Ultrasound Improves Cricothyrotomy Success in Cadavers with Poorly Defined Neck Anatomy: A Randomized Control Trial

DIFFICULTY with tracheal intubation is the most common cause of serious adverse respiratory events for patients undergoing anesthesia.^1,2  Repeated attempts of difficult intubation can result in serious soft tissue injury and rapidly deteriorate into a “cannot intubate-cannot oxygenate” situation that requires a surgical cricothyrotomy as a potentially life-saving procedure. However, cricothyrotomy is an infrequently performed procedure^3,4  with complication rates ranging from 9 to 40%.^5,6  Accurate identification of the cricothyroid membrane is crucial for the success and minimal complications in performing a cricothyrotomy. The current practice to identify relevant anatomical landmarks has relied solely on digital palpation. However, inaccuracy in identifying the cricothyroid membrane using digital palpation is among one of the most common errors that have resulted in device misplacement, leading to failed cricothyrotomies and adverse outcomes.^7–9 

Ultrasonography of the airway is increasingly used in airway management.^10,11  Nicholls et al.¹²  reported a standardized ultrasonography technique to identify the cricothyroid membrane and to evaluate the ability of emergency physicians to apply this technique in a cohort of patients in the emergency department. The findings in this study suggest that the cricothyroid membrane and relevant structures can be quickly identified by emergency physicians by using ultrasonography technique. Furthermore, the use of ultrasonography in identification of the cricothyroid membrane has been described in detail in the study by Kristensen.¹⁰  However, little is known on the outcomes in performing a cricothyrotomy with ultrasonography guidance. We hypothesized that ultrasound-guided identification of the cricothyroid membrane would reduce complications associated with cricothyrotomy. In this study, we aimed to determine the outcomes of cricothyrotomy performed on human cadavers by using ultrasound guidance, compared with conventional digital palpation, of anatomical landmarks. In particular, complication rates, failure to cannulate, correct placement of the device, and insertion time of cricothyrotomy were assessed.

This study was approved by the Research Ethics Board of Mount Sinai Hospital, Toronto, Ontario, Canada, and registered with ClinicalTrials.gov identifier: NCT01475487 (principal investigator: N.S.; date of registration: November 3, 2011). The clinical trial registry has been updated due to an oversight in the initial registration before data collection. We followed Consolidated Standards of Reporting Trials (CONSORT) checklist (appendix).

After written informed consent, 47 anesthesia residents in postgraduate years 1 to 5 and anesthesia fellows at the University of Toronto (Toronto, Ontario, Canada) were recruited in the study over a period of 12 months. Only participants with no previous experience in neck ultrasonography were recruited in the study. Their age, training grade, and years of practice were recorded. Participants who had performed a cricothyrotomy in the previous 6 months were excluded from the study.

The study was conducted on formalin-fixed human cadavers in the anatomy laboratory at the Department of Anatomy, University of Toronto. We used a total of 52 formalin-fixed human cadavers for the study assessment and teaching. To make reliable injury assessment, we used separate cadavers for each performance.

Two study investigators selected the cadavers based on preset criteria that included identification of anatomical landmarks visually and by palpation. The identification of the cricothyroid membrane by digital palpation was performed by using the index and third fingers of the nondominant hand. The thyroid cartilage was first palpated in the midline starting from cephalad and moving caudally until the cricoid cartilage is palpated. The cricothyroid membrane is the space between the inferior border of the thyroid cartilage and the superior border of the cricoid cartilage. The degree of difficulty in identifying the landmarks was graded according to a previously described grading system¹² : grade 1—easy = visual landmarks; 2—moderate = requires light palpation of landmarks; 3—difficult = requires deep palpation of landmarks; and 4—impossible = landmarks not palpable. Cadavers with a grade greater than 2 were recruited in the study.

A 15- to 10-mHz linear probe (MicroMaxx System; SonoSite Canada Inc., Canada) was used to obtain sonographic images of anatomical landmarks on cadaveric necks as described in the study by Kristensen.¹⁰ 

The participants held the linear high-frequency transducer in their nondominant hand and placed themselves on the right side of the cadaver facing toward the head of the cadaver. Then they placed the ultrasound probe transversely over the cadaver’s neck just above the suprasternal notch to visualize the trachea. The transducer was then moved laterally to the patient’s right side until the right border of the transducer was superficial to the midline of the trachea. During this movement, it was ensured that the right end of the transducer was kept in the midline of the trachea while the left end of the transducer was rotated into the sagittal plane resulting in a longitudinal scan of the midline of the trachea.

