Equivalent outcomes after coronary artery bypass graft surgery performed by consultant versus trainee surgeons: A systematic review and meta-analysis

Performance of left internal thoracic artery-left anterior descending artery anastomosis by residents versus attendings and coronary artery bypass grafting outcomes

OBJECTIVES

Performance of a technically sound left internal thoracic artery to left anterior descending artery (LITA-LAD) anastomosis during coronary artery bypass grafting (CABG) is critically important. We used prospectively collected data from the multicentre, randomized REGROUP (Randomized Endograft Vein Perspective) trial to investigate CABG outcomes based on whether a resident or an attending surgeon performed the LITA-LAD anastomosis.

METHODS

This was a post hoc subanalysis of the REGROUP trial, which randomized veterans undergoing isolated on-pump CABG to endoscopic versus open vein harvest from 2014 through 2017. The primary end point was major cardiac adverse events, defined as the composite of all-cause deaths, nonfatal myocardial infarctions or repeat revascularizations.

RESULTS

Among 1,084 patients, 344 (31.8%) LITA-LAD anastomoses were performed by residents and 740 (68.2%), by attending surgeons. Residents (compared to attendings) operated on fewer patients with high tercile SYNTAX scores (22.1% vs 37.4%, P < 0.001), performed fewer multiarterial CABGs (5.2% vs 14.6%, P < 0.001) and performed more anastomoses to distal targets with diameters > 2.0 mm (19.0% vs 10.9%, P < 0.001) and non-calcified landing zones (25.1% vs 21.6%, P < 0.001). During a median observation time of 4.7 years (interquartile range 3.84-5.45), major cardiac adverse events occurred in 77 patients (22.4%) in the group treated by residents and 169 patients (22.8%) in the group treated by attendings (unadjusted HR 1.00; 95% confidence interval, 0.76-1.33; P = 0.99). Outcomes persisted on adjusted analyses.

CONCLUSIONS

Based on this REGROUP trial subanalysis, under careful supervision and with appropriate patient selection, LITA-LAD anastomoses performed by the residents yielded clinical outcomes similar to those of the attendings.

Operative Teaching of Coronary Bypass and Need for Repeat Catheterization: Does it Matter Who is Sewing?

OBJECTIVE

There are no studies to date comparing the patency of coronary bypass grafts constructed by attending surgeons versus trainees and the potential consequences of any such disparities. We explored this issue by comparing the patency of individual anastomoses performed by residents versus the attending surgeon.

DESIGN

We reviewed 765 continuous cases performed by a single surgeon which involved at least 1 coronary bypass anastomosis, totaling 2,173 distal anastomoses. At a median follow-up time of 36 months (interquartile range 20.5-47.3), 83 (10.9%) patients had undergone 110 cardiac catheterization procedures after their original operation for various indications. This angiographic information provided the data for our comparison cohorts.

SETTING

Cardiac surgery practice within an academic setting PARTICIPANTS: Adult patient undergoing coronary bypass grafting RESULTS: Of the 83 patients that underwent repeat catheterization, 23 (27.7%) were resident cases, 25 (30.1%) were attending cases and 35 (42.2%) were mixed. There were 4/83 (4.8%) patients with angiographic evidence of internal mammary artery graft compromise of which 3/4 (75%) had been constructed by the attending surgeon. Angiographic evidence of saphenous vein graft compromise was appreciated in 16/83 (19.3%) patients of which 9/16 (56.3%) of the grafts were constructed by the attending surgeon.

CONCLUSIONS

Liberal involvement of surgical trainees as primary operators in coronary revascularization cases led to equivalent rates of postoperative ischemic complications between the attending and resident groups. The outcome equivalence was also maintained when evaluated at the individual anastomosis patency level between the 2 groups. We conclude that academic programs should continue providing trainees significant experience as primary operating surgeons without fear of clinical outcome compromise.

Are there differences in cardiothoracic surgery performed by trainees versus fully trained surgeons?

OBJECTIVES

We sought to assess the safety of training in cardiothoracic surgery comparing outcomes of cases performed by trainees versus fully trained surgeons.

METHODS

EmBase, Scopus, PubMed, and OVID MEDLINE were searched in August 2021 independently by two authors. A third author arbitrated decisions to resolve disagreements. Inclusion criteria were articles on cardiothoracic surgery reporting on outcomes for trainees. Studies were assessed for appropriateness as per CBEM criteria. Eight hundred and ninety-two results were obtained, 27 represented best evidence (2-meta-analyses, 1-RCT, and 24 retrospective cohort studies).

