Br. J. Surg. 1991. Vol. 78. December, 1448-1 450
S. C. S. Chung, J. W. C. Leung*, H. T. Leong and A. K. C. Li Departments of Surgery and *Medicine, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong Correspondence to: Professor A. K. C. Li
Mechanical lithotripsy of large common bile duct stones using a basket Experience with the Olympus basket mechanical lithotriptor ( B M L - 1Q ) in crushing large common bile duct stones before their endoscopic removal is reported. From January 1988 to January 1990, 68 patients with common duct stones too large to be extracted by Dormia baskets or balloon catheters after sphincterotomy were treated with the B M L system. The largest stones in each patient ranged,from i.0 to 4.9 cm in diameter. Fifty-seven patients required one session of lithotripsy, ten patients two sessions and one patient three sessions; 26 patients required further endoscopic extraction of stone fragments after successful lithotripsy. The stones were successfully crushed by the B M L system and the ducts cleared in 55 patients (81 per cent). In 13 patients mechanical lithotripsy failed because the stones could not be engaged in the lithotriptor basket. i n one patient the stone was crushed with the Soehendra lithotriptor, six patients were successjully managed by electrohydraulic lithotripsy through a ‘mother and baby’ endoscope, indwelling stents were inserted in ,four patients and two patients under went surgery.
Endoscopic papillotomy and stone extraction has become the treatment of choice for patients with common duct stones, especially for the elderly high-risk patient and for those who have had previous biliary surgery. The technical difficulty of stone extraction from the common bile duct increases with the size of the stone’. Because of the risks of perforation and haemorrhage, there is a limit t o the size of the sphincterotomy that can safely be made. In general, stones > 2 cm in diameter need to be fragmented before they can be removed. Large common duct stones are often soft brown pigment stones and can sometimes be crushed by closing the Dormia basket forcibly around them. However, in our experience the duct clearance rate by this method in patients with stones of size > 2 cm is only 50 per cent’. This report records our experience with a mechanical lithotripsy basket in the treatment of stones that are too big for conventional endoscopic extraction with Dormia baskets or balloon catheters. The lithotripsy basket can be inserted into the bile duct through the instrument channel of a therapeutic endoscope.
Patients and methods From January 1988 to January 1990,438 patients with common duct stones were referred for stone extraction to the Combined Endoscopy Unit at the Prince of Wales Hospital. In 68 patients difficulties were encountered using conventional endoscopic extraction techniques, because of either the large size of the stone or a comparatively narrow distal common bile duct. Mechanical lithotripsy using the lithotripsy basket was attempted in these patients. There were 25 men and 43 women with a mean age of 67 (range 21-96) years. The size of the stones and the distal common bile duct were measured on the cholangiogram and corrected for magnification. The largest stones in each patient ranged from 1.0 to 4.9 cm (mean 2.4 cm). Forty-nine patients (72 per cent) had stones > 2 cm in size. The diameter of the common duct 1 cm proximal to the ampulla was 0-5-16 cm (mean 1.1 cm). The basket mechanical lithotriptor (Olympus BML-1Q; Olympus Optical Company, Tokyo, Japan) consists of a strong four-wire basket, a flexible metal sheath and a handle (Figure I ) . The basket together with the metal sheath can be passed through the instrument channel of a therapeutic side-viewing duodenoscope (channel diameter 3.7 or 4.2 mm). An endoscopic sphincterotomy, performed previously or at the same session, is necessary before the lithotriptor basket can be inserted into the bile duct. In the patients studied, 1 1 had had
1448
sphincterotomy performed at a previous admission and had a recurrence of stones, 15 had undergone sphincterotomy at a previous endoscopic retrograde cholangiopancreatography (ERCP) session during the same admission and 36 had had the sphincterotomy performed during the same ERCP session just before lithotripsy. In five patients the lithotriptor basket was inserted through a surgical sphincteroplasty and in one patient through a choledochoduodenostomy. After the stone is engaged in the basket, the handle is tightened, drawing the basket into the flexible sheath. thereby crushing the stone (Figure 2).
