Author + information
- Received July 27, 2016
- Revision received February 9, 2017
- Accepted February 12, 2017
- Published online May 1, 2017.
- Thomas T. Tsai, MD, MSca,b,c,∗ (, )
- Maggie A. Stanislawski, MSa,b,
- Kendrick A. Shunk, MD, PhDd,e,
- Ehrin J. Armstrong, MD, MSca,b,
- Gary K. Grunwald, PhDa,b,
- Alan H. Schob, MDf,g,
- Javier A. Valle, MDa,b,
- Carlos E. Alfonso, MDf,g,
- Brahmajee K. Nallamothu, MD, MPHh,i,
- P. Michael Ho, MD, PhDa,b,
- John S. Rumsfeld, MD, PhDa,b and
- Emmanouil S. Brilakis, MD, PhDj,k
- aDenver VA Medical Center, Denver, Colorado
- bUniversity of Colorado Denver, Denver, Colorado
- cInstitute for Health Research, Kaiser Permanente Colorado, Denver, Colorado
- dSan Francisco VA Medical Center, San Francisco, California
- eUniversity of California, San Francisco, School of Medicine, San Francisco, California
- fMiami Veterans Affairs Healthcare System, Miami, Florida
- gUniversity of Miami Miller School of Medicine, Miami, Florida
- hAnn Arbor VA Medical Center, Ann Arbor, Michigan
- iUniversity of Michigan Medical Center, Ann Arbor, Michigan
- jVA North Texas Health Care System, Dallas, Texas
- kUniversity of Texas Southwestern Medical School, Dallas, Texas
- ↵∗Address for correspondence:
Dr. Thomas T. Tsai, Denver VA Medical Center, Cardiology Section 111B, 1055 Clermont Street, Denver, Colorado 80220.
Objectives The aim of this study was to describe the contemporary incidence of chronic total occlusions (CTOs) and the success rates of CTO percutaneous coronary intervention (PCI), as well as the complications and long-term outcomes of these patients.
Background The contemporary prevalence and management of coronary CTOs is understudied.
Methods Consecutive veterans undergoing coronary angiography at 79 Veterans Affairs sites between 2007 and 2013 were examined. Detailed baseline clinical, angiographic, and follow-up outcomes were evaluated using national data from the Veterans Affairs Clinical Assessment Reporting and Tracking program.
Results Among 111,273 patients with obstructive coronary artery disease, 29,399 (26.4%) had ≥1 CTO, most commonly in the right coronary artery distribution (n = 18,986 [64.6%]). Elective CTO PCI was attempted in 2,394 patients (8.1%), with a procedural success rate of 79.7%. The odds of CTO PCI success increased over the years of the study (odds ratio: 1.08; 95% confidence interval [CI]: 1.01 to 1.16; p = 0.03). Compared with failed CTO PCI, successful CTO PCI was associated with a decreased adjusted risk for mortality (hazard ratio: 0.67; 95% CI: 0.47 to 0.95; p = 0.02) and coronary artery bypass graft surgery (hazard ratio: 0.14; 95% CI: 0.08 to 0.24; p < 0.01) at 2 years but no significant change in the risk for hospitalization for myocardial infarction (hazard ratio: 0.89; 95% CI: 0.58 to 1.36; p = 0.58).
Conclusions Approximately 1 in 4 patients with obstructive coronary artery disease on coronary angiography had CTOs. Among patients who went on to elective CTO PCI, the success rate was 79.7%. Compared with failed CTO PCI, successful CTO PCI was associated with a decreased risk for mortality as well as a decreased need for subsequent coronary artery bypass graft surgery.
The contemporary prevalence of coronary artery chronic total occlusions (CTOs) in patients undergoing coronary angiography has been poorly defined and is heavily dependent on the population studied. Older single-center studies showed that approximately 15% to 30% of patients referred for coronary angiography had CTOs, whereas a more contemporary multicenter report from the National Cardiovascular Data Registry (NCDR) found that approximately 5.5% of all diagnostic angiographic studies performed between January 2004 and March 2005 revealed CTOs (1–4).
The presence of a CTO has been associated with increased mortality, and patients are commonly referred for coronary artery bypass graft (CABG) surgery. Historically, the frequency of CTO percutaneous coronary intervention (PCI) has been very low, accounting for 3.8% of all PCIs performed in patients with stable coronary artery disease (CAD) (5), and when compared with non-CTO PCI has been associated with lower procedural success rates, high technical complexity, high equipment use and cost, and the potential for major periprocedural complications (6). Thus, patients with CTOs are more likely to be referred for CABG surgery and less likely to undergo PCI (2,3,7). However, recent advances in CTO PCI techniques, including the retrograde approach, antegrade dissection and re-entry, and a systematic algorithmic approach, have consistently improved success rates, with low complication rates (8,9).
