Safety of Hypertensive Hypervolemic Therapy With Phenylephrine in the Treatment of Delayed Ischemic Deficits After Subarachnoid Hemorrhage
Background and Purpose Hypertensive hypervolemic therapy has been shown to reverse delayed ischemic deficits after aneurysmal subarachnoid hemorrhage. Concern has been raised about systemic complications of therapy, including pulmonary edema and myocardial ischemia, especially when high doses of vasopressors are used. Patients in whom delayed ischemic deficits were treated with hypervolemia and phenylephrine were prospectively evaluated for signs of systemic toxicity.
Methods Twenty-four consecutive patients treated with hypertensive hypervolemic therapy after aneurysmal subarachnoid hemorrhage were studied. Sixty-seven percent had underlying cardiac disease, vascular disease, or hypertension. No patient was excluded because of age or preexisting cardiac disease. Patients were closely monitored for signs of congestive heart failure (physical examination, chest x-ray films, arterial blood gases, cardiac index, pulmonary artery wedge pressure, and oxygen requirement). Indicators of cardiac ischemia and other extracerebral toxicity that were monitored included cardiac enzymes, electrocardiograms, serum creatinine, electrolyte and lactic acid levels, gastrointestinal motility, and urine output.
Results Volume expansion and phenylephrine infusion produced an increase in several hemodynamic parameters including pulmonary artery wedge pressure, which rose 28% (13±3.6 to 16±1.9 mm Hg), mean arterial blood pressure, which rose 25% (99±12.5 to 123±11.4 mm Hg), and systemic vascular resistance, which rose 46% (1234±294 to 1739±315 dyne · s−1 · cm−5); however, there was no change in cardiac index (3.9±0.9 to 4.0±0.6 L · min−1 · m−2). There were no clinically significant episodes of pulmonary edema requiring a change in vasopressor therapy and no myocardial infarctions. Phenylephrine was stopped in only one patient (incidence, 4%; 95% confidence interval, 0% to 12%), who developed an exacerbation of his preexisting bradycardia. There was no evidence of noncardiac organ system toxicity. Eighty-eight percent of the patients exhibited neurological improvement.
Conclusions Hypertensive hypervolemic therapy with the use of high-dose phenylephrine can be administered with acceptable systemic toxicity, even in patients with previous cardiac disease, provided that close monitoring is performed. To minimize risk, aggressive treatment should probably be reserved for patients with signs of delayed ischemia rather than administered prophylactically.
After aneurysmal SAH, DID remain a leading cause of death and disability. The International Study on the Timing of Aneurysm Surgery estimated that ischemic deficits related to vasospasm led to death or permanent disability in 13.5% of patients.1
Vasospasm decreases the caliber of large cerebral arteries and is thought to lower distal cerebral perfusion pressure. This can reduce cerebral blood flow below the ischemic threshold, leading to disturbed brain metabolism and neurological deterioration.2 In addition, after SAH there is thought to be loss of autoregulation,3 and therefore cerebral blood flow is pressure passive.4 5 Neurological deterioration is potentially reversible if adequate blood flow can be restored before infarction occurs.
HHT is thought to reduce the incidence of DID and remains the cornerstone of medical management of vasospasm. In rare cases, intravascular volume expansion alone is adequate to augment blood pressure and thereby blood flow. However, in most recently published protocols of HHT, vasopressors are used to raise blood pressure.6 7 8
A variety of pharmacological agents have been used to induce hypertension, including dopamine, dobutamine, phenylephrine, and norepinephrine. All of these agents have potential adverse effects. For example, dopamine frequently produces an unacceptable degree of tachycardia or tachyarrhythmias. Dobutamine may lower blood pressure or cause arrhythmias.
