Abstract
Introduction/Aims
Diaphragm ultrasound is increasingly used in the diagnosis of diaphragm dysfunction and to guide respiratory management in patients with neuromuscular disorders and those who are critically ill. However, the association between diaphragm ultrasound variables and demographic factors like age, sex, and body mass index (BMI) are understudied. Such relationships are important for correct interpretation of normative values and comparison with selected patients groups. The aim of this study was to determine the associations between diaphragm ultrasound variables and subject characteristics.
Methods
B‐mode ultrasound was used to image the diaphragm at the zone of apposition in 83 healthy subjects. Diaphragm thickness at resting end‐expiration (T end‐exp), diaphragm thickness at maximal end‐inspiration (T max‐insp), diaphragm thickening ratio (T max‐insp/T end‐exp), and diaphragm echogenicity were measured. Multivariate linear regression was used to explore the associations between diaphragm ultrasound variables and subject characteristics.
Results
T end‐exp, T max‐insp, and thickening ratio do not change with age whereas diaphragm echogenicity increases with age. The thickening ratio had a weak negative association with BMI, while T end‐exp was positively associated with BMI. Men had a larger T end‐exp and T max‐insp than women (T end‐exp 1.6 ± 0.5 and 1.4 ± 0.3 mm; p = .011, T max‐insp 3.8 ± 1.0 and 3.2 ± 0.9 mm; p = .004), but similar thickening ratios.
Discussion
Diaphragm thickness, thickening, and echogenicity measured with ultrasound are associated with factors such as age, BMI, and sex. Therefore, subject characteristics should be considered when interpreting diaphragm ultrasound measurements. In the absence of normative values, matched control groups are a prerequisite for research and in clinical practice.
Keywords: diaphragm, intensive care unit, neuromuscular disorders, normative values, ultrasound
Abbreviations
- AIC
Akaike's information criterion
- BMI
body mass index
- ICC
Intraclass correlation coefficient
- ICU
intensive care unit
- NMD
neuromuscular disorder
- T end‐exp
diaphragm thickness at resting end‐expiration
- T max‐insp
diaphragm thickness at maximal end‐inspiration
1. INTRODUCTION
Diaphragm weakness is a common feature in patients with neuromuscular disorders (NMDs) and those who are critically ill. In many NMDs, weakness of the diaphragm leads to dyspnea, sleep disturbances, and lung infections, and is a major contributor to death. 1 In critically ill patients, mechanical ventilation induces diaphragm injury and atrophy, leading to ventilator dependency and difficult weaning. 2 In both settings, reliable assessment of diaphragm function is to identify early signs of respiratory insufficiency, monitor disease progression, and guide individual respiratory management.
Ultrasound is increasingly used as tool to assess diaphragm function and structure in NMD patients and critically ill patients.3, 4 The costal diaphragm can be visualized at its insertion in the anterior thoracic wall, the so‐called zone of apposition. The change in thickness from expiration to inspiration is used to quantify diaphragm function.3, 4 The assessment of the muscle ultrasound gray level, or echogenicity, reflects structural alterations in muscles due to, for example, fibrosis and inflammation. 5 Echogenicity analysis of skeletal muscles is a well‐known and reliable tool for screening, diagnosing, and follow‐up of NMDs.5, 6
Thickness and echogenicity of peripheral muscles are dependent on subject characteristics such as age, body mass index (BMI), and sex. 7 However, it is unknown if diaphragm thickness and echogenicity also depend on these characteristics. Such relationships are important to identify those subject characteristics that should be taken into account when interpreting diaphragm ultrasound measurements in clinical practice or research. The aim of this study was to determine the association of diaphragm thickness and echogenicity with age, sex, and BMI using a standardized approach.
2. METHODS
This is a retrospective study of data collected as part of a project to determine reference limits for clinical practice at the Radboud University Medical Center, Nijmegen, The Netherlands, and was, therefore, exempt from ethical approval. All subjects provided their informed consent and procedures were performed in accordance with the ethical standards as laid down in the declaration of Helsinki. Ultrasound examinations were carried out by five laboratory technicians, each with 5–10 y of experience in muscle ultrasound.