Successful CTO PCI has been associated with symptom relief (10,11), improvement in left ventricular function (12–14), and improved long-term survival (4,15–17) compared with failed CTO PCI in nonrandomized observational studies. However, skepticism still exists regarding the purported benefits of CTO PCI. Thus, we present the largest cohort of coronary angiography patients with follow-up data and describe the contemporary incidence of CTOs and the success rates of CTO PCI, as well as the complications and long-term outcomes of these patients.
Data for this analysis were obtained from the U.S. Department of Veterans Affairs (VA) Clinical Assessment Reporting and Tracking (CART) program, which is a national clinical quality initiative for all VA cardiac catheterization laboratories launched in 2005. The program has 3 missions: to support clinical care, to promote the quality of care, and to advance knowledge through research. The program’s initial focus is on coronary procedures conducted in all VA cardiac catheterization laboratories nationwide (18).
A key feature of the CART program is a clinical software application designed to collect standardized data on all coronary angiographic studies and PCIs. The software is embedded in the VA electronic health record and allows providers to enter patient and procedural information (pre-procedure assessment, coronary angiography, and PCI) as part of routine clinical work flow. The CART software was designed using standardized definitions that conform to the definitions and standards of the American College of Cardiology’s NCDR PCI registry and incorporates features such as pull-down menus and automated clinical report generation to ensure uniformity of data entry by different providers and in different cardiac catheterization laboratories (19). CART is mandatory and universal in all VA cardiac catheterization laboratories.
Quality checks of the data are periodically conducted for completeness and accuracy (20). Finally, CART data are combined with other VA data sources to create a longitudinal data repository that supports the quality assessment, quality improvement, and clinical research missions of the CART program.
Study design, setting, and population
For this study, we identified all patients undergoing coronary angiography between October 2007 and September 2013 in any of the 79 VA cardiac catheterization laboratories. Obstructive CAD was defined as ≥50% stenosis in the left main coronary artery or ≥70% stenosis in any other coronary artery. Patients without lesion-level stenosis information or without obstructive CAD were excluded. CTO was indicated by the physician in the CART application as defined by a 100% stenosis presumed to be occluded for >3 months and not related to an acute clinical event prompting the procedure. If the physician marked CTO, and the pre-stenosis was <95%, it was assumed that the CTO status was a data entry error and excluded from the analysis. We studied 111,273 patients with obstructive CAD (Figure 1); 26.4% of these patients (n = 29,399) had at least 1 CTO.
The primary exposure variable of interest was successful elective CTO PCI. Procedures involving ST-segment elevation acute myocardial infarction (MI), non–ST-segment elevation acute MI, or cardiogenic shock were considered nonelective and excluded. CTO lesion success was defined as luminal diameter stenosis to ≤50% in the CTO segment. Any patients undergoing elective PCI who did not have lesion-level information about their PCI in the CART data were excluded. CTO PCI success was defined as successful treatment of all attempted CTOs and no major adverse events (death, CABG surgery, or MI) during the procedure. The study cohort included 2,394 CTO PCIs on 2,516 CTO lesions.
Outcomes assessed include both short-term outcomes (procedure-related in-laboratory complications) and 1- and 2-year outcomes (mortality, MI, and subsequent CABG surgery).
Procedure-related in-laboratory complications
We assessed the incidence of death, periprocedural MI, emergent CABG surgery, stroke, vascular access complications, dysrhythmia, acute pulmonary edema, and anaphylactic shock. This information was directly entered as discrete data elements into CART by the treating physician (20).
The VA vital status file was used to assess the mortality outcome. This file has 98.3% sensitivity and 97.6% exact agreement with dates compared with the National Death Index (21).
Rehospitalization for MI
MI was assessed via International Classification of Diseases-Ninth Revision codes 410.xx in VA inpatient and VA fee-based data files (22). A random sample of patients with MI also underwent individual chart review to validate this outcome in accordance with the third universal definition of MI (23). Subsequently, codes for MI hospitalizations during the first 14 days after the PCI discharge date were disregarded because a review of cases showed that the majority of these codes were related to the index hospitalization.
CABG procedures were assessed on the basis of International Classification of Diseases-Ninth Revision procedure codes and Current Procedural Terminology codes in the VA administrative and fee-basis files.
