This study determined seated reference values for Tdi, TF, and excursion in 109 healthy Japanese individuals. This is the largest-ever study to include simultaneous measurement of Tdi, TF, and excursion in seated Asian subjects and provides a benchmark for normal seated values in the Asian population.

The success rate for measurement of Tdi and TF was 100% regardless of side or breathing pattern. The main problem in measuring excursion was the overlapping of the lungs during inspiration, but fat attenuation was not a significant problem. Although diaphragm ultrasonography is more helpful in obtaining measurements during DB than during QB, [8] the success rate for excursion in DB was lower than that for Tdi and TF measurements. Therefore, measurement of TF, which has a higher success rate during DB and is less influenced by sex, height, and BMI, might be more useful than measurement of excursion.

Previous reports have indicated that it is easier to measure excursion than TF [6]. However, our study yielded different results. This inconsistency might be explained by differences in BMI between studies. The mean BMI in our study was 22.1, and the percentage of subjects with BMI ≥ 30 was 1%. Given that subcutaneous fat can hinder visibility with the linear transducer used for measuring TF, our high success rate when measuring Tdi may reflect a low obesity rate. The average BMI varies markedly from country to country. For example, the average adult BMI in Japan is 23.6 for men and 21.8 for women, 28.8 and 28.9 in the US, and 27.3 and 27.0 in the UK [18]. In Japan, only 5.4% of men and 3.6% of women have a BMI ≥ 30; these rates are considerably lower than those in Western countries [18, 19]. In populations where obesity is not a significant issue, as in our study, it might be advisable to first consider measuring the TF.

Past reports of Tdi at FRC range from 1.6 to 3.8 mm in men and 1.4 to 2.7 mm in women [1, 8, 13, 14, 20,21,22,23,24]. The results of the present study are consistent with previous reports. Our results for the LLN in men are close to those of Cardenas et al. and Boussuges et al.; however, the mean Tdi in women was more than 40% less than in the earlier studies [1, 20]. One of the criteria for a diagnosis of diaphragmatic weakness is a Tdi of < 2 mm at end expiration [2, 8, 25]; however, considering past reports and our present findings, many healthy individuals fall below this standard, suggesting that this criterion may be inappropriate for diagnosing diaphragmatic paralysis.

Our finding that the diaphragm was significantly thicker in men than in women is consistent with past reports [1, 8, 24, 26, 27]. However, in multiple linear regression analysis adjusted for sex, height, and BMI, there was no association with sex (right, p = 0.1; left, p = 0.92), but there was a statistically significant relationship with BMI (right, p < 0.01; left, p < 0.01). Other studies have also reported Tdi at FRC showed positive correlations with BMI [1, 26, 28]. Our results support these findings regarding the correlation with BMI but not with sex. It is unlikely that our failure to find a statistically significant correlation with sex was attributable to an insufficient sample size, given that the size of our study population was adequate when compared with those in the previous studies. While it appears that the diaphragm is generally thicker in men than in women, Tdi may be influenced more by BMI than by sex. As previously mentioned, the average BMI varies widely by country [18]. Considering that Tdi is influenced by BMI, standard values developed in Western countries, where the average BMI is significantly higher than that in Asian countries, may be inappropriate for use in Japan and other Asian populations.

During QB, even healthy individuals may have a change in Tdi of < 10% or no change at all [1, 27]. In our study, the TF was < 20% during QB in two cases (2%) on the right and in six cases (6%) on the left, and there was one case (1%) with a TF < 10% on the left. Therefore, TF in DB is more useful than TF in QB for evaluating diaphragmatic function and diagnosing diaphragm paralysis [1, 27].

TF is markedly affected by body position, with changes increasing in the order of supine, seated, and standing [10, 11]. In studies that examined changes in TF according to body position, the average TF values in DB were 60% in the supine position, 97% in the seated position, and 174% in the standing position [11]. Therefore, our results should be compared with previously reported measurements obtained in the seated position. However, there are limited data on normal seated TF values. In the only large-scale study of seated TF [1], which included 200 healthy individuals, the mean TF (LLN) in DB was 106% (40%)/112% (39%) for men and 116% (39%)/121% (48%) for women, with no apparent sex differences. Our mean values are close to those in the previous report. Although we calculated the LLN in the same way as the previous report [1] (i.e., the mean ± a standard deviation of 1.95), our results were about 20% lower than in their report.

A value of TF should be considered as indicating dysfunction if it is greater than the criterion for complete paralysis (TF < 20%) and less than the LLN [2]. However, a consensus has not yet been reached on the cut-off value for dysfunction [6]. Considering that the LLN was in the 20% range in our study, the original criterion for diaphragmatic paralysis (TF < 20%) was calculated from data obtained in the standing position, [25] and the fact that the seated TF is smaller than the standing TF [11] suggests there may be room for reconsideration of application of a TF < 20% as a criterion for diagnosis of diaphragmatic paralysis when measurements are obtained in the seated position.

Our multiple linear regression analysis adjusted for sex, height, and BMI found no association of the TF in DB with any of these three parameters. Tdi is affected by BMI, so it is difficult to refer to Western values in Asians, whose physique differs from that of Westerners. TF in DB is less affected by BMI and other factors, it may be used as a more generalized value.

Diaphragmatic dome excursion is also affected by body position, but, unlike TF, becomes larger in the supine position than in the seated position [8, 9]. Since abdominal compliance is less when seated, diaphragm dome excursion may be less for a given degree of diaphragm activation or tidal volume. Therefore, our results should be compared with those previously obtained in the seated position. However, there are few relevant reports. In the only large-scale study, Boussuges et al. measured excursion in the seated position and reported that the mean (LLN) values in QB were 1.9 cm (0.9 cm) on the right and 2.0 cm (0.9 cm) on the left in men and 1.7 cm (0.9 cm) and 1.7 cm (0.9 cm), respectively, on the right in women. In DB, the mean (LLN) values were 6.6 cm (4.1 cm) on the right and 6.7 cm (4.2 cm) on the left in men and 5.4 cm (3.3 cm) and 5.4 cm (3.2 cm), respectively, in women [12]. Our findings were equivalent to those of Boussuges et al. in QB, but were 1.5–2.0 cm smaller in DB in both sexes. Moreover, the LLN was 40–60% smaller in both QB and DB for both sexes. Multiple linear regression analysis identified a statistically significant association of excursion with female sex (p < 0.01) and BMI (p = 0.03) on the right side in DB but no association on the left side. The lack of a significant association on the left may reflect the inadequate sample size caused by the low rate of successful depiction. One reason why our excursion values in DB were smaller than those in the study by Boussuges et al. could be that the mean BMI was lower in our study subjects than in those in their study (22.5 vs. 25.0 for men and 21.7 vs. 25.0 for women). [12] It has been proposed that the criterion for diaphragm dysfunction with excursion in QB should be < 2 cm [6]. Considering past findings and our present results and that seated excursion is smaller than supine excursion, [8, 9] another criterion should be considered for the seated position.

This study has several limitations. The success rate for measurement of Tdi was higher than that for excursion, which might reflect the lower obesity rate in our study population. Therefore, it is unclear whether our results are applicable to populations with higher obesity rates. Further study is needed on a wider range of age groups, body size, and races.

In summary, we have determined reference values for use when performing diaphragm echography in a seated position. Considering that Tdi and excursion are influenced by BMI, it may not be possible to directly apply the reference values developed for the West in Asian countries, which have different obesity rates. The TF in DB does not appear to be affected by sex, height, or BMI and could be utilized more widely than other criteria. The current standards for diagnosis of diaphragmatic paralysis may need to be reconsidered for use in a seated position, given that many healthy individuals meet the criteria when measured in this manner.