Nonspecific Behaviors As Early Indications of Cerebral Vasospasm

Written By: Sam Savage

Published Date: January 10, 2007 Last Edited: January 10, 2007

By Doerksen, Kathy; Naimark, Barbara

Abstract: This study investigated nonspecific behaviors as early indications of vasospasm following subarachnoid hemorrhage. Although symptoms of vasospasm (e.g., lowered level of consciousness, focal deficits such as hemiplegia or aphasia), are well recognized, the significance of early appearance of nonspecific symptoms such as restlessness, unusual behaviors, and impulsive behavior has not been investigated in detail. The study design included descriptive quantitative elements and a small qualitative component. Nonspecific behaviors were recorded, and the prevalence of those behaviors in individuals developing vasospasm was noted. Of 60 participants, 31 developed vasospasm; 24 of the 31 initially presented with nonspecific behaviors (p

Early detection is critical in allowing prompt intervention and treatment of cerebral vasospasm, with the goal of preventing ischemia or infarction. Although symptoms of vasospasm (e.g., a decreased level of consciousness) or the onset of focal deficits (e.g., facial weakness, limb drift, hemiplegia, aphasia) are well recognized (Qureshi, Sung, et al., 2000), the significance of vague symptoms such as restlessness, unusual behaviors, and impulsive behaviors is less well understood. This study investigated the incidence of vague symptoms exhibited by patients who are at risk of vasospasm following subarachnoid hemorrhage.

Background

Subarachnoid hemorrhage (SAH) resulting from aneurysmal rupture occurs in approximately 30,000 people per year in North America (Ullman & Bederson, 1996). Of people who experience SAH, 22%-40% will develop symptomatic cerebral vasospasm (Haley, Kassell, & Torner, 1993; Qureshi, Suarez, et al., 2000). Cerebral vasospasm is a narrowing of the lumen of one or more cerebral arteries. It can result in ischemia of cerebral tissue, and it is the primary cause of disability and death in individuals with a subarachnoid hemorrhage due to an aneurysm rupture (Cook, 2004; Haley, et al.).

Vasospasm generally occurs 4-14 days after aneurysm rupture, with peak occurrence at 7-10 days (Powsner, O’Tuama, Jabre, & Melhem, 1998; Qureshi, Sung, et al., 2000; Ullman & Bederson, 1996). Symptoms of vasospasm include focal deficits and a decreased level of consciousness (Cook, 2004; Oropello, Weiner, & Benjamin, 1996). Recently, Doerksen, Naimark, and Tate (2002) reported that of 30 study participants admitted with SAH from aneurysmal rupture exhibited vague symptoms such as restlessness, unusual behaviors, and impulsive behaviors; they went on to develop vasospasm 5 hours to 5 days after exhibiting these symptoms. Interestingly, two of the nine patients had a Glasgow Coma Scale (GCS) score indicating a normal level of alertness or consciousness. The literature contains minimal information on this phenomenon. Unterberg, Sakowitz, Sarrafzadeh, Benndorf, and Lanksch (2001) reported that 8 of 60 patients presented with “altered mental status” concurrently with vasospasm. There was no detailed description of the altered mental status.

Research Question

The question for this study was the following: What is the association of nonspecific behaviors with the incidence of cerebral vasospasm?

Design

A descriptive quantitative design was used, with a small qualitative component to provide a more detailed description of nurse-documented behaviors exhibited by the patients. Diagnostic tests (e.g., computed tomography [CT] and serum sodium) were conducted per the standard of care to rule out confounding factors. To link the presence of the examined nonspecific behaviors with recovery, a statement about outcome was completed when the patient was discharged. The University of Manitoba Education and Nursing Research Ethics Board approved the study.

Methodology

Recruitment

All nurses on the neurosurgical unit received an informal presentation. Those willing to participate were asked to submit their name using a box left on the unit. Approximately 30 nurses agreed to participate.

On admission of a patient to the unit, the nurses asked whether the patient or proxy was interested in learning more about the study from the research assistant. When the patient or proxy agreed, the research assistant explained the study, ensured that the patient met the inclusion criteria, and obtained consent. A nurse decided whether a patient was too drowsy or confused to decide on participation and would therefore require proxy consent. If a participant later became competent, consent was obtained directly from the participant.

