Antibiotic-Associated Diarrhea and Clostridioides difficile Infection

Clostridioides difficile (formerly Clostridium difficile) is a spore-forming, toxin-producing anaerobic bacillus and the leading identifiable cause of antibiotic-associated diarrhea and health-care-associated infectious diarrhea.^[1] It colonizes the gut of roughly 3-5% of healthy adults without causing symptoms, and colonization rates are substantially higher in hospitalized and long-term-care populations — colonization alone is not infection and should never be tested for or treated. Disease occurs when antibiotic exposure (or another disruption of the normal colonic flora) allows a toxigenic strain to proliferate and produce toxin A and/or toxin B, causing mucosal inflammation that ranges from mild watery diarrhea through pseudomembranous colitis to fulminant colitis with toxic megacolon.^[1] This chapter is deliberately complementary to the general Acute Diarrhea in Adults chapter: it covers only what is specific to antibiotic-associated diarrhea and CDI — risk-factor identification, the testing methodology, severity-based treatment selection, the recurrence ladder, and the infection-control/stewardship messaging pharmacists are positioned to reinforce.

Typical presentation

New or worsening watery diarrhea — by convention, three or more unformed stools within a 24-hour period^[2] — developing during or after antibiotic exposure (most commonly within the course, but onset up to 8-10 weeks after the antibiotic was stopped is well described), often with lower abdominal cramping, mild fever, and leukocytosis. Symptoms unrelated to another clear cause (laxative use, an enteral feeding change, a new medication known to cause diarrhea) should raise suspicion.

Fulminant presentation

Hypotension or shock, ileus (which can paradoxically present with reduced stool output despite severe colitis — a common diagnostic trap), or toxic megacolon. This is a surgical/critical-care emergency, not an outpatient management scenario.^[3]

• Recent or current antibiotic exposure — the dominant risk factor. Fluoroquinolones, clindamycin, and broad-spectrum cephalosporins carry the highest risk of any antibiotic class because of how broadly they suppress the protective anaerobic colonic flora; penicillins (including amoxicillin/clavulanate) also carry meaningful risk.^[1]
• Proton pump inhibitor use — gastric acid suppression is thought to allow more vegetative C. difficile organisms to survive transit to the colon.^[1]
• Hospitalization or residence in a long-term-care facility, and duration of stay — the main source of both exposure to spores and exposure to broad-spectrum antibiotics.^[1]
• Age 65 years or older.
• Immunosuppression — post-transplant immunosuppressive therapy, active chemotherapy, or another significantly immunocompromising condition.^[1]
• Inflammatory bowel disease, or a history of bowel surgery.^[1]
• A prior episode of CDI — the single strongest predictor of a future episode, and the key factor that should push toward fidaxomicin over vancomycin at the next episode.
• Additional factors specifically associated with more severe disease if CDI occurs: older age, immunosuppressive therapy, recent surgery, and infection with a hypervirulent strain (e.g., NAP1/BI/027, historically responsible for a large share of Ontario outbreaks) — note that infection prevention and control practices are the same regardless of strain.^[1]

Who to test

Why a single NAAT is not enough on its own

A nucleic acid amplification test (NAAT/PCR) is highly sensitive and detects the genes for toxin production — it cannot distinguish a patient whose C. difficile is actively producing toxin and causing colitis from a colonized carrier who happens to harbour a toxigenic strain but is not clinically affected by it. Because of this, most laboratories use a multistep algorithm rather than NAAT alone: a glutamate dehydrogenase (GDH) antigen screen plus a toxin A/B enzyme immunoassay (EIA) run together, with NAAT used only to arbitrate discordant results (GDH-positive/toxin-EIA-negative), or a NAAT-first strategy with reflex toxin EIA on NAAT-positive specimens.^[4] A positive toxin EIA reflects active toxin production and correlates far better with genuine clinical disease than a NAAT-only result; interpret a NAAT-positive/toxin-EIA-negative result in the context of the whole clinical picture rather than as automatic proof of infection needing treatment.

