The correlation of cervical symptoms with endotracheal intubation quality and preoperative airway assessment with radiological methods: A prospective observational study

Abstract

Airway management can be challenging for patients undergoing cervical spinal surgery. This study aimed to evaluate how the duration of the cervical pathology associated with neck pain and numbness affects airway anatomy and airway management, and to correlate radiological measurements. This was a single-center prospective observational study. Eighty-two patients with American Society of Anesthesiologists I to III were selected from those undergoing surgery for cervical pathology. The patients’ symptoms and duration of cervical disease were assessed. Skin-epiglottis, skin-hyoid, epiglottis-vocal cord, and hyomental distance measurements were performed using ultrasonography. The bilateral multifidus and longus colli muscles, presence of prevertebral fluid, thickening of the prevertebral soft tissue, increase in muscle signal intensity, pre-epiglottic depth of field, distance between the vocal cords, and epiglottis were evaluated using magnetic resonance imaging. Patients were intubated using a videolaryngoscope. Intubation times, the percentage of glottic opening, the number of intubation attempts, and complications were recorded during intubation. Intubation quality was evaluated based on intubation time, number of attempts, and the percentage of glottic opening achieved during videolaryngoscopy. The correlation between symptom duration and intubation quality was assessed. Radiological measurements of the airway were compared with the symptom duration. Symptom duration did not affect intubation parameters (P > .05). Skin-hyoid distance showed a positive correlation with pain duration (r = 0.410, P = .001) and numbness duration (r = 0.452, P < .001). The depth of the pre-epiglottic space increased with the duration of numbness (r = 0.325, P = .034). Epiglottis-vocal cord distance correlated with intubation time (r = 0.297, P = .026). Our study is significant in that it focuses on analyzing preoperative airway evaluation through radiological data and comparing these findings to the quality of intubation achieved with videolaryngoscopy. Radiological measurements such as skin-hyoid distance and pre-epiglottic space depth demonstrated measurable correlations with symptom duration, suggesting potential value in preoperative airway assessment. However, these anatomical variations did not significantly influence intubation performance when videolaryngoscopy was used.

1. Introduction

Cervical spine disorders can cause neck pain, with or without neurological symptoms. Cervical degenerative changes are the most common causes of both acute and chronic neck pain. Because of this degeneration, the spinal canal may narrow, the spinal cord may compress, and secondary spinal cord damage or dysfunction may develop. Some studies have detected some changes in the cervical muscles and soft tissue in the cervical region due to cervical pathologies.^[1] Studies have also examined cervical anatomical measurements in chronic neck pain.^[2,3] Magnetic resonance imaging (MRI) is usually a diagnostic tool used to detect these changes.^[1]

Airway management is difficult in patients who undergo cervical spinal surgery. These patients may have osteoarthritis, rheumatoid arthritis, ankylosing spondylitis, and neuromuscular disorders that disrupt airway anatomy and may complicate endotracheal intubation in anesthesia practice.^[1] These patients may have severe cervical spine instability or myelopathy at the spinal cord level, and serious neurological complications may develop with the intubation technique. For these reasons, patients undergoing cervical spine surgery are classified as being at risk for endotracheal intubation. Currently, patients who are expected to have a difficult airway are advised to use a videolaryngoscope.^[4] Conventional airway assessments may not predict difficult intubation in all patients, as highlighted in the most recent expert consensus on difficult airway evaluation.^[5] It is possible to detect airway changes by using radiological methods.

Ultrasonography (USG) measurements have become an evaluation method that can be used to evaluate difficult airways.^[6,7] However, ultrasound is not used sufficiently or frequently in airway management. Previous studies have attempted to detect difficult airways using magnetic resonance measurements. However, it is not as popular or widespread as an ultrasound tool. Routine MRI is not recommended to assess a patient’s airway before airway management.^[8]

The primary objective of this study was to evaluate the effect of cervical symptom duration on airway anatomy and intubation quality.

The secondary objective was to correlate radiological measurements obtained by USG and MRI with intubation-related parameters.

We hypothesized that the duration of cervical symptoms (pain and numbness) may influence radiological airway measurements, but not necessarily affect intubation quality when videolaryngoscopy is used.

2. Materials and methods

2.1. Study design

This was a single-center prospective observational study. It was performed between April 2021 and January 2022 at the Ankara University Faculty of Medicine Department of Anesthesiology and Reanimation. The database used in this study can be shared upon request by contacting the corresponding author.

2.2. Ethics statement

Ethical approval for this study was provided by the Ankara University Faculty of Medicine’s Human Research Ethics Committee, Ankara, Türkiye, on March 12, 2018 (Decision No: 05-312-18). The study was registered at ClinicalTrials.gov (ID: NCT05188807). Written informed consent was obtained from all patients.

2.3. Study population

The study population underwent cervical pathology surgery. Eighty-two patients were included in this study. Inclusion criteria comprised adult patients over 18 years old (American Society of Anesthesiologists [ASA] I to III) scheduled for elective cervical spine surgery, who underwent preoperative MRI. Patients scheduled for emergency surgery, patients with cervical tumors, patients with cognitive dysfunction, patients with cardiac instability (decompensated heart failure, recent myocardial infarction), patients who had undergone previous cervical surgery, and patients with cervical trauma were excluded from the study.

2.4. Data collection

During the preoperative anesthesia evaluation, all patients were informed about the study and provided written consent. Their cervical symptoms, including the presence and duration of neck pain and numbness, were documented.

An experienced anesthesiologist performed all airway assessments, USG measurements, and endotracheal intubations.

The primary variable of the study was the duration of cervical symptoms (pain and numbness). The secondary variables included radiological measurements obtained using USG and MRI (skin-hyoid distance, skin-epiglottis distance, epiglottis-vocal cord distance, hyomental distance, and pre-epiglottic space depth) and their correlations with intubation parameters. The sample size was determined by including all eligible patients who met the inclusion criteria during the designated study period.

Airway examinations of the patients were performed. The Mallampati score, mouth opening, neck movements (extension and flexion angles were measured with a goniometer), thyromental distance, sternomental distance, distance between the anterior incisors, upper lip bite test results, and neck circumference measurements were evaluated. Airway difficulty was determined using the Airway Difficulty Score.^[9] Airway measurements were performed using USG. Skin-hyoid distance (Fig. 1), skin-epiglottis distance (Fig. 2), distance between the epiglottis and midpoint of the vocal cords (Fig. 3), and hyomental distance (Fig. 4) were measured using USG (Fujifilm FC1 ultrasonography device [FUJIFILM Healthcare Corporation, Tokyo, Japan] was used for USG measurements). A 15-6 MHz linear probe (HFL50xp model; FUJIFILM Healthcare Corporation, Tokyo, Japan) was used in the transverse plane for skin-hyoid distance, skin-epiglottis distance, and epiglottis-vocal cord distance measurements. For the hyomental distance, measurements were taken with a 5-2 MHz curve probe (C60xf model; FUJIFILM Healthcare Corporation, Tokyo, Japan) in the sagittal plane. USG measurements were performed by a single anesthesiologist who had received specific training in airway USG. Although the anesthesiologist was aware of the patients’ clinical symptoms as part of the preoperative evaluation, the intubation outcomes were not known at the time of measurement. All USG data were reviewed and confirmed by an experienced radiologist. MRI evaluations were performed independently by another radiologist who was blinded to both clinical symptoms and intubation results. Measurements were made using cervical MRI, which was routinely applied to each patient before cervical surgery. MRI measurements were selected to represent structural and spatial characteristics of the upper airway and surrounding soft tissues that could potentially influence laryngoscopic visualization, as well as to evaluate anatomical alterations that may occur in patients with chronic cervical spine pathologies. At levels C2 to 3, C3 to 4, C4 to 5, C5 to 6, and C6 to 7 in the axial plane, the areas of the bilateral multifidus and longus colli muscles were assessed using T2 gradient echo sequence images. The presence of prevertebral fluid, thickening of the prevertebral soft tissue, and an increase in muscle signal intensity were assessed at these levels using sagittal T2-weighted imaging. The pre-epiglottic depth of the field was measured from the longest part of the craniocaudal region on sagittal fat-free T1-weighted images (Fig. 5). The distance between the vocal cords and epiglottis was calculated from sagittal fat-free T1-weighted images by measuring the length from the proximal end of the epiglottis at right angles to the plane parallel to the vocal cord levels.

Figure 1.

Skin-hyoid distance (shown with white arrow).

Figure 2.

Skin-epiglottis distance (shown with white arrow).

Figure 3.

Distance between epiglottis and midpoint of vocal cords (shown with white arrow).

Figure 4.

Hyomental distance (shown with white arrow).

Figure 5.

Depth of pre-epiglottic space measured by MRI (shown with yellow arrow). The distance between the vocal cords and epiglottis (shown with white arrow). MRI = magnetic resonance imaging.

When the patients came to the premedication unit, pharmacological premedication with intramuscular atropine was administered. Subsequently, the patients were placed in the operating room. In addition to routine ASA monitoring (oxygen saturation, noninvasive blood pressure measurement, electrocardiogram), Bispectral Index and train of four (TOF) monitoring were also performed on the patients. If there were no contraindications, the patients were sedated with 0.5 to 1 mcg kg⁻¹ fentanyl. The patients were preoxygenated with 100% FiO2 for 3 minutes at a flow rate of 5 L min⁻¹. After preoxygenation, general anesthesia was induced with 1 mg kg⁻¹ lidocaine and 2 to 3 mg kg⁻¹ propofol. After taking the reference measurement for TOF following induction, muscle relaxation was provided with 0.6 mg kg⁻¹ rocuronium. If needed, 0.5 to 1 mcg kg⁻¹ remifentanil was added. The patients were intubated when the TOF count was 0. Anesthesia was maintained with total intravenous anesthesia. All patients were intubated using a C-MAC D-blade (Karl Storz) and stylet. Intubation time was assessed as the time taken from the insertion of the blade beyond the incisors until the end-tidal carbon dioxide wave drawing. Intubation quality was evaluated based on intubation time, number of attempts, and the percentage of glottic opening achieved during videolaryngoscopy.

At the end of the procedure, patients were assessed for airway complications. The severity of trauma during intubation, bleeding, blood contamination to the blade tip or the tube at the end of extubation, esophageal intubation, tooth damage, sore throats independent of and associated with swallowing, speaking difficulty, pain and hoarseness, laryngospasm, and desaturation were assessed. Complications were divided into 4 groups according to their severity: mild, moderate, severe, and very severe.

2.5. Statistical analysis

SPSS 29.0 (IBM Corp., Armonk) was used to analyze the mean ± standard deviation, median (minimum–maximum) was used as a descriptor for quantitative variables, and the number of patients (percentage) was used for qualitative variables. Whether there is a difference between the categories of the qualitative variable that has 2 categories in terms of a quantitative variable, Student t test if normal distribution assumptions are met, and the Mann–Whitney U test if not. The Kruskal–Wallis H test was used to determine whether there was a difference between the categories of the qualitative variable with 2 categories in terms of the quantitative variable, as the assumptions of normal distribution were not met. Chi-square and Fisher exact tests were used to examine relationships between the 2 qualitative variables. The Spearman correlation test was used to examine the relationship between the 2 qualitative variables, because the assumptions of a normal distribution were not met. The statistical significance level was set as 0.05. As this was an exploratory analysis, no predefined cutoff values were used for radiological measurements. All distances were analyzed as continuous variables using correlation.

3. Results

This study included a total of 82 participants. Of these, 37 (45.1%) were male, and 45 (54.9%) were female. The mean age of the study population was 47.5 years. Among the participants, 42 patients (51.2%) were classified as ASA I, 37 (45.1%) as ASA II, and 3 (3.7%) as ASA III. All patients presented with cervical pain, and 66 (80.5%) also reported accompanying numbness. The demographic data of the patients are shown in Table 1.

Table 1.

The demographic data of the patients.

Gender, n (%)
Male	37 (45.1)
Female	45 (54.9)
Age, mean ± SD	47.50 ± 12.31
BMI, mean ± SD	28.31 ± 5.18
ASA status, n (%)
1	42 (51.2)
2	37 (45.1)
3	3 (3.7)
ADS, mean ± SD	7.89 ± 1.26
Cervical pain, n (%)	82 (100)
Numbness, n (%)	66 (80.5)

ADS = Airway Difficulty Score, ASA = American Society of Anesthesiologists, BMI = body mass index, SD = standard deviation.

There were no differences in intubation time, number of intubation attempts, or percentage of glottic opening values among the patients. Symptom duration did not affect intubation parameters (P > .05). The relationship between pain and numbness duration and the USG and MRI findings was also examined. Intubation parameters and radiological measurements are presented in Table 2. There was a moderately significant positive correlation between clinical pain duration and skin-hyoid distance (r = 0.410, 95% confidence interval [CI]: 0.17–0.60, P = .001). Similarly, a moderate positive correlation was found between numbness duration and skin-hyoid distance (r = 0.452, 95% CI: 0.24–0.63, P < .001). No significant correlations were found between symptom duration and other USG parameters, including skin-epiglottis distance, epiglottis-vocal cord distance, and hyomental distance (P > .05 for all). MRI analysis revealed a moderate correlation between the duration of numbness and the depth of the pre-epiglottic space (r = 0.325, 95% CI: 0.09–0.53, P = .034; Table 3). No significant correlations were found for the remaining parameters, including prevertebral soft tissue thickness, prevertebral fluid presence, muscle area, signal intensity, and epiglottis-vocal cord distance (P > .05 for all). Regardless of the pathological process, USG and MRI measurements were compared with intubation time. A weak positive correlation was found between intubation time and epiglottis-vocal cord distance measured using USG (r = 0.297, 95% CI: 0.04–0.52, P = .026). No significant correlations were observed for other parameters (P > .05).

