Published in the October 2015 issue of Today’s Hospitalist
STUDIES SHOW THAT perioperative comanagement usually can improve patient outcomes and patient and hospital-staff satisfaction. But there are fears “some of them justified “that comanagement programs can increase costs.
According to experts at the full-day perioperative medicine precourse at this spring’s Society of Hospital Medicine annual meeting, that’s particularly the case when it comes to preop labs and other testing.
“It may sound basic, but I don’t think we have preoperative testing nailed down at many institutions,” said Barbara Slawski, MD, MS, a professor of medicine and orthopaedic surgery at the Medical College of Wisconsin in Milwaukee. The result can be not just wasted money and potential delays, but cancelled operations and possible harm to patients from false positives and unnecessary work-ups.
About “60% of periop lab tests could be avoided if they were based on a patient’s indications,” said Dr. Slawski, who also directs a multidisciplinary preoperative clinic and the inpatient consult services at Froedtert Hospital. That’s particularly true, she added, because most abnormalities found during routine complete blood counts, chest X-rays, basic chemistries and ECGs “aren’t acted on, and they don’t change management.”
To truly lower costs and boost quality, comanagement programs need to understand the evidence (or lack of ) behind many of the most common pre- and postoperative tests they offer. Tweet this.
Re-think routine testing
Dr. Slawski cited several recent studies that looked at preoperative testing in patients undergoing low-risk surgeries such as cataract or plastic surgery and other ambulatory procedures. All have shown, she pointed out, that “routine testing makes no difference in outcomes.”
Even when testing may be indicated by something in a patient’s history and physical, many preop tests are unneeded. According to Dr. Slawski, only between 4% and 9% of routine preop tests detect an abnormality, and even fewer change management. With indicated testing, “about 30% of results are abnormal, but only about 3% have management consequences.”
Instead, she recommended that periop programs spell out standard indications for when to order preop tests. She referred hospitalists to an article in the January 2003 issue of Medical Clinics of North America as a good place to start. (See “What preop tests should you order before elective surgery?“)
But one preop test that can reasonably be ordered routinely, said Dr. Slawski, is a pregnancy test for all women of childbearing age.
“Screening pregnancy tests are positive in up to 2% of cases and, according to the American Society of Anesthesiologists, positive results change management in 100% of cases,” she said. Hospitalists should remember that about half of all pregnancies in the U.S. are unintended and that many of these women “don’t know they are pregnant.”
As with lab testing, “the key is to ask yourself how echocardiography results are going to change your management,” explained Kurt Pfeifer, MD, professor of medicine at the Medical College of Wisconsin where he directs the preoperative evaluation clinic at the Froedtert and Medical College Eye Institute.
Dr. Pfeifer added that it’s important to keep in mind what information echoes can and can’t provide. A good example is assessing patients for inducible ischemia.
“Sometimes you see people getting a resting echocardiogram when they are worried about ischemia but don’t want to go all the way and get a stress test,” said Dr. Pfeifer. “Being half sure and doing the wrong test isn’t the right alternative.”
On the other hand, for patients with known or suspected LV systolic dysfunction, an echocardiogram may provide valuable information. The test may help patients and physicians decide whether to go ahead with elective surgery or guide management of IV fluids and appropriate postop care. But if the results won’t change management, ordering an echo may be a waste of time, money and energy.
As for valvular disease, Dr. Pfeifer said, the American College of Cardiology’s appropriate-use criteria suggest that people with mild disease or with a prosthetic valve don’t need an echo more than once every three years. Patients with moderate or severe disease do not require echocardiography prior to surgery if their last test was less than a year ago.
But for patients with known or suspected severe valvular disease and no echo results within the last year, an echo may help the anesthesiologist choose the appropriate anesthesia and the right type of intra- and postop monitoring. Even for patients undergoing urgent surgery, ordering an echo when there is suspected severe aortic stenosis may be warranted because anesthesiologists will change what they do accordingly, such as using intraoperative transesophageal echocardiography or vasopressor drips.
“Anesthesiologists don’t want to do these unless they feel it’s necessary, and they need the information from the echo to know that,” Dr. Pfeifer noted.
His advice? Communicate. If you are on the fence about whether to order an echo, talk to the anesthesiologist. “Avoid the situation where the patient rolls down to the OR and then surgery gets cancelled because someone thinks an echo should have been ordered,” he said.
The same rule applies to stress testing, Dr. Pfeifer said: Don’t order it preoperatively unless it will change your management.
Typically, test results would not change medical management. Even without them, he pointed out, you can determine if the patient should be on beta-blockers, statins or antiplatelet therapy. Stress testing with the intent to perform revascularization if results are abnormal is also misguided. That’s because preoperative coronary revascularization does not improve patients’ surgical risk in most cases.
Here’s another way to think about the decision to order a stress test, Dr. Pfeifer said: Would you order a stress test for these patients if they weren’t having surgery? If the answer is “no,” you should be very cautious about ordering such a test purely for preoperative evaluation.
The most common scenario in which stress testing would be reasonable for asymptomatic patients would be if patients or surgeons were uncertain about proceeding with a nonurgent, higher-risk procedure. In such cases, they would use the results to make a decision.
