Author + information
- Received May 2, 2013
- Revision received July 1, 2013
- Accepted August 1, 2013
- Published online December 1, 2013.
- Shao-Liang Chen, MD∗∗ (, )
- Ya-Ling Han, MD†∗∗ (, )
- Yao-Jun Zhang, PhD‡,
- Fei Ye, MD∗,
- Hai-Wei Liu, MD†,
- Jun-Jie Zhang, PhD∗,
- Bo Xu, MBBS§,
- Tie-Min Jiang, MD‖,
- Yu-Jie Zhou, MD¶ and
- Shu-Zheng Lv, MD¶
- ∗Cardiology Department, Nanjing First Hospital, Nanjing Medical University, Nanjing, China
- †Cardiology Department, Shenyang Northern Hospital, Shenyang, China
- ‡Cardiology Department, Nanjing Heart Center, Nanjing, China
- §Cardiology Department, Beijing Fuwai Hospital, Beijing, China
- ‖Cardiology Department, Tianjin Armed Policemen Hospital, Tianjin, China
- ¶Cardiology Department, Beijing Anzhen Hospital, Beijing, China
Objectives The present study aimed to establish a risk score using a simple calculation with an enhanced predictive value for major adverse cardiac events (MACE) in patients with unprotected left main coronary artery (UPLMCA) disease after the implantation of a drug-eluting stent (DES).
Background The anatomic-, clinical-, and procedure-based NERS (New Risk Stratification) score was superior to the SYNTAX (Synergy Between Percutaneous Coronary Intervention With TAXUS and Cardiac Surgery) score in predicting MACE after stenting UPLMCA. The complexity of the calculation was its major limitation.
Methods The NERS score II was derived from our previous 2 studies and externally compared with the NERS and SYNTAX scores in 1,463 patients with UPLMCA disease who underwent implantation of a DES in a prospective, multicenter registry trial. The primary endpoint was MACE at 1 year after the index procedure, including myocardial infarction, cardiac death, and target vessel revascularization.
Results The NERS score II system consisted of 16 (7 clinical and 9 angiographic) variables. A NERS score II ≥19 demonstrated enhanced MACE sensitivity and specificity of 84.0% and 76.0% (MACE as the state variable), respectively, which were similar to the NERS score but significantly higher compared with the SYNTAX score. A NERS score II ≥19 was the only independent predictor of cumulative MACE (hazard ratio: 3.27; 95% confidence interval [CI]: 1.86 to 5.23; p ≤ 0.001) and stent thrombosis (odds ratio: 22.15; 95% CI: 12.47 to 57.92; p ≤ 0.001) at follow-up.
Conclusions The NERS score II, similar to the conventional NERS score, is more predictive of MACE than the SYNTAX score in UPLMCA patients after implantation of a DES.
Risk stratifications for patients with unprotected left main coronary artery (UPLMCA) disease have been established as important tools to predict clinical events after stenting UPLMCA (1–3). Our previous study has shown that, with the combination of clinical and procedural variables, the NERS (New Risk Stratification) score is more predictive of major adverse cardiac events (MACE) than the SYNTAX (Synergy Between Percutaneous Coronary Intervention With TAXUS and Cardiac Surgery) score (2). However, the complex calculation of the NERS score limits its routine use in everyday practice.
The purpose of the present study was to enhance the anatomic- and clinical-based NERS score to create the NERS score II. The NERS score II should provide the equivalent power to the conventional NERS score in the prediction of clinical outcomes after stenting UPLMCA, thus aiding appropriately individualized decision making.
Between March 19, 2003, and December 30, 2011, 2,060 patients with UPLMCA who underwent percutaneous coronary intervention (PCI) with implantation of a drug-eluting stent (DES) were prospectively studied at 6 tertiary centers. Of these patients, 597 patients—260 in the DISTAL (Drug-Eluting Stent for the Treatment of Left Main Disease) study (3) and 337 in the NERS study (2)—were excluded from the current study. Finally, 1,463 patients were enrolled in this study. The study protocol was approved by the ethics committee of each participating center.
