Age, creatinine, and ejection fraction (ACEF) score predict clinical outcomes in patients who underwent elective percutaneous coronary intervention (PCI) of nonocclusive coronary stenoses. We aimed at assessing the prognostic value of the ACEF score in patients who underwent successful PCI of chronic total occlusion (CTO). ACEF score was calculated in 587 patients treated with PCI of CTO: successful in 433 (74%; success group) and failed in 154 patients (26%; failure group). Patients were divided in ACEF tertiles: first <0.950, second from 0.950 to 1.207, and third ACEF tertile >1.207. Major adverse cardiac events (MACE = overall death + nonfatal myocardial infarction + clinically driven target vessel revascularization) were assessed in 558 patients (95%) up to 24 months (8 to 24 months). In success group, higher MACE rate was significantly associated with increasing ACEF tertile (first = 7%, second = 13%, third ACEF = 18%, p = 0.02). MACE-free survival was significantly decreased with increasing ACEF tertile (log-rank 5.58, p = 0.018). In the failure group, lower MACE rate was significantly associated with increasing ACEF tertile (p = 0.041). This was mainly driven by significant decreasing rate of target vessel revascularization along the tertiles (first = 34%, second = 19%, third ACEF = 10%, p = 0.007). Compared with success group, in failure group, MACE rate was significantly higher in the first tertile (p <0.001) and similar in the third tertile (p = 0.59). In conclusion, ACEF score represents a simple tool in the prognostication of patients successfully treated with PCI of CTO and identifies those patients who would not derive any significant clinical harm despite failed percutaneous revascularization of the CTO.
Coronary chronic total occlusions (CTOs) are the most challenging coronary lesions for percutaneous coronary interventions (PCI), with a success rate ranging from 55% to 100%. Successful PCI of CTOs is associated with improved long-term clinical outcomes compared with conservative management. Nevertheless, the clinical outcome even after successful recanalization remains worse compared with patients with non-CTO stenoses who underwent PCI. As CTO patients are generally older and frequently affected by other co-morbidities, it is unclear whether this unfavorable clinical outcome is driven by their intrinsic frailty rather than by the challenging coronary anatomy. Several risk scores based on clinical and angiographic features have assessed the probability of procedural success. The clinical SYNTAX score improved decision making between PCI and coronary artery bypass grafting (CABG). The Euroscore, although often used for prediction of early mortality after open-heart surgery, has been recently proved to reliably predict short- and long-term prognosis in complex coronary interventions. More specifically, the selection of CTO vessels subtending large myocardial ischemia or viability contributes to target patients who will derive most of the benefit from percutaneous revascularization. Last, the age, creatinine, and ejection fraction (ACEF) score has been not only validated in the setting of elective CABG but also used to risk stratify patients treated with PCI. We aimed in this multicenter retrospective registry at assessing the prognostic value of the ACEF score in patients who underwent successful PCI of CTO.
Methods
From January 2006 to September 2012, 715 consecutive patients who underwent PCI of CTO in major native coronary arteries were screened at 3 centers (Cardiovascular Center Aalst OLV Clinic, Belgium; Royal Brompton Hospital, United Kingdom; and Oost-Limburg Hospital, Belgium). The ACEF score was calculated retrospectively for 587 patients eventually enrolled in the registry. ACEF score could not be calculated in 128 patients because of unavailability of renal function or ejection fraction (EF) before PCI; therefore, they were excluded. All patients had either angina and/or viability/reversible myocardial ischemia in the territory of the occluded artery, assessed either by stress echocardiography or myocardial perfusion scintigraphy. A CTO was defined as a complete obstruction of a native coronary artery with duration longer than 3 months and Thrombolysis In Myocardial Infarction flow vessel grade 0. A successful CTO procedure was defined as coronary revascularization with final Thrombolysis In Myocardial Infarction flow grade ≥2 and full coverage with stents of the occluded segment. Values of troponins, creatinine, and hemoglobin were collected at baseline and at 12 to 24 hours after the procedure to detect possible periprocedural-myocardial infarction (PMI), acute kidney injury, and major bleedings. The local Ethics Committees approved the use of clinical data for this study, and all patients provided written informed consent.
