Secondary Logo

Journal Logo

Novel Measurements of the Length of the Subglottic Airway in Infants and Young Children

Sirisopana, Metee MD*; Saint-Martin, Christine MD; Wang, Ning Nan*; Manoukian, John MD; Nguyen, Lily H. P. MD; Brown, Karen A. MD*

doi: 10.1213/ANE.0b013e3182991d42
Pediatric Anesthesiology: Research Report
Free

BACKGROUND: To date, the lengths of the subglottic and tracheal airway segments have been measured from autopsy specimens. Images of the head and neck obtained from computerized tomography (CT) provide an alternate method. Our objective in this study was to identify anatomic landmarks from CT scans in infants and young children to estimate the lengths of the subglottic and tracheal airway segments and to correlate these lengths with age.

METHODS: We performed a retrospective analysis of CT images of the neck for various diagnostic indications in children ≤3 years. We obtained planes of reconstruction at the level of the vocal cords (VCs), cricoid cartilage, and carina (C) which were parallel to each other and perpendicular to sagittal long axis of the trachea. The lengths of the subglottic airway (LengthSG) and total length of the laryngotracheal airway (LengthVC–C) were measured from the distance between, respectively, the VC versus cricoid cartilage and the VC versus C planes of reconstruction. Tracheal length was then calculated as the difference between LengthVC–C and LengthSG.

RESULTS: Fifty-six children met the inclusion criteria. There were 29 boys. The median weight was 10.7 kg (range 3.1–19.0 kg). Regression analysis yielded mean LengthSG (mm) = 7.8 + 0.03·corrected age (months), r2 = 0.07, P = 0.056; lower and upper 95% confidence interval for β = 0.03 were −0.001 and 0061. The mean LengthSG was 8.4 mm with an SD of 1.4 mm. The 95th percentile for LengthSG was 10.8 mm, and the 5% to 95% interquartile range was 4.9 mm. The estimate for the 95% confidence interval of the 95th percentile was between 10.2 and 11.3 mm. The LengthVC–C increased with age: mean LengthVC–C (cm) = 5.3 + 0.05·corrected age (months), r2 = 0.7, P < 0.001. Tracheal length also increased with age: mean tracheal length (cm) = 4.5 + 0.05·corrected age (months), r2 = 0.6, P < 0.001.

CONCLUSION: We report a novel estimate method for the lengths of the airway segments between the VC and C in 56 infants and young children and suggest that the growth characteristics of the subglottic and tracheal airway may differ.

Published ahead of print June 11, 2013.

From the Departments of *Pediatric Anesthesia, Diagnostic Radiology, and Otolaryngology, Head and Neck Surgery, McGill University Health Center—Montreal Children’s Hospital, Montreal, Quebec, Canada.

Accepted for publication April 11, 2013.

Published ahead of print June 11, 2013.

Funding: Dr. Brown is supported by the Queen Elizabeth Hospital of Montreal Foundation Chair in Pediatric Anesthesia, McGill University, Montreal, Quebec, Canada. Dr. Sirisopana was supported by the Louis Sessenwein Pediatric Anesthesia Fellowship, Montreal Children’s Hospital, Montreal, Quebec, Canada.

The authors declare no conflicts of interest.

Reprints will not be available from the authors.

Address correspondence to Karen A. Brown, MD, Department of Anesthesia, Montreal Children’s Hospital, 2300 Tupper St., Room C-1118, Montreal, Quebec H3H 1P3, Canada. Address e-mail to karen.brown@mcgill.ca.

Whereas intubation of the trachea with a cuffed tracheal tube (TT) has been standard practice for adult patients for decades, the use of cuffed TTs in children has only recently gained popularity.1,2 Recent modifications in cuffed TTs have incorporated a cuff-free subglottic tube shaft which is designed to position the cuff beyond the cricoid ring.3–5 This design feature is achieved with an age-specific distance between the proximal end of the cuff and a glottic depth mark (henceforth referred to as GDM) to be positioned at the level of vocal cords (VCs). However, both the rationale for a cuff-free subglottic tube shaft and its ideal length remain controversial.3–7 Our interest is the latter controversy, the age-specific length of the subglottic airway, with a focus on infants and young children.

Postmortem measurements of the subglottic airway have provided the reference values for the design of the age-specific cuff-free subglottic tube shaft.3 However, the validity of these of the age-specific lengths has not been tested in vivo. Indeed these validation studies would be difficult to perform as (1) the cricoid cartilage (CC) in the immature airway lacks calcification, (2) the exit from the cricoid ring is obscured by the walls of a TT, and (3) the cuff itself is radiolucent.

In its anterior aspect, the CC includes the cricothyroid membrane, facets for articulation with the thyroid cartilage, and the cricoid ring. The posterior CC (PCC) delimits the segment of the airway between the VC and the exit from the CC. The height of the PCC (henceforth referred to as LengthPCC), therefore, is a measure of the length of the subglottic airway. During fetal life, the LengthPCC increases in a quasilinear fashion to achieve 7.8 to 11 mm at term.8–12 From Table 2 reported by Wysocki et al.,13 in boys aged 0 to 2 years, the mean LengthPCC was 11.9 mm with an SD of 3.3 mm. However, the coefficient of variation for LengthPCC in boys was 29%, a value 3-fold greater compared with men (8%) suggesting a larger variability in LengthPCC in children.

As the number of autopsy specimens in young children is limited, we were interested in alternate methods to measure the length of PCC. Computed tomography (CT) affords one such method. In the axial CT planes, the contour of the air column is influenced by the contiguous anatomical structures. Characteristic configurations have been used to identify airway segments. Hudgins et al.14 reported the typical shape of the supraglottis, rima glottidis, and subglottis to be, respectively, a triangle, a teardrop, and an oval. The contour of the tracheal airway is typically rounded anteriorly and flattened posteriorly.14 This approach has been used to assess the cross-sectional area of the pediatric airway from magnetic resonance images (MRIs).15 To our knowledge, it has not been used to assess the length dimension of the child’s airway. Because our diagnostic imaging department has compiled a large electronic database, our objective was to apply the approach of Hudgins et al.14 in a retrospective analysis of the CT scans of the neck. Our primary objective was to identify the segment of airway between the VC and exit from the CC to estimate the length of the subglottic airway segment. The principle aims were to (1) assess the length of the subglottic airway segment in infants and young children and (2) to assess its growth relationship. A secondary aim was to assess the effect of head position on its length. Additional aims were to identify the segment of airway between the VC and carina (C) and to evaluate the effect of age and head position on the length estimates of this segment of airway.

Back to Top | Article Outline

METHODS

To assess the feasibility of this approach, a query to our medical imaging Picture Archiving and Communication System reported that 72 CT studies of the neck had been performed in nonintubated infants aged younger than 3 years, between March 5, 2009 and May 18, 2012. An initial assessment suggested that 90% of the CT images would allow assessment of the airway between the VC and C. With IRB approval, a retrospective analysis of these CT images and the relevant medical dossiers was performed.

Inclusion criteria were (1) corrected age (see below) younger than 3 years and (2) a CT scan of the airway which included the larynx, trachea, and C. Exclusion criteria were diseases of the laryngeal or tracheal airway, intubation of the trachea, and a compromised image quality due to breathing artifacts.

Allowing for clinical indications, the CT scans were obtained with IV injection of nonionic iodine contrast (Omnipaque 300, GE Health care, Princeton, NJ), maximum dose of 2 mL per kg of body weight, scan delay of 30 seconds, and an injection rate of 1.5 to 2.0 mL per second. Helical acquisition was performed with slice thickness of 2.5 mm, pitch of 1.5 in patients younger than 1 year and a slice thickness of 5 mm, pitch of 1 in patients aged 1 to 3 years. Postprocessed 2.5 mm reconstructions were obtained. Multiplanar reformations were obtained in the sagittal plane with contiguous slices of 1.5-mm thickness. Transverse oblique slices, parallel to each other, and oriented perpendicular to the long axis of the trachea, were obtained at the level of the VC, CC, and C (see below). Each level was identified by the slice location number along the craniocaudal axis of the trachea on sagittal reformation. As the planes of reconstruction obtained at the level of the VC, CC, and C were parallel to each other, this allowed electronic calculation of the distance among the 3 anatomical landmarks by the difference in their location number.

Back to Top | Article Outline

Medical Chart Review

The medical dossiers were reviewed, and the gestational age at birth, birth date, gender and date of radiologic examination, weight at the time of study, medical conditions, and radiologic diagnosis were recorded. The gestational age at birth and the postnatal age at the time of study were recorded to calculate the corrected age. We defined a month as having 28 days.

Back to Top | Article Outline

Analysis of the Radiographic Images

Head Position

Our usual practice is to sedate young children who require CT scans, and we prefer not to disturb a sleeping child. The position of the head had been not standardized in exact degrees of flexion–extension. From the lateral ScoutView, we determined the angle between the lower aspect of the horizontal ramus of the mandible and the posterior spinal line of the cervical spine (AngleMandible–Spine) (Fig. 1).

Figure 1

Figure 1

Back to Top | Article Outline

Identification of Subglottic and Tracheal Airway Segments

The level of the VC was identified as the cranial most level with a teardrop shape below the laryngeal ventricle (Fig. 2A). The lumen within the CC was identified by its ovoid contour. The LengthPCC in infants may exceed 1 cm.8–13 and as the reconstruction slices were 1.5 mm thick, several reconstruction slices exhibited an oval contour. The exit from the CC was defined as the caudal most level with an ovoid contour, (Fig. 2B) above the first tracheal ring (recognized by the flattening of its posterior wall).14 The C was identified in the lung windows by a figure 8 shaped lumen (Fig. 2C). Sagittal reformations were obtained along the airway (Fig. 2D). Axial oblique slices, oriented perpendicular to the long axis of the trachea, were obtained for VC, CC, and C, and each anatomic level was identified by the slice location value along the craniocaudal axis of the sagittal reformatted planes of reconstruction. As these reformatted planes of reconstruction were parallel to each other, this allowed an electronic calculation of the distance between the anatomic levels (see Fig. 2 for example). The length of the subglottis (LengthSG) was calculated as the distance between the VC and CC. The total length of airway between the VC and C (LengthVC–C) was calculated as the distance between the VC and C.

Figure 2

Figure 2

The expert CS-M trained 2 scorers (MS and NNW) to analyze the CT images. In a subset of 12 films, the agreement (bias and 95% limits of agreement (LoA) for LengthSG) between expert CS-M and scorers was determined. The bias ± SD for LengthSG between CS-M and scorer MS was −0.26 ± 0.59 mm (95% LoA −1.4; 0.9 mm). The bias for LengthSG between CS-M and scorer NNW was 0.25 ± 1.38 mm (95% LoA −2.5; 3.0 mm).

The 2 scorers read the films independently. The final value for both LengthSG and LengthVC–C was the averaged value. The intrascorer mean ± SD bias for the LengthSG was −0.17 ± 0.42 mm (95% LoA −0.99; 0.65 mm). Interscorer mean ± SD bias for LengthSG of was 0.29 ± 0.62 mm (95% LoA −0.95; 1.53 mm: Δ2.5 mm). Differences in LengthSG exceeding 1.5 mm were reviewed, and the final value was determined by CS-M.

The intrascorer mean ± SD bias for the LengthVC–C was −0.02 ± 0.42 cm (95% LoA −0.84; 0.79 cm). The interscorer mean ± SD bias for LengthVC–C was 0.15 ± 0.23 cm (95% LoA −0.31; 0.61 cm: Δ0.9 cm). Differences in LengthVC–C exceeding 1.5 cm were reviewed, and the final value was determined by CS-M.

Back to Top | Article Outline

Tracheal Length

The tracheal length between the CC and C was calculated as:

Back to Top | Article Outline

Statistical Analysis

Data were summarized as mean and SD, interquartile range, the 90% prediction interval, and 95% confidence intervals (CIs) for the mean and/or 90th and 95th percentiles.16 The relationships between age and head position versus the mean length of the airway segments were assessed with regression analysis and the coefficient of determination (r2). Statistical significance was defined as P < 0.05.

Back to Top | Article Outline

RESULTS

In total, 74 CT scans were read. Eighteen CT scans were excluded: corrected age >3 years = 8. Four scans were excluded for poor image quality (movement artifact = 1, VC nodule = 1, VC paralysis =1, technically poor film = 1). One child with Trisomy 21 was excluded. Five children had repeat studies before the age of 3 years. The first study was included in the final analysis.

Fifty-six children met the inclusion criteria and were included in the final analysis. The majority were healthy (n = 54); asthma = 1; renal failure = 1. All but 2 had a term birth. There were 29 boys. The median weight was 10.7 kg (range 3.1–19.0 kg). The radiologic diagnoses varied: normal = 12; cervical adenopathy = 17; retropharyngeal or parotid abcess = 9; thyroglossal or branchial cyst/sinus = 7, neck mass = 7; mastoiditis = 2, thrombosis = 2.

Back to Top | Article Outline

Head Position

The AngleMandible–Spine ranged from 45.3° to 88.6°. The mean AngleMandible–Spine was 68.5° with an SD of 9.5°. Clinical studies ascertain the head position with the method reported by Boidin,17 i.e., the angle between a horizontal surface and a line drawn through the external auditory meatus and the superior orbital margin, hereafter referred to as the Boidin Angle. The representative ScoutView in Figure 1 contrasts the methods to determine the Boidin Angle and AngleMandible–Spine. A Boidin Angle >110° corresponds to an AngleMandible–Spine >80°. The mean AngleMandible–Spine was 68.5°, corresponding to a Boidin Angle of 80° < 110°. Note that Jordi Ritz et al.18 defined a flexed head position as a Boidin Angle 80° <110°.

Back to Top | Article Outline

Airway Segments

LengthSG

The LengthSG is reported in millimeters. Regression analysis yielded Equation (1):

From Equation (1), the LengthSG age sensitivity was 0.4 mm per year. The lower and upper 95% CI for β = 0.03 were −0.001 and 0.061, respectively. The upper 95% confidence limit for β = 0.061 predicts that at 1 year LengthSG would increase by 0.8 mm and LengthSG would equal 8.6 mm. By 3 years of age, LengthSG would increase 2.4 mm and LengthSG would be 10.8 mm. We interpret the above to mean that between birth and 3 years, age had a minor effect on LengthSG. Regression of the relationship between AngleMandible–Spine and LengthSG showed mean LengthSG (mm) = 9.69 − 0.19·AngleMandible–Spine (degrees), r2 = 0.02, P = 0.33. In addition, within the range of AngleMandible–Spine observed in the current study, head position minimally influenced the value of LengthSG.

The scatterplot of LengthSG and corrected age is shown in Figure 3. The mean LengthSG was 8.4 mm with an SD of 1.4 mm. At a confidence level of 90%, where α = (1 − confidence level), α = 0.1, in ordered data, the lower and upper end points were found in respectively, patient #3 and patient #54; yielding a 90% prediction interval between 6.0 and 10.6 mm. The 95th percentile for LengthSG was 10.8 mm (5%–95% interquartile range = 4.9 mm). The estimate for the 95% CI of the 90th percentile was between 9.7 and 10.7 mm. The estimate for the 95% CI of the 95th percentile was between 10.2 and 11.3 mm. The maximum value for LengthSG, in 2 infants, aged 9 and 13 months, was 12.5 mm.

Figure 3

Figure 3

Back to Top | Article Outline

LengthVC–C

Forty-six studies exhibited a Figure 8 contour in the lung views. The LengthVC–C is reported in centimeters. The minimum and maximum LengthVC–C were 4.3 and 7.8 cm, respectively. In contrast to LengthSG, age had an important influence on LengthVC–C. The scatterplot of LengthVC–C and corrected age is shown in Figure 4. Regression analysis yielded Equation (2):

Figure 4

Figure 4

Therefore, the LengthVC–C age sensitivity was 0.7 cm per year. The lower and upper 95% CI for β = 0.05 were 0.04 and 0.07, respectively.

The regression for LengthVC–C and weight yielded mean LengthVC–C (cm) = 4.5 + 0.17·weight (kg), r2 = 0.5, P < 0.001. The relationship between LengthVC–C and the AngleMandible–Spine was mean LengthVC–C (cm) = 6.4 − 0.2·AngleMandible–Spine (degrees), r2 = 0.001, P = 0.9 suggesting that within the range of AngleMandible–Spine observed, head position had a small influence on LengthVC–C.

Back to Top | Article Outline

Tracheal Length

Tracheal length was calculated in 46 children. Tracheal length is reported in centimeters. The minimum and maximum tracheal lengths were 3.6 and 6.8 cm, respectively. The scatterplot of tracheal length and age is shown in Figure 5. Regression of the relationship between tracheal length and corrected age yielded Equation (3):

Figure 5

Figure 5

The lower and upper 95% CI for β = 0.05 were 0.04 and 0.06, respectively. The tracheal length and LengthVC–C age sensitivities were similar.

Back to Top | Article Outline

Analysis of the 5 Children with Repeat CT Scans

Five children had a CT scan within 12 days to 18 months of the initial study.(Table 1) In 4 infants, the time interval was within 3 months. Comparing the first and second CT scans, in patients #1 and #2 the AngleMandible–Spine was similar and in patients #3 and #4, the head position differed. In all 5 patients the differences in LengthSG between the first and second CT scans were <2 mm and occurring within the Δ2.5 mm interscorer 95% LoA for LengthSG. These results are consistent with the notion that age and head position minimally influenced the value of LengthSG.

Table 1

Table 1

Back to Top | Article Outline

DISCUSSION

To our knowledge, this is the first large study of infants and young children to measure the length of the subglottic airway from CT images. The mean LengthSG was 8.4 mm with an SD of 1.4 mm. We are 97% confident, given the sample size of 56, that the next future measurement of LengthSG would be between 5.5 and 12.5 mm.

Two limitations of the measurement technique influenced our data. First, because as the AngleMandible–Spine ranged from 45.3° to 88.6°, the measurements were made in a flexed head position. The reported lengths therefore represent minimum values and should approximate measurements obtained from postmortem specimens (see discussion below). Second, the accuracy of the measurements relied on the correct identification of the anatomic structures delimiting the airway segment. The level of the VC was easily identified as the cranial most level with a teardrop shape below the laryngeal ventricle. Similarly, the C was easily identified in the lung window. The exit from the CC was defined as the most caudad oval airway contour. However, the true exit from the CC would lie between CC and the first tracheal ring. In the reformat of the CT images, the true location of the exit from the CC may have been underestimated by 1 slice thickness (1.5 mm). Therefore, we may have underestimated the true length of the subglottic airway and overestimated the true length of the tracheal airway (see discussion below).

Back to Top | Article Outline

LengthSG

Holzki et al.12 dissected the airway longitudinally and reported the distance between the VC and exit from the cricoid ring to be 11, 13, and 14 mm in a neonate, 1-year-old, and 2-year-old child, respectively. In the current study, the LengthSG age sensitivity did not reach statistical significance. This may reflect a limitation of our method as the upper CI for β = 0.03 (Equation 1) was 0.06 mm per month. Therefore, the maximal increase in LengthSG by 3 years of age would be 2.4 mm, a value within the interscorer LoA (Δ2.5 mm) for LengthSG.

The LengthSG delimits the segment of airway between the VC and exit from the cricoid ring. In children younger than 3 years of age, the mean LengthSG was 8.4 mm with an SD of 1.4 mm. The PCC delimits this same segment of airway. In specimens from fetuses and infants, 35 to 40 weeks gestational age, Schild10 reported an average LengthPCC of 8.4 mm. In term infants, 40 to 50 weeks gestational age, Fayoux et al.8 reported that LengthPCC ranged from 8 to 11 mm. In young children (n = 7) younger than 2 years, Wysocki et al.13 reported the mean LengthPCC of 11.9 mm with an SD of 3.3 mm. The LengthPCC is measured from the external aspect of the structures which enclose the subglottic airway. In contrast, LengthSG is estimated from the luminal aspect of the anatomic structures delimiting the cranial (VC) and caudad (exit from the cricoid ring) extent of the subglottic airway. A priori one would expect the LengthPCC to be larger than LengthSG. It is likely that LengthSG underestimated the true length of the subglottic airway by as much as 1.5 mm (see above). If the true length of the subglottic airway in boys aged 0 to 2 years is the mean LengthPCC of 11.9 mm, reported by Wysocki et al.,13 then LengthSG may have underestimated the true length of the subglottic airway by as much as 12%.

Weiss et al.3 report the length of the cuff-free subglottic tube shaft for TTs intended for use in children aged 9 to 13 months to be 9 to 10 mm. However, the 95th percentile for LengthSG was 10.8 mm and the estimate for the 95% CI of the 95th percentile was between 10.2 and 11.3 mm. A cuff-free subglottic tube shaft of 10 mm is too short to position the cuff beyond the exit from the cricoid ring in all children. In infants and young children positioning the cuff beyond the cricoid ring may require a deeper insertion than is currently recommended by manufacturers of these cuffed TTs.

This deeper insertion would be of particular importance in the extended head position. During fiberoptic bronchoscopy, Jordi Ritz et al.18 reported that as the head position moved from flexion to extension (increasing Boidin Angle), the GDM, for cuffed TTs of internal diameter 3.5 and 4.0 mm, moved as much as 0.5 cm in the cranial direction. We interpret this finding to mean that during extension of the head either the length of the subglottic airway shortened or the cuff was pulled in a cranial direction to lie within the lumen of the subglottic airway. Because the cartilaginous structures of the CC are likely to exhibit poor extensibility, it is probable that extension of the head will pull the cuff into the lumen of the subglottic airway. This has important clinical implications as changes in head position occur frequently in intubated patients particularly in an intensive care setting. Although factors associated with injury to the subglottic airway are likely to be multifactorial,6 both the pressure exerted by the cuff and its sharp edges have been linked to injury of the airway mucosa.1,4,12 In young children, a cuff-free subglottic tube shaft length >10 mm is required to guarantee a cuff position below the cricoid ring in all patients and all head positions. In infants and young children, correct positioning of the cuff may require deeper insertion than is currently recommended.

Back to Top | Article Outline

LengthVC–C

Griscom and Wohl20 also measured LengthVC–C from CT images. In 13 children younger than 2 years, they reported a mean LengthVC–C 5.4 cm with an SD of 0.7 cm. To compare our data with these results, we observed children ≤2 years (n = 38). In this group of children, the mean LengthVC–C of 5.9 with an SD of 0.6 cm is in good agreement with the data reported by Griscom and Wohl.20

The LengthVC–C ranged from 4.3 to 7.8 cm. As the mean (SD) AngleMandible–Spine was 68.5° (9.5°), the head position in the current study was flexion. Larger values would be expected in the extended head position.18 Indeed, Fearon and Whalen21 measuring LengthVC–C with rigid bronchoscopy (extended head position) reported values ranging from 5 to 9 cm between birth and 18 months of age (Table 2).

Table 2

Table 2

The LengthVC–C increased with age such that mean LengthVC–C (cm) = 5.3 + 0.05 age (months), r2 = 0.7 (Fig. 4). As the value for β was 0.05 cm per month, by 19 months LengthVC–C age sensitivity exceeded the interscorer LoA for LengthVC–C (0.92 cm). During fiberoptic bronchoscopy, in the neutral head position, Jordi Ritz et al.18 measured the LengthVC–C (cm) = 5.8 + 0.03·age (months), r2 = 0.5 (Fig. 6, Dr. Eva-Maria Jordi Ritz, Spezialärztin Anästhesie, Universitätskinderspital beider Basel, Basel, Switzerland, personal communication, 2012). Inspection of their data (Fig. 6) reveals 3 parallel regression lines for the flexed, neutral, and extended head position. In the flexed head position, the y-intercept is approximately 5.6 cm, suggesting the CT method and the fiberoptic bronchoscopic method yield similar values for LengthVC–C. Weiss et al.3 also measured LengthVC–C with fiberoptic bronchoscopy in the neutral head position but reported lower values: LengthVC–C (cm) = 4.7 + 0.4; age (years), r2 = 0.85. In the current study, we focused on infants and children under 3 years and report a LengthVC–C age sensitivity of 0.7 cm per year. The lower LengthVC–C age sensitivity reported by Jordi Ritz et al.18 and Weiss et al.3 may reflect differences in inclusion criteria. Jordi Ritz et al.18 excluded infants younger than 13 months of age. Weiss et al.3 included older children and adolescents. During development, LengthVC–C age sensitivities likely decrease with age. In addition, the study design may have contributed to the different results as both general anesthesia and neuromuscular blockade decrease lung volume and thereby tracheal length.21

Figure 6

Figure 6

Back to Top | Article Outline

Tracheal Length

Classically, tracheal length is defined as the airway segment between the exit from the cricoid ring and the bifurcation of the trachea. Without the influence of head position and lung volume, measurements from autopsy specimens represent a minimum estimate for tracheal length. In 1955, Hall22 reported a tracheal length at birth of 4.5 cm. In 2008, Fayoux8 reported a tracheal length in term infants of 3.0 to 4.0 cm. As the head position in the current study was flexed, the values for tracheal length should approximate those obtained from postmortem studies. From Equation (3), the tracheal length at term would be 4.5 cm. In the reformat of the CT images, the true location of the exit from the CC may have been underestimated by as much as 1 slice thickness (1.5 mm, see above). Because tracheal length is related to age, the maximal overestimate (i.e., in the newborn) would have been 3%. Our values for tracheal length are in good agreement with measurements from autopsy specimens.

The International Standards Organization specifies that for cuffed TTs of internal diameter 3.5 mm, the allowable distance between the tip and the machine end of the cuff is 35 mm.19Figure 5 supports the notion that even in the flexed head position (shortest tracheal length), the tracheal length in infants younger than 3 years would be of sufficient length to accommodate this dimension. However, if the desired location of the tip of the TT is the mid-trachea, the effective length of trachea available for cuff placement would be shortened by 2 to 3 cm. In young children, there may be insufficient length to allow the tip of the TT to lie in the mid trachea and the cuff to lie beyond the exit of the cricoid ring as illustrated in Figure 7A.

Figure 7

Figure 7

Cuffed TTs that feature a cuff-free subglottic tube shaft include a GDM to be placed at the level of the VC.3–5 Inherent in this feature is the notion that the same GDM may be used to position the cuff and the tip of a TT. This assumption has not been validated. Figure 7B shows that the GDM has been placed between the VC. Whereas the tip of the TT has been correctly positioned, the machine end of the cuff abuts on the inferior edge of the cricoid ring. In the current study, we report a LengthSG age sensitivity of 0.4 mm per year and the LengthVC–C age sensitivity of 0.7 cm per year. In the design and manufacture of TTs with a cuff-free subglottic tube shaft, these different age sensitivities have implications to the optimal distances between the GDM, the cuff, and the tip of the TT. The use of the same GDM to correctly place the cuff and tip of a TT requires validation in clinical practice.

The GDM may be as wide as 2 mm,23 a width representing <4% of LengthVC–C, but 24% of the subglottic airway. The margin of safety for the correct placement of the cuff is therefore smaller than for the correct positioning of the tip of a TT, a concern raised previously by Ho et al.7 In infants and young children, positioning of the cuff beyond the cricoid ring may require a deeper insertion than is currently recommended.3 However, a deeper insertion may risk endobronchial intubation especially if the head is subsequently flexed. This may be of particular importance in neonates as Todres et al.24 reported the tip of the TT shifted 7 to 28 mm when the head was moved between full flexion and full extension. In orally versus nasally intubated neonates, the mean range of movement was 14.3 and 16.8 mm, respectively.24

Back to Top | Article Outline

Conclusion

We report a novel study of airway dimensions from CT images of the head and neck in 56 infants and children younger than 3 years. The mean LengthSG was 8.4 mm with an SD of 1.4 mm. The 95th percentile for LengthSG was 10.8 mm, and the estimate for the 95% CI of the 95th percentile was between 10.2 and 11.3 mm. Whereas, the LengthSG age sensitivity was 0.4 mm per year, the LengthVC–C age sensitivity was 0.7 cm per year.

The CT method offers an alternate method to postmortem studies to assess the dimensions of the subglottic and tracheal airway during development and may therefore be of interest to pediatric anesthesiologists, neonatologists, and otolaryngologists in the study of the airway in children. Improvements in spatial resolution of MRI and combinations of CT assessment of the lumen plus MRI assessment of the laryngeal cartilages may for the future prove useful in the study of the airway during development.

Back to Top | Article Outline

DISCLOSURES

Name: Metee Sirisopana, MD.

Contribution: This author helped design the study and prepare the manuscript.

Attestation: Dr. Sirisopana approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.

Name: Christine Saint-Martin, MD.

Contribution: This author helped design the study and prepare the manuscript.

Attestation: Dr. Saint-Martin approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.

Name: Ning Nan Wang.

Contribution: This author helped design the study and prepare the manuscript.

Attestation: Ning Nan Wang approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.

Name: John Manoukian, MD.

Contribution: This author helped design the study and prepare the manuscript.

Attestation: Dr. Manoukian approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.

Name: Lily H. P. Nguyen, MD.

Contribution: This author helped design the study and prepare the manuscript.

Attestation: Dr. Nguyen approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.

Name: Karen A. Brown, MD.

Contribution: This author helped design the study and prepare the manuscript.

Attestation: Dr. Brown approved the final manuscript, attests to the integrity of the original data and the analysis reported in this manuscript, and is the archival author.

This manuscript was handled by: Peter J. Davis, MD.

Back to Top | Article Outline

ACKNOWLEDGMENTS

We acknowledge and thank Roula Cacolyris for her invaluable help in the preparation and submission of the manuscript. We thank Xun Zhang, PhD, Biostatician, Research Institute McGill University Health Centre and Carlos Robles Rubio, MSc, Department of Biomedical Engineering, McGill University for their advice and expertise in the statistical analysis of the data. Carlos Robles Rubio was supported in part by the Mexican National Council for Science and Technology.

Back to Top | Article Outline

REFERENCES

1. Leong L, Black AE. The design of pediatric tracheal tubes. Paediatr Anaesth. 2009(19 Suppl 1):38–45
2. Weiss M, Dullenkopf A, Fischer JE, Keller C, Gerber ACEuropean Paediatric Endotracheal Intubation Study Group. . Prospective randomized controlled multi-centre trial of cuffed or uncuffed endotracheal tubes in small children. Br J Anaesth. 2009;103:867–73
3. Weiss M, Gerber AC, Dullenkopf A. Appropriate placement of intubation depth marks in a new cuffed paediatric tracheal tube. Br J Anaesth. 2005;94:80–7
4. Salgo B, Schmitz A, Henze G, Stutz K, Dullenkopf A, Neff S, Gerber AC, Weiss M. Evaluation of a new recommendation for improved cuffed tracheal tube size selection in infants and small children. Acta Anaesthesiol Scand. 2006;50:557–61
5. Weiss M, Balmer C, Dullenkopf A, Knirsch W, Gerber ACh, Bauersfeld U, Berger F. Intubation depth markings allow an improved positioning of endotracheal tubes in children. Can J Anaesth. 2005;52:721–6
6. James I. Cuffed tubes in children. Paediatr Anaesth. 2001;11:259–63
7. Ho AMH, Aun CST, Karmakar MK. The margin of safety associated with the use of cuffed paediatric tracheal tubes. Anaesthesia. 2002;57:169–82
8. Fayoux P, Marciniak B, Devisme L, Storme L. Prenatal and early postnatal morphogenesis and growth of human laryngotracheal structures. J Anat. 2008;213:86–92
9. Miklaszewska D, Gawlikowska-Sroka A, Czerwiński F, Dzieciołowska-Baran E, Adamiec E. A morphometric study of prenatal development of the human larynx. Ann Acad Med Stetin. 2010;56:103–6
10. Schild JA. Relationship of laryngeal dimensions to body size and gestational age in premature neonates and small infants. Laryngoscope. 1984;94:1284–92
11. Harjeet K, Aggarwal A, Sahni D, Batra YK, Rakesh SV, Subramanyam R. Anatomical dimensions of larynx, epiglottis and cricoid cartilage in foetuses and their relationship with crown rump length. Surg Radiol Anat. 2010;32:675–81
12. Holzki J, Laschat M, Puder C. Stridor is not a scientifically valid outcome measure for assessing airway injury. Paediatr Anaesth. 2009(19 Suppl 1):180–97
13. Wysocki J, Kielska E, Orszulak P, Reymond J. Measurements of pre- and postpubertal human larynx: a cadaver study. Surg Radiol Anat. 2008;30:191–9
14. Hudgins PA, Siegel J, Jacobs I, Abramowsky CR. The normal pediatric larynx on CT and MR. AJNR Am J Neuroradiol. 1997;18:239–45
15. Litman RS, Weissend EE, Shibata D, Westesson PL. Developmental changes of laryngeal dimensions in unparalyzed, sedated children. Anesthesiology. 2003;98:41–5
16. Hahn GJ, Meeker WQ Statistical Intervals: A Guide for Practitioners. 1991 New York, NY John Wiley & Sons, Inc
17. Boidin MP. Airway patency in the unconscious patient. Br J Anaesth. 1985;57:306–10
18. Jordi Ritz EM, Von Ungern-Sternberg BS, Keller K, Frei FJ, Erb TO. The impact of head position on the cuff and tube tip position of preformed oral tracheal tubes in young children. Anaesthesia. 2008;63:604–9
19. International Standard 5361: Anaesthetic and respiratory equipment—Tracheal tubes and connectors. 2012 Geneva, Switzerland International Organization for Standardization
20. Griscom NT, Wohl ME. Dimensions of the growing trachea related to age and gender. AJR Am J Roentgenol. 1986;146:233–7
21. Fearon B, Whalen JS. Tracheal dimensions in the living infant. Ann Otol Rhinol Laryngol. 1967;76:965–74
22. Hall JE. The physiology of respiration in infants and young children. Proc R Soc Med. 1955;48:761–4
23. Loew A, Thibeault DW. A new and safe method to control the depth of endotracheal intubation in neonates. Pediatrics. 1974;54:506–8
24. Todres ID, deBros F, Kramer SS, Moylan FM, Shannon DC. Endotracheal tube displacement in the newborn infant. J Pediatr. 1976;89:126–7
© 2013 International Anesthesia Research Society