Author + information
- Received March 24, 2010
- Revision received April 29, 2010
- Accepted May 3, 2010
- Published online August 1, 2010.
- Mehul Patel, MD,
- Michael Kim, MD,
- Rucha Karajgikar, MBBS,
- Visali Kodali, MD,
- Dheeraj Kaplish, MBBS,
- Paul Lee, MD,
- Pedro Moreno, MD,
- Prakash Krishnan, MD,
- Samin K. Sharma, MD⁎ ( and )
- Annapoorna S. Kini, MD
- ↵⁎Reprint requests and correspondence:
Dr. Samin K. Sharma, Mount Sinai Hospital, Box 1030, Cardiac Catheterization Lab and Intervention, One Gustave L. Levy Place, New York, New York 10029-6754
Objectives This study evaluated the outcomes of patients discharged the day of percutaneous coronary intervention (PCI) by analyzing the data from a single-center, large, multioperator registry of interventions.
Background Although same-day discharge is likely safe after interventions on low-risk stable patients, there is limited data to guide selection of a broader population of patients. Due to numerous patient variables and physician preferences, standardization of the length of stay after PCI has been a challenge. Most of the reported studies on same-day discharge have strict inclusion criteria and hence do not truly reflect a real-world population.
Methods We analyzed the outcomes of consecutive same-day discharge in 2,400 of 16,585 patients who underwent elective PCI without any procedural or hospital complication. Composite end point included 30-day major adverse cardiac cerebral events and bleeding/vascular complications.
Results The mean age of the study population was 57.0 ± 23.7 years with 12% aged over 65 years. Twenty-eight percent received glycoprotein IIb/IIIa inhibitor with closure devices in 90.5%. Clinical and angiographic success was noted in 97% of all PCIs. The average length-of-stay following PCI was 8.2 ± 2.5 h. The composite end point was reached in 23 patients (0.96%). Major adverse cardiac cerebral events occurred in 8 patients (0.33%) and vascular/bleeding complications in the form of Thrombolysis In Myocardial Infarction minor bleeding in 14 patients (0.58%) and pseudoaneurysm in 1 patient (0.04%).
Conclusions When appropriately selected, with strict adherence to the set protocol, same-day discharge after uncomplicated elective PCI is safe despite using femoral access in a wide spectrum of patients.
Previous studies have demonstrated the safety and outcomes of short-term observation after percutaneous coronary intervention (PCI) with cautious patient selection (1,2). Early ambulation after femoral-approach PCI is now considered safe without an increase in the vascular complications (2). Developments in the interventional cardiology armamentarium with better stents and adjunctive antithrombotic therapy has led to a significant improvement in the PCI outcomes (1,2). Prior studies that investigated same-day discharge were limited by stringent inclusion criteria for patient selection and small numbers (1,2). These studies for example, demonstrated safety of same-day discharge in patients with transradial or transfemoral approach with sparing use of glycoprotein (GP) IIb/IIIa inhibitors, drug-eluting stents (DES), or various closure devices (1–6). Accordingly, we sought to evaluate the safety of same-day discharge after elective PCI among patients, irrespective of age, deemed suitable for same-day discharge with respect to the selection criteria designed by the institution.
This is a retrospective assessment of prospectively collected data on 2,400 patients discharged on the same day after an uncomplicated, elective PCI performed at the Mount Sinai Hospital, New York, from April 2003 through March 2008. All patients had their procedural details recorded and maintained in a large institutional review board–approved database. These included patients scheduled for a diagnostic coronary angiogram undergoing ad hoc PCI as well. The selection criteria were designed to ensure maximum patient safety. The institutional triage criteria were specifically designed for patients in need for an extended in-hospital stay.
