Prevalence and Impact of Treatment Crossover in Cardiac Surgery Randomized Trials: A Meta-Epidemiologic Study

Meta-Analysis in Surgical Research: Methodology and Statistical Application

In evidence-based medicine, systematic review continues to carry the highest weight in terms of quality and reliability, synthesizing robust information from previously published cohort studies to provide a comprehensive overview of a topic. Meta-analysis provides further depth by allowing for comparative analysis between the studied intervention and the control group, providing the most up-to-date evidence on their characteristics and efficacy. We discuss the principles and methodology of meta-analysis, and its applicability to the field of surgical research. The clinical question is defined using PICO framework (Problem, Intervention, Comparison, Outcome). Then a systematic article search is performed across multiple medical databases using relevant search terms, which are then filtered out based on appropriate screening tools. Pertinent data from the selected articles are collected and undergo critical appraisal by at least two independent reviewers. Additional statistical tests may be performed to identify the presence of any significant bias. The data are then synthesized to perform comparative analysis between the intervention and comparison groups. In this article, we discuss specifically the usage of R software (R Foundation for Statistical Computing, Vienna, Austria) for data analysis and visualization. Meta-analysis results of the pooled data are presented using forest plots. Concerns for potential bias may be addressed through the creation of funnel plots. Meta-analysis is a powerful tool to provide highly reliable medical evidence. It may be readily performed by independent researchers with minimal need for funding or institutional approval. The ability to conduct such studies is an asset to budding medical scholars.

GRADE concept 4: rating the certainty of evidence when study interventions or comparators differ from PICO targets

OBJECTIVES

This Grading of Recommendations Assessment, Development and Evaluation (GRADE) concept article offers systematic reviewers, guideline authors, and other users of evidence assistance in addressing randomized trial situations in which interventions or comparators differ from those in the target people, interventions, comparators, and outcomes. To clarify what GRADE considers under indirectness of interventions and comparators, we focus on a particular example: when comparator arm participants receive some or all aspects of the intervention management strategy (treatment switching).

STUDY DESIGN AND SETTING

An interdisciplinary panel of the GRADE working group members developed this concept article through an iterative review of examples in multiple teleconferences, small group sessions, and e-mail correspondence. After presentation at a GRADE working group meeting in November 2022, attendees approved the final concept paper, which we support with examples from systematic reviews and individual trials.

RESULTS

In the presence of safeguards against risk of bias, trials provide unbiased estimates of the effect of an intervention on the people as enrolled, the interventions as implemented, the comparators as implemented, and the outcomes as measured. Within the GRADE framework, differences in the people, interventions, comparators, and outcomes elements between the review or guideline recommendation targets and the trials as implemented constitute issues of indirectness. The intervention or comparator group management strategy as implemented, when it differs from the target comparator, constitutes one potential source of indirectness: Indirectness of interventions and comparators-comparator group receipt of the intervention constitutes a specific subcategory of said indirectness. The proportion of comparator arm participants that received the intervention and the apparent magnitude of effect bear on whether one should rate down, and if one does, to what extent.

CONCLUSION

Treatment switching and other differences between review or guideline recommendation target interventions and comparators vs. interventions and comparators as implemented in otherwise relevant trials are best considered issues of indirectness.

Reappraisal of the Medical Research Council Antiepileptic Drug Withdrawal Study: contamination‐adjusted and dose‐response re‐analysis

Objective

The 1991 Medical Research Council (MRC) Study compared seizure relapse for seizure‐free patients randomized to withdraw vs continue of antiseizure medications (ASMs). We re‐analyzed this trial to account for crossover between arms using contamination‐adjusted intention to treat (CA ITT) methods, to explore dose‐response curves, and to validate predictions against external data. ITT assesses the effect of being randomized to withdraw, as‐treated analysis assesses the confounded effect of withdrawing, but CA ITT assesses the unconfounded effect of actually withdrawing.

Methods

CA ITT involves two stages. First, we used randomized arm to predict whether patients withdrew their ASM (logistic) or total daily ASM dose (linear). Second, we used those values to predict seizure occurrence (logistic).

