RSI: The First Two Years

      It is 17:30 on a sunny fall day when a BLS engine from the Wilmington (DE) Fire Department, a BLS ambulance from Wilmington EMS, and a paramedic response unit from New Castle County EMS are dispatched to a "pedestrian struck." All three units arrive at about the same time to find a 15-year-old male face down in the roadway, not moving. The patient responds only to pain, is breathing irregularly and has contusions on his head. The crew immediately takes C-spine precautions and rolls him onto his back. One paramedic easily manipulates his jaw forward while an EMT applies oxygen via non-rebreather mask.

   After he is moved onto the backboard, the patient becomes combative. Several providers attempt to restrain him, but he continues to move his head and strike it against the backboard. His teeth are now clenched and the oxygen mask will not stay on his face.

   The region's level 1 trauma center is Christiana Hospital in Newark, DE. Ground transport in normal traffic conditions is about 15 minutes, but it is now rush hour and could take up to 30 minutes. The paramedics are concerned that the patient does not have a patent airway and his combativeness could worsen a head or neck injury. For these reasons they decide that this patient is a candidate for rapid sequence intubation (RSI), which was implemented three weeks earlier for New Castle County EMS.

   The patient continues to be combative after he is moved to the back of the ambulance. His pulse is 150 and the monitor shows sinus tachycardia, RR 36 and irregular, BP 160 systolic, and a pulse ox reading of 100% with the non-rebreather mask as close as he will allow it to be from his face. One of the paramedics calls Christiana to request orders for RSI while the other starts an IV in the patient's left arm.

   The paramedic requests permission to administer lidocaine, etomidate and succinylcholine to intubate the patient. He also requests orders to administer midazolam and vecuronium after intubation for longer-lasting sedation and paralysis. The base physician agrees and confirms doses for the medications.

   One paramedic draws up the medications and applies labels to the syringes while the other paramedic prepares an 8.0 ET tube with a stylette and syringe, turns on the suction unit, and places an open Thomas Tube Holder and end-tidal CO₂ circuit next to the patient's head.

   The paramedic who is positioned at the head reports that his equipment is ready, and the paramedic at the patient's side administers the medications in the correct order. Cricoid pressure is applied immediately after they are given. The collar is removed and manual C-spine precautions are taken by one of the EMTs. The patient is successfully intubated on the first attempt approximately 45 seconds later, during which his pulse ox reading remains above 96%. The ambulance then starts for the hospital after being on scene for 14 minutes.

   After the run the paramedics are debriefed by their associate medical director. They discuss the decision process leading up to RSI, how the intubation went smoothly and any problems that may have been encountered.

INTRODUCTION

   RSI is routinely used for emergency airway management in hospitals, but there is a great deal of controversy about its safety in the prehospital environment. This article is not intended to advocate for or against the procedure, but instead discuss why EMS and trauma system leaders determined that certain patients could benefit from RSI in one county's EMS system. It also discusses how the program was implemented, the safeguards that were built into the procedure and paramedic airway performance since implementation.

   During rapid sequence intubation, a patient receives a series of medications for sedation and paralysis and has a tube placed in his trachea for airway protection and ventilation control. It is indicated for patients who are unable to protect their own airway or who breathe adequately without assistance but still have some airway reflexes and would not tolerate an oral intubation attempt.

   While definitive airway and respiratory control are lifesaving, achieving this with RSI comes with a number of risks. Patients stop breathing and have no muscle control after they receive medications for paralysis. If they can not be artificially ventilated after, brain damage or death are likely. No procedure in EMS has more potential for catastrophe than RSI.

   Paralysis and intubation is an extremely frightening experience for patients, so they must receive medications for sedation first. If not, they will be aware of everything happening around them while being unable to move or breathe.

DELAWARE'S EMS SYSTEM

   Delaware's EMS system is a tiered response model. Ground EMS is delivered by several fire, private and hospital-based BLS ambulance services and supplemented by county-wide, third-service ALS agencies that respond in chase vehicles. Helicopter transport from scenes is primarily provided by the Delaware State Police staffed with a cross-trained trooper/paramedic. Ground paramedics from the counties frequently accompany patients who are transported by helicopter.

