Yersinia colitis masquerading as pseudomembranous colitis

Digestive Diseases and Sciences, Vol. 3l, No. 5 (May 1986), pp. 548-551

CASE REPORT

Yersinia Colitis Masquerading as Pseudomembranous Colitis R O N A L D B R O W N , MD, F R A N C I S J. T E D E S C O , MD, R A M Z I T. A S S A D , MD, and R A G H U N A T H A RAO, MD

R e c e n t w o r k has d e m o n s t r a t e d that infectious agents can cause changes of the colonic m u c o s a which can m a s q u e r a d e as ulcerative colitis, C r o h n ' s colitis, and p s e u d o m e m b r a n o u s colitis (1, 2). The following case report describes an example of infectious colitis which was initially misinterpreted endoscopically as p s e u d o m e m b r a n o u s colitis.

CASE REPORT The patient was a 15-month-old black male who had an uneventful birth and medical history until one week prior to admission when the child presented with vomiting and fever. The child was found to have otitis media and was treated with amoxacillin. Mucoid diarrhea began after 24 hr of therapy. The patient was then hospitalized and the amoxicillin was discontinued and ampicillin was begun. He was also treated with intravenous fluids, without improvement. On transfer to the Medical College of Georgia's Hospital and Clinics, his temperature was 40.0 ~ C, pulse 130, respirations 40, and his blood pressure was 80/32 mm Hg. The patient was a well-nourished, well-developed child who appeared ill. The head, eyes, ears, nose, and throat were normal. The lungs and heart were normal. The bowel sounds were normal; tenderness was not elicited. The abdomen was not distended, and no organs or masses were palpated. Rectal examination was remarkable for green, mucoid stool that was positive for occult blood. Gram stain of the stool revealed white blood cells. Urinalysis revealed ketones; the sediment contained 5-10 white cells and occasional red cells per high-powered field. The hematocrit was 31.2%; the white-cell count was 19,100 with 31% neutrophils, 15% band forms, 38% lymphocytes, 11% monocytes, and 4% eosinophils. The erythrocyte sedimentation rate was 35 mm/hr. The urea nitrogen was 2 mg/100 ml, and the glucose 102 mg/100 ml. Serum electroManuscript received June 25, 1984; revised manuscript received December 12, 1984; accepted February 27, 1985.

From the Section of Gastroenterology, Department of Medicine, and Department of Pathology, Medical College of Georgia, Augusta, Georgia 30912. Address for reprint requests: Dr. Francis J. Tedesco, GI Medicine, Medical College of Georgia, Augusta, Georgia 30912.

548

lytes were normal. The serum bilirubin was 0.2 mg/100 ml, serum aminotransferase (SGOT) was 27 units, lactic dehydrogenase (LDH) 187 units, albumin 3.3 g/100 ml, protein 6.1 g/100 ml, and alkaline phosphatase 171 units. An x-ray of the chest was normal. Flat and upright films of the abdomen were normal. Flexible sigmoidoscopic examination to a level of 16 cm disclosed erythematous lesions with a raised central whitish area (Figure 1). The mucosa between these lesions was normal. The colon was involved from the rectum to the extent of examination. Multiple biopsies of the lesions were obtained. The biopsies revealed areas of essentially normal mucosa adjacent to areas of edema and acute inflammation (Figure 2). The acute inflammatory changes were reflected by the presence of edema and an increase in neutrophils without obvious increase in lymphocytes and plasma cells. Superficial crypt abscesses were identified with a background of normal crypt architecture. In view of the patient's history of the onset of diarrhea following antibiotic usage and the abnormal sigmoidoscopic findings, the patient was begun on vancomycin, 125 mg orally every 6 hr for suspected antibiotic-associated pseudomembranous colitis. His temperature fell to 37.4~ C after 48 hr, but his diarrhea continued. The initial reports on stool cultures were negative for Salmonella, Shigella, Clostridium difficile, Campylobacter, and Yersinia species. Stool for ovum and parasite was negative. Stool for rotavirus was negative. Serology for amoeba was also negative. Stool specimens for Clostridium difficile toxin were, however, also negative. Blood and urine cultures were also negative. Because the sigmoidoscopic features were not considered characteristic for pseudomembranous colitis, and the biopsy did not demonstrate pseudomembranes and the stool cultures and stool toxin were negative for Clostridium difficile, the sigmoidoscopic examination was repeated on the tenth day. The examination was endoscopically unchanged since the previous examination. Three days later, Y. enterocolitica was cultured from the initial stool specimens. Vancomycin was discontinued, but no further antibiotic therapy was initiated. The patient's diarrhea gradually decreased, and on the day of discharge, the patient was asymptomatic. He was seen several weeks later as an outpatient and was doing well. Digestive Diseases and Sciences, Vol. 31, No. 5 (May 1986)

0163-2116/86/0500-0548505.00/09 1986PlenumPublishingCorporation

Yersinia COLITIS

Fig 1. Endoscopic picture revealing slightly raised central white lesion on an elevated red base (arrows). Adjacent mucosa is normal or slightly edematous.