RESULTS

In all 474,160 operative outcomes were assessed for 434,535 coronary artery bypass grafting (CABG) (431,329 on-pump vs. 3206 off-pump), 3090 AVR, 1740 MVR/repair, 26,433 mixed, 3565 congenital, and 4797 thoracic procedures. In all 398,058 cases were performed by trainees and 75,943 by consultants. One hundred fifty-nine cases were indeterminate. There were no statistically significant differences in the patients' preoperative risk scores. All studies excluded extreme high-risk patients in emergency setting, patients with poor left ventricular function, and reoperation cases that were undertaken by consultants. There were no differences in cardiopulmonary bypass and clamp times for CABG. Times for valve replacement and repair cases were longer for trainees. There were no differences in the postoperative outcomes including perioperative myocardial infarction, resternotomy for bleeding, stroke, renal failure, intensive therapy unit length of stay, and total length of stay. One study reported no differences on angiographic graft patency at 1 year. There were no differences in in-hospital or midterm mortality out to 5-years.

DISCUSSION

Trainees can perform cardiothoracic surgery in dedicated high-volume units with outcomes comparable to those of fully trained surgeons.

A retrospective observational study of patient analgesia outcomes when regional anaesthesia procedures are performed by consultants versus supervised trainees

At teaching hospitals, consultants must provide effective supervision, including appropriate selection of teaching cases, such that the outcomes achieved by trainees are similar to that of consultants. Numerous studies in the surgical literature have compared patient outcomes when surgery is performed by consultant surgeons or surgical trainees but, to our knowledge, none exist in the field of anaesthesia. We aimed to compare analgesia outcomes of regional anaesthesia when performed by supervised trainees versus consultants. We designed a retrospective observational study using registry data. The primary outcome was inadequate analgesia, defined as a numerical rating scale (NRS) for pain >5 reported at any time in the post-anaesthesia care unit (PACU). Secondary outcomes included the maximum pain NRS, pain experienced in the PACU, and the requirement for systemic opioid analgesia in the PACU. Of the 1814 patients analysed, the primary proceduralist was a consultant for 514 (28.3%) patients and a trainee for 1300 (71.7%) patients. All trainees were supervised by an on-site consultant. There were no statistically significant differences between consultants and supervised trainees in terms of the primary outcome (NRS >5 in 24.9% and 24.5% of patients, respectively; P = 0.84) and secondary outcomes. Compared to trainees, consultants had a slightly higher rate of patients with a body mass index >30 kg/m², an American Society of Anesthesiologists Physical Status Classification of 3 or 4, nerve blocks performed under general anaesthesia, paravertebral/neuraxial blocks and blocks with perineural catheter placement. Regional anaesthesia performed by supervised trainees can achieve similar analgesia outcomes to consultant-performed procedures.

Outcomes of cardiac surgical procedures performed by trainees versus consultants: A systematic review with meta-analysis

OBJECTIVES

Cardiac surgery is highly demanding and the ideal teaching method to reach competency is widely debated. Some studies have shown that surgical trainees can safely perform full operations with equivocal outcomes compared with their consultant colleagues while under supervision. We aimed to compare outcomes after cardiac surgery with supervised trainee involvement versus consultant-led procedures.

METHODS

We systematically reviewed databases (PubMed/MEDLINE, Embase, the Cochrane Central Register of Controlled Trials, ClinicalTrials.gov, Google Scholar) and reference lists of relevant articles for studies that compared outcomes of cardiac surgery performed by trainees versus consultants. Primary end points included: operative mortality, coronary events, neurological/renal complications, reoperation, permanent pacemaker implantation, and sternal complications. Secondary outcomes included cardiopulmonary bypass and aortic cross-clamp times and intensive care/in-hospital length of stay. Random effects meta-analysis was performed.

RESULTS

Thirty-three observational studies that reported on a total of 81,616 patients (trainee: 20,154; consultant: 61,462) were included. There was a difference favoring trainees in terms of operative mortality in the main analysis and in an analysis restricted to propensity score-matched samples, whereas other outcomes were not consistently different in both analyses. Overall cardiopulmonary bypass and aortic cross-clamp times were longer in the trainee group but did not translate in longer intensive care unit or hospital stay.

