Without evidence of benefit, an intervention should not be presumed to be beneficial or safe.

- Rogue Medic

Prehospital use of analgesia for suspected extremity fractures


Also posted over at Paramedicine 101 and at Research Blogging. Go check out the rest of the excellent material at both sites.

This is an older study that puts the prehospital pain management problem into a bit of perspective. While prehospital pain management has improved a lot in some places, other places may still be handling pain as described in this study. This is only ten years old. Attitudes are not changed so easily.

The authors looked at what is probably the least controversial type of pain management. If you were to ask medical directors what they feel most comfortable having paramedics use opioids to treat, the only other choice is likelty to be pain due to burns. Chest pain became a bit controversial after the CRUSADE study, but I will get to that in another post.

Over the last decade, pain and its management have received considerable attention. Most notably members of the medical profession in general and specifically emergency medical professionals undertreat pain to a considerable extent.3 [1]

While I would love to be able to defend everyone from this charge euphemistically referred to as undertreatment, the reality is that a decade later, the problem has not changed that much.

This was an observational study involving a retrospective review of all emergency medical services (EMS) runs for suspected extremity fractures[1]

During the study period, all EMS run reports were evaluated by the fire department’s quality assurance coordinator. Only reports documenting the paramedic’s impression that the patient had sustained a fracture, or suspected fracture, of any extremity were included in the database.[1]

There is no mention of any requirement to document any kind of measurement of pain. This seems to be the most significant problem with pain management in the system studied. How do we assess the quality of pain management if we do not assess pain?

The whole structure of this study revolves around the apparent inability to assess pain. There are a bunch of conclusions drawn. Here is what may be the most important omission of the study.

If we do not assess something, how do we treat it appropriately?

If a medical director does not make it clear that pain assessment and management are taken seriously, then is there much reason to expect the paramedics to be more aggressive than the medical director?

We need to provoke medical directors, emergency physicians, emergency nurses, EMTs, and paramedics to take pain seriously.

It’s not my pain.

Akron Fire EMS employs a two-tiered transport system whereby nonurgent patients who may be safely transported in a private vehicle are deemed code 1, allowing the med unit to be put back into service. Nonurgent patients who require ambulance transport to the hospital become code 2 and their care and transport to the hospital are transferred to a private ambulance. A patient requiring immediate transport, medication, or procedures rendered by a paramedic is transported directly to the hospital as a code 3.[1]

The two-tiered structure of this EMS system may unwittingly serve to negatively affect the administration of pain medication in several ways. Administering pain medication to a code 2 patient, for whom transport to the emergency department would be transferred to a private ambulance, would automatically change the run to a code 3. Upgrade to a code 3 would necessitate transport directly to the emergency department by the treating paramedic squad, extending the time required to complete the run.[1]

Another question is whether code 3 means a lights and sirens transport to the ED. What extremity fractures, other than those cutting off circulation, require such rapid and rough transport? Even fractures cutting off circulation are unlikely to benefit from the slight difference in transport time that lights and sirens would provide. Slightly faster, but much rougher and much more painful in spite of the pain medicine! Why?

How did this service do at using pain medicine to manage suspected extremity fractures, which are expected to have a high correlation with severe pain?

A total of 18 patients (1.8%) received treatment for pain; nitrous oxide was administered to 16 patients (1.6%), and morphine sulfate to two patients (0.2%).[1]

We know that 16 patients (out of a thousand patients with suspected extremity fractures) received nitrous oxide and 2 patients (out of a thousand patients with suspected extremity fractures) received morphine.

We do not know if they were being treated for pain, since there is no indication of any assessment of pain. We expect that the patients with suspected extremity fractures would have a lot of pain. If you have ever had a painful extremity fracture, you might expect all extremity fractures to be painful. You might also desire that many, most, or even all of these suspected extremity fractures be treated with pain medicine. The authors do not provide anything to support this conclusion.

Let’s look at the injuries documented.

All of them seem as if they would be painful fractures. Still, we do not know anything about the pain of these patients.

What else was done that might have acted as pain management?

Supportive medical treatment provided included air splints (25% of patients); full immobilization (19%); ice packs (17%); bandages/dressings (16%); and intravenous lines (9.4%).[1]

Air splints may help to relieve pain by positioning the extremity in a less painful, assuming there is any pain, position. Splinting may temporarily increase pain during application.

Full immobilization is unlikely to provide any pain relief. Full immobilization on a solid long spine board is expected to increase pain.[2]

Ice packs can increase pain, decrease pain, or both.

Bandages/dressings might provide some stabilization, or pressure, that decreases the sensation of pain.

