Non-ST–Elevation Myocardial Infarction in Patients Undergoing Carotid Endarterectomy or Carotid Artery Stent Placement
Background and Purpose—The significance of non-ST–elevation myocardial infarction (NSTEMI) after carotid endarterectomy or carotid angioplasty and stent placement is unknown. We performed this study to identify the frequency of NSTEMI and impact on outcomes related to carotid endarterectomy or carotid artery stent placement in patients treated in a large national cohort.
Methods—We determined the frequency of NSTEMI and associated in-hospital outcomes including mortality and composite of stroke, cardiac events, and mortality using data from the Nationwide Inpatient Survey from 2002 to 2009.
Results—Of 1 083 688 patients who underwent carotid endarterectomy or carotid artery stent placement, 11 341 (1%) patients developed NSTEMI during hospitalization. After adjusting for constitutional variables and risk factors, NSTEMI was associated with higher rates of in-hospital mortality (odds ratio, 8.6; 95% confidence interval, 7.0–10.7; P≤0.0001) and composite end point of stroke, cardiac events, and death (odds ratio, 14.6; 95% confidence interval, 13.0–16.5; P≤0.0001).
Conclusions—Our results contradict the notion that NSTEMI is a relatively benign entity after carotid endarterectomy or carotid artery stent placement.
Perioperative myocardial infarction (MI) is associated with 30-day mortality rates ranging from 12% to 25% in noncardiac surgery: the majority of these are non-ST–elevation MI (NSTEMI).1–3 Although MI has been evaluated as a periprocedural complication of either carotid endarterectomy (CEA) or carotid artery stent placement (CAS) in previous trials, rates of periprocedural NSTEMI were not measured.4–11 We sought to evaluate the frequency of NSTEMI after CEA or CAS in clinical practice and to ascertain the relationship with in-hospital outcomes.
Data were obtained from the National Inpatient Sample (NIS) from 2002 to 2009, which includes all-payer discharge information from a national survey of nonfederal hospitals. A comprehensive synopsis on NIS data is available at http://www.hcup-us.ahrq.gov.
International Classification of Disease, 9th Revision; Clinical Modification (ICD-9-CM) codes and procedure codes 00.63 and 38.12 were used, respectively, to identify patients undergoing CAS or CEA. Patients with both procedure codes were excluded. If a patient’s discharge diagnosis (diagnostic fields 1–15) was either carotid artery stenosis with stroke (433.11) or, if there was carotid artery stenosis without stroke (433.10), with subsequent code for transient ischemic attack (435), patients were classified as symptomatic. ICD-9-CM diagnosis code 410.71 was used to identify cases with NSTEMI. We excluded all the patients with other types of MI using ICD-9 diagnosis codes 410.0 to 410.6 and 410.8 to 410.9.
Study variables include age, sex, race/ethnicity, and comorbidities obtained from Agency for Healthcare Research and Quality comorbidity files (online-only Data Supplement). Primary outcome measures were inhospital mortality neurological complication, cardiac complication, and composite end point including death. ICD-9-CM codes 997.00 to 997.09 and 997.10 were used to identify neurological and cardiac complication, respectively. Cardiac complications included in ICD-9-CM code 997.10 include cardiac arrest, insufficiency, or failure during or secondary to a procedure. Composite end point was defined as neurological or cardiac complication, or in-hospital mortality.
Inpatient lengths of stay and hospital charges were estimated. Discharge status was categorized into routine, home healthcare, short-term hospital, other facility including intermediate care and skilled nursing home, or death in the NIS. Disability at discharge was none-to-minimal disability for routine discharge: any other discharge status constituted moderate-to-severe disability as previously described.12
SAS version 9.1 (SAS Institute, Cary, NC) was used to generate national estimates from NIS weighted counts following Healthcare Cost and Utilization Project protocol.13
Outcomes of interest were compared between patients with and without NSTEMI, after CAS or CEA. The rates of postoperative stroke, cardiac complications, in-hospital mortality, and composite end point were compared. χ2 test for categorical data and analysis of variance for continuous data were used for sensitivity analysis, with P<0.05 statistically significant.
Two multivariate logistic regression models were created to determine the effect of NSTEMI on postoperative stroke, cardiac complications, in-hospital mortality composite end point, and discharge disposition after adjusting for the potential confounders of age (continuous), sex (categorical), and comorbid conditions (categorical) including hypertension, diabetes mellitus, dyslipidemia, atrial fibrillation, congestive heart failure, chronic lung disease, renal failure, nicotine dependence, and alcohol abuse. Model one included all patients, and model 2 included patients discharged alive.
