Increased glycemic variability associated with a poor 30-day functional outcome in acute intracerebral hemorrhage

Full access

The authors analyzed the association between the standard deviation or the coefficient of variation in the glucose value, strong independent indexes for determining glycemic variability, and the prognosis of intracerebral hemorrhage. They found that glycemic variability may be associated with a poor outcome in intracerebral hemorrhage.

ABBREVIATIONS AUC = area under the receiver operating characteristic curve; CV = coefficient of variation; DBP = diastolic blood pressure; DM = diabetes mellitus; FBG = fasting blood glucose; GCS = Glasgow Coma Scale; GluCV = coefficient of variation of glucose; GluSD = standard deviation of glucose; GV = glycemic variability; HbAlc = glycosylated hemoglobin; HDL-C = high-density lipoprotein cholesterol; ICH = intracerebral hemorrhage; IVH = intraventricular hematoma; LDL-C = low-density lipoprotein cholesterol; MI = myocardial infarction; mRS = modified Rankin Scale; NG = normal glucose; SBP = systolic blood pressure; SD = standard deviation; SHG = stress hyperglycemia; TC = total cholesterol; TG = triglyceride; WBC = white blood cell.

Abstract

The authors analyzed the association between the standard deviation or the coefficient of variation in the glucose value, strong independent indexes for determining glycemic variability, and the prognosis of intracerebral hemorrhage. They found that glycemic variability may be associated with a poor outcome in intracerebral hemorrhage.

Abstract

OBJECTIVE

Intracerebral hemorrhage (ICH) is associated with a poor prognosis and high mortality, but no study has elucidated the association between glycemic variability (GV) and functional outcome in ICH. The authors of this study aimed to determine whether GV is a predictor of 30-day functional outcome in ICH patients.

METHODS

The study recruited 366 patients with first-ever acute-onset ICH in the period during 2014 and 2015. Fasting blood glucose was assessed on admission and with 7-day continuous monitoring. Glycemic variability was calculated and expressed by the standard deviation (GluSD) and coefficient of variation (GluCV). Patients were divided into groups of those with diabetes mellitus (DM), stress hyperglycemia (SHG), and normal glucose (NG). Functional outcome was measured using the modified Rankin Scale.

RESULTS

The numbers of patients with DM, SHG, and NG were 108 (29.5%), 127 (34.7%), and 131 (35.8%), respectively. As compared with the DM patients, those with SHG had higher mortality (29.9% vs 15.7%, p < 0.05) and a poorer prognosis (64.6% vs 52.8%, p < 0.05). Poor prognosis was associated with both high GluSD (OR 1.54, 95% CI 1.19–1.99) and high GluCV (1.05, 1.02–1.09), especially in the DM group. The area under the receiver operating characteristic curve was greater for the GluSD (OR 0.929, 95% CI 0.902–0.956) and the GluCV (0.932, 0.906–0.958) model than the original model (0.860, 0.823–0.898) in predicting a poor outcome.

CONCLUSIONS

Stress hyperglycemia may be associated with increased mortality and a poor outcome in ICH, and increased GV may be independently associated with a poor outcome, particularly in ICH patients with DM.

Stroke is the second leading cause of death worldwide and the first in China, and the absolute number of people with a first stroke, who are stroke survivors, who have stroke-related deaths, and who have disability-adjusted life-years lost due to stroke has increased in the past 2 decades.11,40 Intracerebral hemorrhage (ICH) is a serious disease, accounting for 10%–15% of stroke cases, and has a high case fatality rate of about 35%–50% as well as a poor prognosis, with only 10%–20% surviving and living independently at 30 days.20,30 Risk factors for early death and a poor functional outcome in ICH include a low Glasgow Coma Scale (GCS) score at admission, large hematoma volume, intraventricular hematoma (IVH), and hypertension.15,23,26 Early intervention and management of these factors may ameliorate the poor prognosis in ICH.

Hyperglycemia is a common phenomenon in ICH patients and can be caused by diabetes mellitus (DM) or stress hyperglycemia (SHG).27 The latter disorder generally refers to transient hyperglycemia in patients without previous evidence of DM.8 The measurement of glycosylated hemoglobin (HbAlc) has unique application value, and an HbAlc ≥ 6.5% is widely considered the gold standard for DM diagnosis.1 Many studies have investigated the effect of hyperglycemia on the prognosis of ICH but have not measured the HbAlc level to distinguish whether the hyperglycemia resulted from DM or SHG.2,14,32,33 The level of glucose in SHG can revert to normal but continues to be high in DM, so it must be accounted for in continuous glucose monitoring. No studies have observed the relationship between glycemic fluctuation and functional outcome in acute-onset ICH.

Recently, glycemic variability (GV), an important index of glycemic fluctuations measured with continuous glucose monitoring, has been implicated in the disease-associated process of dysglycemia.24 The standard deviation (SD) and coefficient of variation (CV) of the glucose value are considered strong independent indexes for determining GV, although the best method for characterizing GV in hospitalized patients has not been agreed on.29,31 Several studies have demonstrated high GV associated with significantly increased mortality in critically ill patients.9,19 However, no study has reported on the association between GV and clinical outcomes in hospitalized patients, particularly those with acute-onset ICH.

In the present study we emphasize the importance of distinguishing between DM and SHG in hyperglycemia and the fasting blood glucose (FBG) level monitored for 7 continuous days. We evaluated the association of hyperglycemia in DM or SHG and GV as measured by continuous monitoring of FBG for 7 days with functional outcome at 30 days in acute-onset ICH.

Methods

Study Participants

We recruited 366 patients with first-ever acute-onset ICH who were admitted to the Department of Neurology Medicine and Surgery Services in the First Affiliated Hospital of Shantou University Medical College between January 1, 2014, and December 31, 2015. All patients were 18 years of age or older and were admitted to the hospital within 24 hours of the first symptoms of disease, with the diagnosis of ICH confirmed by CT or MRI studies. On the basis of ICH guidelines from the American Heart Association, clinical management was established at the discretion of the treating physician.17 Patients with pituitary tumors, hyperthyroidism, acute pancreatitis, and endocrine tumors that may cause hyperglycemia were excluded. Patients with traumatic hemorrhage or recurrent episodes of hemorrhage (for example, subarachnoid hemorrhage, brain tumor or hemorrhagic transformation of ischemic stroke, and subdural or extradural hemorrhage) were also excluded. This study was conducted according to guidelines in the Declaration of Helsinki, and all procedures involving humans were approved by the Ethics Committee of the First Affiliated Hospital of Shantou University Medical College. Written informed consent was obtained from all participants. The required sample size was determined by assuming an OR 1.5 for group comparison at the 5% significance level; therefore, a sample size of 53 per group was needed to attain 90% power, as calculated by NCSS-PASS 2005.