Phenylephrine is a selective α1-adrenergic receptor agonist with essentially no β-adrenergic receptor activity at the intravenous doses used clinically.9 10 It increases blood pressure by producing marked arterial as well as venous vasoconstriction in normal volunteers.10 Phenylephrine decreases renal and splanchnic blood flow. In the coronary artery circulation it causes a moderate direct vasoconstriction. In septic patients, despite this vasoconstriction, phenylephrine improves coronary blood flow by increasing perfusion pressure. Phenylephrine has also been shown to increase CI in septic patients, in part by increasing venous return to the heart.11 12 Conversely, in nonseptic patients with underlying cardiac disease, it decreases CI.12 13 Phenylephrine bolus administration can cause a rapid increase in arterial blood pressure, leading to a baroreceptor-mediated reflex bradycardia.14 There is little published information regarding the cardiovascular effects of continuous phenylephrine infusion outside the setting of sepsis, nor is there much information regarding the safety and efficacy of the drug in normotensive or hypertensive patients. Noncardiac adverse effects of phenylephrine that have been reported reflect end-organ ischemia, including renal insufficiency and gastrointestinal ischemia from decreased renal and splanchnic blood flow.10
In our experience as well as the experience of others (D.F. Hanley and T.P. Bleck, conversation, 1995), very large doses of vasopressors are frequently needed to produce the desired degree of hypertension in patients with vasospasm. These doses may be several times those used in the treatment of sepsis.11 Concern is often raised regarding the potential for systemic toxicity with these large doses. In the treatment of DID, we routinely use volume expansion with crystalloids and/or colloids and induce hypertension with moderate to high doses of phenylephrine. Therapy is titrated to neurological status. We prospectively analyzed the clinical, hemodynamic, and metabolic effects of volume expansion in combination with phenylephrine infusion to determine whether clinically significant end-organ dysfunction occurs at the high doses of phenylephrine used in the treatment of DID.
Subjects and Methods
The study group consisted of 24 consecutive patients with aneurysmal SAH and DID treated in an NNICU of a university medical center between March 1993 and December 1994. The unit is staffed by a dedicated team of residents and a full-time neurointensivist. Patients with signs of ongoing myocardial ischemia were not candidates for HHT. The exclusion criteria were as follows: (1) >1 mm of ST segment elevation or depression in two or more ECG leads, (2) elevation of serum CPK-MB fraction (the cardiac-specific fraction) of ≥3% of the total, and (3) baseline MAP consistently >135 mm Hg without vasoactive drugs. Age, history of cardiopulmonary disease, or other unclipped aneurysms were not absolute criteria for exclusion. None of the patients with DID admitted to the NNICU during the study period were excluded based on these criteria.
All patients received standard therapy for SAH including nimodipine, dexamethasone, phenytoin, bed rest, and analgesia. Early surgical repair of the aneurysm was performed in the majority of patients, usually on posthemorrhage day 1 to 3. During the preoperative period, colloids or crystalloids were infused at rates calculated to be above maintenance requirements (>35 mL · kg−1 body wt · d−1). In all patients, fluid input was kept greater than output by approximately 0.5 L/d in an attempt to maintain a normal intravascular volume. All patients underwent continuous ECG and blood pressure monitoring by automated cuff or indwelling arterial catheter. Thermodilution Swan-Ganz catheters were placed before surgery and remained in place in the immediate postoperative period. In most stable patients they were removed on postoperative day 2 or 3. Swan-Ganz monitoring was continued in patients who had a history of or concomitant cardiopulmonary dysfunction. In all other patients they were reinserted as clinically indicated during HHT.
Intravascular volume was assessed throughout the period of potential vasospasm by monitoring CVP or PAWP trends, daily and hourly fluid balance, and daily body weight. At least twice a day these parameters were reviewed and adjustments were made in fluids.
DID was diagnosed when new neurological deficits, worsening of existing neurological deficits, or decreased level of consciousness occurred 2 to 21 days after aneurysm rupture. In all cases, other potential causes such as hydrocephalus, rebleeding, or cerebral edema were excluded by CT scan. Toxic and metabolic causes were excluded by appropriate laboratory testing. Cerebral angiography was often performed to confirm vasospasm but was not required for diagnosis of DID. In some cases interventions such as angioplasty or papaverine infusion were performed during angiography.
