Published online before print June 26, 2008, doi: 10.1161/STROKEAHA.107.513499
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2008;39:2231.)
© 2008 American Heart Association, Inc.
Original Contributions |
Serum Calcium as Prognosticator in Ischemic Stroke
From the Stroke Center and Department of Neurology (B.O., S.S., J.L.S.), and the Department of Emergency Medicine (S.S.), UCLA Medical Center, Los Angeles, Calif; the Division of Neurology (P.T.), University of British Columbia, Canada; the UCSD Stroke Center (P.L.), San Diego, Calif; the Department of Neurology (M.K.), Helsinki University Central Hospital, University of Helsinki, Finland; the Department of Neurology (S.M.D.), Royal Melbourne Hospital, University of Melbourne, Australia; the Department of Neurology (W.H.), University of Heidelberg, Germany; and Bayer HealthCare AG (M.F.), Leverkusen, Germany.
Correspondence to Bruce Ovbiagele, MD, Stroke Center and Department of Neurology, University of California at Los Angeles, 710 Westwood Plaza, Los Angeles, CA 90095. E-mail Ovibes{at}mednet.ucla.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background and Purpose— Calcium (Ca2+) plays a role in the cellular and molecular pathways of ischemic neuronal death. We evaluated the impact of both early and delayed Ca2+ levels on clinical outcomes from acute ischemic stroke.
Methods— The relations between blood calcium level obtained early (<4.5 hours), and delayed (72 to 96 hours) after ischemic stroke onset versus clinical outcomes were analyzed in 826 subjects enrolled in an international trial in the Virtual International Stroke Trials Archive. Subjects were categorized into Ca2+ quartiles. Outcome measures analyzed included baseline and 72- to 96-hour stroke severity, as well as 3-month functional and global disability scales. The independent effect of calcium on outcome was evaluated by median and logistic regression analysis.
Results— Six hundred and fifty-nine (80%) of the trial subjects had complete baseline data including Ca2+ levels. Bivariately, the highest delayed Ca2+ quartile (versus lowest) was associated with lesser stroke severity and better 3-month functional and independence scale outcomes (all P<0.001), but no significant outcome differences were noted among early Ca2+ levels. In multivariable analysis, delayed Ca2+ in the highest quartile (versus lowest quartile) was associated with greater 3-month independence score on the Barthel Index scale (76.9 versus 55.4, P=0.006). No other significant outcome differences were noted between highest and lowest quartiles for both early and delayed Ca2+ quartiles.
Conclusions— Elevated 72- to 96-hour serum Ca2+ levels independently predict greater independence 3 months after ischemic stroke, but very early serum Ca2+ appear not to have any prognostic significance.
Key Words: acute care • acute stroke • cerebrovascular disease • outcomes • prognosis • stroke • stroke care
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Given the immense burden that ischemic stroke exerts, the need to develop more precise estimates of a stroke survivor’s prognosis remains an important goal. Ischemic neuronal death engages several terminal pathways including the loss of ionic homeostasis.1 Calcium (Ca2+) ions play a physiological role in the multiple pathomechanisms of cerebral ischemia.2 Cell calcium metabolism during and immediately after a transient period of ischemia influences the cascade of events that leads to subsequent neuronal injury.1 For instance, ischemia/hypoxia triggers rapid translocation of Ca2+ from extracellular to intracellular spaces of cerebral tissues.3,4 Awareness of the participation of Ca2+ in the ischemic cascade has led to the development of several potential neuroprotective agents designed to modify the role of this ion in acute focal brain injury.5
Ca2+ has also been studied with regard to its relationship with stroke risk factors and stroke incidence.6 High dietary intake of Ca2+ has been associated with reduced risk of stroke.6 To our knowledge very few attempts have been made to investigate the impact of serum Ca2+ level on clinical outcomes after ischemic stroke.7–9 A recent study suggested that calcium levels obtained within 24 hours of stroke onset are associated with better hospital discharge clinical outcomes.9 However, this finding requires confirmation and it remains unclear as to whether timing of calcium level is of prognostic significance. In this study we aimed to compare the impact of very early versus delayed calcium levels on clinical outcomes after acute ischemic stroke.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
