SAN FRANCISCO – Patients undergoing colorectal surgery have lower rates of complications and are 20% less likely to die if a fellow is involved in the surgery as compared with a senior resident, new data show.
“Advanced trainee level was associated with improved outcomes in colorectal surgery,” first author Dr. Lilian Chen concluded at the annual clinical congress of the American College of Surgeons, where she presented the data.
“Future studies should attempt to delineate the relative contributions of trainees on intraoperative and postoperative patient care and effects on outcome, and perhaps emphasis on appropriate case selection and resident involvement to optimize patient outcomes,” she recommended.
Using data from the National Surgical Quality Improvement Program (NSQIP), the investigators analyzed outcomes of 68,327 common open and laparoscopic abdominal colorectal operations done during 2005-2012 in which trainees participated.
Outcomes were compared for junior residents in postgraduate years 1 through 3 (involved in 23% of surgeries), senior residents in postgraduate years 4 and 5 (61%), and fellows in postgraduate years beyond year 5 (16%).
The prevalence of many surgical risk factors differed significantly across the three groups of trainees, but most differences were not considered clinically important, according to Dr. Chen, who is a fellow at the Lahey Hospital and Medical Center in Burlington, Mass. “The exceptions were systemic sepsis and emergency cases, where it appears that the fellows are doing proportionately less of these cases,” she noted.
Rates of several 30-day outcomes – septic shock, stroke, unplanned intubation, postoperative pneumonia, wound disruption, acute renal failure, deep incisional surgical site infection, and return to the operating room – differed significantly across groups, being less common for fellows than for either resident group.
The rate of mortality also differed significantly, at 2% for fellows but 4% each for junior and senior residents. In multivariate analysis, patients were significantly less likely to die if a fellow was involved in their surgery as compared with a senior resident (odds ratio, 0.80), but there was no significant difference between junior and senior residents.
“Our limitations include the retrospective analysis, and our inability to adjust for the type of hospital and training program, attending surgeon experience, the presence of subspecialty training in those programs, and the actual level of involvement of the trainee in intraoperative as well as postoperative care of patients,” acknowledged Dr. Chen, who disclosed that she had no relevant conflicts of interest.
Indeed, several session attendees took issue with potential confounders. Dr. Mary T. Hawn, chief of Gastrointestinal Surgery at the University of Alabama at Birmingham, commented, “I think you have to be really careful about understanding the data that you are working with. The first thing is that you can’t adjust for hospital characteristics, so you don’t know that a patient didn’t get transferred to a higher level of care, where there are residents available. ...
“So, I would say you should exclude all emergency cases, you should exclude all patients who are transferred, and things like that, and really try to do propensity adjustment or [seek] some reason that might explain what level of trainee is involved,” she recommended, noting that adjustment does not rule out a role for that factor. “I think this is really concerning because it’s harmful to our community programs to publish data like these, where we don’t really understand the true involvement or the mechanisms of why having the trainee there might be associated with higher mortality.”
Another attendee agreed, saying, “If I’m doing a case with a fellow, I’m going to let the fellow do it, and perhaps assisting, sometimes watching, while I let the fellow do it with the junior resident. If I’m in there with the junior resident, I’m doing the case while the junior resident is assisting me. So I think you need to take things like that into account when you think about what your data actually mean.”
Residents are commonly involved in nighttime emergency cases because the fellow is not at the hospital, a third attendee noted. Additionally, “the fellows will tend to cherry-pick the cases they want to do. The fellow is going to be involved in the cases that are more likely to be quality elective cases. So the diabetic on oxygen who’s coming in for the right colon is potentially going to get handed over to someone else to pick it up. So I would agree, I think that just publishing this and having that as your statement can be very harmful. You really need to be very objective and truthful about owning up to these confounding factors that you can’t really measure.”