- Email: [email protected]
Diabetes Mellitus with Chronic Complications in Relation to Carotid Endarterectomy and Carotid Artery Stenting Outcomes
ARTICLE IN PRESS
Diabetes Mellitus with Chronic Complications in Relation to Carotid Endarterectomy and Carotid Artery Stenting Outcomes Oluwole Adegbala,
MPH,*†
Kimberly D. Martin, PhD,† David Otuada, Tomi Akinyemiju, PhD†
PhD,‡
and
Background: Carotid endarterectomy and carotid artery stenting are effective treatment procedures for carotid artery stenosis. Although diabetes mellitus is highly prevalent among patients undergoing these revascularization procedures, few studies have examined their impact on periprocedural outcomes. Objectives: The study aimed to determine whether perioperative outcomes among patients undergoing carotid artery stenting and carotid endarterectomy varied depending on the presence of diabetes with or without chronic complications. Methods: We examined adults aged 45 and above hospitalized between 2007 and 2011 in U.S. hospitals who underwent carotid artery revascularization procedures. We used data from the Healthcare Cost and Utilization Project’s Nationwide Inpatient Sample and evaluated the influence of diabetes with or without chronic complications on outcomes. Results: Among patients receiving carotid artery stenting, diabetic patients with chronic complications had significantly increased odds of acute kidney injury (odds ratio [OR]: 3.17, 95% confidence interval [CI]: 2.31-4.35) and longer hospital stay (β: 1.98, 95% CI: 1.582.38) compared with nondiabetic patients. Diabetic patients with chronic complications receiving carotid endarterectomy experienced increased odds of myocardial infarction (OR: 1.12, 95% CI: .90-1.40), stroke (OR: 1.29, 95% CI: .97-1.72), perioperative infection (OR: 2.45, 95% CI: 1.29-4.65), mortality (OR: 1.48, 95% CI: 1.01-2.16), and longer hospital stay (β (days): 2.05, 95% CI: 1.90-2.20) compared with nondiabetic patients. No significant increased odds of perioperative outcomes were observed among diabetic patients without chronic complications. Conclusions: Uncomplicated diabetes did not appear to convey a higher odds of perioperative outcomes among patients undergoing revascularization. However, the presence of diabetes with chronic complications is an important risk factor in the carotid endarterectomy category. Key Words: Diabetes mellitus—carotid endarterectomy—carotid artery—stenting—mortality—hospitalization outcomes. © 2016 National Stroke Association. Published by Elsevier Inc. All rights reserved.
Introduction From the *Hugh Kaul Personalized Medicine Institute, University of Alabama at Birmingham, Birmingham, Alabama; †Department of Epidemiology, University of Alabama at Birmingham, Birmingham, Alabama; and ‡Department of Emergency Medicine, University of Alabama at Birmingham, Birmingham, Alabama. Received June 14, 2016; revision received August 16, 2016; accepted September 10, 2016. Address correspondence to Tomi Akinyemiju, PhD, Department of Epidemiology, University of Alabama, Birmingham, AL. E-mail: [email protected]. 1052-3057/$ - see front matter © 2016 National Stroke Association. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2016.09.012
Stroke is the leading cause of disability and the fifth leading cause of death in the United States, with close to 1 million new cases every year.1 More than 85% of all strokes are ischemic in origin, with approximately 15% of those attributable to carotid artery stenosis.2 Large randomized trials have established carotid endarterectomy (CEA) as the standard secondary preventive treatment for symptomatic severe carotid artery stenosis.3,4 In addition, revascularization with CEA is recommended for most medically stable patients with asymptomatic carotid stenosis of 60%-99% and for a subgroup of patients with asymptomatic carotid stenosis who have a particularly
Journal of Stroke and Cerebrovascular Diseases, Vol. ■■, No. ■■ (■■), 2016: pp ■■–■■
1
ARTICLE IN PRESS O. ADEGBALA ET AL.
2 5
high risk of stroke. Carotid artery stenting (CAS) has increasingly gained recognition as a safe and efficacious alternative treatment to CEA, especially in patients considered to be at a high surgical risk for CEA.6-9 Diabetes mellitus (DM) is a major independent risk factor for cardiovascular disease and stroke, and the prevalence is higher among those undergoing carotid revascularization.10,11 The effect of DM on CEA outcomes is equivocal; a number of studies have suggested increased perioperative risks of stroke and death during CEA in this subgroup of patients,8,12,13 whereas other studies have documented similar perioperative outcomes in both diabetic patients and nondiabetic patients undergoing the surgical procedure.14-17 Although there are conflicting reports on the impact of DM on the outcomes of patients undergoing CEA procedures, the impact of diabetes on the outcomes of patient undergoing CAS remains largely unclear. Given the growing emergence of CAS into clinical practice,18 and the increasing prevalence of diabetes as a chronic medical condition among U.S. adults—with 1.4 million new diagnoses every year19,20—we aimed to determine whether the periprocedural outcomes of patients undergoing CAS and CEA differed depending on the presence of DM with or without chronic complications in a large nationally representative sample of hospitalized patients.
Methods Study Population This was a cross-sectional study of patients aged 45 and above hospitalized for a carotid artery revascularization procedure in a U.S. hospital between 2007 and 2011. Inpatient procedure and outcomes data were obtained from the Healthcare Cost and Utilization Project’s Nationwide Inpatient Sample (HCUP NIS). HCUP is the largest publicly available all-payer inpatient healthcare database in the United States, with data on more than 7 million hospital stays each year, representing more than 35 million hospitalizations nationally, including discharges from community hospitals participating in HCUP, excluding rehabilitation and long-term acute care hospitals.21 All patients undergoing carotid revascularization for carotid artery stenosis with either CAS (International Classification of Disease, 9th Revision, Clinical Modification [ICD-9-CM] 00.63) or CEA (ICD-9-CM 38.12) procedures were identified. Patients were classified as symptomatic if they carried a diagnosis of amaurosis fugax (ICD-9-CM 362.34 and 368.12), transient ischemic attack (ICD-9-CM 435.9 and 781.4), and stroke (ICD-9-CM 433.11, 433.31, 433.91, 434.01, 434.11, and 434.91). Diabetic patients were identified as those with an established DM comorbid condition with or without complications at the time of carotid procedures (ICD-9-CM 250). Diabetes with chronic complications was derived from the Agency for Healthcare Research and Quality comorbidity software, and included ICD-9
codes 250.40-250.93, representing diabetic patients with renal, ophthalmic, and neurologic manifestations, peripheral circulatory disorders, and other specified or unspecified complications. Demographic information including age on admission, gender, and race was obtained. Racial and ethnic groups were categorized into the following four groups—non-Hispanic white, Hispanic, non-Hispanic black, and other groups (including American Indians and Asians). We created a modified Deyo comorbidity index using ICD-9 codes for major comorbid conditions. The conditions included hypertension, congestive heart failure, obesity, chronic pulmonary disease, peripheral vascular disease, and chronic renal failure disease. We identified the presence of each condition within each patient and summed up the number of conditions per patient into a single comorbidity score.
Outcome Measures In-hospital primary outcomes included new events occurring during or after the procedure but before discharge. We determined the presence of post-procedural complications of CAS and CEA procedures using ICD-9 codes for in-hospital postoperative stroke (997.02), pulmonary embolism and infarction (415.11), hematoma (998.12), postoperative infection (998.59), renal dysfunction (997.5), postoperative myocardial infarction or cardiac complications (997.1), and in-hospital mortality which was available directly in the NIS dataset and defined as death due to any cause at any point in time during the index hospitalization. The secondary outcome measures for this study included length of stay during the index hospital admission stay, defined as the number of days between admission and discharge, with same day hospitalizations coded as 0.
Statistical Analysis Descriptive statistics were used to compare demographic and clinical characteristics of patients by DM status, using chi-square for categorical variables and t-tests or analysis of variance for continuous variables. Logistic regression analysis was used to compute the association between DM and post-procedural complications and in-hospital mortality for the two procedures combined and for each revascularization procedure separately. Stratified analysis was conducted to examine the impact of DM with complications and DM without complications on study outcomes. Covariates included gender, age, presentation type, year of the procedure, hospital teaching status, hospital bed size, and modified Deyo comorbidity index. Statistical analyses were performed using SAS software (v9.4, Cary, NC).