Table 2.

USG and MRI measurements and intubation parameters.

Skin-epiglottis distance (mm), mean ± SD	1.54 ± 0.51
Skin-hyoid distance (mm), mean ± SD	0.87 ± 0.41
Distance between epiglottis and midpoint of vocal cords (mm), mean ± SD	0.76 ± 0.34
Hyomental distance (mm), mean ± SD	4.89 ± 0.54
Depth of pre-epiglottic space (mm), mean ± SD	19.83 ± 4.81
Intubation time (s), mean ± SD	32.45 ± 16.18
POGO, mean ± SD	87.28 ± 11.7
Number of intubation attempts, mean ± SD	1.01 ± 0.11

MRI = magnetic resonance imaging, POGO = percentage of glottic opening, SD = standard deviation, USG = ultrasonography.

Table 3.

Correlations between cervical symptom duration and radiological airway measurements.

Clinical variable	Radiological parameter*	Correlation coefficient (r)	P-value
Pain duration	Skin-epiglottis distance	0.238	.065
	Skin-hyoid distance	0.410	.001
	Epiglottis-vocal cord distance	−0.049	.734
	Hyomental distance	−0.012	.923
Numbness duration	Skin-epiglottis distance	0.253	.06
	Skin-hyoid distance	0.452	<.001
	Epiglottis-vocal cord distance	0.024	.870
	Hyomental distance	0.004	.974
Numbness duration	Pre-epiglottic space depth	0.325	.034

^*Skin-epiglottis distance, skin-hyoid distance, epiglottis-vocal cord distance, and hyomental distance were measured by ultrasonography. The depth of the pre-epiglottic space was measured by magnetic resonance imaging.

No complications developed in 66.3% of the patients when postoperative intubation-related complications and airway complications were assessed. No esophageal intubation or tooth damage was observed in any of the patients. Blood contamination of the endotracheal tube was observed in 1 patient. Sore throat was observed in 27 patients, 18 of whom were associated with swallowing. One patient had severe symptoms. Additional analgesics were also administered. Mild transient hoarseness was observed in 3 patients. Laryngospasm was observed in 4 patients after extubation. Symptoms regressed without additional intervention. Short-term desaturation was observed in 6 patients after extubation. Oxygen inhalation therapy was continued in these patients.

4. Discussion

In this study, significant correlations were observed between the duration of cervical symptoms and radiological airway parameters, particularly skin-hyoid distance and pre-epiglottic space depth. Despite these measurable anatomical variations, the quality of intubation remained unaffected when videolaryngoscopy was used, underscoring its reliability in patients with cervical pathology.

When the duration of neck pain and numbness in patients was compared with intubation, no significant relationship was found. This suggests that it is not always easy to predict the clinical symptoms of patients with cervical pathology and the effects of the duration of these symptoms on the intubation quality. It is also possible that the use of the videolaryngoscope improved the intubation process and concealed any potential relationships. It has been found that the skin-hyoid distance measured with USG is related to pain and numbness in patients, and this distance increases as the symptoms persist. The skin-hyoid distance measurement increased with the duration of pain and numbness. Upper airway ultrasound allows for the identification of important upper airway structures.^[10] Various studies have used USG to predict difficult airways. Studies have generally focused on the anterior neck soft tissues, distance from the skin to the epiglottis, pre-epiglottic areas, distance from the epiglottis to the vocal cords, hyomental distances, tongue thickness, distance from the skin to the tongue, and tongue volume.^[11] The distance from the skin to the epiglottis, hyomental distance, and tongue thickness are better predictors in recent studies.^[12] Gupta et al found that the ratio of the pre-epiglottic area to epiglottis-vocal cord distance can predict the Cormack–Lehane grade.^[13] The skin-epiglottis distance measurement was found to be the most sensitive (75% sensitivity) and specific (63.6% specificity) parameter for the prediction of difficult laryngoscopy in a study by Parameswari et al.^[14] Adhikari et al found higher anterior neck soft tissue measurements at the hyoid bone level in the difficult laryngoscopy group.^[15] In a study using indirect laryngoscopy, Altun et al determined a cutoff value of 0.835 cm for the hyoid-skin distance and 2.09 cm for the epiglottis-skin distance for the difficult airway.^[16]

In the literature, changes due to both cervical pathology and clinical duration have been observed with MRI. Several studies have investigated whiplash-related diseases.^[17–19] Studies have also examined cervical anatomical measurements in chronic neck pain.^[2,3] Some studies evaluated airway structures using MRI and correlated these findings with the difficulty of laryngoscopy. Münster et al found that laryngoscopy is more difficult when the vocal cords are located more proximal to the head.^[20] According to another study, the thickness of the anterior cervical soft tissue on MRI did not correlate with difficult laryngoscopy.^[21] In patients with cervical spine trauma, the thickness of the retropharyngeal space is associated with difficult intubation.^[22]

In our study, a moderately positive correlation was found between the duration of numbness and magnetic resonance-measured depth of the pre-epiglottic space. Anatomical changes due to cervical pathology that can be measured by USG and MRI were observed in our study. However, the relationship between these anatomical changes and intubation difficulty could not be demonstrated. The insignificant effect of possible anatomical changes on intubation quality can be explained by the use of a videolaryngoscope, which is a documented and effective method for the management of a difficult airway.^[4,23] Videolaryngoscopy, which has become the standard approach for endotracheal intubation, provides an improved glottic view and minimizes inter-operator variability in airway assessment.^[5] On the other hand, radiological measurements and intubation periods of the patients were compared regardless of the duration of cervical pathology. It has been shown that there is a correlation between the epiglottis-vocal cord distance measured by USG and intubation time from radiological evaluations. A positive correlation was found between the epiglottis-vocal cord distance and intubation time. However, a cutoff value could not be determined. This radiological measurement was predictive even in intubation with videolaryngoscopy.

The observed correlations between cervical symptom duration and certain radiological parameters suggest that these measurements may have practical value in preoperative airway assessment. Incorporating parameters such as skin-hyoid distance or pre-epiglottic space depth into airway evaluation could help anesthesiologists anticipate anatomical challenges in patients with cervical pathology. Such data may also guide the choice of airway management technique and equipment to optimize intubation safety.

This study has several limitations. First, it did not include a control group of patients with normal cervical anatomy, which limits the direct comparison of radiological airway parameters between healthy individuals and those with cervical pathology. Second, all patients were evaluated using a single videolaryngoscope model, which may restrict the generalizability of the findings to other airway devices or techniques. However, using a standardized videolaryngoscope ensured methodological consistency and minimized variability in intubation assessment. Third, the sample size was relatively small. Additionally, no specific classification was utilized to assess the complexity of videolaryngoscopy. One of the newly developed scoring systems could be employed to evaluate the level of difficulty associated with videolaryngoscopy. Alternatively, additional scores could be devised to assess the challenges posed by videolaryngoscopy. We did not establish a definitive cutoff point in our investigation as we were contrasting 2 quantitative datasets, namely measurements and intubation time.

Despite these limitations, this study found that intubation quality in cases of cervical pathology was unrelated to the duration of these symptoms. However, cervical disease can affect airway anatomy. For airway safety, this patient group must be considered at risk in terms of airway difficulty and intubation with a videolaryngoscope.

In conclusion, USG measures are being increasingly used and studied to predict difficult airway scenarios. The value of these measurements increases when they are examined in patient groups with similar characteristics. Radiological airway measurements showed significant correlations with the duration of cervical symptoms but did not influence intubation performance when videolaryngoscopy was used. These findings emphasize the clinical value of integrating USG and MRI data into preoperative airway assessment.

Author contributions

Conceptualization: Gülencan Yumuşak Ergin, Zekeriyya Alanoğlu, Melih Bozkurt, Başak Ceyda Meço.

Data curation: Gülencan Yumuşak Ergin, İhsan Doğan, Sena Ünal.

Investigation: Gülencan Yumuşak Ergin, Evren Üstüner, Elif Peker.

Resources: Gülencan Yumuşak Ergin.

Supervision: Zekeriyya Alanoğlu, Evren Üstüner, Elif Peker, Başak Ceyda Meço.

Methodology: Melih Bozkurt, Elif Peker, Başak Ceyda Meço.

Validation: Elif Peker, Başak Ceyda Meço.

Visualization: Sena Ünal.

Writing – original draft: Gülencan Yumuşak Ergin, Başak Ceyda Meço.

Writing – review & editing: Başak Ceyda Meço.