Open access Coronary artery disease Benefit of primary percutaneous coronary interventions in the elderly with ST segment elevation myocardial infarction Daniel Fernández-­Bergés ‍ ‍ ,1,2 Irene R Degano ‍ ‍ ,3 Reyes Gonzalez Fernandez,4 Isaac Subirana,3,5 Joan Vila,3,5 Manuel Jiménez-­Navarro,6 Silvia Perez-­Fernandez,7,8 Mercé Roqué,9 Antoni Bayes-­Genis,10 Francisco Fernandez-­Aviles,11 Antonio Mayorga,12 Vicente Bertomeu-­Gonzalez,13 Juan Sanchis,14 Marcos Rodríguez Esteban,15 Antonio Sanchez-­Hidalgo,16 Esther Sanchez-­Insa,17 Ane Elorriaga,18 Emad Abu Assi,19 Alberto Nuñez,20 Jose Manuel Garcia Ruiz,21 Pedro Morrondo Valdeolmillos,22 Daniel Bosch-­Portell,23 Iñaki Lekuona,24 Andres Carrillo-­Lopez,25 Alberto Zamora,26 Berta Vega-­Hernandez,27 Javier Alameda Serrano,28 Catalina Rubert,29 Luis Ruiz-­Valdepeñas,30 Laura Quintas,31 Luis Rodríguez-­Padial ‍ ‍ ,32 Jessica Vaquero,33 Luis Martinez Dolz,34 Jose A Barrabes,35 Pedro L Sanchez,36 Alessandro Sionis,37 Julio Martí-Almor,38 Roberto Elosua,3,7,39 Rosa-­María Lidon,40 David Garcia-­Dorado,40 Jaume Marrugat,3,7 on behalf of the ATHOS investigators To cite: Fernández-­Bergés D, Abstract Degano IR, Key questions Objective Primary percutaneous coronary intervention Gonzalez Fernandez R, et al. (P-­PCI) has demonstrated its efficacy in patients with Benefit of primary percutaneous What is already known about this subject? ST segment elevation myocardial infarction (STEMI). coronary interventions in the ►► Primary percutaneous coronary intervention (P-­ elderly with ST segment However, patients with STEMI ≥75 years receive less P-­ PCI) significantly decreases mortality of patients elevation myocardial infarction. PCI than younger patients despite their higher in-­hospital presenting with ST elevation myocardial infarction Open Heart 2020;7:e001169. morbimortality. The objective of this analysis was to (STEMI) and constitutes a type I A indication in the doi:10.1136/ determine the effectiveness of P-­PCI in patients with international clinical guidelines for the treatment of openhrt-2019-001169 STEMI ≥75 years. acute coronary syndrome (ACS). Methods We included 979 patients with STEMI ≥75 ►► Despite this recommendation, patients aged years, from the ATención HOspitalaria del Síndrome Received 16 September 2019 ≥75 years receives less PCI-­P to the high preva- coronario study, a registry of 8142 consecutive patients Revised 21 April 2020 lence of comorbidities (eg, like chronic renal failure Accepted 2 June 2020 with acute coronary syndrome admitted at 31 Spanish or chronic obstructive pulmonary disease) and in- hospitals in 2014–2016. We calculated a propensity score creased risk of bleeding and other complications. (PS) for the indication of P-­PCI. Patients that received or not P-­PCI were matched by PS. Using logistic regression, What does this study add? we compared the effectiveness of performing P-­PCI versus ►► Our study points out the significant decrease of ad- non-­performance for the composite primary event, which verse events during the hospitalisation of patients included death, reinfarction, acute pulmonary oedema or aged ≥75 years with STEMI when undergoing P-­PCI cardiogenic shock during hospitalisation. in spite of their comorbidities. Results Of the included patients, 81.5 % received P-­PCI. The matching provided two groups of 169 patients with How might this impact on clinical practice? and without P-­PCI. Compared with its non-­performance, ►► Our results should extend the indication of PCI to pa- P-­PCI presented a composite event OR adjusted by PS of tients aged ≥75 years to reduce their mortality when 0.55 (95% CI 0.34 to 0.89). presenting a STEMI. Conclusions Receiving a P-­PCI was significantly associated with a reduced risk of major intrahospital © Author(s) (or their complications in patients with STEMI aged 75 years or heart disease is decreasing, the number of employer(s)) 2020. Re-­use older. permitted under CC BY. events is increasing due to the ageing of the Published by BMJ. population,1 with an expected increase in For numbered affiliations see acute coronary syndrome (ACS) in the popu- end of article. Introduction lation aged 75 years or older.2 The National Coronary heart disease is the leading indi- Institute of Statistics estimates the Spanish Correspondence to Dr Daniel Fernández-­Bergés; ​ vidual cause of death in Spain as in all population of ≥75 years in 2012 in more than polonibo@​gmail.