This study shows a significant reduction in diaphragmatic pillar thickness at the level of L1 and the celiac artery in most cases, except for right weaknesses. CT examination also revealed a significant increase in cupola height differences irrespective of the side, or type (paralysis or weakness) of dysfunction.
The absence of diaphragmatic pillar thinning in right-sided diaphragmatic weakness may be the consequence of different factors. Among the 7 patients with right-sided diaphragmatic weakness, only 4 had severe weakness (meaning greater muscle dysfunction with less than 40% of maximal thickening on ultrasound), while the weaknesses present on the left were all severe. Indeed, the mean thickening fraction in the case of weakness was 40% on the right, compared with 31% on the left. In addition, the right diaphragmatic pillar is stronger, thicker, and longer than the left pillar [10, 19]. On can hypothesize that the right pillar change is lower in the presence of simple diaphragmatic weakness.
With regard to diagnostic thresholds for diaphragmatic pillars thickness on CT, for right-sided diaphragmatic paralysis, ROC analysis identified a threshold of 3.0mm at the level of the celiac artery, and a threshold of 4.5mm at the level of L1 with good diagnostic performances [20]. Both thresholds showed good sensitivity, specificity, and PPV, but the pillar measurement at the L1 level was more sensitive than at the celiac artery level, with a higher area under the curve. In addition, the NPV was significantly higher at the L1 level. Therefore, for the diagnosis of right diaphragmatic paralysis, pillar measurement at the L1 level seems more relevant than at the celiac artery level. For left paralysis, ROC analysis determined a threshold of 2.6 mm for the diaphragmatic pillar at the level of the celiac artery. This threshold is identical in the left diaphragmatic weaknesses, since in our study we found no significant difference in pillar thickness between the paralysis group and the weakness group at this level. In addition, at the L1 level, weakness can be differentiated from left diaphragmatic paralysis. The threshold for paralysis at L1 was 3.8mm, with good diagnostic performance. For left diaphragmatic weakness, ROC analysis determined a threshold of 5.0mm at the L1 level with excellent diagnostic performance (specificity 100% and PPV 100%), making the diagnosis of left diaphragmatic weakness certain if the thickness is below the threshold.
The study by Sukkasem et al. found a threshold of 2.5mm for the diagnosis of diaphragmatic paralysis, on both the right and left, at the level of the celiac artery and L1 [10]. In this study, diaphragmatic pillar thickness was measured in patients with diaphragmatic paralysis versus patients with normal diaphragmatic function, and diaphragmatic function was assessed by fluoroscopy. The results of these two studies cannot be compared, as they do not use the same gold standard.
In the dysfunction group, the contralateral diaphragm showed compensation by an increase in amplitude during QB and VS in both right and left dysfunctions in ultrasound. This increase in the amplitude of the healthy contralateral diaphragm has been described as a neuronal compensation for the function of the contralateral hemi diaphragm in paralysis [12, 21, 22]. However, we did not find any hypertrophy of either the right or left healthy diaphragmatic pillar. CT scans even revealed thinner pillars at the L1 level on the healthy side of patients with dysfunction, compared with controls (p = 0.01 on the right and p = 0.09 on the left), probably due to global muscle weakness.
In the case of right diaphragmatic paralysis or weakness, ROC curve analysis determined a threshold of height difference between the two domes of 4.4cm and 3.5cm respectively, with perfect diagnostic performances of 100% [20], making the diagnosis of right diaphragmatic paralysis or weakness certain if the height is greater than the threshold.
For left dysfunction, diagnostic performances were good for paralysis but not for weakness with close values of height difference between the two domes (2.7cm and 2.5cm respectively). Thus, the height difference between the domes appears in our study to be the most reliable CT measurement for diagnosing DD. To our knowledge, this is the first time that precise threshold values for the diagnosis of DD based on the difference in cupola height have been reported.
DD is generally associated with a slight decrease in vital capacity, to approximately 75%, while TLC is generally preserved [23]. The greater decrease in volumes found in this study in the weakness group may be explained by causes other than DD itself. Indeed, among the 12 patients with a restrictive pattern in the weakness group, there were 6 patients with a history of COVID-19 (including 5 severe diseases requiring intensive care), one patient with a history of talc pleurisy, 2 patients were overweight, and 6 patients were moderately obese.
This study has certain limitations. The pillar thickness is supposed to vary with breathing (it decreases with expiration) [24]. Only one radiologist performed the measurements, intra- and inter-reproducibility were not evaluated, and CT scan were both enhanced and unenhanced which make more difficult the measurements in unenhanced CT. Thus, if the scan is performed during expiration, DD may be mistakenly assumed. Also, in our study, to validate correct inspiration during thoracic scans, we checked that the posterior part of the trachea did not bulge inwards. This check could not be performed for the single abdominal scan included in our study. Moreover, the scans were performed at different times, before or after the ultrasound. This time variable may affect pillar atrophy. Finally, there were more women in the paralysis group than in the weakness group (p < 0.001), and we can assume that the pillars are thinner in women, as is the case in ultrasound. However, in the study by Dovgan et al. [25], pillar thickness did not vary according to gender.