Results Fifty-seven patients underwent one session of lithotripsy, ten underwent two sessions and one patient three sessions. In 5 5 patients (81 per cent) the stones could be engaged in the lithotriptor and crushed. After successful lithotripsy, 26 of these patients required further endoscopic extraction of stone fragments at another session. In 13 patients mechanical lithotripsy failed because the stones could not be engaged in the lithotriptor basket; in six of these cases the stones were fragmented using electrohydraulic lithotripsy with a ‘mother and baby’ endoscope system. In one patient the stone was crushed using a Soehendra lithotriptor ( Wilson-Cook Medical Incorporated, Winston-Salem, North Carolina, USA). Two
Figure 1 T h e Olympus busker mechunicul lirhotriptor
0007-1323/91/12 1448-03
C
1991 Butterworth Heinemann Ltd
Lithotripsy of large common bile duct stones: S. C. S. Chung et al.
Figure 2 a A giant common bile duct stone enguyed in the lithotripsy busket. b The .stone is crushed by riyhtening the busker
patients underwent successful surgery; four were deemed too old and frail for surgery and were managed by indwelling stents. N o mechanical failures of the lithotriptor or basket wire breakages were encountered. In one patient the lithotriptor was noted on fluoroscopy to be outside the confines of the bile duct. No air was seen in the retroperitoneum on plain abdominal radiography and the patient was managed by withholding oral intake, and institution of nasogastric and nasobiliary drainage and intravenous antibiotics. Apart from right upper quadrant tenderness for 48 h he made an uneventful recovery. Another patient, who presented with obstructive jaundice, was found to have a 1.2-cm diameter stone above a stricture in the mid common bile duct. The distal common duct measured 0.5 cm. Endoscopic sphincterotomy was performed and the stone crushed and removed in fragments. Unfortunately the patient developed necrotizing pancreatitis and died from sepsis 3 months after the procedure.
Discussion After endoscopic papillotomy, common duct stones < 1 cm in size can easily be extracted using balloon catheters or Dormia baskets3. Larger stones must be fragmented before they can be removed. A number of techniques are available for breaking up such stones. Extracorporeal shockwave lithotripsy (ESWL), which is effective for fragmenting renal and gallbladder stones, can also be applied to calculi in the common bile duct4a5. Targeting of the shockwaves may be more difficult, and the technique requires expensive equipment which may not be universally available. Contact lithotripsy using electrohydraulic lithotripsy is less expensive but carries the risk of damaging the duct wall if the probe is fired against it6. To avoid this complication, direct visual control using the 'mother and baby' peroral choledochoscope is necessary'. The method is technically demanding and should be reserved for patients in whom other methods have failed. Laser lithotripsy using the tunable dye laser also requires direct visual control or prior immobilizaton of the stone in a basket*. Large common duct stones are commonly of the brown pigment variety. They are often soft and muddy and can usually be sliced through with a Dormia basket2. If this manoeuvre fails and stone impaction results, the trapped stone can be crushed using the Soehendra lithotripsy basket by cutting off the handle of the basket and passing a metal sheath over the
Br. J. Surg., Vol. 78, No. 12, December 1991
wires. This method destroys the basket, and the endoscope must be removed for lithotripsy and reinserted to remove the fragments. These problems can be avoided by using a mechanical lithotriptor in which the metal sheath can pass through the biopsy channel of the duodenoscope as well as the basket. Prototypes using this design have been reported to be successful in clinical The basket lithotriptor used in the present series (Olympus BML-1Q) has a stiff metal sheath that can not be moulded into a curve. When the basket is introduced through the endoscope it points in the direction of the pancreatic duct rather than at the bile duct. Because of the stiffness of the metal sheath it is difficult to lift sufficiently with the elevator to point it in the right direction; there is also a risk of perforation. The newer lithotriptor baskets (Olympus BML-3Q, B M L 