Same-day discharge (SDD) following percutaneous coronary intervention (PCI) is increasingly used, yet data on its safety after chronic total occlusion PCI (CTO PCI) remain limited. We retrospectively analyzed all CTO PCI procedures performed at a single US center between 2019 and 2023, comparing patients discharged the same day with those who were not. The primary endpoint was 30-day major adverse cardiac and cerebrovascular events (MACCE) or rehospitalization. Analyses included multivariable Cox proportional hazards regression and recursive partitioning. Among 1,273 patients, 352 (27.6%) underwent SDD. SDD patients had lower co-morbidity burden (Charlson Comorbidity Index [CCI] 3.6 ± 3.1 vs 4.2 ± 2.8, p = 0.002) and better renal function (estimated glomerular filtration rate 78.6 ± 24.2 vs 73.7 ± 23.3 ml/min/1.73 m², p = 0.001). In-hospital, MACCE occurred in 0% of SDD versus 2.0% of non-SDD patients (p < 0.001). At 30 days, MACCE/rehospitalization was significantly lower in the SDD group (4.3% vs 7.9%, p = 0.021). Multivariable analysis identified body mass index (adjusted hazard ratio [aHR] 1.09; 95% confidence interval [CI] 1.02 to 1.16; p = 0.010), CCI (aHR 2.16; 95% CI 1.37 to 4.49; p = 0.003), and antegrade dissection and reentry use (aHR 1.86; 95% CI 1.09 to 3.34; p = 0.040) as independent correlates of 30-day adverse events, while SDD itself was not. A recursive partitioning model identified CCI, estimated glomerular filtration rate, and distance to the PCI center (driving miles) as the strongest discriminators of risk within the SDD cohort. In this large contemporary series, SDD after CTO PCI was safe and feasible in carefully selected patients. Comorbidity burden, renal function, procedural complexity, and patient geography may help guide SDD decision-making in modern CTO practice.
Same-day discharge (SDD) following percutaneous coronary intervention (PCI) is both feasible and safe, offering several advantages, including reduced hospitalization costs and improved patient satisfaction. , As healthcare systems continue to prioritize efficiency and value-based care, SDD has become an increasingly attractive strategy in routine PCI practice. However, chronic total occlusion (CTO) PCI remains substantially more resource-intensive than non-CTO PCI, with costs driven by greater procedural complexity, higher complication rates, and increased utilization of specialized equipment. The mean cost of index hospitalization for CTO PCI in the OPEN CTO registry was $17,048 ± $9,904, and national estimates demonstrate rising hospitalization costs for CTO PCI. Despite these higher costs, procedural and total contribution margins remain positive, underscoring the importance of identifying strategies, such as SDD, that may safely streamline resource utilization in this complex patient population.
Although SDD has been increasingly adopted after PCI, evidence specific to CTO PCI is limited and heterogeneous. ,,,,, Early studies assessing SDD feasibility included small cohorts and were conducted during previous eras of CTO practice, ,, limiting their relevance to contemporary techniques characterized by higher success rates and broader use of dissection/reentry and retrograde strategies. Other investigations have focused exclusively on in-hospital outcomes, without evaluation of 30-day safety or rehospitalization. Additional studies have enrolled only select patient groups—for example, individuals undergoing large-bore access PCI —or have been conducted entirely outside the United States, reducing their generalizability to American healthcare environments. , Moreover, previous SDD analyses have not isolated CTO PCI patients, instead combining them within broader PCI cohorts, making it challenging to draw CTO-specific conclusions. As a result, the safety profile, correlates of adverse outcomes, and optimal patient selection criteria for SDD in contemporary CTO PCI remain incompletely defined.
This study aimed to evaluate the feasibility, safety, and correlates of 30-day outcomes associated with SDD after CTO PCI in a large, contemporary, single-center cohort and to identify patient and procedural characteristics that may guide selection of optimal SDD candidates.
