Spine Immobilization in Penetrating Trauma: More Harm Than Good?
Posted Jan 21 2010 6:29pm
A recent study points out some of the problems with EMS (Emergency Medical Services) treatment in some places. Each medical director, or state medical director, is permitted to ignore the evidence that some treatments are harmful. They can use ignorance as an excuse for continuing harmful practices. Rather than ignorance, those familiar with the research will claim that somebody might benefit. Their battle cry is What if . . . ?
If we take that approach, there is no limit to how much we can do to a patient, and I do mean to a patient, not for a patient. This is not patient care. This is alternative medicine. Since alternative medicine is not medicine, but an alternative to medicine, this EMS treatment qualifies as alternative medicine. When our patients need care, they do not need an alternative to medicine that works, they need medicine that works.
Spine immobilization in penetrating trauma: more harm than good? shows that What if . . . ? medicine can double or triple the death rate of our patients.
What if . . . ? we act as if our patients matter enough to be treated as human beings.
What if . . . ? we try to help our patients survive.
The reason for the continued use of What if . . . ? practices seems to be more of a fear of lawyers, than any kind of understanding of medicine. The medical directors appear to presume that they will never get in trouble for doing too much, as long as we are not using pain medicine. They seem to think that anything they recommend is good, or at worst, not harmful. We've got to do something! That is the theme in EMS these days.
We need to continue to harm our patients right up until there is inescapable proof that we are causing harm.
What does the study say?
This study seeks to measure the effect of prehospital spine immobilization on mortality in a large national sample of penetrating trauma patients drawn from the NTDB (National Trauma Data Bank). We hypothesized that penetrating trauma patients who underwent prehospital spine immobilization would have higher mortality than penetrating trauma patients who did not undergo spine immobilization. In addition, we expected that a very small proportion of penetrating trauma patients potentially benefited from prehospital spine immobilization.
Patients were considered to have potentially benefited from prehospital spine immobilization if they had an incomplete spine injury and required an operative spine procedure (including vertebral spine repair, spine fusion, laminectomy, and/or halo placement).
This seems as if it should be the way to determine which patients might have been best treated with immobilization, but there remains a big problem. There is no research to show that without prehospital spinal immobilization, outcomes would be any worse, even for patients with unstable spinal fractures. We presume that this is an effective treatment, but we are only hoping that we are doing the right thing.
On subset analysis of specific patient populations, no group of penetrating trauma patients had any survival benefit with prehospital spine immobilization (Fig. 1) Even for patients with the least severe injuries (ISS <15), spine immobilization was independently associated with significantly decreased survival (OR of death 3.40, 95% CI 1.48–7.81). The OR of death was significantly elevated for GSW patients (OR 2.12; 95% CI 1.33–3.37) and for hypotensive patients (OR of death 2.42, 95% CI 1.37–4.27).
The fatality rate appears to be multiplied, not just increased, by this treatment. Maybe it is time to stop killing so many of our patients. Fortunately, most places stopped this dangerous treatment long ago. This comes from Johns Hopkins. It appears to have been motivated by the continuing attempts by MIEMSS (Maryland Institute for Emergency Medical Services Systems) to stick to the What if . . . ? method of treatment, in spite of evidence of harm. MIEMSS protocols do not appear to differentiate between blunt and penetrating trauma, when determining if immobilization is necessary.
Of these 116 patients, 86 (74%) had complete spinal cord injury and would not have benefitted from spine immobilization. Only 30 (0.01%) of the 30,956 patients had incomplete spinal cord injury and underwent operative spine stabilization. The number needed to treat (NNT) with spine immobilization to potentially benefit one penetrating trauma patient was 1,032. The NNH (Number Needed to Harm) with spine immobilization to potentially contribute to one death was 66.
The NNT is not at all clear. They are using a potential benefit of spinal immobilization that has only been presumed. That benefit has not been demonstrated. Where is the research to show that prehospital spinal immobilization in any way improves outcomes for patients with unstable spinal injuries even due to blunt trauma?