We reviewed data from the modified Repinotan–Randomized Exposure Controlled Trial (mRECT) trial, for this analysis.10 Data from this trial was contributed to the Virtual International Stroke Trials Archive, a vast resource of stroke patient data collated for anonymized analysis and hypothesis testing.11 The methods and results of the mRECT trial have already been described.10 In brief, mRECT was a double-blind, placebo-controlled, parallel group, international multicenter study to evaluate the efficacy, safety, tolerability, and pharmacokinetic/pharmacodynamic effects of a targeted exposure of intravenous repinotan, a serotonin agonist (5HT1A receptor subtype). A total of 681 patients with an ischemic thromboembolic hemispheric stroke were randomized 1:1, to placebo or repinotan. Subjects had to have a National Institute of Health Stroke Scale (NIHSS) score of 8 to 23 with a motor deficit
2, to be eligible for participation in mRECT. Patients with CT evidence of major early infarction involving more than one-third of the middle cerebral artery territory, primary intracerebral hemorrhage, a baseline Rankin score 2, systolic blood pressure 210 or <110 mm Hg, diastolic blood pressure >110 or <60 mm Hg, myocardial infarction within 3 months of enrollment, or who had unstable supraventricular and/or ventricular arrhythmia, severe conduction defect (AV-block grades 2 and 3), bradycardia, or uncompensated heart failure, were excluded. For enrolled mRECT patients, study drug infusion had to be initiated within 4.5 hours from the onset of ischemic symptoms. Recombinant tissue plasminogen activator was permitted as standard medication. The primary intention-to-treat outcome was a Barthel Index score >85 at 3 months post-treatment. Repinotan failed to show clinical benefit, because there was no statistically significant difference between the treatment arms with regard to the primary outcome.10
For the purpose of this analysis, repinotan and placebo groups were combined given the lack of difference in efficacy between them. However, a separate analysis was also performed to assess any potential differences in Ca2+ serum levels between repinotan versus placebo-treated subjects at various study time points. Prespecified outcome measures to be evaluated in relation to very early calcium level (obtained within 4.5 hours after stroke onset) and delayed calcium level (obtained at 72 to 96 hours after stroke onset) included the following: (1) Stroke Severity on Admission (median NIHSS Score); (2) Stroke Severity at 72 to 96 hours (median NIHSS Score); (3) Neurological Improvement 
NIHSS (median baseline NIHSS Score to median NIHSS Score at 72 to 96 hours); (4) Global Disability at 72 to 96 hours (median Barthel Index Score); (5) Functional Activity at 72 to 96 hours (modified Rankin Scale Score 2); (6) Global Disability at 3 months (median Barthel Index Score); (7) Functional Activity at 3 months (modified Rankin Scale Score 2).
Statistical Analyses
Serum Ca2+ level was collapsed into quartiles (see cutpoints in tables below), for both early and 72- to 96-hour measures. These quartile versions were used as predictor variables in modeling. Tables and 
2 tests were used to investigate bivariate relationships between the 2 calcium quartile measures and demographic or outcome measures that were categorical or collapsed into categories. Next descriptive statistics and Kruskal-Wallis rank sum tests were used to investigate bivariate relationships between the calcium quartiles and continuous measures. In addition, Pearson correlations were computed among the several continuous measures, including uncollapsed calcium levels. Multivariable analyses were conducted (with Stata 9.1), using median regression for continuous measures, logistic regression for dichotomous measures, and ordinal logistic regression for measures collapsed into 3 categories. To evaluate the role of possible confounding factors, other potential determinants of incident stroke severity and outcome were also analyzed and so each model included one of the sets of calcium quartiles plus other covariates, including age,12 history of atrial fibrillation,13 history of stroke,14,15 admission glucose,1,16–18 admission temperature,18–20 and admission systolic blood pressure.21 Potential baseline covariate predictors of functional activity and global disability at 72 to 96 hours and at 3 months included all the aforementioned potential determinants of stroke severity as well as admission NIHSS score.22 Statistical tests were made to check whether the coefficients for the calcium quartiles were simultaneously significant.