Methods
Study design and population
This retrospective study included all patients who underwent CTO PCI at the University of Washington Medical Center (UWMC) between January 2019 and December 2023. The analytic cohort consisted of patients who were discharged either on the same day as the CTO PCI procedure or after observation.
Manual chart review and dedicated, purpose-built structured query language code were used to extract and merge data into the institutional PCI database. This automated process integrated relevant information from multiple institutional sources, including the hospital’s electronic health record (Epic, Epic Systems, Verona, Wisconsin), the National Cardiovascular Data Registry (NCDR) CathPCI Registry submission, and additional internal databases. , Variables obtained through the CathPCI Registry adhere to standardized definitions and are subject to routine auditing and quality assurance. Procedural equipment utilization, including intravascular imaging and adjunctive devices, was derived from real-time cath lab inventory scanning at the point of use. Core clinical variables (e.g., laboratory values, left ventricular ejection fraction [LVEF], and medications) were obtained directly from the electronic health record and are either intrinsically validated or routinely confirmed during clinical encounters. This data extraction approach has been used consistently in multiple previous peer-reviewed investigations from our institution. ,,
All angiographic images were independently reviewed by experienced CTO operators (S.T. and L.A.) to verify lesion characteristics, procedural strategy, and outcomes. Angiographic assessment included confirmation of the CTO target vessel, lesion complexity, and crossing strategy, as well as identification of procedural complications such as coronary perforation. Baseline demographic and clinical characteristics, angiographic and procedural variables, and in-hospital outcomes were collected for the full cohort.
The study was conducted in accordance with the principles of the Declaration of Helsinki and good clinical practice. Given the retrospective nature of the analysis, informed consent was waived. The study protocol was approved by the University of Washington Institutional Review Board (STUDY00016614).
Outcomes and definitions
The primary study end points were 30-day major adverse cardiac and cerebrovascular events (MACCE) or 30-day all-cause rehospitalization. MACCE included all-cause death, myocardial infarction (MI), stroke, and target vessel revascularization. Secondary outcomes included individual adverse events—30-day all-cause mortality, MI, stroke, and target vessel revascularization.
Death and MI were defined according to the Academic Research Consortium (ARC)-2 criteria. Periprocedural MI was adjudicated per ARC-2 definitions using biomarker and electrocardiographic criteria.
Technical success was defined as successful CTO recanalization with <30% residual stenosis in the treated segment and restoration of thrombolysis in MI (TIMI) 2 to 3 antegrade flow. Procedural success was defined as technical success without in-hospital MACCE.
Definitions for angiographic and procedural characteristics, including proximal cap ambiguity, lesion calcification, tortuosity, and crossing strategies (antegrade wiring, antegrade dissection and reentry [ADR], and retrograde), followed the Chronic Total Occlusion Academic Research Consortium recommendations. The Japanese CTO (J-CTO) score was calculated.
Procedural variable definitions were based on the NCDR CathPCI Registry v5.0 dictionary. Acute kidney injury was defined and categorized based on the Acute Kidney Injury Network (AKIN) criteria. Bleeding was classified according to Bleeding Academic Research Consortium (BARC) criteria.
The decision regarding SDD, admission for observation, and the use of adjunctive therapies (including plaque-modification techniques and intravascular imaging) was left to the operator’s discretion.
Statistical analysis
For comparative analyses, patients were categorized as SDD or non-SDD following CTO PCI. Continuous variables were reported as mean ± standard deviation and compared using the independent-samples t test or the Mann–Whitney U test, as appropriate. Categorical variables were expressed as absolute counts and percentages and compared using either the chi-square test or Fisher’s exact test, as appropriate.
Given that the primary outcome (30-day MACCE or rehospitalization) is time-dependent, time-to-event analyses were performed using Cox proportional hazards regression. Univariable Cox regression analyses were first conducted to evaluate the association among baseline clinical, angiographic, and procedural variables and the composite 30-day outcome, with results expressed as hazard ratios (HRs) and 95% confidence intervals (CIs).
