A Question about Hypoglyemia

[Ace844]

Hi ALL,

Since this is the latest discussion on this subject I decided to add this data here. For sometime, "dust, Rid, PRPG,& others" have been advocating and explaining to the group here why it's important to know the phys and how your meds you administer at any level will possibly effect your patient. As we have discussed previously in another thread, to blindly give glucose to someone who has a hx of Diabetes, and not do a thorough assessment can hurt pts. Now here's a study which shows this and explains it.

(Acad Emerg Med Volume 13 @ Number 2 174-180,

published online before print January 25, 2006, doi: 10.1197/j.aem.2005.08.009

© 2006 Society for Academic Emergency Medicine

[u)

CLINICAL PRACTICE

Decreased Mortality by Normalizing Blood Glucose after Acute Ischemic Stroke

Nina T. Gentile, MD, Michael W. Seftchick, BS, Tien Huynh, BS, Linda K. Kruus, PhD and John Gaughan, PhD

From the Temple University School of Medicine (NTG, MWS, TH, LKK, JG), Philadelphia, PA.

Address for correspondence and reprints: Nina T. Gentile, MD, Department of Emergency Medicine, Temple University School of Medicine, 1007 Jones Hall, 3401 N. Broad Street, Philadelphia, PA 19140. Fax: 215-707-3494; e-mail: ngentile@temple.edu .

]

ABSTRACT

Objectives: Hyperglycemia after cerebral ischemia exacerbates brain injury and worsens the outcome of stroke patients. The authors sought to examine the effect of glycemic control on mortality after acute stroke.

Methods: This was a retrospective study of patients discharged with a diagnosis of ischemic stroke during a 40-month period from a large urban U.S. health system. Patients were compared by initial blood glucose (BG) levels and by glycemic control during the first 48 hours of hospitalization.

Results: Of 960 patients with thromboembolic stroke, 373 (38.9%) were hyperglycemic (BG 130 mg/dL) on hospital admission. Admission hyperglycemia was associated with a higher mortality rate than was euglycemia (odds ratio [OR] = 3.15; 95% confidence interval [CI] = 1.45 to 6.85; p = 0.004). Persistent hyperglycemia (PerHyp) during 48 hours of hospitalization was associated with even higher mortality rate (unadjusted logistic regression, OR = 6.54; 95% CI = 2.41 to 17.87; p < 0.001). Glycemic control (normalization of BG to < 130 mg/dL) was associated with a 4.6-fold decrease in mortality risk as compared with the case of patients with PerHyp (p < 0.001). Multiple logistic regression showed glycemic control to be a strong independent determinant of survival (OR = 5.95; 95% CI = 1.24 to 28.6; p = 0.026) after acute stroke even after adjustment for age, gender, concomitant hypertension and diabetes, and stroke severity.

Conclusions: Admission hyperglycemia is associated with a worse outcome after stroke than is euglycemia. Normalization of blood glucose during the first 48 hours of hospitalization appears to confer a potent survival benefit in patients with thromboembolic stroke.

Key words: glucose; stroke; cerebral ischemia; cerebrovascular injury

INTRODUCTION

Hyperglycemia that accompanies acute stroke is associated with a worse outcome1,2,3 by exacerbating postischemic brain injury,4 cerebral edema,5 and hemorrhagic transformation.6,7 Even slightly elevated blood glucose (BG) levels of 125 to 130 mg/dL have been associated with a longer hospital length of stay (LOS), higher mortality rate,8 and increased infarct volume by MRI.9 Admission hyperglycemia has been shown to be an independent predictor for intracerebral hemorrhage and for mortality after acute stroke.6,7,8