Results Demographic and baseline characteristics for 128,664 patients who underwent revascularization procedures between 2007 and 2011 are presented in Table 1. The
Table 1. Descriptive analysis of patients admitted to the hospital between 2007 and 2011 with carotid artery revascularization procedures CAS N = 16,306
Diabetic N = 41,120
Nondiabetic N = 87,544
P value
Diabetic N = 35,924
Nondiabetic N = 76,608
P value
Diabetic N = 5246
Nondiabetic N = 11,060
P value
70.37 ± 8.95 3.24 ± 4.63 17,227 (41.90) 4146 (10.08)
71.49 ± 9.61 2.79 ± 4.38 36,505 (41.73) 8891 (10.16)
.0001 .0001 .567 .684
70.41 ± 8.92 3.22 ± 4.53 15,139 (42.14) 3351 (9.33)
71.55 ± 9.54 2.77 ± 4.38 32,163 (42.02) 7328 (9.57)
.0001 .0001 .686 .205
70.07 ± 9.16 3.34 ± 5.27 2111 (40.24) 801 (15.27)
71.04 ± 10.04 2.91 ± 4.44 4401 (39.80) 1592 (14.39)
.0001 .0001 .588 .140
28,362 (68.97) 2065 (5.02) 2113 (5.14) 8580 (20.87)
65,235 (74.52) 2496 (2.85) 2404 (2.75) 17,409 (19.89)
.0001
24,926 (69.39) 1753 (4.88) 1808 (5.03) 7437 (20.70)
57,228 (74.70) 2081 (2.72) 2065 (2.70) 15,234 (19.89)
.0001
3470 (66.15) 314 (5.99) 307 (5.85) 1155 (22.02)
8105 (73.28) 422 (3.82) 340 (3.07) 2193 (19.83)
.0001
35,583 (86.53) 4035 (9.81) 5522 (13.43) 10,224 (24.86) 8811 (21.43) 5169 (12.57)
67,555 (77.17) 4789 (5.47) 5848 (6.68) 19,007 (21.71) 19,431 (22.20) 4305 (4.92)
.0001 .0001 .0001 .0001 .0019 .0001 .0001
31,134 (86.67) 3285 (9.14) 4637 (12.91) 8577 (23.88) 7684 (21.39) 4585 (12.76)
9669 (77.89) 3889 (5.08) 4965 (6.48) 16,005 (20.89) 17,103 (22.33) 3883 (5.07)
.0001 .0001 .0001 .0001 .0004 .0001
4493 (85.65) 755 (14.39) 894 (17.04) 1666 (31.76) 1139 (21.71) 591 (11.27)
7978 (72.13) 912 (8.25) 893 (8.07) 3057 (27.64) 2359 (21.33) 427 (3.86)
.0001 .0001 .0001 .0001 .578 .0001 .0001
2617 (6.36) 30,822 (74.96) 7681 (44.01) 207 (.50)
12,329 (14.08) 65,444 (74.76) 9771 (11.16) 457 (.52)
2319 (6.46) 27,138 (75.54) 6467 (18.00) 163 (.45)
10,546 (13.77) 57,713 (75.34) 8349 (10.90) 341 (.45)
.0001
299 (5.70) 3723 (70.97) 1224 (23.33) 45 (.86)
1796 (16.24) 1823 (70.73) 1441 (13.03) 121 (1.09)
29,843 (71.90) 1533 (3.73) 8360 (20.33) 1384 (3.37)
62,946 (71.90) 2353 (2.69) 19,274 (22.02) 2971 (3.39)
.0001
26,101 (72.66) 1282 (3.57) 7363 (20.50) 1178 (3.28)
55,167 (72.01) 1963 (2.56) 16,943 (22.12) 2535 (3.31)
.0001
3778 (72.02) 255 (4.86) 1006 (19.18) 207 (3.95)
7859 (71.06) 397 (3.59) 2356 (21.30) 448 (4.05)
.0001
11,347 (28.16) 11,919 (29.58) 9668 (23.99) 7358 (18.26)
21,794 (25.42) 25,133 (29.32) 21,485 (25.06) 17,320 (20.20)
.0001
9797 (27.83) 10,531 (29.91) 8438 (23.97) 6443 (18.30)
18,901 (25.20) 22,238 (29.65) 18,850 (25.13) 15,023 (20.03)
.0001
1565 (30.49) 1402 (27.31) 1241 (24.18) 925 (18.02)
2937 (27.09) 2920 (26.93) 2671 (24.64) 2314 (21.34)
.0001
.665
.839
.160
3
Abbreviations: CAS, carotid artery stenting; CEA, carotid endarterectomy; COPD, chronic obstructive pulmonary disease; CHF, congestive heart failure; CKD, chronic kidney disease; HTN, hypertension; PVD, peripheral vascular disease. *Others include American Indians and Asians. †Comorbidity score. ‡In-hospital mortality. §Residential income by county. ‖Chronic conditions.
ARTICLE IN PRESS
Age Length of stay Female Symptoms Race White Black Hispanic Others* C. cond.‖ HTN CHF CKD PVD COPD Obesity Comorbidities† 0 1-2 >2 Hospital mortality‡ Insurance type Medicare Medicaid Private Others Residential income§ Q1 Q2 Q3 Q4
CEA N = 112,532
DIABETES MELLITUS AND CAROTID ARTERY STENOSIS OUTCOMES
All Patients N = 128,664
ARTICLE IN PRESS O. ADEGBALA ET AL.
4
majority of patients, 112,532 (87.5%), received CEA, whereas 16,306 (12.5%) received CAS. The mean age upon admission was 70.37 ± 8.95 years and 41.9% were women. Nearly three quarters of the patients were white. African– Americans, and Hispanics comprised about 4% each of the study sample. About 80% had hypertension, nearly 25% had peripheral vascular disease and chronic obstructive lung disease, and about 10% had chronic renal failure. About 32% of the population had DM. Patients with DM tended to be younger (70.37 ± 8.95 versus 71.49 ± 9.61 years) and had longer hospital stays compared with nondiabetics. Patients with DM were more likely to be black, obese, hypertensive, and with congestive heart failure and chronic renal failure. The prevalence of DM tended to increase with decreasing residential income. The proportion of diabetic versus nondiabetic patients was similar in both CAS and CEA groups, about 32% each. Among patients receiving CAS, diabetic patients were younger, more likely to have symptomatic presentation, history of congestive heart failure, hypertension, peripheral artery disease, and chronic renal failure, but less likely to be in the highest level residential income compared with nondiabetic populations. Similar trends were observed for CEA patients. Table 2 shows the results of the association between DM with in-hospital outcomes among patients receiving both CAS and CEA. Diabetic patients with chronic complications experienced significantly increased odds of postoperative stroke (odds ratio [OR]: 1.44, 95% confidence interval [CI]: 1.131.84), hematoma (OR: 1.37, 95% CI: 1.15-1.64), infection (OR: 2.84, 95% CI: 1.60-5.03), pneumonia (OR: 2.73, 95% CI: 1.61-4.64), acute kidney injury (OR: 6.02, 95% CI: 5.37-
6.76), and in-hospital mortality (OR:1.64, 95% CI: 1.162.31) compared with nondiabetic patients after adjusting for baseline study covariates such as age, gender, race, insurance type, and residential income. Diabetic patients with chronic complications had 64% higher odds of in-hospital mortality (OR: 1.64, 95% CI: 1.16-2.31) and significantly longer hospital stay (β: 2.45, 95% CI: 2.31-2.59) compared with nondiabetic patients. These associations became attenuated after adjusting for modified Deyo comorbidity index; however, length of hospital stay (β: 2.05, 95% CI: 1.90-2.19), postoperative stroke (OR: 1.36, 95% CI: 1.06-1.75), post-procedural infection (OR: 1.20, 95% CI: 1.01-1.44), hematoma (OR: 2.59, 95% CI: 1.43-4.70), and odds of acute kidney injury (OR: 3.45, 95% CI: 3.05-3.90) remained significantly higher compared with patients without DM in the fully adjusted model. There was no significant difference in the risk of myocardial infarction, other cardiac complications, ischemic stroke, or periprocedural mortality by DM with chronic complications status. Diabetic patients without chronic complications experienced slightly reduced odds of myocardial infarction (OR: .84, 95% CI: .78-.94), hematoma (OR: .88, 95% CI: .81-.96), and in-hospital mortality (OR: .80, 95% CI: .67-.96) compared with patients without DM; however, there were no statistically significant differences in length of hospital stay. Similar results were observed when the analysis was restricted to patients who received the CAS procedure only (Table 3). Diabetic patients with chronic complications had significantly increased odds of hematoma (OR: 1.70, 95% CI: 1.08-2.70) and acute kidney injury (OR: 5.35, 95% CI: 3.97-7.21) after adjusting for baseline study covariates such as age, gender, race, insurance type, and
Table 2. Multivariate adjusted odds ratios for the association between diabetes mellitus and in-hospital outcomes among patients receiving both CAS and CEA Diabetes mellitus (without chronic complications)
Both procedures MI or cardiac complications Postoperative stroke Hematoma Postoperative infection Pulmonary embolism Pneumonia Acute kidney injury Length of hospital stay In-hospital mortality
Diabetes mellitus (with chronic complications)
Model 1* AOR/β (95% CI)
Model 2† AOR/β (95% CI)
Model 1* AOR/β (95% CI)
Model 2† AOR/β (95% CI)
.89 (.81-.97) .94 (.83-1.06) .91 (.84-.99) 1.12 (.78-1.60) 1.29 (.51-3.30) .99 (.71-1.38) 1.28 (1.18-1.40) .13 (.08-.19) .86 (.72-1.03)
.84 (.78-.94) .91 (.81-1.03) .88 (.81-.96) 1.10 (.76-1.57) 1.21 (.47-3.13) .86 (.61-1.20) 1.06 (.97-1.16) .01 (−.04-.06) .80 (.67-.96)
1.27 (1.03-1.56) 1.44 (1.13-1.84) 1.37 (1.15-1.64) 2.84 (1.60-5.03) 3.03 (.66-13.91) 2.73 (1.61-4.64) 6.02 (5.37-6.76) 2.45 (2.31-2.59) 1.64 (1.16-2.31)
1.10 (.89-1.34) 1.36 (1.06-1.75) 1.20 (1.01-1.44) 2.59 (1.43-4.70) 2.05 (.42-10.05) 1.60 (.92-2.78) 3.45 (3.05-3.90) 2.05 (1.90-2.19) 1.26 (.88-1.79)
Abbreviations: AOR, adjusted odds ratios; β, as regression coefficient indicating number of days; CAS, carotid artery stenting; CEA, carotid endarterectomy; CI, confidence interval; MI, myocardial infarction. Bold indicates significance at P value ≤ .05. *Adjusted for race, sex, age, insurance type, and median household income national quartile for patient zip code. †Adjusted for race, sex, age, insurance type, median household income national quartile for patient zip code, and comorbidity score.
