Skip Navigation LinksHome > August 2014 - Volume 46 - Issue 4 > Peripheral Immune Response and Infection in First-Time and R...
Journal of Neuroscience Nursing:
doi: 10.1097/JNN.0000000000000061
Article

Peripheral Immune Response and Infection in First-Time and Recurrent Ischemic Stroke or Transient Ischemic Attack

Ross, Amy Miner; Lee, Christopher S.; Brewer, Margaret

Free Access
Article Outline
Collapse Box

Author Information

Questions or comments about this article may be directed to Amy Miner Ross, PhD RN CNS, at rossam@ohsu.edu. She is an Assistant Professor at the School of Nursing, Oregon Health & Science University (OHSU), Portland, OR.

Christopher S. Lee, PhD RN, is an Associate Professor at the School of Nursing, OHSU, Portland, OR.

Margaret Brewer, RN MS, is an Instructor at Rogue Community College, Department of Nursing, School of Nursing, OHSU, Portland, OR, and Rogue Regional Medical Center, Medford, OR.

This study was supported by the Sigma Theta Tau, Beta Psi Chapter Grant.

The authors declare no conflicts of interest.

Collapse Box

Abstract

ABSTRACT: Goals: The aims of this study were to determine if the infection rate differs between the first and recurrent ischemic stroke/transient ischemic attack (TIA), if the pattern of the peripheral immune response (PIR) differs between the first and recurrent ischemic stroke/TIA and if infection further influenced the pattern of the PIR. Methods: Retrospective review of 500 stroke cases with strict exclusion criteria (e.g., hemorrhagic stroke, subarachnoid hemorrhage, or spontaneous intracerebral hemorrhage; history of cancer; on steroids or immune suppressive drugs; recent invasive procedure) resulted in inclusion of 198 cases. Independent variables were first stroke or recurrent stroke and not infected or infected cases. Main-effect dependent variables were the white blood cell (WBC) and differential leukocyte counts (percentages of 100 cell counts for neutrophils, lymphocytes, and monocytes and absolute counts of neutrophils, lymphocytes, and monocytes). Findings: Infection rate was not different between the first versus recurrent stroke (p = .279). The pattern of WBC and differential counts were not different between groups, but addition of the covariate of infection showed group differences (p = .05). A four-group comparison of the dependent variables with the laboratory normal ranges showed lymphocyte percentages below the lower range limit in all four groups. Generalized linear modeling showed a modest rise (15%) in WBC counts in both groups with concomitant infection, a modestly low (−18%) lymphocyte percentage in recurrent stroke with infection, and a more substantial rise (22%–26%) in absolute neutrophil count in both groups with concomitant infection. Conclusions: Infection influences the pattern of the PIR in the first and recurrent stroke/TIA, and this difference can be quantified.

Stroke is the fourth leading cause of death in the United States; each year, stroke affects 795,000 people (76.7% with first-time and 23.3% with recurrent stroke) with a large 87% of strokes because of ischemia. More than 6.5 million survivors of stroke live with disability and dependency (American Heart Association, 2012). The risk of stroke increases 2–3 times for patients with ischemic stroke when preceded by either bacterial or viral infection (Emsley & Hopkins, 2008). The precedent incidence of infection the week before stroke is 10%–35%, and another third of strokes are complicated by infection (i.e., upper respiratory infection = ~35%; urinary tract infection = 27%) in the first 7 days after stroke leading to worsening of neurological out comes (Emsley & Hopkins, 2008; Smith & Tyrrell, 2010). For example, the 30-day mortality rate for stroke is approximately 27% in patients with versus 4% in those without concomitant pneumonia (Emsley & Hopkins, 2008). Oregon has a higher age-adjusted mortality rate because of stroke than the U.S. average, 43.9% and 38.9%, respectively (Kochanek, Xu, Murphy, Minino, & Kung, 2011).