   New Castle County is the most densely populated county in Delaware, with 526,000 people living in 437 square miles. New Castle County EMS responds to 30,000 incidents annually in urban, suburban and rural settings. A BLS ambulance is dispatched to every EMS request. A paramedic unit is dispatched for potentially life-threatening complaints as determined by the Medical Priority Dispatch System or at the request of a BLS unit on scene. This allows paramedics more opportunities to treat critical patients and perform invasive skills. Two paramedics staff each ALS unit, which is supplied with two complete sets of ALS equipment, and two paramedics are dispatched to every ALS call. Run reports for all EMS agencies are completed with a statewide electronic charting program, and data is used to track performance.

THE PARAMEDIC RSI CONTROVERSY

   While RSI is one of the most controversial EMS procedures, some studies have shown that it can be safe and effective with few complications and little increase in scene time. Others appear to show an unacceptably high number of complications and worse patient outcomes.

   A 2004 study by Dr. Henry Wang looked at head-injured patients intubated by paramedics in Pennsylvania. The paramedic-intubated group had a higher rate of mortality and worse neurologic outcomes compared with patients intubated in the emergency department. This was evident even after adjustment for the paramedic group having more severe injuries.¹ The study did not look at the number or duration of intubation attempts by paramedics or how they were ventilated afterward. It did not conclude that paramedic intubation caused the worse outcomes, but did call for more rigorous study of paramedic intubation performance.

   Wang also found that paramedics get little experience intubating. Over a one-year period in Pennsylvania, 67% of paramedics performed two or fewer intubations and 39% performed none.² Another study by Wang showed that more intubation experience on live patients was associated with higher intubation success rates.³ Unfortunately, he also found that access to an operating room to practice intubation is limited for paramedic students. Reasons for this include alternate airway devices being used, competition with students from other disciplines and physicians' fear of litigation.⁴

   Paramedics in Wang's studies were seldom exposed to the procedure, performed it in uncontrolled environments, and got less clinical training than other disciplines. This may explain the recent problems found with paramedic intubation.

   The most controversial evidence against out-of-hospital RSI came from the San Diego Paramedic RSI Study, where RSI was introduced to a large ground-based EMS agency and was intended to be used on patients with severe head injuries. The outcomes of patients who received RSI were then compared to a control group of similar patients who did not receive it by EMS. The results showed that most patients were successfully intubated by EMS, but they experienced a higher mortality rate and fewer "good neurologic outcomes" than the control group.⁵

   Since RSI by paramedics is intended to help head-injured patients, why did this happen? A closer look at the data showed that several patients became bradycardic and hypoxic during the procedure and that several were hyperventilated after. Mortality was higher in cases where these events occurred. Another problem was that several of the intubated patients in the study did not ultimately have a serious head injury. These patients were exposed to a dangerous procedure that they did not need.^6,7,8

   Another possible explanation for the poor outcomes is the lack of continuous wave-form capnography, which provides real-time feedback on a patient's respiratory rate and how much carbon dioxide they exhale. Hyperventilating patients causes them to exhale too much carbon dioxide and appears to worsen head injuries. Capnography helps guide how fast patients should be ventilated and is also the most reliable marker for verification of proper tube placement. It was used in only 59 of the 426 patients enrolled in the study, and fewer episodes of hyperventilation were found in the capnography group.⁹

   Limited training and practice with RSI were noted as problems in the San Diego study as well. Data from the trial showed that each paramedic participated in one RSI case every two years. The lack of practice may have contributed to the poor outcomes.⁸

   Better results have been found in other areas. A study of ground paramedics in Seattle compared the outcomes of head-injured patients who received RSI to those intubated without any medications. The RSI group in this study had a lower mortality rate and higher rate of "good neurologic outcomes."¹⁰