ARTICLE IN PRESS DIABETES MELLITUS AND CAROTID ARTERY STENOSIS OUTCOMES
5
Table 3. Multivariate adjusted odds ratios for the association between diabetes mellitus and in-hospital outcomes among patients receiving CAS Diabetes mellitus (without chronic complications)
CAS MI or cardiac complications Postoperative stroke Hematoma Postoperative infection Pneumonia Acute kidney injury Length of hospital stay In-hospital mortality
Diabetes mellitus (with chronic complications)
Model 1* OR/β (95% CI)
Model 2† OR/β (95% CI)
Model 1* AOR/β (95% CI)
Model 2† AOR/β (95% CI)
.98 (.76-1.25) .87 (.67-1.12) .77 (.59-.99) .24 (.03-1.91) .76 (.36-1.63) 1.32 (1.08-1.64) .13 (−.03-.29) .81 (.57-1.16)
.98 (.76-1.26) .87 (.67-1.13) .71 (.55-.92) .29 (.04-2.33) .63 (.29-1.35) 1.03 (.83-1.28) −.003 (−.16-.16) .77 (.54-1.11)
.88 (.46-1.68) 1.43 (.86-2.37) 1.70 (1.08-2.70) 3.67 (.79-17.21) 2.27 (.67-7.65) 5.35 (3.97-7.21) 2.29 (2.00-2.68) .69 (.25-1.87)
.91 (.47-1.75) 1.51 (.90-2.52) 1.44 (.90-2.30) 3.86 (.76-19.59) 1.45 (.41-5.09) 3.17 (2.31-4.35) 1.98 (1.58-2.38) .55 (.20-1.52)
Abbreviations: AOR, adjusted odds ratios; β, as regression coefficient indicating number of days; CAS, carotid artery stenting; CI, confidence interval; MI, myocardial infarction. Bold indicates significance at P value ≤ .05. *Adjusted for race, sex, age, insurance type, and median household income national quartile for patient zip code. †Adjusted for race, sex, age, insurance type, median household income national quartile for patient zip code, and comorbidity score.
residential income. In addition, those patients had significantly longer hospital stay (β: 2.29, 95% CI: 2.002.68) compared with patients without DM. These associations were attenuated after adjusting for baseline chronic conditions, with only the length of hospital stay (β: 1.98, 95% CI: 1.58-2.38) and acute kidney injury (OR: 3.17, 95% CI: 2.31-4.35) remaining significantly higher compared with patients without DM in the fully adjusted model. Diabetic patients without chronic complications experienced slightly reduced odds of hematoma (OR: .71, 95% CI: .55-.92) and reduced hospital stay compared with
patients without DM; however, there were no statistically significant differences in in-hospital mortality, myocardial infarction, or postoperative stroke outcomes. With regard to the CEA procedure (Table 4), there were significantly increased odds of myocardial infarction (OR: 1.33, 95% CI: 1.07-1.65), postoperative stroke (OR: 1.41, 95% CI: 1.06-1.86), hematoma (OR: 1.32, 95% CI: 1.091.59), infection (OR: 2.74, 95% CI: 1.48-5.07), pneumonia (OR: 2.90, 95% CI: 1.61-5.24), acute kidney injury (OR: 6.14, 95% CI: 5.42-6.96), and in-hospital mortality (OR:1.95, 95% CI: 1.35-2.82) associated with DM with chronic
Table 4. Multivariate adjusted odds ratios for the association between diabetes mellitus and in-hospital outcomes among patients receiving CEA Diabetes mellitus (without chronic complications)
CEA MI or cardiac complications Postoperative stroke Hematoma Postoperative infection Pulmonary embolism Pneumonia Acute kidney injury Length of hospital stay In-hospital mortality
Diabetes mellitus (with chronic complications)
Model 1* AOR/β (95% CI)
Model 2† AOR/β (95% CI)
Model 1* AOR/β (95% CI)
Model 2† AOR/β (95% CI)
.88 (.79-.97) .94 (.83-1.08) .93 (.85-1.01) 1.22 (.85-1.76) 1.19 (.44-3.23) 1.06 (.73-1.54) 1.28 (1.16-1.40) .13 (.08-.19) .88 (.72-1.08)
.83 (.75-.92) .91 (.80-1.05) .90 (.82-.98) 1.18 (.81-1.71) 1.09 (.40-3.00) .93 (.64-1.35) 1.06 (.97-1.17) .01 (−.04-.07) .81 (.66-1.00)
1.33 (1.07-1.65) 1.41 (1.06-1.86) 1.32 (1.09-1.59) 2.74 (1.48-5.07) 1.53 (.19-12.24) 2.90 (1.61-5.24) 6.14 (5.42-6.96) 2.47 (2.31-2.62) 1.95 (1.35-2.82)
1.12 (.90-1.40) 1.29 (.97-1.72) 1.16 (.95-1.41) 2.45 (1.29-4.65) 1.06 (.13-8.96) 1.67 (.90-3.10) 3.49 (3.05-3.99) 2.05 (1.90-2.20) 1.48 (1.01-2.16)
Abbreviations: AOR, adjusted odds ratios; β, as regression coefficient indicating number of days; CEA, carotid endarterectomy; CI, confidence interval; MI, myocardial infarction. Bold indicates significance at P value ≤ .05. *Adjusted for race, sex, age, insurance type, and median household income national quartile for patient zip code. †Adjusted for race, sex, age, insurance type, median household income national quartile for patient zip code, and comorbidity score.
ARTICLE IN PRESS O. ADEGBALA ET AL.
6
complications, after adjusting for baseline study covariates such as age, gender, race, insurance type, and residential income. In addition, those patients had 95% higher odds of in-hospital mortality (OR: 1.95, 95% CI: 1.352.82) and had significantly longer hospital stay (β: 2.47, 95% CI: 2.31-2.62) compared with patients without DM. These associations diminished after adjusting for baseline chronic conditions, and only postoperative infection (OR: 2.45, 95% CI: 1.29-4.65), length of hospital stay (β: 2.05, 95% CI: 1.90-2.20), acute kidney injury (OR: 3.49, 95% CI: 1.69-2.26), and in-hospital mortality (OR: 1.48, 95% CI: 1.01-2.16) remained significantly higher compared with patients without DM in the fully adjusted model. Diabetic patients without chronic complications experienced slightly reduced odds of myocardial infarction (OR: .83, 95% CI: .75-.92) and hematoma (OR: .90, 95% CI: .82-.98) compared with patients without DM; however, there were no statistically significant differences in in-hospital mortality, acute kidney injury, and postoperative stroke.