Procedures

The nonspecific behavior categories and definitions were made available to the nurses in poster format. After a patient consented to participate in the study, a sheet of paper describing the nonspecific behavior categories and their definitions was placed on the patient’s bedside chart. The nurses were instructed to record in detail on the vital signs sheet any of the behaviors exhibited by study participants. Typically, nurses documented their observations on the chart progress notes; some nurses wrote more in-depth notes than others. Having the study form visible on the vital signs sheet was convenient for the nurses and reminded them of the behaviors to be observed and recorded in detail.

Doerksen et al. (2002) found that, following a ruptured aneurysm, some patients displayed restlessness, impulsiveness, or strange behaviors. However, no previous literature provided detailed descriptions of these behaviors. In this study common words whose dictionary definitions are congruent with observed nonspecific behaviors were used to categorize behaviors. The primary types of nonspecific behaviors were restless, impulsive, strange, and other (Table 1).

The nurses were instructed to conduct a routine neurological assessment if a patient demonstrated any of these nonspecific changes in behavior. The assessment primarily involved administration of the GCS, which is the current standard protocol used for neurological assessment. It should be noted that the nurses routinely assessed the patients for limb strength and signs of dysphasia. Patients who were within the usual time frame for vasospasm (i.e., 4-14 days after aneurysm rupture) and exhibited restlessness, impulsiveness, strange behavior, focal deficits, or a decreased level of consciousness underwent a CT angiogram (CTA) or digital subtraction angiogram (DSA), as well as other diagnostic tests, to investigate for vasospasm to rule out other potential problems. A neurosurgeon decided whether a CTA or a DSA was required. It was the standard of care to initiate this type of testing when patients experience signs or symptoms of vasospasm. In keeping with current clinical practice, patients who did not exhibit the studied nonspecific behaviors or any other symptoms of vasospasm routinely underwent a CTA 7-10 days after aneurysm rupture.

Sample Selection

Participants recruited for the study met the following criteria:

* diagnosed with subarachnoid hemorrhage due to ruptured aneurysm

* Hunt and Hess clinical grade 1-4 upon study entry (Table 2)

* before or in the classic time for vasospasm (4-14 days after aneurysm rupture)

* no diagnosis of vasospasm

* native speaker of English.

Selected patients were entered into the study as soon as possible after being admitted to the neuroscience unit. Participation in the study was to be terminated when the participant met one of the following conditions:

* exhibited symptoms and was diagnosed with vasospasm

* passed the 4-14 day time frame in which vasospasm typically occurs

* was discharged from the hospital or transferred from the neurosurgical unit

* exhibited a change in neurological condition not attributable to vasospasm, as diagnosed by a neurosurgeon.

The research assistant gathered demographic data from the chart and recorded the data using a coded number. These data included age, gender, admission date, number of days postbleed on admission, location of aneurysm, surgical procedure and date, medical history, surgical history, Hunt and Hess and Fisher scores on admission, test results (CT scans, CTAs, DSAs, daily serum sodium), and the presence or absence of vasospasm. For the purpose of the study, the presence or absence of vasospasm, rather than its severity, was considered. The research assistant also recorded a statement on the participant’s discharge status as an outcome measure.

Research Instruments

Observed nonspecific behaviors were entered on the vital signs record. Neurological status, measured by the GCS score, was noted on the neurological record.

Data Analysis

Raw data were entered into a Microsoft Excel spreadsheet and analyzed using Statistical Analysis System (SAS) Version 9.1. Statistical analyses included chisquare testing and Fisher’s exact test. The nurses’ observations of all nonspecific behaviors were analyzed using content analysis.

Results

An analysis of demographic and clinical characteristics associated with nonspecific behaviors is presented in the quantitative component of the study. The qualitative findings complement the quantitative data by providing detailed descriptions of the nonspecific behaviors.

Demographics

Sixty patients (39 females and 21 males) met the inclusion criteria and consented directly or by proxy t\o participate in the study. Their ages ranged from 27 to 82 years; the mean age was 53 years 12.2 SD.