The revascularization strategy for each patient with UPLMCA was jointly formulated by cardiologists and cardiac surgeons, based upon clinical and angiographic features. UPLMCA patients undergoing stent implantation were consecutively enrolled, and all data were entered into a dedicated database.
NERS Score II
The NERS score (2) was derived from 260 UPLMCA patients included in the DISTAL study (3) and was externally tested in 337 patients in the NERS study. Of 54 variables (including 17 clinical, 4 procedural, and 33 angiographic variables) in the NERS score, several variables remained repeatedly calculated. For example, “requiring an intra-aortic balloon pump” overlapped with “cardiogenic shock” and “lower left ventricular ejection fraction,” “not taking a statin” overlapped with “a higher low-density lipoprotein level.” Similarly, lesions in non-left main (LM) vessels were grouped according to chronic total occlusion (CTO) and non-CTO lesions, respectively, and lesions specificities in the right coronary artery and the left circumflex coronary artery had similar predictive value of MACE. Therefore, the 54 variables included in the NERS score were regrouped into 7 clinical variables, 5 angiographic variables for LM, and 4 angiographic variables for non-LM vessels. These new variables and the conventional 54 variables were put into the model and analyzed in 597 patients by logistic regression, and the probability score for each variable was then calculated. Variables with a p value <0.01 were considered significant predictors and formed the NERS score II. Finally, the NERS score II consisted of 16 variables (7 clinical, 5 lesion variables for LM, and 4 lesion variables for non-LM vessels) (Table 1). Similar to the NERS score, for ostial and shaft or distal LM stenosis, the highest score for distal LM lesions was considered the final score for the LM. Similarly, tandem stenoses within 1 vessel, such as CTO with bifurcation, were scored as the highest score of any lesion (i.e., CTO).
The primary endpoint was the rate of MACE, defined as cardiogenic death, myocardial infarction (MI), or target vessel revascularization (TVR) at follow-up. Stent thrombosis (ST) was assessed as a secondary event.
Procedural and periprocedural medications
After receiving written informed consent, all interventional procedures were performed according to the current standard guidelines. Decisions regarding glycoprotein IIb/IIIa inhibitors, low-molecular-weight heparin, stent type, pre-dilation, and intravascular ultrasound (IVUS) were at the operator's discretion. All patients were advised to take lifelong aspirin (100 mg daily) therapy. Thienopyridine therapies consisted of 12 months of clopidogrel (75 mg/day).
UPLMCA was defined as diameter stenosis ≥50% in the absence of a patent bypass conduit to the left system. Lesion variables, including isolated UPLMCA disease, UPLMCA with 1- to 3-vessel disease (VD), bifurcation or trifurcation lesions, CTO, severe calcification, severe tortuosity, thrombus, dominance, lesion number, and diseased vessel number, were classified according to the SYNTAX definitions (1). Lesion classifications in the LM, downstream lesions, restenosis in the LM, complete revascularization, stent number, stent length, and stent diameter were calculated on the basis of the criteria in the DISTAL study (3).
All deaths were considered cardiac unless a noncardiac cause was established clinically or at autopsy. MI was diagnosed according to the third universal definition (4). In-stent restenosis was defined as >50% diameter stenosis at follow-up; TVR was considered as repeat PCI or coronary artery bypass graft surgery (CABG) involving the index vessel. Stent thrombosis (ST) was classified according to the Academic Research Consortium definitions as definite, probable, or possible and as early (0 to 30 days), late (31 to 360 days), or very late (>360 days) (5). Angiographic success was defined as residual stenosis <10% with TIMI (Thrombolysis In Myocardial Infarction) flow grade 3 in both branches. Procedural success referred to angiographic success in the absence of in-hospital MACE.
Phone interviews were conducted monthly for 12 months and at 6-month intervals thereafter; clinical assessment was performed for suspected ischemic symptomatology. Clinical events were evaluated by an independent committee from 3 non-participating institutions; all interventionalists and committee members had free access to the database.