To evaluate the role of albumin-adjusted Ca2+, serum Ca2+ was adjusted by serum albumin using the formula: Corrected Calcium=serum calcium+0.8 (4–serum albumin).23 These adjustments were made for both early and delayed Ca2+ levels, matching Ca2+ and albumin on time.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Of 826 ischemic stroke subjects enrolled in the trial, 659 (80%) had complete baseline data including very early calcium levels. Table 1 shows a breakdown of salient baseline demographic and clinical variables by very early Ca2+ level. There were no significant differences in demographic and clinical factors by Ca2+ quartile. A comparison of serum Ca2+ at baseline, 24 hours and 72 hours, as well as change from baseline at 24 and 72 hours, between placebo versus repinotan groups did not reveal any statistically significant differences.
|
Table 2 displays the results of the bivariate analyses evaluating early and delayed Ca2+ levels versus the prespecified clinical outcomes. There were no significant differences with regard to the clinical outcomes studied among quartiles of early Ca2+ (Table 2). Also no significant improvements in median 72- to 96-hour NIHSS scores from baseline (NIHSS) were noted among Ca2+ quartiles (Table 2). For delayed Ca2+, those in the highest quartile had significantly lesser 72- to 96-hour stroke severity, greater 72- to 96-hour independence, and better 72- to 96-hour functional activity than those in the lowest quartile (Table 2). Three-month independence and functional activity outcomes were also significantly better in the highest delayed Ca2+ quartile when compared to the lowest quartile.
|
Results of the correlation analyses showed early Ca2+ correlated weakly with baseline NIHSS score (r=–0.05, P=0.21); 72- to 96-hour NIHSS score (r=–0.10, P=0.01); NIHSS (72- to 96-hour NIHSS score–baseline NIHSS score, r=–0.09, P=0.04); and 72- to 96-hour Barthel Index score (r=0.11, P=0.008). Delayed Ca2+ showed stronger correlations with baseline NIHSS score (r=–0.2, P=0.0001); 72- to 96-hour NIHSS score (r=–0.3, P<0.0001); NIHSS (72- to 96-hour NIHSS score–baseline NIHSS score, r=–0.2, P<0.0001); and with Barthel Index score (r=0.3, P<0.0001). Early Ca2+ and delayed Ca2+ had a correlation of 0.4 (P<0.0001), whereas baseline NIHSS score and 72-hour NIHSS score had a correlation of 0.5 (P<0.0001).
Table 3 shows the multivariable analysis results for baseline and 72- to 96-hour clinical outcomes comparing trends across early and delayed Ca2+ quartiles. Those in the highest delayed calcium quartile had significantly better Barthel Index scores and greater improvements in NIHSS than those in the lowest quartile. Results of the multivariable analyses for the 3-month independence and functional activity outcomes are shown in Table 4. After adjusting for prespecified covariates those in the highest delayed Ca2+ quartile had significantly better median Barthel index scores than the lowest quartile (Table 4). For the early Ca2+ quartiles there was a significant difference across quartiles but not in a linear fashion with the greatest difference being between the first and second quartiles (Table 4). There were no significant differences in the likelihood of a poor functional outcome among early Ca2+ quartiles following multivariable analyses, and although the highest delayed Ca2+ quartile had less of a likelihood of a poor functional outcome compared to the lowest delayed Ca2+ quartile this difference did not reach statistical significance (Table 4). Using the formula noted above we examined the influence of early and delayed albumin-corrected Ca2+ levels on the prespecified clinical outcomes, and did not find any significant differences in outcomes among albumin corrected Ca2+ quartiles with regard to the 3-month end points (data not shown). Furthermore, there were no correlations between baseline albumin levels and the 72-hour and 3-month clinical outcomes (data not shown), but there was a weak correlation of baseline albumin with baseline NIHSS score (–0.0889, P<0.02).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Our analysis suggests that elevated serum calcium levels obtained at 72 to 96 hours after ischemic stroke onset are associated with higher 3-month functional independence scores. Furthermore, higher serum calcium levels at 72 to 96 hours after the index stroke showed a clear pattern of association with better functional activity at 3 months, but this relationship did not reach statistical significance. Associating elevated calcium levels with better clinical outcomes after ischemic stroke has been shown in 2 prior single-center studies,7,9 but our study provided the added advantages of investigating this issue within the rigorous procedures of a clinical trial, including subjects from multiple sites from around the world, evaluating the impact of very early calcium levels after ischemic stroke, and examining an outcome time point (3 months postevent) frequently studied in acute stroke clinical trials. The 2 prior studies which looked at the prognostic significance of calcium levels after ischemic stroke found better hospital discharge functional outcomes in those with elevated admission Ca2+ levels,9 and a significant reduction in Ca2+ level in patients who died during hospitalization compared with survivors.7 Our study, however, suggests that timing is particularly important because we found that serum calcium levels drawn early (within 4.5 hours of ischemic stroke ictus) were not of any prognostic significance.
