Here’s an interesting case that was published recently (not taken care of at my hospital). EMS was called by a family who had returned home to find their son covered with blood. He had a history of mental illness and the prehospital providers found the young man awake but obtunded (GCS 11: eyes 4, verbal 1, motor 6). He was spontaneously breathing with an intact airway. Vital signs were BP 130/80, P110, R22, Sat 99% on O2.
On exam, his right hand was completely amputated. He was covered with blood, and he had bilateral periorbital hematomas and a 2cm laceration under his right eye. Evaluation of the scene could not determine if this was an assault or self-inflicted. There was a history of mental illness.
Bleeding was easily controlled with direct pressure. An IV was inserted and fluids were given. The amputated hand was dressed with moist gauze and placed in a plastic bag, which in turn was placed in a bag of ice. He was then taken to a hospital for further evaluation. He was rapidly taken to the OR for debridement and reimplantation of the hand.
Here are my questions:
- What happened to the patient?
- Was the prehospital care appropriate?
- Is it likely that the hand can be reattached?
- What other diagnostic tests should be performed, and when?
Comments please, or tweets. Hints over the weekend and answers on Monday!
For decades, the standard of care for irrigation and debridement (I&D) of open fractures has been within 8 hours of injury. There is a growing body of orthopedic literature that says this isn’t necessarily so.
A paper being presented at the AAST meeting in Chicago next week retrospectively looked at their experience with early (<8hrs) vs late I&D in a series of 248 patients. They looked at infection rates stratified by time and upper vs lower extremity.
They found that the infection rates overall were not significantly different. However, when subgrouped by extremity and higher Gustilo type >= III, they noted that both delayed I&D and Gustilo type correlated with infection risk. For the upper extremity, only Gustilo type >= III correlated with a higher infection rate.
The authors concluded that all lower extremity open fractures should be dealt with in the 8 hour time frame, whereas upper extremity fractures can be delayed for lower Gustilo classes.
Bottom line: I don’t necessarily buy into all the results from this small study. The orthopedic literature has already refined this concept. At Regions Hospital, we allow up to 16 hours to I&D for open fractures up to and including Gustilo class IIIA. Above that, the 8 hour rule is followed. We periodically review our registry data on all open fracture patients to make sure that the extended time frame patients are not experiencing an increase in wound complications. And they haven’t in our 8 year experience in handling them this way.
Refresher on the Gustilo classification system:
- Class I – open fracture, clean wound, <1cm laceration
- Class II – clean wound, laceration >1cm with minimal soft tissue damage
- Class IIIA – clean wound, more extensive soft tissue damage or laceration, periosteum intact, minimal contamination
- Class IIIB – extensive soft tissue damage with periosteal stripping or bone damage, significant contamination
- Class IIIC – arterial injury without regard for degree soft tissue injury
Reference: Open extremity fractures: does delay in operative debridement and irrigation impact infection rates? AAST 2011 Annual Meeting, Paper 22.
Here’s something I was completely unaware of until just a few years ago. A number of 9-1-1 calls (quite a few, I am told) are made, not for injury or illness, but because the caller needs help getting back into bed, chair, etc. It is also common that prehospital providers are frequently called back to the same location for the same problem, or a more serious one, within hours or days.
Yet another study from Yale looked at the details of lift-assist calls in one city in Connecticut (population 29,000) during a 6 year period. The town has a fire department based EMS system with both basic and advanced life support, and they respond to 4,000 EMS calls per year.
Some interesting results:
- Average crew time was about 20 minutes
- 10% of cases required additional fire department equipment, either for forced entry or for assistance with bariatric patients
- About 5% of all calls were for lift-assist, involving 535 addresses
- Two thirds of all calls went to one third of those addresses (174 addresses)
- There were 563 return calls to the same address within 30 days (usual age ~ 80)
- Return calls were for another lift-assist (39%), a fall (8%), or an illness (47%)
Bottom line: It looks to me that we are not doing our elderly patients any favors by picking them up and putting them back in their chair/bed. Lift-assist calls are really a sentinel event for someone that is getting sick or who has crossed the threshold from being able to live independently to someone who needs a little more help (assisted living, etc). Prehospital personnel should systematically look at and report the home environment, and communities should automatically involve social services to help ensure the health and well being of the elder. And a second call to the same location should mandate a medical evaluation in an ED before return to the home.
