Effect of Intensive Glucose Control on Outcomes of Hyperglycemic Stroke Patients Receiving Mechanical Thrombectomy: Secondary Analysis of the SHINE Trial : Journal of Neurosurgical Anesthesiology

Secondary Logo

Journal Logo

Short Reports

Effect of Intensive Glucose Control on Outcomes of Hyperglycemic Stroke Patients Receiving Mechanical Thrombectomy: Secondary Analysis of the SHINE Trial

Asaithambi, Ganesh MD*; Tipps, Megan E. PhD

Author Information
Journal of Neurosurgical Anesthesiology: October 2022 - Volume 34 - Issue 4 - p 415-418
doi: 10.1097/ANA.0000000000000795
  • Free
  • Infographic


Hyperglycemia is present in 40% to 50% of patients with acute ischemic stroke (AIS) and is associated with poor clinical outcomes.1–3 This can occur not only among diabetic patients but also among those without diabetes due to stress-mediated release of cortisol and norepinephrine, which can lead to an increase in inflammation.2,4 Hyperglycemia exacerbates brain injury via enhancement of acidosis in the ischemic penumbra.4,5 This in turn can result in poor clinical outcomes among AIS patients due to worsening cerebral edema and hemorrhagic transformation of infarcted tissue.1,2,4,5

Acute phase hyperglycemia independently increases the risk of death and symptomatic intracranial hemorrhage (sICH) among AIS patients treated with intravenous thrombolysis.2,4 Admission hyperglycemia has been associated with unfavorable outcomes and infarct growth despite successful reperfusion among AIS patients harboring intracranial large vessel occlusions (LVO) undergoing mechanical thrombectomy (MT).2,4–6 What remains unclear is whether intensive blood glucose reduction can be associated with outcomes among AIS patients presenting with hyperglycemia and who are treated with MT. By examining patient-level data from a randomized clinical trial, we hypothesize that intensive glucose control among LVO AIS patients will improve rates of long-term favorable functional independence and decrease rates of death.


In this nonpreplanned secondary analysis, we accessed publicly available data from the Stroke Hyperglycemia Insulin Network Effort (SHINE) trial,1 which was a multicenter, randomized clinical trial with blinded outcome assessment comparing AIS patients presenting with hyperglycemia receiving intensive treatment by continuous insulin infusion (to achieve target blood glucose concentrations of 80 to 130 mg/dL) to those receiving standard treatment with sliding scale insulin subcutaneously (to achieve blood glucose concentrations of 80 to 179 mg/dL) up to 72 hours after onset of stroke symptoms. The SHINE trial (principal investigator Karen C. Johnston, MD; NCT01369069) data and analyses presented in this manuscript are based on fully deidentified research files downloaded from the Archived Clinical Research website of the National Institute of Neurological Disorders and Stroke (www.ninds.nih.gov/Current-Research/Research-Funded-NINDS/Clinical-Research/Archived-Clinical-Research-Datasets). The SHINE trial was funded by the National Institute of Neurological Disorders and Stroke, and the trial protocol was approved by the institutional review board at each participating site. All randomized patients provided informed consent to participate in the trial. Clinical trial design and methodology for SHINE are available in a previous publication.1 Our secondary analysis of data from the SHINE trial was exempted from IRB approval at our local institution (Allina Health) since the dataset was completely deidentified.

All enrolled patients were randomized within 12 hours of AIS onset between April 2012 and August 2018. Hyperglycemia was defined as capillary glucose concentration >110 mg/dL if patients had diabetes or ≥150 mg/dL in patients without history of diabetes. The trial included 1151 patients and the primary outcome studied was the proportion of patients with favorable outcome based on 90-day modified Rankin Scale (mRS) scores. Patients randomized to the intensive treatment group received continuous intravenous insulin infusions with the assistance of a computer decision support tool (GlucoStabilizer, Medical Decision Network, LLC; Charlottesville, VA), which recommends the insulin infusion rate to maintain glucose values within the target range. Patients randomized to the standard treatment group received sliding scale subcutaneous insulin every 6 hours as needed to maintain glucose values within the target range. For every 3 hours, the median number of glucose measurements was 3 for patients in the intensive treatment arm and 1 for patients in the standard treatment arm.1

