44 Copyright © 2012 Korean Neurotraumatology Society

CLINICAL ARTICLEKorean J Neurotrauma 2012;8:44-47 ISSN 2234-8999

Introduction

Following acute traumatic brain injury, stress induced catecholamine release can result in elevated blood glucose levels. Hyperglycemia has been considered a stress re-sponse and a simple reflection of the injury severity.10) In animal studies of ischemic and hemorrhagic stroke, hyper-glycemia has been associated with increased ischemic brain damage, edema, and peri-lesional cell death. Early hyper-glycemia in rats with traumatic brain injury (TBI) was as-

sociated with a larger contusion area and increased inflam-mation.12,13) Current guidelines for acute ischemic stroke suggest that interventions are indicated if a glucose level is greater than 180 mg/dL.1) Since a landmark trial which de-monstrated a significant decrease in mortality using in-tensive insulin protocol targeting blood glucose levels be-tween 80 and 110 mg/dL in surgical intensive care patients, tight glucose control has been widely adopted in the critical setting.14)

Traditionally, the authors have collaborated with endocri-nologists for the control of hyperglycemia in neurosurgical intensive care unit (NSICU), especially in cases with more than 200 mg/dL of blood glucose level. The different treat-ments affect the clinical outcomes of patients with severe TBI. Therefore, we focus on the craniectomy, a uniform sur-gical intervention that is widely performed and useful for the treatment of TBI. Here, we have evaluated the feasibili-ty of intensive insulin therapy for patients with severe TBI who underwent decompressive craniectomy.

Received: February 14, 2012 / Revised: March 9, 2012Accepted: March 10, 2012Address for correspondence: Seung-Ho Yang, MDDepartment of Neurosurgery, St. Vincent’s Hospital, College of Me-dicine, The Catholic University of Korea, 93 Jungbu-daero, Pal-dal-gu, Suwon 442-723, KoreaTel: +82-31-249-8304, Fax: +82-31-245-5208E-mail: 72ysh@catholic.ac.krThe authors wish to acknowledge the financial support of the Catho-lic Medical Center Research Foundation made in the program year of

2012 Research Fund from the St. Vincent’s Hospital in Suwon, Korea.

Intensive Insulin Therapy after Decompression Surgery for Severe Traumatic Brain Injury

Moon Seok Kim, MD1, Seung Whan Lee, MD2, Seung-Ho Yang, MD1, Jae Taek Hong, MD1, Jae Hoon Sung, MD2, Byung Chul Son, MD1 and Sang Won Lee, MD1

1Departments of Neurosurgery, 2Internal Medicine, St. Vincent’s Hospital, College of Medicine, The Catholic University of Korea,Suwon, Korea

Objective: The purpose of this study was to compare intensive insulin therapy and conventional therapy in terms of venti-lator days, neurosurgical intensive care unit (NSICU) stay, Glasgow Outcome Scale (GOS), and complications for patients with a severe traumatic brain injury (TBI) who underwent decompressive craniectomy. Methods: Patients who had a TBI and a Glasgow Coma Scale (GCS) score ≤8, and who had been treated with a unilateral or bilateral decompressive craniectomy were enrolled. Twenty-three patients were treated with intensive insulin therapy tar-geting 80-120 mg/dL of blood glucose level. For comparison, 17 patients with conventional insulin therapy (＜200 mg/dL) were extracted from the historical data. Results: There was no statistically significant difference in terms of sex, age, GCS at admission, diagnosis of TBI, and histo-ry of diabetes. There was no statistically significant difference between the conventional and intensive groups with respect to total days of mechanical ventilation, NSICU days, GOS, and pneumonia. Hypoglycemic episodes developed more fre-quently in the intensive insulin therapy group than in the conventional therapy group. Conclusion: Intensive insulin therapy with our protocol cannot be recommended over conventional therapy in patients with severe TBI. (Korean J Neurotrauma 2012;8:44-47)

KEYWORDS: Hyperglycemia ㆍTraumatic brain injury ㆍIntensive insulin therapy ㆍDecompressive craniectomy.

online © ML Comm

www.neurotrauma.or.kr 45

Moon Seok Kim, et al.

