I have been very critical of plans to have first responders treat people they suspect of having a heroin (or other) opioid overdose with naloxone.
Would first responders be safer with fentanyl?
It is not really the same question, but it does highlight the differences and why I think fentanyl is safer. The patient will be seen by someone more likely to recognize when the treatment is inappropriate. This study looked at IN (IntraNasal) fentanyl given by basic EMTs prior to transport to the ED (Emergency Department).
Previous studies demonstrate adverse effects in 3.3% to 39% of patients treated with intranasal fentanyl,3, 4 and 5 providing an ambiguous safety profile.
The concentration of fentanyl (Instanyl in this study) is different from what I have available. They use 500 µg/ml, while I only have fentanyl in a concentration of 50 µg/ml. Ten times the volume does make measurement easier, but ten times the volume may impair absorption.
The atomizer contains a single dose with a prefixed quantity of either 50 μg (500 μg/mL) or 100 μg (1,000 μg/mL) fentanyl and has a dose volume of 0.1 mL (lower than the 0.15 mL limit necessary to avoid pharyngeal runoff7). The Instanyl preparation contains fentanyl in no other recipients than purified water and a phosphate buffer to match the physiologic environment of the nasal cavity and to increase bioavailability.9 
Patients were not limited to healthy trauma patients, so these results can be generalized to a variety of patients.
We administered 50 μg to patients younger than 18 years, older than 65 years, with chronic obstructive pulmonary disease, or who were considered generally weakened or malnourished by the attending paramedic/EMT. All others received 100 μg. In patients reporting insufficient analgesia, the initial dose could be repeated once or twice after 10 and 25 minutes, respectively.
The smaller decrease in level of pain suggests that they were more cautious in administering fentanyl to the comorbid patients.
The time between doses did lead to some extended scene times (first dose at 0 minutes, second dose at least 10 minutes later, and the third dose at least 35 minutes after the first dose), but that is usually preferable to causing extreme pain by moving the patient with inadequate pain management, regardless of the proximity of the hospital.
What many people fail to realize, doctors included, is that the hospital may only be five minutes away after we are in the ambulance, but we need to manage the pain before we move toward the ambulance. When I call for orders to give more pain medicine than I can give on standing orders, medical command doctors sometimes ask how far away from the hospital we are. I respond that it depends on when the pain is managed. Unless there is some medical condition that requires us to move the patient more quickly, we should move the patient only when the patient feels the pain is managed.
How effective was the intranasal fentanyl at managing pain?
I would prefer to lower the level of pain by more than they did, but I am accustomed to giving IV (IntraVenous) fentanyl, so I am able to titrate it more accurately.
Patients received 1 (n=526), 2 (n=333), or 3 (n=44) doses of fentanyl, with a mean cumulative dose of 114 μg.
Fewer than 5% of patients required more than two doses.
They did give a variety of total doses of fentanyl. The result seemed to be similar regardless of the total dose. This could indicate that fentanyl is just a placebo (unlikely) or that the EMTs did a good job of titrating the medicine to the response.
This was a safety study, so how safe was intranasal fentanyl?
The criterion for hypotension is a bit different from what I am accustomed to. Even using MAP (Mean Arterial Pressure), I have not considered patients to be hypotensive above a MAP of 60.
We calculated the mean arterial pressure (MAP) and defined hypotension as a MAP reduction greater than or equal to 10 mm Hg and an end MAP less than or equal to 70 mm Hg.14 
How much respiratory depression and hypotension did they have?
We did not observe respiratory depression (respiratory rate less than 11 breaths/min), GCS score reduction to 14 in 5 patients was transient, and there was no use of naloxone or mask ventilation. Ten patients (1%) had measurable hypotension; however, none experienced syncope and only 1 experienced dizziness, suggesting that these events were of low clinical importance. Indeed, pain relief may be partially responsible for the decrease in MAP.
Studies repeatedly show that fentanyl can be given safely to hypotensive patients and half of the hypotensive patients were no longer hypotensive after fentanyl was given in one prehospital trauma study. This suggests that a fluid bolus may be less effective than fentanyl at getting rid of hypotension.
As pointed out by O’Donnell et al,20 out-of-hospital undertreatment of pain in pediatric patients may be due to safety concerns. Our study supports the safety of intranasal fentanyl in children.
Fentanyl is even safe in children and safe in adults with comorbidities even when given by basic EMTs.
It seems that fentanyl is safe and much more effective than not treating the pain. Is IN fentanyl more effective than other pain medicines? We still do not know.
 Safety of intranasal fentanyl in the out-of-hospital setting: a prospective observational study.
Karlsen AP, Pedersen DM, Trautner S, Dahl JB, Hansen MS.
Ann Emerg Med. 2014 Jun;63(6):699-703. doi: 10.1016/j.annemergmed.2013.10.025. Epub 2013 Nov 22.
PMID: 24268523 [PubMed – in process]
 Fentanyl in the out-of-hospital setting: variables associated with hypotension and hypoxemia.
Krauss WC, Shah S, Shah S, Thomas SH.
J Emerg Med. 2011 Feb;40(2):182-7. Epub 2009 Mar 27.
PMID: 19327928 [PubMed – in process]
My review of this paper –
Fentanyl in the out-of-hospital setting: variables associated with hypotension and hypoxemia
Fri, 27 May 2011
Krauss, W., Shah, S., Shah, S., & Thomas, S. (2011). Fentanyl in the Out-of-Hospital Setting: Variables Associated with Hypotension and Hypoxemia The Journal of Emergency Medicine, 40 (2), 182-187 DOI: 10.1016/j.jemermed.2009.02.009
Karlsen AP, Pedersen DM, Trautner S, Dahl JB, & Hansen MS (2014). Safety of intranasal fentanyl in the out-of-hospital setting: a prospective observational study. Annals of emergency medicine, 63 (6), 699-703 PMID: 24268523