Neuroradiological Methods

From CT scans and MR images, we recorded hematoma location (basal ganglia, lobar, brainstem, cerebellum, or thalamus), presence of IVH, and hematoma volume (classified as < 20, 20–40, and ≥ 40 ml) measured using the ABC/2 method (A, greatest hemorrhage diameter; B, diameter perpendicular to A; C, number of slices multiplied by slice thickness).18 Initial neurological deficit was determined using the GCS, usually used to assess coma and impaired consciousness;34 GCS scores were classified as 13–15 (mild disability), 9–12 (moderate disability), and ≤ 8 (severe disability). Functional outcome was measured using the modified Rankin Scale (mRS),4 scored by a designated neurologist to evaluate therapeutic effects and prognosis at 30 days after admission. The mRS scores were divided into good outcome (mRS score < 3) and poor outcome (mRS score ≥ 3), with an mRS Score of 6 considered as death.39

Medical Records

Medical records were examined for previous diseases including myocardial infarction (MI), hypertension, and DM. Diagnosis of prior MI was based on the medical history of the patient. In accordance with the World Health Organization (WHO)/International Society of Hypertension statement,38 hypertension was diagnosed as blood pressure ≥ 140/90 mm Hg measured at least twice or if a patient was taking antihypertensive medication. Diabetes mellitus was defined according to the criteria of the American Diabetes Association—that is, FBG ≥ 7.0 mmol/L, random glucose ≥ 11.1 mmol/L measured at least twice, or HbAlc ≥ 6.5%—or as a history of DM (medical record of DM and/or taking insulin or oral hypoglycemic agents).1 Accordingly, all participants were divided into 3 groups: DM (history of DM or HbAlc ≥ 6.5% and FBG ≥ 7.0 mmol/L), SHG (no history of DM, HbAlc < 6.5%, and FBG ≥ 7.0 mmol/L), and normal glucose (NG; no history of DM, HbAlc < 6.5%, and FBG < 7.0 mmol/L).

We collected venous blood from fasting patients at least 8 hours after ICH onset and at 5:00–6:00 each morning of the 7 consecutive days after admission. All examinations were performed in the clinical laboratory of the hospital. Fasting blood glucose was measured via the enzymatic method using an auto-analyzer (Beckman Coulter Inc.), and HbAlc was measured via the high-performance liquid chromatographic method using an automated glycohemoglobin analyzer (Tosoh Corp.).

Glycemic indices such as SD of 7-day FBG level (GluSD) and CV of 7-day FBG level (GluCV) were calculated; GluCV was calculated using the equation GluCV = 100% × GluSD/(mean glucose).28 Patient age and sex, health habits (for example, smoking status, alcohol intake), systolic blood pressure (SBP), diastolic blood pressure (DBP), pulse pressure (calculated as SBP − SDP), white blood cell (WBC) count (109/L), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and triglycerides (TG) at admission were also collected.

Statistical Analysis

Qualitative variables are expressed as frequencies (%) and continuous variables as the mean ± standard deviation. Continuous variables were analyzed for normality using the Kolmogorov-Smirnov test. Significant differences between groups were assessed with the Mann-Whitney U-test for skewed variables and Pearson’s chi-square test with or without Bonferroni correction for categorical variables. Bivariate analysis was initially performed to identify a significant association between individual variables and 30-day functional outcome. Variables with a p ≤ 0.05 on bivariate analysis were included in the multivariate model (that is, the original model). In the multivariate models, variables with a p < 0.05 were retained by forward logistic regression elimination. Odds ratios with 95% confidence intervals were calculated. We separately added GluSD and GluCV to the original model to create the GluSD model and GluCV models. We also analyzed the area under the receiver operating characteristic curve (AUC) to judge the discrimination ability of various statistical methods.16 Statistical analyses involved the use of SPSS for Windows 13.0 (SPSS Inc.) and R version 3.2.0 (R Foundation for Statistical Computing, http://www.R-project.org). Two-tailed tests of significance were reported, with p < 0.05 considered statistically significant

Results

Study Population

A total of 366 ICH patients were divided into 3 glucose groups, with 108 (29.5%), 127 (34.7%), and 131 (35.8%) in the DM, SHG, and NG groups, respectively (Table 1). The rate of newly diagnosed DM on admission, according to an HbAlc level ≥ 6.5%, was 15.3% (56/366). The variables hypertension, SBP, DBP, initial GCS score, hematoma volume, IVH, WBC count, TC, TG, admission FBG and HbAlc levels, GluSD, and GluCV all significantly differed among the 3 glucose groups (all p < 0.05). Hyperglycemia was frequent in patients with a low initial GCS score (≤ 8), hematoma volume ≥ 40 ml, and IVH (all p < 0.05). Patients in the SHG group had the highest WBC count, and those in the DM group had the highest HbAlc level (both p < 0.05). As compared with the NG group, both the DM and SHG groups had higher SBP, admission FBG level, and GluSD and GluCV for 7-day FBG (all p < 0.05). The dynamic changes in 7-day FBG in the 3 glucose groups are represented in Fig. 1. The level of FBG was higher in both the DM and SHG groups than in the NG group, with a downward trend.

TABLE 1.

Baseline clinical characteristics of 366 patients with ICH, according to glucose group

