Quality of Care and Outcomes in Patients With Diabetes Hospitalized With Ischemic Stroke
Findings From Get With the Guidelines–Stroke
Background and Purpose— Diabetes is a common comorbid disease in stroke patients and has a strong influence on stroke-related outcomes, including stroke recurrence. We sought to examine the quality of care and in-hospital outcomes in patients with diabetes in the Get With the Guidelines–Stroke (GWTG-Stroke) program.
Methods— Data were obtained from 415 926 ischemic stroke patients from 1070 United States hospitals that participated in GWTG-Stroke between 2003 and 2008. We analyzed the relationships between diabetes and quality of care, in-hospital mortality, and discharge home using multivariable logistic regression.
Results— There were 130 817 (31%) ischemic stroke patients with diabetes. Quality of care received by patients with and without diabetes was similar except for intravenous recombinant tissue plasminogen activator (rt-PA) and cholesterol treatment. Fifty-four percent of patients with diabetes who arrived within 2 hours of onset received rt-PA compared to 60.8% of patients without diabetes (adjusted odds ratio [aOR], 0.83; 95% CI, 0.79–0.88). Almost 80% of patients with diabetes were discharged on cholesterol treatment compared to 71% of patients without diabetes (aOR, 1.40; 95% CI, 1.37–1.44). Diabetes patients were less likely to be discharged home (aOR, 0.80; 95% CI, 0.78–0.81) and had a higher risk of in-hospital death (aOR, 1.12; 95% CI, 1.08–1.15).
Conclusions— Quality of care among patients with and without diabetes was similar except for rt-PA and cholesterol treatment. Diabetes was associated with worse stroke-related outcomes. Greater quality-improvement efforts to increase the use of rt-PA and other secondary prevention treatments in patients with diabetes are warranted.
Almost 24 million Americans—7.8% of the population— have diabetes and ≈1.6 million new cases are diagnosed every year.1 The prevalence of diabetes in the United States has been steadily increasing over recent decades.2 Diabetes is the seventh leading cause of death in the United States and is associated with a wide range of chronic conditions, including heart disease, stroke, kidney disease, retinopathy, and peripheral neuropathy.1 Stroke is the third leading cause of death and the leading cause of adult disability in the United States. There are currently an estimated 795 000 stroke cases and 137 000 stroke deaths annually in the United States. Diabetes is an independent risk factor for both incident3 and recurrent stroke.4 Diabetes is a common comorbidity in stroke patients5–7 and is associated with poor outcomes after stroke.5,6
The acute hospital management of stroke patients with diabetes is complicated by the need for aggressive and early interventions to reduce the progression and long-term neurological consequences of cerebral injury, as well as for aggressive treatment of cardiovascular risk factors to reduce stroke recurrence.8 Clinical guidelines relevant to the hospital care of acute ischemic stroke in patients with diabetes have been issued by the American Stroke Association and American Diabetes Association.9–11 Guidelines for the secondary prevention of stroke include the following specific recommendations for patients with diabetes: more rigorous control of blood pressure and lipids, the use of angiotensin-converting enzyme inhibitors or angiotensin receptor blockers as first-line antihypertensive medications, the attainment of near-normoglycemic levels, and a glycohemoglobin (HbA1c) goal of <7%.9
Despite the common occurrence of diabetes in patients presenting with ischemic stroke, data on the quality of stroke-related care among hospitalized patients with diabetes are limited.7 Using data from ischemic stroke admissions in the Get With the Guidelines–Stroke (GWTG-Stroke) program, our objectives were to: (1) compare the quality of stroke-related care (ie, interventions addressing the management of stroke) among patients with and without diabetes; (2) document the quality of diabetes-related care (ie, interventions addressing the management of diabetes) in patients with diabetes; and (3) quantify the association between diabetes and stroke-related outcomes.
