SOMETIMES, hospitalists think they are doing everything right, but things still go wrong. In the realm of perioperative medicine, that kind of problem can develop when anesthesiologists call off surgery for patients they feel are not properly prepared.
At this spring’s Society of Hospital Medicine annual meeting in Las Vegas, Paul Grant, MD, a hospitalist and director of perioperative and consultative medicine at the University of Michigan in Ann Arbor, described several recent cases where just such incidents unfolded. His goal was to figure out what went wrong and determine strategies that hospitalists can use to make sure similar delays don’t happen again.
One common theme, he pointed out, is the need for better communication between the hospitalists clearing patients for surgery and anesthesiologists. Some hospitalists could also use a richer appreciation of what an anesthesiologist’s job entails.”
“In my institution, it is actually mandatory to continue baby aspirin for all patients with a history of a stent.
“They are the ones medically managing the patient during the surgery, in charge of the blood pressure and oxygenation,” Dr. Grant explained. “They have a lot of skin in the game, and our communication with them needs to be better.”
A case of miscommunication
Consider the case of a 96-year-old woman who needed “rather urgent surgery” for a hip fracture. Despite a history of hypertension, asthma, osteoarthritis and limited function, she is in fairly good shape, with no chest pain or shortness of breath. She does have a systolic murmur, but she also has a revised cardiac risk index score of zero. According to the hospitalist, she “is an acceptable risk” for hip fracture surgery.
But when she arrives in the OR later that day, the anesthesiologist cancels the surgery, saying the patient needs an echocardiogram to “assess murmur and inform hemodynamic management during anesthesia.”
So what went wrong? Dr. Grant noted that the hospitalist in this case hadn’t considered the fact that information gleaned from a preop echo could help the anesthesiologist better do her job. That would have been the case even if the ultrasound was “unlikely to have an impact on” the hospitalist’s pre- or postop care.
“Had the hospitalist contacted the anesthesiologist and said, ‘There is a murmur there, but clinically I think this is an important case that we do in the next 24-48 hours and I don’t think we need an echo,’ there may have been a completely different outcome,” Dr. Grant said. “If my anesthesia colleague thinks an echo is going to make a difference, then maybe it’s critical to do, especially given the relative ease of obtaining an echo at most places these days.”
At the University of Michigan, said Dr. Grant, “we are in the process of looking at murmurs and putting together guidance” on when to get preop echocardiograms. “Anesthesiology often wants them before surgery, and despite the ACC/AHA guidelines”—which were posted online by Circulation in August 2014—”rarely advocating for preoperative echoes, I am willing to have a reduced threshold for when to order them.” (See “Which patients with murmurs need echoes?“)
A case of misunderstanding
Dr. Grant presented another example of a frustrating last-minute surgery cancellation. A 62-year-old man with coronary artery disease and hypertension had a drug-eluting stent placed eight months ago after a heart attack. He now has bladder cancer and needs a cystectomy.
After guideline-steered examination and analysis, the hospitalist doing the preop evaluation concludes that the patient is at relatively low risk for a bad outcome from the planned surgery. He directs the patient to stop taking his aspirin and clopidogrel seven days before the operation. But on the day of the surgery, the anesthesiologist calls it off.
When told about the cancellation, the hospitalist pointed to the ACC/AHA guidelines on dual antiplatelet therapy (DAPT), which were published in the September 2016 issue of the Journal of the American College of Cardiology. Those guidelines say to discontinue dual antiplatelet therapy, said Dr. Grant.
Here’s what went wrong: Although the algorithm in the 2016 guideline directly states that such patients should discontinue DAPT during surgery, “that’s not really what this means,” he noted. “What ‘discontinue DAPT’ actually means is stop the P2Y12 inhibitor”—in this case, the patient’s clopidogrel—”but continue the aspirin in these patients, most of who are on aspirin long-term. That shouldn’t change for surgery.”
With the exception of some neurosurgery cases, “most surgeons have become very comfortable operating on patients who have aspirin on board,” Dr. Grant said. “In my institution, it is actually mandatory to continue baby aspirin for all patients with a history of a stent, whether it is a drug-eluting or bare-metal one. And it doesn’t matter if the stent was placed one year ago or 10. Patients stay on aspirin.”