VariableTotal (n = 366)DM (n = 108)SHG (n = 127)NG (n = 131)p Value*
Age ≥65 yrs156 (42.6)53 (49.1)49 (38.6)54 (41.2)0.248
Males217 (59.3)63 (58.3)71 (55.9)83 (63.4)0.463
Previous disease
 Hypertension271 (74.0)96 (88.9)91 (71.6)84 (64.1)<0.001
 MI102 (27.9)31 (28.7)36 (28.3)35 (26.7)0.933
Smoking166 (45.4)47 (43.5)57 (44.9)62 (47.3)0.834
Drinking114 (31.1)39 (36.1)41 (32.3)34 (26.0)0.227
SBP in mm Hg164.6 ± 31.4171.9 ± 28.7168.1 ± 31.9155.7 ± 31.2<0.001
DBP in mm Hg97.2 ± 18.797.9 ± 16.5100.5 ± 20.193.4 ± 18.50.010
PP in mm Hg67.6 ± 20.874.0 ± 19.167.7 ± 22.462.2 ± 19.10.356
Initial GCS score<0.001
 13–15145 (39.6)34 (31.5)37 (29.1)74 (56.5)
 9–12102 (27.9)40 (37.0)28 (22.0)34 (26.0)
 ≤8119 (32.5)34 (31.5)62 (48.8)23 (17.6)
Hematoma location0.966
 Basal ganglia151 (41.2)44 (40.7)54 (42.5)53 (40.4)
 Lobar94 (25.7)26 (24.1)32 (25.2)36 (27.5)
 Brainstem39 (10.6)12 (11.1)15 (11.8)12 (9.2)
 Cerebellum48 (13.1)13 (12.0)16 (12.6)19 (14.5)
 Thalamus34 (9.3)13 (12.0)10 (7.9)11 (8.4)
Hematoma vol in ml<0.001
 <20179 (48.9)47 (43.5)48 (37.8)84 (64.1)
 20–40102 (27.9)39 (36.1)32 (25.2)31 (23.7)
 ≥4085 (23.2)22 (20.4)47 (37.0)16 (12.2)
IVH189 (51.6)57 (52.8)83 (65.4)49 (37.4)<0.001
WBC count in 109/L12.6 ± 7.011.9 ± 5.314.5 ± 10.211.4 ± 4.60.002
TC in mmol/L4.8 ± 1.35.2 ± 1.74.8 ± 1.24.6 ± 1.20.023
LDL-C in mmol/L3.0 ± 1.43.1 ± 1.42.8 ± 1.52.9 ± 1.20.427
HDL-C in mmol/L1.1 ± 0.51.1 ± 0.41.1 ± 0.61.1 ± 0.40.645
TG in mmol/L1.5 ± 1.41.7 ± 1.61.2 ± 1.11.4 ± 1.30.011
Admission FBG level in mmol/L8.7 ± 4.211.3 ± 4.510.2 ± 3.75.2 ± 0.9<0.001
Admission HbAlc level in %6.5 ± 1.47.8 ± 1.96.0 ± 0.35.7 ± 0.2<0.001
GluSD in mmol/L1.9 ± 1.52.7 ± 1.62.4 ± 1.30.7 ± 0.3<0.001
GluCV in %22.5 ± 11.627.5 ± 11.727.4 ± 11.113.74 ± 5.0<0.001

n = number of patients; PP = pulse pressure.

Data are expressed as the mean ± SD or as number (%). Boldface type indicates statistical significance.

Comparison among the DM, SHG, and NG groups.

p < 0.05, compared with the DM group.

p < 0.05, compared with the SHG group.

FIG. 1.
FIG. 1.

Dynamic changes in 7-day FBG level in 3 glucose groups of hospitalized patients with ICH. Figure is available in color online only.

Study End Points

Clinical Characteristics

In total, 46.7% of the ICH patients (171/366) had a poor outcome (mRS score ≥ 3) at 30 days. The categories of age ≥ 65 years, hypertension or MI, smoking, drinking, initial GCS score, hematoma location, IVH, hematoma volume, SBP, LDL-C, HDL-C, TG, admission FBG and HbAlc levels, WBC count, GluSD, and GluCV differed between the good and poor outcome groups (all p < 0.05; Table 2).

TABLE 2.

Baseline clinical characteristics associated with 30-day functional outcome in 366 ICH patients

VariableGood OutcomePoor Outcomep Value
No. of patients195 (53.3)171 (46.7)
Age ≥65 yrs63 (32.3)93 (54.4)<0.001
Males112 (57.4)105 (61.4)0.457
Previous disease
 Hypertension128 (65.6)143 (83.6)<0.001
 MI44 (22.6)58 (33.9)0.016
Smoking76 (39.0)90 (52.6)0.006
Drinking44 (22.6)70 (40.9)<0.001
SBP (mm Hg)158.2 ± 31.2171.8 ± 30.3<0.001
DBP (mm Hg)94.5 ± 16.9100.2 ± 20.10.063
PP (mm Hg)64.1 ± 20.371.6 ± 20.70.051
Initial GCS score<0.001
 13–15117 (60.0)28 (16.4)
 9–1261 (31.3)41 (24.0)
 ≤817 (8.7)102 (59.6)
Hematoma location<0.001
 Basal ganglia74 (38.0)77 (45.0)
 Lobar63 (32.3)31 (18.1)
 Brainstem8 (4.1)31 (18.1)
 Cerebellum31 (15.9)17 (9.9)
 Thalamus19 (9.7)15 (8.8)
Hematoma vol in ml<0.001
 <20139 (71.3)40 (23.4)
 20–4039 (20.0)63 (36.8)
 ≥4017 (8.7)68 (39.8)
IVH58 (29.7)131 (76.6)<0.001
Admission FBG in mmol/L7.5 ± 3.410.16 ± 4.71<0.001
Admission HbAlc in %6.2 ± 0.46.8 ± 1.80.002
WBC (109/L)11.0 ± 4.414.4 ± 9.3<0.001
TC (mmol/L)4.9 ± 1.54.6 ± 1.20.530
LDL-C in mmol/L3.2 ± 1.32.7 ± 1.4<0.001
HDL-C in mmol/L1.2 ± 0.41.1 ± 0.50.015
TG in mmol/L1.6 ± 1.51.3 ± 1.20.030
GluSD in mmol/L1.4 ± 1.12.5 ± 1.6<0.001
GluCV in %19.5 ± 10.126.0 ± 12.3<0.001
Glucose groups<0.001
 NG99 (50.8)32 (18.7)
 SHG45 (23.1)82 (48.0)
 DM51 (26.2)57 (33.3)

Data are expressed as the mean ± standard deviation or as number (%). Boldface type indicates statistical significance.

The poor prognosis rates in the DM, SHG, and NG groups were 52.8%, 64.6%, and 24.4%, respectively (Fig. 2). The poor prognosis rate was highest in the SHG group as compared with the DM (p = 0.045) and NG groups (p < 0.001) and was significantly higher in the DM group than in the NG group (p < 0.001). Overall mortality at 30 days was 16.1% for the entire cohort. Mortality was highest for the SHG group compared to the DM group (29.9% vs 15.7%, p = 0.011) and the NG group (3.1%, p < 0.001), and mortality was significantly higher for the DM group than the NG group (p < 0.001).

FIG. 2.
FIG. 2.

Poor prognosis and mortality rate at 30 days in 3 glucose groups of hospitalized ICH patients.

An independent risk factor for a poor outcome in ICH patients was SHG (OR 2.62, 95% CI 1.07–6.44) but not DM (1.69, 0.64–4.43; Table 3). Poor prognosis was associated with an increased FBG level on admission (OR 1.13, 95% CI 1.05–1.22). Other predictors were an age ≥ 65 years, initial GCS score ≤ 8, brainstem hematoma, and hematoma volume ≥ 20 ml (all p < 0.05).