We studied SHINE trial participants who were treated with MT to determine the effect of intensive blood glucose control on outcomes among those presenting with hyperglycemia; patients not treated with MT were excluded from our analysis. Demographic information collected included age, sex, and race/ethnicity. Baseline clinical characteristics collected included stroke severity upon presentation as defined by the National Institutes of Health Stroke Scale (NIHSS) score, admission glucose concentrations, intravenous thrombolysis utilization, and history of hypertension, diabetes, hyperlipidemia, coronary artery disease, atrial fibrillation, prior ischemic stroke, and heart failure. Primary outcomes of interest included favorable 90-day outcomes, defined as mRS score ≤2, and rates of death. Secondary outcomes were the distribution of mRS scores at 90-day among survivors, and rates of severe hypoglycemic events (defined as glucose concentration<40 mg/dL).

Statistical analyses were performed using SPSS software (version 26; IBM, Armonk, NY). Descriptive data are presented as counts with frequencies (percentages), median values with range, or mean values with SD. Data were analyzed using independent t tests for continuous data or χ2 tests for categorical data. A P<0.05 was considered significant for all tests.


In this post hoc analysis of SHINE trial data, we identified 146 AIS patients treated with MT. Of these, 74 patients (mean age 68.1±12.2 y, 48.6% women) were randomized to the intensive treatment group and 72 patients (mean age 68±15.5 y, 50% women) were randomized to the standard treatment group. Baseline and clinical characteristics of 2 groups are shown in Table 1. The mean admission glucose levels were similar between groups (intensive group 204.1±71.1 mg/dL vs. standard group 192.2±62.2 mg/dL; P=0.29). A higher percentage of patients in the standard treatment group had a history of prior ischemic strokes when compared with the intensive treatment group (20.8% vs. 8.1%, respectively; P=0.035). Stroke severity upon admission was similar; intensive group NIHSS 15 (range: 3 to 22) versus standard group NIHSS 16 (range: 3 to 22; P=0.81). The percentage of patients receiving intravenous thrombolysis was also similar between groups (68.9% vs. 59.7% for intensive and standard groups, respectively; P=0.30).

TABLE 1 - Baseline and Clinical Characteristics of Stroke Hyperglycemia Insulin Network Effort Trial Participants Receiving Mechanical Thrombectomy
Treatment of Hyperglycemia
Intensive (n=74) Standard (n=72) P
Age (y) 68.1 (12.2) 68 (15.5) 0.96
Women 36 (49) 36 (50) 1.0
 Non-Hispanic White 39 (53) 41 (57) 0.62
 Non-Hispanic Black 20 (27) 16 (22) 0.57
 Hispanic 11 (15) 11 (15) 1.0
 Other 4 (5) 4 (6) 1.0
Medical history
 Hypertension 64 (87) 63 (88) 0.55
 Diabetes mellitus 54 (73) 53 (74) 0.58
 Hyperlipidemia 39 (53) 41 (57) 0.50
 Coronary artery disease 22 (49) 22 (47) 0.61
 Atrial fibrillation 24 (53) 24 (51) 0.61
 Prior ischemic stroke 6 (8) 15 (21) 0.035
 Heart failure 14 (31) 17 (36) 0.66
Mean admission blood glucose level (mg/dL) 204.1 (71) 192.2 (62) 0.29
National Institutes of Health Stroke Scale score 15 (3-22) 16 (3-22) 0.81
Intravenous thrombolysis 51 (69) 43 (60) 0.30
Data shown as mean (SD), number (percentage), or median (range).