All participants in the study received a 10-min didactic lecture on cricothyrotomy using the Portex^® cricothyrotomy kit (Smiths Medical, USA) followed by a 3-min video demonstration of ultrasonographic and conventional digital palpation technique to identify the anatomical landmarks and the cricothyroid membrane. All participants were then given hands-on training at least five times with ultrasound using the Portex^® cricothyrotomy device on human cadavers.

After the teaching session, participants were randomized by computer-generated number tables to one of two groups: digital palpation group (n = 23), participants in this group performed cricothyrotomy using the Portex^® device with conventional digital palpation of the cricothyroid membrane, and ultrasound group (n = 24), participants in this group performed cricothyrotomy using the Portex^® device with ultrasonographic identification of the cricothyroid membrane. Only cadavers with grades 2 to 4 landmarks were used in the test session.

All cricothyrotomy procedures were video recorded for analysis. After termination of each cricothyrotomy, two anesthesiologists who were blinded to group allocation assessed the cadaver’s neck. The assessments were done first with a fiber-optic bronchoscope and then followed with dissection of the neck to assess for device position, complications, and severity of any injuries. The primary outcome measure was the complication rate as assessed by the severity of injuries, defined as the incidence and severity of posterior laryngeal and tracheal wall injuries, as graded by two anesthesiologists using the grading system described by Murphy et al.¹³  (none [no injury]; mild [< 5-mm laceration]; moderate [> 5-mm laceration or partial puncture]; and severe [> 10-mm laceration or full puncture]). The secondary outcomes included (1) insertion time, measured in seconds from the time of palpation of the skin to insertion of the Portex^® device in the trachea; (2) failure, with a “failure” defined as any attempt in which the trachea was not cannulated or which required more than 300 s to perform; and (3) correct landmarking, defined as having the Portex^® device inserted via the cricothyroid membrane. In this study, the scale was dichotomized to none–mild and moderate–severe injuries. An injury of none–mild severity would likely have little clinical adverse outcomes, whereas a moderate–severe injury could lead to significant adverse outcomes.

We calculated the sample size based on complication rate as primary outcome. This calculation was based on our pilot cases with the use of ultrasound and review of the literature, to give us an α error of 0.05 and 80% power.¹⁴  A sample size of 21 participants per group was required to detect a 40% absolute difference in the complication rate between digital palpation and ultrasonography-guided techniques assuming a 45% complication rate in the digital palpation (control) group and using the Fisher exact test for the comparison of independent proportions. The sample size was increased by 10% to compensate for potential withdrawals and incompletion. Therefore, a total of 23 and 24 participants were recruited in the digital palpation and ultrasonography group, respectively.

All the recorded variables were entered into a Microsoft Excel (Microsoft Corp., USA) spreadsheet and were analyzed using SAS version 9.01 (SAS Institute, Inc., USA) statistical software. Descriptive statistics including measures of central tendency, dispersion, and 95% CIs were calculated. We reported relative risk (RR) to compare injuries between the two groups. This is a randomized, controlled trial where the exposure precedes the outcome. It is therefore preferable to refer to the risk of an event occurring given exposure (and thus RR as a ratio of risks). The independent-samples Student t test and Fisher exact test were used to compare group means. Categorical data including injuries were compared using chi-square test. A two-tailed P value of less than 0.05 was considered statistically significant for all analyses.

We enrolled 47 participants who completed the training and procedural assessments: 23 procedures under digital palpation group and 24 under ultrasonography group (fig. 1). Although the participants included anesthesia residents and clinical anesthesia fellows, the distribution of residents at different training level (first year through fifth year in the residency program) and fellows was similar among groups (P = 0.70). We analyzed 39 procedures performed by residents and 8 by fellows. The cadavers’ sex, 28 (59.6%) feminine and 19 (40.4%) masculine, was equally distributed among groups (P = 0.14). The anatomical difficulty represented by the cadaver’s neck circumference (overall mean [SD] 21.45 [1.82] cm; P = 0.72) and grading system (overall grade 2, 3, and 4: 20, 16, and 11; P = 0.70) was also similar among study groups. Table 1 demonstrates the characteristics of participants and human cadavers, that is, sex, level of training of residents, cadavers’ sex, grade, and neck size.