TABLE 3.

Multivariate analysis of independent predictors of 30-day poor outcome after ICH (original model)

VariableOR (95% CI)p Value
Age ≥65 yrs4.63 (2.35–9.09)<0.001
Initial GCS score<0.001
 13–151.00
 9–121.17 (0.53–2.54)0.700
 ≤87.09 (2.98–16.82)<0.001
Hematoma location<0.001
 Basal ganglia1.00
 Lobar0.86 (0.36–2.02)0.720
 Brainstem23.68 (6.51–86.16)<0.001
 Cerebellum1.19 (0.43–3.25)0.738
 Thalamus0.60 (0.20–1.81)0.365
Hematoma vol in ml<0.001
 <201.00
 20–406.52 (2.93–14.47)<0.001
 ≥407.14 (2.80–18.26)<0.001
IVH5.14 (2.54–10.42)<0.001
Admission FBG level1.13 (1.05–1.22)0.002
Glucose group0.012
 NG1.00
 SHG2.62 (1.07–6.44)0.036
 DM1.69 (0.64–4.43)0.287

Boldface type indicates statistical significance.

Association Between GV and Outcome

On separately adding GluSD and GluCV to the GluSD and GluCV models, respectively, based on the significant variables on bivariate analysis, we found that an independent risk factor for a poor outcome was a high level of GluSD (OR 1.54, 95% CI 1.19–1.99) or GluCV (1.05, 1.02–1.09); the WBC count was newly included in the model, and admission FBG level was excluded (Table 4). An age ≥ 65 years, initial GCS score ≤ 8, brainstem hematoma, and hematoma volume ≥ 40 ml were still predictors of a poor outcome (all p < 0.05).

TABLE 4.

Multivariate logistic regression analysis of the association between GV (GluSD or GluCV) and 30-day poor outcome after ICH

VariableGluSD ModelGluCV Model
Entire CohortDM SubsetSHG SubsetNG SubsetEntire CohortDM SubsetSHG SubsetNG Subset 
Age ≥65 yrs6.21 (3.02–12.78)9.79 (1.34–71.57)9.32 (2.48–35.08)10.52 (2.45–45.15)6.39 (3.08–13.22)8.92 (1.87–56.85)8.76 (2.36–32.54)10.39 (2.34–44.43)
Initial GCS score
 13–151.001.001.001.001.001.001.001.00
 9–120.95 (0.42–2.13)1.12 (0.32–30.21)1.36 (0.27–6.87)5.31 (0.89–31.81)1.03 (0.46–2.30)2.31 (0.34–32.36)1.48 (0.29–7.47)5.20 (0.88–30.91)
 ≤84.59 (1.86–11.36)2.13 (0.11–0.89)2.16 (0.44–10.65)34.14 (4.36–67.17)5.08 (2.08–12.43)3.09 (0.11–0.98)2.56 (0.56–11.77)27.21 (4.38–66.96)
Hematoma location
 Basal ganglia1.001.001.001.001.001.001.001.00
 Lobar0.72 (0.30–1.73)1.14 (0.11–12.42)1.10 (0.22–5.61)0.43 (0.07–2.81)0.72 (0.30–1.72)1.28 (0.11–15.55)1.17 (0.23–5.89)0.42 (0.06–2.72)
 Brainstem19.29 (5.16–72.16)13.37 (3.85–27.34)26.50 (3.74–66.85)7.69 (0.51–15.01)21.81 (5.81–81.89)58.21 (7.11–93.82)91.34 (4.68–83.47)7.38 (0.50–14.69)
 Cerebellum0.76 (0.26–2.25)0.19 (0.08–4.11)1.66 (0.23–11.70)1.83 (0.20–16.63)0.78 (0.27–2.28)0.14 (0.01–4.10)2.07 (0.32–13.07)1.77 (0.20–15.93)
 Thalamus0.55 (0.18–1.71)0.38 (0.03–4.39)0.90 (0.12–6.59)0.59 (0.05–7.05)0.52 (0.17–1.62)0.29 (0.02–3.62)0.81 (0.12–5.64)0.59 (0.05–7.08)
Hematoma vol in ml
 <201.001.001.001.001.001.001.001.00
 20–400.84 (0.44–2.80)0.26 (0.20–3.49)3.48 (0.67–18.01)3.95 (0.44–35.78)5.72 (2.51–13.06)8.74 (1.21–62.99)3.78 (0.74–19.38)4.01 (0.44–36.64)
 ≥406.21 (3.46–12.72)1.02 (0.03–0.54)10.27 (1.53–69.03)11.35 (2.16–59.53)7.59 (2.83–20.34)22.12 (3.44–80.51)12.72 (2.02–80.15)11.13 (2.13–58.17)
IVH5.87 (2.80–12.30)29.52 (9.84–69.63)4.37 (1.07–17.87)3.50 (0.76–16.16)5.78 (2.77–12.05)22.58 (10.04–96.55)4.06 (0.94–17.56)3.46 (0.75–16.01)
WBC1.07 (1.04–1.16)1.04 (0.87–1.25)1.22 (1.07–1.40)0.95 (0.79–1.16)1.08 (1.03–1.15)1.04 (0.86–1.27)1.22 (1.07–1.39)0.96 (0.79–1.16)
GluSD1.54 (1.19–1.99)1.94 (1.13–3.34)1.24 (0.65–2.36)1.90 (0.11–33.31)Not includedNot includedNot includedNot included
GluCVNot includedNot includedNot includedNot included1.05 (1.02–1.09)1.14 (1.04–1.26)0.98 (0.93–1.05)1.02 (0.88–1.19)

Data expressed as OR (95% CI). Boldface type indicates statistical significance.

On stratification by glucose group, all risk factors in the entire cohort (that is, age, initial GCS score, hematoma location, hematoma volume, IVH, WBC count, and GluSD) associated with 30-day functional outcome were included in the GluSD model. Risk factors for a poor outcome were a high GluSD (OR 1.94, 95% CI 1.13–3.34) in the DM subset and a high WBC count (1.22, 1.07–1.40) in the SHG subset. An age ≥ 65 years and large hematoma volume were still risk factors for a poor outcome in all subsets; brainstem hematoma and IVH were significant predictors in the DM and SHG subsets; and initial GCS score ≤ 8 was a predictor of a poor outcome in the DM and NG subsets (all p < 0.05).

