, reduces tissue damage, improves functional recovery, and may stimulate repair processes in the central nervous system (Hurn & Brass, 2003; Liu, Dziennis, Hurn, & Alkayed, 2009). The absence of E2 has been linked to vasoconstriction, impaired cellular metabolism and cellular death in animal models (Azcoitia et al., 2011; Brown, Suzuki, Jelks, & Wise, 2009; Lin et al., 2009; Liu et al., 2009). In addition, researchers have attributed a lower incidence of ischemic neurologic insults and neurodegenerative diseases as well as improved outcomes after traumatic brain injury (TBI) in premenopausal females compared to men of equal age to higher estrogen levels (Liu et al., 2009; Miller & CroninGolomb, 2010; Sealy-Jefferson et al., 2012). E1 is a physiologic estrogen that is the predominant estrogen remaining after menopause. In experimental studies, E1 has been related to vascular reactivity and vasodilation through nitric oxide pathways (Rauschemberger, Sandoval, & Massheimer, 2011). It has also demonstrated neuroprotective properties in stroke, reperfusion injury, and TBI animal studies (Gatson et al., 2012). In studies of patients with critical illness, associations between estrogen levels and improved outcomes have been inconsistent (Gatson et al., 2012; Selles, Polini, Alvarez, & Massheimer, 2005). No study to date has explored whether estrogen levels are related to the incidence of DCI or outcomes in patients after aSAH. The purpose of this study was to determine the relationship between plasma E1 and E2 concentrations and outcomes (DCI and Modified Rankin Scale [MRS]) following aSAH.
Materials and Methods Patient Sample This study was a secondary analysis of prospectively collected data from aSAH patients admitted to the neurovascular intensive care unit who were enrolled in a National Institute of Health (NIH)-funded study (R01NR004339) exploring the relationships between biomarkers and complications of aSAH. For this analysis, we drew a convenience sample of eligible patients from the parent study between 2006 and 2011. Inclusion criteria for the parent study were (1) 21–75 years old, (2) verified aSAH, and (3) Fisher grade > 1 admitted within 3 days of aneurysm rupture. Exclusion criteria were (1) nonaneurysmal SAH, (2) mycotic aneurysm, or (3) history of preexisting debilitating neurologic disease. The sample was stratified to include women (both pre- and postmenopausal) and men. Patients taking any type of hormone or hormone replacement therapy were excluded. The parent study was approved by the University of Pittsburgh Institutional Review Board and informed consent was obtained from the patient or proxy prior to data collection.
Data Collection We collected sociodemographic data, including gender, age, race, menopausal status, severity of injury (Fisher grade and Hunt and Hess [HH] score), and clinical data from the patient/proxy and medical record. Estrogen. As a part of the parent study, plasma samples were collected each morning after initial hemorrhage, centrifuged and aliquoted into cryovials within 60 minutes of collection, and stored at 80 C until analysis. To capture changes in circulating estrogen levels (E1 and E2) during the time period of highest risk for ischemic complications (Days 3–10 after hemorrhage), we pulled and analyzed three plasma samples per subject: the first available sample after consent (generally acquired Days 1–3 after hemorrhage), a Day 5 sample (or middle sample, Days 5–7), and a Day 10 sample (or last sample, Days 9–10). E1 and E2 were assayed at the University of Pittsburgh (UP) Small Molecule Biomarker Core using liquid chromatographytandem mass spectrometry (LC-MS/MS). The UP LC-MS/MS method employs a validated liquid–liquid extraction, derivatization, and detection with a triple quad mass spectrometer as previously described by Nelson et al. with modifications (Nelson, Grebe, O’Kane, & Singh, 2004). The calibration curves, obtained from extracting known concentrations of E1 and E2 from 2-hydroxypropyl-b-cyclodextrin (0.2% in water), ranged from 0.5 pg/ml (lower limit of quantitation) to 200 pg/ml. All back calculations of calibrators, interday and intraday precision and accuracy, and stability were within acceptable limits (1 point] in score on the National Institute of Health Stroke Scale [NIHSS]; decline [>1 point] in the Glasgow Coma Scale score; deterioration in pupil reaction; or documentation of a focal deficit and/or decline in level of consciousness [lethargy, agitation, and confusion] not associated with seizure, hydrocephalus, fever, or medication administration [e.g., sedatives or paralytics]). Impaired CBF was defined using surrogate measurements including cerebral angiography, transcranial Doppler (TCD), and computerized tomography (CT) or magnetic resonance (MR) perfusion scans. Angiographic vasospasm was determined from cerebral angiograms read and coded by neurosurgeons blinded to participant identity and dichotomized as either ‘‘negative’’ or ‘‘positive’’ (25% narrowing of cerebral blood vessels). Daily TCDs were coded as abnormal when a systolic middle cerebral artery velocity >200 ml/sec and/or a Lindegaard ratio > 3.0 was identified. Finally, head CT/MR and head CT/MR perfusion scans were reviewed by study personnel for the presence of cerebral ischemia, infarction, or low blood flow in the absence of hydrocephalus or rebleed. All patients received the study institution’s standard therapy for
Downloaded from brn.sagepub.com at UNIV OF PITTSBURGH on November 14, 2015
560
Biological Research for Nursing 17(5)
SAH patients including strict blood pressure and central venous pressure parameters, nimodipine, hypertensive and hypervolemic therapy.
Table 1. Demographic Data and Plasma Estrogen Analysis.
Outcomes. Outcomes were measured using the MRS at 3 and 12 months after aSAH. Assessments were completed as a part of the parent study during a face-to-face or telephone interview with the patient or primary caregiver. Mortality was determined from follow-up phone calls or hospital records.
Variable
Data Analysis Neither E1 nor E2 data were normally distributed; therefore, we applied data transformation (natural and square root logarithmic). Since few metabolite measurements were not quantifiable or below limits of detection (E1: 1 sample [