DISCUSSION

Our case" demonstrates an unusual presentation of Yersinia colitis in a child less than 5 years of age. The prese, ntation was initially felt to be consistent with antibiotic-associated diarrhea and colitis. The patient's sigmoidoscopic features were initially misinterpreted as being consistent with pseudomembranous colitis, but on reflection the raised central whitish area on focal erythematous lesions were more characteristic of aphthoid ulcers. The typical sigmoidoscopic features of antibiotic-associated pseudomembranous colitis include slightly raised yellowish-white plaques that vary in size from a few millimeters to more than a centimeter. These plaquelike lesions characteristically do not eminate from focal erythematous lesions. The diagnosis of antibiotic-associated pseudomembranous colitis was excluded in our patient by the Digestive Diseases and Sciences, Vol. 31, No, 5 (May 1986)

negative stool examination for Clostrium difficile toxin, the failure to obtain pseudomembranous colitis on rectal biopsy, and the failure to demonstrate sigmoidoscopic improvement with vancomycin therapy. This case report demonstrates that Yersinia infection in infants may endoscopically be misinterpreted as antibiotic-associated pseudomembranous colitis. Yersinia enterocolitica is a Gram-negative rod easily mistaken for E. coli. It is an important intestinal pathogen that causes a spectrum of clinical illnesses from simple gastroenteritis to invasive ileitis and colitis. The clinical presentation in children appears to be age-related (3). Under 5 years of age, the most common illness is mild self-limiting gastroenteritis virtually indistinguishable from gastroenteritis caused by other invasive enteric pathogens. The duration of illness is generally 2-3 weeks, but less frequently, infants may have chronic diarrhea lasting for several months. Rarely, more severe

549

BROWN ET AL

Fig 2. Lamina propria is edematous with scattered neutrophils but without obvious increase in lymphocytes and plasma cells. The crypt architecture is not distorted. Hematoxylin and eosin, • 126.

enteritis results in intestinal ulceration, perforation, and peritonitis. In children over 5 years of age, abdominal pain is common and clinical presentation can mimic that of acute appendicitis. Findings at surgery include mesenteric lymphadenitis and terminal ileitis. In adults, the most frequent combination of symptoms is abdominal pain, fever, and diarrhea, although up to 40% of adults can present with the symptom complex of acute appendicitis. The sigmoidoscopic findings in 13 adult patients with severe diarrhea and bacteriological confirmation of Y. enterocolitica consisted of normal mucosa in five patients, friable and diffusely erythematous and edematous mucosa in six, and aphthoid ulcers in two (4). The sigmoidoscopic features of our patient were similar to those patients with aphthoid ulcers. The delayed recovery of the pathogen from stool as was seen in our patient is not unusual. Isolation of Yersinia from stool is hampered by their slow growth and by overgrowth of normal fecal flora. With the use of cold enrichment of fecal specimens in phosphatebuffered saline and subculturing on appropriate medium, recovery of Yersinia is facilitated (5). 550

Recently, rectal biopsy was suggested to be helpful in distinguishing acute self-limiting colitis from idiopathic inflammatory bowel disease (9). The histological features that favor the di~tgnosis of idiopathic inflammatory bowel disease include distorted crypt architecture, increased numbers of both round cells and neutrophils in the lamina propria, villous surface, basal lymphoid aggregates, and basally located "isolated giant cells." Superficial crypt abscesses and superficially located "isolated giant cells" favored the diagnosis of self-limited colitis. In our patient, the histological findings of superficial crypt abscess without any histological features of idiopathic inflammatory bowel disease appear to support this concept. SUMMARY

We describe a 15-month-old male who presented with fever and diarrhea 24 hr after receiving antibiotics for otitis media. A flexible sigmoidoscopy was initially interpreted endoscopically as antibiotic-associated pseudomembranous colitis, and the Digestive Diseases and Sciences, Vol. 31, No. 5 (May 1986)

Yersinia COLITIS

patient was treated with vancomycin. The diagnosis of antibiotic-associated colitis was excluded in our patient by the negative stool examination for Clostridium difficile toxin, the failure to obtain supportive features on rectal biopsy, and the failure to demonstrate sigmoidoscopic improvement with vancomycin therapy. Thirteen days later, Y. enterocolitica was cultured from the initial stool specimens. In this case, the raised central whitish area on an erythematous base was misinterpreted as pseudomembranous colitis. REFERENCES 1. Tedesco FJ; Hardin RD, Harper RN, Edwards BH: Infectious colitis endoscopically simulating inflammatory bowel disease. Gastrointest Endosc 29(3):195, 1983

Digestive Diseases and Sciences, Vol. 31, No. 5 (May 1986)

2. Tedesco FJ, Moore S: Infectious diseases mimicking inflammatory bowel disease. Am Surg 48:243, 1982 3. Kohl S: Yersinia enterocolitica infections in children. Pediatr Clin North Am 26:433-443, 1979 4. Vantrappen G, Ponette E, Geboes K, Bertrand P: Yersinia enteritis and enterocolitis. Gastroenterological aspects. Gastroenterology 72:220-227, 1977 5. Boyce JM: Yersinia species. In Principles and Practice of Infectious Diseases. GL Mandel, RG Douglas, JE Bennett (eds). New York, Wiley Medical Publications, 1979, pp 1792-1800 6. Kohl S, Jacobson JA, Nahmsas A: Yersinia enterocolitica infections in children. J Pediatr 89:77, 1976 7. Lachman R, Soong J, Wishon G, Maenza R, Hanelin L, St. Geme J: Yersinia colitis. Gastrointest Radiol 2:133-135, 1977 8. Atkinson JR: Yersinia enterocolitica colitis in infants: Radiographic changes. Radiology 148:113-116, 1983 9. Surawicz C, Belic L: Rectal biopsy helps to distinguish acute self-limited colitis from idiopathic inflammatory bowel disease. Gastroenterology 86:104--113, 1984

55 1