Comparison of baseline characteristics and in-laboratory complications between patients with obstructive CAD with and without CTOs as well as unsuccessful and successful CTO PCIs were performed using Pearson chi-square or Fisher exact tests for categorical variables and Wilcoxon rank sum tests for continuous variables. Generalized estimating equations with binary logistic regression were used to estimate the predictors of CTO PCI success, allowing correlation by hospital and using an exchangeable correlation structure. The following patient and CTO characteristics were included a priori in the model: age, sex, race, ethnicity, diabetes, hypertension, hyperlipidemia, history of smoking, cerebrovascular disease, peripheral artery disease, chronic kidney disease, chronic obstructive pulmonary disease, congestive heart failure, prior PCI, prior MI, prior CABG surgery, vessel calcification, vessel, and year of the procedure (to control for changes in practice over time). We used the Kaplan-Meier method to determine event-free survival and compared the 2 groups using the log-rank test, survival curves, and estimated survival rates at 1 and 2 years. We also fit Cox proportional hazards models with robust sandwich covariance matrix estimates to account for intrahospital dependence. We included successful CTO status in the survival model and controlled for age, sex, race, ethnicity, diabetes, hypertension, hyperlipidemia, history of smoking, cerebrovascular disease, peripheral artery disease, chronic kidney disease, dialysis, chronic obstructive pulmonary disease, congenital heart failure, prior PCI, prior MI, prior CABG surgery, and procedure year.
Of the 111,273 patients found to have obstructive CAD on coronary angiography, 29,399 (26.4%) had at least 1 CTO. The median age was 64.7 years (interquartile range: 60.6 to 71.7 years) (Table 1). Patients with CTOs were older and had higher prevalence rates of hypertension, hyperlipidemia, diabetes, smoking, obesity, congestive heart failure, chronic obstructive pulmonary disease, cerebrovascular disease, peripheral artery disease, chronic depression, sleep apnea, chronic kidney disease, dialysis, prior MI, prior PCI, and prior CABG surgery.
Elective CTO PCI was performed in 2,394 patients. The majority of CTO PCI procedures (79.7%) were successful. On univariate analysis, patients with successful CTO PCI procedure were less likely to be on dialysis, to have CTOs in the right coronary artery, and to have calcified lesions. Other comorbidities and procedural details were not significantly different (Table 2).
Angiographic and procedure characteristics
Table 3 summarizes angiographic and procedural characteristics. Treated CTOs were most commonly in the right coronary artery distribution (42.4%), followed by the left anterior descending coronary artery (23.5%) and then the left circumflex coronary artery (15.9%). The average lesion length was 20.0 mm (interquartile range: 15.0 to 32.0 mm). Drug-eluting stents were used in 80.8% of successful CTOs, and 37.8% of successful procedures required 2 or more stents. Successfully treated CTOs were more often in the left anterior descending coronary artery and less often in the right coronary artery, and they tended to be shorter in length.
The overall procedural CTO PCI success rate was 79.7% (1,908 of 2,394 cases), whereas the technical success rate was slightly higher at 79.8% (1,911 of 2,394 cases). More periprocedural complications occurred in the patients with PCIs involving unsuccessfully treated lesions (7.5% vs. 3.5%), including dissections and perforations (Table 4). Specific complications such as hematomas, cardiac tamponade, cardiogenic shock, emergent CABG surgery, periprocedural MI, stroke, and death were not significantly different between successful and unsuccessful procedures.
Success of elective CTO PCI
In multivariate analysis, most patient comorbidities did not significantly influence the odds of CTO PCI success (Table 5). The odds of success increased over the years of the study (odds ratio: 1.08; 95% confidence interval [CI]: 1.01 to 1.16; p = 0.03), decreased with increasing numbers of CTOs treated (odds ratio: 0.55; 95% CI: 0.38 to 0.81; p < 0.01), and were lower among black patients relative to white patients (odds ratio: 0.68; 95% CI: 0.50 to 0.92; p = 0.01).
Cumulative 2-year survival curves demonstrate slightly higher unadjusted survival among patients with successful versus failed elective CTO PCIs (p = 0.05) (Figure 2). After adjustment for differences in the baseline characteristics, successful CTO PCI remained an independent predictor of 2-year survival (HR: 0.67; 95% CI: 0.47 to 0.95; p = 0.02) (Table 6).
Cumulative 2-year freedom from MI did not differ by CTO PCI success (p = 0.83) (Figure 3). After adjustment, the difference remained nonsignificant (hazard ratio: 0.89; 95% CI: 0.58 to 1.36; p = 0.58) (Table 6). Patients with failed CTO PCI had higher rates of CABG at 2 years compared with patients with successful procedures (p < 0.001) (Figure 4). After adjustment, successful CTO PCI remained associated with lower risk for CABG surgery (hazard ratio: 0.14; 95% CI: 0.08 to 0.24; p < 0.01) (Table 6).
