Surgical or percutaneous coronary revascularization for heart failure: an in silico model using routinely collected health data to emulate a clinical trial

Suraj Pathak, Florence Y Lai, Joanne Miksza, Mark C Petrie, Marius Roman, Sarah Murray, Jeremy Dearling, Divaka Perera, Gavin J Murphy, Suraj Pathak, Florence Y Lai, Joanne Miksza, Mark C Petrie, Marius Roman, Sarah Murray, Jeremy Dearling, Divaka Perera, Gavin J Murphy

Abstract

Aims: The choice of revascularization with coronary artery bypass grafting (CABG) vs. percutaneous coronary intervention (PCI) in people with ischaemic left ventricular dysfunction is not guided by high-quality evidence.

Methods and results: A trial of CABG vs. PCI in people with heart failure (HF) was modelled in silico using routinely collected healthcare data. The in silico trial cohort was selected by matching the target trial cohort, identified from Hospital Episode Statistics in England, with individual patient data from the Surgical Treatment for Ischemic Heart Failure (STICH) trial. Allocation to CABG vs. complex PCI demonstrated random variation across administrative regions in England and was a valid statistical instrument. The primary outcome was 5-year all-cause mortality or cardiovascular hospitalization. Instrumental variable analysis (IVA) was used for the primary analysis. Results were expressed as average treatment effects (ATEs) with 95 confidence intervals (CIs). The target population included 13 519 HF patients undergoing CABG or complex PCI between April 2009 and March 2015. After matching, the emulated trial cohort included 2046 patients. The unadjusted primary outcome rate was 51.1 in the CABG group and 70.0 in the PCI group. IVA of the emulated cohort showed that CABG was associated with a lower risk of the primary outcome (ATE 16.2, 95 CI 20.6 to 11.8), with comparable estimates in the unmatched target population (ATE 15.5, 95 CI 17.5 to 13.5).

Conclusion: In people with HF, in silico modelling suggests that CABG is associated with fewer deaths or cardiovascular hospitalizations at 5 years vs. complex PCI. A pragmatic clinical trial is needed to test this hypothesis and this trial would be feasible.

Keywords: Clinical trial emulation; Coronary artery bypass grafting; Coronary artery disease; Hospital episode statistics; Percutaneous coronary intervention; Revascularization; Trial feasibility.

Conflict of interest statement

Conflict of interest: G.J.M. is supported by the British Heart Foundation (RG/13/6/29947, CH/12/1/29419, and AA/18/3/34220). M.C.P. is supported by the BHF Centre of Research Excellence award BHF RE/18/634217. M.R. is a NIHR Clinical Lecturer. S.P. is supported by the Leicester NIHR Biomedical Research Centre and the Leicester BHF Research Accelerator AA18/3/34220.

The Author(s) 2022. Published by Oxford University Press on behalf of the European Society of Cardiology.

Figures

Structured Graphical Abstract
Structured Graphical Abstract
In silico trial modelling, using variation in the regional rates of CABG vs. PCI as an instrumental variable, suggests that CABG is superior to PCI at 5 years for the primary composite outcome of all-cause-death or cardiovascular hospitalization. M, matched—emulated trial cohort; U, unmatched—targeted trial cohort; CABG, coronary artery bypass graft; PCI, percutaneous coronary intervention; CI, confidence interval.
Figure 1
Figure 1
Trial modelling framework. ① Data cleaning, harmonization and linking. ② An in silico trial protocol is developed as a best approximation of the target trial protocol given the limitations of the routinely collected HES data. ③ Phenotyping algorithms developed using ICD 10 and OPCS-4 codes are applied to the Health Episodes Statistics database to shortlist a targeted patient cohort. ④ The targeted patient cohort is matched to an existing trial, targeting the same population (STICH), based on key patient baseline characteristics, to derive the emulated trial cohort, which approximates a likely trial cohort. ⑤ Statistical and machine learning methods are applied to the emulated trial population to provide information on trial feasibility, outcome event rates and treatment effects sizes.
Figure 2
Figure 2
Modified CONSORT diagram describing how our cohorts were defined and the flow of patients in our analyses once the trial-modelling framework had been applied. HES, Hospital Episode Statistics; ONS, Office for National Statistics; NIH, National Institutes of Health; CABG, coronary artery bypass graft; CONSORT, consolidated standards of reporting trials; PCI, percutaneous coronary intervention; NSTEMI, non-ST elevated myocardial infarction; STEMI, ST-elevated myocardial infarction; LVAD, left ventricular assist device.
Figure 3
Figure 3
Funnel plot showing regional variation in surgical rates among revascularized heart failure patients in England. A total of 320 administrative regions in England plotted. CABG, coronary artery bypass graft; PCI, percutaneous coronary intervention.
Figure 4
Figure 4
Average treatment effects with 95% confidence interval for the primary composite outcome of all-cause mortality or cardiovascular hospitalization at 5 years for both matched and unmatched cohorts. Only primary diagnosis codes were used. M, matched—emulated trial cohort; U, unmatched—targeted trial cohort; CABG, coronary artery bypass graft; PCI, percutaneous coronary intervention; 95% CI, 95% confidence interval; PD, primary diagnosis codes; ATE, average treatment effects.
Figure 5
Figure 5
Kaplan–Meier plots for coronary artery bypass grafting vs. percutaneous coronary intervention in the matched and unmatched cohorts for the primary composite outcome of all-cause mortality or cardiovascular hospitalization. M, matched—emulated trial cohort; U, unmatched—targeted trial cohort; CABG, coronary artery bypass graft; PCI, percutaneous coronary intervention; 95% CI, 95% confidence interval; PD, primary diagnosis codes.

