Low Serum Calcium Levels Contribute to Larger Hematoma Volume in Acute Intracerebral Hemorrhage
Background and Purpose—We investigate whether admission serum calcium levels are associated with hematoma volume, stroke severity, and outcomes in patients with acute intracerebral hemorrhage.
Methods—A total of 273 patients admitted within 24 hours after intracerebral hemorrhage onset was divided into quartiles based on admission serum calcium levels (Q1 [≤9.0], Q2 [9.1–9.3], Q3 [9.4–9.7], Q4 [≥9.8] mg/dL).
Results—Median hematoma volumes for each quartile (Q1 to Q4) were 18, 9, 10, and 9 mL (P=0.005), and median National Institutes of Health Stroke Scale scores were 16, 11, 11, and 9 (P=0.010), respectively. On multivariate analysis, Q1 had larger hematoma volume (P=0.025) and higher National Institutes of Health Stroke Scale score (P=0.020) than Q4. There were fewer patients with modified Rankin Scale scores 0 to 2 in Q1 than Q4 after adjustment for risk factors and comorbidities (odds ratio, 0.31; 95% confidence interval, 0.11–0.84) but not after additional adjustment for hematoma volume and National Institutes of Health Stroke Scale score. There were more patients with modified Rankin Scale scores 5 to 6 (P=0.016) and with fatal outcomes (P=0.048) in Q1 than Q4 as crude values, but not after adjustment.
Conclusions—Low admission serum calcium levels were associated with larger hematoma volume and higher National Institutes of Health Stroke Scale score among patients with acute intracerebral hemorrhage.
Intracerebral hemorrhage (ICH) is associated with poor outcome, a high mortality rate, and little effective treatment.1 In patients with acute ICH, large hematoma volume, the presence of intraventricular bleeding, and prior use of anticoagulants or antiplatelets are reported to be associated with poor outcome.2,3
Several studies have reported that low serum calcium (Ca) levels have an association with large infarcts and poor outcome among patients with ischemic stroke.4–6 In another study involving a majority of patients with ischemic stroke and few patients with ICH, low serum Ca levels were also associated with poor outcome.7 In addition, ionized Ca is an essential cofactor for the coagulation cascade and associated with conversion of prothrombin to thrombin. Association of Ca levels with changes in clotting time and bleeding tendency was shown in the rodent models.8,9 Thus, serum Ca levels might play a pivotal role in hemostasis in acute ICH.
However, associations of serum Ca levels with clinical findings and outcomes of isolated patients with ICH remain unknown. Thus, the aim of the present study was to examine the associations between admission Ca levels and clinical findings and outcomes in patients with acute ICH.
Patients and Methods
Consecutive patients admitted to our department within 24 hours from the onset of nontraumatic ICH were studied using our prospectively collected database on inpatients with stroke. Patients were classified into quartiles based on admission serum Ca levels (Q1 [≤9.0], Q2 [9.1–9.3], Q3 [9.4–9.7], Q4 [≥9.8] mg/dL). Acute outcomes included the following: the National Institutes of Health Stroke Scale (NIHSS) score at admission, initial hematoma volume, and hematoma growth. Chronic outcomes included the following: modified Rankin Scale (mRS) scores of 0 to 1 and 0 to 2, bedridden state or death corresponding to mRS scores of 5 to 6 at 30 days, and mortality. For chronic outcomes, patients with prestroke mRS score ≥2 were excluded from the analyses. Associations between each Ca quartile and outcomes were determined using multivariate regression models adjusted by the baseline characteristics automatically selected in a backward stepwise selection method (see Methods in the online-only Data Supplement).
A total of 273 patients (92 women, 70±11 years old) were studied. There were fewer women (P=0.036), liver dysfunction was more common (P=0.043), and levels of albumin (P<0.001), total cholesterol (P=0.002), low-density lipoprotein–cholesterol (P<0.001), high-density lipoprotein–cholesterol (P=0.027), and hemoglobin (P=0.002) were lower in the lower Ca level quartiles than in the higher Ca level quartiles (see Tables I and II in the online-only Data Supplement).
The lowest Ca quartile had higher hematoma volume (P=0.005; Table 1). After multivariate linear regression analyses, Q1 had larger hematoma volume than Q4 (P=0.015). The lowest Ca quartile had a higher NIHSS score (P=0.010). After multivariate linear regression analyses, Q1 had a higher NIHSS score than Q4 (P=0.010).
The Figure shows the distribution of mRS scores. No patients in Q1 had mRS score 0 to 1, as compared with other quartiles (P<0.001). Patients with mRS score 0 to 2 were less frequent in Q1 than Q4 as crude values (P=0.002) and after adjustment with risk factors and comorbidities (P=0.026), but they were no longer different after additional adjustment with hematoma volume and NIHSS score (Table 2). Patients with mRS score 5 to 6 and those with fatal outcomes were more frequent in Q1 than Q4 as crude values (P=0.016 and 0.048, respectively), but the difference disappeared after multivariate adjustment (Table III in the online-only Data Supplement).
This is the first study that investigated the relationships between serum Ca levels at admission and clinical findings and outcomes of patients with acute ICH. The major new finding was that patients with low Ca levels had larger hematoma volumes and higher NIHSS scores at admission.
Three possible mechanisms may explain why serum Ca levels are related to severity at admission. First, as stated above, ionized Ca is an essential cofactor for the coagulation cascade. Second, low serum Ca levels might contribute to hematoma enlargement through blood pressure elevation in the acute ICH. Ca was reported to induce relaxation of isolated arteries by activating Ca receptors in perivascular nerves.10 Third, low serum Ca levels might reflect poor liver function. The lowest Ca quartile (Q1) had high percentage of liver disease, and these findings suggested another mechanism for poor coagulation and therefore larger hematoma volume.
This study was limited by its design as a single-center, hospital-based, retrospective analysis with exclusion of some patients attributable to lack of serum Ca level data. To simplify reporting of results, data on ionized Ca levels and albumin-corrected Ca levels were not shown. Instead, the serum albumin level was used for multivariate regression analyses and it did not affect the results.
In conclusion, low Ca levels in the hyperacute stage of ICH are indicators of larger hematoma volume and more severe neurological deficit. There is an apparent association between serum Ca levels and hematoma volume. Future studies could incorporate serum Ca in models and test for enhanced predictive power over existing models. The clinical benefit and therapeutic time window for Ca level modification, as well as the Ca threshold level that is beneficial for ICH prevention, remain unknown, and further investigations of these issues will be required.
Sources of Funding
This study was supported in part by grants-in-aid (H23-Junkanki-Ippan-010, H24-Junkanki-Ippan-011) from the Ministry of Health, Labor and Welfare, Japan, and the Intramural Research Fund for Cardiovascular Diseases from the National Cerebral and Cardiovascular Center (H22-4-1, H23-4-3).
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.113.001187/-/DC1.
- Received February 13, 2013.
- Accepted March 26, 2013.
- © 2013 American Heart Association, Inc.
- Broderick JP,
- Brott TG,
- Duldner JE,
- Tomsick T,
- Huster G
- Ovbiagele B,
- Starkman S,
- Teal P,
- Lyden P,
- Kaste M,
- Davis SM,
- et al