CLINICAL STUDIES Matthew J. McGirt, M.D. Department of Neurosurgery, Johns Hopkins School of Medicine, and The Johns Hopkins Neuro-oncology Surgical Outcomes Research Laboratory, Baltimore, Maryland

Kaisorn L. Chaichana, B.S. Department of Neurosurgery, Johns Hopkins School of Medicine, and The Johns Hopkins Neuro-oncology Surgical Outcomes Research Laboratory, Baltimore, Maryland

Muraya Gathinji, M.S. Department of Neurosurgery, Johns Hopkins School of Medicine, and The Johns Hopkins Neuro-oncology Surgical Outcomes Research Laboratory, Baltimore, Maryland

Frank Attenello, M.S. Department of Neurosurgery, Johns Hopkins School of Medicine, and The Johns Hopkins Neuro-oncology Surgical Outcomes Research Laboratory, Baltimore, Maryland

Khoi Than, M.D. Department of Neurosurgery, Johns Hopkins School of Medicine, and The Johns Hopkins Neuro-oncology Surgical Outcomes Research Laboratory, Baltimore, Maryland

Amado Jimenez Ruiz, B.S. Department of Neurosurgery, Johns Hopkins School of Medicine, and The Johns Hopkins Neuro-oncology Surgical Outcomes Research Laboratory, Baltimore, Maryland

Alessandro Olivi, M.D. Department of Neurosurgery, Johns Hopkins School of Medicine, and The Johns Hopkins Neuro-oncology Surgical Outcomes Research Laboratory, Baltimore, Maryland

Alfredo Quiñones-Hinojosa, M.D. Department of Neurosurgery, Johns Hopkins School of Medicine, and The Johns Hopkins Neuro-oncology Surgical Outcomes Research Laboratory, Baltimore, Maryland Reprint requests: Alfredo Quiñones-Hinojosa, M.D., Department of Neurosurgery, Johns Hopkins Hospital, Johns Hopkins University, CRB II, 1550 Orleans Street, Room 247, Baltimore, MD 21231. Email: [email protected] Received, December 4, 2007.

PERSISTENT OUTPATIENT HYPERGLYCEMIA IS INDEPENDENTLY ASSOCIATED WITH DECREASED SURVIVAL AFTER PRIMARY RESECTION OF MALIGNANT BRAIN ASTROCYTOMAS OBJECTIVE: Patients with malignant brain astrocytomas are at high risk for developing hyperglycemia secondary to frequent corticosteroid administration. Several clinical studies have shown that hyperglycemia is associated with poor outcome in multiple disease states. Furthermore, hyperglycemia augments in vitro astrocytoma growth, whereas hypoglycemia attenuates in vitro astrocytoma cell growth. We hypothesized that persistent hyperglycemic states in the outpatient setting may serve as a prognostic marker of decreased survival in patients with malignant brain astrocytomas. METHODS: We retrospectively reviewed 367 cases of craniotomy for malignant brain astrocytomas (World Health Organization Grade III or IV). Persistent hyperglycemia was defined as serum glucose greater than 180 mg/dL occurring three or more times between 1 and 3 months postoperatively. Isolated hyperglycemia was defined as an isolated occurrence of serum glucose greater than 180 mg/dL. The independent association of outpatient glucose levels and recorded clinical and treatment variables with overall survival was assessed via multivariate proportional-hazards regression analysis. RESULTS: A total of 367 craniotomies (209 primary, 158 secondary) were performed for malignant brain astrocytomas (glioblastoma multiforme, 297; anaplastic astrocytomas, 70); 68 (19%) and 28 (8%) of the patients experienced isolated or persistent outpatient hyperglycemia, respectively. Patients experiencing persistent hyperglycemia were older (59 ⫾ 13 versus 51 ⫾ 14 yr), were diabetic more frequently (7 [25%] versus 10 [3%]), continued to receive corticosteroids more frequently (21 [75%] versus 35 [10%]); and received temozolomide less often (4 [14%] versus 116 [34%]). Adjusting for intergroup differences and variables associated with survival in this model, age (P ⫽ 0.001), Karnofsky Performance Scale score (P ⫽ 0.001), tumor grade (P ⫽ 0.001), primary versus secondary resection (P ⫽ 0.008), temozolomide (P ⫽ 0.007), subsequent resection (P ⫽ 0.07), and continued outpatient dexamethasone therapy, persistent outpatient hyperglycemia (relative risk, 1.79; 95% confidence interval, 1.05–3.05, P ⫽ 0.03) remained independently associated with decreased survival. Median survival for persistently hyperglycemic versus normal-glycemic cohorts was 5 and 11 months, respectively. CONCLUSION: In our experience, persistent outpatient hyperglycemia was associated with decreased survival in patients undergoing surgical resection for malignant astrocytomas and was independent of the degree of disability, tumor grade, diabetes, prolonged dexamethasone use, or subsequent treatment modalities. Increased glucose control is warranted in this patient population and may contribute to improved outcomes in the treatment of malignant brain astrocytomas. KEY WORDS: Glioblastoma multiforme, Glucose, Hyperglycemia, Malignant astrocytoma, Survival Neurosurgery 63:286–291, 2008