Intravenous lines are often painful. If I only suspect an extremity fracture, I would not have any other justification to be starting an IV, except to have a route to deliver IV pain medication. 2 patients received IV pain medication. 94 patients received IV lines. 2/94?

Did one medic start a line, while the other medic contacted medical command. Since The care of each patient is discussed with an online medical control emergency physician in a local emergency department, the superfluous medic may use that time to get online medical control emergency physician contact out of the way, so that the medic can do something useful, such as assessing or treating the patient.

Did the online medical control emergency physician give this counter-order to the medics? Do not follow your standing orders for pain management. Do not give any pain medicine.

Were the medics, or was one of the medics, hesitant to provide any pain medicine without first contacting the online medical control emergency physician?

Are the medics routinely yelled at by emergency physicians if they administer pain medicine without requesting permission first, even though protocols allow them to give pain medicine without asking for permission?

Do administrators receive complaints from emergency physicians when medics follow standing orders for pain medicine?

The number of patients receiving morphine is so small, that I want to know what was so bizarre about these patients that these Just say No! paramedics gave morphine.

Was the morphine given on standing orders?

Was the morphine even given intentionally?

Since giving morphine for suspected extremity fractures is such a freak occurrence in this system, is there any evidence to suggest that these were not 2 medication errors?

Is the occurrence of morphine administration any less rare than the system’s occurrence of medication errors?

The use of morphine is so breathtakingly out of the ordinary in this system, that I do not see any reason to conclude that there is any connection between morphine and suspected extremity fracture. Were any other medications, aside from nitrous oxide, given to any of these patients? Were any of those medications given more frequently than morphine? 2/1,000 suspected extremity fractures.

The EMS pain control policy included standing orders for administration of either morphine sulfate (adult dose: 2–5 mg intravenous push [IVP], may repeat x 1; pediatric dose: 0.1 mg/kg) or nitrous oxide (50%), self-administered. The care of each patient is discussed with an online medical control emergency physician in a local emergency department. By protocol, analgesic therapy is contraindicated in patients with the following conditions: altered level of consciousness; alcohol or drug use; allergies to morphine or nitrous oxide; hypotension; head injury; chest injury with suspected pneumothorax; abdominal pain with possible bowel obstruction; symptomatic asthma or chronic obstructive pulmonary disease (COPD); or respiratory distress.[1]

Even with standing orders, patient care must be discussed with a doctor. In that case, are they really standing orders? There are a lot of contraindications. I almost expect to see suspected extremity fracture listed as a contraindication for morphine. How much different would the results be, if that were the case?

2/1,000 vs 0/1,000.

Is this number, 2/1,000, even close to being statistically significant?

We don’t know how many of the 1,000 patients actually had pain that would be appropriate to treat with morphine.

This study examined the use of analgesia in 1,000 prehospital patients with suspected fractures of the extremities who were treated by paramedics. Of the 1,000 patients, only a very few (1.8%) received any pain medication, although morphine sulfate and nitrous oxide were available to the paramedics by both standing order and direct physician order through online medical control.[1]

I think it is misleading to suggest that there was any encouragement by medical command to treat patients with morphine. However, I have no way of knowing if one, or both, of the patients treated with morphine only received morphine because the doctor ordered it.

My experience with online medical command has been one of repeated refusal to give orders for for pain medicine for patients with pain – pain that I would be authorized to treat on standing orders under my current protocols.

What is the difference?

The patients treated with morphine do not suffer as much. The medical command physician does not get to exercise a medical whim to deny pain medicine purely due to the physician’s lack of understanding of pain management.

The mean time spent on the scene for all patients in the study was 23 ±3.4 minutes. Scene times were significantly longer for patients who received pain medication (n = 18) 32.8 ±17.4 minutes, than for those who did not, 22.8 ±10.4 minutes (95% CI 5.22 to 14.58). Transport times to the area hospitals average 7 minutes in this system, with the three main receiving hospitals located in the same geographic area.[1]

Unless a patient is unstable (or at risk of injury if not moved), there is no good reason to transport the patient until after the pain is managed. It does not matter if this means an extra 5 minutes on scene or an extra hour on scene. More aggressive dosing (morphine 0.1 mg/kg followed by 0.05 mg/kg every 5 minutes until significant relief) and more appropriate medication (fentanyl at appropriate doses) will result in less time on scene. We should not be manipulating painful injuries until after the pain is managed, unless there is some good reason. That is rare.

When I call for orders for more pain medicine, because the standing orders have not been appropriate in managing pain, medical command often wants to know how far I am from the hospital. My response is, That depends on how quickly I get orders for appropriate pain management, because the patient is not being moved until the pain is managed. Why isn’t that obvious to everyone?