Immune and inflammatory responses, known as defense mechanisms against infection, are concurrently present in ischemic stroke/transient ischemic attack (TIA). These mechanisms are thought to reflect brain-induced immunodepression through cortisol secretion and sympathetic autonomic nervous system suppression of lymphoid tissue almost immediately after stroke symptoms appear (Doyle, Simon, & Stenzel-Poore, 2008; Emsley & Hopkins, 2008, 2010). The peripheral immune response (PIR) has been reported early in the first 24 hours in ischemic stroke and TIA (Emsley et al., 2003; Vila, Castillo, Davalos, & Chamorro, 2000). In a prior study by Ross et al. (2007), the PIR was defined as neutrophilia, lymphocytopenia, and elevated monocytes. In general, the PIR is thought to upregulate the body’s capacity to mount an inflammatory response of longer duration with all pertinent effector cells mobilized (i.e., neutrophils and monocytes) to ameliorate infection (Janeway, Travers, Walport, & Shlomchik, 2001) and, in the case of stroke, mediate ischemic damage (Baumann & Gauldie, 1994; Katzan, Cebul, Husak, Dawson, & Baker, 2003; McColl, Allan, & Rothwell, 2009). Furthermore, elevated neutrophils but decreased monocytes have been associated with risk factors for stroke (i.e., hypertension, diabetes, obesity, and smoking) in studies during the time preceding stroke (Bergmann, Sidkmeier, Mix, & Jaross, 1998; Berliner et al., 2000; Herishanu, Rogowski, Polliack, & Marilus, 2006; Sela et al., 2004; Vayssier-Taussat et al., 2001; Yasunari, Maeda, Nakamura, & Yoshikawa, 2002). Testing for the PIR is inexpensive and common to the clinical care of patients (Nayak et al., 2011).

The aims of this study were to determine (a) if infection was higher in recurrent ischemic stroke/TIA more so when compared with first-time ischemic stroke/TIA, (b) if there was a difference in pattern of the PIR between first-time and recurrent ischemic stroke/TIA, and (c) if infections further influenced this difference in pattern of the PIR.

Back to Top | Article Outline

Methods

Design

A cross-sectional two-group comparative study design was employed to address the study aims. An electronic record query method was employed to locate the ischemic stroke population from the general population in a regional medical center treating 250–300 cases of stroke per year. International Classification of Disease, Ninth Revision (ICD-9) codes were used, followed by manual chart review for the selection of sample case events of stroke/TIA. Hospitalized incident stroke/TIA cases from the same center are thought to belong to the same population where the factors affecting referral, transport, admission, emergency department care, and inpatient care are similar between and within the groups, bringing homogeneity to the demographic data (Coull, Silver, Giles, Rothwell, & Study on behalf of the Oxford Vascular [OXVASC] Study, 2004).

In our a priori design, we targeted two study groups, first-time and recurrent stroke/TIA. Adult patients over 21 years old with ischemic stroke/TIA were included. Patients were excluded if they (a) were classified as having hemorrhagic stroke, subarachnoid hemorrhage, or spontaneous intracerebral hemorrhage; (b) were diagnosed or have a history of cancer; (c) were on immune suppressive drugs; and (d) had a recent surgery or procedure. Sampling continued until approximately 200 cases were identified for inclusion. This sample size was determined by power analysis and sample size estimates from the Ross et al. (2007) study based on the difference between groups on all PIR dependent variables. An a priori plan was put in place to assure reliability of stroke/TIA event inclusion and exclusion, assignment to study groups, and verification of data and analysis.

Back to Top | Article Outline
Variables

The cases were manually reviewed for the following data: demographics, medical history, signs and symptoms of infection, neurologic deficits, diagnostic results (e.g., computed tomography scan, magnetic resonance imaging, chest x-ray, urinalysis), and clinical evidence of the peripheral inflammatory response. The presence of brain infarction was determined by computed tomography scan or magnetic resonance imaging diagnosed by a board-certified radiologist or by a neurologist. Signs and symptoms of infection included reported infectious illness before admission, fever; infiltrates, atelectasis, or pneumonia on chest x-ray; white blood cells (WBCs), leukocyte ester, bacteria, or pus on the urinalysis; and positive body fluid cultures. Clinical evidence of the PIR was measured by Coulter counter analysis of WBC count and differential counts (i.e., percentages of 100 cell counts for neutrophils, lymphocytes, and monocytes and absolute counts of neutrophils, lymphocytes, and monocytes). These seven variables are the main-effect dependent variables. The clinical laboratory Coulter GEN•S System complied with all standards of the H20-A Reference Leukocytes Differential Count (Proportional) and Evaluation of Instrumental Methods. The laboratory reference ranges are evaluated annually on a minimum of 120 healthy outpatient volunteers (Coulter Corporation, 1988; National Committee of Clinical Laboratory Standards, 1997).

