Abstract
Background:
Trials have shown that for patients on oral anticoagulants (OAC), a short course of dual antiplatelet therapy (DAPT) with OAC reduces post-percutaneous coronary intervention (PCI) bleeding without increasing ischemic events. Adoption of this strategy has been variable. We evaluated the impact of an institutional quality improvement (QI) initiative to reduce the use of triple therapy (TT, OAC + DAPT) and improve discharge communication post-PCI.
Methods:
A hospital-wide QI initiative was developed to minimize time on TT post-PCI. Interventions included institutional guidelines emphasizing discharge on OAC with a P2Y12 inhibitor or reducing TT duration to ≤30 days, changes to the computerized decision-support system and an educational curriculum for house staff. PCI patients 18 months before and after the initiative (2017–2020) were reviewed along with a faculty survey assessing prescribing practices to evaluate efficacy of the interventions.
Results:
Among 2,797 PCIs reviewed, 431 were included based on OAC at discharge: 24.9% female, 80.1% white, and mean age 74 years. The most common indications for OAC were atrial fibrillation (70%) and left ventricular dysfunction (14.3%). Mean duration of TT decreased (58.7 to 37.8 days, p=0.02) and patients discharged on TT≤30 days increased (24% to 37%, p=0.019) after intervention. Of surveyed faculty (n=20), 75.0% reported familiarity with the guidelines and 57.9% reported using them to make therapy decisions.
Conclusions:
Following implementation of a QI initiative, fewer patients were discharged on TT and shorter durations of TT were used. Similar initiatives should be considered at institutions with prevalent use of TT post-PCI.
Introduction:
Long term anticoagulation is indicated in almost 10% of patients who undergo percutaneous coronary intervention (PCI).1 Following intervention, dual antiplatelet therapy (DAPT) consisting of a P2Y12 inhibitor and aspirin is the mainstay of treatment to prevent stent thrombosis. In a subset of patients who have a compelling indication for oral anticoagulation (OAC), the combination of an OAC and DAPT, known as triple therapy (TT), carries a significant increase risk of bleeding over DAPT alone.2 In clinical practice, providers must balance the risk of ischemic events with the risk of bleeding on OAC and antiplatelet therapy following PCI.
In recent years, TT has been evaluated in several landmark trials to better understand the risks and benefits of using an OAC with a single antiplatelet agent (double therapy) following PCI. These trials suggest that double therapy with warfarin or a direct oral anticoagulant (DOAC) is associated with decreased bleeding events when compared with TT.3,4,5 In response, the American College of Cardiology (ACC) expert consensus pathway recommends against the routine use of TT. In patients with high risk for ischemic events, prescribing TT for no more than 30 days is recommended.6
Research has shown that national consensus statements, despite being evidence based, are not always readily translated into clinical practice.7 A continuous quality improvement (QI) program may be helpful to facilitate the adoption of these guidelines into clinical practice. Proposed interventions that enhance provider adherence to guidelines include chart based reminders, computerized decision support programs, the education of physicians through local opinion leaders, and surveys of perceptions and prescribing patterns.8 Overall, a comprehensive approach to implementing a QI program can be successful at realizing practice changes at the local level.
In 2019, Beth Israel Deaconess Medical Center (BIDMC) implemented several interventions to standardize prescribing practices for patients on an OAC following PCI and enhance adherence to the ACC consensus statement. These included adoption of institutional guidelines (see appendix 1), new prescribing prompts in the computerized provider order entry system (CPOE) and a teaching curriculum for house staff on the cardiology service (Figure 1). The purpose of these interventions was to provide decision support for providers when determining therapy following PCI and improve transitions of care at discharge. The purpose of this study was to evaluate the effectiveness of these interventions by assessing discharge prescribing practices, discharge communication and faculty perceptions around therapy selection.
Figure 1.