   While paramedics in this study had high intubation success rates and no significant problems with hypoxia or hyperventilation, a different criterion was used than in San Diego. Patients who were intubated with no medications were more likely to be hypotensive and to have had CPR performed. This group had a higher mortality rate but also had more severe injuries to begin with.¹⁰

   Seattle's EMS system is much different from San Diego County's. In Seattle, a small number of paramedics work in an urban environment. Paramedic airway performance is closely monitored and their RSI program has been in place for many years. Each paramedic intubates an average of 12 patients a year, half of which are performed using RSI. Frequent practice and strong oversight may explain the better outcomes in Seattle.⁸

   Another study looked at a flight paramedic RSI program where RSI was attempted on 175 trauma patients over four years, and 96% were successfully intubated. When intubation failed, the airway was managed with a surgical cricothyroidotomy or bag-mask ventilation. The study measured blood gas analysis of RSI patients on ED arrival and scene time when it was performed. This program has a small number of closely monitored paramedics who are frequently exposed to the procedure. With this type of system in place, the authors concluded that prehospital RSI can be safe and effective, has few complications, and there is minimal delay in transport to do the procedure.¹²

   The Brain Trauma Foundation assembled a panel of experts on paramedic use of RSI to interpret the current literature and offer guidance to EMS systems about it. The panel reported that the results of an EMS RSI program depend on training, experience and the characteristics of individual EMS and trauma systems. It should not be implemented in every system, and patients can be harmed if the procedure is done poorly. The group reported that current literature is inconclusive about which patients, if any, will benefit from it.¹¹

WHY RSI IN NEW CASTLE COUNTY, DE?

   So with the current controversy and inconclusive literature about paramedic intubation, why did NCC EMS move forward with an RSI program? The decision came after years of planning, equipment enhancement, study of the perceived need for it, and airway management without paralytics.

   Delaware's other county-based ALS systems had been using RSI for years before New Castle. The average transport time to community hospitals in the other two counties is longer, and transport to the nearest level 1 trauma center is best facilitated by helicopter. The average ground transport time to a level 1 trauma center in New Castle County is 10 minutes and approximately 30 minutes from the most outlying areas. With shorter transport times and access to a trauma center by ground, there was less of a perceived need for RSI in New Castle than in the other counties. But New Castle County recently experienced a population shift. Densely populated suburbs sprawled into the outlying rural areas. More people now live farther from hospitals, and traffic congestion has extended transport times.

   New Castle County's EMS system has similar characteristics to those with successful RSI programs. Advanced airway management is performed by a relatively small number of paramedics and is closely monitored. Each paramedic is required to intubate three patients a year. While most paramedics intubate more often than this, an operating room clinical rotation is available for any who are short the requisite experience level. NCC EMS also works closely with area hospitals and the state trauma system, which support implementing RSI.

   NCC paramedics have a high success rate intubating patients in cardiac arrest, but good documentation was not available on how long it was taking to intubate or how ventilation was managed after. Trending software was added to the cardiac monitors to measure this, and data on the patient's heart and respiratory rate, pulse ox reading and end-tidal CO₂ reading were captured. Periods of desaturation, bradycardia and hyper/hypoventilation could be detected before, during and after intubation. A trend summary must be downloaded to a database any time intubation is attempted, and the chart is flagged for routine quality improvement review. This software was recently upgraded so that the patient's heart rhythm and capnograph for the entire patient contact can be viewed after download.

   Prior to implementing RSI, NCC paramedics used etomidate to facilitate intubation of patients with intact airway reflexes. This skill was performed approximately once per week and appeared to be successful 80% of the time. However, a comprehensive chart review revealed that several patients received etomidate, but no intubation was attempted. The documented reasons for this included a clenched jaw or inadequately relaxed airway reflexes. These cases brought the etomidate-only intubation success rate down to 65%. If paralytics were added, it was believed that the success rate in this group of patients would increase to 80%.

