Health Status Benefits of Successful Chronic Total Occlusion Revascularization Across the Spectrum of Left Ventricular Function: Insights from the OPEN-CTO Registry

Abstract

Objective:

Describe the association between chronic total occlusion revascularization (CTO-PCI) and health status in patients with and without cardiomyopathy.

Background:

Prior PCI trials for cardiomyopathy have excluded CTO patients. Whether patients with reduced ejection fraction (LVEF) receive similar health status benefit from CTO-PCI compared to patients with normal LVEF is unclear.

Methods:

We assessed health status change, using the Seattle Angina Questionnaire Summary (SAQ-SS), SAQ Angina Frequency (SAQ-AF), and Rose Dyspnea Scale (RDS) scores, among patients undergoing successful CTO PCI in the OPEN-CTO Registry. Participants were classified by LVEF (normal (≥50%), mild-moderate (30–49%), and severe (<30%)), with higher SAQ and lower RDS scores indicating better health status. Differences in one-year outcomes were compared using hierarchical multivariable regression.

Results:

Of 762 patients, 506 (66.4%), 193 (25.3%), and 63 (8.3%) had normal, mild-moderate, and severely reduced LVEF. SAQ-SS improvements were observed in each group (27.1 +/− 20.4, 26.7 +/− 21.2, and 20.3 +/− 18.1, respectively). Compared to patients with LVEF ≥50%, those with LVEF < 30% had less improvement in SAQ-SS (−5.2 pts; 95% CI −9.0, −1.5; p = 0.01) and RDS (+0.5 pts; 95% CI 0.1, 0.8; p = 0.01), with no difference in odds of angina (OR 1.3; 95% CI 0.6, 3.0; p=0.48). Health status improvement was similar between patients with LVEF ≥50% and LVEF 30–49%.

Conclusions:

Although health status improvement was less in patients with severely reduced LVEF compared to those with normal LVEF, each group experienced large health status improvements after CTO-PCI.

CONDENSED ABSTRACT:

We examined the association between CTO-PCI and health status in patients with normal (≥50%), mild-moderate (30–49%), and severely reduced (<30%) LVEF in the OPEN-CTO registry. Health status change at one year was assessed using Seattle Angina Questionnaire Summary (SAQ-SS), SAQ Angina Frequency (SAQ-AF), and Rose Dyspnea Scale (RDS) scores, where higher SAQ and lower RDS scores indicated better health status. We found that patients in each LVEF group experienced large improvements in health status after successful CTO-PCI. These findings are significant for symptomatic patients referred for CTO-PCI, independent of LVEF and survival benefit.

INTRODUCTION

Chronic total occlusions (CTO) of the coronary arteries are detected in roughly one in five patients undergoing coronary angiography (1) and are frequent among those with ischemic cardiomyopathy (2). Although studies have suggested that patients who undergo successful recanalization of a CTO experience improved left ventricular function and survival when compared to those who have unsuccessful procedures (3,4), the impact of successful CTO recanalization on angina burden and quality of life remains unclear.

Chronic total occlusion revascularization has been assigned a Class IIa recommendation for reduction of angina relief in patients with medically refractory angina (5). However, patients with CTOs have traditionally been excluded from or under-represented in trials of percutaneous revascularization in the setting of left ventricular dysfunction (6)^,(7), and the impact of CTO-PCI on patients’ symptoms, function, and quality of life is not well defined. To address this gap in knowledge, we used data from the Outcomes, Patient Health Status, and Efficiency in Chronic Total Occlusion Hybrid Procedures (OPEN-CTO) registry to examine whether patients who underwent CTO-PCI experienced health status benefits independent of ejection fraction (LVEF).