DOI: 10.1227/01.NEU.0000315282.61035.48

www.neurosurgery-online.com

ABBREVIATIONS: GBM, glioblastoma multiforme; KPS, Karnofsky Performance Scale; MRI, magnetic resonance imaging; WHO, World Health Organization

Accepted, March 5, 2008.

286 | VOLUME 63 | NUMBER 2 | AUGUST 2008

www.neurosurgery-online.com

HYPERGLYCEMIA AFTER ASTROCYTOMA RESECTION

G

lioblastoma multiforme (GBM) is the most common and aggressive type of primary brain tumor in adults. Despite advances in surgical technology, chemotherapy, and radiotherapy, the median survival period remains less than 1 year (1, 5, 7). In fact, Tait et al. (29) stated that the survival of patients with GBM has not significantly improved during the past 15 years. Although the survival time for patients with GBM, on average, is short, individual patient survival remains heterogeneous (29). As a result, there is an emphasis on the examination of factors that are prognostic of prolonged survival for patients with GBM, age and Karnofsky Performance Scale (KPS) score being the most significant (16). Interestingly, outpatient hyperglycemia may serve as a prognostic marker in patients with GBM. The authors of several clinical studies report that hyperglycemia is associated with poor outcomes in many disease states (13, 21, 30, 33). As early as 1924, Otto Warburg noted that tumor cells in culture relied preferentially on anaerobic glycolysis rather than on respiration for adenosine triphosphate generation (6, 32). In fact, glioma cell lines exhibit a threefold increase in glycolysis as compared with their normal astrocytic counterparts (20). More recently, Jelluma et al. (11) found that GBM cell lines undergo extensive apoptosis on glucose withdrawal, which is a response that is absent in normal human astrocytes. Although continuous hypoglycemia slows glioma growth, it remains largely unstudied, but plausible, that persistent hyperglycemia may preferentially augment tumor growth. We set out to determine whether persistent hyperglycemia after surgical resection of GBM is associated with decreased survival.

PATIENTS AND METHODS

tively to detect possible hematological complications of adjuvant therapy. The timing of laboratory evaluations was not standardized but typically occurred weekly during the 3-month follow-up period. Laboratory values were frequently available beyond 3 months after surgery; however, they were not recorded for the purposes of this study. For purposes of this study, only serum glucose was retrospectively recorded. Before the initiation of the study, we defined persistent hyperglycemia as a serum glucose value greater than 180 mg/dL occurring three or more times during the 1- to 3-month postoperative period (22). Isolated hyperglycemia was defined as an isolated occurrence of serum glucose greater than 180 mg/dL during this period (22). To determine whether outpatient hyperglycemia was a result of dexamethasone (Decadron; Merck & Co., Inc., Whitehouse Station, NJ) dependence caused by greater tumor burden, it was recorded whether the patient was unable to be weaned off of Decadron by 3 months postoperatively. To determine that hyperglycemia associations were independent of greater tumor burdens and prolonged Decadron use, we adjusted all analyses of hyperglycemia in the multivariate analysis for the need for persistent Decadron therapy.

Statistical Analysis Survival as a function of time after surgical resection was expressed as estimated Kaplan-Meier plots. Parametric data were expressed as mean ⫾ standard deviation. Nonparametric data were expressed as median (interquartile range). Percentages were compared with the χ2 test. Continuous variables were compared using the Student’s t test or the Mann-Whitney U test where appropriate. The independent association of outpatient glucose levels, all recorded variables, and survival was assessed via multivariate proportional-hazards regression analysis (Cox model). Variables associated with survival in univariate analysis (P ⬍ 0.1) were included in the multivariate Cox model. Variables demonstrating P values greater than 0.05 were then removed in a stepwise fashion.