The selection criteria for same-day discharge, on arrival and up to completion of procedure, included:
1. Scheduled or ad hoc elective PCI for stable angina or crescendo angina or asymptomatic but positive stress test or perfusion imaging or stenosis on coronary computerized tomography
2. Successful PCI with or without the use of a bolus dose of intravenous GP IIb/IIIa inhibitors
3. Residual <30% stenosis without type C or higher intimal dissection of National Heart, Lung, and Blood Institute classification
4. Left ventricular ejection function >30%
5. Absence of post-procedural chest pain
6. Absence of vascular complications
7. PCI performed before 3 pm to allow 8 h of observation before discharge on the same day
8. Unsuccessful but uncomplicated PCI
9. Successful completion of a 200-m walk
The institutional triage criteria for in-hospital admission included:
1. Acute coronary syndrome presenting as rest angina or myocardial infarction (MI)
2. Use of full-recommended bolus and infusion of GP IIb/IIIa inhibitors
3. Complex PCI (unprotected left main intervention, 2-stent strategy for bifurcation lesions or interventions on a single vessel supplying the heart)
4. Assisted PCI (intra-aortic balloon pump or Impella [ABIOMED, Inc., Danvers, Massachusetts]-assisted PCI)
5. Significant procedural complications
6. Periprocedural hemodynamic instability
7. Serum creatinine >1.5 mg/dl not on hemodialysis
8. Sustained ventricular or atrial arrhythmia
9. Left ventricular ejection fraction of <30% or decompensated systolic heart failure
10. Uncontrolled diabetes mellitus
11. Patients requiring optimization of medications
12. Patients who experienced difficulties in ambulation
13. Socioeconomic issues regarding placement and post-PCI care
Percutaneous coronary intervention was performed via the femoral approach in 99.5% of patients using 5- to 8-F guiding catheters. All patients were pre-treated with aspirin 81 to 162 mg and clopidogrel 300 to 600 mg orally before the procedure. The anticoagulation protocol included either intravenous bivalirudin (Angiomax, The Medicines Company, Parsippany, New Jersey) or intravenous heparin. Additional doses of intravenous heparin or bivalirudin (if needed) were given if the procedure lasted >30 min to maintain an activated clotting time (ACT) between 300 and 350 s. If GP IIb/IIIa inhibitor was used, ACT was maintained between 200 and 250 s. Activated clotting time was checked after 5 min of intravenous heparin or bivalirudin and every 30 min thereafter.
Vascular access management
Transradial interventions were performed in only 11 patients. The transradial arterial sheath was removed immediately after the guiding catheter was pulled out and an occlusive tourniquet (TR Band, Terumo, Elkton, Maryland) was applied at the puncture site for at least 3 h. Pressure was gradually released until hemostasis was achieved. Closure devices were used in 90.5% of all transfemoral access patients. These included Perclose (Abbott Laboratory, Redwood City, California), Angio-Seal (St. Jude Medical, St. Paul, Minnesota), Mynx (Access Closure, Mountain View, California), and Starclose (Abbott Vascular). The rest of the patients had the intra-arterial sheaths removed 2 to 3 h after PCI. Hemostasis was then achieved by manual compression and maintained with a femoral compression device (FemoStop System, RADI Medical Systems, Inc., Wilmington, Maryland).
All patients had a 12-lead electrocardiogram done following PCI and before discharge. Cardiac enzymes (creatine kinase-myocardial band [CK-MB]), complete blood count with platelets, and Sma7 were measured pre-procedure and at 4 to 8 h after procedure.
After PCI, patients were observed in the post-procedure telemetry care unit by staff that is well trained to manage post-PCI complications. Post-interventional therapy included 81 to 162 mg/day of aspirin and 75 mg/day of clopidogrel for 1 to 12 months in case of a bare-metal stent and for 1 to 3 years in case of a DES.
All patients were ambulated after 2 to 3 h of bed rest if a closure device was used or after 4 to 6 h if the sheath was manually removed. Patients who underwent transradial PCI were ambulated after 2 to 3 h of procedure. We followed our ambulation protocol, which involved ambulation for 5 min, with a walking distance of 200 m before discharge. Any patient who failed to accomplish the set ambulatory targets was deemed unsuitable for same-day discharge. Vital sign check, vascular access site, and distal vascular bed examination were done immediately after ambulation for a comprehensive analysis before discharge.
Suitability for discharge required freedom from symptoms, absence of electrocardiogram changes, no CK-MB or serum creatinine elevation, no drop in hematocrit, absence of puncture site abnormalities and successful ambulation. Written instructions and oral explanation of all possible events were given to all patients. Before discharge, patients were instructed on how to achieve hemostasis by local pressure for puncture-related bleeding. In case of an emergency, patients were instructed to contact the general practitioner, their referring cardiologist, the interventionalist, or the nearest emergency department. All patients received pre-discharge counseling on diet and lifestyle modification. The management of medication compliance was repeatedly highlighted by the interventionalist as well as the nurse practitioner directly involved in the patient care. With successful ambulation in the post-procedure telemetry unit, patients were instructed to climb only 1 flight of stairs and lift not more than 10 lbs for 3 days after the procedure.
All patients received a follow-up phone call from a dedicated catheterization laboratory nurse on the next working day to address any cardiac complaints along with entry site–related problems. Thereafter, patients were scheduled for an office visit with the primary medical doctor or the referring cardiologist at 1 to 4 weeks.