Clinical Characteristics

Although grades 1-4 of the Hunt and Hess scale were represented, 45 (75%) of the patients were categorized as grade 1 or 2, indicating relatively good clinical neurological status on admission. The Fisher score (Fisher, Kisuer, & Davis, 1980) describes the amount and location of blood seen on the admission CT scan and is a predictive tool for vasospasm; scores range from 1 to 4 Table 3 shows the distribution of study participants’ Fisher scores. Although the result did not reach statistical significance, it is interesting to note that there was little variance in Fisher scores between those who experienced vasospasm and those who did not.

The 29 patients who were within the classic time frame for vasospasm and who exhibited restlessness, impulsiveness, or strange behavior underwent a CTA or a DSA to investigate for vasospasm. Six of the 29 patients required a DSA for diagnosis, and the other 23 had a CTA. The odds ratio of 16.477, with a confidence interval of 4.579-59.17, indicated that the incidence of nonspecific behaviors was significant (p

Of 29 patients exhibiting nonspecific behaviors, 22 demonstrated a single nonspecific behavior and 7 demonstrated a combination of nonspecific behaviors. When the three nonspecific behavior types (restlessness, impulsiveness, and strange behaviors) were examined individually, none was significant in discriminating between patients who developed vasospasm and those who did not. However, 9 of 11 patients exhibiting nonspecific behaviors in the other category were diagnosed with vasospasm (p

The time of vasospasm diagnosis is presented in Table 5. It is relevant to note that 74% of all symptoms occurred before 7 days postbleed. The majority of individuals who developed vasospasm had a normal GCS score (Table 6). Of 24 patients who exhibited nonspecific behaviors, 21 had normal serum sodium levels when vasospasm was diagnosed (Table 7). In addition, hydrocephalus was absent in 21 of 24 patients who exhibited nonspecific behaviors at the time of vasospasm diagnosis.

Forty-four of the 60 study participants were discharged home. Of the 16 patients who were not discharged home, 13 experienced nonspecific behaviors and were diagnosed with vasospasm (p

Qualitative Findings

All of the categories of nonspecific behaviors (restlessness, impulsiveness, strange behaviors, and other) occurred among patients in the study. Nurses documented the behaviors in detail as they occurred. Table 1 lists examples of the behaviors, as recorded by the nurses.

It was found that 5 of the 11 patients who experienced other behaviors described a type of hallucination.

Three of those patients were diagnosed with vasospasm, and two were not.

Discussion

The study was designed to explore the possibility that nonspecific behaviors observed in some patients admitted with SAH may be early indicators of cerebral vasospasm. Because vasospasm is the primary cause of disability and death in this population, early detection is critical.

In our study, 24 of 29 patients (83%) who exhibited the defined nonspecific behaviors were later diagnosed with vasospasm (p

Eighteen of the 24 patients diagnosed with vasospasm who experienced nonspecific behaviors had a GCS score of 15 (Table 6). Drowsiness is noted in the literature as a symptom of vasospasm (Cook, 2004; Oropello et al., 1996); if it occurs, it should be reflected in a changed GCS score. The finding that 18 patients demonstrated nonspecific behaviors without a change in the GCS further supports our hypothesis that these symptoms may occur earlier than drowsiness and that the GCS is not a reliable tool for detecting vasospasm.

Seventeen of the 24 patients who demonstrated nonspecific behaviors were diagnosed with vasospasm before the seventh day after hemorrhage. This result is of interest because the literature indicates that the peak time for vasospasm to occur is 7-10 days postbleed (Powsner et al, 1998; Qureshi, Sung, et al., 2000; Ullman & Bederson, 1996).

The literature also suggests that hyponatremia is an early sign of vasospasm (Macdonald & Weir, 2001). The finding that 21 of 23 patients who developed nonspecific symptoms before the seventh day postbleed and were later diagnosed with vasospasm had normal serum sodium levels supports our hypothesis that nonspecific behaviors may precede other recognized signs and symptoms associated with vasospasm.

Three of the 21 patients who experienced nonspecific symptoms and were diagnosed with vasospasm were noted to have hydrocephalus. These three patients were being treated for hydrocephalus via intermittent drainage of cerebrospinal fluid. It is possible that hydrocephalus may have affected their behavior (Qureshi, Suarez, et alv 2000).