A case of misapprehension
Dr. Grant also presented this case: A middle-aged man with known and treated pulmonary hypertension related to his alcohol-induced liver cirrhosis had to cope with “severe hip pain” for a year longer than necessary. His elective hip replacement surgery had been cancelled at the last minute due to a disagreement between the hospitalist and the anesthesiologist over the risks patients with severe pulmonary hypertension face undergoing total hip arthroplasty.
The hospitalist had concluded that the man was “OK to proceed to surgery despite the rather severe but treated pulmonary hypertension.” To back up that argument, the hospitalist noted that the man, a regular swimmer, had “very good functional capacity” and was being followed by specialists who likewise backed the idea of surgery.
But the anesthesiologist held that the man was “at an unacceptably high risk” for mortality and other serious periop complications, Dr. Grant pointed out. In reality, he added, evidence supports both conclusions.
A large trial published in the November 2010 issue of Anesthesia & Analgesia on mortality in patients with pulmonary hypertension undergoing joint replacement documented a risk of perioperative mortality in such patients of “greater than 10%.” However, an article in the June 1, 2013, issue of European Respiratory Journal found that pulmonary hypertension patients with good functional capacity and no signs of right heart failure who are scheduled for non-emergent surgery—like this particular patient—have a lower mortality risk, of perhaps 3.5%.
“This is, again, all about good communication with your anesthesia colleagues,” Dr. Grant said. If the hospitalist and the anesthesiologist had talked about the case in advance, the last-minute cancellation may have been avoided. That’s particularly true because the anesthesiologist had offered this compromise: “If you do elect to proceed with surgery, an anesthesiologist with expertise in pulmonary hypertension should be used.” The anesthesiologist also recommended using invasive monitoring, central venous access and postop ICU care following the elective case.
What happened instead, Dr. Grant said, was that the patient “became scared,” decided to forego surgery and opted to have steroid injections instead. He came back a year later with “just terrible” pain, got the hip replacement “and had a very good outcome and no complications.”
A case of misinterpretation
Sometimes, what goes wrong occurs after surgery, like misinterpreting causes of postop fevers, said Dr. Grant. He presented the case of an 82-year-old woman with a history of poorly-controlled diabetes, smoking and alcohol abuse who fell and broke her left femur. Two days after surgery, she complained of bad pain and rising temperature.
Thinking about alcohol withdrawal and atelectasis, the hospitalist caring for her postoperatively started benzodiazepines, ordered acetaminophen and an extra dose of oxycodone, and recommended redoubled use of the incentive spirometer.
But the patient got worse. Finally, the doctor removed the surgical dressing to find a foul-looking and -smelling surgical site wound. The patient was started on antibiotics and sent back to the OR for debridement.
The lesson, Dr. Grant said, is this: “Although postop fever is common and most of the time it is self-limiting, you have to see your patient. There are certain things you can’t miss,” like an infection. While you shouldn’t automatically order blood cultures on stable patients who don’t have any symptoms other than a fever on postop day 0 or 1, you do “have to know when to work it up and take immediate action.” That includes the disagreeable task of touching the surgical dressing and looking at the wound. “Myonecrosis of the surgical site is a life-threatening condition that cannot be missed.”
Early postop fever is very typical, he explained, and is usually due to pyrogenic cytokines (such as interleukin 6) released as a result of tissue trauma. But fever due to an infection rarely happens early in a patient’s recovery. Instead, fever due to infection is “much more likely to be present in a patient on postop day 3 or later,” Dr. Grant said. A study in the July 1984 issue of Infection Control found that 80% of patients with fever on postop day 1 had no infection, but 90% of those with a fever on postop day 5 had an identifiable one.
Hospitalists should also remember that, contrary to popular opinion, atelectasis does not cause fever, as explained in an April 2011 systematic review in Chest. That means that encouraging use of the incentive spirometer, while beneficial, won’t help a patient’s fever. Pulmonary embolism—another “can’t miss” postop complication—also “doesn’t usually cause a fever,” he said.
A case of missed (or delayed) diagnosis
To highlight another postop complication, Dr. Grant described what had caused a middle-aged woman to develop a large, right femoral vein acute DVT a couple of days after having an open cholecystectomy. That operation occurred just a few weeks after a previous hospital stay, in which clinicians had initially diagnosed her acute cholecystitis.
