J Geriatr Cardiol 2018; 15: 559?566. doi:10./

1 Introduction

In a progressively aging population, elderly patients increasingly account for a relevant number of cardiovascular patients.[1,2] Coronary artery disease (CAD) is one of the main causes of death in the elderly, especially in patients aged ≥ 75 years.[1,3–6]Unfortunately, despite an increase in referrals for percutaneous coronary intervention (PCI), the elderly are often excluded from randomized clinical trials. This leads to scarce insights into interventional outcomes and predictors of mortality in this vulnerable population.[1]

The complexity of coronary lesions is an independent predictor of mortality in younger cohorts,[7–11] but it is un-clear if this aspect is also predictive of mortality in elderly patients who often present with more complex coronary lesions[3] and various comorbidities.[12] The Synergy be-tween Percutaneous Coronary Intervention with TAXUS and Cardiac Surgery (SYNTAX)-score is an angiographic tool grading the complexity of coronary artery disease. It was originally developed for the SYNTAX trial where it proved to be an independent predictor of long-term major adverse cardiac and cerebrovascular events (MACCE) in patients undergoing PCI.[13,14] Numerous subsequent studies have validated the SYNTAX-score and confirmed its pre-dictive value for outcomes after PCI in various clinical set-tings.[15] In stable CAD patients with left main or three- vessel disease, the SYNTAX-score is currently recom-mended as a decision tool to guide revascularization strat-egy with either coronary artery bypass grafting (CABG) or PCI.[16–18] Of note, although there was no upper age limit for inclusion in the SYNTAX trial, the mean age of patients was only 65 ± 10 years.[14]

Due to the fact that the risk-to-benefit-ratio of PCI is less clear in elderly patients, interventional cardiologists are frequently careful in undertaking PCI in this high-risk population. Therefore, it is of high relevance to investigate what impacts the outcome after PCI in elderly patients to better adapt therapeutic strategies in this cohort. The aim of this registry study was to investigate whether age modifies the impact of the SYNTAX-score on one-year and two-year mortality.

2 Methods

2.1 Study population

Data from all consecutive patients undergoing PCI between January 2013 and March 2014 at the Department of Cardiology of the Ludwig-Maximilians University, a tertiary referral center in Munich, Germany, were collected in a registry. Only 145 patients with prior CABG were excluded. Follow-up was performed by telephone calls or structured follow-up letters.

2.2 SYNTAX-score

For each patient, the SYNTAX-score was calculated by means of the online SYNTAX-score calculator (available at included three general variables (dominance, the total number of lesions and vessel segments involved per lesion as well as the presence of diffuse/small vessel disease) and eight anatomical variables (length of stenosis, involvement of bifurcations or trifurcations, aortic ostial localization, chronic occlusion, vessel tortuosity, calcification and thrombus formation) in each lesion with ≥ 50% luminal obstruction in vessels ≥ 1.5 mm. Finally, the score of each lesion was added to obtain the patient’s SYNTAX-score.[13,14] The SYNTAX-score was calculated by two experienced cardiologists blinded to cli-nical data. Discrepancies were resolved by consensus. The limitations of the SYNTAX-score regarding ST-segment elevation myocardial infarction (STEMI) patients are well known. In these cases, we followed previously used methods, scoring an occluded infarct-related artery as an occluded artery of < 3 months duration.[20-22]

2.3 Clinical endpoints

Outcomes of interest were all-cause and cardiac mortality at one-year and two-year after PCI. In deceased patients source documents were solicited for verification of the event.

2.4 Statistical methods

The aim of this study was to assess whether age modifies the impact of the SYNTAX-score on one-year and two-year mortality. Classification of patient ≥ 75 years as elderly was chosen in line with previous publications.[23–25] We assessed the interaction of age and SYNTAX-score on one-year and two-year mortality by entering the interaction term into the respective Cox proportional hazards model. We also assessed the capacity of the SYNTAX-score to appropriately stratify the mortality risk at one year and two years after PCI in elderly and younger patients. Therefore, patients ≥ 75 years and < 75 years were classified according to their tertile distribution of the SYNTAX-score into low, intermediate and high SYNTAX-score. Mortality in these groups was assessed using the Kaplan-Meier method and compared using the log-rank test. Multivariable cox regression analysis was used to assess independent correlates of mortality. Variables that differed with P-value < 0.1 in the univariable analysis were entered into the multivariable model. Normality was tested by means of the Kolmogorov-Smirnov test. Continuous data are expressed as mean ± SD or median [interquartile ranges (IQR)] and compared with the unpaired Student’s t-test or Wilcoxon test, respectively. Categorical data are expressed as numbers and percentages and compared with the chi-squared test, or Fisher’s Exact test in case of cell values < 5. A two-side P-value < 0.05 was considered to indicate statistical significance. Statistical analyses were performed with S-PLUS version 4.5 (Insightful Corporation).