In this large contemporary, real-world study evaluating patients undergoing coronary angiography, 26.4% of patients with obstructive CAD, had at least 1 CTO. Among elective CTO PCI patients, the procedural success rate was 79.7%. Compared with patients with failed CTO PCI, successful CTO PCI was associated with improved survival at 2 years and a decreased likelihood for subsequent CABG procedures. This study represents a broad evaluation of CTO PCI at 79 VA hospitals across the country between 2007 and 2013, providing the treatment and outcomes on the largest contemporary cohort of CTO PCI patients to date.
The prevalence of CTOs in patients with CAD on coronary angiography compares similarly with a smaller single-center VA cohort of patients published by Jeroudi et al. (24) showing a 31% prevalence of CTOs in patients without prior CABG surgery undergoing coronary angiography with obstructive CAD, as well as larger registries from Canada, Germany, and the United States with CTO prevalence rates of 18.4%, 33.0%, and 35.0%, respectively (1,2,25). Also, consistent with the other studies, patients with obstructive CAD and CTOs are more likely to have a heavier burden of comorbidities such as hypertension, hyperlipidemia, diabetes, renal disease, peripheral artery disease, and cerebrovascular disease (5,24).
Despite the high prevalence of CTOs on coronary angiography, CTO PCI rates have historically been very low, constituting only 3.8% of the total PCIs in a recent publication from the NCDR Cath-PCI registry (5). This is due in part to the increased procedural complexity, higher complication rates, and questionable long-term benefits of CTO PCI. Work by Azzalini et al. (26) suggested that CTOs were treated by PCI or CABG surgery less than one-half of the time and that CTO PCI occurred in younger patients with lower SYNTAX (Synergy Between PCI With Taxus and Cardiac Surgery) scores and patients with higher contribution of the overall SYNTAX score from the CTO lesion. However, in recent years, important advances have occurred in CTO PCI techniques and technology, including the retrograde approach, antegrade dissection and re-entry techniques, and a systematic algorithmic approach achieving contemporary success rates of 80% to 90% at high-volume centers (5,27–31).
Most CTO-PCIs in our study were performed before the development and implementation of the systematic algorithmic approach to CTO PCI, with success rates approaching 80%. This compares favorably with other studies, with success rates of 58.5% in a large NCDR Cath-PCI population, 85.5% in a pooled analysis of 3 high-volume CTO centers, and 74.4% at a single center over 20 years (5,15,32). The odds of CTO-PCI success in our study increased over the years of the study, potentially indicating an increased uptake of the algorithmic approach, experience with new CTO devices, or more dedication to CTO PCI. However, increasing volume of CTO PCI procedures by site was associated with a decreased likelihood of CTO PCI success, likely related to a focus on CTO PCIs by the sites and specializing in more complex cases.
In a multivariate analysis, successful versus failed CTO PCI was associated with improved survival and decreased CABG over 2 years of follow-up. This theme has been consistent with other single-center observational registries with long-term follow-up. In a single-center registry from the Mid-American Heart Institute, Suero et al. (15) showed a distinct 10-year survival advantage for successful CTO PCI treatment compared with failed CTO PCI treatment (73.5% vs. 65.1%; p = 0.001). Additionally, 4 other single-center studies have also shown an absolute reduction of mortality between 3.8% and 8.4% (4,10,16,17,33). George et al. (34) reported data on 13,443 patients who underwent 14,439 CTO PCI procedures in the United Kingdom between 2005 and 2009. Procedural success was 70.6%. Successful PCI of at least 1 CTO was associated with lower mortality (hazard ratio: 0.72; 95% CI: 0.62 to 0.83; p < 0.001).
Our study shows similar benefit for successful CTO PCI but also provides findings on subsequent MI and CABG surgery that were not examined in the British series. Although there were no significant differences in MI in our study, we were unable to assess for other cause of cardiac mortality, such as arrhythmias and congestive heart failure, which may be more significant in patients with CTO PCI failure. Furthermore, failure of CTO PCI may be a marker of unmeasured patient burden of disease and worse outcomes. These challenges in retrospective observational studies strengthen the need for randomized control trials to fully understand the potential benefits and risks of successful CTO PCI, unsuccessful CTO PCI, medical therapy, and CABG surgery.
There are several important considerations when interpreting the results of this study. First, patients and hospitals in the VA system may not be representative of all U.S. practice, and nearly all patients are men. However, the CART program represents more than 79 VA hospitals across the United States with physicians who also practice at the affiliated university-based academic medical centers. Therefore, the technical expertise in PCI is likely to be representative of interventional care across the United States.