Quantitative coronary angiographic analyses
Repeat angiography at 6 months after the index procedure was scheduled unless clinically indicated earlier. Quantitative coronary angiographic analysis at baseline, post-stenting, and at follow-up were performed by the China Cardiovascular Research Foundation using edge-detection techniques (CAAS II, version 5.0, Pie Medical, Maastricht, the Netherlands). Angiographic measurements included the stented segment as well as margins 5 mm proximal and distal to the stent edge.
All scores were calculated by 3 technicians (Ms. Tian Xu, Ms. Jing Kan, and Ms. Ling Lin from Nanjing First Hospital, Nanjing Medical University) who were blinded to the study design. The mean score was used, t as the final score. The interobserver and intraobserver variability calculated by Kappa analysis was 6% and 11%, respectively.
Continuous variables were expressed as mean ± SD and were compared using the Student t test. Categorical variables were presented as counts and percentages, and were compared using the chi-square test or Fisher exact test. Survival curves were generated by the Kaplan-Meier method, and survival among groups was compared using the log-rank test. Cox proportional hazards models were used to assess risk factors for adverse events. Multivariate analysis, with all significant variables (defined as p < 0.05 for the association), was performed to adjust for possible confounders and identify independent predictors of adverse events. Sensitivity, specificity, false-positive, and false-negative, and positive and negative predictive values were calculated and analyzed by crosstable analysis after adjusting the model for the specific variables. Receiver-operating characteristic (ROC) curves for the NERS score II, the NERS score, and the SYNTAX scores were generated; the area under the curve (AUC) for each scoring method and endpoint was compared with the null hypothesis true area (0.5), and the 1 − specificity, sensitivity, 95% confidence interval (CI), and significance were determined. The value corresponding to the highest accuracy (i.e., minimal false-negative and false-positive rates) was chosen as the optimal cutoff value. To evaluate differences in the AUCs and standard errors from ROC curves between the 3 scoring systems using the Delong method (2), all variables, including cumulative MACE, ST, TVR for LM, MI, and cardiac death, were entered into Analyse-it (Analyse-it Software, version 3.0, Hearne Scientific Software, Melbourne, Australia). A p value <0.05 was considered statistically significant. All statistical tests were 2-tailed. Statistical analyses were performed using SPSS version 16.0 (SPSS, Chicago, Illinois).
Baseline clinical characteristics
In total, 366 patients (25.0%) had decreased LVEF (<40%), with New York Heart Association functional class III to IV in 191 (13.1%) patients. Renal dysfunction (defined as an estimated glomerular filtration rate [eGFR] <60 ml/min/1.73 m2) was observed in 139 (12.4%) patients (Table 2).
Baseline angiographic and procedural characteristics
Emergency LM stenting was required in 22 patients (6.5%). LM-CTO lesions were rare (2.0%); however, 23.0% had non–LM-CTO lesions (Table 3). Most UPLMCAs (77.1%) were distal; 85.4% of patients had 2-VD or 3-VD and required a greater number of stents. Staged procedures were performed in 79.6% of patients. IVUS-guided procedures were performed in 14.6% of patients. In total, 2-stent techniques were used in 479 (23.7%) patients. Overall, the LM stent diameter and length were 3.47 ± 0.42 mm and 30.09 ± 14.24 mm, respectively. Finally, TIMI flow grade 3 in the LM was achieved in 1,461 (99.9%) patients, and angiographic success for non-LM lesions was achieved in 1,373 (93.9%) patients.
Variables in the NERS II scoring system
The NERS score II system consisted of 7 clinical and 9 (5 for LM and 4 for non-LM) angiographic variables (Table 1).