Discussion Previous studies have shown that DM is associated with increased risk of ischemic stroke, and stroke-related mortality and functional limitations are higher in diabetic patients.10,11,22 Therefore, diabetic patients with carotid artery stenosis may derive considerable benefits from carotid revascularization procedures. The benefits of these procedures, however, should outweigh the risks of perioperative complications. In order to minimize perioperative outcomes in diabetic patients, it is important to identify whether there are significantly increased odds of perioperative outcomes, as this information will be critical in assessing the health benefits of such procedures in this population subgroup. Unfortunately, the literature is not particularly clear on the potentially increased risks that DM poses on carotid revascularization procedures. Several previous studies have demonstrated that diabetic patients undergoing CEA have a higher risk of cardiovascular events than nondiabetic patients.10,11 The literature remains divided on this topic, however. A number of investigators8,12,13,23-25 found that diabetic patients had a significantly higher incidence of death and neurologic or cardiovascular complications after CEA. In one of these studies,25 there was no significant difference in neurologic postoperative morbidity between 341 diabetic patients and 2281 nondiabetic patients, although diabetic patients had a higher 30-day mortality that was mainly cardiacrelated. The increased mortality in this study was attributed to the higher frequency of surgery in diabetic patients for minor stroke, which had a higher perioperative complication rate than other clinical presentations or asymptomatic stenosis. Similarly, in a large series of studies, investigators reported that the presence of insulintreated DM was the strongest independent risk factor for
23,24
cardiac events and death. The increased perioperative risks in diabetic patients on insulin therapy may indicate the presence of chronic complications that are often experienced by type 2 diabetic patients on insulin therapy.26 In contrast to the above findings, other studies observed no significant differences in perioperative stroke, myocardial infarction, mortality, or long-term survival between diabetic and nondiabetic patients undergoing CEA.14-16 In the present analysis, among patients who underwent CAS, there were no significant differences in the risk of myocardial infarction, other cardiac complications, ischemic stroke, or periprocedural mortality. This is similar to the pooled analysis of outcomes from randomized clinical trials of 3454 symptomatic patients27 and analysis of 2196 procedures,8 which found no higher perioperative risk of major adverse cardiac events in diabetic patients who underwent CAS procedure. In contrast, other studies found that diabetic patients were at significantly increased risks of stroke, transient ischemic attack, and death within 30 days.28,29 The conflicting results may be due to significant heterogeneity in provider skills or experience, study population characteristics, measure of diabetes used in the study, and possible inadequate control for other confounders such as the presence of other comorbid conditions, and physician and hospital characteristics. With CEA, diabetic patients with chronic complications were more likely to experience myocardial infarction and postoperative stroke. These findings are comparable with previous studies,12,13 including the recent study of 1116 CEA procedures that found that the perioperative risk of stroke and death was threefold higher (OR: 2.83, 95% CI: 1.05-7.61, P = .04) in diabetic patients as opposed to nondiabetic patients.8 Similarly, in a more recent study, insulin-dependent DM was identified as a high-risk profile significantly impacting on survival of patients who underwent CEA for asymptomatic stenosis.30 In contrast, other studies16 observed no significant differences in perioperative stroke, MI, mortality, or long-term survival between diabetic and nondiabetic patients undergoing CEA. Provider characteristics and patient characteristics including lack of stratification by DM with complications may explain the differing results. Patients with DM are known to be at increased risk of silent myocardial disease,31,32 which could partly account for the lack of observed increase in clinically diagnosed MI and other cardiac complications in diabetic patients undergoing CAS procedure. In addition, DM with or without chronic complications does not confer significantly increased perioperative mortality in CAS population despite the increased incidence of comorbid conditions in the diabetes population. However, in the CEA population, there was a significantly increased risk of MI and other cardiac complications and increased mortality, which was attenuated after adjusting for comorbid conditions. This current study adds support to the previous studies in finding no significant differences in major outcome
ARTICLE IN PRESS DIABETES MELLITUS AND CAROTID ARTERY STENOSIS OUTCOMES
parameters between diabetic and nondiabetic patients undergoing CAS revascularization procedures. 8,27 We confirmed here that although DM with chronic complications has slightly increased risks of MACE periprocedural complications in those who underwent CEA, DM with or without chronic complications does not confer increased periprocedural MACE risk in CAS population. This suggests that patients’ stratification based on the presence of DM with complications may improve the identification of subgroups better suited for the CEA revascularization procedures and who may require targeted interventions to reduce the risk of perioperative complications. This study is not without its limitations. These include lack of data on post-hospitalization outcomes, with the stroke or death events only reflecting inpatient events and not including 30-day stroke or death rates. Also, detailed clinical data on stroke severity, subtype or side of the brain affected, management or specifics of DM chronic complications, actual degree of carotid stenosis, the device characteristics, perioperative medications, and history of coronary artery diseases are lacking. The strengths of this study include its large sample size and detailed administrative claims data that were nationally representative. We controlled for hospital teaching status and bed size as proxies for provider skills and experience, and used clinical diagnoses of both surgical procedures and DM status to improve reliability of measures. This is the first study, to our knowledge, to examine the influence of diabetes, including the presence of chronic complications, on revascularization outcomes. In conclusion, although DM overall does not appear to convey higher risks of perioperative outcomes on patients undergoing revascularization, our data suggest that the presence of DM with chronic complications is an important risk factor for poorer outcomes in CEA category. Future studies will be needed to identify the specific mechanism through which DM with complications increases perioperative risks, with the ultimate goal of minimizing perioperative complications in this group of patients.
References 1. Mozaffarian D, Benjamin EJ, Go AS, et al. Heart disease and stroke statistics-2016 update: a report from the American Heart Association. Circulation 2016;133:e38-e60. 2. Boulanger M, Touze E. Periprocedural risk of myocardial infarction after carotid endarterectomy and carotid angioplasty and stenting. Arch Cardiovasc Dis 2016;109:159-162. 3. North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. N Engl J Med 1991;325:445-453. 4. Randomised trial of endarterectomy for recently symptomatic carotid stenosis: final results of the MRC European Carotid Surgery Trial (ECST). Lancet 1998;351: 1379-1387.
7
5. Mohler ER III, Fairman RM. Management of asymptomatic carotid atherosclerotic disease. https://www .uptodate.com/contents/management-of-asymptomatic -carotid-atherosclerotic-disease?source=search_result &search=Mohler%20ER%20III,%20Fairman%20RM.%20 Management%20of%20asymptomatic&selectedTitle=1~150. UpToDate, November, 2015. 6. Goodney PP, Travis LL, Malenka D, et al. Regional variation in carotid artery stenting and endarterectomy in the Medicare population. Circ Cardiovasc Qual Outcomes 2010;3:15-24. 7. Jalbert JJ, Nguyen LL, Gerhard-Herman MD, et al. Outcomes after carotid artery stenting in Medicare beneficiaries, 2005 to 2009. JAMA Neurol 2015;72:276-286. 8. Parlani G, De Rango P, Cieri E, et al. Diabetes is not a predictor of outcome for carotid revascularization with stenting as it may be for carotid endarterectomy. J Vasc Surg 2012;55:79-89. 9. Timaran CH, Mantese VA, Malas M, et al. Differential outcomes of carotid stenting and endarterectomy performed exclusively by vascular surgeons in the Carotid Revascularization Endarterectomy versus Stenting Trial (CREST). J Vasc Surg 2013;57:303-308. 10. Kannel WB, McGee DL. Diabetes and cardiovascular disease. The Framingham study. JAMA 1979;241:20352038. 11. Kuusisto J, Mykkänen L, Pyörälä K, et al. Non-insulindependent diabetes and its metabolic control are important predictors of stroke in elderly subjects. Stroke 1994;25: 1157-1164. 12. Aziz I, Lewis RJ, Baker JD, et al. Cardiac morbidity and mortality following carotid endarterectomy: the importance of diabetes and multiple Eagle risk factors. Ann Vasc Surg 2001;15:243-246. 13. Dorigo W, Pulli R, Pratesi G, et al. Early and long-term results of carotid endarterectomy in diabetic patients. J Vasc Surg 2011;53:44-52. 14. Ballotta E, Da Giau G, Renon L. Is diabetes mellitus a risk factor for carotid endarterectomy? A prospective study. Surgery 2001;129:146-152. 15. Hamdan AD, Saltzberg SS, Sheahan M, et al. Lack of association of diabetes with increased postoperative mortality and cardiac morbidity: results of 6565 major vascular operations. Arch Surg 2002;137:417-421. 16. Pistolese GR, Appolloni A, Ronchey S, et al. Carotid endarterectomy in diabetic patients. J Vasc Surg 2001; 33:148-154. 17. Rockman CB, Saltzberg SS, Maldonado TS, et al. The safety of carotid endarterectomy in diabetic patients: clinical predictors of adverse outcome. J Vasc Surg 2005; 42:878-883. 18. Akbari CM, Pomposelli FB Jr, Gibbons GW, et al. Diabetes mellitus: a risk factor for carotid endarterectomy? J Vasc Surg 1997;25:1070-1075, discussion 5-6. 19. Centers for Disease Control and Prevention. National Diabetes Statistics Report: Estimates of Diabetes and Its Burden in the United States, 2014, Atlanta, GA: U.S. Department of Health and Human Services; 2014. 20. Geiss LS, Wang J, Cheng YJ, et al. Prevalence and incidence trends for diagnosed diabetes among adults aged 20 to 79 years, United States, 1980-2012. JAMA 2014;312:1218-1226. 21. HCUP Databases. Healthcare Cost and Utilization Project (HCUP). Agency for Healthcare Research and Quality, 2013. 22. Tuomilehto J, Rastenyte˙ D, Jousilahti P, et al. Diabetes mellitus as a risk factor for death from stroke: prospective
ARTICLE IN PRESS O. ADEGBALA ET AL.
8
23.
24.
25.
26.
study of the middle-aged Finnish population. Stroke 1996;27:210-215. Axelrod DA, Upchurch GR Jr, DeMonner S, et al. Perioperative cardiovascular risk stratification of patients with diabetes who undergo elective major vascular surgery. J Vasc Surg 2002;35:894-901. Stoner MC, Abbott WM, Wong DR, et al. Defining the high-risk patient for carotid endarterectomy: an analysis of the prospective national surgical quality improvement program database. J Vasc Surg 2006;43:285-296, e2. Ahari A, Bergqvist D, Troeng T, et al. Diabetes mellitus as a risk factor for early outcome after carotid endarterectomy–a population-based study. Eur J Vasc Endovasc Surg 1999;18:122-126. Nathan DM, Buse JB, Davidson MB, et al. Medical management of hyperglycaemia in type 2 diabetes mellitus: a consensus algorithm for the initiation and adjustment of therapy: a consensus statement from the American Diabetes Association and the European Association for the Study of Diabetes. Diabetologia 2009;52:17-30.
27. O’Neill S, O’Driscoll L. Metabolic syndrome: a closer look at the growing epidemic and its associated pathologies. Obes Rev 2015;16:1-12. 28. Siewiorek GM, Krafty RT, Wholey MH, et al. The association of clinical variables and filter design with carotid artery stenting thirty-day outcome. Eur J Vasc Endovasc Surg 2011;42:282-291. 29. Bersin RM, Stabile E, Ansel GM, et al. A meta-analysis of proximal occlusion device outcomes in carotid artery stenting. Catheter Cardiovasc Interv 2012;80:1072-1078. 30. Wallaert JB, Cronenwett JL, Bertges DJ, et al. Optimal selection of asymptomatic patients for carotid endarterectomy based on predicted 5-year survival. J Vasc Surg 2013;58:112-119. 31. Kannel WB. Lipids, diabetes, and coronary heart disease: insights from the Framingham Study. Am Heart J 1985;110:1100-1107. 32. Shlipak MG, Elmouchi DA, Herrington DM, et al. The incidence of unrecognized myocardial infarction in women with coronary heart disease. Ann Intern Med 2001;134: 1043-1047.