​com Europe. Although mortality from coronary 4 million (9%) and expects that this figure Fernández-­Bergés D, et al. Open Heart 2020;7:e001169. doi:10.1136/openhrt-2019-001169 1 Open Heart will increase—if the estimated demographic growth is Study variables maintained—up to 23% of the total population in 2052.3 Age, sex, tobacco consumption, history of diabetes, At present, different series place the prevalence of ACS hypertension, cardiovascular disease (myocardial infarc- in this population above 30%, exceeding 40% in some tion, angina, heart failure, stroke, PCI and coronary national series.4 artery bypass surgery) and related comorbidities (chronic The treatment of ST segment elevation myocardial obstructive pulmonary disease and chronic renal failure infarction (STEMI) includes performing a percuta- (CRF)) were recorded. Glycaemia and creatinine were neous coronary intervention (PCI) as soon as possible collected in the admission analytical test. From the creati- called primary PCI (P-­ PCI) with indication I-­A (I = nine value, the estimated glomerular filtration rate (GFR) evidence and /or general agreement that a given treat- was calculated in mL/min/1.73 m2 using the formula ment or procedure is beneficial, useful, effective. A = Modification of Diet in Renal disease. data derived from multiple randomized clinical trial or During hospitalisation, coronary angiography and large non-­randomized studies).5 If P-­PCI is not available P-­PCI were recorded, as well as the prescription of medi- within 120 min from the first medical contact, fibrinolysis cation (antiplatelet agents, heparin, ACE inhibitors, beta-­ should be performed, except for contraindications, with blockers and diuretics). The onset of heart failure (acute subsequent angiographic study and, if possible, elective pulmonary oedema and cardiogenic shock) was recorded PCI (I-­A). According to the most recent guide, there is at admission and during hospitalisation by the Killip and no upper age limit for the recommendation of PCI, and Kimball classification.9 As major bleeding complications, therefore, all patients who meet the criteria regardless of the fall of >3 g of haemoglobin, the need for transfusion their age should receive it.5 However, the utility of P-­PCI or surgery and intracranial bleeding were considered. in patients with STEMI ≥75 years is discussed, and this population receives fewer reperfusion treatments, both Events of interest invasive and pharmacological.6 The lower reperfusion of The primary event consisted of a composite variable that elderly patients is probably due to the presence of atypical included death, reinfarction, acute pulmonary oedema symptoms and to the delay in diagnosis, to their frequent or cardiogenic shock during hospitalisation. comorbidity and to the higher rate of complications they may present (contrast nephropathy, major bleedings and Statistical analysis so on).5–7 For the description of the continuous variables, the mean The lower invasive treatment in patients with STEMI and SD were used when the variable followed a normal ≥75 years is also due to the fact that randomised clinical distribution and the median and IQR were used other- trials usually do not include older patients, and if they wise. Categorical variables were described as percentages. do, they are highly selected. Therefore, the claim of the Comparisons of continuous variables between groups 2007 American Heart Association 7 referring to the need according to the performance of P-­PCI were performed for clinical studies that include patients over 65 years has using the Student’s t-­test or the Mann-­Whitney test if the not lost its validity. distribution was not normal. Categorical variables were Our objective in the present study was to analyse compared between the same groups using the χ2 test. whether the performance of a P-­PCI reduces complica- To evaluate the effect of P-­PCI on the primary endpoint tions (death, reinfarction, acute pulmonary oedema or (death, reinfarction or Killip III-­IV during hospitalisa- cardiogenic shock) during hospitalisation in patients tion), a logistic regression model was used in patients with STEMI ≥75 years compared with patients with the matched by propensity score (PS) to receive P-­PCI. A same characteristics to which P-­PCI was not performed. PS was calculated for P-­PCI using a logistic regression model. For the PS model, all variables associated with performing P-­PCI were included as independent vari- Methods ables and P-­PCI as a dependent variable. Subsequently, Design patients with and without P-­PCI were paired from the A prospective, consecutive, multicentric and national logit-­PS with a tolerance of 0.2 times their SD.10 After the ACS registry was designed: the ATención HOspitalaria pairing of the patients, the standardised differences of del Síndrome coronario (ATHOS) registry.8 This registry the variables between paired patients with and without included 8142 patients diagnosed with ACS admitted P-­PCI were analysed. It was assumed that the variables consecutively in 31 Spanish hospitals during 2014–2016. were balanced if the standardised differences were <10.11 For the present study, patients from the ATHOS registry Finally, we included the paired patients in a logistic who had an STEMI on admission and who were ≥75 years regression model, in which the dependent variable was old (n=979 patients) were selected. the event of interest, and the independent variables were the P-­PCI performance and the variables that were not Patient and public involvement well balanced in the process of matching by PS. Patients or the public were not involved in the design, or All analyses were performed with the statistical conduct, or reporting, or dissemination of our research programme R (V.3.2.3) . 