Recurrent Duodenal Ulcer Due toNonsteroidal Anti-InflammatoriesFollowing the Suspension ofAntiulcer MedicationAna Paula Silva

Centro Hospitalar de Vila Nova de Gaia Vila Nova de Gaia, Portugal

Abstract The patient was a 62-year-old woman with a personal history of chronicalcoholism and a medical history of duodenal ulcer, congestive heart failure,atrial fibrillation, dilated cardiomyopathy, and recurring urinary tract in-fections. She had osteoporosis, cirrhosis of the liver and portal hypertension.She had undergone surgery for multiple arm fractures after an accidental fallin the previous year and had received NSAIDs without concomitant gastricprotection. On the fourth day of hospitalization she had an episode of hae-matemesis. The patient continued to take NSAIDs as required, as well ashabitual medication of propranolol 40mg/day, spironolactone 25mg/day,furosemide 40mg/day, pantoprazole 40mg/day and strontium 2 g/day. Upperdigestive endoscopy (UDE) revealed level I/II (Baveno) oesophageal varicoseveins with no signs of haemorrhage and portal hypertensive gastropathy. Theduodenal bulbous was deformed by an extensive ulcer with blood clots andone vessel was visible along the antero-superior portion (Forrest IIa classifi-cation). Numerous superficial ulcers with haemosiderin pigment at D2 wereobserved. Endoscopic haemostasis was achieved with epinephrine and bipo-lar probe. Food was suspended and the patient received continuous in-travenous treatment with pantoprazole 8mg/h. UDE repeated after 48 hoursshowed no signs of haemorrhaging from the ulcer. After 72 hours, the pan-toprazole dose was changed to 40mg every 12 hours and food was allowed.After 12 days with no recurrence of the haemorrhagic incident, the patientwas prescribed oral pantoprazole 40mg once daily and released fromhospital.

Background

The patient was a 62-year-old woman, married,and a retired winery bottler. She had a medicalhistory of duodenal ulcer, congestive heart fail-ure, atrial fibrillation, dilated cardiomyopathy(for which she was being followed by the cardio-

logy service) and recurring urinary tract infec-tions. In 2004 the patient sustained a fracturedhumerus with associated pleural effusion; in 2005she fractured her jaw and in 2006 the neck of thefemur. She had also undergone tonsillectomy andtubal ligation. There was no family history ofpeptic ulcer disease or cirrhosis of the liver.

CASE REPORTClin Drug Investig 2009; 29 Suppl. 2: 23-25

1173-2563/09/0002-0023/$49.95/0

ª 2009 Adis Data Information BV. All rights reserved.

Medical History

The patient had also been followed by liverconsultations since 2005 and was referred by theassisting physician owing to cirrhosis of the liverand portal hypertension (Child A). The patienthad a previous incident of decompensation due toascites, but no history of digestive haemorrhage.She admitted chronic alcoholism (consumptionof 60–80 g/day ethanol between the ages of 20 and50 years and current consumption of 20 g/day).The patient was a nonsmoker.

Habitual medication included propranolol40mg daily, spironolactone 25mg daily, furosemide40mg daily, pantoprazole 40mg daily, strontium2 g daily and NSAIDs as required.

Tests performed during office visits revealedthat the patient was negative for hepatitis virusmarkers, and immunological and metabolic stu-dies were negative. Abdominal ultrasound de-monstrated diffusely heterogeneous liver texturesuggesting chronic liver pathology, no focalnodular lesions, normal gall bladder and bileducts, homogeneously enlarged spleen and nochanges in the pancreas. Bone density measure-ments suggested osteoporosis.

The patient had been hospitalized in 2008 be-cause of a supracondylar fracture of the humerusand fracture of the distal extremity of the bonesof the left forearm following an accidental fall.She underwent surgery to set and pin the frac-tures, during which time she was medicated withNSAIDs without concomitant gastric protection.

On the fourth day of hospitalization she hadan episode of haematemesis.

Additional Tests

Analytical results were as follows: haemoglobin7.6 g/dL, haematocrit 25.1%, mean corpuscularvolume 106.4 fL, mean globular haemoglobin con-centration 30.3 g/dL, leukocytes 6.44· 103/uL, neu-trophils 72%, lymphocytes 20.5%, platelets 102000,urea 42, creatinine 0.7mg/dL, glucose 102mg/dL,sodium 135mEq/L, potassium 4.2mEq/L, AST39U/L, ALT 20U/L, total bilirubin 1.2mg/dL,direct bilirrubin 0.32mg/dL, albumin 3.4 g/dL,total proteins 5.9 g/dL, lactate dehydrogenase

278U/L, prothrombin rate 79%, INR 1.18, acti-vated partial thromboplastin time 38.1 sec.