Clarifying the exact pathophysiological mechanism that may underlie these clinical observations has been challenging, especially because it is unclear whether serum Ca2+ level exerts a primary effect on ischemic stroke, or if it reflects a secondary epiphenomenon of ischemic stroke severity.9 However, our study’s findings of prognostic significance for delayed rather than early Ca2+ suggests that the truth may be closer to the latter. Animal studies have shown that Ca2+ movement from serum to brain occurs primarily via the choroid plexuses,1 and when neurons (and/or glia) are exposed to lipid peroxidation, their intracellular structures lose their protection from the extracellular space and a Ca2+ sink is created. As a result more calcium is extracted from the blood into the brain. In order to pull Ca2+ from the serum, the gradient must be sufficient to reduce the content of Ca2+ in the serum.1 It is thought that total neuronal cell Ca2+ content may increase to 150% of control or more.1 Furthermore, the finding of more substantial decreases in calcium levels of ischemic stroke patients than of transient ischemic attack and controls may also support this hypothesis.7 However, whether the amount would be sufficient to change the serum levels to the degree noted in our study is unknown. A comparison of MR images quantifying the extent of brain injury or measurement of Ca2+ concentration in the cerebrospinal fluid8 versus serum Ca2+ levels would provide insight as to whether greater cerebral damage is associated with lower Ca2+ levels.
Interestingly, the potential role of serum calcium as a clinical prognosticator is not limited to ischemic stroke. Studies of general medical conditions, particularly among the critically ill, have shown that those with hypocalcemia tend to be more severely ill, and have higher mortality rates than those with normocalcemia.24–27 We did not have information on medical complications after the strokes in the mRECT trial, and so it is conceivable that the 72- to 96-hour Ca2+ levels we observed may have been influenced by the subsequent development of intercurrent illnesses, especially given the marked difference in 72- to 96-hour NIHSS scores among the 72- to 96-hour serum Ca2+ quartiles, a difference that was not seen in the baseline NIHSS scores among early Ca2+ quartiles.
Both early and delayed albumin-corrected calcium levels did not have any prognostic significance in this study. In a prior acute stroke study, that compared total Ca2+, ionized Ca2+ and albumin-corrected Ca2+, only the total serum Ca2+ level was shown to be of prognostic value.7 Furthermore, some have questioned the accuracy of widely used albumin-corrected Ca2+ formulas, and have suggested that total calcium may be more accurate than albumin-corrected Ca2+ if ionized Ca2+ itself cannot be directly measured.28
This study has limitations. It was a secondary analysis of a completed randomized trial, not a prospective study designed to address this issue. Thus, although we controlled for clinical and biological factors known to influence ischemic stroke outcomes, we cannot exclude the possibility that unmeasured confounding may explain some of our findings. We did not collect data on ionized Ca2+, which is the physiologically active component of serum Ca2+ levels. Additionally, we were constrained by lack of stroke subtyping information because small-vessel disease stroke subtype may carry a better prognosis than other subtypes. However, given the mRECT NIHSS score inclusion criteria,10 it is likely that mostly moderate-to-severe strokes were enrolled in this trial and small-vessel type strokes were in the minority. We also do not know whether there were any systematic differences in secondary preventative treatments between Ca2+ groups which may have impacted 3 month outcomes, although the relation between Ca2+ levels and 72- to 96-hour clinical outcomes is similar to that of the 3-month outcomes, and so we doubt that postdischarge prevention therapies were substantially different among Ca2+ groups. Finally, we lacked brain-imaging data to investigate the relationship between infarct volume and serum Ca2+ level.