2 Fernández-­Bergés D, et al. Open Heart 2020;7:e001169. doi:10.1136/openhrt-2019-001169 Coronary artery disease Results The elderly population that is part of ATHOS registry8 Table 1 showed the characteristics and evolution in has an average age of 83.1 (5.9) years old with an admission of the 979 patients aged ≥75 years with STEMI important index of comorbidities, constituting a repre- included in the analysis. P-­PCI was performed in 798 sentative sample of the elderly patient of a European (81.5%) of these patients. We observed that compared country. After an adequate characterisation and the with patients who did not receive a P-­PCI, the patients use of a validated statistical procedure for observational who did receive it had significantly younger age, higher studies,12 we have seen that P-­PCI in elderly patients with prevalence of tobacco consumption and lower preva- STEMI reduces the risk of presenting complications lence of angina, heart failure, stroke and previous CRF. during hospital admission. On admission, patients who received P-­PCI had a lower Patients aged ≥75 years old with non-­ STEMI have degree of heart failure and a lower prevalence of glomer- received special attention in recent years,13 14 but more ular filtration rate <60 mL/min/1.73 m2. rare, and in some cases inconclusive, are the studies in In patients who received P-­PCI, the indication of medi- patients with STEMI, particularly assessing the use of cation of proven efficacy in the reduction of cardiovas- P-­PCI.15 If we take into account the important Spanish cular events such as beta-­ blockers, ACE inhibitors or multicentre studies published in the first decade of double antiplatelet treatment were significantly more this century, only the TRIANA16 clinical trial and the used. Patients with P-­PCI received more heparin in acute ESTROFA17 registry included patients aged >75 years phase but less diuretics compared with patients without old who presented with STEMI and in whom P-­PCI was P-­PCI. There were no significant differences in the inci- performed. TRIANA showed a trend towards better dence of major bleedings between both groups. There survival when P-­PCI was used compared with fibrino- were statistically significant differences between both lysis, and ESTROFA described that elderly STEMI groups with respect to the presentation of the combined patients had a high prevalence of renal failure and event during hospitalisation. multivessel disease. Age, presence of heart failure on admission, GFR The CASTUO single-­ centre observational study,4 <60 mL/min at admission, previous angina, previous which included patients with acute myocardial infarc- heart failure and previous stroke, all of them with statisti- tion, found improvement in survival in patients aged cally significant differences, were included in the estima- >75 when PCI was indicated during admission, both tion of the PS. The area under curve of the PS model was in the admission phase and during follow-­u p. The 0.68 (95% CI 0.64 to 0.73). rest of the multicentre Spanish registries that include In the PS model, two groups of 169 patients could patients with STEMI did not study in particular the be included and compared. Table 2 showed the vari- subgroup of elderly patients or the indication of ables included in the PS in paired patients as well as P-­P CI.18 19 Of the international multicentre studies the standardised differences of the variables before that include both STEACS and NSTEACS, ACACIA,20 and after matching. There were no significant differ- AMIS Plus21 and MINAP22 showed a reduction in ences in these variables between the two groups. The mortality associated with invasive treatment in elderly standardised differences of the variables included in patients with ACS both with and without ST segment the PS decreased when matching the patients. Only elevation, although they did not analyse the P-­P CI the variables previous angina and glomerular filtration indication. <60 mL/min at admission were not correctly balanced The impact and use of PCI in STEMI, between 2003and 2012 in Spain, has recently been evaluated by since they presented a standardised difference slightly Cequier et al.23 In this study of 302 471 patients, 116 621 above 10 of the corresponding SD. These two variables received PCI (38.6%), 46 720 fibrinolysis (15.4%) and were included in the final model to predict the risk of 139 130 had no indication of reperfusion ​​(46%). The the event of interest. mean age of the patients who received PCI, fibrinolysis, The