In 1990, Chilaiditi first reported the abnormal sign of the transverse colon interposed between the liver and right diaphragm using early roentgenology, and this phenomenon was therefore named eponymously8. In the present study, CS was seen in three of the seven cases of LOPD (43%), much higher than the frequency of approximately 0.25–0.28% in the general population27.
Although diaphragm weakness, one of the major risk factors of CS, was usually detected in NMD, especially in the advanced phase, few studies have evaluated the prevalence of CS in NMD. Our search only revealed investigations of CS in Duchenne muscular dystrophy (DMD), with a prevalence of 6.2%28, and myotonic dystrophy type 1 (DM1), with a prevalence of 12.5%29. The prevalence of NPPV use was not mentioned in either of those studies.
The reason for the prevalence of CS being much higher in LOPD than in healthy groups, or even in NMD with similarly impaired respiratory muscles, remains unclear because we did not compare CS between LOPD and other NMDs. One possible reason is that the intestine in patients with LOPD may be more easily interposed into the hepatodiaphragmatic space because of the increased elasticity of the intestines, following a gradual accumulation of glycogen in the smooth muscles of the intestines4–6. Also, considering the diaphragm was usually more severely affected in DMD than in LOPD, the high prevalence of CS in LOPD in the present study may be associated with the higher severity of the enrolled patients.
Causes of CS are mainly classified as follows30–31:
Bowel factor
i) huge colon or abnormally increased intestinal gas; or ii) floating colon;
Diaphragmatic factor: i) high diaphragmatic position due to muscle atrophy or degeneration; ii) diaphragmatic paresis due to diaphragmatic nerve distress; or iii) changing pressure in the thorax or dilation of the inferior thoracic aperture; and
Hepatic factor
i) dropping liver; ii) liver atrophy; iii) ligament weakness; or iv) adhesion of the liver.
CS can thus be caused by abnormalities of the liver, colon or right hemidiaphragm.
In comparing risk factors between CS and non-CS groups of LOPD patients, the severity of atrophy and fat infiltration in the right diaphragm, and the frequencies of an abnormal position of the right diaphragm, liver under the left diaphragm, and abnormally dilated bowel were all significantly higher in patients with CS. Those findings may indicate an elevated position of the right diaphragm and lowered position of the liver could widen the hepatodiaphragmatic space, and the dilated intestinal bowel could be more easily interposed into this space.
On the other hand, LBM, TFW, and BMI tended to be higher in the CS group than in the non-CS group, although the differences were not significant. Accumulation of visceral fat around the liver widens the hepatodiaphragmatic space, and may allow easier interposition of the intestine32. In the present study, all subjects showed advanced-stage disease and visceral fat accumulation in each group may have been similar, contributing to the lack of significant difference between groups.
Interestingly, NPPV resulting in aerophagia did not differ significantly between groups, despite being a known risk factor for CS 10. In terms of bowel dilation as a bowel factor for CS, NPPV was estimated to more easily result in development of CS compared to TPPV, because of the continuous entry of air into the stomach 21. In the present study, Patients 1 and 4, as non-CS patients with NPPV, suffered chronic recurring spontaneous pneumothorax. The left chronic hemopneumothorax in Patient 1 and right non-functional lung after recurrent pneumothorax in Patient 4 led to the mediastinum shifting rightward in both cases (Figure 4). Mediastinal shift to the right led to decreased space between the right diaphragm and liver, and may have thus prevented development of CS in these patients.
A number of limitations to this study should be considered. First, the number of subjects was quite small, because of the low prevalence of LOPD. Second, CS was not compared with other NMDs in the present study, and the high prevalence of CS in LOPD compared to other NMDs could have resulted from simple differences in clinical stage. Further investigation to clarify the prevalence of CS in each NMD and compare risk factors between diseases is warranted.
The routine evaluation of CXR should also play an important role in ruling out CS in patients with LOPD. If CS is suspected, further delineation of the liver and diaphragm by CT and referral to a gastroenterologist for further management and follow-up may be worth recommending. Preventing the risks associated with CS requires physician awareness and screening using at least CXR and CT to detect patients with CS.