Results

The trial randomized 503 patients to withdraw and 501 patients to continue ASMs. We found that 316 of 376 patients (88%) who were randomized to withdraw decreased their dose at every pre‐seizure visit, compared with 35 of 424 (8%) who were randomized to continue (p < .01). Adjusted odds ratios of a 2‐year seizure for those who withdrew vs those who did not was 1.3 (95% confidence interval [CI] 0.9–1.9) in the as‐treated analysis, 2.5 (95% CI 1.9–3.4) comparing those randomized to withdraw vs continue for ITT, and 3.1 (95% CI 2.1–4.5) for CA ITT. Probabilities (withdrawal vs continue) were 28% vs 24% (as‐treated), 40% vs 22% (ITT), and 43% vs 21% (CA ITT). Differences between ITT and CA ITT were greater when varying the predictor (reaching zero ASMs) or outcome (1‐year seizures). As‐treated dose‐response curves demonstrated little to no effects, but larger effects in CA ITT analysis. MRC data overpredicted risk in Lossius data, with moderate discrimination (areas under the curve ~0.70).

Significance

CA ITT results (the effect of actually withdrawing ASMs on seizures) were slightly greater than ITT effects (the effect of recommend withdrawing ASMs on seizures). How these findings affect clinical practice must be individualized.

Representation of racial minorities in cardiac surgery randomized clinical trials

BACKGROUND

Racial minorities account for 39.9% of the population in the United States, but are often underrepresented in clinical research. Results from studies predominantly enrolling White patients may not apply to racial minorities. The aim of this analysis is to assess the representation of racial minorities in cardiac surgery randomized clinical trials (RCTs).

METHODS

A systematic search of the literature was performed. All RCTs published from 2000 to 2020 including at least 100 patients and comparing two or more adult cardiac surgery procedures were included. Meta-analytic estimates were calculated.

RESULTS

Among 51 cardiac surgery RCTs published between 2000 and 2020, only 9 (17.6%) reported the race of patients and were included in the final analysis. All of them were multicentric, with a mean of 33 centers included. Six RCTs enrolled patients undergoing coronary artery bypass grafting (66.7%), while the remaining three were on valve surgery (33.3%). Overall, 9193 patients were included; of them, 8034 (87.4%) were White and 1026 (11.2%) nonWhite (386 [4.2%] Black, 191 [2.1%] Hispanic, 274 [3.0%] from other races, and 175 [1.9%] nonWhite patients of unspecified race). The proportion of nonWhite patients did not change over time.

CONCLUSIONS

Only 9 (17.6%) of the 51 cardiac surgery RCTs published between 2000 and 2020 reported the race of the patients enrolled and only 11.2% of them were nonWhite patients. Given the association between race and clinical outcomes, future RCTs should either guarantee a balanced inclusion of racial minorities or be designed to specifically enroll them.

Methodological Standards for the Design, Implementation, and Analysis of Randomized Trials in Cardiac Surgery: A Scientific Statement From the American Heart Association

Cardiac surgery presents specific methodological challenges in the design, implementation, and analysis of randomized controlled trials. The purposes of this scientific statement are to review key standards in cardiac surgery randomized trial design and implementation, and to provide recommendations for conducting and interpreting cardiac surgery trials. Recommendations include a careful evaluation of the suitability of the research question for a clinical trial, assessment of clinical equipoise, feasibility of enrolling a representative patient cohort, impact of practice variations on the safety and efficacy of the study intervention, likelihood and impact of crossover, and duration of follow-up. Trial interventions and study end points should be predefined, and appropriate strategies must be used to ensure adequate deliverability of the trial interventions. Every effort must be made to ensure a high completeness of follow-up; trial design and analytic techniques must be tailored to the specific research question and trial setting.

Association of Radial Artery Graft vs Saphenous Vein Graft With Long-term Cardiovascular Outcomes Among Patients Undergoing Coronary Artery Bypass Grafting: A Systematic Review and Meta-analysis

IMPORTANCE

Observational studies have suggested that the use of radial artery grafts for coronary artery bypass grafting may improve clinical outcomes compared with the use of saphenous vein grafts, but this has not been confirmed in randomized trials.