Institutional review board approval for human subjects’ research was obtained before starting this study. Responsible Conduct of Research (National Institutes of Health; Oregon Health & Science University) and the Health Insurance Portability and Accountability Act of 1996 Privacy Rule were observed. Study events were de-identified and coded before data entry and analysis.

Back to Top | Article Outline
Analysis

All data were entered into and analyzed with SPSS version 13.0 or with Stata version 11.0 (College Station, TX). A significance level of .05 was set for all analyses. The demographic variables of the study groups were compared to determine if there were differences that might influence the dependent variables. Analysis of variance was conducted on age, and χ2 tests were conducted on gender, ethnicity, medical history, brain imaging findings, and neurological examination. After this preliminary analysis, if covariates emerged, these were investigated using multivariate analysis of covariance (MANCOVA) or generalized linear modeling.

To test specific aim 1 hypothesis that the infection rate is higher in recurrent ischemic stroke/TIA events than first-time stroke/TIA, χ2 analysis was conducted. A significant χ2 value would be sufficient evidence to conclude that the infection rate is higher in recurrent stroke/TIA events than in first-time stroke/TIA events. Should infection rate be significantly different between these two groups in specific aim 1, then adding in the presence of infection as a covariate to the analysis of specific aim 2 was warranted. Infection rates were calculated using the ratio of stroke/TIA cases that meet diagnostic criteria for bacterial or viral infection compared with the total number of stroke/TIA cases. The pattern of the PIR in stroke/TIA cases without infection is defined as an elevation in both neutrophil and monocyte counts, a decrease in lymphocyte count, and a normal total while blood cell count (Ross et al., 2007).

To test specific aim 2 hypothesis that there is a difference in the pattern of the PIR as a panel of variables between first-time and recurrent stroke/TIA, a MANCOVA was used. A significant F value would support this hypothesis.

To test specific aim 2a hypothesis that the difference in PIR pattern as a panel of variables between first-time and recurrent stroke/TIA was further influenced by the presence of infection, a MANCOVA was conducted. A significant F value would support this hypothesis.

Although the MANCOVA tests for the difference between groups for the panel of all seven variables of the PIR, using independent analysis of variance on each of the seven dependent variables creates a type I error should the significance not be adjusted (e.g., <.007). Therefore, generalized linear modeling was conducted to investigate all seven variables independently (Gaussian identity for all models and Gaussian-log models with ps < .2). This was done to determine the influence of group membership on each variable with first-time stroke not infected as the referent group.

Back to Top | Article Outline

Results

Sample

A total of 500 stroke/TIA admissions were found using the electronic record query method and were investigated for inclusion in this study. The following cases were excluded: (a) hemorrhagic stroke (n = 65), (b) admission for other medical reasons with stroke-like symptoms (e.g., extremity weakness, confusion; n = 72), (c) cancer diagnosis (n = 59), (d) recent surgery or invasive procedure (n = 52), (e) on steroids or immune suppressive drugs (n = 23), and (f) eligible for the study in all ways and no laboratory data available in the medical record (n = 31). This resulted in 198 cases that were included.

Back to Top | Article Outline
Demographics

After this preliminary analysis, covariates emerged, age and number of women in groups (see Table 1). These were investigated in the MANCOVA model and in the regression analysis. Interestingly enough, statistical significance was found between groups for cognitive and sensory deficits, with there being an inverse relationship between these two variables (see Table 2). These were not entered into the analyses but provide a perspective on the complexity of emerging infectious processes in patients with stroke.