   Because etomidate may worsen a patient's airway status and frequently does not help achieve definitive airway management, a decision had to be made. NCC paramedics would either adopt the full RSI protocol including paralytics or abandon intubation attempts with etomidate alone. Because of the high intubation success rate in cardiac arrest patients and the QA/QI program in place to monitor airway performance, the decision was made to proceed with an RSI protocol.

PREPARING FOR LAUNCH

   Airway equipment carried by paramedics was evaluated in preparation for RSI. The Combitube is used as the primary rescue device after failed intubation. The flex-guide ET tube introducer (also known as the bougie) was added as an additional rescue device, and intubation equipment was also upgraded to disposable fiberoptic, stainless steel laryngoscope blades and handles. The "Event" option on monitors was reconfigured so that time of medication delivery, intubation, ED arrival and patient transfer could be archived.

   All NCC paramedics participated in an eight-hour in-service that reviewed the rationale for RSI, the pharmacology, and the use of rescue airway devices for failed intubation attempts. The training program was based on the National Emergency Airway Management algorithm, which promotes a universal approach to all patients who are candidates for RSI.

RSI INDICATIONS/CONTRAINDICATIONS

   The indications for RSI in the Delaware protocols read: "Patients who require urgent or emergent endotracheal intubation but show evidence of incomplete relaxation, or patients who demonstrate a high probability of airway compromise during transport." It is primarily intended for patients in a coma with clenched teeth from a head injury or stroke, but may be used in other cases as well.

   Absolute contraindications to RSI include:

• Anticipated difficulty ventilating with a bag-valve mask after paralysis
• Entrapped patients with inadequate airway access
• Suspected hyperkalemia
• Degenerative or dystrophic neuromuscular disease (such as ALS or Guillain-Barré disease)

   Relative contraindications to RSI include:

• Severe upper airway trauma
• Stridor or upper airway obstruction
• Morbidly obese patients
• Penetrating eye injuries
• Renal failure
• Small mouth, large tongue or short neck (which predict difficulty).

   While predicted difficulty of intubation is not an absolute contraindication for RSI, concern about the ability to ventilate a patient with a bag-valve mask is. Paramedics must be 100% sure of their ability to ventilate the patient before proceeding in any situation.

   It should be noted that there are no absolute indications for RSI. Some patients who require definitive airway management should not necessarily have it performed in the field. The need for intubation must be balanced by the predicted difficulty and transport time.

REDUNDANCY OF SAFETY

   In his book, The Airway Cam Guide to Intubation and Practical Emergency Airway Management, Dr. Richard Levitan compares RSI to skydiving. He writes that both jumping out of a plane and taking away a patient's respiratory drive have fatal outcomes unless a parachute deploys or the patient is able to be ventilated after paralysis. Skydivers plan to use a primary parachute, but always have a backup available if the primary one fails.

   Following paralysis during RSI, intubation is considered the primary parachute. Just as a skydiver would not jump out of a plane without a working backup parachute readily available, a paramedic should not cause a patient to stop breathing without quick access to working rescue devices in the event intubation fails. NCC EMS has four backup parachutes at their disposal if intubation fails:

1. Ventilation via bag-valve mask with oral and nasal adjuncts
2. Attempted intubation with the flex-guide ET tube introducer (bougie)
3. Insertion of the Combitube
4. Performing a cricothyroidotomy with the Pertrach.

PATIENT PREPARATION

   Once the decision to proceed with RSI has been made, the patient should be preoxygenated before any medications are given. This increases the reserve capacity to maintain adequate oxygenation during paralysis. Preoxygenation is achieved with 100% oxygen delivery via nonrebreather mask for 10 minutes (preferred if the patient is breathing adequately) or eight vital capacity breaths with a bag-valve mask connected to oxygen. Here, basic airway management is to be initiated, employing the jaw thrust or head tilt/chin lift, oral/nasal adjuncts, and suctioning.