Sample size estimation yielded a need for 80 subjects for regression‐based reference limits. 8 A decision was made to aim to enroll 10 healthy subjects, 5 males and 5 females, for each 10 y age category to ensure a balanced age and sex distribution. Subjects with any condition affecting the respiratory or skeletal muscle system were excluded. Demographic factors including age, sex, and BMI were collected. Subjects were recruited via advertisements on social media.
An Esaote MyLab Twice ultrasound machine (Esaote SpA, Genoa, Italy) equipped with a 3–13 MHz LA533 linear transducer was used to assess diaphragm thickness at resting end‐expiration (T end‐exp) and at maximal end‐inspiration (T max‐insp). Measurements were performed according to previously published methodology (See Supporting Information Methods, which are available online). 9 Thickening ratio was calculated as T max‐insp divided by T end‐exp.
Diaphragm echogenicity values were calculated from the images used for measurements of T end‐exp. Using custom developed software in Matlab (R2018a, Mathworks, Natick, MA, USA) and the trace method, a region‐of‐interest of diaphragm muscular tissue was manually selected. 10 Echogenicity was calculated as the mean pixel gray‐value of this region‐of‐interest and averaged over three measurements.
After checking linearity, independency, homoscedasticity, and normality, a stepwise multivariate linear regression was performed for echogenicity, thickening ratio, T end‐exp, and T max‐insp using R software. 11 A full factorial model was explored with age, sex, and BMI as covariates and their quadratic terms. To uphold independence between the covariates and their quadratic terms, age and BMI were centered. Akaike's information criterion (AIC) was used to determine the best fit. 12 This approach allows covariates to be included in the final regression model when they have considerable impact on model fit, even when there is no significant association between the covariate and the dependent variable.
Intraclass correlation coefficients (ICCs) were calculated on the three repeated measures during one session to assess intra‐rater reliability. ICC values below 0.50 were considered as poor reliability, 0.50–0.75 moderate, 0.76–0.90 good, and 0.91–1.00 excellent. 13
3. RESULTS
83 healthy subjects were recruited, as listed in Table 1. T max‐insp was missing in one subject. Four diaphragms, with a T end‐exp <1 mm, were too thin to permit determination of a reliable region‐of‐interest, and were excluded from echogenicity analysis.
TABLE 1.
Subject characteristics and outcomes for different categories
| Category | Sex (M/F) | Age (y) | BMI (kg/m2) | Echogenicity (0–255) | Thickening ratio | T end‐exp (mm) | T end‐insp (mm) |
|---|---|---|---|---|---|---|---|
| Age (y): 0–9 | 7/5 | 6.2 (2.3) | 16.6 (2.8) | 56.9 (9.1) | 2.4 (0.5) | 1.4 (0.4) | 3.2 (0.7) |
| Age (y): 10–19 | 4/6 | 16.5 (3.6) | 19.0 (2.4) | 64.5 (8.9) | 2.6 (0.5) | 1.3 (0.4) | 3.3 (0.6) |
| Age (y): 20–29 | 6/6 | 24.7 (3.1) | 21.6 (2.1) | 67.2 (7.8) | 2.3 (0.6) | 1.6 (0.4) | 3.5 (1.1) |
