diarrhea

A Young Woman with Persistent Diarrhea

Ronald Rubin, MD—Series Editor

A 32-year-old woman presents with diarrhea of about 36 hours duration. Her illness began by awakening her in the middle of the night with cramps followed by several formless, watery stools. There were 2 episodes of emesis as well. By the morning, the emesis had stopped but she continued to have watery bowel movements preceded by cramps at least 4 to 6 times in the following day and night. She did not have a fever and did not pass any blood with her bowel movements. 

History

She is an otherwise healthy woman and takes no chronic medications. She works as a paralegal for a law firm and no one at work has been ill. She is married and thus far her husband is not sick. They both have eaten the same foods over the last several days. However, their 3-year-old son was sent home from his day care center 3 days ago with diarrhea, which had been a recent problem with several other children at the center as well. There is no unusual travel history.

Physical Examination

On exam, she was afebrile and seemed clinically enhydrated. Abdominal exam revealed a soft and nontender abdomen with hyperactive bowel sounds. Rectal exam revealed mildly inflamed external hemorrhoids and a stool specimen tested negative for blood.

Laboratory Testing

Routine labs revealed a normal hemoglobin and white blood count. Basic metabolic panel was entirely within normal limits.

Which of the following is the correct statement in reference to the presented patient? 

A. The patient should be quickly referred to a gastroenterologist for flexible sigmoidoscopy and colonoscopy.

B. The patient’s presentation is most suggestive of a food poisoning with a preformed toxin.

C. Polymerase chain reaction assay of stool specimens is a sensitive and specific method of pathogen identification and is indicated here.

D. In most cases, the determination of the precise cause of diarrhea is costly and not necessary.

The patient fulfills the currently used definition of diarrhea—the passage of more than 3 unformed stools in a day. Her diarrhea is less than 14 days duration so is acute (persistent being 14-29 days and chronic ≥30 days) and has been accompanied by other enteric symptoms of emesis and cramping, and therefore, can be defined as a case of gastroenteritis.1 

This is an extremely common illness in the United States and our evolving technology and armamentarium of microbiologic identification techniques has demonstrated a vast array of etiologic causes. However, using well-defined epidemiologic factors and good clinical evaluations, the clinician should be able to generate a reasonably accurate and tight differential diagnosis without a costly and difficult deluge of testing. And, although our much improved specific therapeutics for many of the microbiologic causes of diarrhea make accurate diagnosis more important than ever, this aspect too can be accomplished in a focused and efficient way.

Diagnosis

First, use readily available statistics of probability of causation together with any specific epidemiologic features of the patient.

In this scenario, noroviruses are the cause of gastroenteritis in about 50% of diarrhea outbreaks in the United States and are the prime suspected cause in cases associated with closed populations (eg, cruise ships, dormitories, hospitals, and day care centers). They are passed from person to person very efficiently among these closed populations.2 

Less common but still very important, diarrhea causes and their associated epidemiological matches include norovirus spread by foodborne epidemics with contaminated green leafy vegetables as the main source, food poisoning due to preformed toxins and characterized by a very short (2-7 hours) incubation period and frequently readily identifiable exposure, and Clostridium difficile-related, severe, almost cholera-like diarrhea in patients, particularly the elderly, with hospital and antibiotic exposure histories and a brisk leukocytosis response.1,3 

Our patient had a rather bland history with the only positive historical fact being a diarrheal illness in her child who was part of an “outbreak” in a closed population day care center. This fact makes the leading statistical, empiric diagnosis of norovirus most likely.

Other important clinical findings in diarrhea cases include the presence or absence of blood in the stool. The presence of blood suggests the presence of colitis; with organisms, such as Campylobacter, Shigella, and Shiga, toxin-producing Escherichia coli and salmonella come into play. The presence of significant temperature is characteristic of bacterial cause and the diarrhea lasting >14 days (considered persistent) is characteristic of giardia. None of these were present in our case.

Differential Diagnosis

Once the initial constellation of epidemiologic factors, history, and initial lab findings has been obtained, a reasonable diagnostic approach can be formulated. In our case, norovirus is the most likely diagnosis and no further initial workup is required; Answer D is the best and correct answer here. 

Her history reveals exposure to a likely highly contagious child who was part of an outbreak in a typical, closed population cluster. Her incubation period was 24 to 48 hours rather than >8 hours and there is no history of any suspicious food exposure; therefore, food poisoning associated with a preformed toxin (Answer B) is unlikely and not correct.

The use of early endoscopy in acute diarrhea has very limited value unless the history and findings suggest a possibility of C difficile or the diarrhea becomes persistent.4,5 Neither of these are present in the presented case, thus Answer A is not correct.

Regarding precise identification of norovirus, this is currently indeed possible with polymerase chain reaction techniques probing for viral nucleic acid. However, this technique has shortcomings, including the inability to always differentiate between pathogenicity and carriage as well as a low specificity in relation to its high sensitivity. Further, there currently is no specific therapy for norovirus in contradistinction to Salmonella/Shigella, Campylobacter, C difficile, and others. Thus, Answer C is not correct here.

Outcome of the Case

An office obtained complete blood count and comprehensive metabolic panel were essentially normal without evidence of significant dehydration or leukocytosis. She was prescribed oral fluids and loperamide to diminish the passage of stools. She was advised to return in the event of passage of blood, presence of a significant fever, or if the diarrhea persisted.

Take-Home Message

Acute infectious diarrhea is extremely common in the United States. A wide variety of etiologies are causative and costly if a wide array of diagnostic testing is needed to identify the specific causative agent. However, the astute clinician can use epidemiologic and clinical clues to ascertain the presumptive cause and its potential risk and response to specific therapies. Noroviruses are by far the most common cause in the United States and as a rule does not usually require aggressive diagnostic workup.

Ronald Rubin, MD, is a professor of medicine at Temple University School of Medicine and chief of clinical hematology in the department of medicine at Temple University Hospital, both in Philadelphia.

References:

1.DuPont HL. Acute infectious diarrhea in immunocompetent adults. N Eng J Med. 2014;370:
1532-1540.

2.Wikswo ME, Hall AJ. Outbreaks of acute gastroenteritis transmitted by person-to-person contact—United States 2009. MMWR Surveil Summ. 2012;61:1-12.

3.DuPont HL. Bacterial diarrhea. N Eng J Med. 2009;361:1560-1569.

4.Shen B, Khan K, Ikenberry SO, et al. The role of endoscopy in the management of patients with diarrhea. Gastrointest Endosc. 2010;71:887-892.

5.Donowitz M, Kokke FT, Saidi P. Evaluation of patients with chronic diarrhea. N Eng J Med. 1995;332:725-729.