RESULTS

Patient Population We retrospectively reviewed 367 cases of craniotomy for resection of malignant brain astrocytomas (World Health Organization [WHO] Grade III or IV astrocytomas) performed at a single institution from 1995 to 2006. Malignant oligodendrogliomas and mixed astrocytomaoligodendrogliomas were not included. Presenting clinical, radiological, operative, and hospital course records were retrospectively reviewed. Preoperative KPS score (4) and postoperative neurological status were recorded for all patients. Outpatient clinic notes were available from both neurosurgical and neuro-oncology follow-up visits and were reviewed in all cases. Demographics, presenting symptoms and signs of patients, degree of resection, perioperative morbidity, adjuvant radiotherapy and chemotherapy regimens, and date of death were recorded. Degree of resection was retrospectively classified on the basis of the neuroradiologist’s interpretation of the postoperative magnetic resonance imaging (MRI) scan obtained less than 48 hours after surgical resection. Degree of resection was defined as gross total resection if no residual nodular enhancement was noted on postoperative MRI or subtotal resection if residual nodular enhancement was noted on postoperative MRI. Tumor grade was histologically confirmed as WHO Grade III or IV in all cases. It was recorded whether patients underwent a secondary resection at a later date. Patients who were not confirmed as having died were classified as lost to follow-up at the time of the last clinic visit. Outpatient laboratory tests (complete blood count, basic metabolic profiles) were routinely obtained between 1 and 3 months postopera-

NEUROSURGERY

Patient Population Three hundred sixty-seven craniotomies performed between 1995 and 2006 were reviewed for this study. A total of 209 (57%) were for primary resection and 158 (43%) were for secondary resection. Tumor grade included WHO Grade IV astrocytomas in 297 cases (81%) and WHO Grade III astrocytomas in 70 cases (19%). The average age was 54 ⫾ 13 years, and the patients’ median KPS score at presentation was 80 (range, 70–90). Ninety-six patients (26%) presented with a motor deficit, 75 (20%) with headaches, 62 (17%) with speech or language difficulty, 53 (14%) with new-onset seizures, 49 (14%) with a visual deficit, and 14 (4%) with changes in mental status. Seventeen patients (5%) had a preoperative diagnosis of diabetes mellitus. Gross total resection was achieved in 228 patients (62%). All patients undergoing primary resection received adjuvant radiotherapy. One-hundred twenty patients (33%; all patients after 2004) received temozolomide (Temodar; Schering Corp., Kenilworth, NJ), and 53 (14%) underwent a subsequent resection. Two-hundred seventy-six patients (75%) died during the review period. For the 91 surviving patients (25%), the median follow-up period was 7 months (range, 3–18 mo). Overall, median survival was 11 months.

VOLUME 63 | NUMBER 2 | AUGUST 2008 | 287

MCGIRT ET AL.

Outpatient Hyperglycemia Sixty-eight patients (19%) experienced isolated outpatient hyperglycemia between 1 and 3 months after surgery. Twentyeight patients (8%) experienced persistent outpatient hyperglycemia between 1 and 3 months after surgery. Persistent outpatient hyperglycemia was secondary to diabetes mellitus in seven patients (25%) and secondary to Decadron use in 21 patients (75%). Patients experiencing persistent hyperglycemia were older (59 ⫾ 13 yr versus 51 ⫾ 14 yr, P ⬍ 0.05); were diabetic more frequently (7 [25%] versus 10 [3%], P ⬍ 0.05), and received Temodar less often (4 [14%] versus 116 [34%], P ⬍ 0.05). By 3 months after surgery, 11 persistently hyperglycemic patients (39%) demonstrated a stable residual contrast-enhancing tumor mass on MRI versus 128 normal-glycemic patients (38%). The incidence of early (⬍3 mo) tumor progression was also similar between the normal-glycemic and persistently hyperglycemic cohorts, 20 (6%) versus 2 (7%), respectively. However, in the persistently hyperglycemic cohort, there was a greater incidence of symptoms arising during adjuvant treatment (headaches, mild speech symptoms, mild motor weakness), prompting physicians to prescribe short courses of corticosteroid treatment (21 [75%] versus 35 [10%], P ⬍ 0.05). Otherwise, all clinical, radiographic, and treatment variables were similar between persistently hyperglycemic and normal-glycemic cohorts.