final logistic regression model (table 3) showed or neither was 63.4, 63.7 and 71.8 years old, respectively. that patients who had received a P-­PCI compared with This shows a significant gradient in the age of PCI indi- those who had not received it, had a lower risk of dying or cation that penalises the elderly patients given that presenting with reinfarction, acute pulmonary oedema the mortality for the PCI group was 4.8% versus 17.3% or cardiogenic shock during the hospitalisation (OR for the group without any reperfusion therapy. The 0.55, 95% CI 0.34 to 0.89). number of elderly persons is rapidly growing in Europe, and their proportion in the population is expected to increase. Elderly STEMI patients are typically treated Discussion less aggressively than younger patients because they Our study shows that the use of a P-­PCI in a population have more comorbidities and risk of bleeding.5–7 For of 75 years or older with a STEMI significantly reduces the same reasons they are also under-­represented in complications during hospitalisation, including clinical trials. mortality, reinfarction, pulmonary oedema and cardio- This last aspect is particularly important because it genic shock. is one of the most important causes when it comes to Fernández-­Bergés D, et al. Open Heart 2020;7:e001169. doi:10.1136/openhrt-2019-001169 3 Open Heart Table 1 Characteristics of patients aged 75 years and older Total n=979 No P-­PCI n=181 P-­PCI n=798 P value Age 83.1 (6.01) 85.0 (6.20) 82.7 (5.89) <0.001 Sex women (%) 39.8 46.4 38.3 0.055 Killip III-­IV admission (%) 15.3 27.6 12.5 <0.001 GFR <60 mL/min (%) 42.4 52.3 40.1 0.004 Glucose <126 mg/dL (%) 65.4 68.2 64.7 0.432 Diabetes mellitus (%) 0.668  No 65.7 68.0 65.2  Yes, diet control 3.47 2.76 3.63  Yes, OADs control 21.5 18.8 22.1  Yes, insulin control 8.89 10.5 8.52  Yes, diabetes in hospitalisation 0.51 0.00 0.63 Smoker (%) <0.004  Never 64.5 75.1 62.0  Actual 11.8 8.29 12.7  Ex-­smoker >30 days 23.7 16.6 25.3 Arterial hypertension 73.0 75.1 72.6 0.539 Previous MI 14.6 14.4 14.7 1.000 Previous angina 11.6 16.6 10.5 0.031 Previous CHF 4.60 10.5 3.26 <0.001 Previous stroke 9.60 13.8 8.65 0.047 Previous PCI 11.0 7.73 11.8 0.151 Previous CABG 2.15 2.76 2.01 0.568 COPD 13.6 14.9 13.3 0.646 CRF 13.8 19.9 12.4 0.012 Antiplatelet (%) <0.001  None 4.19 8.29 3.26  One antiplatelet* 5.52 16.6 3.01  Combination† 90.3 75.1 93.7 Heparin hospitalisation (%) <0.001  None 17.5 34.8 13.5  UFH 41.5 12.2 48.1 LMWH 29.0 48.6 24.6  UFH+LMWH 12.1 4.42 13.8 ACE inhibitors (%) 57.8 47.5 60.2 0.002 Beta-­blockers (%) 55.5 48.6 57.0 0.049 Diuretics (%) 39.3 51.4 36.6 <0.001 Coronariography (%) <0.001  No 10.4 56.4 0.00  Yes 89.1 42.5 99.6 Major bleeding 3.16 2.90 4.42 0.405 Exitus, reinfartcion or Killip III-­IV during hospitalisation 22.7 40.9 18.5 <0.001 Results with and without P-­PCI. *Only aspirin or ticlopidine or clopidogrel or others. †Two antiplatelet drugs. CABG, coronary artery bypass grafting; CHF, chronic heart failure; COPD, chronic obstructive pulmonary disease; CRF, chronic renal failure; GFR, glomerular filtration rate; LMWH, low molecular weight heparin; MI, myocardial Infarction; OADs, oral antidiabetic’s drugs; PCI, percutaneous coronary intervention; P-­PCI, primary percutaneous coronary intervention; UFH, unfractioned heparin. 4 Fernández-­Bergés D, et al. Open Heart 2020;7:e001169. doi:10.1136/openhrt-2019-001169 Coronary artery disease antithrombotic drugs and choosing a radial approach Table 2 Characteristics of patients matched by propensity score and standardised differences in propensity score in the PCI, which results in a lower risk of subsequent variables before and after matched bleeding.26 The exacerbation of CRF by contrast mate- rial and its way of preventing it continues to be a cause Matched n=169 n=169 P value for concern and an unresolved discussion.27 Recently, Age 85.0 (79.0; 88.0) 84.0 (79.0; 88.0) 0.507 the EUROTRACS study28 showed a reduction in Killip III-­IV admission 42 (24.9%) 39 (23.1%) 0.799 mortality during hospital admission in elderly patients GFR <60 mL/min 86 (50.9%) 77 (45.6%) 0.384 with ACS undergoing PCI who presented with CRF or Previous angina 142 (84.0%) 148 (87.6%) 0.436 diabetes. These results, together with the rest of the Previous CHF 153 (90.5%) 155 (91.7%) 0.848 evidence, seem to indicate that, although a careful anal- Previous stroke 148 (87.6%) 147 (87.0%) 1.000 ysis of the life situation of the elderly patient should be made, the usual presence of comorbidities should Standardised Before not be assumed as a contraindication. An adequate clin- differences* matching After matching ical assessment may be sufficient to avoid depriving a Age 36.85 5.89 patient of the benefit of a proven treatment. Killip III-­IV admission 38.29 4.15 The success of P-­ PCI has been so remarkable that GFR <60 mL/min 24.55 10.64 recommendations have been extended to improve the Previous angina 17.71 10.15 