| Age (y): 30–39 | 5/5 | 35.3 (2.8) | 23.0 (3.0) | 69.8 (15.8) | 2.5 (0.5) | 1.6 (0.5) | 3.8 (1.2) |
| Age (y): 40–49 | 5/5 | 45.9 (2.9) | 23.7 (2.9) | 76.7 (13.6) | 2.4 (0.6) | 1.5 (0.4) | 3.4 (1.1) |
| Age (y): 50–59 | 4/5 | 55.9 (2.3) | 24.3 (2.1) | 75.3 (14.2) | 2.6 (0.5) | 1.5 (0.4) | 3.8 (0.9) |
| Age (y): 60–69 | 5/6 | 65.4 (3.0) | 24.3 (4.3) | 81.4 (9.5) | 2.4 (0.7) | 1.4 (0.4) | 3.3 (0.7) |
| Age (y): 70–80 | 4/5 | 74.3 (3.3) | 23.0 (1.8) | 84.3 (14.8) | 2.2 (0.6) | 1.8 (0.3) | 4.1 (1.4) |
| Sex: M | 40/0 | 38.2 (23.0) | 21.8 (4.0) | 70.4 (13.5) | 2.5 (0.6) | 1.6 (0.5) | 3.8 (1.0) |
| Sex: F | 0/43 | 39.7 (23.2) | 21.8 (3.6) | 72.3 (15.0) | 2.3 (0.5) | 1.4 (0.3) | 3.2 (0.9) |
| Age (y): 0–19 | 9/8 | 10.9 (6.0) | 17.7 (2.8) | 60.1 (9.6) | 2.5 (0.5) | 1.4 (0.4) | 3.3 (0.7) |
| Age (y): 20–80 | 31/35 | 49.2 (17.6) | 23.3 (2.9) | 75.5 (13.5) | 2.4 (0.6) | 1.6 (0.4) | 3.6 (1.1) |
| Sex: M, age (y): 0–19 | 9/0 | 11.2 (5.4) | 17.7 (3.2) | 60.3 (6.2) | 2.6 (0.6) | 1.4 (0.5) | 3.5 (0.7) |
| Sex: M, age (y): 20–80 | 29/0 | 48.5 (18.1) | 23.3 (3.1) | 74.2 (13.6) | 2.5 (0.6) | 1.7 (0.5) | 4.0 (1.1) |
| Sex: F, age (y): 0–19 | 0/9 | 10.6 (6.8) | 17.7 (2.5) | 60.0 (12.2) | 2.3 (0.3) | 1.3 (0.2) | 3.0 (0.6) |
| Sex: F, age (y): 20–80 | 0/32 | 49.8 (17.5) | 23.2 (2.8) | 76.7 (13.6) | 2.3 (0.6) | 1.4 (0.3) | 3.3 (1.0) |
| Total | 40/43 | 39.0 (22.9) | 21.8 (3.8) | 71.4 (14.3) | 2.4 (0.6) | 1.5 (0.4) | 3.5 (1.0) |
Note: Data are presented as mean (SD). Total mean values were used to center age and BMI.
Table 2 and Figure 1 present the regression analyses, and the specifications are presented in the supplement. Group average values were different between sexes for T end‐expand T max‐insp. An increase in age was associated with an increase in echogenicity. Age and age squared were not associated with thickening ratio, although they were included in the regression model based on AIC. An increase in BMI was associated with an increase in thickening ratio and a decrease in T end‐exp.
TABLE 2.
Regression models fitted to echogenicity, thickening ratio, T end‐exp, and T max‐insp
| Outcome | Regression formula | |
|---|---|---|
| Echogenicity |
|
|
| Thickening ratio |
|
|
| T end‐exp |
|
|
| T max‐insp |
|
Note: cAge: centered age, calculated by subtracting 39.0 from age in years. cBMI: centered BMI, calculated by subtracting 21.8 from BMI in kg/m2. cAge2: centered age squared.
FIGURE 1.
Graphical representation of fitted regression models. Each dot represent the measurement of a single subject, the line represents predicted values for each subject, where BMI was set fixed at the adult group mean of 23.3 kg/m2, and the shaded area represents the 95% prediction interval. A, Diaphragm echogenicity. B, Diaphragm thickening ratio. C, End‐expiratory diaphragm thickness. D, End‐inspiratory diaphragm thickness
Intra‐observer reliability of the three echogenicity measurements, T end‐expand T max‐insp were all excellent, with ICCs of 0.93 (0.89–0.96), 0.92 (0.89–0.94), and 0.93 (0.91–0.95), respectively.