METHODS

Study Population and Data Collection

Details of the OPEN-CTO registry and its design have been published previously (8). Briefly, 1,000 patients were consecutively enrolled at 12 experienced CTO-PCI centers across the United States. All operators were required to have had at least two years’ experience in CTO revascularization. At each center, all patients scheduled for CTO-PCI by a designated hybrid CTO operator were consecutively screened. Of those screened, all patients ≥18 years of age that consented to participate with serial telephone interviews and had at least one CTO, as defined by Thrombolysis in Myocardial Infarction Score (TIMI) (9) antegrade flow of zero with a lesion that had been present for at 3 months, were enrolled. Non-English-speaking patients or those otherwise unable to participate in follow-up health assessments were excluded. For purposes of this analysis, patients with undocumented LVEF or missing baseline or 12-month health status information were excluded from the analytic cohort. Finally, only participants with successful CTO-PCI were analyzed, to better understand the relative merits of revascularization in this understudied scenario. After patients who did not have a successful procedure (n = 137), were missing LVEF (n = 77), or were missing health status data on enrollment or at 12 months (n = 24) were excluded, the final cohort was comprised of 762 patients (Figure 1).

Figure 1.

Patient Exclusion Flowsheet.

Legend: Patient exclusion flowsheet for the primary analysis, where patients with technical failure, missing baseline left ventricular function, and missing 12-month SAQ scores were excluded

Quality of life (QoL) assessment was performed at 1, 6 and 12-month follow-up using disease-specific instruments (8). Baseline patient QoL information was collected on enrollment and subsequent assessments were acquired through telephone interviews performed by trained study personnel. Clinical information (sociodemographic and medical history) as well as the most recent LVEF assessment was abstracted from patients’ medical records using standardized case record forms at the time of enrollment. OPEN-CTO received local IRB approval from all participating centers, and informed consent was obtained from each patient.

Primary Health Status Outcomes

We used the Seattle Angina Questionnaire (SAQ), a 19-item disease-specific health status measure, to assess coronary disease specific health status (10). The SAQ includes 5 domains: angina frequency, angina stability, physical limitation, quality of life, and treatment satisfaction and the angina frequency, physical limitation and quality of life domains can be combined into an overall summary score (11). For this study, we focused upon the Summary Score (SAQ-SS) and the Angina Frequency score (SAQ-AF), which were felt to provide the most complete capture of angina-specific health status. The SAQ-SS ranges from 0–100, with higher scores indicating less angina, better physical function, and higher quality of life. The Angina Frequency score can also be categorized into daily (0–30), weekly (31–60), monthly (61–99) and no angina (100) symptoms (12). To explicitly measure dyspnea, we used the Rose Dyspnea Scale (RDS) that captures the level of activity with which patients report shortness of breath and ranges from 0–4 (where lower RDS scores indicate less dyspnea). It has also been shown to be sensitive to change in patients with ischemic heart disease (13).

Definitions

Technical procedural success was defined as the positioning of a guidewire in the distal true lumen of the CTO, followed by angioplasty with residual TIMI II or III flow, residual stenosis of 50% or less, and no significant occlusion of visualized side branches. Procedural adjudication was performed at a designated core laboratory.

Prior to angiography, assessment of LVEF was performed using one of several modalities, including echocardiography (41.6%), nuclear imaging (32.8%), angiography (23.0%), and magnetic resonance imaging (1.5%). Patients with documented LVEF were then categorized, using their most recent LVEF, into groups consistent with prior publications (14) as normal (≥ 50%), mild-moderate (30–49%), and severely reduced (<30%).

Statistical Methods

Baseline patient characteristics were compared across strata of LVEF using the linear trend test for continuous variables and the Mantel-Haenszel trend test for categorical variables. All results were reported as a mean ± standard deviation or frequencies and percentages for continuous and categorical data, respectively.

To quantify between-group differences in quality of life benefits following CTO-PCI, we performed hierarchical multivariable linear or logistic regression, as appropriate, with LVEF as a categorical variable and with normal LVEF as the reference category. Health status scores for each domain, at one year, were the primary outcome. We adjusted for potential confounders based upon prior literature (15–17), including baseline health status score, age, sex, body mass index, smoking status, history of diabetes mellitus, history of chronic obstructive lung disease, history of peripheral arterial disease, history of chronic kidney disease, history of coronary artery bypass grafting, and physiologically complete revascularization at the time of the baseline CTO-PCI procedure. An interaction between baseline health status and LV function was also analyzed. Both SAQ-SS and RDS scores were treated as continuous variables, and SAQ-AF was defined as either no (100) versus any (<100) reported angina.