Quality improvement algorithm. This was the algorithm used to implement the quality improvement initiative at Beth Israel Deaconess Medical Center. CPOE=Computerized provider order entry system
Methods
In 2019 a multidisciplinary team including interventional cardiology, critical care cardiology and pharmacy convened to develop a hospital wide QI initiative to optimize therapy post-PCI and minimize time on TT. The QI initiative included several key interventions that encompassed development of institutional guidelines, changes to CPOE, and an educational curriculum for house staff.
This team developed institutional guidelines with recommendations to limit TT to 30 days or less following PCI in patients requiring an OAC and a P2Y12 inhibitor. In concordance with the ACC consensus statement, the institutional guideline also advocated using DOACs, unless otherwise contraindicated, over vitamin K antagonists (VKA) given their advantageous adverse effect profile.9,10 The institutional guideline was reviewed and approved at the interventional cardiology subcommittee and then the hospital wide Pharmacy and Therapeutics committee. Indication-based dosing was added for all P2Y12 inhibitors (clopidogrel, ticagrelor and prasugrel), prompting providers to differentiate between the indications of PCI with or without OAC use in CPOE. If PCI with OAC was selected, providers were automatically directed to an electronic summary of the guidelines and a link to the full document for additional decision support. Decision support was added only to P2Y12-inhibitors given their more limited indications compared with aspirin and OACs. This was done with the intent to enhance provider adherence to the guidelines. Lastly, a one hour case based lecture was added to the general cardiology rotation curriculum which covered topics including antiplatelet and anticoagulation therapy following PCI. House staff providers were educated on the major studies, ACC consensus statements, appropriate ways to navigate CPOE decision support and the overall benefit of TT minimization.
Two years after implementation of this QI initiative, a multidisciplinary research team convened to evaluate effectiveness of the hospital wide interventions. A retrospective observational study consisting of patient chart review and a faculty survey was proposed and approved by the BIDMC institutional review board.
The primary study sample was identified from the BIDMC Clinical Data Repository spanning 2017Q4 to 2020Q4 (18 months before and after implementation of the interventions). Medications at the time of discharge were used to identify patients on OAC. Patients met inclusion criteria if they were age >18, underwent PCI at BIDMC between September 2017 and September 2020, received at least 1 coronary drug eluting stent, and were discharged on an OAC. Documents reviewed included initial cardiac catheterization laboratory procedural report, discharge summary or discharge instructions, and one year cardiology or internal medicine follow up notes if available. PCIs that met criteria for inclusion were cross referenced with the National Cardiovascular Data Registry (NCDR) CathPCI Registry Version 4 (V4) and Version 5 (V5) databases (the CathPCI Registry transitioned from V4 to V5 in April 201811), which were used to abstract baseline characteristics.
Patients were divided into 2 cohorts, a pre-guideline cohort (2017Q4–2019Q1) and a post-guideline cohort (2019Q2–2020Q3). Baseline characteristics, indications for anticoagulation therapy, discharge prescribing patterns and patient outcomes at 12 months were compared between cohorts. Ischemic events were defined as stroke, myocardial infarction or in-stent thrombosis, as documented in the electronic medical record. Major bleeding and non-major clinically significant bleeding events were classified using the Thrombolysis in Myocardial Infarction (TIMI) bleeding criteria.12 The rate of rehospitalization for any reason to the BIDMC and its sister hospitals was also assessed.
Separately, a survey was developed to measure clinicians’ perspectives of the BIDMC institutional guideline (appendix 1). This was developed by 2 physicians (WE and ES) and included multiple choice and free response questions to assess cardiology faculty perceptions of the guidelines and discharge prescribing practices following PCI. The survey was constructed in REDCAP (Nashville, TN) and was distributed electronically to all members of the academic faculty in the cardiology division at BIDMC, regardless of subspecialty.
Statistical analyses were performed using SAS 9.4 analytics software by an independent statistician. Student’s T-tests were used to compare continuous variables and chi-square tests were used to compare categorical variables. A p-value <0.05 was considered statistically significant.