In conclusion, we found Ca2+ levels obtained within 72 to 96 hours to be of prognostic significance after ischemic stroke. Specifically designed prospective studies are needed to confirm these findings, profile the time course of serum Ca2+ alterations in acute stroke, and investigate the pathophysiologic underpinnings for this potential association.
|
Acknowledgments |
|---|
The authors are grateful to Rita Engelhardt, PhD, for her help with statistics, Myzoon Ali for data processing, and to the mRECT trial investigators for contributing their data to the Virtual International Stroke Trials Archive (VISTA).
Disclosures
None.
Received January 8, 2008; accepted January 16, 2008.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Kristian T, Siesjo BK. Calcium in ischemic cell death. Stroke. 1998; 29: 705–718.
2. Siesjo B. Cerebral circulation and metabolism. J Neurosurg. 1984; 60: 883–908.[Medline] [Order article via Infotrieve]
3. Nicholson C, Bruggencate GT, Steinberg R, Stockle H. Calcium modulation in brain extracellular microenvironment demonstrated with ion-selective micropipette. Proc Natl Acad Sci USA. 1977; 74: 1287–1290.
4. Tymianski M, Tator CH. Normal and abnormal calcium homeostasis in neurons: a basis for the pathophysiology of traumatic and ischemic central nervous system injury. Neurosurgery. 1996; 38: 1176–1195.[CrossRef][Medline] [Order article via Infotrieve]
5. Ovbiagele B, Kidwell CS, Starkman S, Saver JL. Neuroprotective agents for the treatment of acute ischemic stroke. Curr Neurol Neurosci Rep. 2003; 3: 9–20.[Medline] [Order article via Infotrieve]
6. Iso H, Stampfer MJ, Manson JE, Rexrode K, Hennekens CH, Colditz GA, Speizer FE, Willett WC. Prospective study of calcium, potassium, and magnesium intake and risk of stroke in women. Stroke. 1999; 30: 1772–1779.
7. D'Erasmo E, Pisani D, Romagnoli S, Ragno A, Acca M. Acute serum calcium changes in transient ischemic attack and cerebral infarction. J Med. 1998; 29: 331–337.[Medline] [Order article via Infotrieve]
8. Lampl Y, Geva D, Gilad R, Eshel Y, Ronen L, Sarova-Pinhas I. Cerebrospinal fluid magnesium level as a prognostic factor in ischaemic stroke. J Neurol. 1998; 245: 584–588.[CrossRef][Medline] [Order article via Infotrieve]
9. Ovbiagele B, Liebeskind DS, Starkman S, Sanossian N, Kim D, Razinia T, Saver JL. Are elevated admission calcium levels associated with better outcomes after ischemic stroke? Neurology. 2006; 67: 170–173.
10. Teal P, Lyden P, Kaste M, Davis S, Hacke W, Fierus M. Effects of Repinotan in Patients with Acute Ischemic Stroke: Phase IIb modified Randomized Exposure Controlled Trial (mRECT) Paper presented at: 14th European Stroke Conference; May 28, 2005; Bologna, Italy.
11. Ali M, Bath PM, Curram J, Davis SM, Diener HC, Donnan GA, Fisher M, Gregson BA, Grotta J, Hacke W, Hennerici MG, Hommel M, Kaste M, Marler JR, Sacco RL, Teal P, Wahlgren NG, Warach S, Weir CJ, Lees KR. The Virtual International Stroke Trials Archive. Stroke. 2007; 38: 1905–1910.