Predictors of Survival Younger age (P ⫽ 0.001), increasing KPS score (P ⫽ 0.001), WHO Grade III versus IV (P ⫽ 0.001), primary versus secondary resection (P ⫽ 0.008), Temodar therapy (P ⫽ 0.007), and a subsequent secondary resection (P ⫽ 0.070) were independently associated with improved survival in our patient population (Table 1). Persistent outpatient hyperglycemia (relative risk, 1.79; 95% confidence interval, 1.05–3.05, P ⫽ 0.03) was associated with decreased survival independent of age, KPS score, or all treatment variables (Table 1). This association of hyperglycemia and poor survival persisted after adjusting for continued 3-month outpatient Decadron therapy. In analysis of all malignant astrocytomas (WHO Grade III and IV), median survival for the persistently hyperglycemic cohort was 5 months compared with 11 months for the normalglycemic cohort (Fig. 1). For patients experiencing persistent hyperglycemia, median survival was 5 months for patients who were weaned off Decadron at the time of hyperglycemia and 6 months for patients with Decadron-associated hyperglycemia (1–3 mo postoperatively). Isolated outpatient hyperglycemia was not associated with survival (Fig. 2; Table 1). In the analysis of GBM patients only, median survival was reduced in the persistently hyperglycemic versus normalglycemic cohort (5 versus 10 mo, P ⬍ 0.01, log-rank analysis). This effect was observed after both primary GBM resection (5 versus 11 mo, P ⬍ 0.05) and revision GBM resection (5 versus 9 mo, P ⬍ 0.05). Even in patients with no residual or progressive tumor mass demonstrated on MRI by 3 months postoperatively, median survival was reduced in persistently hyperglycemic versus normal-glycemic patients (8 versus 13 mo, P ⬍ 0.05, log-rank analysis).

288 | VOLUME 63 | NUMBER 2 | AUGUST 2008

TABLE 1. Variables independently associated with survival after surgical resection of malignant brain astrocytomas (World Health Organization Grade III–IV) via multivariate proportionalhazards regression analysis (Cox model)a RR (95% CI)b

P value

Age

1.03 (1.01–1.04)

0.001c

World Health Organization Grade III

0.42 (0.29–0.61)

0.001c

Secondary resection

1.47 (1.11–1.97)

0.008c

Temodar

0.67 (0.51–0.89)

0.007c

Subsequent resection

0.72 (0.51–1.01)

0.070

Persistent outpatient hyperglycemiad

1.79 (1.05–3.05)

0.031c

Isolated outpatient hyperglycemiad

1.09 (0.85–3.65)

0.710

Continued outpatient Decadron therapyd

1.28 (0.77–2.01)

0.320

Variable

a

RR, relative risk; 95% CI, 95% confidence interval. Persistent outpatient hyperglycemia but not isolated outpatient hyperglycemia 1 to 3 months after surgery was associated with decreased survival independent of a diagnosis of continued outpatient Decadron therapy. c Statistically significant. d During outpatient follow-up period of 1 to 3 months. Persistent outpatient hyperglycemia, serum glucose ⱖ180 mg/dL on three or more occasions; isolated outpatient hyperglycemia, serum glucose ⱖ180 mg/dL once; continued outpatient Decadron therapy, patient still receiving oral Decadron at time of outpatient hyperglycemia. b

DISCUSSION In this study of 367 patients with high-grade gliomas (GBM and anaplastic astrocytomas), 68 (19%) and 28 patients (8%) experienced isolated or persistent outpatient hyperglycemia, respectively. Patients experiencing persistent hyperglycemia were older, were diabetic more frequently, remained on corticosteroids more frequently, and received Temodar less often. Adjusting for intergroup differences and variables associated with survival in this model, age, KPS score, tumor grade, primary versus secondary resection, use of Temodar, subsequent resection, and continued outpatient Decadron therapy, we found that persistent outpatient hyperglycemia remained independently associated with decreased survival. Median survival for persistently hyperglycemic versus normal-glycemic cohorts was 5 and 11 months, respectively. Even in patients with no residual or progressive tumor mass on MRI by 3 months postoperatively, median survival was reduced in persistently hyperglycemic versus normalglycemic patients (8 versus 13 mo). As supported in the multivariate analysis by the independent association of persistent hyperglycemia with survival, decreased survival was observed with persistent hyperglycemia in patients undergoing primary GBM resection or revision GBM resection, or with no residual or progressive tumor 3 months after surgery. However, it remains likely that some patients experienced hyperglycemic events that were not

www.neurosurgery-online.com

HYPERGLYCEMIA AFTER ASTROCYTOMA RESECTION

FIGURE 1. Persistent outpatient (Outpt) hyperglycemia: survival after resection of malignant astrocytomas (WHO Grade III–IV). Patients demonstrating persistent outpatient hyperglycemia (serum glucose ⬎180 mg/dL three or more times) experienced decreased survival compared with patients without persistent hyperglycemia (P ⫽ 0.001). Median survival for the persistently hyperglycemic and normal-glycemic cohorts was 5 and 11 months, respectively.