The mechanisms for such detrimental effects are not clear. Long-term hyperglycemia triggers inflammatory changes in vessel walls and may cause irreversible microangiopathy. Alternatively, acute hyperglycemia may simply reflect a catecholamine-mediated stress response to a more severe stroke.10,11 Regardless, hyperglycemia in animal models has been shown to exacerbate acute ischemic conditions by stimulating vascular inflammation, increasing blood–brain barrier permeability, impairing cellular metabolism, and promoting tissue acidosis.12,13,14

Blood glucose control is important in limiting diabetic complications. Insulin reduces neuronal damage and improves functional outcome and mortality rates after experimental ischemic brain injury.15,16,17 Strict glycemic control has been associated with a reduced mortality in critically ill patients.18 Moreover, BG control, rather than the amount of exogenous insulin given to critically ill patients, appears to account for the mortality benefit.19

Few studies have examined glycemic control in stroke patients. Of those that have been conducted, strict glycemic control has been shown to decrease the incidence of bacteremia, polyneuropathy, and critical illness in patients who were treated in a surgical intensive care unit that included patients with severe stroke.18 One study found no difference between stroke patients treated with either glucose–potassium–insulin infusion or standard therapy20; however, treatment groups did not differ in BG levels, and this may account for the negative findings. Lacking substantial evidence to support more aggressive treatment, the current recommendations suggest treating only severe hyperglycemia (i.e., greater than about 300 mg/dL).21 The present study sought to confirm the relationship of hyperglycemia after acute stroke with mortality and to determine whether glycemic control after stroke improves outcome.

METHODS

Study Design

This was a retrospective study approved by the institutional review board. Medical records of patients admitted for acute stroke were reviewed to determine the effects of hyperglycemia on outcome.

Study Setting and Population

Patients were included if they were discharged from Temple University Hospital or Northeastern Hospital with acute cerebrovascular injury, diagnosis-related group (DRG) Code 14 (Specific Cerebrovascular Disorders Except Transient Ischemic Attack), between May 1999 and December 2002. We identified 1,197 patients with DRG 14; 237 of these patients were assigned a discharge diagnosis of subarachnoid, intracerebral, or other or unspecified intracranial hemorrhage and were excluded from further study. The remaining 960 patients discharged with thromboembolic stroke diagnosis codes made up the study sample. To confirm the reliability of the data, a subset of 80 patients' records was reviewed to determine the correctness of the diagnosis and DRG. In addition, hospital-derived data were cross-referenced with the Atlas query-specific database. This yielded no additional patients, suggesting good disease capture by DRG designation.

Study Protocol

Clinical and laboratory data were extracted from the MediTech Laboratory management system and the Atlas Database (2002 MediQual Systems, Inc., Westborough, MA), including age, gender, and race, and comorbid diabetes, hypertension (HTN), cancer, and heart failure. When not available in the MediTech or Atlas information systems, initial, 24-hour, and 48-hour BG levels were obtained from AccuCheck measurements recorded in the medical chart. Clinical data also included use of insulin, oral hypoglycemic agents (OHAs), heparin, and tissue plasminogen activator (t-PA) during hospitalization. LOS, discharge disposition, and diagnoses also were documented. The Glasgow Coma Scale score and All Patient-Refined Diagnosis-Related Groups (APR-DRGs; 3M Health Information Systems, Wallingford, CT) Risk of Mortality and Severity of Illness measures22 were used to control for baseline injury severity. These measures were used for two reasons. First, APR-DRG admission severity and risk of mortality indices were available through the Atlas administrative database on all hospitalized patients. Second, the APR-DRG incorporates clinical and laboratory variables that contribute to clinical disease severity. APR-DRG measures are constructed from standard discharge abstract data. Admission severity is coded on a scale from 1 to 4, representing minor, moderate, major, and extreme levels, respectively, and Risk of Mortality is coded on a scale of zero to one. Variables included in the equations are DRG diagnosis specific. For DRG 14, the variables included age in years, gender (male = 1), pulse rate (beats/min), respirations >29 per minute, coma score (3-15), presence of acute paresis, blood urea nitrogen (mg/dL), low sodium (mEq/L), high white cell count (109/L), and presence of bleeding on brain imaging. APR-DRG measures have been used to reliably predict death and compare resource consumption and outcomes among hospitalized patients22,23,24 and to adjust for injury severity when studying mortality in stroke patients.25 The Glasgow Coma Scale (GCS) score is a widely used and recognized measure of neurological injury. Although not a sensitive measure of focal neurological impairment, GCS score is reproducible and easily determined; and as such, it has been used as a surrogate marker of injury severity in studies of stroke patients.