At 90 days, the rates of favorable functional outcomes (mRS ≤2) among all patients were similar between the intensive and standard treatment groups (31.1% vs. 30.6%, respectively; P=1.0); odds ratio 1.025, 95% confidence interval 0.51 to 2.07 (Table 2). There was no difference in rates of death between intensive and standard treatment groups (20.3% vs. 22.2%, respectively; P=0.84); odds ratio 0.98, 95% confidence internal 0.40 to 1.97. Among survivors at 90 days, the distribution of mRS scores was also not different between treatment groups (intensive group mRS 0 to 2, 39% and mRS 3 to 5, 61% vs. standard group mRS 0 to 2, 39.2% and mRS 3 to 5, 60.7%; P=0.96). Severe hypoglycemic events only occurred in the intensive treatment group (2.7% vs. 0%; P=0.50).

TABLE 2 - Outcomes for Stroke Hyperglycemia Insulin Network Effort Trial Participants Receiving Mechanical Thrombectomy Based on Treatment of Hyperglycemia
Intensive Standard P
 90 d mRS ≤2 (%) 23 (31) 22 (31) 1.0
  Odds ratio 1.025 (0.51-2.07) Ref
 Death 15 (20) 16 (22) 0.84
  Odds ratio 0.98 (0.40-1.97) Ref
 Severe hypoglycemia 2 (3) 0 0.50
 Distribution of 90 d mRS among survivors 0.96
  0 3 (5) 4 (7)
  1 14 (24) 11 (20)
  2 6 (10) 7 (13)
  3 14 (24) 12 (21)
  4 14 (24) 12 (21)
  5 8 (14) 10 (18)
Data presented as number (percentage) or odds ratio (95% confidence interval).
mRS indicates modified Rankin Scale score; Ref, reference.


Patients harboring LVO ischemic stroke syndromes who undergo successful reperfusion with MT are susceptible to higher rates of long-term unfavorable outcomes, sICH, and larger infarct volumes in the setting of hyperglycemia.2,4–6 In our nonpreplanned analysis of SHINE trial participants who received MT, we did not detect an association between intensive blood glucose control and rates of death or rates long-term favorable outcomes when compared with standard blood glucose control. The absence of clinically significant benefit occurred despite lower mean blood glucose levels in the intensive treatment group compared with the standard treatment group.1 Intensive blood glucose treatment also led to severe hypoglycemic events in patients, which did not occur for any patient randomized to the standard treatment group.

Participants of the SHINE trial were randomized within 12 hours of stroke onset with the intention to start treatment for hyperglycemia before substantial injury, which can occur hours to days after stroke onset.1 The median time from stroke onset to randomization in the SHINE trial was ~7 hours, suggesting that treatment was initiated early enough to potentially disrupt adverse cellular mechanisms.1 However, intensive treatment of blood glucose for up to 72 hours of stroke onset was not associated with a higher rate of favorable functional outcomes at 90 days when compared with standard treatment.

As we observe an increased trend in MT for AIS patients with guideline support,7,8 we aimed to study the importance of glucose control in this specific population. The SHINE trial included a heterogeneous group of patients, including ischemic stroke (treated with intravenous thrombolysis, MT, or intra-arterial drugs), transient ischemic attack, and stroke mimics (including migraine, seizure, and encephalopathy among others).1 For this reason, we aimed to study patients who are at risk for larger infarct volumes due to LVO in the setting of hyperglycemia. In a post hoc analysis of the DEFUSE 3 (Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke) trial, higher blood glucose values were found to be associated with core infarction growth despite successful reperfusion with MT.5 Others have reported that higher blood glucose levels reduced the likelihood of favorable outcomes among AIS despite good collateral circulation and found that each 10 mg/dL reduction in glucose values resulted in an estimated 20 additional AIS patients with favorable outcomes at 90 days.9 Despite these previous findings, our analysis of the SHINE trial data did not support an association between aggressive glucose control and favorable outcomes.

