Published in the July 2013 issue of Today’s Hospitalist
Just because you can do something doesn’t mean you should.
That may be judicious advice when it comes to several big perioperative controversies, according to speakers at a presentation on perioperative medicine at this spring’s annual Society of Hospital Medicine conference.
Consider the question of whether hospitalists should routinely order troponin levels for patients at risk of a heart attack following noncardiac surgery. Convincing evidence continues to mount that the biomarker may help identify patients who have experienced asymptomatic, postop myocardial infarctions. But there really aren’t any data to guide doctors on what to do with that information.
“We need to figure out how the information will help the patient before we start screening everyone with troponin levels,” said Leonard Feldman, MD, assistant professor of medicine and pediatrics at Johns Hopkins, director of its general internal medicine comprehensive consultation service, and one of the session’s two presenters. “Just because we can doesn’t mean we should.”
The same may be true, the presenters made clear, about implementing tight glycemic control in terms of preop insulin dosing for patients with diabetes. And doctors may want to take a much more individualized approach than guidelines suggest when it comes to scheduling elective surgeries for patients who have drug-eluting stents.
The speakers also debated a much broader controversy, one that applies to the majority of hospitalist groups. Just because hospitalists can comanage nearly every surgical patient, should they be doing so?
What should you do with increased troponins?
Research indicates that increased troponin levels, which can rise without any ischemic symptoms such as chest pain, identify a population with a high risk of mortality. But what isn’t clear is whether knowing that troponin levels are elevated will help decrease patient morbidity and mortality.
“In theory, there are going to be patients who will benefit,” pointed out Steven L. Cohn, MD, professor of clinical medicine and director of the UHealth Preoperative Assessment Center and medical consultation service at the University of Miami School of Medicine, and the session’s other speaker. “However, the bottom line is that we don’t have evidence yet that interventions,” such as giving patients with increased troponin levels aspirin or statins, sending them to the cath lab for revascularization, or transferring and treating them in critical care units, “will improve outcomes.”
Without that type of guidance, added Dr. Feldman, checking troponins on all periop patients may just be “opening Pandora’s box. How many more patients are we going to send to the cath lab or CCU without knowing if it will help them? Obviously, we need some intervention trials.”
That said, Dr. Cohn noted that he had more enthusiasm than Dr. Feldman about the potential of troponin screening. Even without ample evidence, he recommended that hospitalists consider ordering the test for patients at particularly high risk of perioperative heart attacks. He said that group includes people “with coronary artery disease and multiple risk factors going for vascular or other high-risk surgery.”
Preop insulin dosing
Another controversy debated concerned preop dosing of insulin for diabetic patients. When deciding how much basal insulin to give diabetic patients on the day before and the day of surgery, hospitalists can control patients’ blood sugar tightly. But should they? Or should the risks of causing hypoglycemia lead doctors to err on the side of cutting patients’ dose of basal insulin and chancing hyperglycemia instead?
The question comes down to whether patients should get the full or reduced dose of their usual long-acting insulin the night before surgery such as a hysterectomy. According to Dr. Cohn, doctors don’t have solid evidence to help guide that decision, only expert opinion. Fairly new guidelines from the British National Health Service and the Society of Ambulatory Anesthesia recommend that patients be continued on the same dose unless they have a history of hypoglycemia or severe diet restrictions.
“I would follow the U.K. or anesthesia guidelines, both of which say no change in dosing,” Dr. Cohn said. “That’s assuming the patient is not on too much to start with and has no history of hypoglycemia.”
But Dr. Feldman argued that hospitalists’ primary concern should be to keep patients from becoming hypoglycemic as a result of taking too much insulin. “We don’t want hyperglycemia, but we can deal with it much more easily if it happens,” he said. “We do want to avoid hypoglycemia.”
Giving patients 50% of their regular long-acting insulin dose on the evening before or morning of surgery, he said, will “prevent hyperglycemia without the dreaded consequence of hypoglycemia.”
And there’s another complication that may argue for giving preop patients less insulin: Doctors who assume that patients come in taking the right dose of insulin may be wrong. Patients who are receiving only basal insulin and not a basal-bolus regimen may be getting too much basal insulin. Without a basal dose reduction, this type of regimen could lead to hypoglycemia in a patient who is then made NPO before surgery.
