Home > Features > Focus on RSI

Focus on RSI

By Cynthia Kelmenson, MD, Josh Salzman, MA, EMT-B, Kent Griffith, RN, EMT-P, Koren Kaye, MD, FACEP, R.J. Frascone, MD, FACEP

A national focus on optimal airway management for prehospital providers is on the rise. Although many feel endotracheal intubation (ETI) is the gold standard of airway management, there is an ongoing discussion about how ETI should be done and whether paramedics should be performing rapid sequence induction (RSI).(1,2)

If successful and properly managed, paramedic RSI has the potential to decrease hypoxia and corresponding hypoxic brain injury associated with acute respiratory distress, as well as facilitate management of trauma patients with decreased Glasgow Coma Score (GCS) who could not otherwise be intubated by the paramedics.

Medications that sedate and/or paralyze a patient can facilitate ETI, and patients could benefit from the earlier intubation by paramedic RSI.(3,4,5,6) In order to be widely accepted, paramedic RSI needs to be done appropriately, safely and with a high rate of success. This involves effective initial and ongoing training; however, the optimal program is not currently known.

There’s scant literature on paramedic intubation training methods and the resultant success rates in practice. During the initiation of paramedic ETI training (without medication assistance), paramedic intubation rates were examined after three training protocols:

1.manikin alone;
2.manikin plus a pig lab; or
3.manikin, pig lab, and operating room (OR) experience.

Although the results suggested improved performance with pig lab and OR experience, the improvement was more strongly correlated to seniority and experience than to training method.7 In a later study, a systematic training program for ETI suggested that manikin-only training was as successful as training with a manikin and human cadavers.8

This study examined the effect of OR intubation experience on RSI success rates for prehospital providers. We were unable to find literature addressing the teaching of RSI to paramedics who are already trained and experienced in ETI.

An RSI training program was developed and delivered to 12 full-time paramedics from a hospital-based, single-role advanced life support (ALS) ambulance provider to suburban and rural communities. The details of this program have been previously reported and included a six-hour classroom didactic session, a simulation session using a manikin, and a practical lab session using adult male sheep.(9)

In addition to these requirements, the participants also completed additional training in the OR where they performed live-human intubations under the supervision of an anesthesiologist. Following completion of all components of the training program, the paramedics were allowed to perform RSI in the field. These paramedics were defined as “OR paramedics.”

An additional set of 16 paramedics from this service were subsequently trained in RSI and completed the same training program as described above—with the exception of the live-human intubation in the OR. These paramedics are defined as “non-OR paramedics.” All of the non-OR and OR paramedics had previous experience in field ETI.

A retrospective review of run reports for RSI performed by these providers between April 1999 and November 2002 was completed. The paramedics who were trained using the protocols described above attempted RSI on 115 patients. An “RSI attempt” was defined as administration of the standard doses of succylcholine and etomidate followed by the passage of the ET tube beyond the patient’s teeth.

The data abstracted from the ambulance run report included the number of RSI attempts and whether each attempt was successful or not successful. “Successful” placement of the ETT was defined as a positive Tube Check B and the presence of bilateral breath sounds on auscultation. The use of an end-tidal carbon dioxide monitor to confirm tube placement was encouraged but not required. Pre- and post-oxygen saturation levels were monitored as part of the protocol but infrequently documented on provider run reports.

The success rates for overall RSI, first attempt and second attempt between OR and non-OR paramedics were compared using a chi-squared test. The first and second attempt success rates within the OR and non-OR paramedic groups were also compared using the chi-squared test to identify a possible learning effect.

Of the 115 patients who underwent RSI, 112 were successfully intubated (97%). In all three instances when intubation was not achieved, the paramedics were able to successfully manage the airway with alternate airway devices (two Combitubes and one bag-valve mask). The OR paramedics performed 90 RSI attempts in 70 patients, with an overall RSI success rate of 74% (67 of 90), compared with the non-OR paramedics who completed 55 RSI attempts on 48 patients with an overall RSI success rate of 82% (45 of 55, p = 0.30).

The first-attempt success rate between OR and non-OR paramedics was not different (76% vs. 82%, p = 0.37). The second attempt success rates did not differ between the two groups (61% versus 75%, p = 0.42). The first and second attempt success rates were examined within OR and non-OR paramedic groups, also resulting in no differences in success rates (p = 0.29, p = 0.74, respectively).

Our results suggested that the non-OR paramedics are performing better than the group that received OR training. Although unexpected, these results should be a stepping stone to further examination of whether OR training is necessary for paramedics to effectively perform RSI. OR training is logistically and financially difficult for many paramedic services, and it has not been conclusively shown to improve success rates. As previous literature suggests, implementing an RSI program with a limited group of specially trained paramedics with intense medical direct involvement appears to be most effective.(1,2,9)

Specifically, we have found the sheep lab to be extremely useful in training paramedics. The airway is similar to the human airway, and the animal lab environment allows for more live scenario experience, critical-thinking skill development and intubation efforts than we can produce in the OR. As we continue to see the explosion in development of sophisticated simulation models and simulation training, it will be important to examine the implementation of these models in a training program and determine what effect they have on RSI success.

The limitations of this study include its retrospective and self-reporting nature, the small number of paramedics in the study sample, the lack of patient-outcome data, the lack of data regarding individual paramedic previous experience, and the lack of recorded SaO2 and EtCO2 values for the intubated patients. Since the trial, the lowest SaO2 and electronic EtCO2 values are both followed as part of the quality assurance (QA) process for all patients intubated under our medical direction.