A 30-day phone call was made to address for repeat chest pain, bleeding, readmission to same or another hospital, repeat intervention, medication compliance, MI, and death. Exact cause of readmission was identified along with overall outcome. Data on all major and minor complications were entered in the interventional database. A questionnaire, along with a self-addressed envelope, was mailed to patients who could not be reached by telephone.
Elective PCI: any coronary revascularization in a low-risk patient who presents to the facility for a planned PCI or for a coronary angiogram followed by ad hoc PCI (3,4).
CK-MB elevation: any elevation above the upper limit of normal (CK-MB by stat >10.4 ng/ml or CK-MB mass >6.3 ng/ml).
Angiographic success: <30% diameter obstruction after procedure with Thrombolysis in Myocardial Infarction flow grade 3 at end of procedure.
Clinical success: angiographic success of at least 1 lesion and no major complications.
ACC/AHA classification: modified American College of Cardiology/American Heart Association classification grading lesions into A, B1, B2, or C categories.
Calcification: readily apparent fluoroscopic densities in the lesion or vessel.
Procedural complication: included prolonged chest pain, transient closure, no-flow or slow-flow phenomenon, hemodynamic instability, persistent electrocardiographic changes, side-branch occlusion of >1.5 mm, or an angiographically suboptimal result.
Major adverse cardiac and cerebral events: clinical events following the interventional procedure that included death, MI, cerebrovascular accident, or repeat target vessel revascularization.
Bleeding: Thrombolysis in Myocardial Infarction criteria were used to define bleeding. Major: Intracranial hemorrhage or a ≥5 g/dl decrease in the hemoglobin concentration or a ≥15% absolute decrease in the hematocrit. Minor: Observed blood loss: ≥3 g/dl decrease in the hemoglobin concentration or ≥10% decrease in the hematocrit. No observed blood loss: ≥4 g/dl decrease in the hemoglobin concentration or ≥12% decrease in the hematocrit.
Vascular complication: occurrence of small (<5 cm), moderate (5 to 12 cm), or large hematoma (>12 cm), pseudoaneurysm, arteriovenous fistula, or any closure device–related complications such as abrupt closure, dissection, thrombosis requiring invasive vascular intervention or surgery.
The end point of the study was the composite of major adverse cardiac, including repeat revascularization and cerebral, events along with access site vascular/bleeding complications at 30 days. The diagnosis of MI was based on symptoms and typical electrocardiogram changes combined with CK-MB isoenzyme elevations >3 times the upper limit of normal.
The data were entered in an Excel database (Microsoft Corp., Redmond, Washington) and transferred to JMP 8.0 (SAS Institute, Cary, North Carolina) for analysis. Continuous variables were described as mean ± SD and categorical variables were expressed as percentages.
During the study period from April 2003 to March 2008, a total of 20,227 PCIs were performed, of which 16,585 were elective. Of the 16,585 elective PCIs, 2,438 fulfilled the criteria for the same-day discharge, and the other 14,147 patients had 1 or more pre-defined set(s) criteria for in-hospital admission, elaborated in detail in Figure 1. Thirty-eight patients for planned same-day discharge (n = 2,438) could not go home due to small hematoma (n = 16), persistent oozing (n = 4), refusal due to different social factors (n = 10), or gait abnormalities or vasomotor instability with difficulty in ambulation (n = 8). A total of 2,400 patients were discharged on the same day, after PCI. Therefore, only 1.6% of the planned patients for same-day discharge were admitted.
Figure 1 shows the 2,400 patients, the selection and triage criteria, along with the study design. The mean age of the total study population was 57.0 ± 23.7 years with 12% aged 65 years and above. Of these, 1,814 patients (75.5%) were men, 1,038 (43.3%) had diabetes mellitus, and 2,096 (87.3%) had systemic hypertension. The mean peak ACT achieved during the intervention was 314.8 ± 60.5 s. The study population included multilesion interventions in 375 (15.6%), complex lesion anatomy such as type B2 and C lesions in 1,617 (67.3%), bifurcation interventions in 350 (14.5%), and total occlusions in 98 (4.1%) patients. Tables 1 and 2⇓⇓ show the baseline characteristics and the angiographic data of the study population.
Interventional procedure results
Interventional procedure was successful in 2,329 of 2,400 (97.0%) same-day discharge patients. In the remaining 3% of cases, PCI was unsuccessful largely due to inability to wire the lesion/vessel (such as in chronic total occlusions) but without any intraprocedural or post-procedural complication. The mean diameter stenosis improved from 81.1 ± 10.3% to 11.7 ± 16.9%. Coronary stents were placed in 1,990 (83.5%) patients, of which 76.5% were DES. There were no closure device–related complications noted at 30 days. Table 3 shows the procedural characteristics.