4 Q ) have an additional Teflon8 (E. I . du Pont de Nemours and Company, Wilmington, Delaware, USA) sheath over the basket. The basket-Teflon sheath assembly is introduced into the bile duct and the stone engaged before the strong spiral metal sheath is advanced over the sheath and basket. This design makes entry into the bile duct easier and may reduce some of the potential hazards such as bile duct perforation". Smaller crushing sheaths, which can be used with standard duodenoscopes, are also available. All the stones that were trapped in the mechanical lithotriptor were successfully crushed. N o cases were encountered in which the basket broke before the stone. In Hong Kong, recurrent pyogenic cholangitis is prevalent and most common duct stones are of the muddy brown pigment variety, which are softer than cholesterol stones. In patients with multiple or giant stones, several sessions of lithotripsy and stone extraction may be necessary. Adequate biliary drainage by nasobiliary tube'* or an indwelling stent before sessions is mandatory in order to prevent cholangitis. We have found the basket mechanical lithotriptor to be very useful in dealing with large common duct stones. Before the use of the basket mechanical lithotriptor in this unit, the success rate for removal of common duct stones 3 2 cm in size was only 50 per cent. Compared with other methods of lithotripsy, the basket mechanical lithotriptor is easy to use, safe and economical. It can be used immediately after identification of large stones without the need to prepare sophisticated equipment. Mechanical lithotripsy should be attempted if the stone is too large to be extracted by conventional means. It can not be used, however, if stone impaction or lack of space to open the basket prevent the stone from being trapped within the basket. Alternative therapies for patients in whom the stones can not be engaged include electrohydraulic lithotripsy using the 'mother and baby' endoscopes, ESWL or surgery. Insertion of an indwelling stent can prevent stone impaction and cholangitis for extended periodst3. Long-term stenting is a last resort for unretrievable stones in patients unfit for surgery.
References 1. 2. 3. 4.
5.
6. 7.
Katon RM. The giant common duct stones: still a hard nut to crack. Gustrointest Endosc 1988; 34: 281-2. Leung JWC, Chung SCS, Mok SD, Li AKC. Endoscopic removal of large common bile duct stones in recurrent pyogenic cholangitis. Gusrrointest Endosc 1988; 34: 23841. Cotton PB. Non-operative removal of bile duct stones by duodenoscopic sphincterotomy. Br J Surg 1980; 67; 1-5. Staritz M, Floth A, Rambow A, Buess G, Wilpert D, Schild F. Extracorporal shock waves (device of the second generation) for therapy of large common bile duct stones: success and problems. Gastroenterology 1987; 92: 1652. Heberer G, Paumgartner G , Sauerbruch T et al. A retrospective analysis of 3 years' experience of an interdisciplinary approach to gallstone disease including shock-waves. Ann Surg 1988; 208: 274-8. Sievert CE, Silvis SE. Evaluation of electrohydraulic lithotripsy on human gall stones. Am J Gustroenterol 1985; 80: 854. Leung JWC, Chung SCS. Electrohydraulic lithotripsy with peroral choledochoscopy. Er Med J 1989; 299: 595-8.
1449
Lithotripsy of large common bile duct stones: S. C. S. Chung et al.
8. 9.
Kozarek RA, Low DE, Ball TJ. Tunable dye laser lithotripsy: in uitro studies and in viuo treatment of choledocholithiasis. Gastroinfesr Endosc 1988; 34: 418-20. Staritz M, Ewe K, Meyer zum Buschenfelde KH. Mechanical gallstone lithotripsy in the common bile duct in-vitro and in-uiuo experience. Endoscopy 1983; 15: 31C18. Schneider M U , Matek W, Bauer R, Domschke W. Mechanical lithotripsy of bile duct stones in 209 patients -effect of technical advances. Endoscopy 1988; 20: 248-53. Higuchi T, Kon Y . Endoscopic mechanical lithotripsy for the treatment of common bile duct stone. Experience with the
12.
-
10. 11.
13.
improved double sheath basket catheter Endoscopy 1987; 19: 216-17. Cotton PB, Burney PGJ, Mason RR. Transnasal bile duct catheterisation after endoscopic sphincterotomy. Method for biliary drainage, perfusion and sequential cholangiography. Gut 1979; 20: 285-7. Cairns SR, Dias L, Cotton PB, Salmon PR, Russell RCG. Additional endoscopic procedures instead of urgent surgery for retained common bile duct stones. Gui 1989; 30: 53540.