CONCLUSIONS

In the right conditions, good outcomes are possible in cardiac surgery with trainee involvement. Carefully designed training programs ensuring graduated hands-on operative exposure as primary operator with appropriate supervision is fundamental to maintain high-quality training in the development of excellent cardiac surgeons.

Training Cardiac Surgeons: Safety and Requirements

To analyze whether cardiac surgical residents can perform their first surgeries without compromising patients' safety or outcomes, by comparing their performance and results to those of senior surgeons. All documented CABGs conducted between 2002 and 2020 were included. Surgeries were divided according to the experience level of the main surgeon (defined by the number of CABG conducted by him/her) using the following thresholds: 1000; 150; 80 and 35. This resulted in 5 groups: senior surgeons (the reference group); attending surgeons; fellow surgeons; advanced residents and new residents. Primary endpoint was 30 day mortality. Secondary endpoints included a list of intra and post-operative parameters (including in-hospital complications). A multivariable analysis was conducted. 16,486 CABG were conducted by 66 different surgeons over a period of 18 years. Multivariable analysis did not find significant differences between both the primary and the secondary endpoints. Skin-to-skin time correlated significantly with experience level, as new residents needed almost 30% more time than senior surgeons (234 vs 180 minutes). With a suitable supervision by experienced surgeons, patient selection and sufficient resources (longer duration of surgery), surgical residents can perform CABGs with good results and without compromising the patient's outcome.

Is It Safe to Let Trainees Operate on High Risk Cardiac Surgery Cases?

Increasing complexity in cardiac operations has raised the discussion on trainee autonomy and the number of cases required to achieve competency. This study compares outcomes among cases done by trainees vs consultants for high risk patients. 696 (trainee=158 vs consultant=438) major high risk cardiac operations (Euroscore >10) were reviewed at a single center. Observations were propensity matched to consultant or trainee based on several baseline characteristics. Euroscore was: Trainee; 12.3 ± 1.6 versus Consultant; 12.8 ± 2.2, p=.036. Multivariable analysis did not identify trainee as a risk factor for worse in-hospital mortality (OR; 0.95, CI; 0.4-2.2, p=.914) or composite outcome of length of stay >30 days, deep sternal infection, new hemodialysis, new stroke or transient ischemic attack, in-hospital death or reoperation (OR; 0.64, CI; 0.39-1.03, p=.069). NYHA class, diabetes and emergency/salvage surgery were predictors of worse composite outcome. After propensity matching (130 pairs), there was no difference in reoperation rates (3.1% versus 4.6%, p=.727), inhospital death (5.4% versus 7.7%, p=.607) or composite outcome (20.8% versus 29.2%, p=.152). There was no statistical difference in cross clamp times (Trainee; 74.0 ± 32.7 min vs Consultant; 82.6 ± 51.1, p=.229) and bypass times (Trainee; 116.3 ± 52.8 min versus Consultant 135.3 ± 72.6 min, p=.055). The length of stay was similar (18.2 ± 13.2 days versus 19.9 ± 15.6 days, p=.302). It is possible for trainees to perform high risk cardiac surgery without compromising the quality of patient care.

Effect of Hospital and Surgeon Procedure Volumes on the Incidence of Intraoperative Conversion During Off-Pump Coronary Artery Bypass Grafting

Intraoperative conversion to cardiopulmonary bypass with its subsequent high mortality is a major concern associated with off-pump coronary artery bypass grafting (OPCAB). The impact of procedure volume on the incidence of intraoperative conversion, however, is poorly defined. This study therefore evaluated the effect of procedure volume on the incidence of conversion in OPCAB using nationwide data. We analyzed 31,361 patients who underwent primary, nonemergent, isolated OPCAB during 2013-2016 reported in the Japan Cardiovascular Surgery Database. Hospitals (n = 548) and surgeons (n = 1315) were divided into tertile categories (low-, medium-, and high volumes) based on the total number of isolated coronary artery bypass grafting (CABG). Hierarchical logistic regression analysis, including 22 preoperative factors and hospital and surgeon CABG volumes, was used to assess the relation between procedure volume and the risk of conversion due to bleeding/hemodynamic instability. There were 797 (2.5%) intraoperative conversions due to bleeding/hemodynamic instability. Risk-adjusted odds ratios for conversion were significantly lower in some combined hospital/surgeon CABG volume categories than in the reference category. Hospital/surgeon volumes and their odds ratio (95% confidence interval) were as follows: low/low 1.00 (reference); medium/low 0.62 (0.39-0.96); high/low 0.47 (0.27-0.81); high/high 0.58 (0.38-0.89). There was a lower risk of conversion in medium- and high-volume than low-volume hospitals, especially among low-volume surgeons. Procedure volume is associated with the incidence of conversion during OPCAB. Among low-volume surgeons, hospital CABG volume significantly reduces conversion in a volume-dependent manner. These findings will be useful for safety training of OPCAB surgeons.