Quantitative coronary angiography and clinical outcomes
An 8-month angiographic follow-up (204 ± 27 days post-procedure) was performed in 1,021 (69.8%) patients (Table 4). One-year follow-up data were available for all patients, and the cumulative rates of MI, TVR, and cardiac mortality were 2.2%, 8.3%, and 3.1%, respectively. The final cumulative MACE rate of 11.6% was driven primarily by subsequent TVR. The overall ST rate was 1.8% (definite 1.0%, probable 0.4%, and possible 0.4%).
Comparison of the NERS score II and the conventional NERS and SYNTAX scores
When in-hospital events were considered as state variables, the AUC by the NERS score II (Table 5) was significant for MI, MACE, and ST; the AUC by the NERS score was significant for MI and MACE; and by contrast, the AUC by the SYNTAX score was only significant for MACE.
At the end of follow-up, there were significant differences in the AUCs for MI, TVR, MACE, and ST between the NERS score II or the conventional NERS score and the SYNTAX score (Fig. 1). However, the AUC for cardiac death using the conventional NERS score was not significant compared with the SYNTAX score (0.75, 95% CI: 0.67 to 0.83 vs. 0.69, 95% CI: 0.60 to 0.71, respectively, p = 0.0543).
Comparisons of subgroups stratified by the NERS and SYNTAX scores
Per the ROC curve, a value of 19 stratified the scores as NERS II-lower (<19) and NERS II-higher (≥19). The Kaplan-Meier analysis demonstrated that in-hospital MACE-free survival differed significantly between the NERS II-lower (99.0%) and NERS II-higher subgroups (88.2%, p < 0.001) (Fig. 2), and between the NERS-lower (96.3%) and NERS-higher (90.4%, p = 0.001) subgroups. This difference was sustained throughout the entire follow-up period. By contrast, in-hospital MACE-free survival was similar between the SYNTAX-lower (96.7%) and SYNTAX-higher (93.3%, p = 0.085) subgroups. The cumulative MACE-free survival rate was significantly different between the SYNTAX-lower (99.1%) and SYNTAX-higher (87.3%, p = 0.039) subgroups.
A regression analysis demonstrated that a NERS score II ≥19 was the only independent predictor of cumulative MACE (hazard ratio: 3.27; 95% CI: 1.86 to 5.23; p ≤ 0.001) and ST (odds ratio: 22.15; 95% CI: 12.47 to 57.92; p ≤ 0.001) at follow-up.
The major findings of the present study are as follows: 1) after several baseline clinical or lesion variables and procedural variables were excluded from the scoring system, the simplified NERS score II had at least equivalent value to the conventional NERS score in predicting clinical outcomes for UPLMCA patients; and 2) the NERS score II and conventional NERS score were superior to the anatomic-based SYNTAX score in predicting clinical events, indicating that the clinical factors had a direct impact on determining clinical predictions.
Direct comparison of the NERS score II with the conventional NERS score
Our previous conventional NERS score (2) highlighted the important role of utilizing clinical, anatomic, and procedural variables in driving outcome prediction for UPLMCA patients after implantation of a coronary stent, a finding similar to that of the ACUITY (Acute Catheterization and Urgent Intervention Triage StrategY) trial (6). In that study, among 6 risk scores, scores incorporating clinical and angiographic variables (CCS [Canadian Cardiovascular Society] and NERS) showed the best tradeoff between discrimination and calibration for most endpoints, with the best discrimination for all endpoints and good calibration for most. However, the complexity of the conventional NERS score and the SYNTAX score limited their extensive use in everyday practice. Another disadvantage of the conventional NERS score was that it was generated from a relatively small UPLMCA patient population (n = 260) and externally validated in a small UPLMCA patient population (NERS study, n = 337), not in a larger UPLMCA patient population. An insight analysis into the NERS score demonstrated that several variables overlapped with others. For example, “multivessel disease” obviously covered “downstream lesions” and “thrombus-containing lesions,” and the variable of “positive biomarkers” was at least partially similar to “acute myocardial infarction (within 12 h).” More importantly, procedural factors (particularly “complex stenting techniques” and “IVUS guidance”) were variables that should have been excluded before the stenting procedures. Similarly, “whole-stem lesions” was actually the same as “distal nonbifurcation lesions.” As a result, with all variables from the NERS study and the DISTAL study retained in the new model, several new variables, such as “eGFR ≤60 ml/min” (thought to be more important than serum creatinine), were incorporated into a new regression model (with 597 patients) for the NERS score II. The latter was externally evaluated in an additional 1,463 UPLMCA patients. Here, an explanation why severe LM calcification (defined as the need of rotablation) was incorporated into NERS score II.