A similar analysis was performed for the GluCV model. Here too risk factors for a poor outcome were a high GluCV (OR 1.14, 95% CI 1.04–1.26) in the DM subset and a high WBC count (1.22, 1.07–1.39) in the SHG subset. An age ≥ 65 years and hematoma volume ≥ 40 ml remained predictors of a poor outcome in all subsets (Table 4).

Assessing Prediction Models

The AUC was 0.860 (95% CI 0.823–0.898) for the original logistic model, 0.929 (0.902–0.956) for the GluSD model, and 0.932 (0.906–0.958) for the GluCV model (Fig. 3). The AUC for the GluSD or GluCV model was greater than that for the original model in predicting a poor outcome; thus, adding GV for 7-day FBG to the model improved prediction of the 30-day functional outcome for patients with acute ICH.

FIG. 3.
FIG. 3.

Receiver operating characteristic curves for the original, GluSD (SD), and GluCV (CV) models for predicting 30-day poor outcome in ICH. Figure is available in color online only.

Discussion

Hyperglycemia has been found to be associated with a poor prognosis in ICH, creating much attention among clinical scientists.2,3,14,32,33 However, there is not enough emphasis on distinguishing between SHG and DM among ICH patients in the clinic. Moreover, the study of the correlation of continuous glucose monitoring and GV with the prognosis of ICH has not been performed. The current study may be the first to focus on the functional outcome associated with continuous glucose monitoring in ICH. We monitored the FBG level on admission and for 7 continuous days after admission in patients with acute-onset ICH. We investigated the associations among admission hyperglycemia, fluctuation in the continuous 7-day FBG level, and 30-day functional outcome.

Hyperglycemia and Outcome After ICH

We found that 64.2% of ICH patients had hyperglycemia at admission, which included 29.5% of patients in the SHG group and 34.7% in the DM group. We found that patients with a low initial GCS score, large hematoma volume, and IVH on admission frequently had hyperglycemia. Moreover, a high admission FBG level was an independent risk factor for an early (30-day) poor outcome in acute ICH, and patients with SHG had a higher mortality and poorer prognosis rate than those with DM. A possible reason is that ICH patients with SHG had lower initial GCS scores, larger hematoma volumes, higher WBC counts, and IVH more often than the patients with DM or normal blood glucose. Related literature has shown that hyperglycemia substantially affects neurological functional recovery and prognosis in ICH, regardless of whether the hyperglycemia is caused by diabetes or not,13 although DM can be characterized by chronic sustained hyperglycemia and the influence of the stress reaction may not be severe.

A study in Finland reported that DM predicted a poor prognosis and an early death in ICH and that admission hyperglycemia was not a risk factor for a poor functional outcome, although the study did not distinguish hyperglycemia in undiagnosed DM from that in SHG by measuring HbAlc level.35 With its unique application value, HbAlc is widely considered the gold standard for DM diagnosis, and hyperglycemia according to this value in undiagnosed DM can affect the analysis of functional outcome.8,37 We measured the HbAlc level and found that 15.3% of ICH patients had newly diagnosed DM, which corresponded well with the 12%–25% reported by Clement and colleagues.6 Moreover, we found that a high FBG level on admission was associated with a 30-day poor functional outcome for ICH patients, particularly those with SHG.

Generally, glucose management for SHG and DM is similar in the acute setting but different later. Glucose management in the enrolled patients with ICH included glucose control, anti-infection agents, nutrient support, treatment of primary disease, and prevention of complications, strategies mainly based on the guidelines of the American Heart Association17 and the standards of medical care in DM.1 Glucose control for DM and SHG was similar when the glucose level was ≥ 10.0 mmol/L; that is, therapy with a dose of 1–2 U of insulin or more per hour via intravenous insulin pump was administered until the glucose level was decreased to 7.8−10.0 mmol/L for at least 2 days. However, when the glucose level was < 10.0 mmol/L in patients with DM, hypoglycemic treatment via subcutaneous insulin injection or oral hypoglycemic agents was continued to keep the glucose level in the normal range (4.4–6.1 mmol/L; diet and exercise can also be suggested); glucose levels only needed to be monitored in the patients with SHG.

Glycemic Mechanism of Outcome

We found a high proportion of ICH patients with acute hyperglycemia on admission, and SHG predicted a poor functional outcome. Hyperglycemia induces an inflammatory response and promotes hemoglobin, ferroheme, and iron release after red blood cell lysis,12 and inflammation progresses in response to various stimuli to produce inflammatory signaling via activated microglia, subsequently releasing proinflammatory cytokines and chemokines that attract peripheral inflammatory infiltrates.41 Furthermore, the inflammatory process seems to enhance glucose toxicity in brain tissue, which also leads to metabolic dysregulation and secondary brain injury.36

The pathophysiological mechanisms of hyperglycemia are different between SHG and DM. Stress hyperglycemia is usually defined as hyperglycemia resolving spontaneously after the dissipation of acute illness. It represents a transient hyperglycemia lasting for a few days or weeks in patients without previous evidence of DM, and it is probably related to the primary disease, underlying type of disease, severity of disease, and stage of illness. With timely and appropriate treatment of the primary disease, effects of the stress reaction such as the release of catecholamines, glucagon, and cortisol may be effectively reduced, thus possibly decreasing the SHG level.8 Stress hyperglycemia is the body’s stress response when a stimulus is greater than the body can tolerate; SHG is thereby generated by the change in homeostasis and neuroendocrine derangements.8 During the response, high levels of inflammatory cytokines such as TNF-α and IL-6 are produced simultaneously, which results in injury to vessel endothelial cells, vessel structure, and the blood-brain barrier and leads to hemorrhagic infarct conversion.10 In contrast, the hyperglycemia in DM mainly results from a combination of insulin resistance and/or β-cell secretory defects1 and may take years to develop. Glycemic control in DM is mainly through stimulating the release of insulin or increasing the sensitivity of insulin.5

A Protective Role for GV

Recently, the GV that refers to fluctuations in glucose values25 has been proposed to be implicated in the disease-associated process of dysglycemia and a marker of glycemic control.29 Several studies have demonstrated that increased GV is negatively associated with a poor prognosis and an early death in critically ill patients9,19 and in patients with congestive heart failure.7 We found GV to be an independent risk factor for a poor prognosis in acute-onset ICH, especially in patients with DM. Furthermore, after adjusting for other related variables, both our GluSD and GluCV models, as compared to our original model, specifically predicted a poor prognosis with no significant difference between the GluSD and GluCV models. This finding did not agree with the results of Mendez et al., who reported that the SD of glucose was a better metric than CV.22

The ICH patients with DM and greater GV had wide fluctuations in glucose levels and poor glycemic control. Although we did not find DM to be an independent risk factor for a poor prognosis in ICH, the DM patients with greater GV tended to have a poor outcome, which means the stress reaction also existed in the patients. However, distinguishing between SHG and DM is difficult in the clinic; thus, GV may be a valuable marker to reflect the stress reaction from severe primary disease and excessive inflammation. Therefore, GV could be a valuable marker for effective glycemic control and outcome in ICH patients with DM. It is important to improve the prognosis in ICH patients with DM by strengthening the management of glucose. Glycemic variability is still not a readily available or well-used metric in ICH in the clinic,21 and more comprehensive research on this factor is needed.