captured on laboratory draws. It also is unclear how patients’ variability in glycemic response to corticosteroids may contribute to the outcomes observed here. Although the incidence of less than 3 months’ progression and residual tumor was similar between cohorts, the increased incidence of corticosteroid use for neurological symptoms arising during adjuvant treatment likely predisposes to worse outcome in the hyperglycemic cohort. However, hyperglycemia remained independently associated with decreased survival after adjusting for Decadron use (Table 1). The prognosis for patients with GBM is dismal. The median survival period is only 14.6 months despite advances in surgical technology, radiotherapy, and chemotherapy (26). Management typically consists of surgery or biopsy to establish diagnosis and reduce the bulk of the tumor, followed by radiotherapy and/or chemotherapy. Even after extensive treatment, residual tumor is inevitable and patients eventually succumb to this devastating disease (1, 5, 7). The median survival is relatively short, but individual patient survival is heterogeneous, with some long-term survivors existing (8). Multiple prognostic factors have been proposed and include age, preoperative KPS score, tumor size, location, and even marital status, among others (2, 3, 15, 17). Currently, age and KPS score are the most significant factors (16). However, an understudied factor that may be prognostic of survival is persistent hyperglycemia after GBM resection. Although our observational study does not allow conclusions on the mechanism of action underlying the poor prognosis observed here in persistently hyperglycemic patients, the authors of many experimental studies have provided possible explanations. Basic science studies demonstrate that hyperglycemia may promote the growth of tumor cell lines, which may contribute to poor outcomes in patients with GBM (11). In the 1920s, Warburg noted that tumor cells rely on anaerobic

NEUROSURGERY

FIGURE 2. Isolated outpatient (Outpt) hyperglycemia: survival after resection of malignant astrocytomas (WHO Grade III–IV). Patients demonstrating isolated outpatient hyperglycemia (serum glucose ⬎180 mg/dL once) experienced similar survival compared with normal-glycemic patients (P ⫽ 0.71). Median survival for the isolated-hyperglycemic and normal-glycemic cohorts was 9 and 11 months, respectively.

glycolysis for the generation of adenosine triphosphate, even in the presence of oxygen (6, 32). The mechanism by which this happens is that glioma cell lines frequently overexpress a transcription factor called hypoxia-inducible factor-1a (24, 25, 34). Hypoxia-inducible factor-1a then activates the expression of several glycolytic enzymes, thus promoting glycolysis (24, 25, 34). In fact, glioma cell lines exhibit a threefold increase in glycolysis compared with nontumor astrocytes (20). Furthermore, the withdrawal of glucose leads to extensive apoptosis in these GBM cell lines but has little effect on nontumor astrocytes (11). In clinical studies, the effect of hyperglycemia on glioma growth and subsequent survival is unknown, but its effects on tumor progression have been observed for various adenocarcinomas, including breast, pancreas, prostate, and colon, among others (14, 19, 27, 28). Insulin resistance and hyperglycemia have been associated with increased risk of cancer development, as well as cancer progression, metastatic spread, and fatal outcome (14, 19, 27, 28). Although it is known that the presence of hypoglycemia slows experimental glioma cell growth (11), it seems intuitive that hyperglycemia may promote glioma cell growth. Hyperglycemia may also contribute to poor outcomes by decreasing the central nervous system’s tolerance to pathology (10, 12, 18, 31). Hyperglycemia leads to increased levels of advanced glycosylation end-products, quenching important vasodilators such as adenosine and nitric oxide, and it potentiates tissue acidosis, contributing to central nervous system injury (9, 23). Furthermore, numerous groups have shown that controlling blood glucose has a significant impact in limiting morbidity and mortality, length of hospital stay, and long-term functional deficits among critically ill patients (13, 21, 30, 33). Our findings suggest that poor glucose control and persistent outpatient hyperglycemia in patients with GBM are independently associated with decreased survival and may serve as prognostic markers in the outpatient management of these patients.

VOLUME 63 | NUMBER 2 | AUGUST 2008 | 289

MCGIRT ET AL.

Although in vitro glioma experiments and clinical studies on various nonneoplastic central nervous system pathologies have offered possible mechanisms underlying our observations, we cannot draw a conclusion on the etiology of hyperglycemiaassociated poor survival. Our observations do not differentiate hyperglycemia as cause or effect. Although the incidence of early tumor progression and 3-month residual tumor mass was not increased in the persistently hyperglycemic cohort, the increased Decadron use in this cohort suggests that the difference in survival may be an epiphenomenon of inherent tumor biology or degree of edema manifesting more clinically rather than radiographically. However, persistent hyperglycemia was associated with poor survival in patients both off and on prolonged 3-month Decadron therapy, supporting the notion that hyperglycemiaassociated poor survival was not entirely an effect of corticosteroid-associated hyperglycemia. It is our opinion that the poorly controlled hyperglycemia cohort represents a patient population of worse overall health status and is multifactorial in origin, thus predisposing them to worse systemic and glioma-specific outcome. Nonetheless, the associations observed in this study warrant investigation into the effects that hyperglycemia may have on patients with GBM as it pertains to tumor progression, recurrence, and survival. On the basis of our findings, we think that the need for a prospective trial of aggressive glucose control may be warranted in patients with malignant astrocytomas.