Notably, female sex was previously shown to have a negative impact on clinical outcomes after either CABG or PCI using DES (2,7–9), primarily because of increased comorbidity and unfavorable plaque characteristics (10–12). However, this “negative female phenomenon” was masked in the NERS score II system. A potential explanation for this discrepancy included, at least partially, the fact that advanced lesions in women overlapped with clinical (peripheral arterial disease, eGFR, diabetes, age, and LVEF) and angiographic variables. Interestingly, diabetes was a significant predictor, although it demonstrated lower scores for the conventional NERS score and the NERS score II. The complex interplay between clinical and angiographic variables may underlie these findings, similar to the other variables (such as hypertension). It is plausible that the NERS score II was not associated with the loss of the predictive value. Additionally, the predictive power of the conventional NERS score was maintained by the NERS score II and enhanced the prediction of in-hospital MACE, although the p value was only marginally significant.
Comparison with the anatomic-based SYNTAX score
Again, in line with our previous study, the present study demonstrated enhanced sensitivity and specificity for the prediction of cumulative MACE, ST, cardiogenic death, TVR, and MI for the conventional NERS score and the NERS score II. Two-level stratifications were used: all patients were divided into lower (SYNTAX <22, NERS <25, and NERS II <19) and higher (SYNTAX >22, NERS ≥25, and NERS II ≥19) risk groups. Similar to the previous study, the SYNTAX score did not predict in-hospital MACE, which had a rate of 96.7% in the SYNTAX-lower subgroup and 93.3% in the SYNTAX-higher subgroup (p = 0.085).
Features of the NERS score II
Two important procedural variables, that is, IVUS guidance and stenting techniques, included in the NERS score were excluded from the NERS score II. The importance of IVUS guidance during PCI has been addressed in several studies (13,14), but controversy still exists (15,16). Chen et al. (14) were the first to report that the IVUS-guided 2-stent technique was associated with significantly reduced late ST in patients with bifurcation lesions, with a resultant reduction in ST-segment elevation MI in 628 patients, which is similar to the findings reported by 2 previous studies (17,18) conducted by Korean teams. However, the impact of IVUS guidance in the present study was minimized, possibly because IVUS guidance was less frequently used (14.6%), and IVUS assessment was primarily performed for 2-stent strategies, which accounted for only 32.7% of patients. However, the predictive power of the NERS score II was similar to the conventional NERS score (incorporating IVUS guidance), indicating that the importance of the IVUS assessment could have been masked by lesion-specific factors and clinical parameters.
Furthermore, 2-stent strategies for UPLMCA patients have been associated with an increased rate of worse clinical events (2,3,19,20) and were also excluded from the NERS score II. Indeed, the 2-stent techniques for the treatment of overall bifurcation lesions (21) and UPLMCA distal bifurcation lesions (22) did not obtain the same clinical results. In our previous study (23), 2-stent techniques for UPLMCA bifurcation lesions had a higher rate of cumulative MACE when compared with a 1-stent technique, but double-kissing crush stenting was not inferior to the 1-stent strategy and was superior to other 2-stent techniques (21,23). Similarly, 2-stent strategies were used only in the first of 3 patients in the present study, which may account for the lack of power of stenting techniques in predicting clinical events.