When DID developed, active volume expansion was accomplished with the infusion of crystalloids and/or colloids. Patients were transfused with packed red blood cells if their hematocrit fell to <30%. In patients with Swan-Ganz catheters, hypervolemic therapy was assessed by PAWP and CI, as well as by daily weight and hourly fluid balance. Fluid administration was assessed and adjusted multiple times per day to attempt to raise PAWP to 14 to 18 mm Hg or to the level that produced the highest CI for each patient. Diuretics were administered as clinically indicated for signs of fluid overload or congestive heart failure in 4 patients. Seven patients were managed during HHT without Swan-Ganz catheters.
Hypertensive therapy was initiated if a rapid bolus of crystalloid or colloid did not result in prompt neurological improvement. In many cases, the two therapies were initiated concurrently. To induce hypertension, phenylephrine was administered as a continuous intravenous infusion into a central vein with a starting dose of 20 μg/min (approximately 0.3 μg · kg−1 · min−1). The infusion was rapidly titrated to initially increase MAP to 20% to 25% above the patient’s baseline. That blood pressure goal was maintained for a period of approximately 2 to 4 hours. Neurological evaluations were performed hourly. Those patients who did not have improvement in their neurological deficits had their MAP further increased, and again an approximately 2- to 4-hour period of observation followed. There was no absolute ceiling for MAP. However, if elevation of MAP by >35% did not produce clinical improvement, patients were considered for adjunct therapies such as cerebral angioplasty or papaverine infusion.
Once an MAP was reached that achieved maximal improvement in the neurological deficits, it was maintained at that level for at least 2 days. If attempts to lower MAP by decreasing the phenylephrine infusion rate were associated with a decline in neurological status, MAP was raise to a level at which neurological function stabilized by increasing the phenylephrine infusion. After a 2- to 3-day period of stable neurological function, weaning of phenylephrine was again attempted.
Before the administration of phenylephrine, the following baseline data were collected: heart rate, MAP, ECG, serum CPK-MB level, creatinine level, Pao2, and lactate level. Patients with Swan-Ganz catheters had baseline CI, PAWP, and SVR recorded. The average MAP, heart rate, PAWP, and CI during a 24-hour period immediately before initiation of phenylephrine therapy was calculated as baseline.
During phenylephrine infusion patients had continuous ECG and oxygen saturation monitoring. Arterial blood pressure was measured by an indwelling arterial catheter in the radial or femoral artery, with MAP continuously displayed. Cardiac ischemia was assessed by observing the ST segment on bedside monitor, daily ECGs, and measurements of CPK-MB levels. Evidence of congestive heart failure was sought by monitoring CI and PAWP at least every 4 hours and continuously monitoring oxygen saturation. In addition, daily chest x-ray films were obtained in all patients with Swan-Ganz catheters or endotracheal tubes to assess the position of both as well as to assess for pulmonary congestion or infiltrates. All chest x-ray films were read by radiologists blinded to the patient’s therapy. When necessary, intravenous fluids were adjusted or diuretics administered (n=4) if a clinical exam, hemodynamic profile, or chest x-ray film indicated a decline in intravascular volume or fluid overload.
Renal function was assessed by daily measurement of serum creatinine level and urine output (<0.5 mL · kg−1 · h−1 was considered to be inadequate). Gastrointestinal function was assessed by tolerance or intolerance of oral or enteral feedings. Systemic ischemia was assessed by daily measurement of serum lactate concentration.
Four patients received dopamine before phenylephrine and dobutamine concurrent with phenylephrine. In a few cases nimodipine was discontinued if, despite evidence of adequate volume status as assessed by PAWP, CI, fluid balance, and weight, significantly larger infusions of phenylephrine were required to maintain MAP goals. Since some of these patients became less responsive during the period of relative hypotension, it was believed that the discontinuation of nimodipine was appropriate.
Phenylephrine infusion was discontinued if evidence of cardiac ischemia developed, defined as >1 mm ST depression or elevation in two or more ECG leads or an elevation in CPK-MB fraction of ≥3% of the total CPK.
Pre- and post-HHT hemodynamic values were normally distributed (coefficient of variation) and compared with paired t tests. A value of P≤.05 was considered statistically significant. All data are expressed as mean±SD.
