The majority of
trauma patients presenting with facial fractures are managed without surgery. Dogma in the facial fracture literature indicates that antibiotics should be
administered for some period of time, typically 7-10 days, for fractures that
involve one of the sinuses.
Although this idea
and practice have been around for a long time, there is no good literature out
there to support it. Most medical professionals are now aware of the downside
to giving unnecessary antibiotics, including allergic reactions, C. Diff infections,
and promotion of antibiotic resistance.
A group at Geisinger
Medical Center in Danville, PA, performed a four-year retrospective review of
their registry data involving nonoperatively managed facial fractures. They
stratified their patients into three groups: no antibiotics, brief antibiotics
(1-5 days), and prolonged antibiotics (>5 days). A total of 289 patients
Here are the factoids:
17% received no antibiotics, 22% received a short course, and 61% got them
for more than 5 days (!)
There were no soft tissue infections in any of the groups
There was 1 C. Diff infection, which occurred in the prolonged
antibiotic group. This was not statistically significant.
line: Granted, this was a small, retrospective study. But absolutely no difference
in soft tissue infections was seen, and fear of infection is the usual
justification for the use of antibiotics in these patients. A single case of C.
Diff colitis was noted, and it just happened to occur in the prolonged
antibiotic group. It’s time that we consider abandoning the blanket use of
antibiotics for facial fractures involving the sinuses.
Reference: Utility of
prophylactic antibiotics for non-operative facial fractures. EAST 2016 Poster
We’re getting pretty handy using ultrasound at
the bedside to tell us some interesting things. It started with FAST exams in
the ED. Then we added a few views and came up with the Extended FAST, which was
helpful in showing potential chest pathology.
Ultrasound made its way into other areas of the
hospital, and is now used routinely to place IV lines, arterial lines, and
central venous catheters. I’ve previously written about using ultrasound to
evaluate volume status by imaging the IVC in the abdomen. And now, the group at
Shock Trauma in Baltimore is trying to reach even further.
They are now using IVC variations and cardiac
stroke volume variations to assist in volume assessment in critically ill
patients. These studies have a learning curve, especially the stroke volume
calculations. They performed a study that evaluated another possible window
into the patient’s volume status, the positional internal jugular change.
The diameter of the IJ was evaluated while the patient
was flat, and again when the head was elevated to 90 degrees. A fluid bolus was
given, and the positional change in diameter was measured again. The results
were then correlated with changes in measured stroke volume of at least 10%.
Here are the factoids:
This prospective, observational study involved 159 patients over 1.5
Positional IJ diameter change was much better than IVC diameter changes
(receiver operating characteristic areas of 0.93 vs 0.67)
The authors tried to use the stroke volume variation during passive leg
raise (odd, but doesn’t involve sitting the patient up), and concluded they
could not accurately assess it. This arm of the study was abandoned.
line: Leave it to the folks at Shock Trauma to come up with more weird yet interesting
stuff. This is very preliminary data, and their analysis is ongoing. Any
application of this study will be somewhat limited, since many patients are not
allowed to sit up due to their injuries or baseline hemodynamic status. We will
see where this technique
ends up: in our armamentarium, or in the trash heap.
Go for the
jugular: assessing volume responsiveness in critically ill surgical patients.
EAST 2016 Oral abstract #32.
Most trauma patients
are considered to be at some risk for deep venous thrombosis (DVT) and/or
venous thromboembolism (VTE) during their hospital stay. Trauma professionals
go to great lengths to screen for, prophylax against, and treat these problems.
One of the tougher questions is, how long do we need to worry about it? For
fractures, we know that the risk can persist for months. But what about head
A group at Brigham
and Women’s Hospital did a large database study looking at the VTE risk in adults
who sustained significant head injury, with only minor injuries to other body
regions. They tried to tease out the risk factors using multivariate regression
Here are the
Patients were only included if their AIS Head
was >3, and all other AIS were <3
Of the over 50,000 patients in the study, overall incidence of VTE was 1.3% during the hospital stay, and 2.8% overall
within 1 year of injury
Risk factors for VTE after discharge included age > 64 (3x), discharge to a skilled nursing facility (3x), and prolonged
hospital length of stay (2x)
Incidence of VTE over time
Bottom line: View this paper as a glimpse of a potential unexpected
issue. The risk of VTE persists for quite some time after head injury (and
probably in most other risky injuries like spine and pelvic fractures. The
three risk factors identified seem to identify a group of more seriously
injured patients who do not return to their baseline soon after injury. We may
need to consider a longer period of screening in select patients, but I believe
further work needs to be done to help figure out exactly who they are.
Reference: How long should we fear? Long-term risk of
venous thromboembolism in patients with traumatic brain injury. EAST 2016 Oral
How not to write your abstract! The full title is this:
Trauma Patients on New Oral Anticoagulation Agents Have Lower Mortality Than Warfarin
Now let’s look at what it really says. This was a retrospective trauma registry review from a single Level I trauma center. Over a 14 month period, 275 of 1994 admitted patients were on anticoagulants.
Here are the (misleading) factoids and my comments:
Patients on warfarin had a higher mortality (13%) than those on new oral agents (NOA) (6%). (I can’t duplicate the statistical significance calculation)
Patients taking any anticoagulant were admitted to an ICU more often (44-50% vs 36%). (Duh! This just shows their usual practice, nothing new)
Patients on warfarin were more statistically likely to receive prothrombin complex concentrate. (Double duh! Because it doesn’t work for NOAs?)
The authors pointed out a trend toward more NOA use in this graph. (Really? It goes from 11 to 14 with wide monthly variations!)
Bottom line: This is why it’s so important to read the entire abstract and think about the stats. And ultimately, it’s even more important to read the whole paper! They don’t always say what you think they say!
Reference: Trauma patients on new oral anti-coagulation agents have lower mortality than those on warfarin. EAST 2016 Oral abstract #24.
The old “scoop and run” vs “stay and play” debate has gone on for years. It would seem to be intuitive that trauma patients, who should be assumed to be bleeding to death, would do better with shorter prehospital times and quicker transport to definitive care.
However, several studies have not shown worse outcomes in the “stay and play” patients. Once again, mortality is a very crude indicator of “worse” outcomes, and may not be a good enough measure. Nonetheless, the debate continues to rage. A group at the University of Pittsburgh used the Pennsylvania Trauma Registry to review a huge number of EMS transports, looking at mortality as the measure of interest.
Recognizing that total prehospital time can be influenced by delays in specific phases (response, scene, or transport), they analyzed the impact of problems in each. If one particular phase represented more that 50% of the total prehospital time, it was considered a delay. Logistical regression was used to match patients to try to control for any confounding issues.
Here are the factoids:
Over 164,000 records with prehospital times were reviewed over a 14 year period.
There was a statistically significant increase in mortality if the scene time phase was prolonged.
No differences in mortality were noted with longer response or transport times.
Prolonged extrication and intubation had a tendency to prolong scene time, and were independently associated with higher mortality.
Lengthy scene time without extrication or intubation was not associated with higher mortality.
Bottom line: This registry-based study has helped us to slice and dice the prehospital time issue a little bit better. As with other studies, the times themselves may not necessarily be the problem. It’s what is causing the delay that matters. Extrication and intubation tend to indicate sicker trauma patients, but they are also somewhat unavoidable. Prehospital trauma professionals will need to focus on tools and exercises that save time during these critical interventions.
Reference: Not all prehospital time is equal: influence of scene time on mortality. EAST 2016 Oral abstract #9, resident research competition.
Home of the Trauma Professional's Blog
Do you want to get a daily email every time there’s a new post? See what I’m up to.