To individualize decision making for treating LM stenosis on the basis of a score, a scoring system free of procedure-related variables is needed. Even though the authors stress that procedure-related variables have been removed in comparison to the NERS score, 1 anatomic, procedure-related variable is still included in the revised NERS score II (severe LM calcification, defined as needing rotablation). Rotablation is an important predictor of MACE from SYNTAX (1) and our previous study (2). According to the definition of rotablation from NERS score II, needing rotablation included that balloon failed to pass or to crack the angiography-defined non-severe calcification and that rotablation was used for angiography-defined severe calcification. Therefore, even though rotablation was only used in 1.2% of patients as in the current study, it indeed indicated the “real” calcified lesions in LM.
The complexity of the calculation of previous scoring systems limited their acceptance. The NERS score II contains only 16 variables and allows the realization of “real-time” scoring. Such convenient scoring is easily performed bedside and would be meaningful to drive the prediction of clinical outcomes. Most importantly, the predictive value of the conventional NERS score was maintained and further enhanced by the simplified NERS score II. With the use of this simplified scoring system, a calculator or smart phone software application could be developed for the NERS score II system.
The current study demonstrated that the composite MACE was acceptable (11.6%). However, even we did not report the baseline characteristics and clinical events in patients who received CABG, our results showed that patients in NERS score II-higher subgroups had significantly higher composite MACE (69.4%), mainly driven by target lesion revascularization/TVR. This result implied that CABG should be recommended for high-risk patients stratified by NERS score II, because the SYNTAX trial indicated that high-risk patients had increased need of revascularization. Furthermore, comparison of simple versus complex stenting techniques for LM bifurcation lesions would be a practical important issue in future studies.
First, its nonrandomized nature likely influenced the power of the results. Second, procedural variables reported to be related to the occurrence of worse clinical events were excluded from the NERS score II. However, our results showed the predictive value of the NERS score II was at least equivalent to the conventional NERS score (with procedural characteristics included). We might boldly postulate that the interaction between clinical and angiographic variables masks the importance of procedural indexes. Finally, patients who received CABG were not followed, and the intracenter MACE rate was not calculated as presented by the SYNTAX trial.
The NERS score II demonstrated improved predictive value for primary endpoints after UPLMCA-PCI compared with the SYNTAX score. Further, randomized studies are needed to elucidate the role of the NERS score II in the decision of PCI versus CABG for UPLMCA and complex anomalies in a more diverse population.
The authors thank all of the participants for their active and long-term cooperation, as well as Ling Lin, Hai-Mei Xu, Yin-Yin Zhang, and Jing Kan for their invaluable work in collecting the data.
Dr. Shao-liang Chen is a fellow at the Collaborative Innovation Center for Cardiovascular Disease Translational Medicine of Jiangsu Province, China. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- area under the curve
- coronary artery bypass graft surgery
- confidence interval
- chronic total occlusion
- drug-eluting stent(s)
- estimated glomerular filtration rate
- intravascular ultrasound
- left main coronary artery
- major adverse cardiac event(s)
- myocardial infarction
- percutaneous coronary intervention
- receiver-operating characteristic
- stent thrombosis
- target vessel revascularization
- unprotected left main coronary artery
- vessel disease
- Received May 2, 2013.
- Revision received July 1, 2013.
- Accepted August 1, 2013.
- American College of Cardiology Foundation
- Chen S.L.,
- Chen J.P.,
- Mintz G.,
- et al.
- Chen S.L.,
- Ye F.,
- Zhang J.J.,
- et al.
- Thygesen K.,
- Alpert J.S.,
- Jaffe A.S.,
- et al.
- Palmerini T.,
- Caixeta A.,
- Genera P.,
- et al.
- Chieffo A.,
- Meliga E.,
- Latib A.,
- et al.
- Morice M.C.,
- Serruys P.W.,
- Kappetein A.P.,
- et al.
- Weintraub W.S.,
- Wenger N.K.,
- Kosinski A.S.,
- et al.
- Palmerini T.,
- Sangiorgi D.,
- Marzocchi A.,
- et al.
- Chen S.L.,
- Santoso T.,
- Zhang J.J.,
- et al.
- Chen S.L.,
- Xu B.,
- Han Y.L.,
- et al.