Why increase the patient’s pain to move the patient to bring the patient to the pain medicine in the hospital, when the patient can be treated just as safely, if not more safely, before being moved?

Second, the administration of nitrous oxide requires that in addition to directly transporting the patient, the paramedics must also exchange the used nitrous tank for a new one. In Akron, the only tank exchange site was located in a remote part of the city, necessitating extended duties and travel time for one of the paramedics. Upon completion of this study, replacement nitrous tanks were placed in each of the 12 fire houses to facilitate more convenient restocking.[1]

It is good to see that they are trying to make things better for patients by eliminating the excuses used by paramedics, when medics rationalize avoiding treating patients appropriately.

Managing pain in the prehospital setting may require a multifaceted approach. Pain experienced by the patient must be evaluated in an objective manner, and once assessed, managed appropriately. Prehospital care providers should be encouraged to appreciate their patients’ pain and given the tools and affirmation needed to provide the most appropriate care.[1]


Prehospital care providers and their medical control supervisors have room to improve the quality of pain control in the prehospital setting. In this review of the use of analgesia for patients with suspected fractures of the extremities, pain medication was rarely used. Improvements in both the recognition and assessment of pain and in treating the pain in the prehospital setting are slow to be implemented. Education, pain control evaluation, protocol development, and quality assurance and audit systems are all measures that can be used to improve the quality of pain management in the prehospital setting.

All good points, but the most important point is not in there.

Pain management is about treating pain, not treating specific medical conditions.

If you look at all of the contraindications to the use of pain management in this study, there appears to be a strong bias against treating many painful conditions that are not medical contraindications. These appear to be just demonstrations of discomfort with pain management and ignorance of appropriate pain management. As critical as I am of this study, at least the authors are working to improve the way their system manages pain. Most systems seem to deny that there is a problem.

We need to educate prehospital providers to be much more aggressive with pain management.

We spend so much time worrying about paramedics being too aggressive with pain management, but nobody seems to be able to come up with any evidence to support this paranoid fantasy.

We need to provoke medical directors, emergency physicians, emergency nurses, EMTs, and paramedics to take pain seriously.


[1] Prehospital use of analgesia for suspected extremity fractures.
White LJ, Cooper JD, Chambers RM, Gradisek RE.
Prehosp Emerg Care. 2000 Jul-Sep;4(3):205-8.
PMID: 10895913 [PubMed – indexed for MEDLINE]

[2] Unnecessary out-of-hospital use of full spinal immobilization.
McHugh TP, Taylor JP.
Acad Emerg Med. 1998 Mar;5(3):278-80. No abstract available.
PMID: 9523943 [PubMed – indexed for MEDLINE]

Standard backboard immobilization is not harmless and can cause significant pain, especially at the occipital prominence and lumbosacral areas. Within 10 minutes of being placed in FSI, Hamilton and Pons12 showed that volunteers developed moderate to severe pain. After 30 minutes in FSI, Chan et al.13 found 100% of volunteers complained of pain, with 55% of the group grading their pain as moderate to severe in quality. Interestingly, 29% of the subjects developed new symptoms over the course of the next 2 days. Chen et al. concluded that “the standard process of immobilization may complicate the evaluation of the trauma patient by generating additional symptoms . . . leading to unnecessary laboratory tests and radiographic studies, time of immobilization, and ultimately, health care costs.” In addition to pain, FSI can cause changes in pulmonary function. can cause pressure ulcers of the buttocks, scalp, or neck, and can increase the risk of aspiration after vomiting.13,14 Because standard FSI can compromise maternal and fetal circulation, it is relatively contraindicated in gravid women.

12 The efficacy and comfort of full-body vacuum splints for cervical-spine immobilization.
Hamilton RS, Pons PT.
J Emerg Med. 1996 Sep-Oct;14(5):553-9.
PMID: 8933314 [PubMed – indexed for MEDLINE]

13 The effect of spinal immobilization on healthy volunteers.
Chan D, Goldberg R, Tascone A, Harmon S, Chan L.
Ann Emerg Med. 1994 Jan;23(1):48-51.
PMID: 8273958 [PubMed – indexed for MEDLINE]

14 A review of spinal immobilization techniques.
De Lorenzo RA.
J Emerg Med. 1996 Sep-Oct;14(5):603-13. Review.
PMID: 8933323 [PubMed – indexed for MEDLINE]

White, L., Cooper, J., Chambers, R., & Gradisek, R. (2000). P REHOSPITAL U SE OF A NALGESIA FOR S USPECTED E XTREMITY F RACTURES Prehospital Emergency Care, 4 (3), 205-208 DOI: 10.1080/10903120090941209