Results
Among 2,797 PCIs reviewed, 431 (15.3%) met criteria for inclusion (Figure 2). Within this sample, 24.9% were female, 80.1% were white, the mean age was 74±10.9 years and the mean CHADS2VASC score was 4. The most common indications for OAC therapy were atrial fibrillation (70.0%), left ventricular apical akinesis (14.3%) and venous thromboembolism (10.9%). The most common indication for PCI was Non-ST segment elevated myocardial infarction (NSTEMI) (48.0%) followed by any form of angina (37.0%) and ST segment elevated myocardial infarction (STEMI) (13.5%).
Figure 2.
Exclusion criteria. PCI=percutaneous coronary intervention. OAC=oral anticoagulation. QI=quality improvement
Of the study cohort, 219 (50.8%) were in the pre-guidelines group and 212 (49.2%) were in the post-guidelines group. Patients in the post-guidelines group had a greater rate of heart failure (58.4% vs 40.5%) and history of stroke (29% vs 20%) and higher average clinical frailty scores (6.6 vs 5.9). All other characteristics were balanced between groups. In addition, there were no significant differences in type of OAC patients were taking between cohorts (Table 1).
Table 1.
Baseline characteristics. P values compare pre and post quality improvement initiative cohorts. Bold values with (*) are statistically significant (p<0.05). A single p value was calculated for race, anticoagulation indications and PCI indications. Clinical frailty scale range 1–9.
| Patient Characteristics | Total | Pre-quality improvement initiative | Post-quality improvement initiative | P values |
|---|---|---|---|---|
| n=431 | n=219 | n=212 | ||
| Age (years), mean (SD) | 74.0 (10.9) | 73.9 (10.7) | 74.1 (11.1) | 0.769 |
| Female | 24.9% | 22.5% | 27.3% | 0.246 |
| Race: | 0.289 | |||
| White | 80.1% | 80.6% | 79.6% | |
| Black | 7.1% | 7.2% | 6.9% | |
| Asian | 2.1% | 2.1% | 1.4% | |
| Hispanic | 5.3% | 3.7% | 7.1% | |
| Other | 5.4% | 7.1% | 5.0% | |
| Body Mass Index (kg/m2), mean (SD) | 29.5 (6.7) | 29.3 (6.4) | 29.8 (6.9) | 0.375 |
| Hypertension | 93.1% | 91.8% | 94.4% | 0.290 |
| Prior myocardial infarction | 38.9% | 39.5% | 38.3% | 0.793 |
| Prior stroke | 24.2% | 20.0% | 29.0% | 0.029* |
| Prior percutaneous coronary intervention | 39.2% | 37.7% | 40.7% | 0.532 |
| Diabetes Mellitus | 41.0% | 40.5% | 41.6% | 0.810 |
| Pre-procedure Creatine (mg/dL), mean (SD) | 1.5 (1.7) | 1.7 (2.0) | 1.4 (1.4) | 0.089 |
| Heart failure | 49.3% | 40.5% | 58.4% | <0.001* |
| Clinical frailty scale | 6.4 + 1.4 | 5.9 ± 0.4 (n=139) | 6.6 ± 1.4 | <0.001* |
| CHADS2VASC score, median [IQR] | 4 [3–5] | 4 [3–5] | 4 [3–5] | |
| Anticoagulation indication | 0.413 | |||
| Atrial fibrillation | 70.1% | 68.5% | 71.8% | |
| Venous thromboembolism | 11.0% | 9.9 % | 12.0% | |
| Left ventricular apical akinesis | 11.4% | 13.1% | 9.7% | |
| Left Ventricular thrombus | 3.0% | 4.1% | 1.9% | |
| Other | 4.5% | 4.4% | 4.6% | |
| Indication for PCI | 0.385 | |||
| STEMI | 13.5% | 12.0% | 15.0% | |
| NSTEMI | 48.5% | 51.0% | 46.0% | |
| Unstable angina | 3.5% | 1.8% | 5.1% | |
| Stable angina | 9.2% | 10.0% | 8.4% | |
| Other indication | 25.1% | 25.0% | 25.2% |
Following implementation of the guidelines, the mean duration of time patients spent on TT (ascertained afterwards) decreased from 58.7 days down to 37.8 days (p=0.02). Patients discharged on TT≤ 30 days increased from 24% to 37% (p=0.019), while patients discharged on dual therapy increased from 26% to 33% (p=0.02). In addition, there was greater documentation in the discharge summary/instructions of the TT duration plan post-PCI following the guidelines intervention, increasing from 80% to 89% (p<0.01; Figure 3).
