FormalPara Key Points

Along with antimicrobial exposure, advanced age has been shown to be a significant risk factor for the development and recurrence of, and mortality from, Clostridium difficile infection (CDI).

Only symptomatic patients should be tested for CDI, and routine surveillance or repeat testing on asymptomatic patients as a test of cure is discouraged.

The selection of therapeutic agent is based on disease severity, history of recurrence, and cost.

Isolation precautions should be initiated at the first suspicion of CDI.

1 Introduction

Clostridium difficile infection (CDI) is the most common cause of infectious diarrhea in healthcare settings, with significant morbidity and mortality in developed countries [1]. In a number of studies, advanced age, along with antimicrobial exposure, has been shown to be a significant risk factor for the development of CDI [2, 3], as well as CDI mortality and recurrence [4, 5]. Since CDI was first recognized as a toxin-induced diarrheal disease in the 1970s [6], there have been remarkable advances in diagnostic tests and management strategies. In recent years, the US has seen dramatic increases in CDI incidence, disease severity, and mortality [7]. These changes have been partially explained by the appearance of the fluoroquinolone-resistant strain, North American PFGE type 1 and PCR-ribotype 027 (NAP1/027). In one study that explored the molecular epidemiology of outbreaks caused by NAP1/027, the strain was found to be isolated more frequently from older patients (median age 75 years) [8]. Increased recurrence of CDI in the elderly is usually not due to the development of antibiotic resistance, but is considered to be related to decreased host immunity or reinfection due to the persistence of spores in the patient’s colon or in the environment. The reason for the substantial burden of CDI in the elderly has not been fully understood; however, it may be related to frequent healthcare exposure, necessity of more medications, altered intestinal microbiota and complicated comorbidities, including decreased renal function or co-existing infections (Table 1). Reported rates of C. difficile colonization are higher in the frail elderly population at long-term care facilities, and when CDI outbreaks occur they can be severe [9]. With the rise in use of feeding tubes to maintain nourishment of elderly, frail residents in long-term care facilities, C. difficile can also be introduced by this route [10]. A case-controlled study from the UK in patients over 65 years of age showed significantly higher mortality in patients aged 90 years and over [11]. Patients with femur neck fractures, who are generally elderly with poor reserve, were reported to have high mortality (47 %) from CDI [12]. Early recognition of the disease and timely management are crucial for optimizing the care of elderly patients with CDI.

Table 1 CDI in the elderly
Full size table

2 Epidemiology

There have been several epidemiologic changes in C. difficile which appear to have been influenced by the antibiotics predominantly used in each era. In the 1980s, the association between amoxicillin, cephalosporins, clindamycin, or combined antibiotic use and the risk of CDI was identified in many studies [13–22]. Around 1990, prospective observational studies using restriction endonuclease analysis (REA) typing revealed that asymptomatic excretors of C. difficile in hospital settings did not have an increased risk of developing subsequent symptomatic disease but could be a source of transmission to other patients who developed diarrhea or colitis from the same REA types of C. difficile [23, 24]. In 2002, Montreal and Quebec started to have an epidemic of CDI, with increased morbidity and mortality [25], and between 2003 and 2004, Quebec had 7004 cases of CDI and 1270 people died [26, 27]. In 2005, the dominant strain of this outbreak was characterized as North American PFGE type 1 and PCR-ribotype 027 (NAP1/027), and its alleged hyperproduction of toxins A and B may explain increased morbidity and mortality [28]. It was found to be strongly associated with fluoroquinolone use during the outbreak in Quebec [29]. Soon after the Quebec outbreak, NAP1/027 also became endemic in Europe [28, 30, 31]. During the period between 2000 and 2003 when NAP1/027 was endemic, the rate of CDI was severalfold higher among persons aged >65 years compared with those aged 45–64 years [32]. Between 2011 and 2012 there was an outbreak of severe CDI in Melbourne, and a new hypervirulent strain, RT244, was identified. It has a different mechanism in terms of toxin production and does not have resistance to fluoroquinolones [33]. Recent surveillance studies from Europe suggested that the prevalence of NAP1/027 decreased significantly, which was paralleled by a decrease in CDI rate and mortality [34, 35], while NAP1/027 remains the most predominant strain in the US [36]. Although higher mortality has been observed in NAP1/027 strains, attempting to predict disease severity based on strain type is not recommended because of the wide disease spectrum seen with this strain, ranging from asymptomatic colonization to fulminant disease [37, 38].