The cricothyrotomy outcomes between digital palpation and ultrasonography groups are presented in table 2. The incidence of moderate–severe injury rate was almost three times greater in the digital palpation group than in the ultrasonography group (17 of 23 = 74% vs. 6 of 24 = 25.0%; P = 0.001) with an RR of 2.88 (95% CI, 1.39 to 5.94). The frequency of distribution for injuries according to original grade classification is presented in table 3.

The mean (SD) insertion time in the ultrasonography group was significantly longer than that in the digital palpation group (196.1 s [60.6 s] vs. 110.5 s [46.9 s]; P< 0.001). The majority of participants in both groups inserted the Portex^® device on the first attempt with nonsignificant difference (digital palpation: 20 of 23 = 87% vs. ultrasound: 19 of 24 = 79%; P = 0.701). However, a greater, but not significant, number of participants in the ultrasonography group, compared with the digital palpation group, inserted the device correctly via the cricothyroid membrane (15 of 24 = 63 vs. 9 of 23 = 39%; P = 0.148).

When cricothyrotomy procedures were stratified to cadaver subjects with grade 3 (difficult) and grade 4 (impossible) landmarks, ultrasonography guided, compared with digital palpation, increased the probability of correct device insertion by 5.6 times (P = 0.043) and significantly reduced the incidence of moderate–severe injuries by three-fold (ultrasound group 5 of 15 = 33.3% vs. digital palpation group 12 of 12 = 100%; P < 0.001; table 4).

This study aimed to assess the impact of ultrasonographic identification of the cricothyroid membrane on cricothyrotomies performed on human cadavers. The most significant findings of our study demonstrate that ultrasonography compared with conventional digital palpation increased the probability of correct insertion of the Portex^® cricothyrotomy device via the cricothyroid membrane by 5.6 times and decreased the incidence of moderate–severe injuries by three-fold in cadavers with difficult or impossible to palpate anatomical landmarks. These results imply that ultrasound-guided identification of cricothyroid membrane significantly improves the success rate and decreases the severity of injuries in performing cricothyrotomy with the Portex^® device on human cadavers.

The accurate identification of the cricothyroid membrane plays a crucial role when performing a surgical cricothyrotomy for correctly inserting the airway device via the cricothyroid membrane.⁵  The fourth National Audit Project on major airway complications reveals that 15 (52%) of the total 29 attempted cricothyrotomies resulted in failure. Of the 25 attempts made by anesthesiologists to establish an emergency surgical airway, only 9 (36%) were successful.⁷  Device misplacement was reported as one of the most common errors resulting in failed cricothyrotomies. The findings in this study suggest that ultrasound-guided identification of cricothyroid membrane may improve correct device insertion and successful cricothyrotomy.

The updated American Society of Anesthesiologist Difficult Airway Algorithm has introduced an additional category of patients with possible difficult surgical airway access.¹⁴  Although the specifics of this group of patients were not described, patients with large body mass index (BMI), previous radiation exposure, neck surgery, neck tumors, and large tissue flaps could potentially create a challenge for identifying landmarks for cricothyroid membrane localization. In this study, we had a subcohort of 27 cadavers with difficult or impossible to palpate landmarks due to various factors such as obesity and short necks. Eisma et al.¹⁵  have demonstrated that formalin treatment of cadavers leads to distortion of skin and tissue and thus hampering tactile feedback for identifying anatomical landmarks. In the remaining 20 cadavers, the landmarks were superficially palpable.