4. DISCUSSION
In healthy subject across a wide age range, we found that diaphragm thickness and thickening do not change with age, whereas diaphragm echogenicity increases with age. Sex and BMI have small effects on diaphragm thickness, but not on diaphragm thickening.
Diaphragm thickness and thickening ratio values presented in this study are comparable to previously reported values.14, 15, 16, 17, 18 Interestingly, Boon and colleagues, using an apparently similar approach reported a T end‐exp of 3.3 mm compared to 1.6 mm in our adult population. 9 Note that BMI in our adult population was considerably lower (23.3 vs. 27.2 kg/m2). Although we and others9, 19 found an association between BMI and diaphragm thickness, this cannot fully explain the relatively large difference. Small differences in methodology may also contribute to differences in thickness. The variability in reported values for diaphragm thickness highlights the importance of collecting center‐specific normative values. The larger variation in T max‐insp compared to T end‐exp is in accordance with other reports. 20 This may be attributed to variation in subjects’ performance of the maneuver to take a deep breath and affects the level of diaphragm recruitment and the anatomical position of the diaphragm relative to the ultrasound probe.
Our study shows that diaphragm thickness is constant over a wide age range. This differs from skeletal limb muscles that generally increase in thickness until 30–40 y, after which muscle thickness decreases (sarcopenia). 7 Apparently, changes in respiratory demand with aging, for example due to decreased respiratory compliance, do not seem to be accompanied by changes in diaphragm muscle mass but other mechanisms, like changes in motor control. 21
In agreement with previous studies, diaphragm thickness was larger in men than women.9, 15, 20 The calculation of thickening ratio cancels out sex differences. We also identified BMI as significant predictor of T end‐exp and thickening ratio, but the size of this effect was small. The latter may result from the narrow range of BMI in our adult population (23.3 ± 2.9 kg/m2). It has been shown that outside this range BMI has larger effects on diaphragm thickness and thickening ratio. 19 Overall, little variance in thickening ratio was explained by our regression model. Thus, the lower limit of normal (5th percentile) of our healthy population, ie, 1.6, can be used as cut off value for a normal diaphragm thickening ratio regardless of the subjects’ age, sex, and BMI. 22
The absence of changes in diaphragm thickening with age is in apparent contrast to age‐related changes in pulmonary function and respiratory muscle strength.23, 24 Furthermore, diaphragm function, as assessed with transdiaphragmatic pressure, is only weakly correlated with diaphragm thickening. 25 Therefore, the relationship between the pressure generating capacity of the diaphragm and diaphragm thickening warrants further research. This may be better reflected by other ultrasound modes such as strain imaging or shear wave elastography.26, 27
Different ultrasound devices produce different grayscale images, meaning that our diaphragm echogenicity values cannot be used with different ultrasound devices.5, 28 However, the observed association with age is expected to exist with values obtained from a different machine, as sarcopenia is an inherent property of aging muscle. 7 Normative values should be collected for each different ultrasound device, and age‐matched controls should be included when normative values are unavailable. For example, a recent study on diaphragm echogenicity compared mechanically ventilated patients, median age 59 y, with healthy controls, median age 27 y. 29 Our data imply that such a comparison is likely not valid. Furthermore, multiple ultrasound devices were used, complicating comparison of echogenicity values.
This study has limitations. First, the range in BMI was narrow. This impairs the interpretation of diaphragm ultrasound data of a patient outside this range. Second, normative data as presented in this study may have low generalizability to other centers. However, the associations with age, sex, and BMI are independent of measurement technique and device settings.
5. CONCLUSIONS
The associations between diaphragm thickness, thickening, and echogenicity and age, BMI, and sex highlight the importance of taking subject characteristics into account when interpreting diaphragm ultrasound measurements in clinical practice and research.
CONFLICT OF INTEREST
Nens van Alfen works as an ultrasound consultant for Dynacure and performs editorial services for Wiley Publishing Inc; all payments go to their employer. The remaining authors have no conflicts of interest.