A total of 24 patients were treated with HHT for DID. Clinical characteristics are presented in Table 1⇓. The mean patient age was 55 years, with a range of 30 to 81 years. Sixty-two percent of patients were female. Sixty-seven percent of patients had underlying medical conditions including hypertension, diabetes, peripheral vascular disease, and congestive heart failure. Clinical condition on admission by the Hunt and Hess classification was as follows: grade I, 21%; grade II, 25%; grade III, 46%; and grade IV, 8%.
All patients except three had unequivocal angiographic confirmation of vasospasm. Two patients had no angiographic vasospasm but were thought to have DID after thorough workup failed to reveal other causes for neurological decline. One patient had equivocal to mild vasospasm on angiogram. All patients except one had focal deficits and/or a decreased level of consciousness as manifestations of DID. The patient without a clinically demonstrable DID was unresponsive and had severe bilateral angiographic vasospasm. It was elected to treat this patient because it was believed that his neurological exam was insensitive to clinical changes from DID.
The infusion rates of phenylephrine used in our patient group were substantial, with a mean maximum dose of phenylephrine of 7.56±4.7 μg · kg−1 · min−1, and the dose range was broad, at 2.1 to 17.0 μg · kg−1 · min−1 (Table 2⇓). These doses are considerably higher than those typically used in hemodynamic support of sepsis.11 The average increase in MAP during maximum doses of phenylephrine was 25% (Fig 1⇓), and the average increase in SVR was 46%. There was no linear relationship between phenylephrine dose and MAP or SVR.
Large volumes of intravenous fluids (9.2±3.7 L/d) were required to sustain volume expansion as substantial diuresis occurred. All patients received a minimum of 5 L of fluid per day. HHT was associated with an average increase in PAWP of 28% (Fig 2⇓). The mean PAWP in our patient group during maximal HHT was 16±1.9 mm Hg. CI decreased by >10% in five patients. In all other patients it was unchanged or mildly elevated.
Twenty-one of 24 patients (88%) had neurological improvement during HHT demonstrated by sustained improvement in Glasgow Coma Scale score by one to two points and/or improvement or resolution of focal deficits (Table 1⇑). In some patients, neurological deficits (eg, aphasia or hemiparesis) clearly fluctuated with change in arterial blood pressure. In two instances, clinical improvement occurred within 30 minutes of initiation of therapy. Four patients did not clinically respond: two were Hunt and Hess grade III, one was Hunt and Hess grade IV, and one had hydrocephalus and improved after lumbar punctures.
Eight patients (33%) had one or more intra-arterial infusions of papaverine during their HHT, which may have contributed to clinical improvement. Two patients in our group had unruptured aneurysms in addition to their surgically treated aneurysms and received more cautious hypertensive therapy (ie, ≤20% increase in MAP) without untoward effects.
The duration of induced hypertension with phenylephrine ranged from 2 to 21 days, with a mean of 7 days. One patient had no clinical improvement, and it was therefore elected to stop phenylephrine after only 2 days. The duration of HHT was determined primarily by a patient’s clinical response to weaning of the therapy. Some patients had worsening neurological status during the taper and therefore required longer therapy. HHT was often continued longer in patients with severe angiographic vasospasm. There was no consistent relationship between maximum dose of phenylephrine and length of therapy.
Complications related to phenylephrine infusion included a one-time elevation of CPK-MB fraction to 3% in one patient with a normal ECG. The CPK-MB concentration normalized after phenylephrine was temporarily discontinued. Reinstitution of the drug within 24 hours was not associated with ECG changes or a repeated rise in CPK-MB levels. A second patient had T-wave inversions in several ECG leads. In this case phenylephrine was cautiously continued because the patient had no chest pain or elevation of CPK-MB. The ECG changes resolved after 4 days. There were no myocardial infarctions during the treatment period. A transient right bundle block pattern that occurred in the ECG of one patient resolved with removal of his Swan-Ganz catheter. Right bundle branch block is reported to be a complication in approximately 5% of patients with Swan-Ganz catheters.15 One patient who had intermittent bradycardia before treatment developed persistent bradycardia in the mid-40s while receiving phenylephrine. A junctional escape rhythm was seen on ECG without compromise of blood pressure. It was elected to rapidly taper and discontinue phenylephrine rather than treat the bradycardia with atropine or use another pressor agent because the patient’s neurological function had improved.
