Statistics from Altmetric.com
An 80 year old woman was admitted in February 1999 with a history of coffee ground vomiting over a 24 hour period. Her past medical history included renal impairment (1993), pancreatitis secondary to “sludge” in the gall bladder and endoscopic retrograde cholangiopancreatography (December 1994), open cholecystectomy (December 1995), and recently diagnosed bladder tumour (January 1999). On examination, she was pale, although well, and cardiovascularly stable. Pulse was 66 beats/min regular, blood pressure was 120/70 mm Hg, and the pre-endoscopy Rockall score was 2. Haemoglobin level was 9.1 g/dl with normal indices. Gastroscopy was carried out the next day and showed the presence of two shallow duodenal ulcers, with a post-endoscopy Rockall score of 3. On the anterior duodenal wall within an ulcer, two surgical metallic clips were protruding and were removed with a snare (fig 1).
Abdominal x ray after the procedure revealed no abnormality and the patient improved with oral proton pump inhibitors. The clips were of the type used by the operating surgeon who performed the cholecystectomy (1995) and were used to ligate the cystic artery. The second day, haemoglobin had not changed significantly (8.9 g/dl). Renal function had deteriorated markedly from previously documented levels. Ultrasound followed by computed tomography scan of the abdomen revealed left sided hydronephrosis, hydroureter, with a bladder tumour extending posteriorly to the rectum and superiorly to involve several loops of the bowel. The urologist and radiologist considered the appearance inoperable. She developed heavy haematuria, was transferred to the care of a urologist, eventually being discharged one month after admission, and lived a further six months before succumbing to the effects of the tumour.
We could find no other descriptions of surgical clips being found in the base of a duodenal ulcer and offer two possibilities as to how they may have arrived at their end point.
The base of the cystic duct and artery lie immediately adjacent to the first part of the duodenum and through direct contact with the bowel the clips may have slowly migrated through the wall, eventually emerging in the base of a small ulcer.
A pre-existing ulcer may have transiently perforated and healed immediately, capturing the clips in the process.
We think the former explanation is more likely.