Prove it: Does endotracheal intubation help children?

Highlights:
Introduction
The Review
What It Means for You

Introduction
In the early morning hours, Medic 34 responds to a motor-vehicle collision on the highway. The dispatch center reports that one vehicle struck another parked on the shoulder. Before the ambulance arrives, first responders report that cardiopulmonary resuscitation (CPR) is in progress.

Upon arrival, firefighters report that they found the 11 month-old infant lying in the floorboard of the rear vehicle. The child was apneic and pulseless. The firefighters gently rolled the patient onto a small backboard, initiated chest compressions, and assisted ventilations using a bag-valve mask (BVM).

The initial assessment in the back of the ambulance reveals a normally developed male with no obvious signs of trauma. A paramedic intern on the ambulance assumes responsibility for the assisted ventilations, which allows that firefighter to relieve the chest compressor. CPR continues at a rate of 15 compressions to 2 ventilations. The student says he wants to do the intubation. Because the children’s hospital is only about four miles away, the paramedic decides to defer intubation to the emergency department (ED) and begin transport immediately.

On the way, there is just enough time to place the patient on a cardiac monitor. The electrocardiogram (ECG) reveals a slow sinus rhythm at a rate of about 20 complexes per minute, however the patient remains pulseless.

In the ED, a pediatric trauma surgeon intubates the child’s trachea as the staff assumes responsibility for the resuscitation attempt. Despite their best efforts, the child dies.
After leaving, the paramedic student asks why the medic did not intubate the child during transport. The medic says that since there was noticeable rise and fall of the patient’s chest with each assisted breath and since the hospital was so close, he did not want to risk a failed attempt.

At the end of the shift, the intern asks the paramedic to sign the ride-out report. The medic notices that in the summary of the call, the student wrote that he believes they should have intubated the child since nothing is more important than the airway.

Review
Researchers in the Netherlands examined what effect paramedic endotracheal intubation (ETI) had on survival in the pediatric population (Gerritse, Th Draaisma, Schalkwijk, van Grunsven, & Scheffer, 2008). The study involved both the ground ambulance and the air medical response components of the Dutch EMS system.

The pathway to become the medical provider on a Dutch ambulance is a long one. Students must first obtain the United States equivalent of a baccalaureate in nursing (Wulterkens, 2007). The nurse must then obtain at least one year of experience and an additional specialty certification in intensive care, coronary care, or anesthesia. After completing this requirement, the nurse is eligible to apply for a yearlong training program to become an ambulance nurse (paramedic). This training involves 24 classroom days followed by on-the-job training with senior personnel for the remainder of the 12-months.

During this training, the paramedic candidate must complete a four-hour pediatric airway management session that combines theory with manikin practice (Gerritse et al., 2008). The Dutch do not mandate live patient ETI training before certification as a paramedic. Prehospital medical protocols in the Netherlands permit paramedics to intubate a child if bag-mask ventilation seems inadequate.
The helicopter medical team (HMT) in the Netherlands includes a physician, a nurse, and a pilot. The helicopter physicians have either an anesthesiology or trauma surgery background. Before assignment to the helicopter, the medical team receives intensive instruction in the field management of both adult and pediatric emergencies. The trauma surgeons also undergo more than six months of additional training in endotracheal intubation using rapid sequence induction.

The primary goal of this investigation was to prospectively determine if critically injured or ill children had better outcomes when paramedics managed the airway with an endotracheal tube compared to bag-mask ventilation. Paramedics arriving on the scene had the option of attempting endotracheal intubation or managing the pediatric airway with a BVM while deferring intubation to the arriving helicopter physicians. Paramedics who choose intubation confirmed placement and subsequently placed the children on portable ventilators using protocol driven settings. Once the helicopter arrived, the physician confirmed placement and effective ventilation using auscultation, capnography, direct laryngoscopy, and inspection of the materials and ventilator settings. A single pediatric trauma center received each of the patients where physicians confirmed tube position with a chest X-ray.

During a 69-month evaluation between January 2001 and September 2006, the HMT responded to 463 field requests for assistance with pediatric patients (Figure 1). Paramedics on the scene cancelled about one-third of the responses before the helicopter could arrive on the scene, either because the patient was not as critical as originally thought (131 calls), because the child died (21 calls), or for other reasons (11 calls). Once on the scene, the HMT evaluated and cared for 300 pediatric patients with a mean age of 6.8 years (S.D. = 5.4). The majority of the patients (83%) were injured as the result of traumatic forces. In every one of the 300 cases, EMS personnel arrived on the scene and initiated care before helicopter arrival. More than half of the children (n=155) required ETI. Seventy-nine of those children died, 21 on the scene, 12 en route to the hospital or in the emergency department, and 44 after admission to the intensive care unit.

The target population for this study was children under the age of 16 years who required ETI because of a critical illness or injury - defined as an initial Glasgow Coma Scale (GCS) score of 3 or 4. Ninety-five children met these criteria (Figure 2). Overall, survival in this group was 38%, similar to other studies of traumatized children with very low GCS scores (Grewal & Sutcliffe, 1991; White et al., 2001).

Paramedics elected to intubate 41 of the 95 patients before the HMT arrived on the scene. Once the helicopter arrived, the HMT physician performed 15 emergency corrections of tube position or ventilator settings (37% complication rate). Six children (14.6%) had unrecognized esophageal placements, seven children (17%) were intubated with an uncuffed tube that was too small thereby making positive pressure ventilation impossible, and two children (4.9%) had potentially lethal ventilator settings defined as 300% greater than recommended. Survival to hospital discharge for the children intubated by paramedics was 4.9%.

In 54 of the cases, the paramedics ventilated with a BVM and waited for the HMT physician to perform the intubation. Every one of the HMT intubations was correctly placed with proper ventilator settings. Survival in the group receiving BVM ventilation by the paramedics and endotracheal intubation by the HMT physician was 63%. There was no difference in survival if an anesthesiologist or a trauma surgeon intubated the child.

There was no statistical difference between the two groups in average age, Revised Trauma Score (RTS), incidence of trauma vs. non-trauma, or incidence of prehospital CPR (Table 1). There was however, a statistically significant increase in the probability of survival if paramedics ventilated the child with a BVM and waited for the HMT physician to perform the endotracheal intubation. The authors conclude that the current standards for training prehospital providers in the skill of pediatric ETI demands reevaluation and that EMS systems should focus greater attention on BVM management of the pediatric airway.

S. D. = standard deviation; n = number in group; ETI = endotracheal intubation; ns = not significant - did not meet criteria for statistical significance

What it means for you
Pediatric ETI offers unique challenges not seen in the adult patient. The glottic opening in a child is small and lies further anterior making direct visualization and successful placement more difficult. This difficulty could result in longer scene time and delayed transport. Adding to the problem, the educational requirements that paramedics need to master the skill may be inadequate (Stewart, Paris, Pelton, & Garretson, 1984; Stratton et al., 1991; Warner et al., 2010; Youngquist et al., 2008). All of these factors combine to make first time success rate of pediatric intubation very low (Ehrlich, Seidman, Atallah, Haque, & Helmkamp, 2004).

Researchers also found an unacceptably high rate of complications associated with pediatric ETI. In an evaluation of 355 pediatric intubations in the Northwest United States, researchers found about one-third of the placed tubes were too small, one-fifth had complications, with slightly less than half of the complications categorized as serious, including a 2% overall unrecognized esophageal placement rate (Brownstein, Shugerman, Cummings, Rivara, & Copass, 1996). Another study found the major complication rate associated with out-of-hospital pediatric ETI to be about 25% (Nakayama, Gardner, & Rowe, 1990).

High complication rates are not unique to urban paramedics. A level one trauma center in West Virginia receiving pediatric trauma patients from 13 surrounding counties with an average transport interval from the scene of 61 minutes found that over two-thirds of the patients who had an intubation attempt in the field had a complication (Ehrlich, Seidman, Atallah, Haque, & Helmkamp, 2004).

Multiple studies fail to demonstrate survival advantages associated with pediatric ETI. A Los Angeles prehospital intubation study in children under the age of 12 years could not demonstrate that prehospital intubation offered survival advantages or improvements in neurological outcome compared to airway management with a BVM (Gausche et al., 2000). In addition, over half of the children who paramedics believed they correctly intubated had complications including esophageal placement (1.6%), unrecognized dislodgement (6%), recognized dislodgement (8%), mainstem intubation (18%) and incorrect tube size (24%). A northern California study found no statistical differences in outcome when paramedics performed ETI in children under the age of 19 years compared to airway management with a BVM (Aijian, Tsai, Knopp, & Kallsen, 1989). The authors also found the complication rate to be 46%. A Spanish study of pediatric cardiac arrest in children under the age of 16 years found the risk of death was 2.5 times higher if paramedics performed ETI (Lopez-Herce et al., 2005). Finally, a retrospective evaluation of data from the National Pediatric Trauma Registry could find no survival advantage offered by paramedic ETI for pediatric patients with severe head injury compared to treatment with a BVM (Cooper et al., 2001).

Although this Dutch study has the strength offered by a prospective study, it suffers from a lack of randomization. Randomization helps reduce inadvertent bias or error from affecting the outcome. This study allowed the paramedic to decide whether or not to intubate, thereby determining each patient’s treatment group assignment. Although the medics, in theory made the determination based on a perception that BVM ventilation was suboptimal, it is possible that the medics selected for intubation, patients they believed had more serious problems. The researchers attempted to correct for this possible bias in two ways.

First, the researchers compared the mean (average) Revised Trauma Score (RTS) of the two groups of patients. Medics calculate the RTS by assigning a numeric value to the child’s GCS score, the systolic blood pressure, and the respiratory rate. The final score results from adding those three values. In theory, the healthcare team can use the RTS to predict outcome owing to the belief that lower scores will result in greater mortality. However, studies suggests that the RTS is an accurate predictor of the need for intensive care (Narci et al., 2009; Yian, Gullahorn, & Loder, 2000) but a poor predictor of mortality (Narci et al., 2009).

Next, the researchers only compared patients with a GCS score of 3 or 4. Individual components of the GCS appear to reliably predict mortality in pediatric patients, especially the pupil and motor response scores (Burd, Jang, & Nair, 2006). However, researchers who use a formal randomization protocol provide greater confidence that any difference between the treatment and the control groups is the result of the study variable and not error.

Generalization of the results also suffers from a very small number of patients in the analysis. The sample represented only about 60% of all the pediatric patients who required ETI during the study period. Although the survival difference between the two groups met the requirements for statistical significance and is very compelling, the small sample size lacks sufficient power to definitively determine cause and effect.

The study is also limited by the lack of outcome data. Survival-to-hospital discharge only provides a glimpse of the true efficacy of an intervention. The ultimate goal of any medical procedure is to return the patient to a functional state. If the medical community employs an intervention that only results in comatose and vegetative survivors, one should certainly question the utility of the intervention.

Finally, one major criticism of the study is that a Dutch paramedic may not be analogous to a paramedic in the United States, thereby limiting the generalizability of the results. Dutch paramedics are essentially baccalaureate prepared nurses with two additional years of critical and prehospital care training. Many areas of the United States do not require any college-level education as a prerequisite to paramedic certification. It is beyond the scope of this review to comment on possible advantages that college education for paramedics might have on patient outcome. This author will save that for another time.

However, the relevant point is the similarities in pediatric airway control training that Dutch “paramedics” receive compared to United States paramedics. The Dutch system requires four-hours training in pediatric airway management that combines theory with manikin practice (Gerritse et al., 2008). The National Standards Curriculum for paramedics in the United States does not establish a minimum amount of training time for pediatric airway control and it is reasonable to believe that many U. S. training programs do not exceed the training requirements in the Netherlands. In addition, neither the Dutch nor the U. S. systems mandate live endotracheal training on pediatric patients before certification as a paramedic.

Pediatric ETI by paramedics remains controversial in the United States. No evidence exists that demonstrates improved outcomes when paramedics perform the skill. As suggested by the Dutch authors, the EMS community must critically examine the educational standards for pediatric airway management in this country and reevaluate whether paramedics should continue to perform ETI.

References
Aijian, P., Tsai, A., Knopp, R., & Kallsen, G. W. (1989). Endotracheal intubation of pediatric patients by paramedics [abstract]. Annals of Emergency Medicine, 18(5), 489-494. doi:10.1016/S0196-0644(89)80830-3

Brownstein, D., Shugerman, R., Cummings, P., Rivara, F., & Copass, M. (1996). Prehospital tracheal intubation of children by paramedics. Annals of Emergency Medicine, 28(1), 34-39.
Burd, R. S., Jang, T. S., & Nair, S. S. (2006). Predicting hospital mortality among injured children using a national trauma database. Journal of Trauma, 60(4), 792-801. doi:10.1097/01.ta.0000214589.02515.dd

Cooper, A., DiScala, C., Foltin, G., Tunik, M., Markenson, D., & Welborn, C. (2001). Prehospital endotracheal intubation for severe head injury in children: A reappraisal. Seminars in Pediatric Surgery 10(1), 3-6.