12. Fiorelli M, Alperovitch A, Argentino C, Sacchetti ML, Toni D, Sette G, Cavalletti C, Gori MC, Fieschi C; Italian Acute Stroke Study Group. Prediction of long-term outcome in the early hours following acute ischemic stroke. Arch Neurol. 1995; 52: 250–255.
13. Dulli D, Stanko H, Levine RL. Atrial fibrillation is associated with severe acute ischemic stroke. Neuroepidemiology. 2003; 22: 118–123.[CrossRef][Medline] [Order article via Infotrieve]
14. Wegener S, Gottschalk B, Jovanovic V, Knab R, Fiebach JB, Schellinger PD, Kucinski T, Jungehulsing GJ, Brunecker P, Muller B, Banasik A, Amberger N, Wernecke KD, Siebler M, Rother J, Villringer A, Weih M; MRI in Acute Stroke Study Group of the German Competence Network Stroke. Transient ischemic attacks before ischemic stroke: preconditioning the human brain? A multicenter magnetic resonance imaging study. Stroke. 2004; 35: 616–621.
15. Schaller B. Ischemic preconditioning as induction of ischemic tolerance after transient ischemic attacks in human brain: its clinical relevance. Neurosci Lett. 2005; 377: 206–211.[CrossRef][Medline] [Order article via Infotrieve]
16. Bruno A, Biller J, Adams HP, Clarke WR, Woolson RF, Williams LS, Hansen MD; Trial of ORG 10172 in Acute Stroke Treatment (TOAST) Investigators. Acute blood glucose level and outcome from ischemic stroke. Neurology. 1999; 52: 280–284.
17. Blair J, Warner DS, Todd MM. Effects of elevated plasma magnesium versus calcium on cerebral ischemic injury in rats. Stroke. 1989; 20: 507–512.
18. Muir K. Magnesium in stroke treatment. Postgrad Med J. 2002; 78: 641–645.
19. Hajat C, Hajat S, Sharma P. Effects of poststroke pyrexia on stroke outcome: a meta-analysis of studies in patients. Stroke. 2000; 31: 410–414.
20. Castillo J, Davalos A, Marrugat J, Noya M. Timing for fever-related brain damage in acute ischemic stroke. Stroke. 1998; 29: 2455–2460.
21. Dawson S, Manktelow BN, Robinson TG, Panerai RB, Potter JF. Which parameters of beat-to-beat blood pressure and variability best predict early outcome after acute ischemic stroke? Stroke. 2000; 31: 463–468.
22. Frankel M, Morgenstern LB, Kwiatkowski T, Lu M, Tilley BC, Broderick JP, Libman R, Levine SR, Brott T. Predicting prognosis after stroke: a placebo group analysis from the National Institute of Neurological Disorders and Stroke rt-PA Stroke Trial. Neurology. 2000; 55: 952–959.
23. Albumin-corrected Calcium Formula. Available at www.globalrph.com. Last accessed July 29th, 2007.
24. Zaloga G. Hypocalcemia in critically ill patients. Crit Care Med. 1992; 20: 251–262.[Medline] [Order article via Infotrieve]
25. Chernow B, Zaloga G, McFadden E, Clapper M, Kotler M, Barton M, Rainey TG. Hypocalcemia in critically ill patients. Crit Care Med. 1982; 10: 848–851.[Medline] [Order article via Infotrieve]
26. Carlstedt F, Lind L, Rastad J, Stjernstrom H, Wide L, Ljunghall S. Parathyroid hormone and ionized calcium levels are related to the severity of illness and survival in critically ill patients. Eur J Clin Invest. 1998; 28: 898–903.[CrossRef][Medline] [Order article via Infotrieve]
27. Zivin J, Gooley T, Zager RA, Ryan MJ. Hypocalcemia: a pervasive metabolic abnormality in the critically ill. Am J Kidney Dis. 2001; 37: 689–698.[Medline] [Order article via Infotrieve]
28. Clase C, Norman GL, Beecroft ML, Churchill DN. Albumin-corrected calcium and ionized calcium in stable haemodialysis patients. Nephrol Dial Transplant. 2000; 15: 1841–1846.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||