References

    1. NICE . Resource impact report: Chronic heart failure in adults: diagnosis and management (NG106). NICE. 2018.
    1. Savarese G, Lund LH. Global public health burden of heart failure. Cardiac Fail Rev 2017;3:7–11. 10.15420/cfr.2016:25:2
    1. Velazquez EJ, Lee KL, Jones RH, Al-Khalidi HR, Hill JA, Panza JA, et al. . Coronary-artery bypass surgery in patients with ischemic cardiomyopathy. N Engl J Med 2016;374:1511–1520. 10.1056/NEJMoa1602001
    1. Head SJ, Milojevic M, Daemen J, Ahn JM, Boersma E, Christiansen EH, et al. . Mortality after coronary artery bypass grafting versus percutaneous coronary intervention with stenting for coronary artery disease: a pooled analysis of individual patient data. Lancet 2018;391:939–948. 10.1016/S0140-6736(18)30423-9
    1. Lee CW, Ahn JM, Cavalcante R, Sotomi Y, Onuma Y, Suwannasom P, et al. . Coronary artery bypass surgery versus drug-eluting stent implantation for left main or multivessel coronary artery disease: a meta-analysis of individual patient data. JACC Cardiovasc Interv 2016;9:2481–2489. 10.1016/j.jcin.2016.10.008
    1. Serruys PW, Morice MC, Kappetein AP, Colombo A, Holmes DR, Mack MJ, et al. . Percutaneous coronary intervention versus coronary-artery bypass grafting for severe coronary artery disease. N Engl J Med 2009;360:961–972. 10.1056/NEJMoa0804626
    1. Marui A, Kimura T, Nishiwaki N, Mitsudo K, Komiya T, Hanyu M, et al. . Comparison of five-year outcomes of coronary artery bypass grafting versus percutaneous coronary intervention in patients with left ventricular ejection fractions ≤50% versus >50% (from the CREDO-Kyoto PCI/CABG Registry Cohort-2). Am J Cardiol 2014;114:988–996. 10.1016/j.amjcard.2014.07.007
    1. Murphy G. National Cardiac Surgery Clinical Trial Initiative. University of Leicester. 2021.
    1. Sarah Murray JC, Collis P, Dearling J, Haywood P, Keys A, Leeman I, et al. . National PPI Group. University of Leicester.
    1. Perera D, Clayton T, O’Kane PD, Greenwood JP, Weerackody R, Ryan M, et al. . Percutaneous revascularization for ischemic left ventricular dysfunction. N Engl J Med 2022;387:1351–1360. 10.1056/NEJMoa2206606
    1. Zhang Y, Wang Y, Xu J, Zhu B, Chen X, Ding X, et al. . Comparison of prediction models for acute kidney injury among patients with hepatobiliary malignancies based on XGBoost and LASSO-logistic algorithms. Int J Gen Med 2021;14:1325–1335. 10.2147/IJGM.S302795
    1. Chen Z, Luo H, Xu L. Machine learning models of ischemia/hemorrhage in moyamoya disease and analysis of its risk factors. Clin Neurol Neurosurg 2021;209:106919. 10.1016/j.clineuro.2021.106919
    1. Coban M. Redistribution preferences, attitudes towards immigrants, and ethnic diversity, IAB Discussion Paper 2020/23. 2020.
    1. Leuven E, Sianesi B.. PSMATCH2: Stata module to perform full Mahalanobis and propensity score matching, common support graphing, and covariate imbalance testing. Statistical Software Components S432001, Boston College Department of Economics, revised 01 Feb 2018. 2003.
    1. Chen H, Harinen T, Lee J-Y, Yung M, Zhao Z.. CausalML: Python package for causal machine learning, 2020, eprint: 2002.11631, arXiv.