Diabetes Mellitus with Chronic Complications in Relation to Carotid Endarterectomy and Carotid Artery Stenting Outcomes Oluwole Adegbala,
MPH,*†
Kimberly D. Martin, PhD,† David Otuada, Tomi Akinyemiju, PhD†
PhD,‡
and
Background: Carotid endarterectomy and carotid artery stenting are effective treatment procedures for carotid artery stenosis. Although diabetes mellitus is highly prevalent among patients undergoing these revascularization procedures, few studies have examined their impact on periprocedural outcomes. Objectives: The study aimed to determine whether perioperative outcomes among patients undergoing carotid artery stenting and carotid endarterectomy varied depending on the presence of diabetes with or without chronic complications. Methods: We examined adults aged 45 and above hospitalized between 2007 and 2011 in U.S. hospitals who underwent carotid artery revascularization procedures. We used data from the Healthcare Cost and Utilization Project’s Nationwide Inpatient Sample and evaluated the influence of diabetes with or without chronic complications on outcomes. Results: Among patients receiving carotid artery stenting, diabetic patients with chronic complications had significantly increased odds of acute kidney injury (odds ratio [OR]: 3.17, 95% confidence interval [CI]: 2.31-4.35) and longer hospital stay (β: 1.98, 95% CI: 1.582.38) compared with nondiabetic patients. Diabetic patients with chronic complications receiving carotid endarterectomy experienced increased odds of myocardial infarction (OR: 1.12, 95% CI: .90-1.40), stroke (OR: 1.29, 95% CI: .97-1.72), perioperative infection (OR: 2.45, 95% CI: 1.29-4.65), mortality (OR: 1.48, 95% CI: 1.01-2.16), and longer hospital stay (β (days): 2.05, 95% CI: 1.90-2.20) compared with nondiabetic patients. No significant increased odds of perioperative outcomes were observed among diabetic patients without chronic complications. Conclusions: Uncomplicated diabetes did not appear to convey a higher odds of perioperative outcomes among patients undergoing revascularization. However, the presence of diabetes with chronic complications is an important risk factor in the carotid endarterectomy category. Key Words: Diabetes mellitus—carotid endarterectomy—carotid artery—stenting—mortality—hospitalization outcomes. © 2016 National Stroke Association. Published by Elsevier Inc. All rights reserved.
Introduction From the *Hugh Kaul Personalized Medicine Institute, University of Alabama at Birmingham, Birmingham, Alabama; †Department of Epidemiology, University of Alabama at Birmingham, Birmingham, Alabama; and ‡Department of Emergency Medicine, University of Alabama at Birmingham, Birmingham, Alabama. Received June 14, 2016; revision received August 16, 2016; accepted September 10, 2016. Address correspondence to Tomi Akinyemiju, PhD, Department of Epidemiology, University of Alabama, Birmingham, AL. E-mail: [email protected]. 1052-3057/$ - see front matter © 2016 National Stroke Association. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2016.09.012
Stroke is the leading cause of disability and the fifth leading cause of death in the United States, with close to 1 million new cases every year.1 More than 85% of all strokes are ischemic in origin, with approximately 15% of those attributable to carotid artery stenosis.2 Large randomized trials have established carotid endarterectomy (CEA) as the standard secondary preventive treatment for symptomatic severe carotid artery stenosis.3,4 In addition, revascularization with CEA is recommended for most medically stable patients with asymptomatic carotid stenosis of 60%-99% and for a subgroup of patients with asymptomatic carotid stenosis who have a particularly
Journal of Stroke and Cerebrovascular Diseases, Vol. ■■, No. ■■ (■■), 2016: pp ■■–■■
1
ARTICLE IN PRESS O. ADEGBALA ET AL.
2 5
high risk of stroke. Carotid artery stenting (CAS) has increasingly gained recognition as a safe and efficacious alternative treatment to CEA, especially in patients considered to be at a high surgical risk for CEA.6-9 Diabetes mellitus (DM) is a major independent risk factor for cardiovascular disease and stroke, and the prevalence is higher among those undergoing carotid revascularization.10,11 The effect of DM on CEA outcomes is equivocal; a number of studies have suggested increased perioperative risks of stroke and death during CEA in this subgroup of patients,8,12,13 whereas other studies have documented similar perioperative outcomes in both diabetic patients and nondiabetic patients undergoing the surgical procedure.14-17 Although there are conflicting reports on the impact of DM on the outcomes of patients undergoing CEA procedures, the impact of diabetes on the outcomes of patient undergoing CAS remains largely unclear. Given the growing emergence of CAS into clinical practice,18 and the increasing prevalence of diabetes as a chronic medical condition among U.S. adults—with 1.4 million new diagnoses every year19,20—we aimed to determine whether the periprocedural outcomes of patients undergoing CAS and CEA differed depending on the presence of DM with or without chronic complications in a large nationally representative sample of hospitalized patients.
Methods Study Population This was a cross-sectional study of patients aged 45 and above hospitalized for a carotid artery revascularization procedure in a U.S. hospital between 2007 and 2011. Inpatient procedure and outcomes data were obtained from the Healthcare Cost and Utilization Project’s Nationwide Inpatient Sample (HCUP NIS). HCUP is the largest publicly available all-payer inpatient healthcare database in the United States, with data on more than 7 million hospital stays each year, representing more than 35 million hospitalizations nationally, including discharges from community hospitals participating in HCUP, excluding rehabilitation and long-term acute care hospitals.21 All patients undergoing carotid revascularization for carotid artery stenosis with either CAS (International Classification of Disease, 9th Revision, Clinical Modification [ICD-9-CM] 00.63) or CEA (ICD-9-CM 38.12) procedures were identified. Patients were classified as symptomatic if they carried a diagnosis of amaurosis fugax (ICD-9-CM 362.34 and 368.12), transient ischemic attack (ICD-9-CM 435.9 and 781.4), and stroke (ICD-9-CM 433.11, 433.31, 433.91, 434.01, 434.11, and 434.91). Diabetic patients were identified as those with an established DM comorbid condition with or without complications at the time of carotid procedures (ICD-9-CM 250). Diabetes with chronic complications was derived from the Agency for Healthcare Research and Quality comorbidity software, and included ICD-9
codes 250.40-250.93, representing diabetic patients with renal, ophthalmic, and neurologic manifestations, peripheral circulatory disorders, and other specified or unspecified complications. Demographic information including age on admission, gender, and race was obtained. Racial and ethnic groups were categorized into the following four groups—non-Hispanic white, Hispanic, non-Hispanic black, and other groups (including American Indians and Asians). We created a modified Deyo comorbidity index using ICD-9 codes for major comorbid conditions. The conditions included hypertension, congestive heart failure, obesity, chronic pulmonary disease, peripheral vascular disease, and chronic renal failure disease. We identified the presence of each condition within each patient and summed up the number of conditions per patient into a single comorbidity score.
Outcome Measures In-hospital primary outcomes included new events occurring during or after the procedure but before discharge. We determined the presence of post-procedural complications of CAS and CEA procedures using ICD-9 codes for in-hospital postoperative stroke (997.02), pulmonary embolism and infarction (415.11), hematoma (998.12), postoperative infection (998.59), renal dysfunction (997.5), postoperative myocardial infarction or cardiac complications (997.1), and in-hospital mortality which was available directly in the NIS dataset and defined as death due to any cause at any point in time during the index hospitalization. The secondary outcome measures for this study included length of stay during the index hospital admission stay, defined as the number of days between admission and discharge, with same day hospitalizations coded as 0.
Statistical Analysis Descriptive statistics were used to compare demographic and clinical characteristics of patients by DM status, using chi-square for categorical variables and t-tests or analysis of variance for continuous variables. Logistic regression analysis was used to compute the association between DM and post-procedural complications and in-hospital mortality for the two procedures combined and for each revascularization procedure separately. Stratified analysis was conducted to examine the impact of DM with complications and DM without complications on study outcomes. Covariates included gender, age, presentation type, year of the procedure, hospital teaching status, hospital bed size, and modified Deyo comorbidity index. Statistical analyses were performed using SAS software (v9.4, Cary, NC).