There was no incidence of any major complications during the 6- to 8-h triage period. Fourteen patients developed a small hematoma, which was timely diagnosed and effectively reduced by manual compression. There was no drop in the hematocrit and all were discharged on the same day. One patient developed a pseudoaneurysm of the superficial femoral artery after 2 days, requiring thrombin injection in a day care setting. There was no death, MI, or cerebrovascular accident at day 30 in any of the patients. One patient died after 30 days due to MI secondary to stent thrombosis. This was notably due to clopidogrel resistance with 0 platelet inhibition despite good medication compliance. Eight patients required readmission within 30 days of the procedure for planned staged nontarget-vessel revascularization. Overall incidence of 30-day readmission was low (28 patients, 1.17%) with 19 due to cardiac reasons (4 coronary artery bypass grafts, 4 target-vessel revascularizations, 8 planned staged PCI, and 3 recurrent chest pain with unremarkable noninvasive/catheterization findings) and 9 due to noncardiac reasons (3 vascular, including 1 pseudoaneurysm, 2 pulmonary, 3 gastrointestinal, and 1 near-syncope episode due to dehydration) (Table 4).
With improvements in the surgical techniques and safety over the past several years, there has been a movement toward performing a large and wide spectrum of noncardiac surgeries in the outpatient setting. It is now estimated that over 65% of surgeries performed in North America are done so in the outpatient setting. On the contrary, despite significant improvement in interventional technologies in recent years, there has been only marginal reduction in the post-PCI length of stay (LOS). The LOS following PCI is one of the major determinants of hospital cost and quality-of-care assessment. Reimbursement policies governing the LOS for elective PCI may possibly contribute to unwarranted hospitalizations causing logistic constraints on health care resources.
There have been few appraisals of same-day discharge in a real-world spectrum of practice due to extreme variations in the LOS between countries, regions, and hospitals. Length of stay shows a decreasing trend over time, and shorter LOS does not appear to affect health outcomes adversely as demonstrated in various other day care procedures (7–26).
The first study on same-day discharge reported by Kiemeneij et al. (18) clearly demonstrated safety of early ambulation after transradial PCI. This was followed by the study on 922 patients reported by Koch et al. (19) that showed short-term triage of 4 h as sufficient and safe for same-day discharge. However, this was a highly selective study with guiding catheter size restricted to 6-F and only 20% use of stents. Because these patients underwent PCI using the femoral approach without the use of closure devices, a sizable number of patients were discharged back to the referring hospital for overnight care. Slagboom et al. (21) later reported safety of transradial PCI with 6-F guiding catheters and 40% usage of stents in the OUTCLAS (Outpatient Coronary Low-Profile Angioplasty Study) trial. Subsequent studies have demonstrated safety of the use of bivalirudin and GP IIb/IIIa inhibitors when discharging patients in the same day (20–30).
To the best of our knowledge, this is the largest study of same-day discharge after elective PCI using the femoral approach in a wide range of patients. The current study has a limited number of exclusions and used more device variations such as larger guide catheters, complex interventions requiring rotablation, a uniform use of stents, and various closure devices. The patients included in our study had diverse demographics, comorbidities, and risk factors with complex coronary lesions that match the real-world setting of a tertiary referral center. There were very few post-PCI cardiac events or vascular access complications even in the elderly population (>65 years). However, this comprised only 12% of the entire PCI population. The 6- to 8-h observation in the recovery unit with the necessary post-PCI care and education seems adequate to triage patients for either same-day discharge or extended overnight in-hospital admission (12–15). Our study shows that early physical activity after PCI does not lead to a higher rate of puncture-related complications despite femoral approach. A walking distance of 200 m must be strictly enforced to effectively screen for potential access site complications during triage. The telemetry unit dropout rate during the 4- to 8-h post-procedure observation was 1.6%, excluding only 38 of 2,438 patients. These patients required extended overnight observation for reasons listed in Figure 1. This underscores the usefulness and suitability of the selection criteria in the current study for same-day discharge after uneventful coronary intervention in a selected group of elective patients. The 30-day major adverse cardiac cerebral events along with 30-day vascular complications for same-day discharge PCIs were low and comparable to the other studies reported to date. This is the largest study to assess the outcomes of same-day discharge using the femoral approach with similar access site complication rates as reported with brachial or radial approach (1). The lower incidence of major adverse cardiac cerebral events may be explained by the exclusion of acute coronary syndrome patients and further reconfirms the appropriateness of the selection criteria implemented in the current study. Our study in a diverse patient population supports the importance of careful selection guidelines for determining the suitability of same-day discharge so that in-hospital resources can be predictably allocated.