Variables entered into the Cox proportional hazards model were selected based on clinical relevance and plausibility. This approach was chosen to minimize data-driven overfitting and to ensure inclusion of clinically meaningful confounders. Candidate variables included demographic characteristics, comorbid conditions, measures of cardiac and renal function, anatomic and procedural complexity metrics, and procedural strategies.
To assess for multicollinearity among covariates, variance inflation factors (VIFs) were calculated for all variables included in the final multivariable model. A VIF threshold of >5 was prespecified to indicate significant collinearity. No variables exceeded this threshold, indicating the absence of clinically meaningful multicollinearity and supporting the stability of the final model estimates.
Candidate correlates were SDD group, age, sex, body mass index (BMI), Charlson Comorbidity Index (CCI), diabetes mellitus, hypertension, hyperlipidemia, distance to PCI center (driving miles), estimated glomerular filtration rate (eGFR), previous MI, previous PCI, previous coronary artery bypass graft, previous stroke, peripheral arterial disease, tobacco use, LVEF, use of ADR crossing strategy, use of retrograde crossing strategy, intravascular imaging use, atherectomy/intravascular lithotripsy use, femoral access site, and J-CTO score. The driving distance in miles was derived by geocoding addresses to latitude/longitude pairs and computing distance from UWMC using the R geosphere package.
To investigate the patterns of 30-day MACCE and rehospitalization risk in the SDD patient population, a recursive partitioning tree model was fitted using the R rpart package, with the sum of squared errors criterion to select candidate binary splits based on baseline patient clinical and demographic characteristics at each node in the tree. ,
A p value ≤0.05 was considered statistically significant. Analyses were conducted using R version 4.4.2 (R Core Team, Vienna, Austria) and MedCalc Statistical Software version 12.7.7 (MedCalc Software, Ostend, Belgium).
Results
Study cohort
Among 1,273 patients who underwent CTO PCI during the study period, 352 (27.6%) patients were discharged the same day, whereas 921 (72.4%) patients were non-SDD. Over the study period, the frequency of SDD increased significantly from 2019 to 2023 (11.7%–44.3%, p < 0.001) ( Figure 1 ).
Same-day discharge after CTO PCI incidence by year. Annual frequency of same-day discharge following CTO PCI from 2019 to 2023. CTO = chronic total occlusion; PCI = percutaneous coronary intervention.
Clinical characteristics
As shown in Table 1 , SDD patients were younger (65.0 ± 10.8 vs 68.6 ± 9.1 years, p < 0.001) and had a lower co-morbidity burden (CCI 3.6 ± 3.1 vs 4.2 ± 2.8, p = 0.002) than non-SDD patients. SDD patients also lived closer to the PCI center (185.5 ± 389.0 vs 260.0 ± 484.1 miles, p = 0.021). Cardiovascular co-morbidities were less common among SDD patients, and SDD patients also demonstrated better cardiac and renal function.