Trainee surgeons and patient outcomes in carotid endarterectomy: a retrospective cohort study

BACKGROUND

In Australia and New Zealand, more than 2000 carotid endarterectomies are performed annually. The major morbidities arising from this procedure are post-operative stroke, cranial nerve injury and death. Carotid endarterectomy surgery is a key component of the vascular surgical training programme. We assessed the impact of having a surgical trainee perform a major component of this procedure on the post-operative rates of stroke, cranial nerve injury and mortality.

METHODS

We performed a retrospective cohort study of vascular surgical patients undergoing carotid endarterectomy, with data obtained from the Australasian Vascular Audit database between January 2010 and December 2014. The dataset comprised of 6528 carotid endarterectomies performed during this time. The collected data were stratified into two categories - consultant-led cases, and those in which trainee surgeons performed at least a major component of the surgery under consultant supervision. The results were analysed for differences in post-operative stroke, cranial nerve injury and inpatient mortality. Differences between groups were assessed using multivariate analysis, adjusting for potentially confounding covariables.

RESULTS

On multivariate analysis, there was no statistically significant difference in the rates of post-operative stroke (odds ratio (OR) 0.88, 95% confidence interval (CI) 0.57-1.36, P = 0.55), cranial nerve injury (OR 0.68, 95% CI 0.39-1.21, P = 0.19) or inpatient mortality (OR 0.78, 95% CI 0.29-2.13, P = 0.63) between the two cohorts.

CONCLUSION

Having surgical trainees perform components of carotid endarterectomies under supervision is not associated with an increased rate of post-operative stroke, cranial nerve injury or mortality.

Publication and related bias in quantitative health services and delivery research: a multimethod study

Background

Bias in the publication and reporting of research findings (referred to as publication and related bias here) poses a major threat in evidence synthesis and evidence-based decision-making. Although this bias has been well documented in clinical research, little is known about its occurrence and magnitude in health services and delivery research.

Objectives

To obtain empirical evidence on publication and related bias in quantitative health services and delivery research; to examine current practice in detecting/mitigating this bias in health services and delivery research systematic reviews; and to explore stakeholders’ perception and experiences concerning such bias.

Methods

The project included five distinct but interrelated work packages. Work package 1 was a systematic review of empirical and methodological studies. Work package 2 involved a survey (meta-epidemiological study) of randomly selected systematic reviews of health services and delivery research topics (n = 200) to evaluate current practice in the assessment of publication and outcome reporting bias during evidence synthesis. Work package 3 included four case studies to explore the applicability of statistical methods for detecting such bias in health services and delivery research. In work package 4 we followed up four cohorts of health services and delivery research studies (total n = 300) to ascertain their publication status, and examined whether publication status was associated with statistical significance or perceived ‘positivity’ of study findings. Work package 5 involved key informant interviews with diverse health services and delivery research stakeholders (n = 24), and a focus group discussion with patient and service user representatives (n = 8).

Results

We identified only four studies that set out to investigate publication and related bias in health services and delivery research in work package 1. Three of these studies focused on health informatics research and one concerned health economics. All four studies reported evidence of the existence of this bias, but had methodological weaknesses. We also identified three health services and delivery research systematic reviews in which findings were compared between published and grey/unpublished literature. These reviews found that the quality and volume of evidence and effect estimates sometimes differed significantly between published and unpublished literature. Work package 2 showed low prevalence of considering/assessing publication (43%) and outcome reporting (17%) bias in health services and delivery research systematic reviews. The prevalence was lower among reviews of associations than among reviews of interventions. The case studies in work package 3 highlighted limitations in current methods for detecting these biases due to heterogeneity and potential confounders. Follow-up of health services and delivery research cohorts in work package 4 showed positive association between publication status and having statistically significant or positive findings. Diverse views concerning publication and related bias and insights into how features of health services and delivery research might influence its occurrence were uncovered through the interviews with health services and delivery research stakeholders and focus group discussion conducted in work package 5.