Table 1
Table 1
Image Tools
Table 2
Table 2
Image Tools

A comparison is shown to illustrate those variables that are outside the range of the laboratory normal for the laboratory at the research setting. The WBCs were within normal limits for those groups with infection, the neutrophil percentages were above the upper limit for those groups with infection, and the lymphocyte percentages were below the lower limit of normal for all groups. The absolute counts of neutrophils were above the upper limit for those groups with infection, and this is the only variable that reaches significance when adjusted to avoid type I error (i.e., p < .007; see Table 3).

Table 3
Table 3
Image Tools
Back to Top | Article Outline
Main Effects Analysis
Aim 1

The infection rate was not statistically significantly higher in recurrent stroke (p = .279). However, the number/percentage of cases that were infected in recurrent stroke (n = 56, 28.3%) versus first-time stroke (n = 38, 19.2%) was higher, and this may be clinically significant, specifically because infection was associated with cognitive deficits (Pearson r = .201; p = .005).

Back to Top | Article Outline
Aims 2 and 2a

The between-study-group (i.e., first-time vs. recurrent stroke) differences in the seven dependent variables were not significant (F value = .495, p = .838). However, with the covariate of infection, the between-group differences were significant (F value = 2.062, p = .05, observed power = .784). This evidence was enough to conclude that infection significantly influences the PIR when compared between groups (i.e., first-time vs. recurrent stroke) more so than the influence of group membership alone.

Back to Top | Article Outline
Influence of Group Membership on Individual Variables

The WBCs were significantly different in the recurrent infected group in comparison with the referent group (p = .049). In the log linear model, this is a difference of 15% in the WBCs between these group comparisons. Lymphocyte percentage approached significance with the same between-group comparisons in the log linear model (p = .070). The magnitude of this difference showed that the lymphocyte percentage was 18% lower in the recurrent group with infection than in first-time stroke without infection. Finally, the absolute neutrophils were significantly different in the recurrent infected group in comparison with the referent group (p = .012). In the log linear model, this is a difference of 26% in the absolute neutrophils between these group comparisons (p = .013). Although approaching statistical significance (p = .055), absolute neutrophils were 22% higher in the first-time stroke group with infection than in the first-time stroke group without infection (see Table 4).

Table 4
Table 4
Image Tools
Back to Top | Article Outline

Discussion

We found that evidence of infection in this study ranged from 43% to 52% and was higher than 10%–35% incidence of infection within 1 week before stroke and the strokes complicated by urinary tract infection and upper respiratory infection as reported in the literature (Emsley & Hopkins, 2008; Smith & Tyrrell, 2010). This may be because of the determination of infection between studies, verbal report versus clinical evidence of infection on admission. We also found that the rate of infection was higher in recurrent stroke, and this may be influenced by the association between cognitive deficits and presence of infection. For example, those with cognitive deficits may not be able sense and communicate that they are ill with an infectious process, and the infection may go undetected until it is late in the course of the infection.

The PIR was found to be the same as reported earlier by Ross et al. Meaning that the WBC counts were within normal limits, the neutrophil percentages and absolute counts of neutrophils were elevated above normal, and the lymphocyte percentages were the only variable in all groups that were below the lower limit of normal. Previously, in the Ross et al. study, a MANCOVA showed that the PIR was not different between patients with TIA and with stroke when draconian exclusions were applied (Ross et al., 2007). Subsequent reports have suggested that the PIR in stroke is an epiphenomenon (Becker, 2009). The PIR, when tested as a panel of variables, was found to be the same between first-time and recurrent stroke, suggesting that this finding is stable in both of these contexts, a possible epiphenomenon. Furthermore, the stability may provide a clinical parameter to monitor and use for the timing of interventions. However, there was an effect of infection on the panel of variables of the PIR between these two groups, and the exploration of this finding has some clinical importance.