application of this treatment, which is closely linked to the time of completion.29 In Europe, the Stent for Life Previous CHF 28.84 4.15 Initiative30 has become a very active platform supporting Previous stroke 16.38 1.77 the indications and the use of clinical guidelines in the *Differences between groups based on the % of the SD of each management of myocardial infarction, defining actions individual variable. that improve compliance and helping to identify barriers CHF, chronic heart failure; GFR, glomerular filtration rate. that may arise. In Spain, the use of reperfusion strate- gies has been improving year after year and, with this, justifying the lower indication of catheterisation and PCI mortality due to STEMI has decreased, and the incor- in elderly patients. poration of this European initiative has significantly This scenario constitutes a therapeutic challenge in the increased the PCI rate in general and especially P-­PCI.23 management of coronary artery disease elderly patients, a If we take into account that age is one of the variables major concern for cardiac interventionists and surgeons. that best predict mortality in ACS and that P-­PCI is the Our study shows the feasibility and benefits of intensive most effective intervention to avoid it, it is clear to what management with current clinical guidelines recommen- extent it is important to establish its use and usefulness in dations of this population, despite their high comorbidity the patients aged ≥75 years old who present with STEMI. prevalence (eg, chronic kidney failure, COPD, diabetes We must also take into account the use of fibrinolysis as a mellitus and previous coronary heart disease or stroke). valid alternative when times or distance prevent the real- Among these comorbidities, the most prominent are isation of a P-­PCI, although there are greater bleeding usually the existence of diabetes, anaemia and CRF. risks in this population.31 32 There is a potential risk of exacerbation of CRF when Among the reasons that have traditionally been put the patient is submitted to a procedure that includes forward to exclude those over 75 years of age and that iodinated contrast. In addition, double antiplatelet have already been exposed, there is one,33 which we have therapy in the elderly diabetic patient and with CRF not been able to explore, referring to the late arrival at increases the risk of bleeding after PCI.24 These risks the hospital of the elderly patient ‘outside the optimal may explain why the indication for PCI is lower in window’ for perform a P-­PCI. The European Society of patients aged >75 years old due to fear of complications. Cardiology guidelines5 established that ‘a routine primary Savonitto et al25 analyse the importance of emphasising PCI strategy should be considered in patients presenting CRF and anaemia by carefully adjusting the dose of late (12–48 hours) after symptom onset’. This strategy should certainly be explored in depth. In short, the elderly population with STEMI, due to Table 3 Risk of death, reinfarction, acute pulmonary the increase in prevalence, the benefits of P-­PCI and the oedema or cardiogenic shock during hospitalisation in increased risk of bleeding secondary to the use of fibrino- patients with STEMI aged ≥75 years undergoing P-­PCI lytics, constitutes a group of special interest to be treated OR 95% CI P value with invasive procedures. The European Society of Cardiology Guidelines5 P-­PCI versus not done 0.552 0.342 to 0.892 0.015 states that ‘There is no upper age limit with respect to reper- GFR <60 mL 2.971 1.832 to 4.818 0.001 fusion, especially with primary PCI’ but is unusual that Previous angina 1.610 0.842 to 3.081 0.150 the decision is based only in one study16 that could not GFR, glomerular filtration rate; P-­PCI, primary percutaneous complete the estimated sample size due to the diffi- coronary intervention. culty of including patients older than 75 years. This Fernández-­Bergés D, et al. Open Heart 2020;7:e001169. doi:10.1136/openhrt-2019-001169 5 Open Heart 22 situation, in addition to the difficulty of having suffi- Medicina Intensiva, Hospital Universitario Donostia, San Sebastián, San Sebastian, ciently statistically powered randomised clinical trials, Spain 23 Department of Cardiology, Hospital Universitari de Girona Doctor Josep Trueta, shows that doctors themselves have an inclusion bias, Girona, Spain finding intangible reasons to excluding elderly patients 24 Department of Cardiology, Hospital Galdakao-­Usansolo, Galdacano, Spain in P-­PCI prospective protocols. 