Results Demographic and baseline characteristics for 128,664 patients who underwent revascularization procedures between 2007 and 2011 are presented in Table 1. The
Table 1. Descriptive analysis of patients admitted to the hospital between 2007 and 2011 with carotid artery revascularization procedures CAS N = 16,306
Diabetic N = 41,120
Nondiabetic N = 87,544
P value
Diabetic N = 35,924
Nondiabetic N = 76,608
P value
Diabetic N = 5246
Nondiabetic N = 11,060
P value
70.37 ± 8.95 3.24 ± 4.63 17,227 (41.90) 4146 (10.08)
71.49 ± 9.61 2.79 ± 4.38 36,505 (41.73) 8891 (10.16)
.0001 .0001 .567 .684
70.41 ± 8.92 3.22 ± 4.53 15,139 (42.14) 3351 (9.33)
71.55 ± 9.54 2.77 ± 4.38 32,163 (42.02) 7328 (9.57)
.0001 .0001 .686 .205
70.07 ± 9.16 3.34 ± 5.27 2111 (40.24) 801 (15.27)
71.04 ± 10.04 2.91 ± 4.44 4401 (39.80) 1592 (14.39)
.0001 .0001 .588 .140
28,362 (68.97) 2065 (5.02) 2113 (5.14) 8580 (20.87)
65,235 (74.52) 2496 (2.85) 2404 (2.75) 17,409 (19.89)
.0001
24,926 (69.39) 1753 (4.88) 1808 (5.03) 7437 (20.70)
57,228 (74.70) 2081 (2.72) 2065 (2.70) 15,234 (19.89)
.0001
3470 (66.15) 314 (5.99) 307 (5.85) 1155 (22.02)
8105 (73.28) 422 (3.82) 340 (3.07) 2193 (19.83)
.0001
35,583 (86.53) 4035 (9.81) 5522 (13.43) 10,224 (24.86) 8811 (21.43) 5169 (12.57)
67,555 (77.17) 4789 (5.47) 5848 (6.68) 19,007 (21.71) 19,431 (22.20) 4305 (4.92)
.0001 .0001 .0001 .0001 .0019 .0001 .0001
31,134 (86.67) 3285 (9.14) 4637 (12.91) 8577 (23.88) 7684 (21.39) 4585 (12.76)
9669 (77.89) 3889 (5.08) 4965 (6.48) 16,005 (20.89) 17,103 (22.33) 3883 (5.07)
.0001 .0001 .0001 .0001 .0004 .0001
4493 (85.65) 755 (14.39) 894 (17.04) 1666 (31.76) 1139 (21.71) 591 (11.27)
7978 (72.13) 912 (8.25) 893 (8.07) 3057 (27.64) 2359 (21.33) 427 (3.86)
.0001 .0001 .0001 .0001 .578 .0001 .0001
2617 (6.36) 30,822 (74.96) 7681 (44.01) 207 (.50)
12,329 (14.08) 65,444 (74.76) 9771 (11.16) 457 (.52)
2319 (6.46) 27,138 (75.54) 6467 (18.00) 163 (.45)
10,546 (13.77) 57,713 (75.34) 8349 (10.90) 341 (.45)
.0001
299 (5.70) 3723 (70.97) 1224 (23.33) 45 (.86)
1796 (16.24) 1823 (70.73) 1441 (13.03) 121 (1.09)
29,843 (71.90) 1533 (3.73) 8360 (20.33) 1384 (3.37)
62,946 (71.90) 2353 (2.69) 19,274 (22.02) 2971 (3.39)
.0001
26,101 (72.66) 1282 (3.57) 7363 (20.50) 1178 (3.28)
55,167 (72.01) 1963 (2.56) 16,943 (22.12) 2535 (3.31)
.0001
3778 (72.02) 255 (4.86) 1006 (19.18) 207 (3.95)
7859 (71.06) 397 (3.59) 2356 (21.30) 448 (4.05)
.0001
11,347 (28.16) 11,919 (29.58) 9668 (23.99) 7358 (18.26)
21,794 (25.42) 25,133 (29.32) 21,485 (25.06) 17,320 (20.20)
.0001
9797 (27.83) 10,531 (29.91) 8438 (23.97) 6443 (18.30)
18,901 (25.20) 22,238 (29.65) 18,850 (25.13) 15,023 (20.03)
.0001
1565 (30.49) 1402 (27.31) 1241 (24.18) 925 (18.02)
2937 (27.09) 2920 (26.93) 2671 (24.64) 2314 (21.34)
.0001
.665
.839
.160
3
Abbreviations: CAS, carotid artery stenting; CEA, carotid endarterectomy; COPD, chronic obstructive pulmonary disease; CHF, congestive heart failure; CKD, chronic kidney disease; HTN, hypertension; PVD, peripheral vascular disease. *Others include American Indians and Asians. †Comorbidity score. ‡In-hospital mortality. §Residential income by county. ‖Chronic conditions.
ARTICLE IN PRESS
Age Length of stay Female Symptoms Race White Black Hispanic Others* C. cond.‖ HTN CHF CKD PVD COPD Obesity Comorbidities† 0 1-2 >2 Hospital mortality‡ Insurance type Medicare Medicaid Private Others Residential income§ Q1 Q2 Q3 Q4
CEA N = 112,532
DIABETES MELLITUS AND CAROTID ARTERY STENOSIS OUTCOMES
All Patients N = 128,664
ARTICLE IN PRESS O. ADEGBALA ET AL.
4
majority of patients, 112,532 (87.5%), received CEA, whereas 16,306 (12.5%) received CAS. The mean age upon admission was 70.37 ± 8.95 years and 41.9% were women. Nearly three quarters of the patients were white. African– Americans, and Hispanics comprised about 4% each of the study sample. About 80% had hypertension, nearly 25% had peripheral vascular disease and chronic obstructive lung disease, and about 10% had chronic renal failure. About 32% of the population had DM. Patients with DM tended to be younger (70.37 ± 8.95 versus 71.49 ± 9.61 years) and had longer hospital stays compared with nondiabetics. Patients with DM were more likely to be black, obese, hypertensive, and with congestive heart failure and chronic renal failure. The prevalence of DM tended to increase with decreasing residential income. The proportion of diabetic versus nondiabetic patients was similar in both CAS and CEA groups, about 32% each. Among patients receiving CAS, diabetic patients were younger, more likely to have symptomatic presentation, history of congestive heart failure, hypertension, peripheral artery disease, and chronic renal failure, but less likely to be in the highest level residential income compared with nondiabetic populations. Similar trends were observed for CEA patients. Table 2 shows the results of the association between DM with in-hospital outcomes among patients receiving both CAS and CEA. Diabetic patients with chronic complications experienced significantly increased odds of postoperative stroke (odds ratio [OR]: 1.44, 95% confidence interval [CI]: 1.131.84), hematoma (OR: 1.37, 95% CI: 1.15-1.64), infection (OR: 2.84, 95% CI: 1.60-5.03), pneumonia (OR: 2.73, 95% CI: 1.61-4.64), acute kidney injury (OR: 6.02, 95% CI: 5.37-
6.76), and in-hospital mortality (OR:1.64, 95% CI: 1.162.31) compared with nondiabetic patients after adjusting for baseline study covariates such as age, gender, race, insurance type, and residential income. Diabetic patients with chronic complications had 64% higher odds of in-hospital mortality (OR: 1.64, 95% CI: 1.16-2.31) and significantly longer hospital stay (β: 2.45, 95% CI: 2.31-2.59) compared with nondiabetic patients. These associations became attenuated after adjusting for modified Deyo comorbidity index; however, length of hospital stay (β: 2.05, 95% CI: 1.90-2.19), postoperative stroke (OR: 1.36, 95% CI: 1.06-1.75), post-procedural infection (OR: 1.20, 95% CI: 1.01-1.44), hematoma (OR: 2.59, 95% CI: 1.43-4.70), and odds of acute kidney injury (OR: 3.45, 95% CI: 3.05-3.90) remained significantly higher compared with patients without DM in the fully adjusted model. There was no significant difference in the risk of myocardial infarction, other cardiac complications, ischemic stroke, or periprocedural mortality by DM with chronic complications status. Diabetic patients without chronic complications experienced slightly reduced odds of myocardial infarction (OR: .84, 95% CI: .78-.94), hematoma (OR: .88, 95% CI: .81-.96), and in-hospital mortality (OR: .80, 95% CI: .67-.96) compared with patients without DM; however, there were no statistically significant differences in length of hospital stay. Similar results were observed when the analysis was restricted to patients who received the CAS procedure only (Table 3). Diabetic patients with chronic complications had significantly increased odds of hematoma (OR: 1.70, 95% CI: 1.08-2.70) and acute kidney injury (OR: 5.35, 95% CI: 3.97-7.21) after adjusting for baseline study covariates such as age, gender, race, insurance type, and
Table 2. Multivariate adjusted odds ratios for the association between diabetes mellitus and in-hospital outcomes among patients receiving both CAS and CEA Diabetes mellitus (without chronic complications)
Both procedures MI or cardiac complications Postoperative stroke Hematoma Postoperative infection Pulmonary embolism Pneumonia Acute kidney injury Length of hospital stay In-hospital mortality
Diabetes mellitus (with chronic complications)
Model 1* AOR/β (95% CI)
Model 2† AOR/β (95% CI)
Model 1* AOR/β (95% CI)
Model 2† AOR/β (95% CI)
.89 (.81-.97) .94 (.83-1.06) .91 (.84-.99) 1.12 (.78-1.60) 1.29 (.51-3.30) .99 (.71-1.38) 1.28 (1.18-1.40) .13 (.08-.19) .86 (.72-1.03)
.84 (.78-.94) .91 (.81-1.03) .88 (.81-.96) 1.10 (.76-1.57) 1.21 (.47-3.13) .86 (.61-1.20) 1.06 (.97-1.16) .01 (−.04-.06) .80 (.67-.96)
1.27 (1.03-1.56) 1.44 (1.13-1.84) 1.37 (1.15-1.64) 2.84 (1.60-5.03) 3.03 (.66-13.91) 2.73 (1.61-4.64) 6.02 (5.37-6.76) 2.45 (2.31-2.59) 1.64 (1.16-2.31)
1.10 (.89-1.34) 1.36 (1.06-1.75) 1.20 (1.01-1.44) 2.59 (1.43-4.70) 2.05 (.42-10.05) 1.60 (.92-2.78) 3.45 (3.05-3.90) 2.05 (1.90-2.19) 1.26 (.88-1.79)
Abbreviations: AOR, adjusted odds ratios; β, as regression coefficient indicating number of days; CAS, carotid artery stenting; CEA, carotid endarterectomy; CI, confidence interval; MI, myocardial infarction. Bold indicates significance at P value ≤ .05. *Adjusted for race, sex, age, insurance type, and median household income national quartile for patient zip code. †Adjusted for race, sex, age, insurance type, median household income national quartile for patient zip code, and comorbidity score.