Study Advantages

There are certain advantages in our work. First, by observing ICH patients in regular hospital units rather than the intensive care unit, we provided a novel view of the implication of GV for these specific patients. Second, by relying on both medical history and an examination of the HbAlc level for nondiabetic patients, rather than basing our observations only on medical history, we could determine whether hyperglycemia was associated with DM or SHG. Third, we collected FBG data at admission and with 7-day continuous monitoring in order to observe dysglycemia immediately, thereby avoiding influential factors of food metabolism. Fourth, we analyzed admission FBG level and GV over 7 days of continuous monitoring as well as other factors as they related to functional outcome in ICH through multivariate analysis. We eliminated the influence of other risk factors to better understand the relationship between glycemic index and prognosis. In brief, the study provided an improved understanding of hyperglycemia and GV in ICH that could help to elucidate the pathophysiology of ICH and lead to more effective management.

Study Limitations

Several limitations are acknowledged. First, we did not collect data on factors such as nutritional status, fluids, and time of administration of insulin, which may affect the assessed GV. Second, we did not attempt to answer whether early insulin treatment for hyperglycemia could be beneficial because our patients were not observing strict glycemic control. Third, we did not evaluate other glycemic indexes of GV such as mean amplitude of glycemic excursion and mean absolute glucose rate of change, which may indicate that our results are unilateral. Further studies should consider other glycemic indexes of GV.

Conclusions

Stress hyperglycemia may be associated with increased mortality and a poor outcome in ICH, and increased GV may be independently associated with a poor outcome, particularly in ICH patients with DM. Both GluSD and GluCV improved the prediction of 30-day functional outcome in acute ICH.

Acknowledgments

This study was supported by grants from the Department of Education, Guangdong Government under the Top-tier University Development Scheme for Research and Control of Infectious Diseases (Grant Nos. 2015022 and 2015023) and the Science and Technology Plans of Shenzhen (Grant Nos. JCYJ201504033000426 and JCYJ20150403150555632). We thank Laura Smales (BioMedEditing, Toronto, Canada) for English language editing.

Disclosures

The authors report no conflict of interest concerning the materials and methods used in this study or the finding specified in this paper.

Author Contributions

Conception and design: QY Zhang. Acquisition of data: Y Wu, Wang. Analysis and interpretation of data: Y Wu, Ding. Drafting the article: QY Zhang, Y Wu. Critically revising the article: QY Zhang, Y Wu. Reviewed submitted version of manuscript: QY Zhang, Y Wu. Approved the final version of the manuscript on behalf of all authors: QY Zhang. Statistical analysis: Ding, Wen, Dong. Administrative/technical/material support: QY Zhang, J Wu, SC Zhang. Study supervision: QY Zhang.

References

  • 1

    American Diabetes Association: (2) Classification and diagnosis of diabetes. Diabetes Care 38 (Suppl):S8S162015

  • 2

    Appelboom GPiazza MAHwang BYCarpenter ABruce SSMayer S: Severity of intraventricular extension correlates with level of admission glucose after intracerebral hemorrhage. Stroke 42:188318882011

  • 3

    Béjot YAboa-Eboulé CHervieu MJacquin AOsseby GVRouaud O: The deleterious effect of admission hyperglycemia on survival and functional outcome in patients with intracerebral hemorrhage. Stroke 43:2432452012

  • 4

    Bloch RF: Interobserver agreement for the assessment of handicap in stroke patients. Stroke 19:14481988

  • 5

    Brownlee M: The pathobiology of diabetic complications: a unifying mechanism. Diabetes 54:161516252005

  • 6

    Clement SBraithwaite SSMagee MFAhmann ASmith EPSchafer RG: Management of diabetes and hyperglycemia in hospitals. Diabetes Care 27:5535912004 (Erratum in Diabetes Care)

  • 7

    Dungan KMBinkley PNagaraja HNSchuster DOsei K: The effect of glycaemic control and glycaemic variability on mortality in patients hospitalized with congestive heart failure. Diabetes Metab Res Rev 27:85932011

  • 8

    Dungan KMBraithwaite SSPreiser JC: Stress hyperglycaemia. Lancet 373:179818072009

  • 9

    Egi MBellomo RStachowski EFrench CJHart G: Variability of blood glucose concentration and short-term mortality in critically ill patients. Anesthesiology 105:2442522006

  • 10

    Esposito KNappo FMarfella RGiugliano GGiugliano FCiotola M: Inflammatory cytokine concentrations are acutely increased by hyperglycemia in humans: role of oxidative stress. Circulation 106:206720722002

  • 11

    Feigin VLForouzanfar MHKrishnamurthi RMensah GAConnor MBennett DA: Global and regional burden of stroke during 1990–2010: findings from the Global Burden of Disease Study 2010. Lancet 383:2452542014

  • 12

    Fogelholm RMurros KRissanen AAvikainen S: Admission blood glucose and short term survival in primary intracerebral haemorrhage: a population based study. J Neurol Neurosurg Psychiatry 76:3493532005

  • 13

    Freire AXBridges LUmpierrez GEKuhl DKitabchi AE: Admission hyperglycemia and other risk factors as predictors of hospital mortality in a medical ICU population. Chest 128:310931162005

  • 14

    Godoy DAPiñero GRSvampa SPapa FDi Napoli M: Hyperglycemia and short-term outcome in patients with spontaneous intracerebral hemorrhage. Neurocrit Care 9:2172292008

  • 15

    Han JLee HKCho TGMoon JGKim CH: Management and outcome of spontaneous cerebellar hemorrhage. J Cerebrovasc Endovasc Neurosurg 17:1851932015

  • 16

    Hanley JAMcNeil BJ: The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 143:29361982

  • 17

    Hemphill JC IIIGreenberg SMAnderson CSBecker KBendok BRCushman M: Guidelines for the management of spontaneous intracerebral hemorrhage: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 46:203220602015