Materials and Methods
The GWTG-Stroke program is a voluntary, national, quality-improvement program sponsored by the American Heart Association and American Stroke Association that has been in development since 2000. Details of the design and conduct of the GWTG-Stroke program have been previously described.12 Whereas the GWTG-Stroke program is overrepresented with larger academic teaching hospitals, the patient demographics and comorbidites are similar to those described in other stroke registries13 and administrative databases.14 Outcome Sciences serves as the data collection and coordination center for GWTG. The Duke Clinical Research Institute serves as the data analysis center and has an agreement to analyze the aggregate de-identified data for research purposes.
Case Identification and Data Abstraction
Trained hospital personnel were instructed to ascertain consecutive patients admitted with the principal clinical diagnosis of acute stroke or TIA by either prospective clinical identification, retrospective identification using discharge codes, or a combination of these. Methods used for the prospective clinical identification of cases involved regular review of a combination of data sources, including emergency department admission logs, ward census logs, intensive care unit logs, and neurology service consultations. Methods used for the retrospective clinical identification of cases included regular surveillance of discharge codes, specifically ICD-9 433.xx, 434.xx, and 436 for ischemic stroke and 435.xx for TIA. The eligibility of all acute stroke admissions was confirmed before chart abstraction.12
Patient data including demographics, medical history, in-hospital treatment and procedures, discharge treatments and counseling, in-hospital mortality, and discharge destination were abstracted by trained hospital personnel. All patient data were de-identified before submission. Data on hospital-level characteristics (ie, bed size, academic or nonacademic status, annual stroke volume, and geographical region) were obtained from the American Hospital Association.15
Between April 1, 2003 and September 30, 2008, 716 107 acute stroke admissions were submitted by 1211 participating hospitals. We excluded 288 230 (40.2%) nonischemic stroke cases (56.5% had TIA, 26.4% had intracerebral hemorrhage, 8.4% had subarachnoid hemorrhage, and 8.7% had unknown stroke type). Of the remaining 427 877 ischemic stroke cases, we excluded 8481 (2.0%) because of missing hospital-level data and 376 (0.09%) because of missing information on gender. Diabetes status was determined based on a medical history of diabetes mellitus or the use of diabetes-related medications before admission. We excluded 3094 (0.7%) cases because we could not determine their diabetes status. The final analysis consisted of 415 926 admissions from 1070 hospitals, which represented 97.2% of the available ischemic stroke cases.
Quality of Stroke-Related Care
Seven performance measures, selected by the GWTG-Stroke program as primary targets for stroke quality-improvement efforts,12 were used to compare the quality of stroke-related care between ischemic stroke admissions with and without diabetes. Acute ischemic stroke performance measures are: (1) intravenous tissue plasminogen activator in patients who arrive <2 hours after symptom onset and with no contraindications to treatment; (2) antithrombotic medication (includes any aspirin, aspirin/dipyridamole, ticlopidine, clopidogrel, unfractionated heparin, low-molecular-weight heparin, and warfarin) administered within 48 hours of admission; and (3) deep vein thrombosis prophylaxis (includes heparins, heparinoids, other anticoagulants, or pneumatic compression devices) within 48 hours of admission in nonambulatory patients. Discharge ischemic stroke performance measures were: (1) antithrombotic (includes any aspirin, aspirin/dipyridamole, ticlopidine, clopidogrel, unfractionated heparin, low-molecular-weight heparin, and warfarin) medication; (2) anticoagulation (includes therapeutic doses of warfarin, heparin(oid), or other anticoagulants such as direct thrombin inhibitors) for patients with atrial fibrillation or flutter identified during the hospital admission; (3) cholesterol treatment (includes statins, fibrates niacin, binding resins, or selective cholesterol absorption inhibitors) if low-density lipoprotein cholesterol (LDL-C) >100 mg/dL or if LDL-C is not documented; and (4) counseling or medication for smoking cessation.
The following exclusions were applied to all discharge measures: died in hospital (5.5%); discharged to hospice (3.3%); transferred out (2.8%); left against hospital advice (0.5%); and discharge destination missing (1.3%). To summarize the overall quality of stroke-related care, we calculated a defect-free measure of care,16 which is a binary variable calculated as the proportion of patients who received all of the interventions for which they were eligible.
