CASE I~EPORT Differentiation between sigmoid volvulus and free air on supine abdominal radiographs: the 'liver overlap sign' versus the 'football sign' AbstractWFC van Gelderen Two cases are presented which emphasize the difficulty of d ifferenti ati ng between a sigmoid volvu Ius, where the 'Iiver overlap sign' is the on ly sign present, and perforation of a hollow viscus where the only sign on a supine abdominal radiograph may be the 'football sign' simulating a 'liver overlap sign'. In the case of sigmoid volvulus described in' this report, the correct diagnosis was MBChB, FFRAD(D)(SA), Consultant Radiologist, Wanganul Base Hospital, Wanganui, New Zealand 24 SA JOURNAL OF RADIOLOGY. June 1997 established only with much difficulty and after fu rther conventional rad iographs and contrast studies. Case reports An 89 year old man presented with a three day history of marked abdomi- nal distention and pain. Clinical diag- noses of subacute intestinal obstruc- tion or perforation were considered. A supine abdominal radiograph re- vealed a curvilinear demarcation, con- vex to lateral, overlying the liver with the area lateral the demarcating line uniformly dense and the area medial to this of relatively low attenuation (Figure 1a). This was thought to rep- resent the superolateral aspect of the 'football sign' overlying the liver, but upright chest radiographs failed to re- veal any subdiaphragmatic free air. Figure 1a: Supine abdominal radiograph with 'liver overlap sign' (arrows) of sigmoid volvulus masquer- ading as 'football sign' of intraperitoneal free air. Confirmed as sigmoid volvulus (cf. Figure 1cl. to page 25 Differentiation bct\Ncen signloid volvulus and free air on supine abdorninal radiographs: the 'I iver overlap sign' versus the 'footba" sign' of faecal material following the gastrografin enema. The sigmoid vol- vulus was repaired surgically on an elective basis with good results. A supine abdominal radiograph of a 63 year old man revealed intraperi- toneal free air due to a perforated duodenal ulcer, also demon- strated on chest radiographs, and is shown as comparison to the previous case (Figure 2). The similarity of free air with a 'football sign' overlying the liver to the previous case is re- markable. Discussion Approximately one-third of 40 confirmed cases of sig- moid volvulus were considered to be difficult to diagnose on conven- tional radiographs' and the above case belongs to this category. The 'liver overlap sign' was found to be of con- siderable diagnostic aid in those cases where the greater part of the sigmoid loop was obscured by proximal co- lonic distension. The sign was positive in 27 out of 40 patients with the right side of the distended sigmoid loop over- lying the liver in curvilinear fashion. In the above case of sigmoid vol-I.~~~;;;;:===;::=======:::;Ivulus the 'liver overlap sign' was the only obvious sign of sigmoid volvulus and even in retrospect the other de- scribed signs'? were difficult to detect. The absence of sub- diaphragmatic free air on upright chest radiographs should have militated strongly against the 'football sign' on the supine abdomi- nal film, but the findings on decubitus radiographs with air on both sides of bowel from page 24 A left side down decubitus x-ray did not show any free air, but a right side down decubitus film demon- strated apparent air outlining bowel wall (Figure 1b) which retrospectively was due to a large distended loop of overlying large bowel simulating the Figure lb: Right side down decubitus radiograph of abdomen with free air on both sides of small bowel wall (arrows) being simulated by overlying distended loop of large bowel. appearance of air on both sides of the walls of small bowel loops. The 'trian- gle sign' was also present with multi- ple triangles of'free air' between bowel loops masquerading in a similar way. A gastrografin meal did not reveal any evidence of perforation. A subse- quent gastrografin enema, however, demonstrated the typical appearance of a partially obstructed sigmoid vol- vulus (Figure 1c). The patient felt most relieved after a massive evacuation Figure 1c: Gastrografln enema demonstrating typical signs of partially obstructed sigmoid volvulus. 