As described above, 7 of the 31 patients diagnosed with vasospasm did not experience nonspecific symptoms before the occurrence of expected symptoms signaling neurological deterioration (Table 4). The initial clinical symptoms of these 7 patients were detected through routine nursing assessments. The specific symptoms included expressive dysphasia (2 patients); weakness of arm, leg, or both (3 patients); generalized drowsiness (1 patient); and confusion (1 patient). These symptoms are commonly noted in the literature and are recognized by healthcare professionals as indicating vasospasm.

Five patients experienced nonspecific behaviors in the absence of vasospasm. The explanations included atrial fibrillation and subsequent subacute infarctions in the frontal and occipital lobe, air over the frontal lobes demonstrated on CT, temporary hydrocephalus, and a serum sodium level of 128. No obvious physiological explanation was recorded for the fifth participant, except that vasospasm was ruled out. It is possible that the participant was experiencing delirium, intensive care unit psychosis, or sleep deprivation.

Outcomes Related to Nonspecific Behaviors

Significantly, 13 of the 16 patients with vasospasm who could not be discharged home had experienced early nonspecific symptoms. Given the nature of these nonspecific symptoms and the vascular territory of the vasospasm, cognitive impairment might account for difficulty in functioning independently.

Qualitative Findings

The quantitative analysis coding was based on the descriptions of the predefined nonspecific behaviors. The nurses’ documentation of behaviors in the restless and impulsive categories matched the definitions provided at the study’s outset. Behavior categorized as strange also corresponded well with the definition; however, the definition of confusion needs to be considered in relation to these behaviors. In particular, the patient who described concern about her husband in the garage could reasonably be considered confused, although she was oriented according to the GCS. On the GCS, confusion is assessed based on a patient’s orientation to name, place, and time; this formulation may be too limited to assess confusion.

The other category applied to 5 of 11 patients who described visual hallucinations. A possible explanation for one patient was that lorazepam had been administered to treat anxiety. Visual hallucinations could be explained by the presence of delirium. Delirium involves a variety of symptoms, including hallucinations. It can be caused by any of several factors (e.g., fever, intoxication; Hickey, 2002). Given the significance in the quantitative results of the presence of vasospasm and incidence of the other symptoms (p

Limitations

The relatively small study cohort resulted in small individual- category numbers. Therefore, despite the association between nonspecific behaviors and the presence of vasospasm, small numbers in the individual categories of restless, impulsive, and strange failed to predict vasospasm. Behaviors in the other category demonstrated statistical significance, but the information provided in this category and possible confounding factors require further investigation. Information regarding drug or alcohol abuse was not directly available for evaluation. The presence of pain, headache in particular, was not considered a factor that would influence nonspecific behaviors. The standard assessment of level of orientation and confusion using the GCS may be too limited and may influence certain qualitative findings related to the GCS score. A further limitation might be that the research was conducted at a single site. However, the site is a tertiary care hospital and the only center in the province where patients can receive this type of care.

Nursing Implications

Nurses need to be educated about the importance of observing nonspecific behaviors following subarachnoid hemorrhage. They should watch for these behaviors when conducting routine neurological assessments. In this study, the nonspecific behaviors were observed in some patients when the GCS was still within normal limits and before the peak time for vasospasm diagnosis. Nurses’ recognition of these behaviors can lead to earlier initiation of effective treatment.

Future Research

The findings of this study can lead to an important line of inquiry in the early detection of cerebral vasospasm. Future studies should be undertaken in multiple sites, and larger numbers of patients should be studied, particularly in relation to individual n\onspecific behaviors. Future research must take into account unrelated conditions that could be associated with nonspecific behaviors, such as hydrocephalus, medical therapies, uncontrolled pain, or drug or alcohol abuse. In administering the GCS, confusion should be well-articulated to clarify its role in the context of nonspecific behaviors.

If recognition of nonspecific behaviors as an early warning of vasospasm is incorporated into nursing practice on neurological units, patient outcomes should be improved. Therefore, future outcomes research concerning patients diagnosed with ruptured cerebral aneurysm who have experienced nonspecific behaviors should include measures of quality of life and physical functioning.

Summary

This objective of this study was to determine the significance of nonspecific behaviors associated with the onset of cerebral vasospasm. The quantitative results demonstrated a significant correlation between the nonspecific behaviors, collectively, and a diagnosis of vasospasm. The onset of these symptoms was seen to occur before the peak time frame for vasospasm and before occurrence of the signs and symptoms commonly recognized as indicating vasospasm.