Further prospective, randomized research with a larger number of patients will be necessary to make conclusive statements as to whether OR training is necessary for paramedic RSI training; however, this pilot study indicates it may not be essential.

Although providing optimal oxygenation, ventilation, and airway protection is clearly the goal of airway management, the means to accomplishing these goals for prehospital providers is not always clear. Previous studies show prehospital RSI success rates greater than 90%.(10,11) One study indicated paramedic RSI was actually harmful for patients with severe traumatic brain injury but did show a marked increase in overall intubation success rates with the introduction of the RSI program.(2) As a result, the authors proposed that RSI be the domain of small, specially trained groups of paramedics. But, as previously stated, there is little data on what this training should entail.

In our study, 112 out of 115 patients underwent successful ETI with the use of RSI medications. The overall success rate was higher than reported in earlier studies and compares favorably with recommendations for an acceptable intubation success rate.(12) However, when examined in the context of first and second RSI attempt success, these results should be interpreted with caution with regard to hypoxia during RSI.(11,13,14)

In this pilot study of otherwise similarly trained paramedics, no significant difference between RSI success rates (overall, first attempt or second attempt) was found based on the presence of OR training. This suggests that a program of paramedic RSI training may be successfully accomplished with didactic, manikin and animal lab sessions without the addition of OR training.

1. Bochicchio GV , Ilahi O , Joshi M , et al. “Endotracheal intubation in the field does not improve outcome in trauma patients who present without an acutely lethal traumatic brain injury.” The Journal of Trauma. 2003;54(2):307–311.
2. Davis DP , Hoyt DB , Ochs M , et al. “The effect of paramedic rapid sequence intubation on outcome in patients with severe traumatic brain injury.” Journal of Trauma . 2003;54(3):444–453.
3. Talucci RC , Shaikh KA , Schwab CW. “Rapid sequence induction with oral endotracheal intubation in the multiply injured patient.” American Surgery. 1988;54(4):185–187.
4. Yamamoto LG , Yim GK , Britten AG. “Rapid sequence anesthesia induction for emergency intubation.” . Pediatric Emergency Care. 1990;6(3):200–213.
5. Li J , Murphy-Lavoie H , Bugas C , et al. “Complications of emergency intubation with and without paralysis.” . American Journal of Emergency Medicine. 1999;17(2):141–143.
6. Lowe L , Sagehorn K , Madsen R. “The effect of a rapid sequence induction protocol on intubation success rate in an air medical program.” . Air Medical Journal. 1998;17(3):101–104.
7. Stewart RD , Paris PM , Pelton GH , et al. “Effect of varied training techniques on field endotracheal intubation success rates.” Annals of Emergency Medicine. 1984;13(11):1032–1036.
8. Stratton SJ , Kane G , Gunter CS , et al. “Prospective study of manikin-only versus manikin and human subject endotracheal intubation training of paramedics.” Annals of Emergency Medicine . 1991;20(12):1314–1318.
9. Kaye K , Frascone RJ , Held T. “Prehospital rapid-sequence intubation: A pilot training program.” Prehospital Emergency Care. 2003;7(2):235–240.
10. Bozeman WP , Kleiner DM , Huggett V. “A comparison of rapid-sequence intubation and etomidate-only intubation in the prehospital air medical setting.” Prehospital Emergency Care. 2006;10(1):8–13.
11. Bulger EM , Copass MK , Maier RV , et al. “An analysis of advanced prehospital airway management.” . Journal of Emergency Medicine. 2002;23(2):183–189.
12. Garza AG , Gratton MC , Coontz D , et al. “Effect of paramedic experience on orotracheal intubation success rates.” Journal of Emergency Medicine . 2003;25(3):251–256.
13. Wang HE , Kupas DF , Paris PM , et al. “Preliminary experience with a prospective, multi-centered evaluation of out-of-hospital endotracheal intubation.” Resuscitation. 2003;58(1):49–58.
14. Doran JV , Tortella BJ , Drivet WJ , et al. “Factors influencing successful intubation in the prehospital setting.” Prehospital Disaster Medicine. 1995;10(4):259–264.

Cynthia Kelmenson, MD, is an emergency department physician at the Medical Clinic of Aurora in Aurora, Colo.

Josh Salzman, MA, EMT-B, & Kent Griffith, RN, EMT-P, are employed with the EMS program in St. Paul, Minn.

Koren Kaye, MD, FACEP, & R.J. Frascone, MD, FACEP, are board-certified emergency physicians and co-medical directors of the EMS program at Regions Hospital in St. Paul, Minn.

This article originally appeared in March 2008 JEMS.


  1. RSI is a valuable tool that must be carefully monitored. Safeguards, such as a dual provider requirement and 100% QA/QI, as well as in-depth medical director participation are needed to ensure quality.

  2. I am curious as to how much experience this group of paramedics have? Is it all 911? Is it Critical Care Transport? Are they working in an urban fast pace system or a rural extended transport system? How many intubations have each of them performed prior to this study? How many of those intubations where performed via sedation and paralysis? These studies are important, but are incomplete to the point that they already tell us what we know.


    Matt King FP-C

  3. With p-values of 0.30 and up, I find none of their data are statistically significant. You can’t really legitimately draw any conclusions from this data set. I still believe that OR time is a valuable thing for a new paramedic before turning them loose with RSI in the tool-kit.

Leave a Reply

Your email address will not be published. Required fields are marked *