The Trainee Effect on Early Postoperative Surgical Outcomes: Reflects the Effect of Resident Involvement or Hospital Capacity to Overcome Complications?^*

^*Commentary of Elucidating Trainee Effect on Outcomes for General, Gynecologic, and Urologic Oncology Procedures

One of the most important challenges of medical education is the society's ongoing need for well-trained physicians. This could only be possible thru remuneration for hands-on learning activities of trainees, but can theoretically harm the individual patient. Patients seeking high quality surgical care are knowingly admitted to teaching hospitals, where some diagnostic and therapeutic interventions are conducted by surgical residents and fellows. Although surgical residents are ideally expected to take part in surgical procedures only after they have satisfactorily improved their operative skills through training box- and simulator-based work, this is almost never the case. Instead, they begin to participate in surgical procedures in operation theatre quite early during their residency. This reality has led to an increase in the number of studies investigating the effect of surgical trainee participation in healthcare services on perioperative patient outcomes.

Based on a very large-scale American College of Surgeons' National Surgical Quality Improvement Program (ACS-NSQIP) dataset, a very recent study published in this issue of the Journal of Investigative Surgery has investigated the effect of the experiences of surgical residents as operating or assisting physicians on early postoperative outcomes in major surgical interventions [1]. The authors concluded that trainee involvement and surgical outcomes had a complex relationship, with seemingly inconsistent increase and decrease in mortality and morbidity rates when residents were involved.

I would like to congratulate the authors on performing such a large-scale study to elucidate the resident effect on surgical outcomes. They tried their best to interpret the statistical results, of which some were not expected to ensue. For example, resident involvement increased morbidity in major abdominal surgeries but decreased mortality in abdominal surgical procedures performed for malignancy. The authors' explanation was that, although residents are likely to make mistakes during surgical procedures, high volume centers that train a higher number of residents have more resources and experience to overcome potential complications, which, in turn, decrease mortality rates. The results of this study seem to be consistent with some of the earlier high quality studies on the same topic, which imply higher morbidity but conversely lower mortality rates in surgical procedures that involve trainees [2, 3]. I believe that the term failure-to-rescue rate, which is described as the rate of mortality among patients who developed complications following surgery, can best explain this phenomenon. Tracking and recording postoperative complications in a busy daily routine is a demanding task, and many hospitals lack the ability to record all postoperative adverse events. However, payers increasingly refuse paying for preventable complications, such as pulmonary embolism and surgical site infections, and tend to tie reimbursements to outcomes. Therefore, it is logical to say that hospitals will need to start recording and reporting more detailed information of their early postoperative outcomes in the near future. This will result in a more in-depth analysis of organization-related factors leading to complications, including hospital volume, surgeon volume, case-mix, workload, logistical support, and trainee involvement. From this standpoint, the present study by the authors of over 150,000 cancer patients provides an insight into an important aspect of the scope.

Trainee involvement in a surgical procedure has been shown in previous studies to lengthen surgical time and hospital stay with conflicting results when including the resident postgraduate year (PGY) level in the analyses [4–6]. The present work demonstrates a linear correlation between resident PGY level and morbidity rate. As the authors suggest, this may have resulted from the fact that senior residents are more likely to attend complex surgical procedures where there are inherent risks for developing complications. Another explanation is that senior residents are allowed to perform some important parts of complex surgical procedures, which decreases the portion of the surgery done by the attending surgeon. The key point relating to this finding, however, is that the majority of relevant studies, including the present one, lack the ability to clearly analyze the degree of resident participation in a surgical procedure. The term “resident involvement” is a general definition and does not give information about to what extent the trainee is in the operating position. The attendance may vary from holding the camera throughout the procedure to performing considerably large portion of the procedure [7]. Future projects should have clear data on this issue in order to interpret statistical results in a more conclusive manner.

The authors of the present study were unable to include the data on whether experience and annual volume of attending surgeons in NSQIP hospitals had an effect on outcomes. As we all know, volume of an attending surgeon for a certain procedure might affect his/her attitude to have the procedure performed by a trainee. A low volume surgeon is less likely to allow a resident to perform considerable portion of the procedure. I believe that looking at the experience and volume of attending surgeons at academic centers and community hospitals would help us understand the complex statistical relationships that emerged in the study, and eliminate potential confounding factors.

Lastly, the present study was based on a NSQIP dataset that included the greatest number of cancer patients analyzed until now in the relevant literature. While this significantly increases the generalizability of the results obtained, it would render the findings more vulnerable to the effects of case-mix. For example, combining data on thyroidectomy and Whipple's procedure increases the number of patients included in the analysis but attenuates homogeneity, as each of these two procedures have completely different perioperative morbidity and mortality features. Some previous papers have relied solely on the data from patients undergoing hysterectomy [5], appendectomy [6], or Roux-en-y gastric bypass [8], which can more precisely determine the effect of surgeon volume or resident involvement on the outcomes.

In summary, using a very large database, the present study aims to elucidate the trainee effect on surgical early postoperative outcomes in patients with general surgical, gynecological, and urological malignant diseases. The study further emphasizes that trainee involvement increases morbidity but decreases mortality, most likely reflecting the capacity of teaching hospitals to overcome potential complications rather than a direct effect of trainee involvement in the surgical procedure. The effect of PGY level on the outcomes does not appear to have a clinically meaningful consequence. The future research is expected to consider the experience of the attending surgeon, and should be investigating whether trainee effect really results from the involvement of trainee in the procedure or reflects high potential of teaching hospitals to decrease mortality when complications occur.

DECLARATION OF INTEREST

The author reports no conflicts of interest. The author alone is responsible for the content and writing of the paper.

Notes

^* Commentary of Elucidating Trainee Effect on Outcomes for General, Gynecologic, and Urologic Oncology Procedures