ARTICLE IN PRESS DIABETES MELLITUS AND CAROTID ARTERY STENOSIS OUTCOMES
5
Table 3. Multivariate adjusted odds ratios for the association between diabetes mellitus and in-hospital outcomes among patients receiving CAS Diabetes mellitus (without chronic complications)
CAS MI or cardiac complications Postoperative stroke Hematoma Postoperative infection Pneumonia Acute kidney injury Length of hospital stay In-hospital mortality
Diabetes mellitus (with chronic complications)
Model 1* OR/β (95% CI)
Model 2† OR/β (95% CI)
Model 1* AOR/β (95% CI)
Model 2† AOR/β (95% CI)
.98 (.76-1.25) .87 (.67-1.12) .77 (.59-.99) .24 (.03-1.91) .76 (.36-1.63) 1.32 (1.08-1.64) .13 (−.03-.29) .81 (.57-1.16)
.98 (.76-1.26) .87 (.67-1.13) .71 (.55-.92) .29 (.04-2.33) .63 (.29-1.35) 1.03 (.83-1.28) −.003 (−.16-.16) .77 (.54-1.11)
.88 (.46-1.68) 1.43 (.86-2.37) 1.70 (1.08-2.70) 3.67 (.79-17.21) 2.27 (.67-7.65) 5.35 (3.97-7.21) 2.29 (2.00-2.68) .69 (.25-1.87)
.91 (.47-1.75) 1.51 (.90-2.52) 1.44 (.90-2.30) 3.86 (.76-19.59) 1.45 (.41-5.09) 3.17 (2.31-4.35) 1.98 (1.58-2.38) .55 (.20-1.52)
Abbreviations: AOR, adjusted odds ratios; β, as regression coefficient indicating number of days; CAS, carotid artery stenting; CI, confidence interval; MI, myocardial infarction. Bold indicates significance at P value ≤ .05. *Adjusted for race, sex, age, insurance type, and median household income national quartile for patient zip code. †Adjusted for race, sex, age, insurance type, median household income national quartile for patient zip code, and comorbidity score.
residential income. In addition, those patients had significantly longer hospital stay (β: 2.29, 95% CI: 2.002.68) compared with patients without DM. These associations were attenuated after adjusting for baseline chronic conditions, with only the length of hospital stay (β: 1.98, 95% CI: 1.58-2.38) and acute kidney injury (OR: 3.17, 95% CI: 2.31-4.35) remaining significantly higher compared with patients without DM in the fully adjusted model. Diabetic patients without chronic complications experienced slightly reduced odds of hematoma (OR: .71, 95% CI: .55-.92) and reduced hospital stay compared with
patients without DM; however, there were no statistically significant differences in in-hospital mortality, myocardial infarction, or postoperative stroke outcomes. With regard to the CEA procedure (Table 4), there were significantly increased odds of myocardial infarction (OR: 1.33, 95% CI: 1.07-1.65), postoperative stroke (OR: 1.41, 95% CI: 1.06-1.86), hematoma (OR: 1.32, 95% CI: 1.091.59), infection (OR: 2.74, 95% CI: 1.48-5.07), pneumonia (OR: 2.90, 95% CI: 1.61-5.24), acute kidney injury (OR: 6.14, 95% CI: 5.42-6.96), and in-hospital mortality (OR:1.95, 95% CI: 1.35-2.82) associated with DM with chronic
Table 4. Multivariate adjusted odds ratios for the association between diabetes mellitus and in-hospital outcomes among patients receiving CEA Diabetes mellitus (without chronic complications)
CEA MI or cardiac complications Postoperative stroke Hematoma Postoperative infection Pulmonary embolism Pneumonia Acute kidney injury Length of hospital stay In-hospital mortality
Diabetes mellitus (with chronic complications)
Model 1* AOR/β (95% CI)
Model 2† AOR/β (95% CI)
Model 1* AOR/β (95% CI)
Model 2† AOR/β (95% CI)
.88 (.79-.97) .94 (.83-1.08) .93 (.85-1.01) 1.22 (.85-1.76) 1.19 (.44-3.23) 1.06 (.73-1.54) 1.28 (1.16-1.40) .13 (.08-.19) .88 (.72-1.08)
.83 (.75-.92) .91 (.80-1.05) .90 (.82-.98) 1.18 (.81-1.71) 1.09 (.40-3.00) .93 (.64-1.35) 1.06 (.97-1.17) .01 (−.04-.07) .81 (.66-1.00)
1.33 (1.07-1.65) 1.41 (1.06-1.86) 1.32 (1.09-1.59) 2.74 (1.48-5.07) 1.53 (.19-12.24) 2.90 (1.61-5.24) 6.14 (5.42-6.96) 2.47 (2.31-2.62) 1.95 (1.35-2.82)
1.12 (.90-1.40) 1.29 (.97-1.72) 1.16 (.95-1.41) 2.45 (1.29-4.65) 1.06 (.13-8.96) 1.67 (.90-3.10) 3.49 (3.05-3.99) 2.05 (1.90-2.20) 1.48 (1.01-2.16)
Abbreviations: AOR, adjusted odds ratios; β, as regression coefficient indicating number of days; CEA, carotid endarterectomy; CI, confidence interval; MI, myocardial infarction. Bold indicates significance at P value ≤ .05. *Adjusted for race, sex, age, insurance type, and median household income national quartile for patient zip code. †Adjusted for race, sex, age, insurance type, median household income national quartile for patient zip code, and comorbidity score.
ARTICLE IN PRESS O. ADEGBALA ET AL.
6
complications, after adjusting for baseline study covariates such as age, gender, race, insurance type, and residential income. In addition, those patients had 95% higher odds of in-hospital mortality (OR: 1.95, 95% CI: 1.352.82) and had significantly longer hospital stay (β: 2.47, 95% CI: 2.31-2.62) compared with patients without DM. These associations diminished after adjusting for baseline chronic conditions, and only postoperative infection (OR: 2.45, 95% CI: 1.29-4.65), length of hospital stay (β: 2.05, 95% CI: 1.90-2.20), acute kidney injury (OR: 3.49, 95% CI: 1.69-2.26), and in-hospital mortality (OR: 1.48, 95% CI: 1.01-2.16) remained significantly higher compared with patients without DM in the fully adjusted model. Diabetic patients without chronic complications experienced slightly reduced odds of myocardial infarction (OR: .83, 95% CI: .75-.92) and hematoma (OR: .90, 95% CI: .82-.98) compared with patients without DM; however, there were no statistically significant differences in in-hospital mortality, acute kidney injury, and postoperative stroke.