Mild to moderate interstitial infiltrates occurred on the chest x-ray films of nine patients (38%) at some time during the course of therapy. However, this was of minimal clinical significance in only four patients who had transient increases in oxygen requirements (in all cases, fractional inspired oxygen of ≤50%) and required diuretic therapy on one or more occasions. No patient required intubation to maintain adequate oxygenation. Three patients had evidence of fulminant pulmonary edema on admission to the hospital, with marked bilateral pulmonary infiltrates and need for mechanical ventilation and/or fractional inspired oxygen of >50%. In all cases the pulmonary edema resolved within 2 to 3 days and did not preclude later aggressive HHT.
Three patients had a single elevation of serum lactate level during phenylephrine therapy but demonstrated no other evidence of end-organ ischemia. Phenylephrine infusion was maintained at the same dose, and the lactate levels returned to normal within 24 hours. Renal toxicity did not occur in any patient. There were no instances of gastrointestinal ischemia.
One patient who had a femoral artery 7F introducer left in place for several days after a cerebral angiogram developed ischemia in two toes of the ipsilateral leg. This was believed to be embolic in origin, and there was no evidence that phenylephrine was a contributing factor. There were no other complications, such as pneumothorax or bacteremia, due to the use of invasive arterial or central venous catheters.
There is a convincing body of evidence that total intravascular volume is decreased after SAH and that volume contraction is a risk factor for vasospasm.16 17 Hypervolemic therapy can ameliorate this decrease in intravascular volume but is not always effective in preventing DID. Reversal of ischemic deficits with the use of hypervolemia in conjunction with vasopressors to induce hypertension has been demonstrated.6 7 8 18 Choice of therapy and aggressiveness of treatment are tempered by concern about causing systemic complications such as cardiac ischemia or pulmonary edema.
In this study we describe the hemodynamic profile produced by HHT using high-dose phenylephrine in SAH patients with DID. Phenylephrine was selected as a vasopressor agent for several reasons: it is a pure α1-adrenergic receptor agonist and therefore does not produce tachycardia or tachyarrhythmias as can dopamine or dobutamine, it can be titrated rapidly, and it produces sustained hypertension. Since cerebral blood vessels have been shown to have a very low density of α1-receptors19 and cerebral blood vessels are thought to be relatively insensitive to adrenergic agents,20 phenylephrine should not produce significant direct cerebral vasoconstriction.
Our patient population was similar to patients in other studies of HHT with regard to age and Hunt and Hess grade.8 21 22 Unlike some previous studies,22 however, we did not exclude elderly patients or patients with preexisting cardiopulmonary disease.
Eighty-eight percent of our patients demonstrated neurological improvement during phenylephrine infusion. Similar to previous investigators, we found clinical improvement more frequently in patients with Hunt and Hess grades I to III.8 Attributing clinical improvement to HHT is problematic in that one cannot definitively conclude that outcome after treatment differed from the natural history of vasospasm without treatment. Also, in several instances multiple therapies were administered simultaneously (dobutamine or papaverine/angioplasty with HHT). This could be further evaluated by measurement of cerebral blood flow.
Phenylephrine produced significant and sustained elevation in MAP and SVR. Cardiac function, as assessed by CI, was never significantly compromised or augmented throughout therapy. Despite the increased afterload on the heart and the direct coronary artery constriction effect of phenylephrine, there was no myocardial ischemia or pulmonary edema severe enough to alter pressor therapy. This is in contrast to earlier investigators who, using aggressive prophylactic hypervolemia in patients monitored with Swan-Ganz catheters, found that 26% of patients developed pulmonary edema sufficient to require a change in management.21 Similarly, Coyne et al22 reported pulmonary edema in 34% of their patients, one half of whom required mechanical ventilation. One patient required long-term treatment for heart failure. Both groups concluded that prophylactic hypervolemia did not lead to a decreased incidence of neurological morbidity and was associated with an unacceptably high incidence of complications.