We performed two sensitivity analyses, where we examined (i) only those patients with single vessel disease undergoing CTO-PCI and (ii) patients with both successful and unsuccessful procedures. These were done in order to provide a clearer understanding of the role of the CTO when assessing health status change following CTO-PCI as well as overall procedure effectiveness in real-world practice, respectively. Similarly, an interaction between baseline health status and LV function was also assessed.

Few patients were missing data for baseline covariates (<2%). Of 762 patients, 699 (91.7%) had LVEF documented within 12 months of CTO-PCI and 63 (8.3%) had documentation at ≥12 months or were missing the date at which LVEF was assessed. Due to low rates of missing data, no data imputation methods were used. A pre-set alpha level of 0.05 was considered statistically significant and a 5-point change in the SAQ summary score and angina frequency score was considered clinically meaningful. All statistical analyses were performed using SAS version 9.4 software (SAS Institute, Inc., Cary, NC). Analyses were performed at Saint Luke’s Mid-America Heart Institute, and the lead authors take responsibility for guiding data analysis and the interpretation of the results.

RESULTS

Baseline Patient Characteristics, Angiographic Characteristics, and Health Status

The baseline characteristics of the primary analytic cohort are described in Table 1. Of the analysis sample, 506 (66.4%), 193 (25.3%), and 63 (8.3%) patients had normal, mild-moderate, and severely reduced LVEF, respectively. Patients with normal cardiac function were less likely to have had a documented history of myocardial infarction (p < 0.001), non-CTO revascularization (p = 0.002), chronic kidney disease (p = 0.001), and heart failure (p < 0.001) than those with mild-moderate and severely reduced function. There were no statistically significant differences in baseline SAQ-SS (p = 0.30), SAQ-AF (p = 0.11), and RDS (p = 0.92) scores between groups.

Table 1.

Baseline Characteristics of the Primary Study Population According to LVEF (n = 762)

	No Dysfunction	Mild-Moderate Dysfunction	Severe Dysfunction	P-Value
	N = 506	N = 193	N = 63
Age	64.8 ± 10.3	66.4 ± 10.4	64.7 ± 10.5	0.975
Male – no. (%)	388 (76.7%)	159 (82.4%)	54 (85.7%)	0.032
Race – no (%)				0.587
White	451 (89.1%)	183 (94.8%)	55 (87.3%)
Black	21 (4.2%)	5 (2.6%)	1 (1.6%)
Other	34 (6.7%)	5 (2.6%)	7 (11.1%)
History of MI	199 (39.3%)	125 (64.8%)	42 (66.7%)	<0.001
History of COPD	73 (14.4%)	27 (14.0%)	11 (17.5%)	0.684
History of Diabetes Mellitus	198 (39.1%)	77 (39.9%)	26 (41.3%)	0.729
BMI	30.7 ± 6.1	30.7 ± 6.4	28.5 ± 5.1	0.006
History of PAD	79 (15.6%)	40 (20.7%)	10 (15.9%)	0.365
History of CKD	49 (9.7%)	38 (19.7%)	11 (17.5%)	0.001
History of CKD and Dialysis	9 (1.8%)	8 (4.1%)	2 (3.2%)	0.141
History of Heart Failure	56 (11.1%)	77 (39.9%)	42 (66.7%)	<0.001
Current Smoker	71 (14.2%)	31 (16.1%)	7 (11.5%)	0.949
Prior Viability Testing	76 (15.0%)	38 (19.7%)	24 (38.1%)	<0.001
Any Non-CTO PCI	58 (11.5%)	30 (15.5%)	16 (25.4%)	0.002
Viability in Region of CTO	76 (100%)	34 (89.5%)	21 (87.5%)	0.004
History of CABG	155 (30.6%)	78 (40.4%)	20 (31.7%)	0.146
Total CTO Lesions	1.04 ± 0.20	1.08 ± 0.29	1.05 ± 0.21	0.839
Total Coronary Lesions
1	428 (84.6%)	153 (79.3%)	45 (71.4%)	<0.001
2	75 (14.8%)	33 (17.1%)	15 (23.8%)
3	3 (0.6%)	7 (3.6%)	3 (4.8%)
Complete Revascularization	431 (85.7%)	159 (82.4%)	55 (87.3%)	0.740

Abbreviations: MI (myocardial infarction); BMI (body mass index); PAD (peripheral arterial disease); CKD (chronic kidney disease); CABG (coronary artery bypass grafting); PCI (percutaneous intervention)

There were, also, no significant differences in total number of CTO lesions between patients with normal, mild-moderate, and severely reduced LVEF groups as assessed during angiography (p = 0.839). However, patients with normal cardiac function were less likely to have had two or three angiographically significant lesions (CTO and non-CTO) compared to those with mild-moderate and severely reduced function (p < 0.001).