Ehrlich, P. F., Seidman, P. S., Atallah, O., Haque, A., & Helmkamp, J. (2004). Endotracheal intubations in rural pediatric trauma patients. Journal of Pediatric Surgery, 39(9), 1376-1380. doi:10.1016/j.jpedsurg.2004.05.010

Gausche, M., Lewis, R. J., Stratton, S. J., Haynes, B. E., Gunter, C. S., Goodrich, S. M., Poore, P. D., McCollough, M. D., Henderson, D. P., Pratt, F. D., & Seidel, J. S. (2000). Effect of out-of-hospital pediatric tracheal intubation on survival and neurological outcome. Journal of the American Medical Association, 283(6), 783-790. doi: 10.1001/jama.283.6.783

Gerritse, B. M., Th Draaisma, J. M., Schalkwijk, A., van Grunsven, P. M., & Scheffer, G. J. (2008). Should EMS-paramedics perform paediatric tracheal intubation in the field? Resuscitation, 79(2), 225—229. doi:10.1016/j.resuscitation.2008.05.016

Grewal, M., & Sutcliffe, A. J. (1991). Early prediction of outcome following head injury in children: An assessment of the value of Glasgow Coma Scale score trend and abnormal plantar and pupillary light reflexes. Journal of Pediatric Surgery, 26(10), 1161-1163. doi:10.1016/0022-3468(91)90323-L

Lopez-Herce, J., Garcıa, C., Dominguez, P., Rodrıguez-Nunez, A., Carrillo, A., Calvo, C., Delgado, M. A., & Spanish Study Group of Cardiopulmonary Arrest in Children. (2005). Outcome of out-of-hospital cardiorespiratory arrest in children. Pediatric Emergency Care, 21(12), 807-815.

Nakayama, D. K., Gardner, M. J., & Rowe, M. I. (1990). Emergency tracheal intubation in pediatric trauma. Annals of Surgery, 211(2), 218-223.

Narci, A., Solak, O., Turhan-Haktanir, N., Ayçiçek, A., Demir, Y., Ela, Y., Ozkaraca, E., & Terzi, Y. (2009). The prognostic importance of trauma scoring systems in pediatric patients. Pediatric Surgery International, 25(1), 25-30. doi:10.1007/s00383-008-2287-5

Stewart, R. D., Paris, P. M., Pelton, G. H., & Garretson, D. (1984). Effect of varied training techniques on field endotracheal intubation success rates. Annals of Emergency Medicine, 13(11), 1032-1036.

Stratton, S. J., Kane, G., Gunter, C. S., Wheeler, N. C., Ableson-Ward, C., Reich, E., Pratt, F. D., Ogata, G., & Gallagher, C. (1991). Prospective study of manikin-only versus manikin and human subject endotracheal intubation training of paramedics. Annals Emergency Medicine, 20(12), 1314-1318. doi:10.1016/S01960644(05)81073-X

Warner, K. J., Carlbom, D., Cooke, C. R., Bulger, E. M., Copass, M. K., & Sharar, S. R. (2010). Paramedic training for proficient prehospital endotracheal intubation. Prehospital Emergency Care, 14(1), 103-108. doi: 10.3109/10903120903144858

White, J. R. M., Farukhi, Z., Bull, C., Christensen, J., Gordon, T., Paidas, C., & Nichols, D. G. (2001). Predictors of outcome in severely head-injured children. Critical Care Medicine, 29(3), 534 –540.
Wulterkens, D. (2007). EMS in the Netherlands: A Dutch treat? Retrieved from http://www.jems.com/print/4067

Yian, E. H., Gullahorn, L. J., & Loder, R. T. (2000). Scoring of pediatric orthopedic polytrauma: correlations of different injury scoring systems and prognosis for hospital course. Journal of Pediatric Orthopedics, 20(2), 203–209. doi:10.1097/00004694-200003000-00014

Youngquist, S. T., Henderson, D. P., Gausche-Hill, M., Goodrich, S. M., Poore, P. D., & Lewis, R. J. (2008). Paramedic self-efficacy and skill retention in pediatric airway management. Academic Emergency Medicine, 15(12), 1295-303. doi: 10.1111/j.1553-2712.2008.00262.x