Readmission for puncture-related complications was exceedingly rare, occurring in only 1 patient, who developed a pseudoaneurysm of the superficial femoral artery, presenting after 48 h of the procedure. This femoral access site was closed using the Mynx device. The pseudoaneurysm was effectively treated by thrombin injection without the need for vascular surgery. Importantly, our study markedly differs from the randomized trial reported by Heyde et al. (2) by analyzing 3 times the number of patients with markedly diverse procedural characteristics. We included patients with ad hoc PCI in our analysis unlike the EPOS (Elective PCI in Outpatient Study) trial (2,7–10). The current study has a higher usage of DES and closure devices that were sparingly used in the EASY (Early Discharge After Transradial Stenting of Coronary Arteries) and EPOS studies (1,2). In the current study, 139 patients (5.8% of the study population) requiring 7- and 8-F guides for intervention were successfully discharged on the same day, unlike in the EPOS study in which such patients were excluded. Our study used a variety of closure devices unlike the study reported by Carere et al. (31) that showed safety of only suture-based closure devices. Despite the advent of radial interventions, femoral approach is still the favored technique worldwide. Based on the current study, the factors considered favorable for same-day discharge are outlined in Table 5, which can safely be applied to a large number of patients undergoing elective PCI. Our criteria for ambulatory PCI are in line with the recent recommendations made by the Society of Coronary Angiography and Interventions (also endorsed by the American College of Cardiology) in this field (32).
Moreover, the present study demonstrates that this can be achieved without compromising the quality-of-care or safety in the patient population with a higher-risk profile. Also, once the protocol for ambulatory PCI is set in place, it does not incur extra costs on the hospital system to follow these patients and it does not add any costs to the physicians.
As a retrospective study, we are unable to account for the differences in baseline characteristics that may influence outcome. Our study was not primarily designed to identify predictors of PCI-related complications for same-day discharge. Asymptomatic elevations in CK-MB and serum creatinine 6- to 8-h after the procedure may have been missed. A possible selection bias due to involvement of multiple providers and operators cannot be ruled out. Because closure device was used in most of the same-day discharge patients, the results of this study should not be translated when only manual compression or FemoStop devices are used.
With strict adherence to the set protocol as outlined in the current study, we conclude that same-day discharge after uncomplicated elective PCI is safe in a wide range of patients despite using femoral access. Same-day discharge is likely to improve patient comfort and reduce utilization of hospital resources.
The authors have reported that they have no relationships to disclose.
- Abbreviations and Acronyms
- activated clotting time
- creatine kinase-myocardial band
- drug-eluting stent(s)
- length of stay
- myocardial infarction
- percutaneous coronary intervention
- Received March 24, 2010.
- Revision received April 29, 2010.
- Accepted May 3, 2010.
- American College of Cardiology Foundation
- Bertrand O.F.,
- De Larochelliere R.,
- Rodes-Cabau J.,
- et al.,
- Early Discharge After Transradial Stenting of Coronary Arteries Study Investigators
- Heyde G.S.,
- Koch K.T.,
- de Winter R.J.,
- et al.
- Krone R.J.,
- Shaw R.E.,
- Klein L.W.,
- et al.,
- American College of Cardiology-National Cardiovascular Data Registry
- Smith S.C. Jr..,
- Feldman T.E.,
- Hirshfeld J.W. Jr..,
- et al.
- Patel M.R.,
- Dehmer G.J.,
- Hirshfeld J.W.,
- et al.
- Wu C.,
- Hannan E.L.,
- Walford G.,
- et al.
- Ellis S.G.,
- Vandormael M.G.,
- Cowley M.J.,
- et al.
- Koch K.T.,
- Piek J.J.,
- de Winter R.J.,
- et al.
- Nikolsky E.,
- Mehran R.,
- Halkin A.,
- et al.
- Clarke A.,
- Rosen R.
- Kiemeneij F.,
- Laarman G.J.,
- Slagboom T.,
- van der Wieken R.
- Koch K.T.,
- Piek J.J.,
- Prins M.H.,
- et al.
- Banning A.P.,
- Ormerod O.J.,
- Channon K.,
- et al.
- Porto I.,
- Blackman D.J.,
- Nicolson D.,
- et al.
- Kumar S.,
- Anantharaman R.,
- Das P.,
- et al.
- Chambers C.E.,
- Dehmer G.J.,
- Harrington R.A.,
- et al.,
- Society for Cardiovascular Angiography and Interventions