Figure 3.
Triple therapy (TT) prescription categories at discharge. P<0.001 for all prescription categories at discharge. TT regimen included aspirin, oral anticoagulant (OAC) and P2Y12 inhibitor. Following 30 days of TT, aspirin was discontinued and P2Y12 inhibitor was continued with OAC. In patients where TT was continued beyond 30 days, aspirin was continued until TT was stopped. Of note 2 patients were discharged on aspirin and OAC alone, which are not represented in the chart above. QI=quality improvement.
Post guidelines patients discharged on warfarin decreased from 50.2% to 29.2%, while patients discharged on apixaban increased from 20.5% to 47.6% (p<0.01; Figure 4). The percentage of patients discharged on clopidogrel decreased from 96.4% to 90.3% (p=0.01), while patients discharged on ticagrelor increased 2.7% to 8.8% (p=0.01). Aspirin usage decreased by a non-statistically significant amount (77.5% to 73.1%, p=0.29).
Figure 4.
Oral anticoagulants (OAC) on admission and discharge pre/post quality improvement (QI) initiative. Combined p-value compares discharge medications pre and post QI initiative. P-value <0.05 considered significant.
Among outcomes ascertained from patients who had follow up care obtained at BIDMC or a partner hospital during the 1st year after stenting (78.2%, n=337), there was a decrease in ischemic events from pre-guidelines intervention to post-guidelines intervention (17.0% vs 8.4%, p=0.03). Other outcomes including major bleeding events, clinically significant non-major bleeding events and rehospitalization rates were not significantly different (Figure 5).
Figure 5.
Adverse outcomes at 12 months pre and post quality improvement (QI) initiative. Total patients with follow up total n=337 (78.2%), 72.1% in pre-QI initiative cohort, 84.4% in the post-QI initiative cohort. Patients who did not follow up at Beth Israel Deaconess Medical Center or a clinical partner were lost to follow up. Stratification of bleeding events was defined by Thrombolysis in Myocardial Infarction (TIMI) bleeding criteria. P value compares ischemic events in pre and post QI initiative cohorts. P<0.05 considered significant.
Among the surveyed faculty (total n=78) there was a 26% response rate (n=20). The most common subspecialties were general cardiology (n=11; 55.0%), interventional cardiology (n=5; 25.0%) and imaging (n=4; 20.0%). The majority of respondents (75.0%) reported familiarity with the BIDMC institutional guidelines and 57.9% reported using them to make post-PCI therapy decisions. Of the respondents, 55.0% reported typically prescribing TT for exactly 30 days, while 40.0% reported prescribing for less than 30 days. Also, 65% of the respondents reported that discharge therapy decisions were made by the interventional cardiologist at the time of PCI, while 35% reported that therapy decisions were made primarily by the inpatient cardiology attending after assuming care. Lastly, faculty reported the most common reasons for changing therapy were bleeding event, bleeding risk (without a bleeding event) and medication cost.
Discussion
We report a hospital wide QI initiative that included adoption of institutional guidelines, new prescribing prompts in CPOE and a teaching curriculum for house staff on the cardiology service. The BIDMC QI initiative, which included the central recommendations for the 2020 ACC expert consensus pathway, was associated with a reduction in the use and duration of TT following PCI.