2G SA JOURNAL OF RADIOLOGY. June 1997 wall and multiple triangles of air be- tween bowel loops being simulated, added to the confusion. Gastrografin studies excluded a source of perfora- tion. Radiological signs of free air in the abdomen on supine radiographs have been well documented in the litera- ture, but it is of interest that the 'foot- ball sign' (a large loculus of air in the shape of a rugby or soccer ball, situ- ated anteriorly in the abdomen with Figure 2: Part of 'football sign'seen overlying liver (arrows) on supine abdominal radiograph, due to intraperitoneal free air which was shown to be subdiaphragmatic in position on chest x-rays. No distended large bowel loops to suggest sigmoid volvulus. The similarity of the 'liver overlap sign' of sigmoid volvulus (Figure la) and the 'football sign' of intraperitoneal free air (Figure 2) on supine abdominal radiographs is remarkable. the string of the football represented by air outlining the falciform liga- ment), is reported as an uncommon feature in adults'. Other signs de- scribed include the 'inverted V sign' (with air outlining the lateral umbili- cal ligaments overlying the sacrum) and a sign with air outlining a full uri- nary bladder. to page 26 Differentiation betvvcen sigrnoid volvulus and free air on supine abdonlinal radiographs: the 'I iver overlap sign' versus the 'football sign' from page 25 In conclusion it must be stated that having seen many cases of free air with a 'football sign' and a few cases of sig- moid volvulus with a 'liver overlap sign', the former diagnosis was fa- voured. This was also a provisional clini- cal diagnosis.The fact that free air was conclusively excluded on chest radio- graphs and a left side down decubitus abdominal radiograph, should have made gastrografin studies of the upper and lower gastrointestinal tract to search for a perforation unnecessary. The gastrografin enema was, however, of considerable benefit in the diag- nosis and treatment of the sigmoid volvulus. If only a supine abdominal radio- graph had been available,which might sometimes be the case, it would have been very difficult to distinguish be- tween a ruptured hollow viscus and sigmoid volvulus. Other signsof pneu- moperitoneum or sigmoid volvulus also proved to be unhelpful. (In the first case of possible free air they added to the confusion). The clinical features did not aid in differentiating between CASE REPORT Unusual foreign bodies in the oesophagus WFC van Gelderen FFRad(D), Consultant Radiologist KSCheng FRCS(Ed),FRACS, Consultant Surgeon, Department of Radiology, Wanganul Base Hospital, New Zealand. A 70 year old man with a known long segment benign stricture of the lower oesophagus, presented again with dysphagia and the stricture was dilated. Subsequent chest radiographs demonstrated signs of a pneumome- diastinum. An Ultravist 300 non-ionic con- trast (Schering AG, Germany) swal- low demonstrated a localized, con- tained perforation of the left lower these two conditions and without fur- ther conventional radiographs and contrast studies the diagnosis could not have been made. This case illus- trates that neither the 'football sign' nor the 'liver overlap sign' are invari- able, infallible indicators. References I. Young WS, Engelbrecht HE, Stoker A. Plain film analysis in sigmoid volvulus. Cli" Radio11978; 29: 553- 560. 2. Janzen DL, Heap Sw. Organo-axial volvulus of the sigmoid colon. Australas Radiol; 1992: 332-333. 3. Rice RP,Thompson WM, Gedgaudas RK.The diagnosis and significance of extraluminal gas in the abdomen. Radiol Cli" North Am 1982: 20(4): 819-837. oesophagus. Half a dozen rec- tangular low attenuation foreign bodies were noted within the area of rupture and within the distal oesophagus (Figure 1). The patient had not been permitted to have anything by mouth and denied having had anything to eat or drink. On more persistent questioning, however, he admitted to being rather partial to lozenges and had not thought that these would be contraindicated. He had surreptitiously sucked the lozenges one by one in the ward, and subsequently the nucleus of each lozenge, a thin hard square wafer, had found its way down to the oesophagus fully intact 26 SA JOURNAL OF RADIOLOGY. June 1997 in shape, though reduced in size. These wafers were eloquently dem- onstrated on Ultravist swallow. The remains of the lozenges were removed at subsequent oesophagos- copy and the patient made an unre- markable recovery with a follow-up contrast swallow demonstrating reso- lution of the contained rupture. Figure 1: Localized contained rupture left lower oesophagus shown at contrast swallow. Note multiple rectangular low attenuation foreign bodies (arrows) which are the residue of lozenges sucked by the patient over the preceding hours.