Quality of Diabetes-Related Care
To document the quality of diabetes-related care (ie, interventions with special relevance to patients with diabetes), we used available GWTG-Stroke data from the 130 817 patients with diabetes to develop the following 6 indicators or metrics that reflect current American Stroke Association or American Diabetes Association guideline recommended standards of care for secondary prevention.9–11 First, in-hospital HbA1c level was measured. This measure directly addresses the recommendation that hospitalized patients with diabetes have an HbA1c measured unless test results are available from the previous 2 to 3 months.11 Second, in-hospital LDL-C level was measured or value from previous 30 days was documented. This measure addresses the recommendations for rigorous control of lipids,9 with target LDL-C levels of <100 mg/dL.10,11 Third, discharge on cholesterol treatment if LDL-C >100 mg/dL or LDL-C was not documented. The rationale for this measure is the same as for the previous measure.9–11 Fourth, discharge on angiotensin-converting enzyme inhibitors or angiotensin receptor blockers medication if the patient has a medical history of hypertension. This measure directly addresses the recommendation that these medications be used as first-line agents for the treatment of hypertension in patients with diabetes.9 Fifth, discharge on an antihypertensive agent (defined as angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, beta blocker, calcium channel blocker, diuretic, or other antihypertensive agent) if the patient has a medical history of hypertension. This measure reflects the recommendations for rigorous control of blood pressure in patients with diabetes with target blood pressure levels of 130/80 mm Hg.9,11 Sixth, discharge on any diabetes-related medication (defined as any oral agent, insulin, or other subcutaneous or injectable agent). This measure reflects the recommendations for attainment of near-normoglycemic levels in stroke patients with diabetes,9 with target HbA1c levels of ≤7%.9,11
The guidelines also include specific recommendations with respect to target blood pressure and blood glucose levels. However, because the GWTG-Stroke program does not capture these data, we chose not to report an overall defect-free measure for diabetes-related care.
Patient demographic and clinical variables, hospital-level characteristics, and compliance with the individual and summary quality-of-care measures were compared between patients with and without diabetes. Percentages and means±SD were reported for categorical and continuous variables, respectively. Pearson χ2 test and Wilcoxon rank-sum tests were used to compare the categorical and continuous variables, respectively, between patients with and without diabetes.
The relationship between diabetes status and compliance with individual performance measures, as well as the defect-free summary measure of care were further examined using multivariable logistic regression models. To account for within-hospital clustering, generalized estimating equations were used to generate unadjusted and adjusted models.17 The adjusted models included the following prespecified potential confounders: age, race, gender, medical history (including atrial fibrillation, prosthetic heart valve, previous stroke/TIA, coronary heart disease, or previous myocardial infarction [coronary artery disease/previous MI], carotid stenosis, peripheral vascular disease, hypertension, dyslipidemia, heart failure, and current smoking), hospital size, region, teaching status, and the number of annual stroke discharges from each hospital.
Similar multivariable logistic regression analyses were performed to explore the relationship between diabetes status and 2 other binary outcome measures (ie, in-hospital mortality and discharge status [home vs other]). The relationship between diabetes status and length of stay (LOS) was explored using multivariable linear regression analyses using generalized estimating equations. Because the LOS variable was highly skewed, it was logarithm-transformed. We included the same set of prespecified potential confounders in all 3 of these outcomes-based models, and we chose not to adjust for differences in performance measures because of the inherent problem of confounding by indication (ie, the tendency for patients with inherently poorer prognosis to receive less care).
All tests are 2-tailed with P<0.05 considered as the level of statistical significance. All statistical analyses were performed using SAS version 9.1 software (SAS Institute).