CONCLUSION In our experience, persistent outpatient hyperglycemia was associated with decreased survival in patients undergoing surgical resection for malignant astrocytomas, independent of the degree of disability, tumor grade, diabetes, prolonged Decadron use, or subsequent treatment modalities. A prospective trial of aggressive glucose control may be warranted in patients with malignant astrocytomas.

REFERENCES 1. Buckner JC: Factors influencing survival in high-grade gliomas. Semin Oncol 30:10–14, 2003. 2. Chang SM, Barker FG 2nd: Marital status, treatment, and survival in patients with glioblastoma multiforme: A population based study. Cancer 104:1975– 1984, 2005. 3. Chang SM, Parney IF, McDermott M, Barker FG 2nd, Schmidt MH, Huang W, Laws ER Jr, Lillehei KO, Bernstein M, Brem H, Sloan AE, Berger M, Glioma Outcomes Investigators: Perioperative complications and neurological outcomes of first and second craniotomies among patients enrolled in the Glioma Outcome Project. J Neurosurg 98:1175–1181, 2003. 4. Crooks V, Waller S, Smith T, Hahn TJ: The use of the Karnofsky Performance Scale in determining outcomes and risk in geriatric outpatients. J Gerontol 46:M139–M144, 1991. 5. Curran WJ Jr, Scott CB, Horton J, Nelson JS, Weinstein AS, Fischbach AJ, Chang CH, Rotman M, Asbell SO, Krisch RE, Nelson DF: Recursive partitioning analysis of prognostic factors in three Radiation Therapy Oncology Group malignant glioma trials. J Natl Cancer Inst 85:704–710, 1993. 6. Dang CV, Semenza GL: Oncogenic alterations of metabolism. Trends Biochem Sci 24:68–72, 1999. 7. DeAngelis LM: Brain tumors. N Engl J Med 344:114–123, 2001.