Materials and Methods

Patients eligible for enrollment were patients who had a TBI and Glasgow Coma Scale (GCS) score ≤8, and who were treated with decompressive craniectomy unilaterally and/or bilaterally. Between January 2007 and December 2008, 23 patients were enrolled for the study and treated with in-tensive insulin therapy. Blood glucose levels were obtained via fingerstick (capillary) testing every 1 hr initially and re-duced to every 2 hr as levels stabilized. A continuous insu-lin infusion (250 units regular insulin in 500 milliliters of 0.9% sodium chloride solution) was delivered via a central or peripheral venous catheter and titrated for a goal blood sugar of 80 to 120 mg/dL (Table 1). Patients deemed to be approaching death based on the following criteria were ex-cluded: GCS of 3 without sedation; systolic blood pressure ＜50 mmHg or heart rate ＜40 despite vasoactive drugs;

oxygen saturation of less than 90% despite FiO2 of 100% oxygenation. Blood glucose levels were measured at ad-mission to NSICU and at least every 2 hr thereafter during intensive care unit (ICU) stay. Intensive insulin therapy continued throughout the study. Data on demographic and clinical characteristics were obtained at baseline.

Historic control (n=17) was extracted from medical re-cords and radiological reviews of TBI patients (GCS ≤8) who had undergone decompressive craniectomy, between 2004 and 2006. The blood glucose was measured every 6 hr via fingerstick (capillary) testing. Treatment was initiat-ed with a goal blood sugar of less than 200 mg/dL, when-ever the blood sugar level was greater than 200 mg/dL. Re-gular insulin (RI) was administered subcutaneously with 4 units of RI for a blood sugar of 201 to 249 mg/dL and in-creased stepwise by 4 units for every 50 mg/dL higher ran-ge, thereafter.

All patients had been fed an enteral diet via a nasogastric tube, as soon as possible after NSICU admission. When pa-tients were discharged from NSICU, conventional man-agement of blood sugar was performed by blood sugar measurements every 8-hr via fingerstick to target a blood sugar for less than 200 mg/dL.

For the statistical comparison, we performed Student’s t-test and chi-square test using the SPSS 12 software pack-age for Windows (SPSS, Inc., Chicago, IL, USA). A p-value of less than 0.05 was considered statistically significant.

Results

Patients’ characteristics are shown in Table 2. Twenty-three patients were treated with intensive insulin therapy following decompressive craniectomy for the treatment of severely traumatic brain injury. For the comparison be-tween the intensive insulin therapy and conventional con-trol (＜200 mg/dL of blood sugar), 17 patients with conven-tional insulin therapy were extracted from the historical data. There was no statistically significant difference in terms of sex, age, GCS at admission, diagnosis of TBI, and a history of diabetes.

Clinical outcomes are summarized in Table 3. There was

TABLE 1. Protocol for intensive insulin therapy

Blood glucose (mg/dL) Infusion rate (mL/hr)*451- 50351-450 12301-350 10251-300 8201-250 6151-200 4101-150 3

81-100 2

*mixture for infusion: 250 units regular insulin in 500 milliliters of 0.9% sodium chloride solution

TABLE 2. Demographic dataITT

groupHistoric control p-value

Number of patients 23 17Male : Female 15 : 8 11 : 6 1.00Age 53.8±18.4 47.8±21.3 0.403Glasgow coma scale 6.2± 1.3 6.1± 0.9 0.874Neurological diagnosis 0.373

Subdural hematoma 22 14Epidural hematoma 1 2Brain contusion 5 7

History of diabetes 3 3 1.00Mean±SD. ITT: intensive insulin therapy, NS: not significant