One possible explanation for the differing results in the present study is the timing of initiation of therapy. We did not use HHT prophylactically; rather, we attempted to maintain normal intravascular volume and did not treat moderate elevations of blood pressure. Aggressive therapy was only initiated at the onset of DID, which usually occurred on posthemorrhage day 5 to 10. It appears that in some patients during the immediate few days after SAH, there exists a depression of myocardial function.23 Hypervolemia and hypertension during that time may cause a higher propensity to congestive heart failure. There are well-documented cases demonstrating reversible neurogenic cardiac injury with manifested ECG changes,24 cardiac enzyme elevation,25 and myocardial wall motion abnormalities on echocardiography.26 27 28
Phenylephrine was permanently discontinued because of cardiac concerns in only one patient: a 44-year-old man with a heart rate of 50 to 60 beats per minute before initiation of therapy. While the patient received phenylephrine, his heart rate decreased over several days to the low 40s, with the development of junctional escape beats. The patient was asymptomatic and had no changes on ECG or cardiac enzymes elevation suggestive of ischemia. Therapeutic options would have included treating the bradycardia with anticholinergic agents such as atropine or glycopyrrolate while continuing phenylephrine or switching to another pressor with β-activity such as dopamine or norepinephrine. Since the patient had stabilized neurologically, it was elected to rapidly taper and discontinue phenylephrine. The patient remained neurologically stable during the taper, and his heart rate rose to 60 to 70 beats per minute within 2 hours of discontinuation of phenylephrine.
Very high doses of phenylephrine were required to maintain the target MAP and SVR in some patients. In the absence of sepsis or hypovolemia, it is possible that this represents tachyphylaxis (not previously reported with phenylephrine). Alternatively, at higher doses, β-receptor activation may occur. In some of these cases we attempted to improve CI and reduce phenylephrine requirement by concurrently administering dobutamine. This strategy often led to a fall in MAP, probably due to the β2 effects of dobutamine. An alternative approach to the escalating doses would be to add or change to another pressor such as norepinephrine; however, this agent may be more likely to produce systemic ischemia.
Hypervolemic hyperdynamic therapy that includes the use of intravascular volume expansion and dobutamine to enhance cardiac output with minimal elevation of blood pressure has been advocated by some investigators.29 There have been no recent published reports on the use of hypervolemia alone to treat delayed ischemic deficits due to vasospasm.
HHT is made safer when coupled with meticulous monitoring that can be best delivered in a specialized ICU staffed with a core of neurointensivists and specialized ICU nursing staff. Intravascular volume status should be assessed every few hours and fluid adjustments made when necessary.
Based on our experience, monitoring for toxicity should probably include the following: continuous monitoring of ECG and oxygen saturation, daily chest x-ray films for patients with Swan-Ganz catheters and those in whom pulmonary edema is a concern, daily ECGs, daily measurement of weight, hourly measurement of fluid balance, and daily measurement of CPK-MB levels. We did not find lactate or creatinine levels to be helpful. Swan-Ganz catheters should be placed and cardiovascular parameters monitored every 2 to 4 hours in patients with or at risk for cardiopulmonary disease. While CVP monitoring by itself is not a reliable means of assessing intravascular volume status, in some patients without cardiopulmonary disease, PAWP and CVP are found to be roughly correlated. In those instances Swan-Ganz catheters may be removed, and CVP monitoring (with its lower morbidity and cost) can be used in addition to fluid balance and weight for assessment of intravascular volume.
While hypovolemia should always be avoided, the most advantageous time to use HHT may be at the onset of neurological deficits. Shortening the duration of aggressive HHT may result in decreased complications. Prophylactic HHT during the period of reversible neurogenic cardiac injury may result in higher morbidity. Finally, recent data suggest that prophylactic hypervolemia does not alter cerebral blood flow30 and may offer no clinical advantage over normovolemia.