Paper accepted 5 August 1991
Case report Br. J. Surg. 1991, Vol. 78. December, 1450-1451
Bleeding duodenal varices N.D. Heaton, H. Khawaja and E. R. Howard Department of Surgery, King's College Hospital, Denmark Hill, London SE5 8RX. UK Correspondence to: Mr N. D. Heaton
Oesophageal varices are a common cause of gastrointestinal bleeding in patients with portal hypertension. Ectopic varices are less common but may occur anywhere along the gastrointestinal tract. Tiyo patients who presented with gastrointestinal bleeding from duodenal varices are described. The aetiology and management of ectopic varices in this site are discussed.
Case reports Patient I In 1986 a 15-year-old boy presented with melaena and a haemoglobin level o f 3 6 g/dl. At the age of 18 months he had presented with vomiting and an abdominal mass after a fall. At laparotomy, there was a large mass in the head of the pancreas and a dilated second part of the duodenum with enlarged lymph nodes around the superior mesenteric artery. An antecolic gastrojejunostomy was performed. Biopsy of the pancreas revealed granulation tissue. He subsequently developed duodenal obstruction and, at the age of 3 years, the distal duodenum was resected with restoration of continuity by anastomosis to the jejunum behind the superior mesenteric vessels. He remained well until the age of 12 years when he presented with melaena and a haemoglobin level of 4.2 g/dl. Investigation failed to show the site of the bleeding and he remained well for another 3 years until further haemorrhage required a blood transfusion of 12 units. Liver function tests were normal apart from the alkaline phosphatase of 139 units/) (normal range 30-85). The prothrombin ratio was 1.4, HBsAg and autoantibodies were negative. Upper gastrointestinal endoscopy revealed a leash of varices at the site of the previous duodenal resection. Ultrasonography of the liver and biliary tree was normal, but the venous phase of visceral angiography showed replacement of the superior mesenteric vein by multiple collaterals which drained into a patent portal vein. The spleen was enlarged, measuring 15 cm in length. A splenic venogram showed a patent proximal splenic vein and occlusion 3 cm to the left of the midline. The patient had a further bleed and required a further 6 units of blood. At subsequent laparotomy there was an area of scarring around the central pancreas with evidence of segmental portal hypertension affecting the area normally drained by the superior mesenteric vein. There were large anastomotic varices. The gastrojejunostomy was taken down and the duodenojejunal anastomosis was resected and resutured. The patient has remained well for 4 years following surgery.
Figure 1 Patient 2: venous phase of' aortoportogram. Exlensive collaterals are seen around the portal vein ( 7 ) . Varices are seen ascending towards the gastro-oesophageal junction ( A). The superior mesenteric fiein is paten!, hut there is no filling of the splenic vein
1450
%07-1323/91,12145&02
Patient 2 The second patient had a duodenojejunostomy performed on the second day of life for duodenal atresia and presented at 2 years of age with bleeding from oesophageal varices secondary to extrahepatic portal vein thrombosis. The oesophageal varices were controlled by injection sclerotherapy for 7 years, but he then began to bleed from another site in the gastrointestinal tract. A splenectomy and lienorenal shunt was performed and bleeding stopped for 9 months before further haemorrhage occurred. Repeated blood transfusions were necessary. Endoscopy revealed bleeding varices at the duodenojejunal anastomosis. At laparotomy, large collaterals were found in very vascular adhesions to the duodenal anastomosis. The anastomosis was resected and bleeding stopped. The patient remained well for a further 2 years, when he presented with bleeding from varices in the gastric fundus. Aortoportography confirmed patency of the superior mesenteric vein but showed thrombosis of the splenic and portal veins (Figure I ). He was treated successfully by a mesocaval shunt using an internal jugular vein conduit and has had no further bleeding for 3 years.