According to the National Hospital Discharge Survey 1996–2009, C. difficile rates for hospitalized persons aged ≥65 years increased 200 %, with increases of 175 % for those aged 65–74 years, 198 % for those aged 75–84 years, and 201 % for those aged ≥85 years [39]. A recent epidemiologic study showed that, in 2011, incidence estimates, recurrence rate, and death rate were higher among persons 65 years of age than those less than 65 years of age [40]. Although studies showed up to 40–50 % of CDI cases were acquired in long-term care facilities, most occurred at subacute rehabilitation facilities within 30 days of hospitalization and therefore the number of CDI cases for long-term care facilities may have been overestimated [41–43].

3 Diagnosis

A diagnosis of CDI is based on evidence of toxin, or the C. difficile organism itself, in a stool sample in the presence of compatible clinical signs and symptoms. The usual symptoms are watery, loose, or unformed stools generally more than three times within 24 h, often accompanied by fever. Radiographic evidence of ileus or toxic megacolon can also provide clues to the diagnosis, particularly in the absence of diarrhea.

It is recommended that only symptomatic patients be tested for CDI. The sensitivity, specificity, and positive predictive value of the non-culture, laboratory-based assays are lower in asymptomatic patients. Although asymptomatic carriers can be the source of transmission and cause symptomatic disease in other patients, routine surveillance or repeat testing on asymptomatic patients as a test of cure is discouraged because many patients asymptomatically shed C. difficile spores for weeks even after CDI is cured [44–46].

A number of laboratory tests are available for the detection of C. difficile toxins or organism, and they are currently categorized into six modalities: (1) enzyme immunoassay (EIA) for toxins A and B; (2) EIA for glutamate dehydrogenase (GDH; a protein produced by all C. difficile strains); (3) cell cytotoxicity; (4) culture for toxigenic C. difficile; (5) nucleic acid amplification test (NAAT; polymerase chain reaction or loop-mediated isothermal amplification); and (6) colonoscopy. Table 2 shows the characteristics of the diagnostic tests [47].

Table 2 Characteristics of diagnostic tests for Clostridium difficile infection [42, 44, 45]
Full size table

Repeating EIA for toxins A and B used to be recommended to increase its sensitivity; however, a retrospective study using a large number of samples suggested that 90.7 % of patients who had multiple stool samples tested for CDI were accurately diagnosed based on the first stool sample [48]. Due to the variable sensitivity in some of the toxin assay kits, and potential overdetection of colonization by NAAT, multistep test algorithms have been adopted in many institutions to increase the accuracy of CDI diagnosis. Although these algorithms consist of different numbers and types of tests, they share the basic concept of using inexpensive screening tests first (EIA for GDH or toxins A or B, or both), followed by more expensive tests (NAAT, toxigenic culture or cell cytotoxicity) for either discrepant results between two EIA tests or for a negative EIA test.

4 Management

4.1 Discontinuation of Antibiotics

Any antibiotic can disrupt the intestinal microbiota, which has barrier effects to prevent overgrowth of pathogens such as C. difficile. Antibiotic discontinuation alone can improve or resolve symptoms in some patients with CDI [49], and withdrawing the offending antibiotics is critically important in treating CDI. Concomitant use of antibiotics is associated with both decreased initial response to CDI therapy and durability of response [50]. When discontinuation of antibiotic therapy is not possible due to concurrent infections, fidaxomicin showed higher clinical cure rates compared with vancomycin [50].