Data Analysis

The primary endpoint was in-hospital mortality. Analyses were conducted by using Statistical Package for the Social Sciences (SPSS, Chicago, IL) software, versions 11.5 and 12.0. All statistical analyses were two-tailed and considered significant at the p 0.05 level. Descriptive statistics were generated for the demographic and clinical characteristics of the study sample. The relation of mortality and patient demographic and clinical factors, as well as initial, 24-hour, and 48-hour glucose levels were investigated first by using analyses without adjustment for confounding variables; significant predictors then were entered into a multivariate logistic regression equation.

Next, patient BG levels on hospital admission were compared. Initial BG was analyzed as both a continuous and a dichotomous variable. Two statistical methods were used to assess the ability of initial BG to predict mortality: linear regression analysis and the measurement of the area under the receiver operating characteristics curve. We chose to dichotomize BG for two reasons. First, distribution of initial BG values is asymmetrical, with a significant degree of skew (skewness statistic, 2.8; SE ± 0.08). Second, clinical effects of hyperglycemia are likely to be by status rather than by degree. Hyperglycemia was defined as BG 130 mg/dL. Cutoffs to define hyperglycemia have ranged from 108 mg/dL to 180 mg/dL (6 to 10 mmol/L) in previous studies.26 However, hyperglycemia more often has been defined as BG of 130 mg/dL.2,3

Patients with hyperglycemia on admission were compared with euglycemic patients (BG = 60 to 129 mg/dL) in terms of demographics, disposition, and outcome using Student's t-test, analysis of variance, and Fisher's exact test. Logistic regression analyses were used to examine the relationship of mortality to demographic and clinical variables described above. In the multivariate analyses, all variables were entered into the model simultaneously. To control for severity of illness and risk of death, both APR-DRG risk adjustment systems were included in the multivariate model at the same time and separately, to explore their effects on the results. Final results were consistent with the use of either risk adjustment system or both measures when entered simultaneously.

Patients with hyperglycemia on presentation were grouped in terms of whether their BG remained elevated (PerHyp) or decreased to < 130 mg/dL (Cont) within 48 hours after admission. Similarly, euglycemic patients on admission whose BG remained < 130 mg/dL during 24 to 48 hours after admission were considered persistently euglycemic (PerEug), and initially euglycemic patients with BG 130 mg/dL within 48 hours were considered uncontrolled (Uncont). For consistency, glycemic control groups were compared with PerHyp as the referent group. Multiple logistic regressions were used to compare glycemic control groups in terms of demographic and clinical characteristics. Model fit was assessed by using the Hosmer and Lemeshow test. The statistic ( 2 = 4.33, df = 8; p = 0.83) showed a good fit for the multivariate regression model used. Cross-products of potentially influential covariates were analyzed.