If a preop stress test is ordered for such a purpose and it comes back as abnormal, hospitalists should first “take a deep breath. A lot of those are false positives,” Dr. Pfeifer said. When referring these patients to cardiology, make it clear that you don’t expect these patients to be rushed to the cath lab and that you would be happy to have them referred back to you for medical management.
There has also been a lot of anticipation about the value of ordering pre- or postop B-type natriuretic peptides (BNP), proBNPs or troponins to help predict which patients are most likely to suffer cardiac complications during noncardiac surgery. But Dr. Pfeifer said the data still aren’t conclusive.
According to the most recent meta-analysis on BNP published in the Jan. 20, 2014, issue of the Journal of the American College of Cardiology, preop BNPs do improve risk prediction. That’s particularly the case when results are combined with clinical risk scoring systems such as a revised cardiac risk index (RCRI).
“But again, the question is: How will this change what we do?” Dr. Pfeifer said. “Clearly all these biomarkers provide some prognostic value, although the troponins are less specific for cardiac complications. But there are still conflicting data on what is the incremental improvement in risk classification. Do these really add to my own assessment?”
Ask yourself that question before you order the test and get the results, Dr. Pfeifer reminded his audience. “Otherwise, you are going to be stuck with abnormal labs and have no idea what to do with them.”
Deborah Gesensway is a freelance writer who covers U.S. health care from Toronto.
WHILE SUCCESSFUL perioperative comanagement programs share several key characteristics, few look alike.
“You cannot take my perioperative program and move it to your institution and expect it to work,” said Barbara Slawski, MD, MS, a professor of medicine and orthopaedic surgery at the Medical College of Wisconsin in Milwaukee, speaking at this year’s Society of Hospital Medicine’s perioperative medicine precourse.
Instead, comanagement programs succeed or fail based on how well they meet the “needs and culture” of their institutions. Nonetheless, Dr. Slawski told hospitalists, comanagement programs should try to meet the following goals:
Create efficiency and savings
“There isn’t an administrator who doesn’t ask why we are paying two doctors to care for the same patient,” explained Dr. Slawski, who directs a multidisciplinary preoperative clinic and the inpatient consult services at Froedtert Hospital. To answer that perennial question, programs need to collect data and feed regular results back to hospital administrators.
In addition to traditional metrics, such as mortality, length of stay, ICU utilization and cost, it can be helpful to develop innovative program metrics.
“We got creative and measured the number of rapid responses called on the surgical floor before and after the comanagement service was implemented,” she said. It can also be helpful to align comanagement outcome measures with other goals of your institution. One initiative at Froedtert Hospital, for example, focused on the appropriate use of blood transfusion resources, so the comanagement team measured whether their service affected blood utilization rates in surgical patients.
“Traditional and unique metrics,” said Dr. Slawski, “can demonstrate the value of a perioperative program.”
Change how you communicate
The expectations that surgeons and other referring physicians have of hospitalist consultants “are often very different than the expectations internists have when we are asking other doctors to consult,” said co-presenter Leonard Feldman, MD, associate professor of internal medicine and pediatrics at Johns Hopkins School of Medicine and director of its general medicine consultation service.
Something that galls many surgeons, he explained, is being wishy-washy when making recommendations or including too many literature citations to back up your plan. “The worst recommendations are the ones that start with ‘Consider, ‘ ” Dr. Feldman said. “Do not use the C word, especially with surgeons. They don’t want to hear it.”
Surgeons also want hospitalists comanaging patients to take a very broad approach to medical issues. Studies “including well-known research published in the Feb. 12, 2007, issue of the Archives of Internal Medicine “have shown that surgeons want medical consultants to address any and all issues a patient may have.
Dr. Feldman also urged hospitalists to be specific. “Don’t say, ‘Start some anti-hypertensives,’ ” he said. “Surgeons want to know what anti-hypertensive at what dose and regimen. They want active involvement and for you to think broadly.”
He also noted that internists tend to be more wary of giving up control of patients when they request a consult as compared to surgeons.
“We want control of our patients, but surgeons say, ‘Just do whatever you think is necessary,’ ” he pointed out.
Surgeons also want medicine consultants to have ongoing involvement with patients, not just make a one-time visit to write a note or set of orders. According to that 2007 study, both nonsurgeons and surgeons want medicine consultants to see patients every day and write a daily note, regardless of patient acuity.
Plan to expand
While successful programs may start with limited scope, they need to have an expansion plan.
“You don’t want to say, ‘I will see hip fractures and that is it,’ ” Dr. Slawski cautioned. “You may find that surgical groups suddenly really appreciate the care you provide and want you to admit and discharge every hip fracture patient in the hospital. Anticipate that upfront, and make a staged plan to accommodate the care you think is reasonable to deliver later.”
Choose partners wisely
Finally, be sure to partner with the right group of surgeons. That means finding not only a group that you can communicate well with, she said, but one that “needs the outcomes you can produce.”
You won’t be able to help a group of surgeons with poor surgical outcomes, Dr. Slawski said. You can help a group whose quality data indicate that it has issues with postoperative MIs in medically complex patients. “Look for the low-hanging fruit in a group that is willing to work with you,” she advised.
How often do common lab tests influence your perioperative management of patients? Here’s a look at data presented at this year’s Society of Hospital Medicine perioperative precourse: