Airway & Respiratory, Columns, Documentation & Patient Care Reporting, Patient Care

Good Reasons Not To Perform an RSI (And Why I Am No Fan of Naloxone)

Local EMS crews responded emergently at 3:10 p.m. to an approximately 65-year-old female. On arrival, the patient was comatose. She was found in her bed with a variety of differently shaped pills surrounding her. The scene was located approximately 12 minutes from the destination hospital.

As the crew evaluated the patient, the patient’s husband advised them that she had had back surgery less than one week before the call. In fact, she had just been discharged from the hospital the day prior. The patient had filled her prescription for acetaminophen/hydrocodone tablets. The morning of the call, the patient ate breakfast and then took a nap at about 10:30 a.m. At 3:05 p.m., the husband was unable to waken the patient. He noted that she was not wearing her continuous positive airway pressure (CPAP) breathing device as she was supposed to when sleeping. He also discovered that the patient appeared blue in color and was snoring deeply.

The past medical history was positive for chronic back pain, depression, hypertension, sleep apnea, hypothyroidism, obesity, and the recent back surgery.

Medications included verapamil, synthroid, paroxetine (Paxil), bupropion (Wellbutrin), methocarbamol (Robaxin), gabapentin (Neurontin), tegaserod (Zelnorm), and hydromorphone in addition to the recently prescribed acetaminophen/hydrocodone. She was also receiving injections of enoxaparin (Lovenox) for reasons that were unclear possibly as prevention for post-operative deep vein thrombosis.

Physical assessment at the scene revealed an unconscious woman appearing consistent with her age. She was obese with an estimated weight of 80 kg and a height of 5’3″. There was no response to verbal stimuli and minimal response to pain. The pulse rate was 120 and regular, blood pressure 134/86, and the respirations estimated to be between six to 12 per minute and irregular. She was extremely cyanotic with snoring respirations.

The pupils were pinpoint. The Patient Care Report (PCR) describes the woman as “having no neck” along with jugular venous distension. The oral examination suggested Mallampati four findings (see figure 1). She was also noted to have trismus. Heart tones were distant and the rate tachycardic, but regular. Lung sounds demonstrated rhonchi throughout. Bruising was noted on the abdomen consistent with enoxaparin injections. Distal pulses were strong and equal in all four extremities. Bruising was also seen extensively over the extremities, thought to be due to recent IV line insertions in during the hospitalization.

Data included a finger stick glucose of 249 mg/dL and an initial pulse oximetry of 62% on room air. Monitor strip demonstrated sinus tachycardia at a rate of 130 with normal appearing QRS complexes (see figure 2). Nasal determination of end-tidal carbon dioxide (EtCO2) was not attempted.

The patient was initially treated with bag-valve-mask (BVM) ventilation and 100% oxygen. Shortly after this, a nasal pharyngeal airway (NPA) was placed. The combination of the NPA with BVM resulted in improved ventilation and oxygenation, evidenced by chest rise and less cyanosis. Pulse oximetry was reported to be 97% during these maneuvers.

Intravenous (IV) access proved to be very difficult. Several attempts failed. Finally, an external jugular line was placed.

The patient was then given 1 mg of naloxone (Narcan) intravenously. In response, she began flailing her arms and head. Shortly after this, the woman began vomiting copiously. The trismus continued, making oral suctioning impossible. The patient was rolled onto her side during this episode and suctioning around the teeth continued.

A nasoendotracheal tube (NETT) was considered, but because of the bruising and the history of enoxaparin, there were great concerns about the potential development of uncontrolled epistaxis associated with a NETT attempt. Therefore, a NETT was not utilized.

The RSI paramedic on the call then elected to perform a Rapid Sequence Induction or Intubation (RSI) to “clear the patient’s airway of emesis and stop the patient from aspirating any further emesis.” He was aware the woman presented with an extremely difficult airway and that there was a good probability that he would not be able to orally intubate the patient. He felt that, with the elimination of trismus after the administration of the paralytic medication he would be much better able to suction the mouth and pharynx, thereby removing the vomitus. He also reasoned that he could likely place a CombiTube in the event of an unsuccessful intubation.

The RSI was carried out although the crew was still on scene at the patient’s home. The patient was given 16 mg of etomidate followed by 160 mg of succinylcholine. After paralysis was achieved, two oral intubation attempts failed. During the second attempt, the oxygen saturation fell to as low as 40%, although it was not clear exactly why. Whether aspiration of emesis was the cause of the desaturation or some other explanation was responsible was unknown. In any event, when BVM assistance was resumed, the oxygen saturation promptly returned to over 90%.

At this point a CombiTube was successfully placed into the esophagus. Symmetrical chest rise was observed, and bilateral breath sounds were auscultated although bagging through tube one of the CombiTube. Initial EtCO2 confirmation was obtained via a colorimetric device. Subsequently, an EtCO2 capnograph was added to the circuit with an excellent waveform noted (see figure 2). The initial peak EtCO2 reading was 70. Subsequent peak EtCO2 levels, after continued ventilations via the CombiTube, remained between 40 and 45. Oxygen saturations were 97% or higher after placement.

Suctioning of the oropharynx continued. Suctioning of the esophagus and stomach commenced via tube two of the CombiTube with good results. The range of vital signs from this point and throughout transport remained as follows:

  • Blood pressure: 102 108 mmHg (systolic); 64-78 (diastolic);
  • Pulse: 116 130; and
  • Respirations: 16 18.

At this point, the patient began biting down on the CombiTube prior to moving her to the ambulance. The crew attempted to sedate her with diazepam they had with them in the home, but it was somehow spilled before administration. As an alternative, they gave the patient another dose of etomidate (8 mg). When they got to the ambulance, another vial of diazepam was located and this was administered to the woman, with good success in reduction of agitation.

The patient continued to be well ventilated and oxygenated after arrival in the emergency department (ED). The CombiTube continued in place for several hours in the ED due to the difficult airway features exhibited by the patient. At some point later in the intensive care unit, the patient was successfully intubated by an anesthesiologist. The woman remained hospitalized for two weeks, but was discharged at that time completely neurologically intact. The diagnoses were, in part, acute drug overdose complicated by vomiting and aspiration.

Napoleon is reported to have said, “I would rather have a lucky general than a good general.” I would modify this statement slightly to read, “I would rather have a good general who is lucky.” Or alternatively, “Good generals breed good luck.”

This case illustrates my versions of Napoleon’s comment. This particular RSI paramedic was an excellent practitioner and highly trained. Nevertheless, he encountered serious problems with this patient and was very lucky the patient did not die or suffer devastating brain damage. Luck should never be substituted for high quality training, skill and judgment. But sometimes, if these educational goals are met, good luck may follow.

RSI in the ground ambulance setting was first described in medical literature in the late 1980s and early 1990s. It’s defined by the use of a paralytic drug (most commonly succinylcholine) to achieve a brief period of paralysis. During paralysis, the muscles of the face and mouth are completely immobile and the gag reflex is lost. The result is, theoretically, a much improved chance of successful oral intubation of the trachea. Overall expected success or tracheal tube placement with RSI, based on published reports, is between 92 95%.

Although actual results in the field may vary, our experience in Colorado Springs with RSI parallels the success rate described in the literature. Consequently, there is no doubt in my mind that RSI can be successfully performed by ground paramedics provided that there is very stringent training and medical oversight.

What continues to be a source of great debate nationally is patient outcome when field RSI is utilized. In my opinion, the jury is still out on this very critical question.

The indications for field RSI may vary somewhat between agencies. Typically, acute respiratory distress (not resolving with conservative treatment), head injury (or other cause of critical intracranial pathology), accompanied by combativeness and/or trismus, certain severe drug overdoses and status seizures constitute good reasons to consider RSI.

In this case, although the patient did suffer a drug overdose, the stated indication for the procedure was to “clear the patient’s airway of emesis and stop the patient from aspirating any further emesis.” That, in and of itself, is not an acceptable indication for RSI.

Once an appropriate indication has been established, patients must be very carefully selected prior to commencement of the procedure in order to achieve correct tracheal tube placement 92 95% of the time. In other words, unlike the cardiac arrest patient, RSI intubation is a semi-emergent intervention. Usually other options are available to the provider. The best analogy to a potential candidate for RSI is a patient who might be nasally intubated.

So, in a patient who is not a “good selection candidate” for RSI, some other form of airway management would be expected to be safer thus negating any thought of RSI and oral intubation attempts. Accordingly, the bulk of our agency’s training focuses on the identification of selection features with the goal of avoiding RSI in patients not possessing good selection criteria.

We do not have enough space in this column to review all of the factors that might lead to “selecting out” potential RSI candidates, but I have developed three rules that summarize important selection characteristics in hopes of making them easier for crews to remember.

The three rules are really questions that a paramedic should be asking regarding a patient undergoing selection consideration. A “no” answer to any one of the three questions should exclude a patient from the procedure. The rules (or questions) are:

  1. Can I get a good seal with the BVM?
  2. Is the airway patent?
  3. Do I think I can intubate this patient?

In regard to this specific case, two factors were noted in the PCR that had great potential to impact Rule #3. One of these was the comment that there “was no neck.” This may or may not have been truly the case, and by itself may not have eliminated the RSI attempt. But the paramedic should expect the likelihood of a difficult intubation. The second, and most critical, factor was the comment that the patient was a “Mallampati 4” (see figure 1). The Mallampati classification estimates the distance between the top of the tongue and either the palate or the uvula, as well as provides information about mouth opening. A Mallampati Class 3 or 4 would suggest a difficult, if not impossible, oral intubation. In fact, at our agency, the presence of a Mallampati Class 4 directly impacts Rule #3 and is an absolute contraindication to RSI.

Additionally, the patient is reported to be obese and the height and weight noted in the PCR would support that impression. Obesity alone is not a risk for a difficult airway, but it should suggest to the paramedic the potential presence of a short neck and other associated limitations on selection.

In any event, RSI should never be undertaken with the thought that “if I can’t place the endotracheal tube, I can always drop a CombiTube or a Laryngeal Mask Airway and call it good.” The single goal of an RSI is to safely place a tube in the trachea (the definitive airway).

Despite the reasons strongly favoring withholding RSI, the procedure was carried out. The team was lucky that the paramedic was able to insert the CombiTube as a bail-out airway and very effectively oxygenate and ventilate the patient with this modality and that the patient ultimately survived.

Finally, the agent that likely caused the significant vomiting and associated narcotic withdrawal leading to the paramedic’s quandary over an RSI to deal with emesis was naloxone. It seems likely that although the patient may have continued to have trismus, she could have been managed successfully with a BVM, at least prior to the onset of profuse vomiting.

The effects of naloxone in chronic narcotic users have been widely underestimated for years by both physicians and EMS crews. Rather than being on the way out, its use seems to be on the rise. Recently, some in our state have advocated nasal atomization of the drug. It’s a knee-jerk medication given to many patients with altered mental status of any cause. It’s commonly given because it doesn’t require much thought and “it’s something we can do for the comatose patient.”

Although it may seem innocuous, naloxone is in fact a dangerous drug when administered to the wrong patient, specifically one who is chronically narcotic dependent. Precipitous narcotic withdrawal can lead to a number of complications and risks for both patients and providers. Why convert a somnolent patient who can be easily managed with a BVM into a violent, combative, flailing mess, although increasing the chances of emesis and other problems? This rationale makes no sense to me.

In my opinion, the only current indication for naloxone is in the narcotic drug overdosed patient with severe, persistent hypotension (shock) that cannot be managed with fluid boluses. Most EMS providers from the basic level up have the capability to manage the obtunded, overdose patient who is hypoventilating without resorting to naloxone. Needless to say, in my world, naloxone use would be almost nonexistent.

In summary, what can we in EMS learn from this case?

  • This is a scary, sphincter-tightening situation. (If your sphincters did not respond appropriately check your pulse).
  • RSI can be performed safely and with a high degree of success. But it requires in-depth initial and continuing training, as well as tight quality assurance. All of this requires time and money to implement and maintain, and is not for every EMS agency.
  • Adherence to established indications and selection criteria are critical in maximizing success with an RSI procedure.
  • Naloxone’s effects in the chronic user of narcotics is markedly underappreciated by most providers and may lead to a constellation of unanticipated emergent problems. If given at all, the lowest possible dose should be selected 0.4 mg maximum. Some sources advocate 0.1 to 0.2 mg maximum in these cases. The best approach is to avoid naloxone entirely.
  • This case emphasizes the importance of having adequate backup, or bail-out, airways besides the BVM available to prehospital providers. Incorporated into this availability is the mandate to be appropriately trained to use them.