All patients were pretreated with aspirin and clopidogrel before the procedure. Anterograde or retrograde PCI approaches were used at operator’s discretion depending on the coronary anatomy, lesion morphology, and previous failed procedures. After successful CTO revascularization, coronary stenting was uniformly performed. During the procedure, heparin was administered and activated clotting time regularly checked to maintain an activated clotting time ≥250 to 300 seconds. All patients with successful vessel revascularization received dual antiplatelet therapy with aspirin indefinitely and clopidogrel for at least 12 months after procedure. Quantitative coronary angiography was performed using the computer-based analysis system Siemens QuantCor QCA (ACOM.PC 5.01; Siemens Medical Systems Inc, Malvern, Pennsylvania), based on the CAAS II system (Pie Medical Imaging, Maastricht, The Netherlands).
The ACEF score was calculated according to the following formula: ACEF = age/left ventricular EF (+1 if creatinine was >2.0 mg/dl). Analysis per tertile of ACEF score and clinical follow-up evaluation were performed in all patients. Primary end point of the study was the 2-year cumulative incidence of major adverse cardiovascular events (MACEs), defined as the composite of overall death, nonfatal myocardial infarction, and clinically driven target vessel revascularization (TVR). Secondary end points were in-hospital incidence of PMI, major bleedings, and acute kidney injury and also overall death, nonfatal myocardial infarction, and clinically driven TVR at 2 years. PMI and major bleeding were defined as previously described. TVR was defined as any attempted percutaneous or surgical revascularization of the target vessel after the index procedure in the presence of recurrent angina or proved ischemia. Clinical follow-up was collected through outpatient clinic evaluations, hospital records, or telephone interviews.
All variables were stratified according to ACEF tertiles. Continuous variables expressed as mean ± SD were compared using 1-way analysis of variance. Categorical data presented as frequency (percentages) were compared using the Fisher’s exact test or the Pearson chi-square test. Normal distribution was assessed by the Kolmogorov-Smirnov test. Two-year cumulative rates of MACE were estimated by the Kaplan-Meier method. The Cox proportional hazards model was used to compare clinical outcomes between the groups. C-statistics and receiver operating characteristic curves were constructed to assess the ability of ACEF to predict events. A p <0.05 was considered significant, and all tests were 2 tailed. Data were analyzed with SPSS, version 17.0, software (SPSS Inc., Chicago, Illinois) and GraphPad Prism 5.00 (GraphPad Software Inc., La Jolla, California).
Results
PCI of CTO was successful in 433 patients (74%; success group), whereas it failed in 154 patients (26%; failure group; Table 1 ). Patients of the 2 groups were stratified according to the ACEF score tertiles as follows: (a) success group, first ACEF <0.950 (n = 147), second ACEF from 0.950 to 1.207 (n = 136), and third ACEF >1.207 (n = 150); (b) failure group, first ACEF <0.950 (n = 47), second ACEF from 0.950 to 1.207 (n = 55), and third ACEF >1.207 (n = 52). As expected by the stratification, in the success group, significant differences were observed among tertiles in age, creatinine levels, left ventricular EF, and ACEF score. Rate of previous CABG was also significantly different among tertiles. Likewise in the failure group, the same significant differences were observed. In addition, rate of previous PCI and CABG was also significantly different among tertiles. Significant differences were also observed between the success and failure group in the second (for age, EF, Peripheral Artery Disease, and previous CABG) and third ACEF tertile (for age).