RESULTS

Patient Characteristics

The study sample (n = 960) included patients identified as African American (68%), Hispanic (13%), Caucasian (11%), and female (55%), and the patients ranged in age from 20 to 101 years (mean ± SD = 65.7 ± 13.6). Mean BG level at admission was 151.8 (SD ± 94.4 mg/dL); mean 24-hour and 48-hour BG levels were 143.7 (SD ± 74.8 mg/dL) and 140.2 (SD ± 74.6 mg/dL), respectively. Comorbid diagnoses included diabetes (n = 374, 39.0%), HTN (n = 708, 73.8%), heart failure (n = 137, 14.3%), and cancer (n = 47, 4.9%). Six patients (0.5%) had a concurrent acute myocardial infarction. Patients were administered t-PA (n = 7, 0.7%), heparin (n = 129, 13.4%), insulin without an oral hypoglycemic agent (OHA; n = 162, 16.9%), one or more OHAs without insulin (n = 44, 4.6%), and the combination of insulin and one or more OHAs (n = 272, 28.3%). Seventeen (1.8%) patients had spontaneous intracerebral hemorrhage during hospitalization. The median LOS was six days (range: 1 to 96 days). The majority of patients returned home after discharge (72.6%); 24.0% transferred to another facility, and 3.4% died during hospitalization.

Independent Predictors of Mortality

Table 1 shows unadjusted univariate logistic regression analyses of patient demographic and clinical variables as they relate to mortality. Age, initial and 48-hour BG as well as comorbid HTN and injury severity were significant predictors of mortality. Patients with higher initial and 48-hour BG levels were at increased risk of death than patients with lower BG levels. Stroke severity measurements were strong predictors of mortality, and the presence of HTN was associated with lower death rates.

Admission Hyperglycemia as a Predictor of Mortality

Of the 960 patients, 2.1% (n = 20) did not have documented initial BG and were excluded; thus, data for 940 patients were available for us to use in examining the relationship between initial BG level and mortality. Figure 1 shows the attrition of patients throughout the study. Linear regression analysis showed initial BG to be a sufficiently strong predictor of mortality (F = 6.2; p = .019). In addition, a receiver operating characteristic curve of initial BG was plotted as a function of mortality (Figure 2). The area under the curve was 0.68 ± 0.04 (95% CI = 0.598 to 0.758; p = 0.001) and suggests that initial BG is a moderately good predictor of mortality. When we used a cutoff of 130 mg/dL, 373 (38.9%) patients were hyperglycemic, and 567 (59.1%) were euglycemic, with BG between 60 and 129 mg/dL. The mean age of patients in the euglycemic group (65.2 ± 14.3 yr) was similar to that of hyperglycemic patients (66.3 ± 12.1 yr; p = 0.19). More African Americans presented as euglycemic (404/567; 71.3%) than as hyperglycemic (242/373; 65.1%; 2 = 4.02; p < 0.05), and more women presented as hyperglycemic (220/373, 59.0%) than as euglycemic (294/567, 51.9%, 2 = 4.62; p < 0.05). Hyperglycemic patients had higher admission severity scores (OR = 1.54; 95% CI = 1.085 to 2.196; p < 0.02). However, risk of mortality assessments were similar between groups (OR = 2.4; 95% CI = 0.22 to 26.6), as were GCS scores (OR = 0.98; 95% CI = 0.91 to 1.07). Hyperglycemic patients had a slightly longer hospital LOS (8.03 ± 8.8 days) than euglycemic patients (7.00 ± 7.4 days; p = 0.05). Five (0.8%) euglycemic patients, as compared with 12 (3.2%) hyperglycemic patients, had concomitant intracerebral hemorrhage ( 2 = 5.66; p < 0.01). In-hospital mortality was 1.9% (11/567) in euglycemic and 5.6% (21/373, 2 = 9.32; p < 0.01) in hyperglycemic patients. Univariate logistic regression showed that patients with admission hyperglycemia were more likely to die during that hospitalization than were patients presenting with BG levels in the normal range (OR = 3.01; 95% CI = 1.44 to 6.33; p = 0.004). This relationship persisted after controlling for age, gender, race, comorbid HTN, and diabetes as well as for injury severity (OR = 3.29, 95% CI = 1.34 to 8.06; p = 0.009).

