An Update on Total Knee Replacement (Sort Of)

CMS open door forum reveals clarity of policy still lacking for total knee replacements.

 We are two months into 2018, and the level of clarity regarding how to determine the status of Medicare patients undergoing elective total knee replacements is no better than it was on the day in November 2017 when the Outpatient Prospective Payment System Final Rule was released. Very little useful guidance was added with the Feb. 27 Centers for Medicare & Medicaid Services (CMS) Open Door Forum call.

Before I summarize the new information, we have received in recent weeks, I’ll remind you this only applies to traditional Medicare patients. For every other circumstance, get the procedure pre-certified and get the status from the payer. I’ll also remind you that with those payers, you really should not care about the status; what you should care about is the reimbursement – so check your contracts and fee schedules. If an inpatient total knee replacement from an insurer paid $25,000 and an outpatient paid $30,000, which status would you request?

Why is this still confusing? I was made keenly aware of the first issue we are all facing a week ago. I visited five hospitals in Los Angeles, and at three of them, 95 percent of Medicare total knee replacement patients go home on the second post-op day, passing the second midnight. At the other two, 95 percent of patients go home on the day after the surgery. For these hospitals, each group has to look at the status issue completely differently; the first group can admit most patients as inpatients since they will be staying two midnights, but they must justify keeping the patients those two days. The second group has either figure out how to justify inpatient status for patients with an expectation of under two midnights or decide if the surgery should be performed as outpatient (with inpatient admission only occurring if a second midnight is needed). I would refer readers to an excellent article by Dr. Lisa Banker in ICD10monitor.com for a look at hospitals that have adopted this latter approach.

As I discuss this with staff at hospitals around the country, I see similar patterns. Many hospitals, often those with younger surgeons, have been reducing their lengths of stay, with the majority of patients discharged on the first post-operative day, while other hospitals, especially those with more senior staff, routinely keep patients until the second day. This makes total knee replacement unique in that we cannot state that there is a definitive standard of care for post-operative length of stay – as we can with other surgeries not on the Medicare inpatient-only list, such as laparoscopic cholecystectomy or laminectomy. If the standard is to keep such patients for two days and to admit them as inpatients because of that, are we not financially rewarding less-efficient, less cutting-edge hospitals by allowing this, while the more-efficient hospitals that have worked hard to improve lengths of stay are being financially penalized?

The second issue is the two-midnight rule itself and how to interpret it. Many argue that if most of the patients at a hospital go home on the day after surgery, the default status would be outpatient “because of the two-midnight rule.” In other words, they claim that without a two-midnight expectation or actual stay of two midnights, the requirement for inpatient admission has not been met. But that is not correct; the two-midnight rule has two parts.

They are describing the first part: the two-midnight expectation. But the second part is equally applicable: the two-midnight exception. There are a set of patients whose expected length of stay is under two midnights, but who can be compliantly admitted as inpatients. This includes patients undergoing inpatient-only surgery, patients who unexpectedly require mechanical ventilation, and other patients as determined by the physician on a case-by-case basis, based on risk of an adverse event. And in the applicable Final Rule, CMS specifically states that this case-by-case exception can be used for patients undergoing total knee replacement.

Financially, this is an important decision. One of the hospitals I visited, whose patients generally go home the day after surgery, performs more than 600 surgeries a year, and the difference between inpatient and outpatient reimbursement is about $4,000 per patient – so there is about $2.4 million at stake. I am confident that the chief financial officer of that hospital, and every hospital, wants to be sure that they are capturing all compliant revenue. I will add that with the prospect of joint replacements, cardiac interventions, and electrophysiologic procedures being allowed at ambulatory surgery centers in the next year or two, revenue preservation is as important as ever.

As you may recall, in the Final Rule, CMS indicated that it would not produce guidelines, but instead suggest that professional societies and clinical staff develop guidelines to determine correct patient status, since they have “the specialized knowledge and experience” to do so. Well, that finally happened. First, the American Academy of Orthopedic Surgeons (AAOS) put out a frequently asked questions document on Jan. 31. While it is not a position statement, the group did state that “there is no need to justify why a procedure is not being performed as an outpatient.” In other words, they seem to be saying that the default status is inpatient, no questions asked.