Confirmed case definition

A confirmed case is diarrhea meeting the frequency definition above with laboratory confirmation of toxin A or B (by a validated EIA, NAAT for the toxin genes, or cytotoxicity assay); or visualization of pseudomembranes on sigmoidoscopy/colonoscopy; or histological/pathological diagnosis of pseudomembranous colitis; or a diagnosis of toxic megacolon.^[2]

Severity classification

Severity determines the treatment pathway (see Choose Initial Treatment).

These criteria are used consistently by Canadian antimicrobial stewardship protocols and the IDSA/SHEA framework alike.^[3][4]

• Resolve diarrhea and colitis while minimizing further disruption of the colonic microbiota.
• Identify and stop the inciting antibiotic as early as clinically possible.
• Prevent progression to fulminant disease, and recognize it immediately when it occurs.
• Prevent recurrence — the leading long-term challenge in CDI management, occurring in roughly 1 in 5 patients after a first episode and more often after each subsequent episode.
• Prevent transmission to other patients, residents, or household members through consistent infection-control practice.
• Avoid unnecessary antibiotic and testing exposure in patients who are colonized but not infected, or whose diarrhea has another clear explanation.

General measures

• Avoid antimotility agents (loperamide, diphenoxylate/atropine) and promotility agents throughout active CDI — slowing transit or increasing motility can worsen toxin retention and precipitate toxic megacolon.^[3]
• Discontinue the proton pump inhibitor if it is not clearly indicated — acid suppression is a modifiable CDI risk factor.^[3]
• Discontinue any standing bowel protocol/stimulant laxative for the duration of active diarrhea.^[3]
• For oral rehydration and dietary measures, see the Acute Diarrhea in Adults chapter — the general supportive-care approach there applies equally here.

Infection prevention and control

Counselling patients and household/long-term-care contacts

• After discharge home, person-to-person transmission within a household is uncommon — reassure patients this is not a high risk to family members with routine hygiene.^[1]
• Clean the home bathroom thoroughly with a regular household cleaner; reinforce thorough handwashing with soap and water, particularly after using the bathroom and before eating, for the patient and household/facility contacts.^[1]
• In long-term care specifically, reinforce that alcohol-based hand sanitizer alone does not reliably remove spores — soap-and-water handwashing at a non-patient sink remains important for both staff and the resident when feasible.
• Counsel that diarrhea recurring within about 8 weeks of finishing treatment is a recognized pattern (relapse/recurrence) and should prompt a call to the prescriber rather than self-treatment with an over-the-counter antidiarrheal.

Pregnancy

Metronidazole is not associated with congenital malformations and can be used when indicated. Oral vancomycin has not shown adverse pregnancy effects (minimal systemic absorption) and can be used. Fidaxomicin has limited pregnancy-specific data — reserve for when it is otherwise clearly preferred (e.g., significant recurrence risk) and involve obstetric/specialist input.

Breastfeeding

Metronidazole is transferred into breast milk — many sources recommend suspending breastfeeding for the duration of a metronidazole course and resuming once treatment is complete; vancomycin is considered probably compatible given minimal oral absorption.

Older adults / long-term care

The population at highest risk for both CDI itself and for severe/fulminant disease.^[1] Have a lower threshold to classify as severe and to seek prompt medical assessment, since atypical or blunted presentations (reduced stool output with ileus, confusion rather than classic peritoneal signs) are more common. Reinforce environmental cleaning and hand-hygiene practices specifically in shared long-term-care bathrooms and commodes.

Renal impairment

Oral vancomycin needs no renal dose adjustment — it acts locally in the gut. IV metronidazole (used for fulminant disease) needs no specific renal adjustment but active metabolites can accumulate in significant hepatic impairment (see below). Fidaxomicin needs no renal dose adjustment.

Hepatic impairment

Metronidazole is hepatically metabolized — use with caution and consider dose reduction in severe hepatic impairment, particularly for the IV regimen used in fulminant disease.

Immunocompromised patients

Higher risk of both initial infection and recurrence — this alone is a reason to prefer fidaxomicin over vancomycin for an initial episode.^[5]

Monitor clinical response — stool frequency and consistency, resolution of fever and abdominal pain — rather than repeat laboratory testing. Routine "test of cure" is not recommended: a toxin or NAAT-based stool test can remain positive for weeks after a clinically successful course, and a positive result at that point reflects residual colonization, not treatment failure.^[4] For severe or fulminant disease, monitor white blood cell count, serum creatinine, and lactate until clearly trending toward normal. Reinforce infection-control precautions and antimicrobial-stewardship messaging (see Common Errors to Avoid) at every follow-up contact. Because recurrence is common — most often within 8 weeks of completing treatment — counsel patients explicitly to seek reassessment rather than self-treat if diarrhea returns in that window, and confirm the inciting antibiotic was actually discontinued.

1. Confirm the diagnosis is still correct: reconsider testing timing/interpretation, and rule out an alternative or superimposed cause of ongoing diarrhea rather than assuming the CDI regimen simply needs more time.
2. Failure to improve after 7 days of appropriate therapy strongly suggests an alternative diagnosis (or, less commonly, a complication) rather than a need to simply extend the same course — reassess and involve infectious diseases and/or gastroenterology.
3. Reassess severity at any clinical deterioration — a patient started on a non-severe regimen who develops severe or fulminant features needs immediate escalation, not a wait-and-see approach.
4. Confirm the inciting antibiotic was actually stopped, and reassess the full medication list for anything (a new antibiotic, an antimotility agent, an unstopped PPI) that could be undermining response.
5. Manage a genuine recurrence using the recurrence ladder — the specific next step depends on which agent was used for the prior episode(s) and how many recurrences have now occurred (see Clinical Decision Algorithm and Drug Tables).
6. After two or more recurrences, especially following a failed vancomycin taper, refer for fecal microbiota transplantation rather than cycling through further antibiotic courses indefinitely.

• Testing an asymptomatic patient, or a patient whose loose stool is fully explained by a laxative given in the past 48 hours, "just to be sure" — this produces false-positive-for-disease results from colonization and leads to unnecessary treatment.^[1]
• Treating a positive test in a patient who has no clinically significant diarrhea, or whose symptoms have already resolved before the result returns.^[1]
• Defaulting to metronidazole as first-line therapy — it is no longer preferred for any severity except a genuine access/cost-limited non-severe case.^[3]
• Using loperamide or another antimotility agent for symptomatic relief during active CDI.^[3]
• Missing a fluoroquinolone, clindamycin, or cephalosporin course in the medication history because it "was just a short course" — these remain the highest-risk classes even for brief exposures.^[1]
• Continuing an unnecessary proton pump inhibitor through and after a CDI episode without reassessing the indication.^[3]
• Ordering repeat stool testing to confirm cure — a positive result can persist for weeks after successful treatment and does not by itself indicate treatment failure.^[4]
• Waiting for a positive test result before starting contact precautions in a patient with clinically suspected CDI.^[1]
• Cleaning a CDI patient's room or bathroom with a standard (non-sporicidal) disinfectant.^[1]
• Recommending a probiotic as a substitute for antibiotic therapy in active CDI — no probiotic has a role as CDI monotherapy.
• Not recognizing that reduced stool output in a patient with worsening abdominal pain and distension can mean ileus/fulminant colitis rather than "improving" CDI.^[3]
• Cycling through repeated same-agent antibiotic courses for a second or later recurrence instead of referring for fecal microbiota transplantation once that threshold is met.^[6]

How to use this table: Doses are adult doses and require patient-specific confirmation, including renal/hepatic adjustment. This table covers CDI-specific antibacterial and adjunct therapy only — see the Acute Diarrhea in Adults chapter's Drug Tables for oral rehydration solution and general antidiarrheal agents. Select any drug name to open its complete Pharmasist Drug Database page.

Initial episode, non-severe or severe disease

Fulminant disease (hypotension/shock, ileus, or toxic megacolon)

Not an outpatient scenario — obtain urgent infectious diseases, general surgery, and/or critical care consultation.

First recurrence

Choice depends on what was used for the initial episode.

Second or multiple recurrences