Of the 1 083 688 patients who underwent CEA or CAS, 11 341 (1%) patients developed NSTEMI during hospitalization. Patients with NSTEMI were older than patients without NSTEMI: patients <75 years old comprised 59% of those with versus 64% without NSTEMI, patients 76 to 80 years old comprised 21% of those with versus 19% of those without NSTEMI, and patients >80 years old comprised 20% of those with versus 17% of those without NSTEMI (P<0.001). The proportions of patients with diabetes mellitus (31% versus 26%), atrial fibrillation (24% versus 8%), renal failure (16% versus 5%), or congestive heart failure (24% versus 7%) were significantly higher among those who developed NSTEMI than those who did not. Symptomatic carotid stenosis patients comprised 7.2% in patients with NSTEMI versus 5.2% in patients without NSTEMI (P<0.002; Table I in the online-only Data Supplement).
In-hospital complications including pneumonia, deep venous thrombosis, and urinary tract infections were significantly higher among patients who developed NSTEMI. Patients who developed NSTEMI were more likely to undergo gastrostomy, intubation, mechanical ventilation, and transfusions. In-hospital mortality (6.2% versus 0.4%; P<0.0001) and neurological complications (6% versus 1.4%; P<0.0001) were significantly higher among patients with NSTEMI. Hospitalization charges ($113 317 versus $29 160; P<0.0001) and mean (±SD) hospital days (12.2±9.0 versus 2.8±4.0; P<0.0001) were significantly higher among patients with NSTEMI.
After adjusting for age, sex, presence of hypertension, diabetes mellitus, hyperlipidemia, atrial fibrillation, congestive heart failure, and renal failure, NSTEMI was associated with higher rates of in-hospital mortality (odds ratio, 8.6; 95% confidence interval, 7.0–10.7; P≤0.0001) and composite end point of stroke, cardiac events, and death (odds ratio, 14.6; 95% confidence interval, 13.0–16.5; P≤0.0001; Table II in the online-only Data Supplement). Similarly, NSTEMI was associated with higher rates of moderate-to-severe disability (odds ratio, 6.2; 95% confidence interval, 5.5–6.8; P≤0.0001).
We found that ≈1% of patients who have perioperative NSTEMI after CEA or CAS have significantly higher rates of neurological and cardiological complications, mortality, and healthcare resource use before hospital discharge. The increased mortality associated with periprocedural NSTEMI parallels the findings of the 8351-patient Perioperative Ischemic Evaluation (POISE) trial.1
We found a surprising disparity in the postprocedural duration of hospitalization between those with and without periprocedural NSTEMI. In our unadjusted analysis, patients with NSTEMI were admitted for an average of 12.2 days, whereas those without NSTEMI remained inpatient for only 2.8 days on average. By contrast, the average length of stay in the United States for patients admitted with NSTEMI decreased from 5.3 days to 5.1 days between 2000 and 2009.14 The ≈1 week longer inpatient course in our study is likely partially reflective of delay in perioperative MI diagnosis, which has been previously described and time incurred in stabilization of other associated inhospital events.1
Our study has several notable limitations. First, NIS sample data are based on ICD-9-CM diagnoses and procedures codes, such that miscoding is a potential source for error. Second, the temporal sequence of diagnoses during hospital admission is not discernible in NIS data, so any coding of NSTEMI using only cardiac marker elevation could possibly categorize poststroke enzyme elevation as periprocedural NSTEMI. Additionally, the proportion of patients who had concurrent coronary artery bypass surgery and CEA could not be identified because of the absence of a unique identifying procedure code. Similarly, the effect of variables such as general anesthesia (versus conscious sedation), delay to carotid intervention in symptomatic patients, and periprocedural antithrombotic regimens could not be determined because of the absence of this information in the NIS database.
Our findings demonstrate that although perioperative NSTEMI after CEA or CAS is relatively infrequent, patients with NSTEMI are more likely to have in-hospital adverse outcomes and a significantly increased burden of healthcare resource use. Although our findings do not exclude NSTEMI as a cause of adverse outcomes after CEA or CAS, they strongly suggest that NSTEMI is a marker of overall postprocedural illness severity.
All authors have read and approved the article submission. Drs Khan, Adil, and Qureshi equally assisted in the synthesis and discussion of ideas, and share equal responsibility for the information written in the article.
Dr Qureshi has received funding from NIH R01-NS44976-01A2, AHA established investigator award 0840053N. The other authors report no conflicts.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.113.003728/-/DC1.
- Received October 2, 2013.
- Accepted October 9, 2013.
- © 2013 American Heart Association, Inc.
- Devereaux PJ,
- Xavier D,
- Pogue J,
- Guyatt G,
- Sigamani A,
- Garutti I,
- et al
- Halliday A,
- Mansfield A,
- Marro J,
- Peto C,
- Peto R,
- Potter J,
- et al
- Yadav JS,
- Wholey MH,
- Kuntz RE,
- Fayad P,
- Katzen BT,
- Mishkel GJ,
- et al
- 13.↵HCUP Databases. Healthcare Cost and Utilization Project (HCUP) Rockville, MD, 2011.http://www.hcup-us.ahrq.gov/home.jsp.
- Sridhar A,
- Jeevanantham V,
- Patel D,
- Gupta K,
- Dawn B