Upper digestive endoscopy (UDE) revealedlevel I/III (Baveno) oesophageal varicose veins,with no signs of haemorrhage and no lesions inthe stomach. The bulbous was deformed by anextensive ulcer with blood clots and one vesselwas visible along the antero-superior portion(Forrest IIa classification). Numerous superficialulcers were observed with haemosiderin pigmentat D2.

Diagnosis

Digestive haemorrhage due to duodenal ulcer(Forrest IIa) was diagnosed.

Treatment

The visible vessel was treated endoscopicallywith epinephrine and coagulated with a bipolarprobe without any recurrence. Food was sus-pended and the patient was medicated with con-tinuous intravenous proton pump inhibitor (PPI)[pantoprazole] 8mg/h for 72 hours.

UDE was repeated after 48 hours with nosigns of haemorrhaging from the ulcer, and spe-cimens were removed from the stomach to biopsyfor Helicobacter pylori; the results of the biopsywere negative. After 72 hours the dose of panto-prazole was changed to 40mg every 12 hours andthe patient started on an oral diet. There was norecurrence of the haemorrhagic incident and thepatient was released from hospital on the 12thday with an outpatient prescription for oral pan-toprazole 40mg once daily.

Discussion

Upper digestive haemorrhaging in patients withcirrhosis is most often due to complications ofportal hypertension, specifically gastro-oesophagealvaricose veins and hypertensive gastropathy.However, other causes must also be considered,such as peptic ulcer and erosive gastritis, espe-cially in patients taking NSAIDs, as was the casefor this patient. The use of NSAIDs is known tobe associated with lesions to the gastroduodenal

24 Silva

ª 2009 Adis Data Information BV. All rights reserved. Clin Drug Investig 2009; 29 Suppl. 2

mucosa due to either topical or systemic effectsand with increased risk of severe gastrointestinalcomplications such as haemorrhaging in elderlypatients and those with a history of ulcers, bothof which were the case here. In addition, livercirrhosis alone brings an increased risk of hae-morrhaging and the use of NSAIDs should, as arule, be avoided by these patients.

The treatment of NSAID-induced ulcers andtheir complications is identical to the treatmentof peptic ulcers in general. PPIs are the treatmentof choice owing to their increased effectivenessand, whenever possible, NSAIDs should be sus-pended. Endoscopic therapy is indicated forbleeding peptic ulcers (Forrest I) or recent hae-morrhage with the risk of recurrence (Forrest IIaand b). Endoscopic haemostasis techniques in-clude injection and thermal and mechanicalmeans. In this case a combination of two techni-ques (injection and thermal) was used to increasethe effectiveness of haemostasis. In such situa-tions it is also recommended that patients betested for H. pylori and treated if it is present, asthis is an added risk for peptic ulcer.

Preventing the gastroduodenal toxicity ofNSAIDs has a fundamental role in minimizingthe associated morbidity and mortality. This pa-

tient had been receiving long-term therapy withPPIs in the hepatology department since 2005and took NSAIDs for osteoarticular pain only asneeded, with no evidence of toxicity seen in theoutpatient clinic until the day she was hospita-lized.

Conclusions

This case exemplifies the risk of gastro-duodenal toxicity associated with NSAIDs andthe importance of preventing it. In patients withcirrhosis of the liver there is an increased risk ofcomplications such as haemorrhaging, so as arule the use of NSAIDs should be avoided. IfNSAIDs must be used the concomitant use ofgastric protection such as a PPI is recommended.

Acknowledgements

Editorial assistance was provided by Andrea Bothwellof Wolters Kluwer. Funding for this assistance was providedby Tecnimede Portugal. The author reports no conflicts ofinterest and did not receive any funding for this paper.

Correspondence: Ana Paula Silva, Department of Gastro-enterology, Centro Hospitalar de Vila Nova de Gaia,Portugal.

Recurrent Duodenal Ulcer 25

ª 2009 Adis Data Information BV. All rights reserved. Clin Drug Investig 2009; 29 Suppl. 2