Discussion Previous studies have shown that DM is associated with increased risk of ischemic stroke, and stroke-related mortality and functional limitations are higher in diabetic patients.10,11,22 Therefore, diabetic patients with carotid artery stenosis may derive considerable benefits from carotid revascularization procedures. The benefits of these procedures, however, should outweigh the risks of perioperative complications. In order to minimize perioperative outcomes in diabetic patients, it is important to identify whether there are significantly increased odds of perioperative outcomes, as this information will be critical in assessing the health benefits of such procedures in this population subgroup. Unfortunately, the literature is not particularly clear on the potentially increased risks that DM poses on carotid revascularization procedures. Several previous studies have demonstrated that diabetic patients undergoing CEA have a higher risk of cardiovascular events than nondiabetic patients.10,11 The literature remains divided on this topic, however. A number of investigators8,12,13,23-25 found that diabetic patients had a significantly higher incidence of death and neurologic or cardiovascular complications after CEA. In one of these studies,25 there was no significant difference in neurologic postoperative morbidity between 341 diabetic patients and 2281 nondiabetic patients, although diabetic patients had a higher 30-day mortality that was mainly cardiacrelated. The increased mortality in this study was attributed to the higher frequency of surgery in diabetic patients for minor stroke, which had a higher perioperative complication rate than other clinical presentations or asymptomatic stenosis. Similarly, in a large series of studies, investigators reported that the presence of insulintreated DM was the strongest independent risk factor for
23,24
cardiac events and death. The increased perioperative risks in diabetic patients on insulin therapy may indicate the presence of chronic complications that are often experienced by type 2 diabetic patients on insulin therapy.26 In contrast to the above findings, other studies observed no significant differences in perioperative stroke, myocardial infarction, mortality, or long-term survival between diabetic and nondiabetic patients undergoing CEA.14-16 In the present analysis, among patients who underwent CAS, there were no significant differences in the risk of myocardial infarction, other cardiac complications, ischemic stroke, or periprocedural mortality. This is similar to the pooled analysis of outcomes from randomized clinical trials of 3454 symptomatic patients27 and analysis of 2196 procedures,8 which found no higher perioperative risk of major adverse cardiac events in diabetic patients who underwent CAS procedure. In contrast, other studies found that diabetic patients were at significantly increased risks of stroke, transient ischemic attack, and death within 30 days.28,29 The conflicting results may be due to significant heterogeneity in provider skills or experience, study population characteristics, measure of diabetes used in the study, and possible inadequate control for other confounders such as the presence of other comorbid conditions, and physician and hospital characteristics. With CEA, diabetic patients with chronic complications were more likely to experience myocardial infarction and postoperative stroke. These findings are comparable with previous studies,12,13 including the recent study of 1116 CEA procedures that found that the perioperative risk of stroke and death was threefold higher (OR: 2.83, 95% CI: 1.05-7.61, P = .04) in diabetic patients as opposed to nondiabetic patients.8 Similarly, in a more recent study, insulin-dependent DM was identified as a high-risk profile significantly impacting on survival of patients who underwent CEA for asymptomatic stenosis.30 In contrast, other studies16 observed no significant differences in perioperative stroke, MI, mortality, or long-term survival between diabetic and nondiabetic patients undergoing CEA. Provider characteristics and patient characteristics including lack of stratification by DM with complications may explain the differing results. Patients with DM are known to be at increased risk of silent myocardial disease,31,32 which could partly account for the lack of observed increase in clinically diagnosed MI and other cardiac complications in diabetic patients undergoing CAS procedure. In addition, DM with or without chronic complications does not confer significantly increased perioperative mortality in CAS population despite the increased incidence of comorbid conditions in the diabetes population. However, in the CEA population, there was a significantly increased risk of MI and other cardiac complications and increased mortality, which was attenuated after adjusting for comorbid conditions. This current study adds support to the previous studies in finding no significant differences in major outcome
ARTICLE IN PRESS DIABETES MELLITUS AND CAROTID ARTERY STENOSIS OUTCOMES
parameters between diabetic and nondiabetic patients undergoing CAS revascularization procedures. 8,27 We confirmed here that although DM with chronic complications has slightly increased risks of MACE periprocedural complications in those who underwent CEA, DM with or without chronic complications does not confer increased periprocedural MACE risk in CAS population. This suggests that patients’ stratification based on the presence of DM with complications may improve the identification of subgroups better suited for the CEA revascularization procedures and who may require targeted interventions to reduce the risk of perioperative complications. This study is not without its limitations. These include lack of data on post-hospitalization outcomes, with the stroke or death events only reflecting inpatient events and not including 30-day stroke or death rates. Also, detailed clinical data on stroke severity, subtype or side of the brain affected, management or specifics of DM chronic complications, actual degree of carotid stenosis, the device characteristics, perioperative medications, and history of coronary artery diseases are lacking. The strengths of this study include its large sample size and detailed administrative claims data that were nationally representative. We controlled for hospital teaching status and bed size as proxies for provider skills and experience, and used clinical diagnoses of both surgical procedures and DM status to improve reliability of measures. This is the first study, to our knowledge, to examine the influence of diabetes, including the presence of chronic complications, on revascularization outcomes. In conclusion, although DM overall does not appear to convey higher risks of perioperative outcomes on patients undergoing revascularization, our data suggest that the presence of DM with chronic complications is an important risk factor for poorer outcomes in CEA category. Future studies will be needed to identify the specific mechanism through which DM with complications increases perioperative risks, with the ultimate goal of minimizing perioperative complications in this group of patients.
References 1. Mozaffarian D, Benjamin EJ, Go AS, et al. Heart disease and stroke statistics-2016 update: a report from the American Heart Association. Circulation 2016;133:e38-e60. 2. Boulanger M, Touze E. Periprocedural risk of myocardial infarction after carotid endarterectomy and carotid angioplasty and stenting. Arch Cardiovasc Dis 2016;109:159-162. 3. North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. N Engl J Med 1991;325:445-453. 4. Randomised trial of endarterectomy for recently symptomatic carotid stenosis: final results of the MRC European Carotid Surgery Trial (ECST). Lancet 1998;351: 1379-1387.
7
5. Mohler ER III, Fairman RM. Management of asymptomatic carotid atherosclerotic disease. https://www .uptodate.com/contents/management-of-asymptomatic -carotid-atherosclerotic-disease?source=search_result &search=Mohler%20ER%20III,%20Fairman%20RM.%20 Management%20of%20asymptomatic&selectedTitle=1~150. UpToDate, November, 2015. 6. Goodney PP, Travis LL, Malenka D, et al. Regional variation in carotid artery stenting and endarterectomy in the Medicare population. Circ Cardiovasc Qual Outcomes 2010;3:15-24. 7. Jalbert JJ, Nguyen LL, Gerhard-Herman MD, et al. Outcomes after carotid artery stenting in Medicare beneficiaries, 2005 to 2009. JAMA Neurol 2015;72:276-286. 8. Parlani G, De Rango P, Cieri E, et al. Diabetes is not a predictor of outcome for carotid revascularization with stenting as it may be for carotid endarterectomy. J Vasc Surg 2012;55:79-89. 9. Timaran CH, Mantese VA, Malas M, et al. Differential outcomes of carotid stenting and endarterectomy performed exclusively by vascular surgeons in the Carotid Revascularization Endarterectomy versus Stenting Trial (CREST). J Vasc Surg 2013;57:303-308. 10. Kannel WB, McGee DL. Diabetes and cardiovascular disease. The Framingham study. JAMA 1979;241:20352038. 11. Kuusisto J, Mykkänen L, Pyörälä K, et al. Non-insulindependent diabetes and its metabolic control are important predictors of stroke in elderly subjects. Stroke 1994;25: 1157-1164. 12. Aziz I, Lewis RJ, Baker JD, et al. Cardiac morbidity and mortality following carotid endarterectomy: the importance of diabetes and multiple Eagle risk factors. Ann Vasc Surg 2001;15:243-246. 13. Dorigo W, Pulli R, Pratesi G, et al. Early and long-term results of carotid endarterectomy in diabetic patients. J Vasc Surg 2011;53:44-52. 14. Ballotta E, Da Giau G, Renon L. Is diabetes mellitus a risk factor for carotid endarterectomy? A prospective study. Surgery 2001;129:146-152. 15. Hamdan AD, Saltzberg SS, Sheahan M, et al. Lack of association of diabetes with increased postoperative mortality and cardiac morbidity: results of 6565 major vascular operations. Arch Surg 2002;137:417-421. 16. Pistolese GR, Appolloni A, Ronchey S, et al. Carotid endarterectomy in diabetic patients. J Vasc Surg 2001; 33:148-154. 17. Rockman CB, Saltzberg SS, Maldonado TS, et al. The safety of carotid endarterectomy in diabetic patients: clinical predictors of adverse outcome. J Vasc Surg 2005; 42:878-883. 18. Akbari CM, Pomposelli FB Jr, Gibbons GW, et al. Diabetes mellitus: a risk factor for carotid endarterectomy? J Vasc Surg 1997;25:1070-1075, discussion 5-6. 19. Centers for Disease Control and Prevention. National Diabetes Statistics Report: Estimates of Diabetes and Its Burden in the United States, 2014, Atlanta, GA: U.S. Department of Health and Human Services; 2014. 20. Geiss LS, Wang J, Cheng YJ, et al. Prevalence and incidence trends for diagnosed diabetes among adults aged 20 to 79 years, United States, 1980-2012. JAMA 2014;312:1218-1226. 21. HCUP Databases. Healthcare Cost and Utilization Project (HCUP). Agency for Healthcare Research and Quality, 2013. 22. Tuomilehto J, Rastenyte˙ D, Jousilahti P, et al. Diabetes mellitus as a risk factor for death from stroke: prospective
ARTICLE IN PRESS O. ADEGBALA ET AL.
8
23.
24.
25.
26.
study of the middle-aged Finnish population. Stroke 1996;27:210-215. Axelrod DA, Upchurch GR Jr, DeMonner S, et al. Perioperative cardiovascular risk stratification of patients with diabetes who undergo elective major vascular surgery. J Vasc Surg 2002;35:894-901. Stoner MC, Abbott WM, Wong DR, et al. Defining the high-risk patient for carotid endarterectomy: an analysis of the prospective national surgical quality improvement program database. J Vasc Surg 2006;43:285-296, e2. Ahari A, Bergqvist D, Troeng T, et al. Diabetes mellitus as a risk factor for early outcome after carotid endarterectomy–a population-based study. Eur J Vasc Endovasc Surg 1999;18:122-126. Nathan DM, Buse JB, Davidson MB, et al. Medical management of hyperglycaemia in type 2 diabetes mellitus: a consensus algorithm for the initiation and adjustment of therapy: a consensus statement from the American Diabetes Association and the European Association for the Study of Diabetes. Diabetologia 2009;52:17-30.
27. O’Neill S, O’Driscoll L. Metabolic syndrome: a closer look at the growing epidemic and its associated pathologies. Obes Rev 2015;16:1-12. 28. Siewiorek GM, Krafty RT, Wholey MH, et al. The association of clinical variables and filter design with carotid artery stenting thirty-day outcome. Eur J Vasc Endovasc Surg 2011;42:282-291. 29. Bersin RM, Stabile E, Ansel GM, et al. A meta-analysis of proximal occlusion device outcomes in carotid artery stenting. Catheter Cardiovasc Interv 2012;80:1072-1078. 30. Wallaert JB, Cronenwett JL, Bertges DJ, et al. Optimal selection of asymptomatic patients for carotid endarterectomy based on predicted 5-year survival. J Vasc Surg 2013;58:112-119. 31. Kannel WB. Lipids, diabetes, and coronary heart disease: insights from the Framingham Study. Am Heart J 1985;110:1100-1107. 32. Shlipak MG, Elmouchi DA, Herrington DM, et al. The incidence of unrecognized myocardial infarction in women with coronary heart disease. Ann Intern Med 2001;134: 1043-1047.