Discussion Ectopic varices are defined as those arising in sites other than the oesophagus and cardia of the stomach. They may occur in isolation or in association with varices at other sites. They
~~
f 1991 Butterworth-Heinemann Ltd
S. C. S. Chung, J. W. C. Leung*, H. T. Leong and A. K. C. Li Departments of Surgery and *Medicine, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong Correspondence to: Professor A. K. C. Li
Mechanical lithotripsy of large common bile duct stones using a basket Experience with the Olympus basket mechanical lithotriptor ( B M L - 1Q ) in crushing large common bile duct stones before their endoscopic removal is reported. From January 1988 to January 1990, 68 patients with common duct stones too large to be extracted by Dormia baskets or balloon catheters after sphincterotomy were treated with the B M L system. The largest stones in each patient ranged,from i.0 to 4.9 cm in diameter. Fifty-seven patients required one session of lithotripsy, ten patients two sessions and one patient three sessions; 26 patients required further endoscopic extraction of stone fragments after successful lithotripsy. The stones were successfully crushed by the B M L system and the ducts cleared in 55 patients (81 per cent). In 13 patients mechanical lithotripsy failed because the stones could not be engaged in the lithotriptor basket. i n one patient the stone was crushed with the Soehendra lithotriptor, six patients were successjully managed by electrohydraulic lithotripsy through a ‘mother and baby’ endoscope, indwelling stents were inserted in ,four patients and two patients under went surgery.
Endoscopic papillotomy and stone extraction has become the treatment of choice for patients with common duct stones, especially for the elderly high-risk patient and for those who have had previous biliary surgery. The technical difficulty of stone extraction from the common bile duct increases with the size of the stone’. Because of the risks of perforation and haemorrhage, there is a limit t o the size of the sphincterotomy that can safely be made. In general, stones > 2 cm in diameter need to be fragmented before they can be removed. Large common duct stones are often soft brown pigment stones and can sometimes be crushed by closing the Dormia basket forcibly around them. However, in our experience the duct clearance rate by this method in patients with stones of size > 2 cm is only 50 per cent’. This report records our experience with a mechanical lithotripsy basket in the treatment of stones that are too big for conventional endoscopic extraction with Dormia baskets or balloon catheters. The lithotripsy basket can be inserted into the bile duct through the instrument channel of a therapeutic endoscope.
Patients and methods From January 1988 to January 1990,438 patients with common duct stones were referred for stone extraction to the Combined Endoscopy Unit at the Prince of Wales Hospital. In 68 patients difficulties were encountered using conventional endoscopic extraction techniques, because of either the large size of the stone or a comparatively narrow distal common bile duct. Mechanical lithotripsy using the lithotripsy basket was attempted in these patients. There were 25 men and 43 women with a mean age of 67 (range 21-96) years. The size of the stones and the distal common bile duct were measured on the cholangiogram and corrected for magnification. The largest stones in each patient ranged from 1.0 to 4.9 cm (mean 2.4 cm). Forty-nine patients (72 per cent) had stones > 2 cm in size. The diameter of the common duct 1 cm proximal to the ampulla was 0-5-16 cm (mean 1.1 cm). The basket mechanical lithotriptor (Olympus BML-1Q; Olympus Optical Company, Tokyo, Japan) consists of a strong four-wire basket, a flexible metal sheath and a handle (Figure I ) . The basket together with the metal sheath can be passed through the instrument channel of a therapeutic side-viewing duodenoscope (channel diameter 3.7 or 4.2 mm). An endoscopic sphincterotomy, performed previously or at the same session, is necessary before the lithotriptor basket can be inserted into the bile duct. In the patients studied, 1 1 had had
1448
sphincterotomy performed at a previous admission and had a recurrence of stones, 15 had undergone sphincterotomy at a previous endoscopic retrograde cholangiopancreatography (ERCP) session during the same admission and 36 had had the sphincterotomy performed during the same ERCP session just before lithotripsy. In five patients the lithotriptor basket was inserted through a surgical sphincteroplasty and in one patient through a choledochoduodenostomy. After the stone is engaged in the basket, the handle is tightened, drawing the basket into the flexible sheath. thereby crushing the stone (Figure 2).