Of 415 926 ischemic stroke admissions, less than one-third (31.5%; n=130 817) met the definition of diabetes (95.5% had a medical history of diabetes and 4.5% were using diabetes-related medications at admission). Table 1⇓ compares the demographic and clinical characteristics of ischemic stroke patients with and without diabetes. Patients with diabetes were younger (mean, 69.8 vs 71.6 years), with far fewer being 80 years or older, and were more likely to be male, black, or Hispanic (Table 1⇓). Patients with diabetes were more likely to have a medical history of stroke/TIA, coronary artery disease/previous MI, hypertension, and dyslipidemia, but they were less likely to be current smokers or to arrive within 2 hours of stroke onset (Table 1⇓).
Compliance with most individual ischemic stroke performance measures was similar between patients with and without diabetes (Figure). The largest discrepancy was seen in the intravenous recombinant tissue plasminogen activator (rt-PA) (<2 hours) measure: 54.0% of diabetes patients who arrived within 2 hours of onset and had no contraindications received rt-PA within 3 hours of onset compared to 60.8% of patients without diabetes. In contrast, cholesterol treatment was higher in patients with diabetes (79.3%) compared to those without diabetes (71.2%). Overall, patients with diabetes received defect-free care slightly more often compared to patients without diabetes (72.2% vs 69.0%; Table 2), which is primarily a reflection of the higher cholesterol treatment rates in this population.
After multivariable adjustment, the discrepancies observed between patients with and without diabetes in rt-PA and cholesterol treatment remained (Table 3). Compared to patients without diabetes, the adjusted OR [aOR] for receiving rt-PA was 0.83 (95% CI, 0.79–0.88) and the aOR for cholesterol treatment was 1.40 (95% CI, 1.37–1.44) among diabetes patients. Adjusted results for the other individual performance measures showed that patients with diabetes were slightly less likely to receive a given treatment compared to patients without diabetes, although these differences were quite modest (ie, aOR ranged between 0.91 and 0.93). However, the adjusted OR for the defect-free summary measure of care was higher in patients with diabetes (aOR, 1.06; 95% CI, 1.04–1.07; Table 3).
The results of the 6 diabetes-related quality metrics calculated among the 130 817 ischemic stroke patients with diabetes are shown in Table 4. Less than half of the patients with diabetes had their HbA1c measured while in the hospital, whereas only 70% underwent lipid testing. Compliance with discharge medications was high for any hypertensive medications (92.2%), any diabetes medications (80.3%), and cholesterol treatment (79.3%). However, <60% of diabetes patients with a history of hypertension were discharged on angiotensin-converting enzyme inhibitors or angiotensin receptor blockers.
Significant differences were observed in stroke-related outcomes between patients with and without diabetes. Patients with diabetes were slightly less likely to be discharged home and were more likely to be discharged to a skilled nursing facility or to inpatient acute rehabilitation (Table 1⇑). Patients with diabetes were also less able to ambulate independently at discharge. Because LOS was highly skewed, the mean LOS was slightly longer in patients with diabetes (6.2 days vs 5.6 days), but the median and interquartile ranges were identical (ie, 4 days and 3.0–7.0 days, respectively) in the 2 groups. The crude (unadjusted) in-hospital case fatality rates were identical in the 2 groups (5.8%; Table 1⇑).
Adjustment for potential confounding variables had an important impact on in-hospital case fatality; the unadjusted OR was 1.02 (95% CI, 0.99–1.05), but after adjustment the odds of in-hospital mortality were significantly elevated in patients with diabetes (aOR, 1.12; 95% CI, 1.08–1.15; Table 3) Adjusted analyses also showed that the odds of discharge home were 0.80 in patients with diabetes (Table 3). The multivariable linear regression analysis of the log-transformed LOS outcome indicated that patients with diabetes had a 1.07-fold longer in-hospital stay compared to patients without diabetes (adjusted ratio, 1.07; 95% CI, 1.07–1.08).
The results from this large-scale, hospital-based stroke registry showed that diabetes is common among hospitalized acute ischemic stroke patients—almost one-third of cases had diabetes diagnosed or treated. This figure is higher than in most previous reports,5–7,18 which may be a reflection of the increasing prevalence of diabetes in the United States.2 The characteristics of the diabetes patients included in this registry were similar to those found in a large, population-based, United States stroke study.3 Diabetes patients were younger, more likely to be black or Hispanic, and had a higher rates of previous cardiovascular disease, hypertension, and hypercholesterolemia.
