Diaphragm ultrasound varies by intercostal space, and the differences are measurable
A cross-sectional study in 337 healthy adults shows that diaphragm ultrasound is not a one-size-fits-all examination. Measurements shifted across the 7th, 8th, and 9th intercostal spaces, and the magnitude of those shifts was large enough to matter for routine practice, especially when clinicians rely on thickness metrics to infer respiratory muscle function.
The study examined diaphragmatic thickness, thickening fraction, and shear-wave elastography using both 2D and M-mode ultrasonography. End-inspiratory thickness and thickening fraction differed significantly between intercostal spaces, and the same measures also differed between imaging methods. In practical terms, a diaphragm measured at one rib space may not be directly interchangeable with the same diaphragm measured one or two spaces away. The muscle may be the same; the number is not.
Thickness and thickening fraction cannot be treated as interchangeable across levels
The clearest signal from the cohort is level dependence. Echogenicity classification did not differ significantly between the 8th and 9th intercostal spaces, but the quantitative metrics did. End-inspiratory thickness and thickening fraction showed statistically significant variation both by intercostal space and by whether the scan used 2D or M-mode imaging. For a muscle as thin and motion-sensitive as the diaphragm, even small technical differences can propagate into large interpretive differences at the bedside.
That matters because diaphragm ultrasound is often used to estimate respiratory effort, neuromuscular integrity, and mechanical performance. Thickening fraction is especially vulnerable to measurement conditions because it combines the diaphragm’s resting state and inspiratory excursion into a single derived number. If the baseline thickness changes with probe position, the downstream fraction changes too. This study makes the case for level-specific reference ranges rather than a universal threshold applied everywhere along the lower intercostal spaces.
The physiology is straightforward. The diaphragm is dome-shaped, anisotropic, and sensitive to probe angulation, rib shadowing, and local acoustic windows. Slight changes in scanning level alter the visible portion of the muscle and the quality of its fascial margins, which in turn changes how the software or operator defines thickness. In a structure this small, submillimeter differences can be clinically consequential.
The 9th intercostal space offered the best elastography image quality
Shear-wave elastography added another layer of nuance. Image quality was superior at the 9th intercostal space, with a mean score of 76.55 ± 7.17% compared with 74.97 ± 9.15% at the 8th intercostal space. Elastic modulus values themselves did not differ significantly between the two sites, so the advantage of the 9th space was technical rather than biologic. Better image quality can improve confidence in acquisition, reduce repeat scans, and make stiffness assessment more reproducible in practice.
This is the kind of result that changes workflow without changing pathology. If the elastography signal is similar but the image is cleaner at the 9th intercostal space, then sonographers have a rational default site for stiffness assessment. In a field where acoustic quality often determines whether a measurement is usable at all, that kind of operational guidance is valuable.
Routine scanning may favor 2D imaging when quality is limited
A particularly pragmatic finding was the observation that 2D imaging may be preferable in routine practice because it depends less on exceptionally high image quality. That preference does not mean 2D is inherently superior to M-mode for every question. It means that when the image window is imperfect, 2D may be more forgiving and more feasible across a broader range of patients and scanning conditions.
This has direct implications for respiratory and critical care ultrasound. Patients with altered body habitus, limited cooperation, or suboptimal windows may not provide consistently pristine images, and a method that tolerates modest degradation is more likely to be adopted. In real-world terms, the best technique is often the one that can be repeated reliably at the bedside.
A standardized protocol now has a stronger evidence base
Investigators developed a standardized protocol for diaphragmatic ultrasonography and outlined it in detail, but the more important point is why such standardization is necessary. Without a fixed approach to intercostal level, imaging mode, and measurement timing, diaphragm ultrasound risks producing numbers that look precise while actually reflecting operator variation. The present data argue for consistent acquisition rules before clinicians lean too heavily on any single cutoff.
The study enrolled healthy volunteers rather than patients with respiratory disease, which keeps the physiology clean but limits direct extrapolation to diseased diaphragms. A normal diaphragm and a fatigued, paralyzed, or mechanically disadvantaged diaphragm do not behave identically. Even so, normative work is the foundation for clinical interpretation. Reference standards cannot be built on unstable measurement practices.
Healthy-adult data define the measurement problem before disease enters the picture
This cohort of 337 healthy adults establishes a practical map for future use. The 8th and 9th intercostal spaces both appear feasible for routine diaphragmatic assessment, yet thickness measurements require level-specific interpretation because inter-level variability was significant. For SWE, the 9th intercostal space seems preferable on technical grounds. For day-to-day scanning, the study also suggests that 2D imaging may be the more robust default when image quality is less than ideal.
The authors performed paired statistical comparisons and found no significant difference in echogenicity classification between the 8th and 9th spaces, but the quantitative measures told a different story. That split between qualitative appearance and quantitative output is a familiar problem in sonography. A diaphragm can look acceptable at more than one level while still yielding non-equivalent measurements.
For clinicians who use diaphragm ultrasound to track ventilatory mechanics, this work is a reminder that the instrument and the site are part of the result. A thickness value is not just a property of the patient; it is also a property of where, how, and with what mode the scan was obtained. Until patient-based disease studies validate the same protocol in respiratory weakness, critical illness, and ventilator dependence, the safest approach is discipline in acquisition and caution in interpretation.