OBJECTIVE

To compare clinical outcomes between patients receiving radial artery vs saphenous vein grafts for coronary artery bypass grafting after long-term follow-up.

DESIGN, SETTING, AND PARTICIPANTS

Patient-level pooled analysis comparing radial artery vs saphenous vein graft in adult patients undergoing isolated coronary artery bypass grafting from 5 countries (Australia, Italy, Serbia, South Korea, and the United Kingdom), with enrollment from 1997 to 2009 and follow-up completed in 2019.

INTERVENTIONS

Patients were randomized to undergo either radial artery (n = 534) or saphenous vein (n = 502) grafts for coronary artery bypass grafting.

MAIN OUTCOMES AND MEASURES

The primary outcome was a composite of death, myocardial infarction, or repeat revascularization and the secondary outcome was a composite of death or myocardial infarction.

RESULTS

A total of 1036 patients were randomized (mean age, 66.6 years in the radial artery group vs 67.1 years in the saphenous vein group; 376 [70.4%] men in the radial artery group vs 351 [69.9%] in the saphenous vein group); 942 (90.9%) of the originally randomized patients completed 10 years of follow-up (510 in the radial artery group). At a median (interquartile range) follow-up of 10 (10-11) years, the use of the radial artery, compared with the saphenous vein, in coronary artery bypass grafting was associated with a statistically significant reduction in the incidence of the composite outcome of death, myocardial infarction, or repeat revascularization (220 vs 237 total events; 41 vs 47 events per 1000 patient-years; hazard ratio, 0.73 [95% CI, 0.61-0.88]; P < .001) and of the composite of death or myocardial infarction (188 vs 193 total events; 35 vs 38 events per 1000 patient-years; hazard ratio, 0.77 [95% CI, 0.63-0.94]; P = .01).

CONCLUSIONS AND RELEVANCE

In this individual participant data meta-analysis with a median follow-up of 10 years, among patients undergoing coronary artery bypass grafting, the use of the radial artery compared with the saphenous vein was associated with a lower risk of a composite of cardiovascular outcomes.

Prevalence and Impact of Treatment Crossover in Cardiac Surgery Randomized Trials: A Meta-Epidemiologic Study

Background Crossover dilutes treatment effect and reduces statistical power of intention-to-treat analysis. We examined incidence and impact on cardiac surgery randomized controlled trial (RCT) outcomes of crossover from experimental to control interventions, or vice versa. Methods and Results MEDLINE, EMBASE, and Cochrane Library were searched, and RCTs (≥100 patients) comparing ≥2 adult cardiac surgical interventions were included. Crossover from the initial treatment assignment and relative risks (RRs) for each trial's primary end point and mortality at longest available follow-up were extracted. All RRs were calculated as >1 favored control group and <1 favored experimental arm. Primary outcome was the effect estimate for primary end point of each RCT, and secondary outcome was all-cause mortality; both were appraised as RR at the longest follow-up available. Sixty articles reporting on 47 RCTs (25 440 patients) were identified. Median crossover rate from experimental to control group was 7.0% (first quartile, 2.0%; third quartile, 9.7%), whereas from control to experimental group, the rate was 1.3% (first quartile, 0%; third quartile, 3.6%). RRs for primary end point and mortality were higher in RCTs with higher crossover rate from experimental to control group (RR, 1.01 [95% CI, 0.94-1.07] versus RR, 0.80 [95% CI, 0.66-0.97] and RR, 1.02 [95% CI, 0.95-1.11] versus RR, 0.94 [95% CI, 0.82-1.07], respectively). Crossover from control to experimental group did not alter effect estimates for primary end point or mortality (RR, 0.82 [95% CI, 0.63-1.05] versus RR, 0.95 [95% CI, 0.86-1.04] and RR, 0.88 [95% CI, 0.73-1.07] versus RR, 1.02 [95% CI, 0.95-1.09], respectively). Conclusions Crossover from experimental to control group is associated with outcomes of cardiac surgery RCTs. Crossover should be minimized at designing stage and carefully appraised after study completion.