    1. Hou N, Li M, He L, Xie B, Wang L, Zhang R, et al. . Predicting 30-days mortality for MIMIC-III patients with sepsis-3: a machine learning approach using XGBoost. J Transl Med 2020;18:462. 10.1186/s12967-020-02620-5
    1. Wei C, Zhang L, Feng Y, Ma A, Kang Y. Machine learning model for predicting acute kidney injury progression in critically ill patients. BMC Med Inform Decis Mak 2022;22:17. 10.1186/s12911-021-01740-2
    1. Setty VMVA. XGBoost Algorithm. Towards Data Science. 2019.
    1. Barthel FMS, Babiker A, Royston P, Parmar MKB. Evaluation of sample size and power for multi-arm survival trials allowing for non-uniform accrual, non-proportional hazards, loss to follow-up and cross-over. Stat Med 2006;25:2521–2542. 10.1002/sim.2517
    1. Farkouh ME, Domanski M, Sleeper LA, Siami FS, Dangas G, Mack M, et al. . Strategies for multivessel revascularization in patients with diabetes. N Engl J Med 2012;367:2375–2384. 10.1056/NEJMoa1211585
    1. Holm NR, Mäkikallio T, Lindsay MM, Spence MS, Erglis A, Menown IBA, et al. . Percutaneous coronary angioplasty versus coronary artery bypass grafting in the treatment of unprotected left main stenosis: updated 5-year outcomes from the randomised, non-inferiority NOBLE trial. Lancet 2020;395:191–199. 10.1016/S0140-6736(19)32972-1
    1. Burns EM, Rigby E, Mamidanna R, Bottle A, Aylin P, Ziprin P, et al. . Systematic review of discharge coding accuracy. J Public Health 2012;34:138–148. 10.1093/pubmed/fdr054
    1. Campbell SE, Campbell MK, Grimshaw JM, Walker AE. A systematic review of discharge coding accuracy. J Public Health 2001;23:205–211. 10.1093/pubmed/23.3.205
    1. Völz S, Redfors B, Angerås O, Ioanes D, Odenstedt J, Koul S, et al. . Long-term mortality in patients with ischaemic heart failure revascularized with coronary artery bypass grafting or percutaneous coronary intervention: insights from the Swedish Coronary Angiography and Angioplasty Registry (SCAAR). Eur Heart J 2021;42:2657–2664. 10.1093/eurheartj/ehab273
    1. McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M, et al. . 2021 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure: developed by the task force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC) with the special contribution of the Heart Failure Association (HFA) of the ESC. Eur Heart J 2021;42:3599–3726. 10.1093/eurheartj/ehab368
    1. Neumann FJ, Sousa-Uva M, Ahlsson A, Alfonso F, Banning AP, Benedetto U, et al. . 2018 ESC/EACTS guidelines on myocardial revascularization. Eur Heart J 2019;40:87–165. 10.1093/eurheartj/ehy394
    1. Deo SV, Reddy YNV, Zakeri R, Karnib M, Selvaganesan P, Elgudin Y, et al. . Revascularization in ischaemic heart failure with preserved ejection fraction: a nationwide cohort study. Eur J Heart Fail 2022;24:1427–1438. 10.1002/ejhf.2446
    1. Velazquez EJ, Lee KL, Deja MA, Jain A, Sopko G, Marchenko A, et al. . Coronary-artery bypass surgery in patients with left ventricular dysfunction. N Engl J Med 2011;364:1607–1616. 10.1056/NEJMoa1100356

Source: PubMed

Szponzorok és közreműködők

Egészségi állapot

Kábítószer-beavatkozások

3
Iratkozz fel