Previous studies have confirmed an association between hyperglycemia and sICH among AIS patients treated with intravenous thrombolysis.2,4,7 The higher rates of sICH among those treated with intravenous thrombolysis in the setting of hyperglycemia are thought to be related to an acceleration of downstream microvascular thromboinflammation, which impairs reperfusion and precipitates neurovascular damage and hemorrhagic transformation.3 However, and despite successful reperfusion, the association between hyperglycemia and sICH has been confirmed among AIS patients receiving MT.2,4,6

Our study has important limitations. It was a secondary analysis of the SHINE trial, which was not designed to answer the clinical question in the current study. We were unable to calculate rates of sICH, which may have contributed to unfavorable outcomes. Reperfusion data and imaging data measuring volumes of ischemia were also not available for review. Because the SHINE trial began to enroll patients in 2012 before landmark MT studies,10 only 12.6% of SHINE trial participants underwent MT and this resulted in a small sample size for our analysis. As a consequence of this small sample size, we cannot exclude an association between glucose control and outcomes from the wide confidence intervals noted in our study. Moreover, advancements in MT devices evolved during the study period and that could have contributed to reperfusion rates and long-term functional outcomes. Finally, the locations of LVO for which patients received MT are unknown.

In conclusion, intensive blood glucose control within the first 72 hours among hyperglycemic AIS patients who underwent MT was not associated with lower rates of death or higher rates of long-term favorable outcomes when compared with standard treatment. Our findings do not suggest an association between aggressive blood glucose control among patients undergoing MT and neuroprotection, but larger studies are needed to study the association between aggressive hyperglycemia treatment and clinical outcomes in this population.


1. Johnston KC, Bruno A, Pauls Q, et al. Intensive vs Standard Treatment of Hyperglycemia and Functional Outcome in Patients With Acute Ischemic Stroke: The SHINE Randomized Clinical Trial. JAMA. 2019;322:326–335. doi:10.1001/jama.2019.9346
2. Li F, Ren Y, Cui X, et al. Postoperative hyperglycemia predicts symptomatic intracranial hemorrhage after endovascular treatment in patients with acute anterior circulation large artery occlusion. J Neurol Sci. 2020;409:116588. doi:10.1016/j.jns.2019.116588
3. Desilles J, Syvannarath V, Ollivier V, et al. Exacerbation of thromboinflammation by hyperglycemia precipitates cerebral infarct growth and hemorrhagic transformation. Stroke. 2017;48:1932–1940. doi:10.1161/STROKEAHA.117.017080
4. Goyal N, Tsivgoulis G, Pandhi A, et al. Admission hyperglycemia and outcomes in large vessel occlusion strokes treated with mechanical thrombectomy. J Neurointerv Surg. 2018;10:112–117. doi:10.1136/neurintsurg-2017-012993
5. Yaghi S, Dehkharghani S, Raz E, et al. The effect of hyperglycemia on infarct growth after reperfusion: an analysis of the DEFUSE 3 trial. J Stroke Cerebrovasc Dis. 2021;1:105380. doi:10.1016/j.jstrokecerebrovasdis.2020.105380
6. Huo X, Liu R, Gao F, et al. Effect of hyperglycemia at presentation on outcomes in acute large artery occlusion patients treated with solitaire stent thrombectomy. Front Neurol. 2019;10:71. doi:10.3389/fneur.2019.00071
7. Asaithambi G, Tong X, Lakshminarayan K, et al. Current trends in the acute treatment of ischemic stroke: analysis from the Paul Coverdell National Acute Stroke Program. J Neurointerv Surg. 2020;12:574–578. doi:10.1136/neurintsurg-2019-015133
8. Powers WJ, Rabinstein AA, Ackerson T, et al. Guidelines for the Early Management of Patients With Acute Ischemic Stroke: 2019 Update to the 2018 Guidelines for the Early Management of Acute Ischemic Stroke: A Guideline for Healthcare Professions From the American Heart Association/American Stroke Association. Stroke. 2019;50:e344–e418. doi:10.1161/STR.0000000000000211
9. Kim J, Liebeskind DS, Jahan R, et al. Impact of hyperglycemia according to the collateral status on outcomes in mechanical thrombectomy. Stroke. 2018;49:2706–2714. doi:10.1161/STROKEAHA.118.022167
10. Goyal M, Menon BK, van Zwam WH, et al. Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016;387:1723–1731. doi:10.1016/S0140-6736(16)00163-X

hyperglycemia; mechanical thrombectomy; outcomes; stroke

Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.