With the exception of rt-PA and cholesterol treatment, we found that quality of care was similar between ischemic stroke patients with and without diabetes. The rt-PA measure included only patients who arrived within 2 hours and had no contraindications to treatment. A comparison of the relative frequencies of documented contraindications and warnings to rt-PA treatment between patients with and without diabetes found them to be almost identical (data not shown). Diabetes per se is not a contraindication to rt-PA treatment, although hyperglycemia is known to be a poor prognostic factor in stroke patients.19 Low (<50 mg/dL) or high (>400 mg/dL) glucose levels were listed as a reason for not administering rt-PA in only a minority of cases (0.7% and 0.03% of patients with and without diabetes, respectively). The lower rt-PA treatment rate in patients with diabetes may reflect physician concerns regarding the increased risk of intracerebral hemorrhage after rt-PA treatment in patients with diabetes or hyperglycemia.20,21 Reluctance to treat diabetes patients with rt-PA also could reflect physician awareness that stroke outcomes are worse in patients with diabetes.5,6 In a subgroup analysis of the NINDS rt-PA stroke trial, diabetes was associated with poorer outcomes, but there was no evidence that rt-PA was less effective in patients with diabetes.22 Severe stroke is also listed as a reason for no treatment; however, again, there was no difference between patients with and without diabetes in the GWTG-Stroke program (2.9% and 2.8%, respectively). Whether stroke events are more severe in patients with diabetes is still unclear.8 Although differences in stroke subtype are widely reported in patients with diabetes,6–8,18 a population-based study that measured initial stroke severity did not show it to be higher in diabetes patients.18 Unfortunately, stroke severity (as measured by the NIHSS) is incompletely documented in the GWTG-Stroke program, so we are unable to compare stroke severity in persons with and without diabetes.
Because we could find no rationale or justification for the less frequent use of rt-PA in patients with diabetes in our study, and because rt-PA treatment is associated with better long-term outcomes in ischemic stroke patients, including those with diabetes, greater quality-improvement efforts should be undertaken to increase rt-PA treatment in patients with diabetes. These efforts could focus on greater education of health care professionals concerning the guideline-based evidence for rt-PA use in patients with diabetes, as well as the poorer outcomes in this vulnerable population, which should increase, not decrease, rt-PA treatment rates. The use of the GWTG-Stroke program may be used to help facilitate such process of care efforts through the use of targeted clinical support tools, including feedback reports specific to diabetes patients.
Patients with diabetes had significantly higher rates of cholesterol treatment at discharge (Table 3). The performance measure for cholesterol treatment at discharge is carefully constructed to identify all subjects who are eligible for treatment. The measure denominator includes those subjects who are already using lipid-lowering treatment at admission, those subjects who are eligible for treatment based on LDL-C values >100 mg/dL, and those subjects who did not undergo lipid testing. Therefore, these criteria account for the fact that diabetes patients were more likely to be using lipid therapy at admission (49% of patients with diabetes were using lipid therapy at admission compared to 30% without diabetes), which would increase the number of patients with diabetes who had an indication for lipid therapy at discharge. However, these criteria also account for the fact that diabetes patients were no more likely to undergo in-hospital lipid testing than those without diabetes (ie, 75% vs 76%, respectively), and for the fact that diabetes patients were less likely to have high LDL levels (>100 mg/dL) than those without diabetes (ie, 46% vs 54%, respectively). This latter finding is likely a reflection of the higher rates of preadmission cholesterol therapy in patients with diabetes. Given the objective definitions used to define the eligible population for the cholesterol treatment performance measure, the fact that cholesterol treatment rates at discharge were higher in patients with diabetes warrants further explanation. These data may perhaps be a reflection of a greater awareness on behalf of the treating physicians regarding either the importance of cholesterol and atherosclerosis in diabetes patients7,8 or the knowledge of the greater efficacy of statins among patients with diabetes.23,23 Although a medical history of coronary heart disease or stroke/TIA was more common in patients with diabetes, these 2 comorbidities were not associated with higher cholesterol treatment rates at discharge, whereas the presence of diabetes was (data not shown).