The more sensitive regression and log linear model using the standard and also significant covariates of age and gender showed differences between first-time stroke uncomplicated by infection and the other groups: first-time stroke with infection, recurrent stroke uncomplicated by infection, and recurrent stroke with infection. This analysis showed off the differences that could be detected clinically using this inexpensive test. First, WBCs were highest in recurrent stroke with infection and second highest in first-time stroke with infection. Second, the lymphocyte percentages were lowest in recurrent stroke with infection and second lowest in first-time stroke with infection. Interestingly, the magnitude of the WBC increases and the lymphocyte decreases are very similar and support the theory that this lymphocyte decrease may influence the WBC count enough to result in the WBC count remaining within normal limits. The magnitude of the lymphocyte decrease to well below the lower limit of the normal range may be because of the effects of epinephrine on T-cell-specific lymphoid tissue (Doyle et al., 2008; Emsley & Hopkins, 2008, 2010). Finally, the absolute neutrophils were highest in recurrent stroke with infection and second highest in first-time stroke with infection. Because the absolute neutrophil counts are the product of the WBC count and the neutrophil percentage, this finding may explain the strength of the magnitude differences in the WBC count being in a positive direction reflecting the effect of the neutrophil percentages rather than conforming to the direction suggested by the decrease in lymphocyte percentages. Taken together, these findings support the theory that immunodepression immediately after stroke occurs probably through suppression of lymphoid tissue by epinephrine through sympathetic autonomic nervous system signaling and that there is a smaller response of neutrophil elevation through activation of the innate immune system when faced with infection (Janeway et al., 2001).

In a comparison with laboratory normal range limits, the WBC count was normal even in the face of infection in first-time and recurrent stroke in this study. Lymphocytes were found to be decreased less than the lowest limit of normal. An explanation for this finding may be the sequestration of lymphocytes in the margin, producing pavementation on vascular walls. Or, the lymphocytopenia reported here may reflect the mechanisms associated with brain-induced immunodepression, cortisol secretion, and sympathetic autonomic nervous system suppression of lymphoid tissue, which occur almost immediately after stroke symptoms appear and therefore would be detected on admission (Doyle et al., 2008; Emsley & Hopkins, 2008, 2010). The finding that the WBCs are not elevated in the face of infection may reflect the influence of the percentage of lymphocytes that are well below the lower limit of normal in all groups, whether infected or not.

The neutrophil percentage and also the absolute counts of neutrophils were elevated in the presence of infection possibly reflecting the need to have these innate immune cells mobilized to fight infection (Janeway et al., 2001). Or, as an alternate explanation, these elevations may be a residual finding in patients with elevated risk factors for stroke just before admission (Bergmann et al., 1998; Berliner et al., 2000; Herishanu et al., 2006; Ross et al., 2007; Sela et al., 2004; Vayssier-Taussat et al., 2001; Yasunari et al., 2002).

The limitations of this study are the retrospective design; the powered but relatively small sample size; and the basis of the PIR, which is the available data in the medical record. Previously, a retrospective design was used to describe the PIR in stroke and TIA using strict exclusion criteria. This resulted in knowing more about the PIR uncomplicated by confounders, but this did not fully describe the PIR in all stroke occurrences and did not reflect the changes in the PIR with infection, a frequent finding on stroke admission. It may be that the PIR is present because of other influences, like a stress response, based on either the secretion of cortisol and the effect on neutrophils and the secretion of epinephrine and the effect on lymphocytes or the presence of the PIR may be solely related to the size of the ischemic infarction and has no other clinical meaning. This response has been investigated by several research groups, and the worth has not been laid to rest; these possible explanations have not been confirmed.

To describe the worth of the inexpensive test of the WBC and differential count in stroke/TIA, a retrospective design was considered to be appropriate, even in the face of the lack of generalizability. This study was a preliminary step in investigating the PIR as a valuable test in determining group differences when infection is and is not present. The sample size was appropriately powered to detect significant differences in the PIR without “overpowering” the study and making clinically insignificant changes appear significant. The frequency that laboratory work was not in the medical record is of concern when considering the possible utility of the WBC and differential count when developing monitoring schemes of key indicators related to stroke outcome.