Change in Health Status After CTO PCI

All patients undergoing successful CTO-PCI demonstrated large, clinically meaningful improvements in QoL scores (Table 2). Patients with normal cardiac function, on average, reported an unadjusted mean improvement of 27.1 ± 20.4 points in SAQ-SS, 23.4 ± 25.3-point improvement in SAQ-AF, and −1.1 ±1.5-point improvement in RDS. Patients with mild-moderately reduced cardiac function experienced a mean 26.7 ± 21.2-point improvement in SAQ-SS score, 24.6 ± 26.2-point improvement in SAQ-AF score, and −1.0 ± 1.7-point improvement in RDS score. Finally, patients with severely reduced LV function demonstrated a 20.3 ± 18.1-point improvement in SAQ-SS, 17.2 ± 29.8-point improvement in SAQ-AF, and −0.8 ± 1.5-point improvement in RDS score. Change in score, between baseline and 12 months, was similar between patients with normal, mild-moderate, and severely reduced LVEF for SAQ-AF (0.10) and RDS (0.11), but not for SAQ-SS (0.02).

Table 2.

Mean Health Status Scores at Baseline, 12-Months, and Change Between Baseline and 12-Month Follow-Up Across LVEF Groups (n = 762)

	None (LVEF (≥50%)	Mild-Moderate (LVEF 30–49%)	Severe LVEF (<30%)	P-Value
SAQ AF Baseline	70.00 ± 26.56	70.57 ± 27.52	75.71 ± 26.13	0.11
SAQ AF 12-Months	93.36 ± 16.67	95.06 ± 14.51	92.64 ± 17.78	0.76
SAQ AF Score ∆	23.36 ± 25.33	24.61 ± 26.21	17.17 ± 29.77	0.10
SAQ Summary Baseline	61.40 ± 22.33	61.87 ± 22.78	64.49 ± 20.96	0.30
SAQ Summary 12-Months	88.79 ± 14.63	88.89 ± 13.45	84.12 ± 17.25	0.03
SAQ Summary Score ∆	27.13 ± 20.35	26.73 ± 21.20	20.26 ± 18.14	0.02
RDS Baseline	2.22 ± 1.49	2.33 ± 1.53	2.24 ± 1.42	0.92
RDS 12-Months	1.08 ± 1.34	1.25 ± 1.41	1.46 ± 1.60	0.06
RDS Score ∆	−1.13 ± 1.51	−1.00 ± 1.65	−0.76 ± 1.54	0.11

Mean health status scores (SAQ-SS, SAQ-AF, and RDS) at baseline and 12-month assessment as well as change between baseline and 12 months for patients with normal, mild-moderate, and severely reduced LVEF.

In a final, parsimonious model, patients with mild-moderately reduced LVEF demonstrated similarly robust improvements in QoL compared to those with a normal LVEF following CTO-PCI. Accordingly, between the two cohorts, there were no statistically significant differences in SAQ-SS improvement (0.00-point higher, 95% CI −2.2, 2.3; p = 0.99), odds of experiencing any angina (OR 0.8; 95% CI 0.5, 1.3; p = 0.37), and RDS improvement (0.1-point less improvement, 95% CI −0.1, 0.4; p = 0.22; Figures 2–3).

Figure 2.

Adjusted Differences in SAQ-SS and SAQ-AF Score for Mild-Moderate and Severely Reduced LVEF Groups (Compared to Normal LVEF Group) Between Baseline and 12-Month Follow-Up (n = 762).

Legend: Multivariable-adjusted differences in SAQ-SS and SAQ-AF improvement between patients with mild-moderate or severely reduced LVEF and those with normal LVEF

Figure 3.

Adjusted Differences in RDS Score for Mild-Moderate and Severely Reduced LVEF Groups (Compared to Normal LVEF Group) Between Baseline and 12-Month Follow-Up (n = 762).

Legend: Multivariable-adjusted differences in RDS improvement between patients with mild-moderate or severely reduced LVEF and those with normal LVEF

Following adjustment, CTO-PCI still conveyed a clinically large improvement in QoL for patients with severely reduced LVEF, though the magnitude of the effect was slightly less robust for overall health status (SAQ-SS −5.2; 95% CI −9.0, −1.5; p = 0.01) and dyspnea relief (RDS score + 0.5; 95% CI 0.1, 0.8; p = 0.01) compared to patients with a normal LVEF at baseline. However, there were no significant differences in presence of any angina (OR 1.3; 95% CI 0.6–3.0; p = 0.48) between the best and worst LVEF cohorts (Figures 2–3). All interactions between baseline health status and LVEF categories were non-significant.

Health Status Benefit of CTO-PCI by LVEF in Sensitivity Analyses

Results of our sensitivity analyses were both qualitatively consistent with our primary findings. When considering only those patients with CTOs (i.e. without other obstructive coronary disease), there were 448, 163, and 47 patients with normal, mild-moderately reduced, and severely reduced LVEF, respectively. Compared to patients with normal LVEF, those with severely reduced function had significantly less improvement in overall health status (SAQ SS −6.8; p = 0.001) and dyspnea resolution (RDS +0.4; p = 0.04) without any difference in residual odds of angina presence (SAQ-AF OR 1.8; p = 0.17). However, those with mild-moderately reduced LVEF had no differences in score improvement when compared to patients with normal LVEF as measured by SAQ-SS (p = 0.55), SAQ-AF (p = 0.55), and RDS (p = 0.09) (Supplemental Tables 1–3). Similar findings were observed across a combined cohort of patients with successful and unsuccessful CTO-PCI (Supplemental Tables 4–6).

DISCUSSION

We studied change in patients’ health status from baseline to 12-months after successful CTO-PCI among patients with normal, mild-moderate, and severely reduced LVEF in the OPEN-CTO Registry. We found that: (i) regardless of baseline LVEF, there were large improvements in health status by 12 months after the index CTO-PCI; and (ii) patients with severely reduced LVEF had slightly less improvement in generic and disease-specific health status measures than those with normal cardiac function, despite similar improvements in angina relief. These findings were consistent across unadjusted and multivariable-adjusted analyses. This study, of CTO-PCI among patients in a real-world setting, represents an important contribution on the health status benefits of CTO-PCI and highlights the utility of this procedure.

The OPEN-CTO registry is the largest observational cohort of CTO revascularization to date that documents disease-specific health status before and after treatment, and its inclusion of patients with LVSD contributes important insight to a literature comprised primarily of patients with normal cardiac function (18). Though the effects of CTO-PCI on change in LVEF and ventricular strain have been examined (19,20), the relationship between ventricular function and health status remains poorly defined (21,22). However, there is growing evidence describing the health status benefits of CTO-PCI among patients with reduced LVEF (18,19,23–27). Cardona et al. prospectively studied the influence of CTO revascularization on health status outcomes in patients with evidence of CTO territorial viability and documented a beneficial reduction in angina (19). While clinically significant, those conclusions were diminished by a small sample size, an unclear definition of “angina,” and lack of classification according to LVSD. Importantly, this complements the information obtained using the SAQ and RDS questionnaires in OPEN-CTO, which have proven to be both responsive and sensitive to change in health status over time. Wijeysundera and colleagues conducted a multicenter prospective analysis to compare the impact of PCI or surgical-guided CTO revascularization versus standard-of-care medical treatment on change in SAQ-SS between baseline and 12-months (24). Patients treated with CTO revascularization, as compared with the medical therapy arm, experienced significant improvements in multiple SAQ domains at 12 months, but were not categorized by conventional gradations of cardiac dysfunction. Finally, the FACTOR trial investigators collected 1-month SAQ health status on patients referred for CTO-PCI and reported that the benefits of intervention applied primarily towards those with symptomatic angina (25). However, while its results were limited by short-term data collection and exclusion of patients with LVEF < 30%, our findings document 12-month health status benefits in patients with normal to severely reduced ventricular function.