Our study found that diabetes had important independent associations with adverse stroke outcomes. Although crude in-hospital mortality rates were similar between patients with and without diabetes, after adjusting for confounding variables we found that the odds of in-hospital mortality were 1.12-times higher in patients with diabetes (the change after adjustment is likely a reflection of the younger age of patients with diabetes). The finding that diabetes was associated with an increase in short-term mortality is concordant with some18,24 but not all studies.3,5,7 Patients with diabetes were also less likely to be discharged home, were more likely to be nonambulatory at discharge, and were more likely to have a slightly longer LOS. These observations fit with previous studies that showed that diabetes resulted in increased poststroke disability,6,25 longer LOS,26 and increased long-term mortality.3,5,24 It should be noted for the 3 outcomes-based analyses (mortality, discharge home, and LOS) that we chose not to adjust for any differences in the quality of care between patients with and without diabetes. We do not believe that it is appropriate to link the care patients received during their hospitalization with their immediate outcomes because of the inherent problem of “confounding by indication” (ie, the tendency for patients with inherently poorer prognosis to receive less defect-free care). Also, the 7 performance measures tracked by the GWTG-Stroke program are based on interventions that, although supported by substantial evidence, have not previously been directly linked with reduced in-hospital mortality in randomized trials.
We found evidence of several important deficiencies in the quality of diabetes-related care. Compliance was poorest with respect to HbA1c testing (only 45% were tested) and treatment with angiotensin-converting enzyme inhibitors or angiotensin receptor blockers medications (only 58% were treated). Although only 70% of diabetes patients underwent lipid testing while hospitalized, this was similar to the rate in patients without diabetes (72%). Overall, these results suggest there is considerable room for improvement in diabetes-related care during the hospitalization of acute ischemic stroke cases.
This study has several limitations. First, the GWTG-Stroke program is voluntary and the hospitals that participate tend to be larger, are teaching institutions, and have an interest in stroke and quality improvement. It is possible that stroke cases treated at GWTG hospitals represent a more severe case mix, which could influence our findings. However, the stroke cases included in GWTG-Stroke tend to be similar to other hospital-based stroke databases,12 and the diabetes patients in this registry were similar to those in other population-based stroke studies.3 Second, although hospitals are instructed to include all consecutive admissions or to take a random sample, these processes are not audited so the potential for selection bias exists. Third, several variables of potential importance to this study were not collected in the GWTG-Stroke program. These included blood glucose, blood pressure, presence of diabetes-related comorbidities, such as retinopathy or nephropathy, and admission source, including the place of residence before the stroke event. Thus, the impact of these data on quality of care or outcomes could not be assessed. Additionally, although the 6 diabetes-related quality metrics addressed specific guideline recommendations, they do not cover all of the standards relevant to diabetes care in stroke patients and have not been validated by GWTG-Stroke program. Fourth, the determination of diabetes was based on a medical history or the use of diabetes-related medications at admission; therefore, an unknown proportion of the patients classified as not having diabetes in the registry actually have unrecognized diabetes, and previous studies have estimated that up to 16% to 24% of stroke admissions have unrecognized diabetes.27 Finally, unfortunately, no data on stroke-related outcomes after discharge are currently collected in the GWTG-Stroke program; therefore, the longer-term impact of diabetes or of the differences in quality of care identified in this study on stroke-related outcomes cannot be determined.
In summary, among patients hospitalized with ischemic stroke, most stroke performance measures were applied with similar frequency to patients with and without diabetes. However, we did find an important difference in thrombolysis treatment and identified several potential opportunities for improvements in diabetes-related care. Given that almost one-third of ischemic stroke patients have diabetes, even modest differences in the quality of care and clinical outcomes can have important consequences at the population level. Further quality improvement efforts are needed to improve care for all stroke patients and to eliminate the differences in quality and outcomes among patients with diabetes.