A strength of the design was in the combination of stroke and TIA cases based on the finding of the first study by Ross et al. and the planned stepwise inclusion of infected stroke/TIA cases using the four-group design. The findings that are presented here are not complicated by recent invasive intervention and presence or recent history of cancer and include all other comorbid conditions related to traditional risk factors for ischemic stroke. This study is a first description of the PIR in the general ischemic stroke population because they would present acutely for clinical care without specific subclassification of cases by presence of risk factors (e.g., hypertension, diabetes, smoking, hyperlipidemia), either singularly or in combination. This, to some extent, allows for generalization to the population with traditional risk factors for ischemic stroke.

The implications of these findings include both research and practice aspects. First, the practice implications are for stroke care providers, nurses, and physicians to monitor for infection in both first-time and recurrent stroke and to educate family and other caregivers about the importance of early recognition of infection because this has been associated with an increase in the risk for recurrent stroke and may contribute to a worsening of stroke outcome (Haeusler et al., 2008). Because all seven dependent variables are interrelated to the point of strong correlation, then examining why the WBCs were normal in all groups is of interest. Practitioners should be advised to not use the WBC count as the only guide to determine that infection is present, because the WBC count is within the normal range in stroke/TIA with and without concomitant infection in this study and may signal a stable epiphenomenon. Rather, because the numbers of stroke/TIA cases that might have infection are large, this should then guide the investigation of possible infectious processes in all cases of stroke on acute clinical admission presentation. And second, the research implications are to (a) elucidate the influence of traditional risk factors for ischemic stroke, either singularly or in combination, on the PIR; (b) determine if the PIR trajectory differs over time for these groups and if these trajectories are associated with worsening stroke outcomes; and (c) determine if the PIR trajectories can be used to identify treatment windows. Furthermore, biological investigation into the triggers to lymphocytopenia in stroke is important because suppression of lymphocytes in animal models has produced neuroprotection (Hurn et al., 2007). Studies that focus inclusively on the whole stroke population are warranted to test the clinical utility of the PIR in timing treatments. Both the practice and research implications are worthy of study, because the PIR can be monitored and interpreted by both nurses and physicians with a relatively inexpensive test, the WBC count and differential.

To conclude, infections appear to complicate recurrent stroke more frequently than first-time stroke and that the PIR differs with the presence of infection.

Back to Top | Article Outline

References

American Heart Association and American Stroke Association. (2012). Heart disease and stroke statistics—2012 update. Retrieved on from http://circ.ahajournals.org/content/125/1/e2

Baumann H., Gauldie J. (1994). The acute phase response. Immunology Today, 15 (2), 74–80.

Becker K. J. (2009). Sensitization and tolerization to brain antigens in stroke. Neuroscience, 158, 1090–1097.

Bergmann S., Sidkmeier R., Mix C., & Jaross W. (1998). Even moderate cigarette smoking influences the pattern of circulating monocytes and the concentration of sICAM-1. Respiration Physiology, 114, 69–275.

Berliner S., Rogowski O., Rotstein R., Fusman R., Shapira I., Bornstein N. M., Zeltser D. (2000). Activated polymorphonuclear leukocytes and monocytes in the peripheral blood of patients with ischemic heart and brain conditions correspond to the presence of multiple risk factors for atherothrombosis. Cardiology, 94, 19–25.

Coull A. J., Silver L. E., Giles M. F., Rothwell P. M., & Study, on behalf of The Oxford Vascular (OXVASC) Study. (2004). Direct assessment of completeness of ascertainment in a stroke incidence study. Stroke, 35, 2041–2047.

Coulter Corporation. (1988). GENñS system reference manual. Miami, FL: Author.

Doyle K. P., Simon R. P., & Stenzel-Poore M. P. (2008). Mechanisms of ischemic brain damage. Neuropharmacology, 55, 310–318.

Emsley H. C., & Hopkins S. J. (2008). Acute ischaemic stroke and infection: Recent and emerging concepts. Lancet, 7, 341–353.

Emsley H. C., & Hopkins S. J. (2010). Post-stroke immunodepression and infection: An emerging concept. Infectious Disorders: Drug Targets, 10, 91–97.