While hypertensive therapy can be used most safely only in those patients with all of their aneurysms surgically excluded from the circulation, we induced a modest degree of hypertension in two patients with unruptured aneurysms. We believe that it may be reasonable to accept the unknown risk of aneurysmal rupture in the face of ongoing cerebral ischemia.
In summary, we find that with careful hemodynamic monitoring, volume expansion and hypertension induced with high-dose phenylephrine for up to 3 weeks is associated with minimal systemic complications.
Selected Abbreviations and Acronyms
|CVP||=||central venous pressure|
|DID||=||delayed ischemic deficits|
|HHT||=||hypertensive hypervolemic therapy|
|MAP||=||mean arterial pressure|
|NNICU||=||Neurology/Neurosurgery Intensive Care Unit|
|PAWP||=||pulmonary artery wedge pressure|
|SVR||=||systemic vascular resistance|
The authors gratefully acknowledge the assistance of Allison Thornton in the preparation of this manuscript.
- Received May 18, 1995.
- Revision received August 4, 1995.
- Accepted August 22, 1995.
- Copyright © 1995 by American Heart Association
Pickard JD, Boisvert DPJ, Graham DI. Late effects of subarachnoid hemorrhage on the response of primate cerebral circulation to drug induced changes in arterial blood pressure. J Neurol Neurosurg Psychiatry. 1979;42:899-903.
Awad IA, Carter LP, Spetzler RF, Medina M, Williams FW. Clinical vasospasm after subarachnoid hemorrhage: response to hypervolemic hemodilution and arterial hypertension. Stroke. 1987;18:365-372.
Heerdt PM, Forstot RM. Heart rate control. In: Chernow B, ed. Essentials of Critical Care Pharmacology, 2nd ed. Baltimore, Md: Williams & Wilkins Co; 1993:297-312.
Starkey K, Docherty JR. Alpha-1 and alpha-2 adrenoreceptors: pharmacology and clinical implications. J Cardiovasc Pharmacol. 1981;3(suppl 1):514-516.
Iberti TJ, Silverstein JH. Complications of pulmonary artery catheterization. In: Sprung CL, ed. The Pulmonary Artery Catheter. 2nd ed. Closter, NJ: Critical Care Research Associates, Inc; 1993.
Olesen J. The effect of intracarotid epinephrine, norepinephrine, and angiotensin on the regional cerebral blood flow in man. Neurology. 1972;22:978-987.
Coyne TJ, Wallace MC, Sinclair F. Pulmonary edema following prophylactic hypervolemia after subarachnoid hemorrhage. Stroke. 1994;25:248. Abstract.
Mayer SA, Fink ME, Homma S, Sherman D, Li Mandri G, Lennihan L, Solomom RA, Klebanoff LM, Beckford A, Raps EC. Cardiac injury associated with neurogenic pulmonary edema following subarachnoid hemorrhage. Neurology. 1994;44:815-820.
Cruickshank JM, Neil-Dwyer G, Brice J. ECG changes and their prognostic significance in SAH. J Neurol Neurosurg Psychiatry. 1974;37:755-759.
Norris JW, Hachinski VC, Myers MG. Serum cardiac enzymes in stroke. Stroke. 1979;10:548-553.
Hart GK, Humphrey L, Weiss J. Subarachnoid hemorrhage: cardiac complications. Crit Care Report. 1989;1:88-92.
Davies KR, Gelb AW, Manninen PH, Boughner DR, Bisnaire D. Cardiac function in aneurysmal subarachnoid hemorrhage: a study of ECG and echocardiographic abnormalities. Br J Anaesth. 1991;67:58-63.
Levy ML, Rabb CH, Zelman V, Giannotta SL. Cardiac performance enhancement from dobutamine in patients refractory to hypervolemic therapy for cerebral vasospasm. Neurosurgery. 1993;79:494-499.
Lennihan L, Solomon RA, Mayer S, Prohovnik I, Fink M, Klebanoff L, Beckford A, Paik M, Wu Y. Effect of volume therapy on cerebral blood flow after subarachnoid hemorrhage. Neurology. 1994;44:A345. Abstract.