In this case, the NNH is not the Number Needed to Harm. What is reported is the Number Needed to Kill, because they are only looking at fatal harm in calculating NNH. When looking at benefit, if they were to look for lives saved by spinal immobilization for penetrating injuries, they would still be looking. If they had a unicorn to guide them, they might find something.
According to one study of the harm due to spinal immobilization, the NNH is less than 2.
In this population of alert and cooperative patients with no obvious distracting injuries or clinical signs of intoxication, 52% had no complaints of neck pain or back pain yet were transported to the ED using FSI (Full Spinal Immobilization), which increased both their level of discomfort and their EMS charges.
The number needed to treat is 1,032 - assuming there is any benefit from prehospital spinal immobilization. Only 1 study has looked at this. It found no evidence of any benefit. Even those with unstable spinal fractures did worse with spinal immobilization. Unfortunately the study was too small to be statistically significant.
The number needed to harm is less than 2.
The number needed to kill is 66.
The extremely optimistic number needed to treat is 1,032.
Although the intention behind conservative prehospital spine immobilization protocols is to protect the minority of patients who suffer spine injuries, this study demonstrates that spine immobilization is associated with higher mortality in penetrating trauma patients and may harm more penetrating trauma patients than it helps. Prehospital spine immobilization was associated with higher odds of death in all penetrating trauma patients, and this association was qualitatively robust across all subsets of penetrating trauma patients.
The merits of IV ﬂuid administration, endotracheal intubation, and now spine immobilization (in penetrating trauma patients) have been called into question, because their clinical benefit may not be worth the extra time on scene.
This is an excellent example of narrative fallacy.
We know that spinal immobilization leads to worse outcomes for patients with penetrating injuries. That is the part that is important to know. Then there is an attempt at an explanation - because their clinical benefit may not be worth the extra time on scene. This explanation is where we make a mistake. I have written about narrative fallacy here, here, here, here, here, and here.
In the limitations, they do acknowledge this to some extent.
Our conservative estimate of the benefit is possibly exaggerated as not all patients with an incomplete spinal cord injury who underwent surgery truly benefitted from spinal immobilization.
Elsewhere in the limitations, they write this.
This retrospective study suffers some significant limitations, mainly because of the data available. The NTDB does not report prehospital scene or transport times or differentiate urban versus rural care. Thus, we could not demonstrate that the excess mortality in patients who underwent spine immobilization was associated with delays in transport to definitive care.
There is not really a good reason to presume that extra time on scene is the reason for the dramatic increase in death among those immobilized. While it is possible that time does contribute to the result, it is a mistake to claim that a study that does not have the ability to examine prehospital times at all is capable of providing evidence that extra time on scene is the cause.
Another recent study showed that there is no reason to believe that prehospital times significantly affect outcomes even for the most unstable trauma patients.
Where research is not being used, we need to find ways to get the medical directors to understand research. Then we need to get them to apply the research.
^1Spine immobilization in penetrating trauma: more harm than good? Haut ER, Kalish BT, Efron DT, Haider AH, Stevens KA, Kieninger AN, Cornwell EE 3rd, Chang DC. J Trauma. 2010 Jan;68(1):115-20; discussion 120-1. PMID: 20065766 [PubMed - in process]
^2Spine immobilization in penetrating trauma: more harm than good? Haut ER, Kalish BT, Efron DT, Haider AH, Stevens KA, Kieninger AN, Cornwell EE 3rd, Chang DC. J Trauma. 2010 Jan;68(1):115-20; discussion 120-1. PMID: 20065766 [PubMed - in process]
^7Emergency Medical Services Intervals and Survival in Trauma: Assessment of the "Golden Hour" in a North American Prospective Cohort. Newgard CD, Schmicker RH, Hedges JR, Trickett JP, Davis DP, Bulger EM, Aufderheide TP, Minei JP, Hata JS, Gubler KD, Brown TB, Yelle JD, Bardarson B, Nichol G; Resuscitation Outcomes Consortium Investigators. Ann Emerg Med. 2009 Sep 22. [Epub ahead of print] PMID: 19783323 [PubMed - as supplied by publisher]