STATEMENT ON ETHICAL PUBLICATION
We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
Supporting information
Appendix S1 Supporting Information.
Table S1 Regression model specifications.
ACKNOWLEDGMENTS
Several authors of this publication are members of the Netherlands Neuromuscar Center (NL‐NMD) and the European Reference Netwerk for rare neuromuscular diseases (EURO‐NMD).
van Doorn JLM, Wijntjes J, Saris CGJ, Ottenheijm CAC, van Alfen N, Doorduin J. Association of diaphragm thickness and echogenicity with age, sex, and body mass index in healthy subjects. Muscle & Nerve. 2022;66(2):197‐202. doi: 10.1002/mus.27639
DATA AVAILABILITY STATEMENT
Research data are not shared.
REFERENCES
- 1. Laghi F, Tobin MJ. Disorders of the respiratory muscles. Am J Respir Crit Care Med. 2003;168(1):10‐48. [DOI] [PubMed] [Google Scholar]
- 2. Goligher EC, Brochard LJ, Reid WD, et al. Diaphragmatic myotrauma: a mediator of prolonged ventilation and poor patient outcomes in acute respiratory failure. Lancet Respir Med. 2019;7(1):90‐98. [DOI] [PubMed] [Google Scholar]
- 3. Tuinman PR, Jonkman AH, Dres M, et al. Respiratory muscle ultrasonography: methodology, basic and advanced principles and clinical applications in ICU and ED patients‐a narrative review. Intensive Care Med. 2020;46(4):594‐605. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. van Doorn JLM, Pennati F, Hansen HHG, van Engelen BGM, Aliverti A, Doorduin J. Respiratory muscle imaging by ultrasound and MRI in neuromuscular disorders. Eur Respir J. 2021;58:2100137. [DOI] [PubMed] [Google Scholar]
- 5. Wijntjes J, van Alfen N. Muscle ultrasound: present state and future opportunities. Muscle Nerve. 2021;63(4):455‐466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. van Alfen N, Gijsbertse K, de Korte CL. How useful is muscle ultrasound in the diagnostic workup of neuromuscular diseases? Curr Opin Neurol. 2018;31(5):568‐574. [DOI] [PubMed] [Google Scholar]
- 7. Pillen S, Arts IM, Zwarts MJ. Muscle ultrasound in neuromuscular disorders. Muscle Nerve. 2008;37(6):679‐693. [DOI] [PubMed] [Google Scholar]
- 8. Virtanen A, Kairisto V, Uusipaikka E. Regression‐based reference limits: determination of sufficient sample size. Clin Chem. 1998;44(11):2353‐2358. [PubMed] [Google Scholar]
- 9. Boon AJ, Harper CJ, Ghahfarokhi LS, Strommen JA, Watson JC, Sorenson EJ. Two‐dimensional ultrasound imaging of the diaphragm: quantitative values in normal subjects. Muscle Nerve. 2013;47(6):884‐889. [DOI] [PubMed] [Google Scholar]
- 10. Sarwal A, Parry SM, Berry MJ, et al. Interobserver reliability of quantitative muscle sonographic analysis in the critically ill population. J Ultrasound Med. 2015;34(7):1191‐1200. [DOI] [PubMed] [Google Scholar]
- 11. R Core Team . R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing; 2019. [Google Scholar]
- 12. Halsey LG. The reign of the p−value is over: what alternative analyses could we employ to fill the power vacuum? Biol Lett. 2019;15(5):20190174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Koo TK, Li MY. A guideline of selecting and reporting Intraclass correlation coefficients for reliability research. J Chiropr Med. 2016;15(2):155‐163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Hiwatani Y, Sakata M, Miwa H. Ultrasonography of the diaphragm in amyotrophic lateral sclerosis: clinical significance in assessment of respiratory functions. Amyotroph Lateral Scler Frontotemporal Degener. 2013;14(2):127‐131. [DOI] [PubMed] [Google Scholar]