Following implementation, there was a significant increase in the number of patients discharged on double therapy compared with TT. When TT was prescribed, a greater number of patients were discharged on 30 days or less and there was a lower incidence of incomplete documentation at discharge of intended duration. A literature search from 2020–2022 yielded no previous QI initiatives in this high-risk population. Our findings suggest that the BIDMC targeted interventions provided effective clinical decision support and standardized care. This was further supported by the statistically significant increase in the number of patients discharged on a DOAC in post-intervention cohort. This finding is consistent with the ACC consensus statement which endorses DOAC use over VKA agents. Clopidogrel was the most commonly prescribed P2Y12 inhibitor, as it was the most commonly used in the landmark trials and is endorsed by our institution’s specific guideline.iiiivv Faculty members were more likely to select clopidogrel because of its advantageous safety profile and lower rates of bleeding when coupled with an OAC agent compared with the other P2Y12 inhibitors. Although this study was not powered to detect differences in patient outcomes, there was a statistically significant decrease in the number of ischemic events at one year. There were also non-significant decreases in the one year rate of rehospitalization and non-actionable bleeding.
Lastly, faculty providers reported using the institutional guidelines to aid in their clinical decision making, aligning with the findings of this study. Although 95% of faculty members specified that they would prescribe TT for 30 days or less, this study found a higher proportion of patients were prescribed TT for longer than 30 days, suggesting that faculty underestimate the prevalence of extended TT use or that respondents were not representative of the faculty practice as a whole. Additionally, faculty reported that they preferred prescribing a DOAC over a VKA and preferred clopidogrel.
This study has several limitations. First, both populations were overwhelmingly white, elderly and with multiple comorbidities, which may limit generalizability. Second, the sample size in this study was small, and there was limited power to detect differences in outcomes between groups. Third, follow up data at one year was only available for patients who were seen in the BIDMC health system or in one of its sister hospitals. As such, clinical follow-up data were not available for all study participants. Fourth, house staff education may not have impacted prescribing patterns or discharge documentation for patients managed under the cardiology nurse practitioner service, which cares for approximately 60% of PCI patients overall, but only 25% of inpatient PCIs. Fifth, angiographic and lesion characteristics were not included in our data collection, which may have impacted the selection of antithrombotic therapy. Additionally, despite the majority of clinical faculty participating in the study, our survey had a low response rate. Not all faculty members do the same amount of inpatient service time or clinic time, so their likelihood of managing post-PCI OAC and antiplatelet therapies differs. Lastly, this study was performed at an academic tertiary care hospital, with sufficient resources to implement this QI initiative. At institutions without comparable resources, it may be challenging to implement similar initiatives.
Conclusions
A multidisciplinary QI initiative involving adoption of institutional guidelines, new prescribing prompts in CPOE and a teaching curriculum for house staff led to a reduction in the frequency and duration of prescribed TT after PCI. At other academic tertiary institutions where prescribing practices post-PCI for patients on OACs remains heterogeneous, similar approaches should be considered to incorporate consensus recommendations into clinical practice.
Supplementary Material
Acknowledgements:
The authors would like to acknowledge Kevin Kennedy for his contributions to this manuscript.
Funding:
Dr. Secemsky is funded in part by NHLBI K23HL150290.
Footnotes
Conflicts of Interest
The authors state that they have no conflict of interest.
An abstract of this work was presented at the SCAI Scientific Sessions in Atlanta, Georgia, USA on 5/21/22.
Data Availability Statement
The data that support the findings of this study are not publicly available because they contain protected health information that could compromise the privacy of research participants. Deidentified data are available on reasonable request from the corresponding author, ES.