Sources of Funding
The Get With the Guidelines Program (GWTG) is funded by the American Heart Association and the American Stroke Association. The program is also supported in part by unrestricted educational grants to the American Heart Association by Pfizer (New York, NY) and the Merck-Schering Plough Partnership (North Wales, PA), who did not participate in the design, analysis, or manuscript preparation, and did not require approval of this manuscript for submission.
Dr Reeves receives salary support from the Michigan Stroke Registry and serves as a member of the American Heart Association’s Get With the Guidelines Quality Improvement Subcommittee. Dr Vaidya has no disclosures to report. Dr Fonarow chairs the American Heart Association GWTG Steering Committee; serves as a consultant to Pfizer, Merck, Schering Plough, Bristol Myers Squibb, and Sanofi-Aventis; receives speaker honoraria from Pfizer, Merck, Schering Plough, Bristol Myers Squibb, and Sanofi-Aventis; and receives research support from Pfizer and National Institutes of Health. Dr Liang is a member of the Duke Clinical Research Institute, which serves as the American Heart Association GWTG data coordinating center. Dr Smith serves as a member of the Get With the Guidelines Science Subcommittee and receives research support from the NIH (NINDS R01 NS062028) and the Canadian Stroke Network, and received salary support from the Heart and Stroke Foundation of Canada and the Canadian Institute for Health Research. Mr Matulonis has no disclosures to report. Dr Olson is a member of the Duke Clinical Research Institute, which serves as the American Heart Association GWTG data coordinating center. Dr Schwamm serves as vice chair of the American Heart Association GWTG Steering Committee; serves as a consultant to the Research Triangle Institute, CryoCath, and the Massachusetts Department of Public Health; and has provided expert medical opinions in malpractice lawsuits regarding stroke treatment and prevention.
- Received November 10, 2009.
- Revision received December 31, 2009.
- Accepted January 5, 2010.
Centers for Disease Control and Prevention. National diabetes fact sheet: general information and national estimates on diabetes in the United States, 2007. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention; 2008.
Cowie CC, Rust KF, Byrd-Holt DD, Eberhardt MS, Flegal KM, Engelgau MM, Saydah SH, Williams DE, Geiss LS, Gregg EW. Prevalence of diabetes and impaired fasting glucose in adults in the U.S. population: National Health and Nutrition Examination Survey 1999–2002. Diabetes Care. 2006; 29: 1263–1268.
Kissela BM, Khoury J, Kleindorfer D, Woo D, Schneider A, Alwell K, Miller R, Ewing I, Moomaw CJ, Szaflarski JP, Gebel J, Shukla R, Broderick JP. Epidemiology of ischemic stroke in patients with diabetes: The Greater Cincinnati/Northern Kentucky Stroke Study. Diabetes Care. 2005; 28: 355–359.
Hankey GJ, Jamrozik K, Broadhurst RJ, Forbes S, Burvill PW, Anderson CS, Stewart-Wynne EG. Long-term risk of first recurrent stroke in the Perth community stroke study. Stroke. 1998; 29: 2491–2500.
Megherbi SE, Milan C, Minier D, Couvreur G, Osseby GV, Tilling K, Di Carlo A, Inzitari D, Wolfe CD, Moreau T, Giroud M. Association between diabetes and stroke subtype on survival and functional outcome 3 months after stroke: data from the European Biomed Stroke Project. Stroke. 2003; 34: 688–694.
Sacco RL, Adams R, Albers G, Alberts MJ, Benavente O, Furie K, Goldstein LB, Gorelick P, Halperin J, Harbaugh R, Johnston SC, Katzan I, Kelly-Hayes M, Kenton EJ, Marks M, Schwamm LH, Tomsick T. Guidelines for Prevention of Stroke in Patients with Ischemic Stroke or Transient Ischemic Attack—a Statement for Healthcare Professionals from the American Heart Association/American Stroke Association Council on Stroke—Co-Sponsored by the Council on Cardiovascular Radiology and Intervention—the American Academy of Neurology Affirms the Value of This Guideline. Stroke. 2006; 37: 577–617.