During both hospitalizations, the patient had been given prophylactic heparin. When she developed some postop pulmonary emboli, no one had picked up on the fact that her platelet count was also dropping. A day later, when she reported pain, swelling and erythema in her right thigh, the hospitalist finally diagnosed heparin-induced thrombocytopenia with thrombosis (HITT), stopped the heparin and started argatroban.
“This is an uncommon diagnosis, but you should suspect this when platelets drop by 50% or if they are less than 150,000 platelets per microliter of blood,” Dr. Grant said, especially “for patients exposed to heparin products.” Patients taking heparin—unfractionated more than low-molecular weight heparin—less than 30 days before receiving the drug again can see “platelets drop very quickly,” sometimes in hours. In patients with no or remote heparin exposure, on the other hand, “you will see their platelets drop five or 10 days after being exposed to heparin.”
And thrombosis is common, he added: Clots form between 25% and 50% of the time. “HITT should be on your radar if a patient presents with a DVT or a PE.”
But making the diagnosis, said Dr. Grant, “is difficult and is best done in real time. It’s very difficult to do retrospectively, if not impossible.” He recommended that hospitalists use the “4T” scoring system, which assigns points to Thrombocytopenia, Timing of the platelet drop, Thrombosis, and other causes. (Physicians can access 4T calculators online.)
To treat HITT, Dr. Grant said the first step is to stop all heparin products and start an alternative anticoagulant, usually argatroban. After patients’ platelet count recovers, you can then put them on warfarin if indicated for clots.
Deborah Gesensway is a freelance writer who covers U.S. health care from Toronto.
Which patients with murmurs need echoes?
HOSPITALISTS AND ANESTHESIOLOGISTS at the University of Michigan in Ann Arbor are working to fill a guideline gap that increasingly causes frustration around perioperative care. Their “work in progress” is to create an informal, local protocol to help them decide which patients with heart murmurs need preop echocardiograms.
The guidelines are “a little bit fuzzy, and there is variation within the medicine and anesthesia worlds, so we are not always on the same page,” said Paul Grant, MD, a hospitalist and director of perioperative and consultative medicine at the University of Michigan. Dr. Grant spoke about periop medicine during this year’s Society of Hospital Medicine meeting in Las Vegas. “The echo is the one test that anesthesia tends to like because it can help them in the operating room.” But “medicine folks say, ‘I don’t need this to do my assessment of these patients to make sure they are OK for surgery.’ ” Generally speaking, he noted, findings from a preop echocardiogram, “wouldn’t change what we would do.”
Meanwhile, as the availability of echocardiograms continues to improve and the costs come down, the controversy keeps cropping up.
“I think we will see a trend toward more preop echocardiograms than we did 10 years ago,” he said. “That’s probably OK because there really isn’t much of a downside, but we do need to give it some thought.” At the University of Michigan, as at most mid-sized and larger community and academic hospitals, echoes can usually be performed and read in a day or two.
According to Dr. Grant, there are two general groups of patients planning elective or urgent (not emergent) surgery where guidance and agreement would be helpful.
One is patients with unexplained murmurs that previously either were not known or weren’t mentioned in the medical record, especially when those patients have poor functional capacity. The other group is patients with a history of congestive heart failure who have had echoes in the past, but not recently. “How long should it be before we get another echo before moving forward with surgery?” he asked. According to Dr. Grant, within a year would be optimal, but up to two years may be reasonable.
The bottom line, Dr. Grant said, is that hospitalists “should have a low threshold to reach out to anesthesia and communicate with them,” especially if they think a patient about to have surgery could have their cardiac workup delayed, given the urgency of surgery.
“We probably need more verbal communication,” he explained, “even though it can be hard sometimes. Some guidance—maybe a list of the types of patients we think that having an echo before surgery makes sense for—could be helpful.”Published in the July 2017 issue of Today’s Hospitalist
“Had the hospitalist contacted the anesthesiologist” In Case 1, in my opinion, the fault is with the anesthesiologist. I believe the anesthesiologist could have simply paged the hospitalist who did the consult or the one on call to sort out the murmur issue. Couldn’t the anesthesiologist read the echo report if it was part of the system? I’d assume it would be difficult for the hospitalist to know who the anesthesiologist participating in the surgery is and so difficult to page that doctor. The responsibility of communication is the entire team’s and not just the hospitalist’s. However, it’s the hospitalist… Read more »