Persistent Hyperglycemia and Glycemic Control as Predictors of Mortality

The majority of patients with normal BG on admission continued to be euglycemic (PerEug) 24 and 48 hours after presentation (n = 396/511, 77.5%). Likewise, most patients with hyperglycemia on admission continued to have BG 130 mg/dL during 48 hours of admission (246/357, 68.9%). Thirty-one percent (111/357) of patients with admission hyperglycemia had BG that fell into the normal range (Cont) within 48 hours, and 22% (115/511) of initially euglycemic patients became uncontrolled (Uncont) with BG 130 mg/dL during the following 48 hours.

Clinical and demographic characteristics of glycemic control groups are presented in Table 2. Unadjusted logistic regression analyses showed that Cont patients were significantly less likely to die than were PerHyp patients (OR = 0.22; 95% CI = 0.05 to 0.96; p < 0.05). This relationship strengthened when adjusting for age, gender, concomitant HTN and diabetes, and for stroke severity (OR = 0.17; 95% CI = 0.03 to 0.81; p = 0.026). In-hospital mortality rates of Uncont patients were similar to PerHyp patients (4.3% vs. 7.7%, respectively; p = 0.26, Fisher's exact test). Likewise, Cont patients had mortality rates similar to PerEug patients after stroke (1.8% vs. 1.3%, respectively; p = 0.65, Fisher's exact test).

There were no significant interactions between the effects of BG and race or gender on mortality, but the effect of glycemic control appeared to be contingent to a small degree on the patient's age (OR = 1.2; 95% CI = 1.01 to 1.04). Comorbid diabetes was positively associated with mortality after stroke. However, diabetics with consistently normal BG levels had lower mortality odds (OR = 0.28; 95% = CI = 0.10 to 0.75; p = 0.01).

DISCUSSION

Blood glucose normalization after ischemic stroke was associated with decreased mortality. Hyperglycemia, especially when it persisted during the first 48 hours after admission, increased the risk of mortality. Glycemic control (i.e., spontaneous or intentional normalization of BG to < 130 mg/dL in patients with admission hyperglycemia) was an independent predictor of survival even after controlling for age, stroke severity, and other potentially confounding conditions.

The overall rate of in-hospital death in this study was 3.4%, increasing from 0.8% in patients younger than 55 years to 5.2% in patients aged 75 years or older. In-hospital mortality ranges from as low as 3.3%27 to more than 10% among stroke patients who are considered for intravenous t-PA.25 The relatively low mortality rate in this study may be due to the younger age of our cohort (65 years, as compared with 72 years in patients treated in community hospitals25) and its predominance of African Americans.28

In-hospital mortality was associated with increasing age, stroke severity, and BG concentration. Although hyperglycemia occurs in diabetics and nondiabetics, it has been suggested that BG rises as a result of early hormonal response to the stress of the cerebral ischemia in nondiabetics.10,11 If high BG is a physiological adaptation to injury, it would only be expected with severe stroke, and it actually could be beneficial in stroke patients without diabetes.29 In this study, hyperglycemia was associated with a higher Admission Severity score. However, there was no association between BG and Risk of Mortality assessment or with GCS score, and admission hyperglycemia was an independent predictor of mortality even after controlling for disease severity.

Acute hyperglycemia has been associated with hemorrhagic transformation and intracerebral hemorrhage after ischemic infarct.6,7 In this study, 12 (76%) of 17 patients with intracerebral hemorrhage after thromboembolic stroke were hyperglycemic on admission, and 83% were PerHyp at 48 hours. There is evidence that hyperglycemia and/or hyperinsulinemia affects several coagulation factors.30 Hypercoagulation in combination with alterations in thrombolysis may, in part, explain the increase in both intracerebral hemorrhage and mortality in hyperglycemic patients after stroke.