TABLE 3. Patient outcomes

ITT group Historic control p-valueVentilator days (number of patients with mechanical ventilation) 9.7± 5.6 (11) 6.3±3.3 (7) 0.162ICU days 20.0±12.8 20.3±8.1 0.971Glasgow outcome scale 3.5± 0.6 2.7±1.3 0.240Patients with pneumonia (%) 11 (47.8) 10 (58.8) 0.538Hypoglycemic episodes (＜60 mg/dL) 9 2 0.079Mean±SD. ITT: intensive insulin therapy, ICU: intensive care unit

46 Korean J Neurotrauma 2012;8:44-47

Intensive Insulin Therapy for Severe Traumatic Brain Injury

no statistically significant difference between the conven-tional and intensive groups with respect to the total days of mechanical ventilation, ICU days, Glasgow Outcome Scale (GOS) and pneumonia. Hypoglycemic episodes developed more frequently in the intensive insulin therapy group rath-er than in the conventional therapy group. Hypoglycemic events were accompanied by diaphoresis, arrhythmia, or increased respiratory rate and heart rate. The intensive in-sulin therapy was finished in 2 patients within 5 days after ICU admission, because of repeated episodes of hypogly-cemia. Fluid overloading developed in 2 patients during our protocol of intensive insulin therapy.

Discussion

In this study, we showed that in patients admitted to a N- SICU after decompressive craniectomy for the treatment of severe TBI, intensive insulin therapy titrated to maintain strict blood glucose control (blood glucose levels at 80-120 mg/dL) has no significant benefits on the clinical outcomes and increases the risk of hypoglycemia and fluid overload-ing.

Previous studies indicate that systemic hyperglycemia is harmful because it contributes to anaerobic metabolism in the brain following acute injury, resulting in brain tissue lactic acidosis and secondary neuronal injury.16,18) However, it is not determined to what degree glucose should be tar-geted in TBI patients to improve clinical outcomes. It is re-ported that the admission glucose ＞200 mg/dL was asso-ciated with poor outcomes.17) A glucose value of 200 mg/dL was chosen as the cut-off because it was accepted as a treat-able number in clinical practice. A different study demon-strated that patients with severe TBI had significantly high-er glucose levels compared to those with mild TBI and found that blood glucose levels ＞150 mg/dL on admission were associated with poorer neurologic outcomes.9) Additionally, it was observed that blood admission glucose levels ＞150 mg/dL at admission was associated with higher in-hospital mortality.15) On the other hand, it was reported that blood glucose levels ＞170 mg/dL were associated with poor out-comes.8) Hyperglycemia is transient following acute TBI, and is indicative of poor outcomes. It is suggested that ther-apeutic intervention for controlling hyperglycemia needs to be initiated within the first 24 h of admission.3,5) Total dosage of insulin administered for intensive insulin thera-py during the early period should be calculated to convert to intermittent insulin administration.

Hypoglycemic episodes were more frequent in the inten-sive insulin therapy group. A recent randomized trial that

examined intensive insulin therapy in critically ill patients was halted early due to a lack of efficacy and increased in-cidence of hypoglycemia.4) In neurological patients here may be more of a concern with having even low normal blood glucose levels and its impact on intracerebral glucose availability and metabolism and, ultimately, on neurologic recovery.2,11) In our series, the lack of benefit of intensive insulin therapy could be related with the higher risk of hy-poglycemia. Additionally, our protocol (Table 1) has the risk of fluid overloading for patients with intractable hy-perglycemia. Given these results, intensive insulin therapy with our protocol (targeting 80-120 mg/dL of blood glu-cose level) cannot be recommended over conventional ther-apy (targeting ＜200 mg/dL of blood glucose level).6,7) Cur-rently, intensive insulin therapy is performed in selected TBI patients whose hyperglycemia is not controlled with conventional therapy and have tolerable cardiopulmonary function.

Conclusion

There was no benefit to intensive insulin therapy for se-vere TBI patients in terms of ventilator days, ICU stay, GOS, and pneumonia according to the single institution’s experi-ence. Moreover, the risk of hypoglycemic episodes and flu-id overloading was increased.

■ The authors have no financial conflicts of interest.

REFERENCES1) Adams HP Jr, Adams RJ, Brott T, del Zoppo GJ, Furlan A, Gold-

stein LB, et al. Guidelines for the early management of patients with ischemic stroke: A scientific statement from the Stroke Coun-cil of the American Stroke Association. Stroke34:1056-1083, 2003

2) Bilotta F, Caramia R, Cernak I, Paoloni FP, Doronzio A, Cuzzone V, et al. Intensive insulin therapy after severe traumatic brain inju-ry: a randomized clinical trial. NeurocritCare9:159-166, 2008

3) Bilotta F, Giovannini F, Caramia R, Rosa G. Glycemia manage-ment in neurocritical care patients: a review. JNeurosurgAnes-thesiol21:2-9, 2009

4) Brunkhorst FM, Engel C, Bloos F, Meier-Hellmann A, Ragaller M, Weiler N, et al. Intensive insulin therapy and pentastarch re-suscitation in severe sepsis. NEnglJMed358:125-139, 2008

5) Godoy DA, Di Napoli M, Rabinstein AA. Treating hyperglycemia in neurocritical patients: benefits and perils. NeurocritCare13: 425-438, 2010

6) Graffagnino C, Gurram AR, Kolls B, Olson DM. Intensive insulin therapy in the neurocritical care setting is associated with poor clinical outcomes. NeurocritCare13:307-312, 2010

7) Green DM, O’Phelan KH, Bassin SL, Chang CW, Stern TS, Asai SM. Intensive versus conventional insulin therapy in critically ill neurologic patients. NeurocritCare13:299-306, 2010

8) Jeremitsky E, Omert LA, Dunham CM, Wilberger J, Rodriguez A. The impact of hyperglycemia on patients with severe brain in-

www.neurotrauma.or.kr 47

Moon Seok Kim, et al.

jury. JTrauma58:47-50, 20059) Lam AM, Winn HR, Cullen BF, Sundling N. Hyperglycemia and

neurological outcome in patients with head injury. JNeurosurg75:545-551, 1991

10)Liu-DeRyke X, Collingridge DS, Orme J, Roller D, Zurasky J, Rho-ney DH. Clinical impact of early hyperglycemia during acute phase of traumatic brain injury. NeurocritCare11:151-157, 2009

11) Oddo M, Schmidt JM, Carrera E, Badjatia N, Connolly ES, Presciutti M, et al. Impact of tight glycemic control on cerebral glucose metabolism after severe brain injury: a microdialysis study. CritCareMed36:3233-3238, 2008

12)Pulsinelli WA, Waldman S, Rawlinson D, Plum F. Moderate hy-perglycemia augments ischemic brain damage: a neuropathologic study in the rat. Neurology32:1239-1246, 1982

13)Song EC, Chu K, Jeong SW, Jung KH, Kim SH, Kim M, et al. Hy-perglycemia exacerbates brain edema and perihematomal cell death after intracerebral hemorrhage. Stroke34:2215-2220, 2003

14)van den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M, et al. Intensive insulin therapy in critically ill patients.

NEnglJMed345:1359-1367, 200115)Vespa P, Boonyaputthikul R, McArthur DL, Miller C, Etchepare

M, Bergsneider M, et al. Intensive insulin therapy reduces micro-dialysis glucose values without altering glucose utilization or im-proving the lactate/pyruvate ratio after traumatic brain injury. CritCareMed34:850-856, 2006

16)Welsh FA, Ginsberg MD, Rieder W, Budd WW. Deleterious effect of glucose pretreatment on recovery from diffuse cerebral isch-emia in the cat. II. Regional metabolite levels. Stroke11:355-363, 1980

17)Young B, Ott L, Dempsey R, Haack D, Tibbs P. Relationship be-tween admission hyperglycemia and neurologic outcome of se-verely brain-injured patients. AnnSurg210:466-472; discussion 472-473, 1989

18)Zygun DA, Steiner LA, Johnston AJ, Hutchinson PJ, Al-Rawi PG, Chatfield D, et al. Hyperglycemia and brain tissue pH after trau-matic brain injury. Neurosurgery55:877-881; discussion 882, 2004