Second, the CART program does not capture CTO-specific data regarding crossing strategies (antegrade dissection re-entry, retrograde, and so on) or the angiographic variables to calculate a SYNTAX score to ascertain angiographic complexity. However, CTO PCI success is the exposure variable of interest and allows the assessment of this variable with outcomes regardless of the method of achieving successful revascularization.
Third, as in all observational studies, unmeasured confounders that could influence the likelihood of CTO PCI success or failure must be considered (e.g., unmeasured comorbid illness affecting the intensity of CTO PCI efforts). However, all available measured variables were used in the multivariate regression.
Fourth, we were unable to compare outcomes of the CABG and PCI patients because of different covariate profiles and confounding by indication bias.
Fifth, we did not have follow-up information on ejection fraction to assess any potential differences in patients with successful CTO PCI versus unsuccessful CTO PCI.
In a large national cohort study, we found that approximately 1 in 4 patients with obstructive disease on coronary angiography had a least 1 CTO. Elective CTO PCI had a high rate of successful revascularization. Compared with failed CTO PCI, successful CTO PCI was associated with increased survival and decreased need for CABG surgery. These results are consistent with prior studies suggesting improved survival with successful CTO PCI procedures. However, these findings need to be validated in prospective randomized control trials.
WHAT IS KNOWN? Older single-center studies over a long follow-up period suggest low CTO PCI rates with variable success rates. Furthermore, there is skepticism regarding the benefits of successful CTO PCI, such as mortality.
WHAT IS NEW? This study is the largest contemporary cohort of CTO PCI patients representing the contemporary management and outcomes of successful CTO PCI. Our findings suggest that 1 in 4 patients with CAD have CTOs. Of those, approximately 8% undergo CTO PCI, with a success rate approaching 80%. Compared with failed CTO PCI, successful CTO PCI is associated with a decreased risk for mortality as well as a decreased need for subsequent CABG.
WHAT IS NEXT? Randomized controlled trials are needed to fully understand the potential benefits and risks of successful CTO PCI, unsuccessful CTO PCI, medical therapy, and CABG surgery.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- coronary artery bypass graft
- coronary artery disease
- Clinical Assessment Reporting and Tracking
- chronic total occlusion
- myocardial infarction
- National Cardiovascular Data Registry
- percutaneous coronary intervention
- U.S. Department of Veterans Affairs
- Received July 27, 2016.
- Revision received February 9, 2017.
- Accepted February 12, 2017.
- Fefer P.,
- Knudtson M.L.,
- Cheema A.N.,
- et al.
- Grantham J.A.,
- Marso S.P.,
- Spertus J.,
- House J.,
- Holmes D.R. Jr..,
- Rutherford B.D.
- Hoye A.,
- van Domburg R.T.,
- Sonnenschein K.,
- Serruys P.W.
- Brilakis E.S.,
- Banerjee S.,
- Karmpaliotis D.,
- et al.
- Shah P.B.
- Patel V.G.,
- Brayton K.M.,
- Tamayo A.,
- et al.
- Olivari Z.,
- Rubartelli P.,
- Piscione F.,
- et al.
- Cheng A.S.,
- Selvanayagam J.B.,
- Jerosch-Herold M.,
- et al.
- Melchior J.P.,
- Doriot P.A.,
- Chatelain P.,
- et al.
- Sirnes P.A.,
- Myreng Y.,
- Molstad P.,
- Bonarjee V.,
- Golf S.
- Suero J.A.,
- Marso S.P.,
- Jones P.G.,
- et al.
- Valenti R.,
- Migliorini A.,
- Signorini U.,
- et al.
- Brindis R.G.,
- Fitzgerald S.,
- Anderson H.V.,
- Shaw R.E.,
- Weintraub W.S.,
- Williams J.F.
- Byrd J.B.,
- Vigen R.,
- Plomondon M.E.,
- et al.
- Sohn M.W.,
- Arnold N.,
- Maynard C.,
- Hynes D.M.
- Thygesen K.,
- Alpert J.S.,
- Jaffe A.S.,
- et al.
- Azzalini L.,
- Jolicoeur E.M.,
- Pighi M.,
- et al.
- Brilakis E.S.,
- Grantham J.A.,
- Rinfret S.,
- et al.
- Galassi A.R.,
- Sianos G.,
- Werner G.S.,
- et al.
- Karmpaliotis D.,
- Michael T.T.,
- Brilakis E.S.,
- et al.
- Teramoto T.,
- Tsuchikane E.,
- Matsuo H.,
- et al.
- George S.,
- Cockburn J.,
- Clayton T.C.,
- et al.