Conclusions

This study provided prima facie evidence on publication and related bias in quantitative health services and delivery research. This bias does appear to exist, but its prevalence and impact may vary depending on study characteristics, such as study design, and motivation for conducting the evaluation. Emphasis on methodological novelty and focus beyond summative assessments may mitigate/lessen the risk of such bias in health services and delivery research. Methodological and epistemological diversity in health services and delivery research and changing landscape in research publication need to be considered when interpreting the evidence. Collection of further empirical evidence and exploration of optimal health services and delivery research practice are required.

Study registration

This study is registered as PROSPERO CRD42016052333 and CRD42016052366.

Funding

This project was funded by the National Institute for Health Research (NIHR) Health Services and Delivery Research programme and will be published in full in Health Services and Delivery Research; Vol. 8, No. 33. See the NIHR Journals Library website for further project information.

Safety of tympanoplasty and ossiculoplasty performed by otorhinolaryngology trainees

OBJECTIVE

This study aimed to examine the impact of trainee involvement in performing tympanoplasty or tympano-ossiculoplasty on outcomes.

METHODS

A retrospective analysis was performed of a prospective database of all patients undergoing tympanoplasty and tympano-ossiculoplasty in a single centre during a three-year period. Patients were divided into three primary surgeon groups: consultants, fellows and residents. The outcomes of operative time, surgical complications, length of hospital stay, and air-bone gap improvement were compared among the groups.

RESULTS

The study included 398 tympanoplasty and tympano-ossiculoplasty surgical procedures, 71 per cent of which were performed by junior trainees (residents). The junior trainee group was associated with a significantly longer surgical time, without adverse impact on outcomes.

CONCLUSION

Trainee participation in tympanoplasty and tympano-ossiculoplasty surgery was associated with longer surgical time, but did not negatively affect the peri-operative course or hearing outcome. Therefore, resident involvement in these types of surgery is safe.

Education and the pediatric surgeon

There have been many changes in the domain of clinical surgery over the last 2 decades, but probably none more so than in education. The British Association of Pediatric Surgeons recognizes the importance of this topic and has commissioned a special session on this subject jointly with the International Pediatric Endosurgery Group. Free papers presented on a number of educational subjects and specific topics of recruitment and simulation were covered in invited lectures. The session was completed with a keynote lecture from the renowned educationalist Professor Roger Kneebone.

LEVEL OF EVIDENCE

5 (Expert Opinion).

Influence of Specialty Training and Trainee Involvement on Perioperative Outcomes of Esophagectomy

BACKGROUND

Hospitals' and surgeons' volume-outcome relationship have been reported in several esophagectomy studies with an inverse association of mortality and volume. The purpose of our study was to evaluate the outcomes of esophagectomy in the United States relative to the surgeon's specialty.

METHODS

This was a retrospective analysis using the American College of Surgeons National Surgical Quality Improvement Program database (2006 to 2013). All patients (18 years of age and older) who underwent esophagectomy were divided into 2 groups according to whether the operation was performed by a general surgeon (GS) or a cardiothoracic surgeon (CTS). A comparison of intraoperative and postoperative outcomes between the groups was conducted. The primary outcome was 30-day mortality. Secondary outcomes included overall and serious morbidity, discharge destination, and length of hospital stay.

RESULTS

Of the 5,142 esophagectomies identified, 70.3% were performed by GS and 29.7% by CTS. Overall, CTS patients had significantly higher comorbidities and cancer rates (61% versus 53%). Both specialties preferred the transthoracic approach (59.41% for CTS versus 44.90% for GS). Trainee involvement was higher for CTS. There was no significant difference in mortality or overall morbidity. Patients operated on by GS had higher rates of wound infection, sepsis, shock, prolonged or unplanned intubation, and a longer hospital stay, whereas patients operated on by CTS had higher chance for bleeding and return to the operating room. Trainees' involvement in esophagectomy was not associated with worse outcome.

CONCLUSIONS

Our study showed that a large number of esophagectomies in the United States are performed by GS, with the transthoracic approach being the most popular among both specialties. Trainees' involvement in esophagectomy did not significantly affect patients' outcomes. However CTS specialty was associated with lower incidence of infection and a shorter hospital stay.