| Variable | Success | p Value | Failure | p Value | p Values (Success vs. Failure) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| First Tertile ACEF | Second Tertile ACEF | Third Tertile ACEF | First Tertile ACEF | Second Tertile ACEF | Third Tertile ACEF | First Tertile ACEF | Second Tertile ACEF | Third Tertile ACEF | |||
| 147 (25%) | 136 (23%) | 150 (26%) | — | 47 (24%) | 55 (29%) | 52 (26%) | — | ||||
| Age (yrs) | 55 (50–60) | 64 (58–70) | 69 (64–77) | <0.001 | 60 (52–65) | 70 (63–75) | 75 (66–79) | <0.001 | 0.34 | <0.001 | 0.014 |
| Men | 116 (79%) | 109 (80%) | 119 (79%) | 0.97 | 39 (83%) | 43 (78%) | 41 (78%) | 0.81 | 0.68 | 0.84 | 1.00 |
| Ejection fraction (%) | 71 (62–80) | 60 (55–65) | 50 (41–55) | <0.001 | 70 (64–80) | 66 (56–70) | 50 (44–57) | <0.001 | 0.93 | 0.002 | 0.14 |
| Smoker | 76 (52%) | 56 (41%) | 63 (42%) | 0.16 | 27 (57%) | 30 (54%) | 19 (36%) | 0.07 | 0.76 | 0.24 | 0.57 |
| Hypertension | 78 (53%) | 84 (62%) | 89 (59%) | 0.30 | 28 (60%) | 32 (58%) | 35 (67%) | 0.59 | 0.50 | 0.74 | 0.33 |
| Hyperlipidemia | 111 (75%) | 105 (77%) | 102 (68%) | 0.17 | 30 (64%) | 42 (76%) | 42 (81%) | 0.14 | 0.13 | 1.00 | 0.11 |
| Diabetes | 27 (18%) | 36 (26%) | 43 (29%) | 0.10 | 13 (28%) | 15 (27%) | 16 (31%) | 0.91 | 0.21 | 1.00 | 0.86 |
| Peripheral artery disease | 13 (9%) | 10 (7%) | 11 (7%) | 0.86 | 5 (11%) | 11 (20%) | 8 (15%) | 0.43 | 0.77 | 0.019 | 0.10 |
| Creatinine, (mg/dl) | 0.96 (0.80–1.06) | 1.01 (0.84–1.17) | 1.12 (0.95–1.29) | 0.001 | 0.85 (0.77–1.01) | 1.03 (0.82–1.22) | 1.08 (0.96–1.23) | 0.001 | 0.05 | 0.53 | 0.57 |
| Previous PCI | 55 (37%) | 66 (48%) | 57 (38%) | 0.10 | 17 (36%) | 28 (51%) | 14 (27%) | 0.036 | 1.00 | 0.87 | 0.18 |
| Previous CABG | 16 (11%) | 30 (22%) | 25 (17%) | 0.040 | 7 (15%) | 20 (36%) | 11 (21%) | 0.033 | 0.45 | 0.047 | 0.53 |
| Atypical chest pain | 7 (5%) | 8 (6%) | 7 (5%) | 0.73 | 2 (4%) | 3 (5%) | 3 (6%) | 0.63 | 0.63 | 0.62 | 0.87 |
| Stable angina | 140 (95%) | 128 (94%) | 143 (95%) | 0.88 | 45 (96%) | 52 (94%) | 49 (94%) | 0.94 | 1.00 | 1.00 | 0.72 |
| ACEF score | 0.81 (0.71–0.88) | 1.05 (1.00–1.13) | 1.45 (1.31–1.67) | <0.001 | 0.83 (0.76–0.89) | 1.07 (1.00–1.14) | 1.41 (1.32–1.63) | <0.001 | 0.54 | 0.58 | 0.68 |
In the success group, a significant difference among tertiles was observed in the number of vessels and rate of multivessel disease, distribution of the target vessel, success rate at first attempt, and contrast amount used. In addition, PMI was also significantly different among tertiles ( Table 2 ). In the failure group, a significant difference among tertiles was only observed in the rate of anterograde and retrograde approach. Significant differences were observed between the success and failure group in the first (for number of vessel diseased, anterograde and retrograde approach) and second ACEF tertile (for number and rate of multivessel disease and x-ray time). At discharge, in the success group, the rate of dual antiplatelet therapy was higher (350 [81%] vs 78 [51%], p <0.001), whereas the rate of β blockers was lower (242 [56%] vs 103 [67%], p = 0.032) compared with the failure group.