The qualitative analysis provided detailed information supporting the definitions of restkss and impulsive behavior. The prevalence of patients with hallucinations indicates a need for further investigation into implications for nursing management.

Acknowledgments

This study was funded by the Health Science Centre Foundation’s Dolly and Michael Gembey Nursing Research Award. We would like to acknowledge our appreciation of the patients and family members who agreed to participate in our study. We would also like to thank the nurses on the neurosurgical unit for volunteering their time and effort in support of this study. Ann Huggard, RN, and Janice Nesbitt, RN, are particularly thanked for their assistance in obtaining consents and coUecting data. Dr. Michael West, section head, neurosurgery, is thanked for his ongoing support of clinical research in neurosurgery.

References

Cook, N. F. (2004). Subarachnoid haemorrhage and vasospasm: Using physiological theory to generate nursing interventions. Intensive and Critical Care Nursing, 20, 163-173.

Doerksen, K., Naimark, B. J., & Tate, R. B. (2002). A comparison of a standard neurological tool to a stroke scale for detecting symptomatic vasospasm. Journal of Neuroscience Nursing, 34, 320- 325.

Fisher, C. M., Kistler, J. P., & Davis, J. M. (1980). Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic screening. Neurosurgery, (SCl), 1-9.

Haley, E. C., Jr., Kassell, N. F., & Torner, J. C. (1993). A randomized controlled trial of high-dose intravenous nicardipine in aneurysmal subarachnoid hemorrhage:. A report of the Cooperative Aneurysm Study. Journal of Neurosurgery, 78, 537-547.

Hickey, J. V. (2002). The clinical practice of neurological and neurasurgical nursing (5th ed.). Philadelphia: Lippincott Williams & Wilkins.

Hunt, W. E., & Hess, R. M. (1968). Surgical risk related to time of intervention in the repair of intracranial aneurysms. Journal of Neurosurgery, 2S(I), 14-20.

Macdonald, R. L, & Weir, B. (2001). Medical aspects of vasospasm. In Cerebral Vasospasm (pp. 353-416). San Diego: Academic Press.

Oropello, J. M., Weiner, L, & Benjamin, E. (1996). Hypertensive, hypervolemic, hemodilutional therapy for aneurysmal subarachnoid hemorrhage: Is it efficacious? No. Critical Care Clinics, 12, 709- 730.

Powsner, R. A., OTuama, L. A., Jabre, A., & Melhem, E. R. (1998). SPECT imaging in cerebral vasospasm following subarachnoid hemorrhage. Journal of Nuclear Medicine, 39, 765-769.

Qureshi, A. I., Suarez, J. I., Bhardwaj, A., Yahia, A. M., Tamargo, R. J., & Ulatowski, J. A. (2000). Early predictors of outcome in patients receiving hypervolemic and hypertensive therapy for symptomatic vasospasm after subarachnoid hemorrhage. Critical Care Medicine, 28, 824-829.

Qureshi, A. I., Sung, G. Y., Razumovsky, A. Y1 Lane, K., Straw, R. N., & Ulatowski, J. A. (2000). Early identification of patients at risk for symptomatic vasospasm after aneurysmal subarachnoid hemorrhage. Critical Care Medicine, 28, 984-990.

Ullman, J. S., & Bederson, J. B. (1996). Hypertensive, hypervolemic, hemodilutional therapy for aneurysmal subarachnoid hemorrhage. Is it efficacious? Yes. Critical Care Clinics, 12(3), 697-707.

Unterberg, A. W., Sakowitz, O. W., Sarrafzadeh, A. S., Benndorf, G., & Lanksch, W. R. (2001). Role of bedside microdialysis in the diagnosis of cerebral vasospasm following aneurysmal subarachnoid hemorrhage. Journal of Neurosurgery, 94, 740-749.

Questions or comments about this article may be directed to Kathy Doerksen, MN RN CNN(c), at [email protected]. She is a clinical nurse specialist in neurosurgery at the Health Sciences Centre, Winnipeg, MB, Canada.

Barbara Naimark, PhD RN, is an associate professor, Senior Scholar Faculty of Nursing, at the University of Manitoba, Winnipeg, MB, Canada.

0047-2606/06/3806/0409$5.00

Comments

comments