25 Hospital Universitari Son Espases, Palma de Mallorca, Spain Because of this, we believe that it is necessary to rein- 26 Hospital Comarcal de Blanes, Blanes, Spain force in daily clinical practice—that the person suffering 27 Hospital Universitario de Cabueñes, Gijon, Spain 28 from STEMI should always be considered to perform Cardiología, Hospital Clínico Universitario Lozano Blesa, Zaragoza, Spain 29 a P-­PCI and balance the cost–benefit equation without Hospital Son Llàtzer, Palma de Mallorca, Spain 30 Hospital General Universitario de Ciudad Real, Ciudad Real, Spain excluding anyone by age. It is likely that in the immediate 31 Hospital de Mendaro, Mendaro, Spain future we must refer to the elderly population as those 32 Complejo Hospitalario de Toledo, Toledo, Spain over 85 years of age and include nonagenarians and even 33 Hospital Universitario Araba sede Txagorritxu, Vitoria-­Gasteiz, Spain older. 34 Hospital Universitario La Fe, Valencia, Spain 35 The results of our study would support the indication of Department of Cardiology, University Hospital Vall d'Hebron, Barcelona, Spain 36 P-­PCI as the treatment of choice in patients with STEMI Department of Cardiology, Hospital Universitario Salamanca, Salamanca, Spain 37 ≥75 years old. Intensive Cardiac Care Unit, Cardiology Department, Hospital de la Santa Creu i Sant Pau, Biomedical Research Institute IIB-­Sant Pau, Universitat de Barcelona, Barcelona, Spain Limitations 38 Department of Medicine, Department of Cardiology, Hospital del Mar. Universitat The present study has the limitations of observational Autónoma de Barcelona, Barcelona, Spain 39 studies that explore the effect of the exposure to treat- Universitat de Vic, Barcelona, Spain 40 ments (eg, confounding, selection bias and potential Hospital Universitari Vall d' Hebron, Barcelona, Spain reverse causation), which have been overcome by PS. The Twitter Esther Sanchez-­Insa @Estherzukis information brought up in our study is otherwise diffi- cult to obtain in elderly patients because they tend to be Acknowledgements The authors would like to thank the ATención HOspitalaria del Síndrome coronario (ATHOS) study investigators and collaborators (​www.​ excluded from clinical trials. redheracles.​net/​athos_​inv). The relatively small sample size is another limitation. Contributors DF-­B: contributed to acquisiton, analysis and interpretation of the The difficulty of recruiting elderly patients is commonly data for the work; drafted the manuscript; gave final approval; and agrees to referred to in the literature. be accountable for all aspects of work in ensuring integrity and accuracy. IRD: contributed to conception and interpretation of the data for the work; critically revised the manuscript; gave final approval; and agrees to be accountable Conclusions for all aspects of work in ensuring integrity and accuracy. MR: contributed to The performance of P-­PCI is significantly associated with interpretation of the data for the work; drafted the manuscript; gave final approval; and agrees to be accountable for all aspects of work in ensuring integrity and a lower risk of major complications in patients ≥75 years accuracy. IS: contributed to design, analysis and interpretation of the data for the old admitted for a STEMI. work; drafted the manuscript; gave final approval; and agrees to be accountable for all aspects of work in ensuring integrity and accuracy. JV: contributed to Author affiliations conception; critically revised the manuscript; gave final approval; and agrees to 1 Unidad de Investigación, Hospital Don Benito-­Villanueva, Don Benito, Spain be accountable for all aspects of work in ensuring integrity and accuracy. SP-­F: 2 Instituto Universitario de Investigación Biosanitaria de Extremadura, Badajoz, Spain contributed to analysis of the data for the work; critically revised the manuscript; 3 Institut Hospital del Mar d' Investigacions Mediques, Barcelona, Spain gave final approval; and agrees to be accountable for all aspects of work in 4 Department of Cardiology, Infanta Cristina University Hospital, Badajoz, Spain ensuring integrity and accuracy. MJ-­N, MR, AB-­G, AM, VBG, JS, MR, AS-­H, ES-­I, 5 CIBERESP de Epidemiologia y Salud Publica, Barcelona, Spain AE, EAA, AN, PMV, DB-­P, IL, AC-­L, AZ, BV-­H, JAS, CR, LR-­V, LQ, LR-­P, JV, LMD, PLS, 6 Cardiology Department, Virgen de la Victoria University Hospital, Málaga, Spain AS, JM-­A and R-­ML: contributed to acquisiton of the data for the work; drafted 7 CIBER Enfermedades Cardiovasculares (CIBERCV), Barcelona, Spain the manuscript; gave final approval; and agrees to be accountable for all aspects 8 Institut Hospital del Mar d'Investigacions Mediques, Barcelona, Spain of work in ensuring integrity and accuracy. FF-­A, RE and DG-­D: contributed to 9 conception and design of the work; critically revised the manuscript; gave final Hospital Clinic, Barcelona, Spain 10 approval; and agrees to be accountable for all aspects of work in ensuring integrity Department of Cardiology, Hospital Universitari Germans Trias i Pujol, Badalona, and accuracy. JMGR: contributed to acquisiton of the data for the work; drafted Spain 11 the manuscipt; gave final approval; and agrees to be accountable for all aspects of Servicio de Cardiologia, University Hospital