Emsley H. C., Smith C. J., Garvin C. M., Georgiou R. F., Vail A., Barberan E. M., Hopkins S. J. (2003). An early and sustained peripheral inflammatory response in acute ischemic stroke: Relationships with infection and atherosclerosis. Journal of Neuroimmunology, 139, 93–101.

Haeusler L. G., Schmidt W. U., Fohring F., Meisel C., Helms T., Jungehulsing G. J., Volk H. D. (2008). Cellular immunodepression preceding infectious complications after acute ischemic stroke in humans. Cerebrovascular Disease, 25 (1–2), 50–58.

Herishanu Y., Rogowski O., Polliack A., & Marilus R. (2006). Leukocytosis in obese individuals: Possible link in patients with unexplained persistent neutrophilia. European Journal Haematology, 76, 516–520.

Hurn P. D., Subramanian S., Parker S. M., Afentoulis M. E., Kaler L. J., Vandenbark A. A., & Offner H. (2007). T- and B-cell-deficient mice with experimental stroke have reduced lesion size and inflammation. Journal of Cerebral Blood Flow and Metabolism, 27, 1798–1805.

Janeway C. A., Travers P., Walport M., & Shlomchik M. (2001). Immunobiology: the Immune system in health and disease (5th ed.). New York, NY: Garland.

Katzan I. L., Cebul R. D., Husak S. H., Dawson N. V., & Baker D. W. (2003). The effect of pneumonia on mortality among patients hospitalized for acute stroke. Neurology, 60, 620–625.

Kochanek K. D., Xu J., Murphy S. L., Minino A. M., & Kung H-C. (2011). Division of Vital Statistics Deaths: Final data for 2009. National Vital Statistics Reports, 60 (3), 85.

McColl B. W., Allan S. M., & Rothwell N. J. (2009). Systemic infection, inflammation and acute ischemic stroke. Neuroscience, 158, 1049–1061.

National Committee for Clinical Laboratory Standards. (1997). Reference leukocyte (WBC) differential count (proportional) and evaluation of instrumental methods; Approved standard H20-A2 (2nd ed.). Wayne, PA: National Committee for Clinical Laboratory Standards.

Nayak A. R., Kashyap R. S., Kabra D., Deoras P., Purohit H. J., Taori G. M., & Daginawala H. F. (2011). Evaluation of routinely performed hematological and biochemical parameters for the prognosis of acute ischemic stroke patients. Neurological Science, 32, 855–860.

Ross A. M., Hurn P., Perrin N., Wood L., Carlini W., & Potempa K. (2007). Evidence of the peripheral inflammatory response in patients with transient ischemic attack. Journal of Stroke and Cerebrovascular Diseases, 16 (5), 203–207.

Sela S., Mazor R., Amsalam M., Yagil C., Yagil Y., & Kristal B. (2004). Primed polymorphonuclear leukocytes, oxidative stress, and inflammation antecede hypertension in the Sabra rat. Hypertension, 44, 764–769.

Smith C. J., & Tyrrell P. J. (2010). Current and emerging treatments for acute stroke: Relationship with infection. Infectious Disorders: Drug Targets, 10, 112–121.

Vayssier-Taussat M., Camilli T., Aron Y., Meplan C., Haunaut P., Polla B. S., & Weksler B. (2001). Effects of tobacco smoke and benzo[a]apyrene on human enothelial cell and monocyte stress responses. American Journal of Physiology: Heart and Circulation Physiology, 280, H1293–H1300.

Vila N., Castillo J., Davalos A., & Chamorro A. (2000). Proinflammatory cytokines and early neurological worsening in ischemic stroke. Stroke, 31 (10), 2323–2329.

Yasunari K., Maeda K., Nakamura M., & Yoshikawa J. (2002). Oxidative stress in leukocytes is a possible link between blood pressure, blood glucose, and C-reactive protein. Hypertension, 39, 777–780.

Keywords:

infection; ischemic stroke; peripheral immune response; recurrent ischemic stroke

© 2014 American Association of Neuroscience Nurses

Login

Article Level Metrics

Search for Similar Articles
You may search for similar articles that contain these same keywords or you may modify the keyword list to augment your search.