Table 1
Clinical characteristics
|
Total Cohort
( N = 1273) |
SDD
( N = 352) |
Non-SDD
( N = 921) |
p-value | |
|---|---|---|---|---|
| Age, years | 67.6 ± 9.7 | 65.0 ± 10.8 | 68.6 ± 9.1 | < 0.001 |
| Body mass index, kg/m 2 | 30.4 ± 5.9 | 30.5 ± 5.8 | 30.4 ± 5.9 | 0.790 |
| Female | 199 (15.6) | 53 (15.1) | 146 (15.9) | 0.760 |
| Race | ||||
| Caucasian | 1043 (82.0) | 288 (81.8) | 755 (82.0) | 0.932 |
| African American | 40 (3.1) | 11 (3.1) | 29 (3.1) | 0.989 |
| Asian | 93 (7.3) | 31 (8.8) | 62 (6.7) | 0.207 |
| Other | 97 (7.6) | 22 (6.3) | 75 (8.1) | 0.265 |
| Hispanic Ethnicity | 37 (2.9) | 8 (2.3) | 29 (3.1) | 0.480 |
| Distance to PCI Center, driving miles | 206.4 ± 418.7 | 185.5 ± 389.0 | 260.0 ± 484.1 | 0.021 |
| Charlson Comorbidity Index | 4.0 ± 2.9 | 3.6 ± 3.1 | 4.2 ± 2.8 | 0.002 |
| Tobacco Use | 91 (7.1) | 25 (7.1) | 66 (7.2) | > 0.999 |
| Hypertension | 1135 (89.2) | 301 (85.5) | 834 (90.6) | 0.009 |
| Hyperlipidemia | 1247 (98.0) | 346 (98.3) | 901 (97.8) | 0.589 |
| Diabetes Mellitus | 542 (42.6) | 138 (39.2) | 404 (43.9) | 0.128 |
| Chronic Kidney Disease | 310 (24.4) | 65 (18.5) | 245 (26.6) | 0.002 |
| Previous myocardial infarction | 578 (45.4) | 144 (40.9) | 434 (47.1) | 0.043 |
| Previous CABG | 544 (42.7) | 118 (33.5) | 426 (46.3) | < 0.001 |
| Previous PCI | 953 (74.9) | 251 (71.3) | 702 (76.2) | 0.080 |
| Previous stroke | 42 (3.3) | 13 (3.7) | 29 (3.1) | 0.610 |
| Peripheral Arterial Disease | 190 (14.9) | 33 (9.4) | 157 (17.0) | < 0.001 |
| Atrial Fibrillation | 147 (11.5) | 36 (10.2) | 111 (12.1) | 0.318 |
| LVEF, % | 48.8 ± 13.3 | 50.9 ± 12.6 | 47.9 ± 13.5 | < 0.001 |
| eGFR, mL/min/1.73 m 2 | 75.1 ± 23.7 | 78.6 ± 24.2 | 73.7 ± 23.3 | 0.001 |
Abbreviations : CABG = coronary artery bypass grafting; eGFR = estimated glomerular filtration rate; LVEF = left ventricular ejection fraction; PCI = percutaneous coronary intervention; SDD = same-day discharge.
Angiographic features and procedural data
As described in Table 2 , the distribution of CTO target vessels was similar between groups, with no statistically significant differences. Angiographic complexity was similar between SDD and non-SDD patients (J-CTO score 2.0 ± 1.1 vs 2.0 ± 1.3, p = 0.960).
Table 2
Angiographic and lesion features
|
Total Cohort
( N = 1273) |
SDD
( N = 352) |
Non-SDD
( N = 921) |
p-value | |
|---|---|---|---|---|
| CTO Target Vessel | ||||
| Left Main | 18 (1.4) | 7 (2.0) | 11 (1.2) | 0.286 |
| Left Anterior Descending Artery | 268 (21.2) | 84 (23.9) | 184 (20.0) | 0.123 |
| Circumflex Artery | 305 (24.0) | 88 (25.0) | 217 (23.6) | 0.616 |
| Right Coronary Artery | 669 (52.5) | 170 (48.3) | 499 (54.2) | 0.054 |
| Ramus | 13 (1.0) | 3 (0.9) | 10 (1.1) | 0.999 |
| Proximal Cap Ambiguity | 749 (58.8) | 211 (59.9) | 538 (58.4) | 0.629 |
| Moderate to Severe Calcification | 525 (41.3) | 144 (40.9) | 381 (41.4) | 0.867 |
| Bending | 177 (13.9) | 54 (15.3) | 123 (13.4) | 0.357 |
| Lesion Length > 20 mm | 835 (65.6) | 227 (64.5) | 608 (66.0) | 0.627 |
| Reattempt CTO PCI | 196 (15.4) | 63 (17.9) | 133 (14.4) | 0.148 |
| J-CTO score | 2.0 ± 1.2 | 2.0 ± 1.1 | 2.0 ± 1.3 | 0.960 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