Results Fifty-seven patients underwent one session of lithotripsy, ten underwent two sessions and one patient three sessions. In 5 5 patients (81 per cent) the stones could be engaged in the lithotriptor and crushed. After successful lithotripsy, 26 of these patients required further endoscopic extraction of stone fragments at another session. In 13 patients mechanical lithotripsy failed because the stones could not be engaged in the lithotriptor basket; in six of these cases the stones were fragmented using electrohydraulic lithotripsy with a ‘mother and baby’ endoscope system. In one patient the stone was crushed using a Soehendra lithotriptor ( Wilson-Cook Medical Incorporated, Winston-Salem, North Carolina, USA). Two
Figure 1 T h e Olympus busker mechunicul lirhotriptor
0007-1323/91/12 1448-03
C
1991 Butterworth Heinemann Ltd
Lithotripsy of large common bile duct stones: S. C. S. Chung et al.
Figure 2 a A giant common bile duct stone enguyed in the lithotripsy busket. b The .stone is crushed by riyhtening the busker
patients underwent successful surgery; four were deemed too old and frail for surgery and were managed by indwelling stents. N o mechanical failures of the lithotriptor or basket wire breakages were encountered. In one patient the lithotriptor was noted on fluoroscopy to be outside the confines of the bile duct. No air was seen in the retroperitoneum on plain abdominal radiography and the patient was managed by withholding oral intake, and institution of nasogastric and nasobiliary drainage and intravenous antibiotics. Apart from right upper quadrant tenderness for 48 h he made an uneventful recovery. Another patient, who presented with obstructive jaundice, was found to have a 1.2-cm diameter stone above a stricture in the mid common bile duct. The distal common duct measured 0.5 cm. Endoscopic sphincterotomy was performed and the stone crushed and removed in fragments. Unfortunately the patient developed necrotizing pancreatitis and died from sepsis 3 months after the procedure.
Discussion After endoscopic papillotomy, common duct stones < 1 cm in size can easily be extracted using balloon catheters or Dormia baskets3. Larger stones must be fragmented before they can be removed. A number of techniques are available for breaking up such stones. Extracorporeal shockwave lithotripsy (ESWL), which is effective for fragmenting renal and gallbladder stones, can also be applied to calculi in the common bile duct4a5. Targeting of the shockwaves may be more difficult, and the technique requires expensive equipment which may not be universally available. Contact lithotripsy using electrohydraulic lithotripsy is less expensive but carries the risk of damaging the duct wall if the probe is fired against it6. To avoid this complication, direct visual control using the 'mother and baby' peroral choledochoscope is necessary'. The method is technically demanding and should be reserved for patients in whom other methods have failed. Laser lithotripsy using the tunable dye laser also requires direct visual control or prior immobilizaton of the stone in a basket*. Large common duct stones are commonly of the brown pigment variety. They are often soft and muddy and can usually be sliced through with a Dormia basket2. If this manoeuvre fails and stone impaction results, the trapped stone can be crushed using the Soehendra lithotripsy basket by cutting off the handle of the basket and passing a metal sheath over the
Br. J. Surg., Vol. 78, No. 12, December 1991
wires. This method destroys the basket, and the endoscope must be removed for lithotripsy and reinserted to remove the fragments. These problems can be avoided by using a mechanical lithotriptor in which the metal sheath can pass through the biopsy channel of the duodenoscope as well as the basket. Prototypes using this design have been reported to be successful in clinical The basket lithotriptor used in the present series (Olympus BML-1Q) has a stiff metal sheath that can not be moulded into a curve. When the basket is introduced through the endoscope it points in the direction of the pancreatic duct rather than at the bile duct. Because of the stiffness of the metal sheath it is difficult to lift sufficiently with the elevator to point it in the right direction; there is also a risk of perforation. The newer lithotriptor baskets (Olympus BML-3Q, B M L 