   All airway equipment must be prepared and checked before any medications are administered. The suction device should be turned on, checked and attached to a rigid suction catheter. The laryngoscope and ET tube are prepared, and rescue devices are placed within reach of the provider performing the intubation. Syringes are to be labeled immediately after medications are drawn to ensure they are administered in the correct order.

   These steps may seem time-consuming when treating a critical patient, and it is very tempting to take shortcuts. It is stressed that a basic level of airway control must be established before any medications are given, and overlooking any part of the procedure may leave the patient in worse condition than if RSI was not attempted.

THE PROCEDURE

   Two NCC paramedics are required to be with the patient during RSI. The paramedic who is not intubating is generally responsible for obtaining orders from the base physician, then preparing and delivering the medications. Head-injured patients are pretreated with lidocaine (1.5 mg/kg) and children are pretreated with atropine (0.02 mg/kg). Etomidate (20 mg for adults) is given next for sedation, followed by succinylcholine for paralysis (2 mg/kg, max 200 mg for adults).

   Cricoid pressure is to be applied immediately after medications are administered. This reduces the risk of aspiration and improves airway landmark visualization. The person assigned this task assumes the role of "safety officer," and must also watch the patient's pulse ox reading and heart rate. Any intubation attempt should be abandoned if the patient's pulse ox reading drops below 90% or they become bradycardic. The patient should then be ventilated with a bag-valve mask.

   A successful intubation is confirmed with multiple techniques, including visualization of the tube passing airway landmarks, the presence of lung sounds, absence of sounds over the epigastrium and waveform capnography.

POST-INTUBATION MANAGEMENT

   Once intubated, the patient must be monitored to confirm respiratory rate, tidal volume and sedation level. Because worse outcomes have been noted after ventilating a patient too fast and too forcefully, the capnography reading should be continuously monitored and maintained at 35-45 mmHg. If the patient's blood pressure is above 100 systolic, midazolam may be given for longer-lasting sedation and vecuronium may be considered for longer-lasting paralysis. Fentanyl may also be considered for pain.

FAILED INTUBATION

   In Delaware, an intubation attempt is defined as the laryngoscope passing the patient's teeth, even if no ET tube is introduced. Up to four intubation attempts are permitted for each patient and up to two attempts are permitted for each paramedic. In the event of a failed intubation attempt, the patient must be ventilated so the pulse ox reading remains above 90%. In the rare instance that an adequate pulse ox reading cannot be obtained with bagging, the Combitube or Pertrach must be inserted immediately.

OVERSIGHT AND PROBLEMS

   RSI cases are closely monitored by the medical directors in Delaware and the agency's QA/QI officer. For the first 18 months of the program, NCC paramedics involved in an RSI case were required to speak directly with one of the medical directors before going off shift. This is still required if there are any complications or if it is the paramedic's first RSI case. The Delaware Office of EMS closely monitors RSI data, and the continuation of each program remains under review.

   Early in the program there were issues with multiple intubation attempts, ventilation rates and documentation. These were usually discovered during the post-incident debriefing with one of the medical directors. Feedback was then relayed to the division via e-mail and at continuing education sessions. These issues have improved since.

   Only one case in the first two years was judged to have had sub-optimal management. It occurred within the first three months of the program and was utilized as a teaching case for all NCC EMS paramedics. Procedures were clarified, specifically contraindications to RSI, earlier use of the Combitube, and the decision to perform surgical airway in a patient who was not adequately oxygenated or ventilated, were reviewed.

   Initially, paramedics seemed reluctant to use the bougie or Combitube. In a few instances, the medications wore off after initial intubation attempts were unsuccessful. While state protocol allows four intubation attempts for each patient, use of the Combitube was encouraged after fewer attempts.

   There were also problems with documentation. Some trend summaries showed large gaps of low or missing pulse ox readings that were not explained in the chart. On follow-up it was discovered that the probe fell off the patient's finger or did not register. Paramedics were reminded to explain any change in pulse ox reading and the reason for unsuccessful intubation attempts.