The incidence rate of correctly inserting the Portex^® device via the cricothyroid membrane in all cadavers studied was 62.5 and 39.1% by digital palpation and ultrasound, respectively. However, when stratified by the cadavers with difficult and impossible landmarks, conventional digital palpation resulted in only 8.3% of correct device insertion. In contrast, ultrasound guidance significantly increased correct device insertion by almost six-fold to 46.7%. Our findings are consistent with the literature showing that conventional digital palpation was inaccurate in identifying the cricothyroid membrane. Elliott et al.¹⁶  demonstrated that only 30% of digital palpation by anesthesiologists correctly identified the cricothyroid membrane of patients with various BMI. Furthermore, Aslani et al.¹⁷  highlighted inaccurate identification of the cricothyroid membrane in 14 of 15 obese patients (BMI > 30 kg/m²). There is evidence that increased pretracheal soft tissue and greater neck circumference can be associated with difficult laryngoscopy.¹⁸  These features could potentially lead to difficult palpable landmarks similar to the cadavers in our study.

Cricothyrotomies have been shown to lead to major complications such as esophageal perforation, subcutaneous emphysema, and hemorrhage due to improper device insertion.³  In this study, 15 cadavers had difficult or impossible neck landmarks, and the incidence rate of moderate–severe injuries using ultrasound-guided cricothyrotomy was only 33%. This was statistically significant when compared with the 100% incidence of moderate–severe injuries in the 12 cadavers of the digital palpation group.

The mean (SD) insertion time in the ultrasound group (196.1 [60.6]) was significantly longer than that in the digital palpation group (110.5 [46.9]). A possible explanation might be that the ultrasound group participants were novices at the technique for ultrasound-guided identification of the cricothyroid membrane. In addition, they only had one didactic and hands-on teaching session. A study performed by Nicholls et al.¹²  demonstrated a mean time of less than 30 s (24.3 s [20.2 s]) to visualize the cricothyroid membrane using ultrasound on patients with a variety of neck circumference and BMI. The physician participants in the study by Nicholls et al. who scanned the patients were not formally trained in this technique but regularly attended monthly didactic ultrasound teaching sessions and used ultrasound in their daily practice. Thus, it can be suggested that with routine practice and teaching, there may be room for improvement in this technique. An interesting finding of this previous study was the shorter time to visualization of the cricothyroid membrane in patients with higher BMI and difficult palpable neck landmarks. The justification suggested by these authors was similar to that provided by Aslani et al.¹⁷ ; that is, the additional tissue overlying the membrane could improve focus of the ultrasound beam to provide a better postimage acquisition resolution.

This study has some limitations. Percutaneous cricothyrotomy can be performed using a variety of techniques. In this study, we used only the trocar-over-needle technique with the Portex^® device, which appears to be more traumatic than other surgical airway techniques such as the Seldinger or scapel.¹³  Future studies comparing ultrasound guided versus conventional digital palpation on cricothyrotomy performances using different techniques merit further research. The formalin-fixed cadavers used in this study have firm pallid tissues, are not pliable, and do not reflect normal tissue handling characteristics. These changes could have made digital palpation of landmarks more difficult and challenging in comparison with ultrasound guidance. Fresh cadavers with normal tissue characteristics are ideal models for research; however, they are not readily available and cannot be preserved for longer duration. The limited sample size prevented potentially important findings, such as the greater than 20% difference in correct landmarking that could not be concluded as statistically different. Although this difference was large and probably clinically relevant, sample size was not large enough to be conclusive. It may be argued that, in a cohort of patients with difficult palpable landmarks, ultrasound guidance may lead to faster localization, increased accuracy, and less injuries when performing a cricothyrotomy in an emergency situation. It should be emphasized that we are not proposing to use of ultrasound guidance of the cricothyroid membrane in emergency airway crisis situations but rather to localize the cricothyroid membrane before airway management in patients with difficult airways and neck anatomy landmarks that are not well defined.

In conclusion, ultrasound guidance of the cricothyroid membrane improves correct device insertion and reduces the severity of injuries when performing cricothyrotomy using the Portex^® device on human cadavers. We propose that ultrasound guidance of the cricothyroid membrane and neck landmarks should be performed before airway management, particularly in patients with difficult palpable neck landmarks and difficult airways. There may be scope for studies in real-life patients who present with risk factors for accessing the cricothyroid membrane in an emergency although there might be challenges for adequately powering the study and an ethical aspect.

Support was provided solely from institutional and/or departmental sources.

The authors declare no competing interests.

Raw data and full protocol are not available.