Reference: A descriptive study of the “lift-assist” call. Prehospital Emergency Care, online ahead of print, September 2012.
Pain relief is important for two reasons: it’s the humane thing to do for someone who is suffering, and just as importantly, it assists in the physiologic response to trauma. There are several papers that have shown that prehospital providers may not use pain medications as much as they should. Why would this be?
Researchers at Yale released a paper describing a number of interviews with prehospital providers to get the answers to this question. They did individual and group interviews with five EMS agencies in the states of New Hampshire, Massachusetts and Connecticut. Eight individual and 2 group interviews were conducted, with a total of 15 paramedics in the study.
The results were very interesting and several themes emerged:
- There was a reluctance to give opioids unless objective signs were present (deformity, hypertension)
- There was a preoccupation that patients might be malingering
- Paramedics were not clear on what the pain control target should be (complete relief vs “taking the edge off”)
- Fear of masking symptoms with pain medicine
- Reluctance to use large doses (e.g. using no more than 5mg morphine)
Bottom line: This study is very small, which is a problem. But it also used face to face interviews, so a lot of information was obtained. It’s hard to say if this work is representative of other agencies or countries, but it is thought provoking. My take is this: trauma hurts like hell. Patients really do need the medication. And they are not going to get addicted from a few doses while enroute to the hospital. Whether the cause of their injury was truly accidental or the result of poor choices, it’s not our place to judge because we don’t know the full story. Give pain medication and be generous. You’re not going to make the symptoms go away. But do use judgment to make sure they keep breathing all the way to the emergency department.
I’m very interested in EMS comments about this study. Please comment or tweet!
Reference: Paramedic attitudes regarding prehospital analgesia. Prehospital emergency care; Online ahead of print, Sep 2012.
EMS policy and the trauma center verification process requires that all trauma patients delivered to a trauma center must have a copy of the EMS run sheet. Two parameters that are commonly used to monitor performance improvement (PI) in EMS are:
- accurate record of scene physiology (SBP, HR, RR, GCS)
- request by on-scene BLS for ALS assistance
A study looked at the impact of those criteria on patient survival. A total of 4744 patients from the National Trauma Data Bank were analyzed.
Physiologic data: About 28% had at least one missing physiologic data point, with respiratory rate being most commonly missed. They found that the mortality in the group with missing data was over twice as high (10.3%) as it was in the group with complete date (4.5%).
BLS call for ALS assistance: This assist was called for in 17% of cases. These cases were less likely to involve penetrating injuries and more likely to involve car or motorcycle crashes. Injury Severity Score was the same. Eventual patient mortality was the same for BLS calling ALS and ALS response alone.
So why does failure to record physiologic data translate into higher mortality? The initial response may be that the patient was sicker, and so they needed more intense care during transport with less time to record vitals. However, the researchers controlled for this and found it was not a factor. Other issues that may be a factor are EMS training and proficiency, leadership at the scene and enroute, and available staff and resources, among other things.
The researchers speculate that documentation might be a good global measure of appropriate or inappropriate prehospital care that rolls all of these possible factors into one easily identifiable audit filter. They recommend that this be used to focus performance improvement efforts and hopefully improve survival.
I recommend that the results of this study be taken to heart and used to help persuade EMS programs to get religious about recording complete vital signs and leaving the run sheet at the trauma center every time a patient is delivered. Documentation should be evaluated regularly, and all cases with any missing vital signs should be reviewed closely. Trauma Center PI programs should work with EMS to analyze this data and look for the patterns that increase mortality.
Reference: Lack of Emergency Medical Services documentation is associated with poor patient outcomes: a validation of audit filters for prehospital trauma care. Journal of the American College of Surgeons, 210(2):220-227, 2010.