Randomized controlled trial comparing an open surgical technique and a Seldinger technique for cricothyrotomy performed on a simulated airway

Abstract

Objectives

Emergency cricothyrotomy is a lifesaving procedure performed when intubation fails and oxygenation cannot occur. There are multiple techniques and kits to perform this procedure. However, current evidence does not provide a definitive answer as to which method is superior. Two techniques in common use are a surgical technique and a percutaneous Seldinger‐based cricothyrotomy kit. The objective was to determine which of these two methods was quickest to perform and to determine which was most preferred by participants.

Methods

A prospective randomized controlled crossover trial was conducted involving emergency physicians and trainees. Each participant performed both cricothyrotomy techniques in succession on an airway model, with the technique performed first being randomized for each participant. The primary outcome was time to first insufflation of the artificial lung. A survey was completed by participants asking their comfort with each technique on a 5‐point scale from 1 (not at all comfortable) to 5 (very comfortable) and which technique they preferred.

Results

Twenty‐one emergency physicians and nine emergency medicine trainees were recruited. The surgical technique was performed the fastest, with a mean (±SD) time of 51.6 (±16.3) s versus 66.6 (±14.9) s for the Seldinger technique, with a statistically significant difference of 15.0 s (95% confidence interval = 8.5 to 21.5, p < 0.001). The surgical technique was rated the most comfortable to perform, with a median rating of 5 (interquartile range [IQR] = 4–5) versus 4 (IQR = 3–5) for the Seldinger technique. The surgical technique was most preferred by participants (80% vs 20%).

Conclusion

The surgical technique was the fastest to perform and was rated the most comfortable to perform and the most preferred technique.

INTRODUCTION

The final can't intubate can't oxygenate (CICO) option in all airway management algorithms is insertion of a tracheal tube or other device via a cricothyrotomy. ¹ , ² , ³ , ⁴ , ⁵  This situation arises when attempts have failed to manage the airway using intubation, facemask ventilation or supraglottic airway device. ⁶ , ⁷ Although rarely performed, it is vital that doctors are confident in performing the procedure and that they are using the best technique available.

There are many different techniques for performing a cricothyrotomy. ⁸ However, generally there are two broad techniques: the open/surgical/scalpel technique (surgical) and the percutaneous/needle over wire/Seldinger technique (Seldinger). Multiple studies have been conducted to compare these techniques. Some show a superiority of Seldinger ⁷ , ⁹ whereas others show the superiority of surgical. ¹⁰ , ¹¹ , ¹² , ¹³

A systematic review examining all the techniques available for cricothyrotomy summarized the evidence available at the time ⁸ and found a significant problem with the available literature. A large majority of the studies were too small to demonstrate statistically significant differences, and the available evidence was of low or very low quality with large heterogeneity between studies. The main conclusion of this review was that no technique had been proven to be superior to the others, regarding success rate or timeliness. For comparisons of Seldinger with surgical, no significant difference in the success rates between the two was identified. However, studies indicated that surgical was generally faster than Seldinger.

Complicating this is the attitudes of practitioners to the two techniques. It is generally acknowledged that practitioners may be more comfortable with Seldinger rather than surgical ¹⁴ and evidence shows that practitioners, and nonspecialists especially, prefer the Seldinger technique. ⁹ , ¹² , ¹³ , ¹⁵ However, there is also conflicting evidence that surgical is preferred. ¹⁰

There is also very little evidence describing the difficulties of cricothyrotomy. ⁹ Both techniques are technically difficult in some respects, using equipment that may not be familiar to the operator. ⁴ , ⁵ , ⁸

Therefore, the aim of this study was to determine which of two methods for emergency cricothyrotomy is the most time‐efficient, to determine which of these techniques is the most preferred, and to document any equipment or procedural difficulties for each technique.

METHODS

The Department of Emergency Medicine at Nepean Hospital sees approximately 75,500 patients per year and is classified as a regional trauma center by the NSW Institute of Trauma and Injury Management. Airway management is performed by emergency medicine physicians and trainees. Regular training sessions are held covering airway management, including difficult airway algorithms and surgical airway techniques. Both surgical and Seldinger are taught for cricothyrotomy.

Study participants were emergency medicine physicians and trainees who have been trained to use both surgical and Seldinger to perform cricothyrotomy. Participants were excluded if they had not been trained in either or both techniques or they were unwilling to participate. We approached all Nepean Hospital emergency medicine physicians and trainees. A brief study overview was provided by mass email. During the study period, 30 emergency physicians and 16 emergency medicine trainees were approached.

The equipment required for both procedures is standard equipment found in the difficult airway trolley in the resuscitation bay at Nepean Hospital Emergency Department. The equipment available for the surgical technique was a scalpel, forceps, airway bougie, and size 6 cuffed endotracheal tube (Figure 1). Seldinger was performed using Cook Melker cuffed emergency cricothyrotomy catheter set (Cook Inc.; Figure 1). Both techniques were performed on a Frova Crico‐Trainer airway model (VBM Medizintechnik GmbH; Figure 2), which was familiar to participants because it is available in the Nepean Hospital Emergency Department and is used for training and simulation. One author would act as an airway assistant to the procedure. In this role the assistant would take a very passive role and only perform tasks that were requested by the participant. This was explained to the participant during the consent process.

FIGURE 1.

Equipment available for cricothyrotomy. The equipment for the surgical technique is on the left and the Melker kit is on the right

FIGURE 2.

The Frova Crico‐Trainer airway model used for the study

A randomized controlled crossover trial was performed comparing surgical and Seldinger. Participants were not provided with any refresher training or information regarding the equipment they would be using. Each participant performed both techniques and the technique that was performed first was randomized. Randomization was implemented via block randomization with random block sizes. The techniques were performed in quick succession in the same session. Although there was a risk of a carry‐over effect to the second procedure, this was thought preferable to performing the second procedure at a later date and avoiding the possibility of the participant preparing for it, knowing which procedure was yet to be performed, and was also mitigated by the randomized crossover design of the study.

The time to complete each procedure was measured via a stopwatch: timing began at the direction of the researcher to commence the procedure and ended at the first successful insufflation. The participant had a short time to become familiar with the equipment and palpate the relevant anatomy prior to commencing each procedure, to mimic an optimized real‐life situation. ³ , ⁴ , ⁵ , ¹⁶

Data were recorded regarding whether the participant was an emergency physician or trainee and when they last performed a cricothyrotomy (either real or simulated). The time to perform each procedure was recorded, as were any technical issues or difficulties with the procedure. After performing both procedures, participants filled out an attitude survey that asked to rate how comfortable they would be performing each technique, recorded on a 5‐point Likert scale, and their preferred method.

Primary outcome was which technique was performed quickest. Secondary outcomes were which technique was rated as most comfortable, which was the most preferred technique, whether there was a difference in preferred technique between emergency physicians and trainees, whether there was a difference in times for surgical and Seldinger between emergency physicians and trainees, whether there was a difference in times for surgical and Seldinger between participants who had performed a cricothyrotomy within the last year compared to those who had not, and what technical difficulties were encountered by participants for each technique.

Times to perform each technique were compared using a two‐tailed paired‐samples t‐test. The comfort rating for each technique was compared using a two‐tailed Wilcoxon signed‐rank test. The proportion of participants who preferred surgical versus Seldinger and the differences between emergency physicians’ and trainees’ preferred technique was compared by calculating a z‐score for two population proportions.

The difference in times for surgical and Seldinger between emergency physicians and trainees and the difference in times for surgical and Seldinger between participants who had performed a cricothyrotomy within the past year compared to those who had not were examined using a two‐tailed two‐samples t‐test using unequal variances. Technical difficulties with each technique were collated and examined for any patterns. The t‐test was performed using Microsoft Excel for Microsoft 365. The Wilcoxon signed‐rank test was performed using the Web resource VassarStats. ¹⁷  The calculation of z‐scores for the difference between two population proportions was performed using the Web resource Social Science Statistics. ¹⁸

The null hypothesis was that there was no difference in mean time to perform each technique. We expected a mean time for each technique of around 60 s based on our experience in surgical airway training locally. Studies with similar times showed a standard deviation (SD) of 15 s. ⁸ We thought a difference of 10 s would be clinically significant considering the clinical situation. To detect a difference of 10 s between techniques, with an alpha of 0.05, a beta of 0.2, and a SD of 15 s, 20 participants were required. We aimed to recruit 30 participants: first, to ensure the validity of the results, and second, a study of 30 participants performing 60 procedures would be larger than most previous studies reviewed. ⁸

Ethics approval was granted by the Nepean Hospital Human Research and Ethics Committee as a low‐ and negligible‐risk study and participants gave written consent prior to participation.

RESULTS

There were 30 participants, including 21 emergency physicians and nine emergency medicine trainees. Sixteen participants had performed a cricothyrotomy in the past year. Two emergency physicians had performed cricothyrotomy in a real clinical situation.

Surgical was performed fastest, with a mean (±SD) time to perform each technique of 51.6 (±16.3) s for surgical and 66.6 (±14.9) s for Seldinger, with a statistically significant difference (95% CI) of 15.0 (6.5) s (p < 0.001). Surgical was rated the most comfortable to perform, with a median (IQR) rating of 5 (4–5) for surgical and 4 (3–5) for Seldinger (p = 0.017). Surgical was most preferred by participants (24/30; p < 0.001).

There was no difference in the preferred technique between emergency physicians and trainees (17/21 vs 7/9, p = 0.841). There was no significant difference in times for surgical and Seldinger between emergency physicians and trainees and between participants who had performed a cricothyrotomy within the past year compared to those who had not (Table 1).

TABLE 1.

Differences in time for the two techniques between emergency physicians and trainees and between participants who had performed a cricothyrotomy within the past year compared to those who had not

Technique	Variable	Participants	Time (s)	p‐value
Surgical	Physician	21 (70)	48.6 (±16.6)	0.100
	Trainee	9 (30)	58.8 (±13.9)
Seldinger	Physician	21 (70)	64.6 (±12.8)	0.364
	Trainee	9 (30)	71.2 (±19.2)
Surgical	<1 year	16 (53)	47.8 (±16.1)	0.174
	>1 year	14 (47)	56.0 (±16.0)
Seldinger	<1 year	16 (53)	64.4 (±16.2)	0.400
	>1 year	14 (47)	69.1 (±13.5)

Values are number (%) or mean (±SD).

A number of technical difficulties were identified (Table 2). These included the incision for either surgical or Seldinger being too small and a failure to hold the dilator in Seldinger delaying insertion. A post hoc analysis was performed comparing the times for Seldinger with and without a failure to hold the dilator, with mean (±SD) of 62.9 (13.2) s holding the dilator and 72.1 (±16.2) s when not holding the dilator. However, this difference was not statistically significant on a two‐tailed two‐sample t‐test (p = 0.12).

TABLE 2.

List of technical problems documented for each procedure

Technical difficulty
Incision too small
Seldinger	7 (23)
Surgical	5 (17)
Did not hold dilator during insertion (Seldinger)	12 (40)
Wire came out with needle (Seldinger)	1 (3)
Cut finger (surgical)	1 (3)
Device not fully inserted (beyond the balloon)
Seldinger	1 (3)
Surgical	0 (0)

Values are number (%).

DISCUSSION

The ideal cricothyrotomy technique should result in a high success rate and a low complication rate; it should be easy to master, involve only a few steps, provide protection against aspiration, and allow adequate ventilation. ¹⁵  This study in a task trainer model demonstrated surgical was performed faster than Seldinger. This result adds to the literature suggesting that surgical should be the preferred approach to achieve airway access in a CICO scenario.

Many caregivers have little experience in real CICO situations in life‐threatened patients; therefore, maximizing time efficiency is crucial. ² , ¹⁰ It is therefore imperative that the method used for securing such an airway should be rapid, highly effective, and intrinsically simple to perform. ¹⁰

The open surgical technique has a number of advantages. ¹¹ , ¹³ , ¹⁶ , ¹⁹ A scalpel, forceps, bougie, and tracheal tube are common implements in critical care environments and their use should hold no surprises. While a wire‐guided technique may be a reasonable alternative for operators who are experienced with this method, the evidence suggests that a surgical cricothyroidotomy is both faster and more reliable. ⁵

There are theoretical advantages to the Seldinger technique. It is a standard technique used across a number of different procedures, from central venous access to intercostal catheters. It is generally considered that anesthetists are more comfortable with Seldinger than use of a scalpel. ¹⁴ , ¹⁵ Studies have also shown that consultants have preferred surgical whereas nonconsultants preferred Seldinger. ¹² In this study, surgical was the preferred technique and was rated the most comfortable to perform. There was also no difference in the proportion of emergency physicians and trainees who preferred surgical.

The mean time for Seldinger in this study was more rapid than the majority of papers that have studied the Melker device. ⁸ For 12 papers that examined the Melker device, times ranged from 38 to 126 s, with nine of 12 papers quoting times slower than the mean of 66.6 s in this study. Two of the papers used a different starting point that may have led to an artificially lower mean time to perform the procedure compared to the methods used in this study. Importantly, in contrast to our study, a number of these studies involved some form of training prior to the participant performing the procedures. ⁷ , ⁸ , ¹⁰ , ¹¹ , ¹² In this respect, the faster time for Seldinger in our study is significant. However, the mean time to perform surgical was comparable to the available literature. ⁸ Despite the comparatively rapid Seldinger time in our study, it was still slower than surgical.

There was a small but nonsignificant difference in times to perform each procedure related to experience. A concerning aspect of this study was that 14 of 30 participants had not performed a cricothyrotomy, either real or simulated, within 1 year of participating in the study. Previous training will likely be remote in time and possibly poorly remembered in this highly stressful situation. ¹⁰  The Difficult Airway Society guidelines suggest that training should be repeated at regular intervals to ensure skill retention. However, the optimal frequency of training for cricothyrotomy is yet to be determined.