Adams RJ, Albers G, Alberts MJ, Benavente O, Furie K, Goldstein LB, Gorelick P, Halperin J, Harbaugh R, Johnston SC, Katzan I, Kelly-Hayes M, Kenton EJ, Marks M, Sacco RL, Schwamm LH, American Heart Association, American Stroke Association. Update to the AHA/ASA recommendations for the prevention of stroke in patients with stroke and transient ischemic attack. Stroke. 2008; 39: 1647–1652.
American Diabetes Association. Standards of medical care in diabetes—2008. Diabetes Care. 2008; 31: S12–S54.
Schwamm LH, Fonarow GC, Reeves MJ, Pan W, Frankel MR, Smith EE, Ellrodt G, Cannon CP, Liang L, Peterson E, Labresh KA. Get with the Guidelines-Stroke is associated with sustained improvement in care for patients hospitalized with acute stroke or transient ischemic attack. Circulation. 2009; 119: 107–115.
Reeves MJ, Arora S, Broderick JP, Frankel M, Heinrich JP, Hickenbottom S, Karp H, LaBresh KA, Malarcher A, Mensah G, Moomaw CJ, Schwamm L, Weiss P. Acute stroke care in the US: results from 4 pilot prototypes of the Paul Coverdell National Acute Stroke Registry. Stroke. 2005; 36: 1232–1240.
Bateman BT, Schumacher HC, Boden-Albala B, Berman MF, Mohr JP, Sacco RL, Pile-Spellman J. Factors associated with in-hospital mortality after administration of thrombolysis in acute ischemic stroke patients: an analysis of the nationwide inpatient sample 1999 to 2002. Stroke. 2006; 37: 440–446.
American Hospital Association (AHA). AHA Hospital Statistics 2007. Chicago, IL: American Hospital Association; 2007.
Kleinbaum D, Klein M. Logistic regression—A self learning text, II ed. New York, NY: Springer; 2002.
Jorgensen H, Nakayama H, Raaschou HO, Olsen TS. Stroke in patients with diabetes. The Copenhagen Stroke Study. Stroke. 1994; 25: 1977–1984.
Bruno A, Biller J, Adams HP Jr, Clarke WR, Woolson RF, Williams LS, Hansen MD. Acute blood glucose level and outcome from ischemic stroke. Trial of Org 10172 in Acute Stroke Treatment (Toast) Investigators. Neurology. 1999; 52: 280–284.
Demchuk AM, Morgenstern LB, Krieger DW, Linda Chi T, Hu W, Wein TH, Hardy RJ, Grotta JC, Buchan AM. Serum glucose level and diabetes predict tissue plasminogen activator-related intracerebral hemorrhage in acute ischemic stroke. Stroke. 1999; 30: 34–39.
Generalized efficacy of t-PA for acute stroke. Subgroup Analysis of the NINDS t-PA Stroke Trial. Stroke. 1997; 28: 2119–2125.
Tuomilehto J, Rastenyte D, Jousilahti P, Sarti C, Vartiainen E. Diabetes mellitus as a risk factor for death from stroke. Prospective study of the middle-aged Finnish population. Stroke. 1996; 27: 210–215.
Desmond DW, Moroney JT, Sano M, Stern Y. Recovery of cognitive function after Stroke. Stroke. 1996; 27: 1798–1803.
Williams LS, Rotich J, Qi R, Fineberg N, Espay A, Bruno A, Fineberg SE, Tierney WR. Effects of admission hyperglycemia on mortality and costs in acute ischemic stroke. Neurology. 2002; 59: 67–71.
Gray CS, Scott JF, French JM, Alberti KG, O'Connell JE. Prevalence and prediction of unrecognised diabetes mellitus and impaired glucose tolerance following acute stroke. Age Ageing. 2004; 33: 71–77.