   Transient episodes of desaturation have been found during intubation attempts, but there have been no issues with patients becoming bradycardic. This indicates that there were no prolonged, unrecognized periods of hypoxia as found in the San Diego study.

   Through the second quarter of 2009, the NCC EMS intubation success rate with RSI was 89%. This is equal to the non-RSI intubation success rate, most of which is for cardiac arrest patients. For those not intubated, 6% were managed with a Combitube and 5% with a bag-valve mask.

   These numbers meet the program's expectations, even though not all patients who underwent paralysis were intubated. Combitube placement is considered a successfully managed airway if the patient remained oxygenated and ventilated appropriately. Most patients in the BVM-only group had RSI attempted during transport with a single unsuccessful attempt and a very short remaining transport time precluding further attempts or use of a rescue device. After medical director review of each case's trend summary, with the one exception previously discussed, all patients who received RSI medications arrived at the hospital with an adequately managed airway.

CONCLUSION

   After two years of an RSI protocol, NCC EMS has been pleased with the program. Its success depends on a redundancy of safety measures built into the procedure and close attention to details. Close oversight is in place to maintain patient safety and to detect areas for improvement. The NCC EMS RSI program will evolve as more EMS airway studies are published and new equipment is developed. Cases continue to be closely monitored, and protocols may be adjusted based on trends.

References

1. Wang HE, Peitzman AB, Cassidy LD, et al. Out-of-hospital endotracheal intubation and outcome after traumatic brain injury. Ann Emerg Med 44(5):439-450, 2004.

2. Wang HE, Kupas DF, Hostler D, et al. Procedural experience with out-of-hospital endotracheal intubation. Crit Care Med 33(8):1718-1721, 2005.

3. Wang HE, Seitz SR, Hostler D, Yealy DM. Defining the learning curve for paramedic student endotracheal intubation. Prehosp Emerg Care 9(2):156-162, 2005.

4. Johnston BD, Seitz SR, Wang HE. Limited opportunities for paramedic student endotracheal intubation training in the operating room. Acad Emerg Med 13:1051-1055, 2006.

5. Davis DP, Hoyt DB, Ochs M, et al. The effect of paramedic rapid sequence intubation on outcome in patients with severe traumatic brain injury. J Trauma 54(3):444-453, 2003.

6. Davis DP, Hwang JQ, Dunford JV Rate of decline in oxygen saturation at various pulse oximetry values with prehospital rapid sequence intubation. Prehosp Emerg Care 12(1):46-51, Jan-Mar, 2008.

7. Davis DP, Dunford JV, Poste JC, et al. The impact of hypoxia and hyperventilation on outcome after paramedic rapid sequence intubation of severely head-injured patients. J Trauma 57(1):1-8, Jul 2004.

8. Davis DP, Fakhry SM, Wang HE, et al. Paramedic rapid sequence intubation for severe traumatic brain injury: Perspectives from an expert panel. Prehosp Emerg Care 11(1):1-8, Jan-Mar 2007.

9. Davis DP, Dunford JV, Ochs M, et al. The use of quantitative end-tidal capnometry to avoid inadvertent severe hyperventilation in patients with head injury after paramedic rapid sequence intubation. J Trauma 56(4):808-814, Apr 2004.

10. Bulger EM, Copass MK, Sabath DR, et al. The use of neuromuscular blocking agents to facilitate prehospital intubation does not impair outcome after traumatic brain injury. J Trauma 58(4):718-723, Apr 2005; discussion 723-724.

11. Fakhry SM, Scanlon JM, Robinson L, et al. Prehospital rapid sequence intubation for head trauma: Conditions for a successful program. J Trauma 60(5):997-1001, 2006.

   Robert J. Sullivan, BA, NREMT-P, is a paramedic with New Castle County (DE)EMS and teaches with the paramedic program at Delaware Technical and Community College. Contact him via e-mail at Rsullivan@nccde.org.

   Timothy Shiuh, MD, FACEP, is the associate EMS medical director for New Castle County and an attending emergency physician at Christiana Care Health System.