Glycemic control was associated with a marked reduction in in-hospital mortality when compared with persistent hyperglycemia. Insulin has been shown to be protective after experimental ischemic brain injury.16,17 Its effectiveness, at least as shown in animals, is predominantly via peripheral hypoglycemia rather than central cytoprotection.16 In the present study, glycemic control was associated with an improvement in outcome in untreated patients, in patients treated with either insulin or OHA, and in patients treated with both insulin and OHA. As with critically ill patients,19 outcome after stroke may depend on peripheral BG levels rather than on insulin administration. The presence of diabetes in the patients with PerEug was associated with a 3.5-fold decrease in mortality risk. However, diabetes did not impact mortality in PerHyp or Contr patients. This finding also suggests that it is the peripheral BG level rather than the presence of diabetes that affects outcome after stroke.

When multiple logistic regressions were used, glycemic control remained an independent predictor of in-hospital mortality even after adjusting for the other potentially confounding variables. Lowering BG concentration may reduce membrane permeability and cerebral edema, restore endothelium-dependent vascular flow dysfunction, and reverse the alterations in coagulation and oxidative injury that are caused by hyperglycemia after stroke.5,30,31

Typical in-hospital management of hyperglycemia results in relatively poor BG control. Details of therapy administered during the hospitalization that affect glycemic control are provided in Table 2. Most patients who were hyperglycemic on admission were treated with insulin, an OHA, or both. Yet two thirds of patients with elevated BG levels on admission remained hyperglycemic during the subsequent 48 hours. Conventional in-hospital management of hyperglycemia includes the combination of the patient's outpatient regimen (an OHA, insulin, or both) and sliding-scale insulin titrated for BG levels of 180 to 200 mg/dL or higher. Protocols using sliding-scale insulin are widely used. However, the current study confirms prior reports that glycemic control using this method is generally poor,31 and our results support lower glycemic targets and the growing use of continuous insulin infusion.32,33

LIMITATIONS

The present study was limited by its retrospective design and by use of a data set from a single urban health system. The predominance of African Americans also may limit the ability to generalize the results to a more heterogeneous population. Although attempts were made to control for demographic and clinical characteristics between groups, the retrospective nature of the study precludes the ability to control for all potential confounders. For example, there are other conditions that could impact glycemic control, including infection, renal failure, and so on. Specifically controlling for these conditions and for the medications used for their treatment was not feasible in this retrospective study. However, adjustment for APR-DRG injury severity measures that incorporate clinical and laboratory studies and for comorbid conditions, such as diabetes and HTN, help to control for factors that potentially affect glycemic control. Despite its limitations, this study does provide the basis for further study of the mechanisms underlying the detrimental effects of hyperglycemia and provides support for the prospective study of glycemic control in stroke.

CONCLUSIONS

Admission hyperglycemia is associated with a worse outcome after stroke than is euglycemia. Normalization of BG during the first 48 hours of hospitalization appears to confer a potent survival benefit in patients with thromboembolic stroke

FOOTNOTES

Supported in part by the Return of Overhead Incentive Grant Program from Temple University School of Medicine.

REFERENCES

TOP

ABSTRACT

INTRODUCTION

METHODS

RESULTS

DISCUSSION

LIMITATIONS

CONCLUSIONS

REFERENCES

1. Weir CJ, Murray GD, Dyker AG, Lees KR. Is hyperglycaemia an independent predictor of poor outcome after acute stroke? Results of a long-term follow-up study. BMJ. 1997; 314:1303–6.[Abstract/Free Full Text]

2. Williams LS, Rotich J, Qi R, et al. Effects of admission hyperglycemia on mortality and costs in acute ischemic stroke. Neurology. 2002; 59:67–71.[Abstract/Free Full Text]

3. Bruno A, Levine SR, Frankel MR, et al.and the NINDS rt-PA Stroke Study Group. Admission glucose level and clinical outcomes in the NINDS rt-PA Stroke Trial. Neurology. 2002; 59:669–74.[Abstract/Free Full Text]

4. Pulsinelli WA, Levy DE, Sigsbee B, Scherer P, Plum F. Increased damage after ischemic stroke in patients with hyperglycemia with or without established diabetes mellitus. Am J Med. 1983; 74:540–4.[CrossRef][Medline]

5. Berger L, Hakim AM. The association of hyperglycemia with cerebral edema in stroke. Stroke. 1986; 17:865–71.[Abstract/Free Full Text]

6. de Courten-Myers GM, Kleinholz M, Holm P, et al. Hemorrhagic infarct conversion in experimental stroke. Ann Emerg Med. 1992; 21:120–6.[Medline]

7. Demchuk AM, Morgenstern LB, Krieger DW, et al. Serum glucose level and diabetes predict tissue plasminogen activator-related intracerebral hemorrhage in acute ischemic stroke. Stroke. 1999; 30:34–9.[Abstract/Free Full Text]

8. Bruno A, Biller J, Adams HP Jr, et al.for the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) Investigators. Acute blood glucose level and outcome from ischemic stroke. Neurology. 1999; 52:280–4.[Abstract/Free Full Text]

9. Baird TA, Parsons MW, Phanh T, et al. Persistent poststroke hyperglycemia is independently associated with infarct expansion and worse clinical outcome. Stroke. 2003; 34:2208–14.[Abstract/Free Full Text]

10. Tracey F, Stout RW. Hyperglycemia in the acute phase of stroke and stress response. Stroke. 1994; 25:524–5.[Medline]

11. O'Neill PA, Davies I, Fullerton KJ, Bennett D. Stress hormone and blood glucose response following acute stroke in the elderly. Stroke. 1991; 22:842–7.[Abstract/Free Full Text]

12. Widmer H, Abiko H, Faden AI, James TL, Weinstein PR. Effects of hyperglycemia on the time course of change in energy metabolism and pH during global cerebral ischemia and reperfusion in rats. J Cereb Blood Flow Metab. 1992; 12:456–68.[Medline]

13. Dietrich WD, Alonso O, Busto R. Moderate hyperglycemia worsens acute blood-brain barrier injury after forebrain ischemia in rats. Stroke. 1993; 24:111–6.[Abstract/Free Full Text]

14. Folbergrova J, Memezawa H, Smith ML, Siesjo BK. Focal and perifocal changes in tissue energy state during middle cerebral artery occlusion in normo- and hyperglycemic rats. J Cereb Blood Flow Metab. 1992; 12:25–33.[Medline]

15. Zhu CZ, Auer RN. Intraventricular administration of insulin and IGF-1 in transient forebrain ischemia. J Cereb Blood Flow Metab. 1994; 14:237–42.[Medline]

16. Hamilton MG, Tranmer BI, Auer RN. Insulin reduction of cerebral infarction due to transient focal ischemia. J Neurosurg. 1995; 82:262–8.[Medline]

17. LeMay DR, Gehua L, Zelenock GB, D'Alecy LG. Insulin administration protects neurologic function in cerebral ischemia in rats. Stroke. 1988; 19:1411–9.[Abstract/Free Full Text]

18. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. N Engl J Med. 2001; 345:1359–67.[Abstract/Free Full Text]

19. Finney SJ, Zekveld C, Elia A, Evans TW. Glucose control and mortality in critically ill patients. JAMA. 2003; 290:2041–7.[Abstract/Free Full Text]

20. Scott JF, Robinson GM, French JM, O'Connell JE, Alberti KG, Gray CS. Glucose potassium insulin infusions in the treatment of acute stroke patients with mild to moderate hyperglycemia. Stroke. 1999; 30:793–9.[Abstract/Free Full Text]

21. Adams HP Jr, Adams RJ, Brott T, et al. Guidelines for the early management of patients with ischemic stroke. Stroke. 2003; 34:1056–9.[Free Full Text]

22. Edwards N, Honemann D, Burley D, Navarro M. Refinement of Medicare diagnosis-related groups to incorporate a measure of severity. Health Care Fin Rev. 1994; 16:45–64.[Medline]

23. Iezzoni LI, Shwartz M, Ash AS, Hughes JS, Daley J, Mackiernan YD. Using severity-adjusted stroke mortality rates to judge hospitals. Int J Qual Health Care. 1995; 7:81–94.[CrossRef][Medline]

24. Iezzoni LI, Ash AS, Shwartz M, Landon BE, Mackiernan YD. Predicting in-hospital deaths from coronary artery bypass graft surgery: do different severity measures give different predictions? Med Care. 1998; 36:28–39.[CrossRef][Medline]

25. Reed SR, Cramer SC, Blough DK, Meyer K, Jarvik JG. Treatment with tissue plasminogen activator and inpatient mortality rates for patients with ischemic stroke treated in community hospitals. Stroke. 2001; 32:1832–40.[Abstract/Free Full Text]

26. Capes SE, Hunt D, Malmberg K, Pathak P, Gerstein HC. Stress hyperglycemia and prognosis of stroke in nondiabetic and diabetic patients. Stroke. 2001; 32:2426–32.[Abstract/Free Full Text]

27. Chen ZM, Sandercock P, Pan HC, et al. Indications for early aspirin use in acute ischemic stroke. Stroke. 2000; 31:1240–9.[Abstract/Free Full Text]

28. Gordon HS, Harper DL, Rosenthal GE. Racial variation in predicted and observed in-hospital death: a regional analysis. JAMA. 1996; 276:1639–44.[Abstract]

29. Oppenheimer SM, Hoffbrand BI, Oswald GA, Yudkin JS. Diabetes mellitus and early mortality from stroke. BMJ. 1985; 291:1014–5.[Medline]

30. Rao AK, Chouhan V, Chen X, Sun L, Boden G. Activation of the tissue factor pathway of blood coagulation during prolonged hyperglycemia in young healthy men. Diabetes. 1999; 48:1156–61.[Abstract]

31. Davi G, Ciabattoni G, Consoli A, Mezzetti A, Falco A, Santarone S. In vivo formation of 8-iso-prostaglandin F2alpha and platelet activation in diabetes mellitus: effects of improved metabolic control and vitamin E supplementation. Circulation. 1999; 99:224–9.[Abstract/Free Full Text]

32. Moghissi E. Hospital management of diabetes: beyond the sliding scale. Cleve Clin J Med. 2004; 71:801–8.[Medline]

33. Abourizk NN, Vora CK, Verma PK. Inpatient diabetology. The new frontier. J Gen Intern Med. 2004; 19:466–71.[CrossRef][Medline]

Hope this helps,

ACE844

[cosgrojo]

Cos ......To answer the Glucagon question our standing order at the NH for low BS is give it first if unresponsive ....I aggressive treat my residents because unlike some of you guys that think that Nursing home nurses are stupid and lazy I am not...I was not trained as nurse or a EMT I to not do what is needed to save a life ...I do what needs to be done when it needs to be done. I work at a boondocksville NH and honey you can bet your sweet ass that i know what cardiac arrest looks like and how to initate treatment on it.

I did not say all Nursing home nurses are stupid and lazy, that would be like saying all EMT's are intelligent, reasonable, and moderately educated. I was not taking a pot-shot at you, I was simply making an observation (as wrong, crass, and incredibly insensitive it may have been), about many of my own personal experiences. I have met many amazingly talented and intelligent nursing home nurses in my day, who on a daily basis raise the level of care for an entire facility. But I've also seen some very bad, non-chalant care being given in a lot of these facilities. I don't know which one you work at.... who knows you might work for one that is amazing with an amazing staff.... but don't tell me that you haven't seen dreadfully sub-standard care in the SNF's in our area. You don't say that, and I won't say that I've never seen an EMT give poor care in the pre-hospital setting.

I don't know who you are.... but judging by the comment about my sweet ass.... I'm wagering you know me.... :wink: