Displaced Proximal Humerus Fractures – To Operate Or Not?

Sometimes you may look at a fracture on an x-ray and cringe. It just seems intuitive that an orthopaedic surgeon should do something about it.

image

But as with many things in the field of medicine, what seems intuitive is not always right. For quite some time, many of these fractures have been managed without surgery, but there has been a more recent trend toward operation. 

Is it the right thing to do? Many patients with this injury are old, with pre-existing medical problems that increase the risk of surgery. A consortium of hospitals in the UK decided to examine this issue with a multi-center trial. They recruited patients from 33 hospitals that could provide either operative or nonoperative management of these fractures.

Here are the factoids:

  • 1250 patients were recruited, and 1000 were excluded. The patients were predominantly elderly (average age 66), and many had comorbidities, insufficient mental capacity, an associated dislocation or clear indication for surgery, or fracture of the other upper extremity.
  • The remaining 250 patients were equally randomized between operative and nonoperative groups.
  • All patients received surgery or a sling, as well as rehabilitation therapy afterwards.
  • After attrition over the 2 year followup period, 114 patients remained in the surgical group and 117 in the nonsurgical group. Overall, they remained well matched.
  • There was no difference in shoulder function or physical well-being between the 2 groups at any time during the 2 years as measured by the Oxford Shoulder Score of the SF-12 Health Survey and the SF-12. Both are self-reported and were administered by a separate examiner.
  • There were 30 complications in the surgical group vs 23 in the nonsurgical group (not significant). One in each group required later surgery.

Bottom line: This is a decent randomized study of a specific clinical question. The large exclusion numbers are a little bothersome, but the authors believed that they had adequate statistical power with their final number of patients. Interestingly, a Cochrane review from 2012 showed similar results with self-reported functional scores, but found a significant number of the nonop patients went on to require surgery. But note that the Cochrane review was an analysis of 6 separate studies, which may weaken their conclusions a bit.

Ultimately, I think that we don’t have any solid conclusions yet. But given the quality of this study, we should start to seriously question whether patients with this fracture, especially elderly ones, really need operative treatment.

Related posts:

Reference: Surgical vs nonsurgical treatment of adults with displaced fractures of the proximal humerus. JAMA 313(10):1037-1047, 2015.

Are “Routine” Vital Signs Really So Routine?

Trauma professionals, particularly physicians, tend to take vital signs for granted on patients who are admitted to the hospital. And we tend to assume that our patients won’t ask questions, either. Unfortunately, they usually don’t.

“Routine” vital signs tend to get measured by the nurses once a shift. But think about that for a minute. In the US, the typical shifts run from 7 am to 3 pm, 3 pm to 11 pm, and 11 pm to 7 am. This means that at some point in the night, they will be disturbed to take their blood pressure and pulse. At least! And what if they need to have a neuro exam, pulse checks, or to have that beeping pulse oximeter hooked up?

And even though the shift runs from 11 pm to 7 am, does that mean the vitals will be take at the beginning or end of shift? No way! The nurse has to receive report for a safe handoff and get organized at the start of the shift. And how many patients does he or she have? They may not be able to check vitals on everybody until after midnight. And what if vitals are ordered to be taken more than once a shift? How can any patient get decent sleep? 

Bottom line: Once again, think carefully about the orders! It’s no wonder some of our elderly patients sundown when they are admitted to the hospital. How can anyone get a good night’s sleep there? 

Don’t just reflexively write for a frequency. Think about how often your patient really needs to be disturbed, especially at night. If they are recovering uneventfully from an orthopedic procedure, why bother them at all at night? And nurses, make it your responsibility to advocate for your patient and bring up these crazy orders so they can be fixed.

A Simple Tool To Predict The Need To Operate On A Subdural Hematoma

Trauma centers in the US are seeing lots of elderly patients, and falls are a major mechanism in the patient group. A significant number sustain a traumatic brain injury. Extra-axial bleeding is fairly common, but because of the increased space available inside the skull, the patient may not become overtly symptomatic. 

So what objective criteria can be used to determine if evacuation of a subdural hematoma (SDH)is needed? A study from the University of Manchester in the UK sought to figure this out. They speculated that the size of the lesion and the amount of displacement it caused might be objective enough. So they set out to see if any specific numbers would provide a reliable method.

Here are the factoids:

  • Two neurosurgeons reviewed four years of head CT scans and determined if they should be treated surgically or nonsurgically.
  • Measurements of the maximum thickness of the lesion, its volume, and the degree of midline shift were taken.
  • Reasonable attempts were made to ensure inter-rater reliability.
  • The total pool of scans studied was 483. 44% were judged to need surgical management.
  • Maximum SDH thickness of 10mm or more, or a midline shift of 1mm or more were found to accurately predict 100% of surgical lesions.
  • The best predictor of the need for surgery was midline shift.
  • Adding hematoma thickness did not significantly improve the ROC curve.

Bottom line: This study is somewhat limited because it is the experience of only one hospital, and the number of clinicians involved in decision making is small. It does echo other similar studies, but in my opinion it omits the use of the mental status exam.

Using a lesion thickness of 10mm or shift of 1mm does not necessarily mean the patient needs surgery if there mental status is completely normal. But these criteria can certainly identify a subset of patients who are at risk, and should be monitored very carefully for any deterioration. A change in GCS by even a single point should then send them straight to OR.

Related posts:

Reference: A simple tool to identify elderly patients with a surgically important acute subdural haematoma. Injury 46(1):76-79, 2015.

What Is: A Morel-Lavallee Lesion?

Anyone who takes care of blunt trauma has seen the Morel-Lavallee lesion (M-L). Here’s an obvious one because it’s acute:

The M-L lesion is essentially a closed degloving injury in which the skin remains intact. The subcutaneous tissue is sheared off of the underlying fascia, and typically blood accumulates in the potential space that is created. This picture shows a less acute lesion; the bruising and ecchymosis on the surface have resolved. Note the collection on the lateral thigh:

These injuries may take a very long time to resolve and may leave some residual deformity. The definitive management has never been very clear: needle drainage vs incision, timing, compression wraps, etc.

The Mayo Clinic reviewed their 8 year experience with 87 of these lesions to try to shed some light on proper management. They treated their patients in four different ways: needle drainage, incision and drainage, compression wraps, and debridement with vacuum drainage devices. Here are the factoids from their study:

  • Motor vehicle crash was the most common etiology for this lesion, which makes sense due to the energy needed to shear the tissues
  • The most common locations were thigh, hip and flank
  • The incidence of pre-existing conditions that might influence outcome (diabetes, obesity, smoking history, use of anticoagulants) did not seem to influence outcomes
  • Lesion location did not change the recurrence rate (even over joints)
  • Aspiration suffered the highest recurrence rate (56%) vs only 15-19% in the other groups
  • Aspiration of more than 50cc of fluid was more common in lesions that recurred (83%) vs those that did not (33%)

Their experience led them to develop the following practice guideline:

Bottom line: The Morel-Lavallee lesion can be challenging to treat. Although this study has limited numbers, it provides enough guidance to suggest a consistent way of managing it. I recommend adopting this algorithm to provide a standard pathway for dealing with it.

Reference: The Mayo Clinic experience with Morel-Lavallee lesions: establishment of a practice management guideline. J Trauma 76(2):493-497, 2014.

Do We Need All Those Trauma Centers In The US?

There are a lot of trauma centers in the US. Unfortunately, they are not very evenly distributed. An example of this disparity can be found in Washington state. Harborview Medical Center is the only Level I trauma center serving all of Washington, Alaska, Montana, and Idaho. Yet in other metropolitan areas, there can be multiple Level I’s, II’s, and III’s. And in some other areas, new centers seem to be popping up right and left.

Unfortunately, there is such a thing as too many trauma centers. Opening a new center is a zero sum game, however. No more trauma patients will miraculously appear. They will only get redistributed from other centers, decreasing the number of their trauma admissions. Until the next one opens and begins to take patients away from the last new one, as well. Frequently, the “need” for the new center is strictly an economic one for its parent organization, not an actual population need.

The American College of Surgeons Committee on Trauma (ACSCOT) released a position statement on this phenomenon early this year. They promote the following guidelines:

  • Designation responsibility falls to the governmental agency that oversees the regional trauma system. This body needs leadership and statutory authority to enforce reasonable guidelines on how many centers may exist.
  • Trauma professionals must advocate for their patients in educating the lead agency about what the needs really are. The interests of the patients must supersede the interests of the providers and their health care organizations.
  • The designation process should be guided by a concrete regional trauma plan.
  • Needs should be assessed using concrete measures like the number of centers per 100,000 people, population location with respect to these centers, EMS transport times, trauma mortality, and frequency of diversion status.
  • Trauma center allocation should be reassessed on a regular basis.
  • Regional variability must be taken into account.

Bottom line: A super-abundance of trauma centers already exists in several cities around the US (and you know who you are). Unfortunately, the cat is out of the bag, and few if any designating agencies have stepped up to the plate to deal with this. The sad truth is that little will happen until hastily and poorly resourced centers start to close unexpectedly, straining established trauma centers and jeopardizing patient safety. When this crisis finally hits, our state and regional trauma systems will finally seek and wield the authority to designate more intelligently.

Reference: Statement on trauma center designation based upon system need. ACSCOT January 2015.