4 Q ) have an additional Teflon8 (E. I . du Pont de Nemours and Company, Wilmington, Delaware, USA) sheath over the basket. The basket-Teflon sheath assembly is introduced into the bile duct and the stone engaged before the strong spiral metal sheath is advanced over the sheath and basket. This design makes entry into the bile duct easier and may reduce some of the potential hazards such as bile duct perforation". Smaller crushing sheaths, which can be used with standard duodenoscopes, are also available. All the stones that were trapped in the mechanical lithotriptor were successfully crushed. N o cases were encountered in which the basket broke before the stone. In Hong Kong, recurrent pyogenic cholangitis is prevalent and most common duct stones are of the muddy brown pigment variety, which are softer than cholesterol stones. In patients with multiple or giant stones, several sessions of lithotripsy and stone extraction may be necessary. Adequate biliary drainage by nasobiliary tube'* or an indwelling stent before sessions is mandatory in order to prevent cholangitis. We have found the basket mechanical lithotriptor to be very useful in dealing with large common duct stones. Before the use of the basket mechanical lithotriptor in this unit, the success rate for removal of common duct stones 3 2 cm in size was only 50 per cent. Compared with other methods of lithotripsy, the basket mechanical lithotriptor is easy to use, safe and economical. It can be used immediately after identification of large stones without the need to prepare sophisticated equipment. Mechanical lithotripsy should be attempted if the stone is too large to be extracted by conventional means. It can not be used, however, if stone impaction or lack of space to open the basket prevent the stone from being trapped within the basket. Alternative therapies for patients in whom the stones can not be engaged include electrohydraulic lithotripsy using the 'mother and baby' endoscopes, ESWL or surgery. Insertion of an indwelling stent can prevent stone impaction and cholangitis for extended periodst3. Long-term stenting is a last resort for unretrievable stones in patients unfit for surgery.
References 1. 2. 3. 4.
5.
6. 7.
Katon RM. The giant common duct stones: still a hard nut to crack. Gustrointest Endosc 1988; 34: 281-2. Leung JWC, Chung SCS, Mok SD, Li AKC. Endoscopic removal of large common bile duct stones in recurrent pyogenic cholangitis. Gusrrointest Endosc 1988; 34: 23841. Cotton PB. Non-operative removal of bile duct stones by duodenoscopic sphincterotomy. Br J Surg 1980; 67; 1-5. Staritz M, Floth A, Rambow A, Buess G, Wilpert D, Schild F. Extracorporal shock waves (device of the second generation) for therapy of large common bile duct stones: success and problems. Gastroenterology 1987; 92: 1652. Heberer G, Paumgartner G , Sauerbruch T et al. A retrospective analysis of 3 years' experience of an interdisciplinary approach to gallstone disease including shock-waves. Ann Surg 1988; 208: 274-8. Sievert CE, Silvis SE. Evaluation of electrohydraulic lithotripsy on human gall stones. Am J Gustroenterol 1985; 80: 854. Leung JWC, Chung SCS. Electrohydraulic lithotripsy with peroral choledochoscopy. Er Med J 1989; 299: 595-8.
1449
Lithotripsy of large common bile duct stones: S. C. S. Chung et al.
8. 9.
Kozarek RA, Low DE, Ball TJ. Tunable dye laser lithotripsy: in uitro studies and in viuo treatment of choledocholithiasis. Gastroinfesr Endosc 1988; 34: 418-20. Staritz M, Ewe K, Meyer zum Buschenfelde KH. Mechanical gallstone lithotripsy in the common bile duct in-vitro and in-uiuo experience. Endoscopy 1983; 15: 31C18. Schneider M U , Matek W, Bauer R, Domschke W. Mechanical lithotripsy of bile duct stones in 209 patients -effect of technical advances. Endoscopy 1988; 20: 248-53. Higuchi T, Kon Y . Endoscopic mechanical lithotripsy for the treatment of common bile duct stone. Experience with the
12.
-
10. 11.
13.
improved double sheath basket catheter Endoscopy 1987; 19: 216-17. Cotton PB, Burney PGJ, Mason RR. Transnasal bile duct catheterisation after endoscopic sphincterotomy. Method for biliary drainage, perfusion and sequential cholangiography. Gut 1979; 20: 285-7. Cairns SR, Dias L, Cotton PB, Salmon PR, Russell RCG. Additional endoscopic procedures instead of urgent surgery for retained common bile duct stones. Gui 1989; 30: 53540.