Then the American Association of Hip and Knee Surgeons (AAHKS) issued a formal position statement on Feb. 21. The Association based its position on the fact that CMS stated that the majority of patients will remain inpatient. The AAHKS went on to say that “when a standard status is expected by the overwhelming majority, the burden of proof should fall on the exception, not the standard … (and) all relevant parties agree that the burden of proof is on the surgeon to clearly state, not why this patient requires inpatient designation, but rather what criteria are present that suggest that inpatient resources are not expected to be utilized.”

In other words, the AAHKS position is also that the default status is inpatient – except, perhaps, for the rare Medicare beneficiary who would be categorized as an American Society of Anesthesiologists Class I: a normal, healthy patient who doesn’t smoke, has minimal alcohol use, is not obese, and has no medical conditions.

Likewise, many hospitals have taken the CMS directive to heart, formed interdisciplinary teams, and developed internal guidelines for determining which patients should be admitted as inpatients based on comorbid conditions and increased risk of surgery, regardless of expected length of stay.

So, what did CMS have to say about all this on the Open Door Forum? Unfortunately, they did not give us the answers we all sought. But they did provide some insight into their thinking. First, they stressed several times that the admission decision is the physician’s decision to make, but that there must be documentation to support that decision. Many on the call felt that their emphasis on documenting justification for inpatient admission makes the positions of AAOS and AAHKS unacceptable; inpatient admission as a default status with no justification would not pass scrutiny.

CMS also clearly stated that although it asked professional societies and medical staff to develop guidelines, it would not endorse any guidelines, nor could they be used without supporting evidence documented in the medical record. This does not rule out the use of such guidelines by physicians to guide them to a status decision, but again, the factors considered must be documented in the medical record.

CMS also stated that these patients should be treated like any other patients under the two-midnight rule, and that although the discussion of the removal of the surgery from the inpatient-only list was 20 times longer than the discussion of the removal of laparoscopic prostatectomy, that should not be interpreted as a sign that the surgery will get any special dispensation in reviews.

But that may be a good thing. Rather than saying that the case-by-case exception must be applied only sparingly and only to critically ill patients with a one-midnight expectation, CMS may have been implying that since the implementation of the case-by-case exception in January 2016, we have been underusing it. Most hospitals have reserved it for “sick” patients presenting with diseases such as acute myocardial infarction or complete heart block, wherein treatment could result in a “cure” in one midnight. They hinted that not only would it apply to such cases, but also to patients having a scheduled non-inpatient-only surgery, wherein the patient was at higher risk of perioperative complications. For example, could we admit as an inpatient an 82-year-old female with diabetes, hypertension, depression, and arthritis, who was taking seven prescribed medications, was classified as ASA (American Society of Anesthesiologists) Class III with an increased risk of anesthesia and of surgery, and was undergoing a simple mastectomy, despite the expectation that the patient would go home on the first post-operative day?

CMS also refused to directly address the issue of the patient who requires care at a Skilled Nursing Facility (SNF) after surgery – although they once again implied that if the documentation supports a physician’s decision to admit a patient as an inpatient, with the need for skilled nursing care a factor in the decision, it would be considered.

What can hospitals and providers conclude from the events of the last two weeks? Nothing that occurred has changed my recommendation on how to handle total knee replacements. As discussed earlier, if an orthopedist presents a copy of the AAHKS position statement, I would feel comfortable telling the doctor that CMS has rejected their position.

If a hospital wants to perform all such surgeries as outpatient and only upgrade patients who require a second midnight, I would support their decision to proceed that way. But I continue to feel that if a patient is at a higher surgical risk by virtue of their comorbid conditions (even if most patients at that hospital are discharged on the first post-operative day), he or she can be admitted as an inpatient as long as those comorbid conditions are delineated and the increased risk is described. I’d suggest more documentation specific to the patient rather than the use of smart phrases or templated documentation.

It is easy for me to express an opinion when I don’t have to implement it. But I am convinced that for hospitals with significant revenue at stake, if there is a will, there’s a way. Work with your anesthesiologists to evaluate patients earlier in the process, and get a narrative description of risk from them. Talk to the primary care doctors and enlist their help. Work with your orthopedists and their extenders, who often do much of the documentation.

And finally, take a few moments to consider my argument that we have been too conservative with admitting other patients having non-inpatient-only surgery. CMS unlocked the door for us; perhaps we should be doing more than opening it just a crack.

Comment on this article