Our finding that, despite a robust reduction in angina burden for all patients with CTO, patients with severely compromised cardiac function experienced less overall health status improvement at one year compared to those with normal cardiac function warrants further discussion. For one, it is clear that patients with LVSD and heart failure have other comorbidities that may impact quality of life – namely, anemia, atrial fibrillation, chronic kidney disease, chronic obstructive pulmonary disease, and depression (28–30). It is also possible that this difference in improvement reflects residual severe cardiomyopathy, and that residual cardiomyopathy results in ongoing dyspnea symptoms and reduced functional capacity that may be difficult to parse out from angina as a “dyspnea equivalent” when using generic health instruments such as the RDS. To that end, this might explain why there was no significant differences in the magnitude of improvement in SAQ-AF domain scores across groups, as opposed to differences in overall health status benefit. Third, it is noteworthy that viability testing was performed in only 38.1% of patients with severely reduced LVEF and those patients had less viability (and, perhaps, less potential for health status improvement) in the CTO territory (87.5%) compared to those with mild-moderate (89.5%) and normal (100%) LVEF, albeit without a consistent viability testing strategy. Finally, it is important to emphasize that despite a smaller temporal health status benefit in patients with reduced, compared to normal, LVEF, this between-group difference is overwhelmed by large health status benefits observed across all LVEF groups.

Our findings are limited by the observational nature of this registry, and residual or unmeasured confounding cannot be excluded. Although there is always a possibility of confounding in observational studies, the finding of similar benefits across a range of LVEF categories suggests that successful CTO-PCI can improve health status regardless of patients’ baseline LVEF. Second, this study did not address major adverse cardiac events or mortality. Although improving patients’ health status is one of multiple primary goals of treatment with CTO-PCI, these important outcomes should be evaluated in future studies of the impact of CTO-PCI on outcomes in patients with LVSD. Third, OPEN-CTO was an observational study of patients selected for management with PCI. Without a comparison group of patients treated with surgery or medical therapy alone (and exclusion of those with an unsuccessful procedure), the incremental benefit of CTO-PCI compared to these alternative management strategies cannot be defined. While this comparison is important and warrants further evaluation in future trials, given recent randomized trial observations of incremental SAQ improvement in patients treated with optimal medical therapy, alone (31), this was not a goal of the current analysis or the OPEN-CTO study. Fourth, while patients with both low and intermediate to high-risk findings on stress testing may experience health status benefit from CTO-PCI (27), viability testing was performed in less than half of patients within each LVEF category. Finally, follow-up echocardiogram data were not collected in OPEN-CTO. Accordingly, we were unable to measure the incremental change in LVEF following CTO-PCI, and it is unknown whether there was greater symptomatic benefit in patients with improved ventricular function.’

CONCLUSION:

The incidence of a CTO in patients with angina poses a clinical treatment dilemma to physicians whose patients’ symptoms are refractory to medical therapy. Our findings suggest that, although patients with LVEF < 30% have less overall health status improvement at one year compared to those with normal cardiac function, patients may experience large, sustained benefits in angina symptom burden, dyspnea, and overall quality of life following CTO-PCI, regardless of LVEF.

Clinical Perspectives.

What is Known?

Prior studies have shown that patients undergoing successful CTO revascularization experience improved left ventricular function and survival, but the impact of successful CTO-PCI on long-term angina burden and quality of life in patients is unclear.

What is New?

We found that patients experience large, sustained improvements in angina symptom burden, dyspnea, and overall quality of life following CTO-PCI across the spectrum of baseline ventricular function.

What is Next?

Future studies are needed to confirm the effects of this procedure on long-term health status in patients with LVSD.

ABBREVIATIONS LIST:

CTO

Chronic Total Occlusion

LVEF

Left Ventricular Ejection Fraction

LVSD

Left Ventricular Systolic Dysfunction

HF

Heart Failure

QoL

Quality of Life

SAQ

Seattle Angina Questionnaire

RDS

Rose Dyspnea Scale

PCI

Percutaneous Coronary Intervention