Gregorio Marañon, Madrid, Spain 12 work in ensuring integrity and accuracy. JAB: contributed to acquisiton of the data Department of Cardiology, Hospital Universitario Gregorio Marañón, Madrid, for the work; gave final approval; and agrees to be accountable for all aspects of Madrid, Spain work in ensuring integrity and accuracy. JM: contributed to conception and design 13 Department of Cardiology, Hospital Universitario de San Juan, Alicante, Spain of the work, analysis, interpretation of the data for the work; critically revised the 14 Cardiology Department, Hospital Clinico Universitario, INCLIVA, Universitat de manuscript; gave final approval; and agrees to be accountable for all aspects of València, Valencia, Spain work in ensuring integrity and accuracy. 15 Department of Cardiology, Hospital Universitario Nuestra Senora de la Candelaria, Funding Supported by: MARATO TV3 (081630), de AGAUR (2014SGR240); del Santa Cruz de Tenerife, Spain 16 Instituto de Salud Carlos III: Red de Investigación Cardiovascular RD12/0042 Consorci Sanitari de Terrassa, Terrassa, Spain 17 (Programa HERACLES); Red RedIAPP RD06/0018; CP12/03287; CIBER Department of Cardiology, Hospital Universitario Miguel Servet, Zaragoza, Spain 18 Epidemiología y Salud Pública; CIBERCV de enfermedades Cardiovasculares, Department of Cardiology, Hospital Universitario de Basurto, Bilbao, Bilbao, Spain 19 Fondo Europeo de Desarrollo Regional (FEDER) (European Regional Development Department of Cardiology, Hospital Álvaro Junqueiro de Vigo, Pontevedra, Vigo, Funds -ERDF-); FIS CP12/03287, FIS 14/00449, FIS PI081327, FIS INTRASALUD Spain PI1101801. 20 Department of Cardiology, Hospital Universitario Fundación de Alcorcón, Madrid, Spain Competing interests None declared. 21 Hospital Universitario Central de Asturias, Oviedo, Spain 6 Fernández-­Bergés D, et al. Open Heart 2020;7:e001169. doi:10.1136/openhrt-2019-001169 Coronary artery disease Patient consent for publication Not required. 16 Bueno H, Betriu A, Heras M, et al. Primary angioplasty vs. fibrinolysis in very old patients with acute myocardial infarction: Ethics approval The study was approved by the clinical research ethics committee TRIANA (TRatamiento del Infarto Agudo de miocardio eN Ancianos) of IMIM-­Hospital del Mar (reference 2014/5491/1), and was declared by the Medicines randomized trial and pooled analysis with previous studies. Eur and Healthcare Products Regulatory Agency to be a study in which the medicine is not Heart J 2011;32:51–60. the fundamental exposure factor investigated (registration: 2258/RG 4274). 17 de la Torre Hernández JM, Brugaletta S, Gómez Hospital JA, et al. Angioplastia primaria en mayores de 75 años. Perfil de pacientes y Provenance and peer review Not commissioned; externally peer reviewed. procedimientos, resultados y predictores pronósticos en el registro Data availability statement No data are available. ATHOS is a collaborative ESTROFA IM + 75. Rev Esp Cardiol 2017;70:81–7. observational study with participation of more than 30 hospitals. To share 18 Ferreira-­González I, Permanyer-­Miralda G, Marrugat J, et al. the complete database all Hospital PIs would have to acknowledge the data Estudio MASCARA (Manejo del Síndrome Coronario Agudo. Registro Actualizado). Resultados globales. Rev Esp Cardiol sharing, which is not impossible but quite difficult under normal conditions. For 2008;61:803–16. exceptionally important projects all PI’s acceptance could be sought. 19 Arós F, Cuñat J, Loma-­Osorio Ángel, et al. Tratamiento del infarto Open access This is an open access article distributed in accordance with the agudo de miocardio en España en el año 2000. El estudio PRIAMHO Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits II. Rev Esp Cardiol 2003;56:1165–73. others to copy, redistribute, remix, transform and build upon this work for any 20 Malkin CJ, Prakash R, Chew DP. The impact of increased age on outcome from a strategy of early invasive management and purpose, provided the original work is properly cited, a link to the licence is given, revascularisation in patients with acute coronary syndromes: and indication of whether changes were made. See: https://​creativecommons.​org/​ retrospective analysis study from the ACACIA registry. BMJ Open licenses/​by/​4.​0/. 2012;2:e000540. 21 Schoenenberger AW, Radovanovic D, Windecker S, et al. Temporal ORCID iDs trends in the treatment and outcomes of elderly patients with acute Daniel Fernández-­Bergés http://​orcid.​org/​0000-​0002-​9561-​2901 coronary syndrome.. Eur Heart J 2016;37:1304–11. Irene R Degano http://​orcid.​org/​0000-​0003-​1914-​6934 22 Zaman MJ, Stirling S, Shepstone L, et al. The association between Luis Rodríguez-­Padial http://​orcid.