290 | VOLUME 63 | NUMBER 2 | AUGUST 2008

8. Dehdashti AR, Sharma S, Laperriere N, Bernstein M: Coincidence vs cause: Cure in three glioblastoma patients treated with brachytherapy. Can J Neurol Sci 34:339–342, 2007. 9. Dempsey RJ, Baskaya MK, Combs DJ, Donaldson D, Rao AM, Prasad MR: Effect of hyperglycemia on reperfusion-associated recovery of intracellular pH and high energy phosphates after transient cerebral ischemia in gerbils. Neurol Res 18:546–552, 1996. 10. Harik SI, LaManna JC: Vascular perfusion and blood-brain glucose transport in acute and chronic hyperglycemia. J Neurochem 51:1924–1929, 1988. 11. Jelluma N, Yang X, Stokoe D, Evan GI, Dansen TB, Haas-Kogan DA: Glucose withdrawal induces oxidative stress followed by apoptosis in glioblastoma cells but not in normal human astrocytes. Mol Cancer Res 4:319–330, 2006. 12. Kawai N, Keep RF, Betz AL, Nagao S: Hyperglycemia induces progressive changes in the cerebral microvasculature and blood-brain barrier transport during focal cerebral ischemia. Acta Neurochir Suppl 71:219–221, 1998. 13. Krinsley JS: Effect of an intensive glucose management protocol on the mortality of critically ill adult patients. Mayo Clin Proc 79:992–1000, 2004. 14. Krone CA, Ely JT: Controlling hyperglycemia as an adjunct to cancer therapy. Integr Cancer Ther 4:25–31, 2005. 15. Lacroix M, Abi-Said D, Fourney DR, Gokaslan ZL, Shi W, DeMonte F, Lang FF, McCutcheon IE, Hassenbusch SJ, Holland E, Hess K, Michael C, Miller D, Sawaya R: A multivariate analysis of 416 patients with glioblastoma multiforme: Prognosis, extent of resection, and survival. J Neurosurg 95:190–198, 2001. 16. Lamborn KR, Chang SM, Prados MD: Prognostic factors for survival of patients with glioblastoma: Recursive partitioning analysis. Neuro Oncol 6:227–235, 2004. 17. Laws ER, Parney IF, Huang W, Anderson F, Morris AM, Asher A, Lillehei KO, Bernstein M, Brem H, Sloan A, Berger MS, Chang S, Glioma Outcomes Project: Survival following surgery and prognostic factors for recently diagnosed malignant glioma: Data from the Glioma Outcomes Project. J Neurosurg 99:467–473, 2003. 18. Matchar DB, Divine GW, Heyman A, Feussner JR: The influence of hyperglycemia on outcome of cerebral infarction. Ann Intern Med 117:449–456, 1992. 19. Nilsen TI, Vatten LJ: Prospective study of colorectal cancer risk and physical activity, diabetes, blood glucose and BMI: Exploring the hyperinsulinaemia hypothesis. Br J Cancer 84:417–422, 2001. 20. Oudard S, Arvelo F, Miccoli L, Apiou F, Dutrillaux AM, Poisson M, Dutrillaux B, Poupon MF: High glycolysis in gliomas despite low hexokinase transcription and activity correlated to chromosome 10 loss. Br J Cancer 74:839–845, 1996. 21. Pittas AG, Siegel RD, Lau J: Insulin therapy for critically ill hospitalized patients: A meta-analysis of randomized controlled trials. Arch Intern Med 164:2005–2011, 2004. 22. Preiser JC, Devos P: Clinical experience with tight glucose control by intensive insulin therapy. Crit Care Med 35:S503–507, 2007. 23. Sappey-Marinier D, Chileuitt L, Weiner MW, Faden AI, Weinstein PR: Hypoglycemia prevents increase in lactic acidosis during reperfusion after temporary cerebral ischemia in rats. NMR Biomed 8:171–178, 1995. 24. Semenza GL, Artemov D, Bedi A, Bhujwalla Z, Chiles K, Feldser D, Laughner E, Ravi R, Simons J, Taghavi P, Zhong H: ‘The metabolism of tumours’: 70 years later. Novartis Found Symp 240:251–264, 2001. 25. Semenza GL, Roth PH, Fang HM, Wang GL: Transcriptional regulation of genes encoding glycolytic enzymes by hypoxia-inducible factor 1. J Biol Chem 269:23757–23763, 1994. 26. Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ, Belanger K, Brandes AA, Marosi C, Bogdahn U, Curschmann J, Janzer RC, Ludwin SK, Gorlia T, Allgeier A, Lacombe D, Cairncross JG, Eisenhauer E, Mirimanoff RO, European Organisation for Research and Treatment of Cancer Brain Tumors and Radiotherapy Groups, National Cancer Institute of Canada Clinical Trials Group: Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 352:987–996, 2005. 27. Suba Z, Ujpál M: Correlations of insulin resistance and neoplasms. Magy Onkol 50:127–135, 2006. 28. Suba Z, Barabás J, Szabó G, Takács D, Ujpál M: Increased prevalence of diabetes and obesity in patients with salivary gland tumors. Diabetes Care 28:228, 2005.

www.neurosurgery-online.com

HYPERGLYCEMIA AFTER ASTROCYTOMA RESECTION

29. Tait MJ, Petrik V, Loosemore A, Bell BA, Papadopoulos MC: Survival of patients with glioblastoma multiforme has not improved between 1993 and 2004: Analysis of 625 cases. Br J Neurosurg 21:496–500, 2007. 30. van den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M, Vlasselaers D, Ferdinande P, Lauwers P, Bouillon R: Intensive insulin therapy in the critically ill patients. N Engl J Med 345:1359–1367, 2001. 31. Voll CL, Auer RN: The effect of postischemic blood glucose levels on ischemic brain damage in the rat. Ann Neurol 24:638–646, 1988. 32. Warburg O: On the origin of cancer cells. Science 123:309–314, 1956. 33. Woodworth GF, Chaichana KL, McGirt MJ, Sciubba DM, Jallo GI, Gokaslan Z, Wolinsky JP, Witham TF: Predictors of ambulatory function after surgical resection of intramedullary spinal cord tumors. Neurosurgery 61:99–106, 2007. 34. Zundel W, Schindler C, Haas-Kogan D, Koong A, Kaper F, Chen E, Gottschalk AR, Ryan HE, Johnson RS, Jefferson AB, Stokoe D, Giaccia AJ: Loss of PTEN facilitates HIF-1-mediated gene expression. Genes Dev 14:391–396, 2000.

