Volume 7, Issue 11 , Page A18, November 2009
Chronic Diarrhea, Weight Loss, and Refractory Epilepsy Three Years After Percutaneous Endoscopic Gastrostomy
Article Outline
A 46-year-old man presented to our hospital with a 2-month history of chronic diarrhea and weight loss. Moreover, his long-standing epilepsy had become refractory to therapy even with a combination of several antiepileptic drugs.
The patient was known to have severe congenital brain damage with spastic paresis and epilepsy. Because of increasing dysphagia, a percutaneous endoscopic gastrostomy (PEG) tube was placed uneventfully 3 years ago. Physical abdominal examination was normal; the PEG tube had no signs of inflammation in its typical location; it could be moved and flushed normally. Laboratory findings were notable for a low albumin level (30 g/L; normal, 35–52 g/L), low potassium level (2.5 mmol/L; normal, 3.7–4.7 mmol/L), low magnesium level (0.57 mmol/L; normal, 0.70–1.0 mmol/L), and low phosphate level (0.42 mmol/L; normal, 0.80–1.50 mmol/L). Stool examination for parasites and bacteria was negative. During the work-up of his chronic diarrhea we performed a colonoscopy, during which we surprisingly found the baseplate of the PEG tube in the transverse colon (Figure A). We removed the PEG tube and the baseplate with a snare and closed the mucosa with 5 clips (Figure B). Furthermore, antibiotic therapy with tazobactam was initiated. The postinterventional course was uneventful and after 5 days a new PEG was placed carefully without any problems.
Colocutaneous fistula is a rare complication of PEG tube placement.1, 2 In our case the PEG tube was removed by colonoscopy and the hole was sealed with resolution clips. Conservative management can be tried by clipping the hole and initiating antibiotic therapy, especially in patients with a high risk for surgical intervention. The diarrhea in our patients was an extreme example of osmotic diarrhea owing to the nutrient solution delivered directly into the transverse colon. In addition, the antiepileptic medication obviously was not applied at the right place and therefore was not absorbed properly. A possible explanation for the displaced PEG is that during the initial insertion the needle must have passed through the transverse colon before it entered the stomach. Then, by continuous traction, the bumper migrated through the gastric wall into the colonic lumen. Because the patient was asymptomatic for the first 3 years after PEG insertion and presented with diarrhea only for the past few months, we consider this to be the most likely explanation. This case illustrates once more that PEG insertion can have serious complications and that diaphanoscopy is mandatory.
References
Conflicts of interest The authors disclose no conflicts.
PII: S1542-3565(08)01249-4
doi:10.1016/j.cgh.2008.12.020
© 2009 AGA Institute. Published by Elsevier Inc. All rights reserved.
Volume 7, Issue 11 , Page A18, November 2009