  • 18

    Kothari RUBrott TBroderick JPBarsan WGSauerbeck LRZuccarello M: The ABCs of measuring intracerebral hemorrhage volumes. Stroke 27:130413051996

  • 19

    Krinsley JS: Glycemic variability: a strong independent predictor of mortality in critically ill patients. Crit Care Med 36:300830132008

  • 20

    Lovelock CEMolyneux AJRothwell PM: Change in incidence and aetiology of intracerebral haemorrhage in Oxfordshire, UK, between 1981 and 2006: a population-based study. Lancet Neurol 6:4874932007

  • 21

    Martin WGGalligan JSimpson S JrGreenaway TBurgess J: Admission blood glucose predicts mortality and length of stay in patients admitted through the emergency department. Intern Med J 45:9169242015

  • 22

    Mendez CEMok KTAta ATanenberg RJCalles-Escandon JUmpierrez GE: Increased glycemic variability is independently associated with length of stay and mortality in noncritically ill hospitalized patients. Diabetes Care 36:409140972013

  • 23

    Meretoja AStrbian DPutaala JCurtze SHaapaniemi EMustanoja S: SMASH-U: a proposal for etiologic classification of intracerebral hemorrhage. Stroke 43:259225972012

  • 24

    Monnier LColette COwens DR: Glycemic variability: the third component of the dysglycemia in diabetes. Is it important? How to measure it? J Diabetes Sci Technol 2:109411002008

  • 25

    Monnier LMas EGinet CMichel FVillon LCristol JP: Activation of oxidative stress by acute glucose fluctuations compared with sustained chronic hyperglycemia in patients with type 2 diabetes. JAMA 295:168116872006

  • 26

    Palm FHenschke NWolf JZimmer KSafer ASchröder RJ: Intracerebral haemorrhage in a population-based stroke registry (LuSSt): incidence, aetiology, functional outcome and mortality. J Neurol 260:254125502013

  • 27

    Piironen KPutaala JRosso CSamson Y: Glucose and acute stroke: evidence for an interlude. Stroke 43:8989022012

  • 28

    Rodbard D: Characterizing accuracy and precision of glucose sensors and meters. J Diabetes Sci Technol 8:9809852014

  • 29

    Rodbard D: New and improved methods to characterize glycemic variability using continuous glucose monitoring. Diabetes Technol Ther 11:5515652009

  • 30

    Sacco RLKasner SEBroderick JPCaplan LRConnors JJCulebras A: An updated definition of stroke for the 21st century: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 44:206420892013

  • 31

    Siegelaar SEHolleman FHoekstra JBDeVries JH: Glucose variability; does it matter? Endocr Rev 31:1711822010

  • 32

    Stead LGJain ABellolio MFOdufuye AGilmore RMRabinstein A: Emergency Department hyperglycemia as a predictor of early mortality and worse functional outcome after intracerebral hemorrhage. Neurocrit Care 13:67742010

  • 33

    Tapia-Pérez JHGehring SZilke RSchneider T: Effect of increased glucose levels on short-term outcome in hypertensive spontaneous intracerebral hemorrhage. Clin Neurol Neurosurg 118:37432014

  • 34

    Teasdale GJennett B: Assessment of coma and impaired consciousness. A practical scale. Lancet 2:81841974

  • 35

    Tetri SJuvela SSaloheimo PPyhtinen JHillbom M: Hypertension and diabetes as predictors of early death after spontaneous intracerebral hemorrhage. J Neurosurg 110:4114172009

  • 36

    Wang J: Preclinical and clinical research on inflammation after intracerebral hemorrhage. Prog Neurobiol 92:4634772010

  • 37

    Wexler DJNathan DMGrant RWRegan SVan Leuvan ALCagliero E: Prevalence of elevated hemoglobin A1c among patients admitted to the hospital without a diagnosis of diabetes. J Clin Endocrinol Metab 93:423842442008

  • 38

    Whitworth JA: 2003 World Health Organization (WHO)/International Society of Hypertension (ISH) statement on management of hypertension. J Hypertens 21:198319922003

  • 39

    Wilson JTHareendran AGrant MBaird TSchulz UGMuir KW: Improving the assessment of outcomes in stroke: use of a structured interview to assign grades on the modified Rankin Scale. Stroke 33:224322462002

  • 40

    Yang GWang YZeng YGao GFLiang XZhou M: Rapid health transition in China, 1990–2010: findings from the Global Burden of Disease Study 2010. Lancet 381:198720152013

  • 41

    Zhou YWang YWang JAnne Stetler RYang QW: Inflammation in intracerebral hemorrhage: from mechanisms to clinical translation. Prog Neurobiol 115:25442014

Article Information

Correspondence Qing-Ying Zhang, Department of Preventive Medicine, Shantou University Medical College, 22 Xinling Rd., Shantou, Guangdong 515041, China. email: qyzhang@stu.edu.cn.

INCLUDE WHEN CITING Published online November 3, 2017; DOI: 10.3171/2017.4.JNS162238.

Disclosures The authors report no conflict of interest concerning the materials and methods used in this study or the finding specified in this paper.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Dynamic changes in 7-day FBG level in 3 glucose groups of hospitalized patients with ICH. Figure is available in color online only.

  • View in gallery

    Poor prognosis and mortality rate at 30 days in 3 glucose groups of hospitalized ICH patients.

  • View in gallery

    Receiver operating characteristic curves for the original, GluSD (SD), and GluCV (CV) models for predicting 30-day poor outcome in ICH. Figure is available in color online only.

References

1

American Diabetes Association: (2) Classification and diagnosis of diabetes. Diabetes Care 38 (Suppl):S8S162015

2

Appelboom GPiazza MAHwang BYCarpenter ABruce SSMayer S: Severity of intraventricular extension correlates with level of admission glucose after intracerebral hemorrhage. Stroke 42:188318882011

3

Béjot YAboa-Eboulé CHervieu MJacquin AOsseby GVRouaud O: The deleterious effect of admission hyperglycemia on survival and functional outcome in patients with intracerebral hemorrhage. Stroke 43:2432452012

4

Bloch RF: Interobserver agreement for the assessment of handicap in stroke patients. Stroke 19:14481988

5

Brownlee M: The pathobiology of diabetic complications: a unifying mechanism. Diabetes 54:161516252005

6

Clement SBraithwaite SSMagee MFAhmann ASmith EPSchafer RG: Management of diabetes and hyperglycemia in hospitals. Diabetes Care 27:5535912004 (Erratum in Diabetes Care)