References
- 1.Secemsky EA, Butala NM, Kartoun U, Mahmood S, Wasfy JH, Kennedy KF, et al. Use of chronic oral anticoagulation and associated outcomes among patients undergoing percutaneous coronary intervention. J Am Heart Assoc. 2016;5(10) [DOI] [PMC free article] [PubMed] [Google Scholar]; A large retrospective assessment of the prevalence of triple therapy prescription in patients undergoing percutaneous coronary intervention. Demonstrates an increased association with in-hospital bleeding, readmission, and decreased long-term survival in those patients receiving oral anticoagulation
- 2.Hansen ML, Sørensen R, Clausen MT, et al. Risk of bleeding with single, dual, or triple therapy with warfarin, aspirin, and clopidogrel in patients with atrial fibrillation Arch Intern Med, 170 (2010), pp. 1433–1441 [DOI] [PubMed] [Google Scholar]
- 3.Gibson CM, Mehran R, Bode C, Halperin J, Verheugt F, Wildgoose P, et al. An open-label, randomized, controlled, multicenter study exploring two treatment strategies of rivaroxaban and a dose-adjusted oral vitamin K antagonist treatment strategy in subjects with atrial fibrillation who undergo percutaneous coronary intervention (PIONEER AF-PCI). Am Heart J. 2015;169(4):472–8.e5. [DOI] [PubMed] [Google Scholar]
- 4.Cannon CP, Bhatt DL, Oldgren J, Lip GYH, Ellis SG, Kimura T, et al. Dual antithrombotic therapy with dabigatran after pci in atrial fibrillation. N Engl J Med. 2017;377(16):1513–24. [DOI] [PubMed] [Google Scholar]
- 5.Lopes RD, et al. “Antithrombotic therapy after acute coronary syndrome or PCI in atrial fibrillation”. The New England Journal of Medicine. 2019. 380(16):1509–1524. [DOI] [PubMed] [Google Scholar]
- 6.Tomaselli GF, Mahaffey KW, Cuker A, Dobesh PP, Doherty JU, Eikelboom JW, … & Wiggins BS (2020). 2020 ACC expert consensus decision pathway on management of bleeding in patients on oral anticoagulants: a report of the American College of Cardiology Solution Set Oversight Committee. Journal of the American College of Cardiology, 76(5), 594–622. [DOI] [PubMed] [Google Scholar]
- 7.Cabana MD, Rand CS, Powe NR, et al. Why don’t physicians follow clinical practice guidelines? A framework for improvement. JAMA 1999;282:1458–65. [DOI] [PubMed] [Google Scholar]
- 8.Bottorff MB, Nutescu EA, & Spinler S (2007). Antiplatelet Therapy in Patients with Unstable Angina and Non–ST-Segment-Elevation Myocardial Infarction: Findings from the CRUSADE National Quality Improvement Initiative. Pharmacotherapy: The Journal of Human Pharmacology and Drug Therapy, 27(8), 1145–1162. [DOI] [PubMed] [Google Scholar]
- 9.Granger CB, Alexander JH, McMurray JJ, Lopes RD, Hylek EM, Hanna M, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011;365(11):981–92. [DOI] [PubMed] [Google Scholar]
- 10.Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365(10):883–91. [DOI] [PubMed] [Google Scholar]
- 11.Jennings HS, Kutcher MA & Messenger JC. How to Incorporate New Version 5 of the NCDR CathPCI Registry Into Your Institution and Clinical Practice. Society for Cardiovascular Angiography and Interventions. Published online 11/13/18. https://scai.org/how-incorporate-new-version-5-ncdr-cathpci-registry-your-institution-and-clinical-practice
- 12.Bovill EG, Terrin ML, Stump DC, Berke AD, Frederick M, Collen D, Feit F, Gore JM, Hillis LD, Lambrew CT, et al. Hemorrhagic events during therapy with recombinant tissue-type plasminogen activator, heparin, and aspirin for acute myocardial infarction. Results of the Thrombolysis in Myocardial Infarction (TIMI), Phase II Trial. Ann Intern Med. 1991. Aug 15;115(4):256–65. doi: 10.7326/0003-4819-115-4-256.. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are not publicly available because they contain protected health information that could compromise the privacy of research participants. Deidentified data are available on reasonable request from the corresponding author, ES.