We identified a number of technical issues with both procedures (Table 2). A particular problem for the Seldinger catheter was around holding the dilator during insertion. If the catheter (rather than the dilator) was held during attempted insertion, there was a tendency for the dilator to telescope up the catheter. This delayed the insertion of the device and therefore the first insufflation, due to the operator trying to penetrate catheter alone into the trachea, rather than the more streamlined dilator. Another problem was the incision for either technique being too small, requiring extension of the initial incision and delaying insertion.

This has particular implications for training. Having identified a number of issues that can potentially delay insertion, these issues need to be specifically addressed in the training of both techniques. NAP4 concluded the technique of cannula cricothyroidotomy needs to be taught and performed to the highest standards to maximize the chances of success ⁴ and these findings reinforce this.

Some commentary suggests that a wire‐guided technique should be avoided, ¹¹ , ¹³ , ¹⁶ , ¹⁹ with the Difficult Airway Society advocating a didactic scalpel technique to promote standardized training. ⁵ However, there are potential difficulties with the surgical technique. It uses standard intubating equipment not designed specifically for this procedure. The use of an assistant is necessary to control the bougie and to railroad the tracheal tube onto it. This contrasts with the Seldinger technique where the catheter was specifically designed for the procedure.

An open surgical technique using the Melker cricothyrotomy catheter has been described. ²⁰ , ²¹ It is a straightforward technique that can be performed by a single operator, potentially within 20 s. ²² It has the advantages of the open surgical technique as well as using a catheter specifically designed for cricothyrotomy. Because surgical has been proven quickest and most preferred compared to Seldinger in this study, a study comparing surgical to the open technique using a cricothyrotomy catheter is required.

LIMITATIONS

This study used a simulated airway model and did not replicate potential real‐life phenomena such as difficult anatomy and excessive bleeding. Previous studies also tended to use airway models or cadavers. ⁵ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹² It is possible that the lack of bleeding in the majority of the study models may—especially for the surgical techniques—result in artificially short procedure times, ⁸ and this study has the potential for this same bias. Also, participants would have known the two techniques that were to be studied from the recruitment email and could have refreshed themselves with the techniques prior to participation, which could have affected success rates and performance times.

Success rate was 100%. Airway models and cadavers may also result in artificially high success rates. ⁸ A study performed on anesthetized sheep had success rates of only 91% and 73% for two open surgical techniques studied. ²³ Accuracy for identifying the cricothyroid membrane may be as low as 30% in live human subjects ²⁴ and research suggests that there is still a role for live tissue models for training of surgical airway as those trained on synthetic models had more critical fails when tested on live tissue models. ²⁵

Again, this has significant implications for training. Airway models are appropriate for the initial training of surgical techniques. They allow training in a purposeful manner that allows useful repetitions of the skill before working up to stressful scenarios. ²⁶ However, to increase fidelity, and therefore difficulty and stress, models need to approximate real‐life conditions as much as possible. Cadavers and live animal models are not a sustainable source of training models. An appropriate model allows for the palpation of anatomical landmarks, introduces complicating factors or common sources of error such as skin and subcutaneous tissue, and produces both visual and tactile feedback. ²⁶  Very little is available commercially and it is left to the trainer to devise a model that approximates real clinical conditions.

The results are not translatable to other equipment and techniques, even though they might broadly be classified as either an open surgical or a needle over guidewire technique. There was the possibility of a learning bias for the second technique performed but this was mitigated by randomization of the technique performed first. The airway assistant during each procedure was one of the study authors and it may be that using a different assistant, such as a trained airway nurse, may affect the results. The fact that one of the authors was an emergency physician may also bias results considering that some of the participants were trainees. The participants were all emergency physicians or trainees so the results may not translate to other specialty groups.

CONCLUSIONS

This study in a task trainer model demonstrated that the surgical technique was performed faster than the Seldinger technique, was rated the most comfortable to perform, and was the most preferred by participants. We also identified a number of technical issues with each technique, such as the incision being too small and a failure to hold the Melker catheter dilator adequately leading to delays in insertion. These issues need to specifically addressed in the training of these techniques.

CONFLICT OF INTEREST

The authors have no potential conflicts to disclose.

AUTHOR CONTRIBUTIONS

James L. Mallows contributed to the study protocol, submitted application to ethics committee, assisted with data collection, assisted with the literature search, assisted with writing the manuscript, and performed the statistical analysis. Perry A. Tyler contributed to the study protocol, assisted with data collection, assisted with the literature search, and assisted with writing the manuscript.

Mallows JL, Tyler PA. Randomized controlled trial comparing an open surgical technique and a Seldinger technique for cricothyrotomy performed on a simulated airway. AEM Educ Train. 2021;5:e10699. doi: 10.1002/aet2.10699