Protamine Does Not Increase Risk of Stroke in Patients With Elective Carotid Stenting
Background and Purpose—Reversal of anticoagulation with protamine might predispose to a higher risk of stroke in patients with carotid stenting. We evaluated a national, multihospital patient database to examine the risk of stroke in patients with carotid stenting receiving protamine compared with those who did not.
Methods—The Premier Perspective database was used to identify patients who were electively hospitalized for carotid stenting from 2006 through 2011. The incidence of in-hospital mortality and morbidity was compared between patients who did and did not receive protamine after propensity score adjustment via 1:1 matching to reduce selection bias.
Results—Of 6664 patients with carotid stenting treated at 193 hospitals, 556 (8%) received protamine on the day of the procedure. After matching by propensity score, patients who received protamine had a similar likelihood of stroke or transient ischemic attack (P=0.77), in-hospital mortality (P=0.12), discharge to long-term care (P=0.15), and access site complications (P=0.90) as compared with patients who did not receive protamine.
Conclusions—Protamine administration is not associated with additional risk of adverse events after carotid stenting.
Administration of heparin to achieve anticoagulation during carotid angioplasty and stenting is a routine measure to avoid thromboembolic complications. Protamine is sometimes administered to promote hemostasis after femoral sheath removal. Although such use of protamine might reduce femoral access site bleeding complications, it is reasonable to consider that it might be associated with ischemic complications related to a carotid stent. To investigate this potential risk, we evaluated the Premier Perspective database from 2006 to 2011 to compare rates of stroke, death, and access site complications among patients with carotid stenting who received protamine and those who did not.
Data Source and Study Population
The Perspective database is a voluntary, fee-supported collection of data to assess quality and resource use.1 As of 2011, the Perspective database consisted of ≈15% of hospitalizations nationwide and represented >600 US hospitals. Procedures and outcomes were identified using International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9CM) and Current Procedural Terminology codes.
Patients were included if they were electively admitted for carotid stenting (ICD-9 code 00.63). Patients who received protamine on the day of the procedure were identified from the billing record.
The primary outcome variables of this study were stroke or transient ischemic attack during hospitalization. Secondary outcomes were in-hospital mortality, discharge to long-term care, subarachnoid hemorrhage or intracranial hemorrhage, acute myocardial infarction, iatrogenic cerebrovascular infarction/hemorrhage, blood transfusion, access site complications (hemorrhage or hematoma, contusion/hematoma of trunk/lower limb, embolism/thrombosis/aneurysm of lower extremity artery, injury to retroperitoneum/femoral artery, accidental procedural puncture/laceration, peripheral vascular complication, and iatrogenic infection), and complications requiring intervention (hemorrhage control, vessel suturing or other repair, percutaneous treatment of extremity pseudoaneurysm, and exploration for postoperative extremity hemorrhage, thrombosis, or infection). Procedural codes used to identify outcomes are listed in Table I in the online-only Data Supplement.
Propensity Score Analysis
A propensity score representing the likelihood of receiving protamine was calculated for all patients using a logistic regression model as previously described.2 Protamine recipients and patients who did not receive protamine underwent 1:1 nearest neighbor (Greedy-type) matching of the propensity score logit. Covariates and outcome incidences were compared between protamine and no protamine groups using Wilcoxon rank-sum test or Fisher exact test. Odds ratiosof outcomes were calculated using Fisher exact test.
There were 6664 patients identified who underwent carotid stenting at 193 hospitals between 2006 and 2011, with 556 patients (8%) receiving protamine on the day of procedure (Table II in the online-only Data Supplement). Patients who received protamine were more likely to be female, more likely to be a transfer patient, more likely to be treated at Midwest hospitals, and less likely to be treated at West hospitals compared with patients that did not receive protamine. The majority of carotid stenting cases were performed at urban medical centers (95% overall), and slightly more than half were performed at centers classified as teaching hospitals (54% overall).
After propensity score generation, 555 patients who received protamine and 555 patients who did not receive protamine were matched based on similarities in their demographic and clinical characteristics (Table III in the online-only Data Supplement). All covariates were statistically indistinguishable between the 2 patient groups after matching. In this matched cohort, the likelihood of stroke or transient ischemic attack was similar between patients who did and did not receive protamine (Table; P=0.77). The likelihood of all secondary outcomes, including in-hospital mortality (P=0.12), discharge to long-term care (P=0.15), iatrogenic cerebrovascular infarction or hemorrhage (P=0.26), access site complications (P=0.90), and complications requiring intervention (P=0.57), were also similar between the protamine and no protamine groups. The incidence of blood transfusions was slightly higher in the protamine group compared with the control group (1.4% versus 0.2%); however, this difference bordered on significance (P=0.0500).
Our study did not find an association between protamine administration and risk of femoral access site complications, neurological complication, discharge to long-term care, or death in patients with elective carotid stent. We found a higher rate of blood transfusion in the protamine group compared with patients not given protamine; however, this may be attributable to protamine that was administered after a bleeding complication rather than directly attributable to the protamine itself.
A prior study evaluating the use of protamine in carotid stent patients reported an increased risk of stroke and death in patients receiving protamine with borderline statistical significance (P=0.052).3 However, only 171 of 1989 patients received protamine in that study. Our study improves on these findings by comparing 555 patients treated with protamine to a control group of patients using propensity score adjustment.
The risk of protamine administration has been studied extensively in patients undergoing carotid endarterectomy. Early studies noted a higher risk of stroke with postoperative protamine reversal,4–6 but subsequent larger studies found no increased risk of stroke with protamine in patients with carotid endarterectomy.7–10 Several of these studies reported a benefit to protamine use via reduced risk of bleeding complications.6,9–11
Protamine use after percutaneous coronary interventions has also been studied extensively. Protamine administration did not predispose to stent thrombosis,12 myocardial infarction,13 or coronary restenosis14 but was found to be associated with reduction in major femoral access site bleeding complications.13–15
Our study has several limitations. First, selection bias and unmeasured confounders may remain that could affect our results. Second, the nonstandardized protamine protocols in our study may limit the generalizability of our results. Third, we do not know the precise sequence of events for patients in the database, and it is possible that protamine was administered after a complication occurred rather than before. This might limit our ability to detect a reduction in major femoral access complications that others have reported with protamine.13–15 Finally, coding inaccuracies occur in administrative databases; however, these inaccuracies are likely distributed similarly between the protamine and no protamine groups.
Protamine administration is not associated with additional risk of adverse events after carotid stenting.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.113.001188/-/DC1.
- Received February 13, 2013.
- Revision received March 15, 2013.
- Accepted April 4, 2013.
- © 2013 American Heart Association, Inc.
- 1.↵Premier Research Services. Premier Perspective Database. Charlotte, NC. 2012.
- McDonald JS,
- McDonald RJ,
- Fan J,
- Lanzino G,
- Kallmes DF,
- Cloft HJ
- Aronow HD,
- Gray WA,
- Ramee SR,
- Mishkel GJ,
- Schreiber TJ,
- Wang H
- Hannan EL,
- Popp AJ,
- Feustel P,
- Halm E,
- Bernardini G,
- Waldman J,
- et al
- Parodi G,
- De Luca G,
- Moschi G,
- Bellandi B,
- Valenti R,
- Migliorini A,
- et al