“Most patients are not on the basal-bolus nutritional kind of insulin regimen that many of us recommend,” said Dr. Feldman. “Many of them may be getting more basal insulin than they actually require because patients are also trying to cover the nutritional aspects of their insulin needs.” Given that reality, he said, ordering patients their full dose of insulin glargine before surgery can be worrisome. “I do worry about patients getting a lot more basal insulin than they should,” he added.
Surgery and drug-eluting stents
How soon after receiving a drug-eluting stent (DES) can patients safely have elective surgery? Just because many guidelines have been saying that patients should wait at least a year doesn’t mean that delaying surgery is always the best course.
Drs. Cohn and Feldman presented the case of a 50-yearold obese woman with end-stage renal disease on dialysis who wants laparoscopic bariatric surgery so she can lose enough weight to be eligible for a renal transplant. Does the fact that she had a drug-eluting stent placed eight months ago mean she should wait?
A number of studies are starting to show that the timing of surgery does not necessarily correlate with who does or does not suffer a major cardiac event with major noncardiac surgery. Undergoing surgery between six and 12 months after DES placement may not be riskier than waiting more than 12 months.
Although the current AHA/ACC guidelines recommend delaying noncardiac surgery for at least a full year, some associations, including the American College of Chest Physicians and the European Society of Cardiology, now state that surgery may be reasonable for patients who have had a DES placed between six and 12 months before.
“I don’t think there is any solid evidence to suggest that waiting 12 months rather than eight months is any better,” Dr. Cohn said. “You need to individualize it for each patient.” In the case of the woman wanting bariatric surgery to be eligible for a kidney transplant, he thinks the benefits may outweigh the risks.
“The average wait time for a transplant is four to six years, and the longer she is on the waiting list, the worse she is going to do, even if she gets the transplant,” he said.
“So I would let the patient proceed to surgery.” But Dr. Feldman noted that he would come down on the other side of that question and recommend that the patient delay the surgery for four more months.
While the new studies are encouraging, “there are not a lot of data points to feel confident with the findings,” Dr. Feldman pointed out. “Are you ready to go against the guidelines based on limited data?”
Which patients should you comanage?
Drs. Cohn and Feldman also debated what for many hospitalist groups is a basic bread-and butter issue: comanagement. Just because you can manage perioperative patients, they noted, doesn’t mean that hospitalist comanagement is always the right answer.
“My concern with comanagement is that there is no benefit for straightforward, uncomplicated patients,” Dr. Cohn said. “It increases complexity, and if it is not set up the right way at the beginning, you don’t know who does what.”
That can lead to what he called “confusion and a misallocation of authority, with people not knowing who to call and who is taking care of what. And that presents a potential risk for patients.” Moreover, he added, what happens in some places is “mission creep. The more we take over, the less the surgeons will want to do.”
Another big downside is a “misuse of a scarce resource, which is us, the hospitalists, who can be better utilized elsewhere.”
The only consistent evidence of the benefits of hospitalist comanagement as it relates to hip fracture patients, Dr. Cohn explained, is that it may decrease the average time it takes patients to get to the operating room. However, this may be clinically insignificant, such as reducing that time from about 36 hours to less than 24 hours.
But a dozen or so studies have looked at other outcomes, and results appear to be mixed. Research is split, for instance, on whether comanagement reduces length of stay or improves functional status.
As for mortality and readmission rates, “there doesn’t seem to be any consistent improvement,” said Dr. Cohn. While the evidence does suggest that comanagement may decrease complications, most of those identified were “minor complications.”
Notably, however, there is evidence that another care model does reduce complications in hip fracture patients: using clinical pathways, protocols and standardized order sets. Putting these in place seems to decrease DVTs, pressure ulcers, surgical site infections and UTIs, Dr. Cohn noted.
At the same time, Dr. Feldman said that multiple studies find that medical personnel “especially nurses, but also specialists, surgeons, pharmacists, residents and hospitalists “all like comanagement and feel patients are getting better care because of it.
Nonetheless, “there are opportunity costs to comanagement,” Dr. Feldman said, “and we may be needed elsewhere. When we do comanage, it should be for a high-risk population and not all comers. It’s something we need to think about.”
Deborah Gesensway is a freelance writer who covers U.S. health care from Toronto.