Paper accepted 5 August 1991
Case report Br. J. Surg. 1991, Vol. 78. December, 1450-1451
Bleeding duodenal varices N.D. Heaton, H. Khawaja and E. R. Howard Department of Surgery, King's College Hospital, Denmark Hill, London SE5 8RX. UK Correspondence to: Mr N. D. Heaton
Oesophageal varices are a common cause of gastrointestinal bleeding in patients with portal hypertension. Ectopic varices are less common but may occur anywhere along the gastrointestinal tract. Tiyo patients who presented with gastrointestinal bleeding from duodenal varices are described. The aetiology and management of ectopic varices in this site are discussed.
Case reports Patient I In 1986 a 15-year-old boy presented with melaena and a haemoglobin level o f 3 6 g/dl. At the age of 18 months he had presented with vomiting and an abdominal mass after a fall. At laparotomy, there was a large mass in the head of the pancreas and a dilated second part of the duodenum with enlarged lymph nodes around the superior mesenteric artery. An antecolic gastrojejunostomy was performed. Biopsy of the pancreas revealed granulation tissue. He subsequently developed duodenal obstruction and, at the age of 3 years, the distal duodenum was resected with restoration of continuity by anastomosis to the jejunum behind the superior mesenteric vessels. He remained well until the age of 12 years when he presented with melaena and a haemoglobin level of 4.2 g/dl. Investigation failed to show the site of the bleeding and he remained well for another 3 years until further haemorrhage required a blood transfusion of 12 units. Liver function tests were normal apart from the alkaline phosphatase of 139 units/) (normal range 30-85). The prothrombin ratio was 1.4, HBsAg and autoantibodies were negative. Upper gastrointestinal endoscopy revealed a leash of varices at the site of the previous duodenal resection. Ultrasonography of the liver and biliary tree was normal, but the venous phase of visceral angiography showed replacement of the superior mesenteric vein by multiple collaterals which drained into a patent portal vein. The spleen was enlarged, measuring 15 cm in length. A splenic venogram showed a patent proximal splenic vein and occlusion 3 cm to the left of the midline. The patient had a further bleed and required a further 6 units of blood. At subsequent laparotomy there was an area of scarring around the central pancreas with evidence of segmental portal hypertension affecting the area normally drained by the superior mesenteric vein. There were large anastomotic varices. The gastrojejunostomy was taken down and the duodenojejunal anastomosis was resected and resutured. The patient has remained well for 4 years following surgery.
Figure 1 Patient 2: venous phase of' aortoportogram. Exlensive collaterals are seen around the portal vein ( 7 ) . Varices are seen ascending towards the gastro-oesophageal junction ( A). The superior mesenteric fiein is paten!, hut there is no filling of the splenic vein
1450
%07-1323/91,12145&02
Patient 2 The second patient had a duodenojejunostomy performed on the second day of life for duodenal atresia and presented at 2 years of age with bleeding from oesophageal varices secondary to extrahepatic portal vein thrombosis. The oesophageal varices were controlled by injection sclerotherapy for 7 years, but he then began to bleed from another site in the gastrointestinal tract. A splenectomy and lienorenal shunt was performed and bleeding stopped for 9 months before further haemorrhage occurred. Repeated blood transfusions were necessary. Endoscopy revealed bleeding varices at the duodenojejunal anastomosis. At laparotomy, large collaterals were found in very vascular adhesions to the duodenal anastomosis. The anastomosis was resected and bleeding stopped. The patient remained well for a further 2 years, when he presented with bleeding from varices in the gastric fundus. Aortoportography confirmed patency of the superior mesenteric vein but showed thrombosis of the splenic and portal veins (Figure I ). He was treated successfully by a mesocaval shunt using an internal jugular vein conduit and has had no further bleeding for 3 years.
Discussion Ectopic varices are defined as those arising in sites other than the oesophagus and cardia of the stomach. They may occur in isolation or in association with varices at other sites. They
~~
f 1991 Butterworth-Heinemann Ltd