COMMENTS

I

n this interesting retrospective analysis, the authors reported that persistent hyperglycemia (defined as a serum glucose level ⬎180 mg/dL) was independently associated with worse survival outcome in patients with malignant gliomas. Over the years, a variety of prognostic factors (age, tumor grade, Karnofsky Performance Scale score, and others) have been shown to correlate with outcome in this group of patients. The association of hyperglycemia with glioma growth may have a biological rationale in accelerating the metabolism of the neoplasm although this remains speculative. This study does imply that perhaps improved glucose control would be a benefit although a prospective randomized trial would be required to prove this hypothesis. One may ask whether the observation is related to the inability to wean in patients who have residual tumor or patients who have tumors that are faster growing off steroids. However, McGirt et al. note that by 3 months after surgery, the percentage of patients with stable residual contrast-enhancing tumor was similar between the normoand hyperglycemic groups. Similarly, the incidence of early (⬍3 mo) tumor progression was similar between these two groups. In addition, the median survival of patients with hyperglycemia receiving or not receiving steroids was well below that of normoglycemic patients. Although the number of patients in each of these groups is small, the study implicates an elevated serum glucose level as an independent and possible marker of poorer prognosis. Although not addressed in this study, it may have been interesting to determine whether the persistently hyperglycemic patients were also obese. In summary, McGirt et al. have persistently tried to determine whether an elevated serum glucose level was associated with poorer outcome. Although we commonly use the Karnofsky Performance Scale as a predictor of prognosis in patients with gliomas, one wonders whether measures of overall health (serum glucose level, body mass index, and cardiovascular status) are also significant predictors of the extent of survival. E. Antonio Chiocca Columbus, Ohio

I

n this retrospective review of 367 patients with malignant gliomas, McGirt et al. showed that repeated glucose levels greater than 180 mg/dL in the 3-month postoperative period independently correlated with decreased survival in these patients. In general, the authors’ conclusions tend to confirm what has previously been reported for other intracranial pathological conditions (i.e., head

NEUROSURGERY

trauma, stroke, and others): the belief that excess hyperglycemia is a bad prognostic indicator. However, as presented in this article, hyperglycemia cannot be directly linked to mortality from tumor recurrence and may reflect overall medical condition. The correlation here was only between hyperglycemia and shorter overall survival, and there are no longitudinal data regarding tumor status (i.e., time to tumor progression or progression-free survival). Patients with severe, chronic, uncontrolled hyperglycemia may be less able to tolerate systemic chemotherapy, require prolonged steroid use, or might be dying from causes not directly resulting from tumor recurrence. Therefore, the direct impact of hyperglycemia on tumor control is unknown. Nevertheless, this is an interesting article, which suggests that we should exercise tighter glucose control postoperatively in our patients with malignant gliomas. Linda M. Liau Los Angeles, California

T

his is an important observation that should change the management of patients with malignant gliomas. Although the reasons for the association between persistent hyperglycemia and survival remain speculative, the reality is that attention to serum glucose levels after hospital discharge can potentially improve survival. The anticipated outcome for these patients remains dismal; therefore, anything we can do to improve on this is a welcomed finding. Joseph M. Piepmeier New Haven, Connecticut

M

cGirt et al. retrospectively reviewed 367 craniotomies for malignant brain astrocytomas (World Health Organization Grade III or IV) and defined persistent hyperglycemia as a serum glucose level greater than 180 mg/dL occurring three or more times in a 1- to 3-month postoperative period. They reported that persistent hyperglycemia was associated with decreased survival in these patients, and this result was independent of the degree of disability, tumor grade (World Health Organization Grade III or IV), diabetes, prolonged steroid use, or other treatment modalities. Intuitively, we would expected that sicker patients, with a greater residual tumor mass requiring higher dose steroid therapy, would have a worse prognosis and that these patients would have higher serum glucose levels. However, in their analysis in this retrospective study, McGirt et al. showed that hyperglycemia was an independent variable, and even in patients with no residual or progressive tumor on magnetic resonance imaging scans 3 months postoperatively, the median survival was reduced in patients with persistent hyperglycemia versus those with normoglycemia. Their observation does have a scientific basis, with hyperglycemia being known to promote the in vitro growth of tumor cell lines and withdrawal of glucose leading to apoptosis in glioblastoma multiforme cell lines. Nevertheless, they acknowledge that their observations are not conclusive in the causal etiology of hyperglycemia-associated poor survival and do not differentiate hyperglycemia as a cause or effect. However, I agree with their point that poorly controlled hyperglycemia is more likely to be a marker of a patient population that has an overall worse health status and is likely to have a poorer prognosis. This is an excellent and interesting study. Andrew H. Kaye Melbourne, Australia

VOLUME 63 | NUMBER 2 | AUGUST 2008 | 291