​org/​0000-​0002-​5783-​3922 older age and receipt of care and outcomes in patients with acute coronary syndromes: a cohort study of the myocardial ischaemia national audit project (MINAP). Eur Heart J 2014;35:1551–8. References 23 Cequier Ángel, Ariza-­Solé A, Elola FJ, et al. Impacto en la 1 Timmis A, Townsend N, Gale C, et al. European Society of mortalidad de diferentes sistemas de asistencia en red para cardiology: cardiovascular disease statistics 2017 on behalf of the el tratamiento del infarto agudo de miocardio con elevación atlas writing group. Eur Heart J 2018;39:508–59. del segmento ST. La experiencia de España. Rev Esp Cardiol 2 Dégano IR, Elosua R, Marrugat J. Epidemiología del síndrome 2017;70:155–61. coronario agudo en España: estimación del número de casos Y La 24 O'Gara PT, Kushner FG, Ascheim DD, et al. 2013 ACCF/AHA tendencia de 2005 a 2049. Rev Esp Cardiol 2013;66:472–81. guideline for the management of ST-­elevation myocardial infarction: 3 Sabaté M. «No es país para viejos» con infarto agudo de miocardio y a report of the American College of cardiology Foundation/American elevación del segmento ST. Rev Esp Cardiol 2017;70:70–1. heart association Task force on practice guidelines. J Am Coll 4 Fernández-­Bergés D, Félix-­Redondo FJ, Consuegra-­Sánchez L, Cardiol 2013;61:e78–140. et al. Infarto de miocardio en mayores de 75 años: Una población en 25 Savonitto S, Morici N, De Servi S. El tratamiento de síndromes aumento. estudio CASTUO. Rev Clín Española 2015;215:195–203. coronarios agudos de ancianos y pacientes con comorbilidades. Rev 5 Ibañez B, James S, Agewall S, et al. 2017 ESC guidelines for the Esp Cardiol 2014;67:564–73. management of acute myocardial infarction in patients presenting with ST-­segment elevation. Eur Heart J 2018;39:119–77. 26 Romagnoli E, Biondi-­Zoccai G, Sciahbasi A, et al. Radial versus 6 Toleva O, Ibrahim Q, Brass N, et al. Treatment choices in elderly femoral randomized investigation in ST-­segment elevation acute patients with ST: elevation myocardial infarction-­insights from the coronary syndrome: the RIFLE-­SETACS (radial versus femoral vital heart response registry. Open Heart 2015;2:e000235. randomized investigation in ST-­elevation acute coronary syndrome) 7 Alexander KP, Newby LK, Armstrong PW, et al. Acute coronary care study. J Am Coll Cardiol 2012;18:2481–9. in the elderly, part II. Circulation 2007;115:2570–89. 27 Navarese EP, Gurbel PA, Andreotti F, et al. Prevention of contrast-­ 8 Subirana I, Fernández Avilés F, Elosua R, et al. Interhospital induced acute kidney injury in patients undergoing cardiovascular variability in acute coronary syndrome management in the ATHOS procedures-­a systematic review and network meta-­analysis. PLoS study. Rev Esp Cardiol 2019;72:691–3. One 2017;12:e0168726. 9 Killip T, Kimball JT. Treatment of myocardial infarction in a coronary 28 Dégano IR, Subirana I, Fusco D, et al. Percutaneous coronary care unit. Am J Cardiol 1967;20:457–64. intervention reduces mortality in myocardial infarction patients with 10 Austin PC. Optimal caliper widths for propensity-­score matching comorbidities: implications for elderly patients with diabetes or when estimating differences in means and differences in proportions kidney disease. Int J Cardiol 2017;249:83–9. in observational studies. Pharm Stat 2011;10:150–61. 29 Kristensen SD, Laut KG, Fajadet J, et al. Reperfusion therapy for ST 11 Heinze G, Jüni P. An overview of the objectives of and elevation acute myocardial infarction 2010/2011: current status in 37 the approaches to propensity score analyses. Eur Heart J ESC countries. Eur Heart J 2014;35:1957–70. 2011;32:1704–8. 30 STENT-­Save A LIfe. Available: https://www.​stentsavealife.​com/ 12 D'Agostino RB. Propensity scores in cardiovascular research. [Accessed 10 Jul 2019]. Circulation 2007;115:2340–3. 31 Widimsky P. Primary angioplasty vs. thrombolysis: the end of the 13 Angeli F, Verdecchia P, Savonitto S, et al. Early invasive versus controversy? Eur Heart J 2010;31:634–6. selectively invasive strategy in patients with non-­ST-­segment 32 Brass LM, Lichtman JH, Wang Y, et al. Intracranial hemorrhage elevation acute coronary syndrome: impact of age. Catheter associated with thrombolytic therapy for elderly patients with acute Cardiovasc Interv 2014;83:686–701. myocardial infarction: results from the cooperative cardiovascular 14 Morici N, Savonitto S, Murena E, et al. Causes of death in patients project. Stroke 2000;31:1802–11. ≥75 years of age with non-­ST-­segment elevation acute coronary 33 Peiyuan H, Jingang Y, Haiyan X, et al. The comparison of the syndrome. Am J Cardiol 2013;112:1–7. outcomes between primary PCI, fibrinolysis, and NO reperfusion 15 Shah P, Najafi AH, Panza JA, et al. Outcomes and quality of life in in patients ≥ 75 years old with ST-­segment elevation myocardial patients>or=85 years of age with ST-­elevation myocardial infarction. infarction: results from the Chinese acute myocardial infarction Am J Cardiol 2009;103:170–4. (CAMI) registry. PLoS One 2016;11:e0165672. Fernández-­Bergés D, et al. Open Heart 2020;7:e001169. doi:10.1136/openhrt-2019-001169 7