- 15. Carrillo‐Esper R, Pérez‐Calatayud ÁA, Arch‐Tirado E, et al. Standardization of sonographic diaphragm thickness evaluations in healthy volunteers. Respir Care. 2016;61(7):920‐924. [DOI] [PubMed] [Google Scholar]
- 16. Sartucci F, Pelagatti A, Santin M, Bocci T, Dolciotti C, Bongioanni P. Diaphragm ultrasonography in amyotrophic lateral sclerosis: a diagnostic tool to assess ventilatory dysfunction and disease severity. Neurol Sci. 2019;40(10):2065‐2071. [DOI] [PubMed] [Google Scholar]
- 17. Henke C, Spiesshoefer J, Kabitz HJ, et al. Characteristics of respiratory muscle involvement in myotonic dystrophy type 1. Neuromuscul Disord. 2020;30(1):17‐27. [DOI] [PubMed] [Google Scholar]
- 18. Ruggeri P, Lo Monaco L, Musumeci O, et al. Ultrasound assessment of diaphragm function in patients with late‐onset Pompe disease. Neurol Sci. 2020;41:2175‐2184. [DOI] [PubMed] [Google Scholar]
- 19. Kooragayalu S, Lim L, Aldrich TK. Diaphragm thickness and function in obese individuals. B29. Respiratory muscle at work: is it ‘all about that bass’? American Thoracic Society International Conference Abstracts: American Thoracic Society; 2015. p. A2676‐A.
- 20. Boussuges A, Rives S, Finance J, et al. Ultrasound assessment of diaphragm thickness and thickening: reference values and limits of normality when in a seated position. Front Med (Lausanne). 2021;8(1982):742703. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Fogarty MJ, Mantilla CB, Sieck GC. Breathing: motor control of diaphragm muscle. Physiology (Bethesda). 2018;33(2):113‐126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Culver BH. How should the lower limit of the normal range be defined? Respir Care 2012;57(1):136–45; discussion 43–5. [DOI] [PubMed] [Google Scholar]
- 23. Quanjer PH, Stanojevic S, Cole TJ, et al. Multi‐ethnic reference values for spirometry for the 3–95‐yr age range: the global lung function 2012 equations. Eur Respir J. 2012;40(6):1324‐1343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Laveneziana P, Albuquerque A, Aliverti A, et al. ERS statement on respiratory muscle testing at rest and during exercise. Eur Respir J. 2019;53(6):1801214. [DOI] [PubMed] [Google Scholar]
- 25. Poulard T, Bachasson D, Fossé Q, et al. Poor correlation between diaphragm thickening fraction and transdiaphragmatic pressure in mechanically ventilated patients and healthy subjects. Anesthesiology. 2021;136(1):162‐175. [DOI] [PubMed] [Google Scholar]
- 26. Oppersma E, Hatam N, Doorduin J, et al. Functional assessment of the diaphragm by speckle tracking ultrasound during inspiratory loading. J Appl Physiol (1985). 2017;123(5):1063‐1070. [DOI] [PubMed] [Google Scholar]
- 27. Bachasson D, Dres M, Niérat MC, et al. Diaphragm shear modulus reflects transdiaphragmatic pressure during isovolumetric inspiratory efforts and ventilation against inspiratory loading. J Appl Physiol (1985). 2019;126(3):699‐707. [DOI] [PubMed] [Google Scholar]
- 28. Pillen S, Boon A, Van Alfen N. Muscle ultrasound. Handb Clin Neurol. 2016;136:843‐853. [DOI] [PubMed] [Google Scholar]
- 29. Coiffard B, Riegler S, Sklar MC, et al. Diaphragm echodensity in mechanically ventilated patients: a description of technique and outcomes. Crit Care. 2021;25(1):64. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Appendix S1 Supporting Information.
Table S1 Regression model specifications.
Data Availability Statement
Research data are not shared.