| Variable | Success | p Value | Failure | p Value | p Values (Success vs. Failure) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| First Tertile ACEF | Second Tertile ACEF | Third Tertile ACEF | First Tertile ACEF | Second Tertile ACEF | Third Tertile ACEF | First Tertile ACEF | Second Tertile ACEF | Third Tertile ACEF | |||
| No. vessel disease | <0.001 | 0.16 | 0.015 | 0.041 | 0.18 | ||||||
| 1 | 93 (63%) | 74 (54%) | 42 (28%) | 27 (57%) | 19 (34%) | 21 (40%) | |||||
| 2 | 44 (30%) | 37 (27%) | 58 (39%) | 10 (21%) | 20 (37%) | 14 (27%) | |||||
| 3 | 10 (7%) | 25 (19%) | 50 (33%) | 10 (22%) | 16 (29%) | 17 (33%) | |||||
| Multivessel disease | 54 (37%) | 62 (46%) | 108 (72%) | <0.001 | 20 (43%) | 36 (65%) | 31 (60%) | 0.06 | 0.49 | 0.016 | 0.12 |
| CTO target vessel | 0.040 | 0.06 | 0.36 | 0.16 | 0.54 | ||||||
| Left anterior descending | 40 (27%) | 33 (24%) | 59 (39%) | 8 (17%) | 8 (14%) | 16 (31%) | |||||
| Circumflex coronary | 31 (21%) | 24 (18%) | 27 (18%) | 12 (25%) | 7 (13%) | 11 (21%) | |||||
| Right coronary | 76 (52%) | 79 (58%) | 64 (43%) | 27 (57%) | 40 (73%) | 25 (48%) | |||||
| Success at first attempt | 118 (92%) | 98 (80%) | 114 (81%) | 0.015 | — | — | — | ||||
| In-stent restenosis | 12 (8%) | 6 (4%) | 14 (9%) | 0.26 | 2 (4%) | 3 (5%) | 2 (4%) | 0.92 | 0.52 | 0.52 | 0.25 |
| Anterograde approach | 136 (92%) | 117 (86%) | 133 (89%) | 0.19 | 38 (81%) | 41 (74%) | 50 (96%) | 0.008 | 0.045 | 0.19 | 0.17 |
| Retrograde approach | 11 (7%) | 19 (14%) | 17 (11%) | 0.21 | 9 (19%) | 14 (25%) | 2 (4%) | 0.012 | 0.029 | 0.09 | 0.17 |
| Multivessel PCI | 28 (19%) | 26 (19%) | 29 (19%) | 1.00 | 7 (15%) | 15 (27%) | 7 (13%) | 0.13 | 0.66 | 0.24 | 0.40 |
| Stent used per patient, n | 2.4 ± 1.3 | 2.4 ± 1.1 | 2.5 ± 1.2 | 0.73 | — | — | — | ||||
| Drug-eluting stent | 139 (95%) | 131 (97%) | 138 (93%) | 0.23 | — | — | — | — | — | — | |
| Stent diameter (mm) | 3.0 (2.8–3.4) | 3.0 (2.8–3.4) | 3.0 (2.8–3.3) | 0.75 | — | — | — | — | — | — | |
| Stent length (mm) | 54 (28–77) | 48 (34–69) | 49 (31–69) | 0.74 | — | — | — | — | — | — | |
| X-ray time (min) | 29 (18–44) | 28 (20–49) | 33 (21–47) | 0.52 | 29 (19–55) | 40 (26–68) | 36 (25–57) | 0.17 | 0.42 | 0.007 | 0.08 |
| Contrast medium (ml) | 350 (250–475) | 300 (220–420) | 280 (200–400) | 0.015 | 330 (221–451) | 300 (225–400) | 270 (200–350) | 0.09 | 0.50 | 0.84 | 0.16 |
| In-hospital outcome | |||||||||||
| Peri-procedural infarction | 33 (22%) | 47 (35%) | 53 (35%) | 0.028 | 9 (19%) | 19 (34%) | 14 (27%) | 0.22 | 0.69 | 1.00 | 0.31 |
| TIMI major bleedings | 4 (3%) | 1 (1%) | 0 (0%) | 0.24 | 1 (3%) | 2 (5%) | 1 (3%) | 0.88 | 1.00 | 0.27 | 0.31 |
| Acute kidney injury | 2 (1%) | 3 (2%) | 4 (3%) | 0.73 | 0 (0%) | 2 (4%) | 3 (6%) | 0.26 | 1.00 | 0.63 | 0.38 |
Follow-up at 24 months was obtained in 558 of 587 patients (95%) with a median of 24 months (8 to 24 months). The cumulative incidence of the end points across the ACEF tertiles is listed in Table 3 . In the success group, a significant increasing rate of death and MACE was observed along the ACEF tertiles. After adjustment, higher MACE rate remained significantly associated with increasing ACEF tertile. MACE-free survival was significantly decreased with increasing ACEF tertile ( Figure 1 ). In the failure group, a significant decreasing rate of TVR was observed along the ACEF tertiles. After adjustment, lower MACE rate was significantly associated with increasing ACEF tertile. A trend to lower MACE-free survival was observed in the first ACEF tertile ( Figure 1 ).