7

Dungan KMBinkley PNagaraja HNSchuster DOsei K: The effect of glycaemic control and glycaemic variability on mortality in patients hospitalized with congestive heart failure. Diabetes Metab Res Rev 27:85932011

8

Dungan KMBraithwaite SSPreiser JC: Stress hyperglycaemia. Lancet 373:179818072009

9

Egi MBellomo RStachowski EFrench CJHart G: Variability of blood glucose concentration and short-term mortality in critically ill patients. Anesthesiology 105:2442522006

10

Esposito KNappo FMarfella RGiugliano GGiugliano FCiotola M: Inflammatory cytokine concentrations are acutely increased by hyperglycemia in humans: role of oxidative stress. Circulation 106:206720722002

11

Feigin VLForouzanfar MHKrishnamurthi RMensah GAConnor MBennett DA: Global and regional burden of stroke during 1990–2010: findings from the Global Burden of Disease Study 2010. Lancet 383:2452542014

12

Fogelholm RMurros KRissanen AAvikainen S: Admission blood glucose and short term survival in primary intracerebral haemorrhage: a population based study. J Neurol Neurosurg Psychiatry 76:3493532005

13

Freire AXBridges LUmpierrez GEKuhl DKitabchi AE: Admission hyperglycemia and other risk factors as predictors of hospital mortality in a medical ICU population. Chest 128:310931162005

14

Godoy DAPiñero GRSvampa SPapa FDi Napoli M: Hyperglycemia and short-term outcome in patients with spontaneous intracerebral hemorrhage. Neurocrit Care 9:2172292008

15

Han JLee HKCho TGMoon JGKim CH: Management and outcome of spontaneous cerebellar hemorrhage. J Cerebrovasc Endovasc Neurosurg 17:1851932015

16

Hanley JAMcNeil BJ: The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 143:29361982

17

Hemphill JC IIIGreenberg SMAnderson CSBecker KBendok BRCushman M: Guidelines for the management of spontaneous intracerebral hemorrhage: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 46:203220602015

18

Kothari RUBrott TBroderick JPBarsan WGSauerbeck LRZuccarello M: The ABCs of measuring intracerebral hemorrhage volumes. Stroke 27:130413051996

19

Krinsley JS: Glycemic variability: a strong independent predictor of mortality in critically ill patients. Crit Care Med 36:300830132008

20

Lovelock CEMolyneux AJRothwell PM: Change in incidence and aetiology of intracerebral haemorrhage in Oxfordshire, UK, between 1981 and 2006: a population-based study. Lancet Neurol 6:4874932007

21

Martin WGGalligan JSimpson S JrGreenaway TBurgess J: Admission blood glucose predicts mortality and length of stay in patients admitted through the emergency department. Intern Med J 45:9169242015

22

Mendez CEMok KTAta ATanenberg RJCalles-Escandon JUmpierrez GE: Increased glycemic variability is independently associated with length of stay and mortality in noncritically ill hospitalized patients. Diabetes Care 36:409140972013

23

Meretoja AStrbian DPutaala JCurtze SHaapaniemi EMustanoja S: SMASH-U: a proposal for etiologic classification of intracerebral hemorrhage. Stroke 43:259225972012

24

Monnier LColette COwens DR: Glycemic variability: the third component of the dysglycemia in diabetes. Is it important? How to measure it? J Diabetes Sci Technol 2:109411002008

25

Monnier LMas EGinet CMichel FVillon LCristol JP: Activation of oxidative stress by acute glucose fluctuations compared with sustained chronic hyperglycemia in patients with type 2 diabetes. JAMA 295:168116872006

26

Palm FHenschke NWolf JZimmer KSafer ASchröder RJ: Intracerebral haemorrhage in a population-based stroke registry (LuSSt): incidence, aetiology, functional outcome and mortality. J Neurol 260:254125502013

27

Piironen KPutaala JRosso CSamson Y: Glucose and acute stroke: evidence for an interlude. Stroke 43:8989022012

28

Rodbard D: Characterizing accuracy and precision of glucose sensors and meters. J Diabetes Sci Technol 8:9809852014

29

Rodbard D: New and improved methods to characterize glycemic variability using continuous glucose monitoring. Diabetes Technol Ther 11:5515652009

30

Sacco RLKasner SEBroderick JPCaplan LRConnors JJCulebras A: An updated definition of stroke for the 21st century: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 44:206420892013

31

Siegelaar SEHolleman FHoekstra JBDeVries JH: Glucose variability; does it matter? Endocr Rev 31:1711822010

32

Stead LGJain ABellolio MFOdufuye AGilmore RMRabinstein A: Emergency Department hyperglycemia as a predictor of early mortality and worse functional outcome after intracerebral hemorrhage. Neurocrit Care 13:67742010

33

Tapia-Pérez JHGehring SZilke RSchneider T: Effect of increased glucose levels on short-term outcome in hypertensive spontaneous intracerebral hemorrhage. Clin Neurol Neurosurg 118:37432014

34

Teasdale GJennett B: Assessment of coma and impaired consciousness. A practical scale. Lancet 2:81841974

35

Tetri SJuvela SSaloheimo PPyhtinen JHillbom M: Hypertension and diabetes as predictors of early death after spontaneous intracerebral hemorrhage. J Neurosurg 110:4114172009

36

Wang J: Preclinical and clinical research on inflammation after intracerebral hemorrhage. Prog Neurobiol 92:4634772010

37

Wexler DJNathan DMGrant RWRegan SVan Leuvan ALCagliero E: Prevalence of elevated hemoglobin A1c among patients admitted to the hospital without a diagnosis of diabetes. J Clin Endocrinol Metab 93:423842442008

38

Whitworth JA: 2003 World Health Organization (WHO)/International Society of Hypertension (ISH) statement on management of hypertension. J Hypertens 21:198319922003

39

Wilson JTHareendran AGrant MBaird TSchulz UGMuir KW: Improving the assessment of outcomes in stroke: use of a structured interview to assign grades on the modified Rankin Scale. Stroke 33:224322462002

40

Yang GWang YZeng YGao GFLiang XZhou M: Rapid health transition in China, 1990–2010: findings from the Global Burden of Disease Study 2010. Lancet 381:198720152013

41

Zhou YWang YWang JAnne Stetler RYang QW: Inflammation in intracerebral hemorrhage: from mechanisms to clinical translation. Prog Neurobiol 115:25442014

TrendMD

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 22 22 22
Full Text Views 414 414 206
PDF Downloads 320 320 167
EPUB Downloads 0 0 0

PubMed

Google Scholar