4.2 Treatment

Metronidazole, vancomycin, and fidaxomicin are the therapeutic agents currently available for CDI, with the selection of therapeutic agents being based on disease severity, history of recurrence, and cost (Table 3). Capsule vancomycin is expensive; however, oral vancomycin can be inexpensive by formulating intravenous vancomycin for oral use. Treatment recommendations do not change with the age of the patient. However, according to pharmaceutical company data, following a single 500 mg oral or intravenous dose of metronidazole, subjects aged >70 years with no apparent renal or hepatic dysfunction had a 40–80 % higher mean area under the curve (AUC) of metronidazole active metabolite, with no apparent increase in the mean AUC of the parent compound of metronidazole compared with young healthy controls aged <40 years. In geriatric patients, monitoring for metronidazole-associated adverse events is recommended.

Table 3 Recommended medical therapy for Clostridium difficile infection [39, 62]
Full size table

4.2.1 Mild to Moderate Disease

CDI is considered to be mild to moderate when the white blood cell count is 15,000 cells/mL or lower and the serum creatinine level is less than 1.5 times the baseline level [51]. Oral metronidazole is often chosen for mild to moderate disease because of its low cost, and many studies have indicated slow clinical response or decreased efficacy of metronidazole [52, 53]. In a recent study, vancomycin showed better treatment success compared with metronidazole [54]. For these reasons—faster response, lower relapse rates, and increased toxicity—vancomycin has replaced metronidazole as the drug of choice.

4.2.2 Severe Disease

CDI is considered severe when the white blood cell count is greater than 15,000 cells/mL and serum creatinine is greater than 1.5 times baseline. Multiple studies have suggested that oral vancomycin is superior to oral metronidazole in treating severe CDI [46, 53, 55]. As an alternative, fidaxomicin can be used for severe CDI as it showed equivalent treatment response and a lower rate of later recurrence compared with vancomycin [56]. Fidaxomicin is not recommended for mild disease due to its high cost; however, it is recommended for use when the risk of recurrence is high.

4.2.3 Severe Complicated Disease

Severe complicated disease is defined as CDI with ileus, toxic megacolon, or shock. Elderly patients are at greater risk of severe disease and often have comorbid conditions that imperil their recovery. Oral vancomycin remains the drug of choice; however, oral administration may not be feasible in those with severe complicated CDI. Rectal administration of vancomycin and/or intravenous administration of metronidazole are often used as adjunctive therapies, although evidence of efficacy is limited due to the lack of randomized trials. Patients with severe complicated CDI should be evaluated promptly for surgical intervention. Studies regarding surgical management have been limited by the small sample size and lack of a standardized approach. While early surgical intervention can reduce mortality [57, 58], the risks associated with surgical procedures are high in those with severe complicated CDI, and the optimal timing of surgical intervention has been very difficult to determine [59]. The advocated surgical approach has been total or subtotal colectomy with end ileostomy. In one study, loop ileostomy with colonic lavage followed by postsurgical vancomycin administration via ileostomy also achieved reduced mortality and preservation of the colon [60].

4.2.4 Recurrent Disease

CDI recurrence can be due to either reinfection or relapse. Risk factors for recurrence include advanced age, concomitant or post-CDI antibiotic use, prolonged or recent stay in a healthcare facility, proton-pump inhibitor use, presence of comorbidities, absence of an antitoxin A antibody response, and infection with the NAP1/027 strain [61–63]. It is believed that recurrence is more likely induced by a failure of host immune response to mount protective immunity, or a failure in reconstituting the microbiota rather than antimicrobial treatment failure. The immunological senescence that accompanies aging may lead to impaired immune responses to C. difficile and contribute to the significant association between advancing age and increased risk of CDI recurrence [64]. Recurrence occurs in 20–30 % of patients with CDI after initial treatment. The Infectious Diseases Society of America (IDSA) guidelines and the European Society of Clinical Microbiology and Infectious Diseases (ESCMID) recommend treating the first recurrence of CDI with the same therapeutic agent as used in the initial episode. For subsequent recurrences, both guidelines recommend vancomycin in a tapered and/or pulsed regimen, and the ESCMID guideline also recommends fidaxomicin as an option. Tapered or pulsed vancomycin therapy is often used in clinical practice, although there are no randomized trials and there is no standardized regimen or duration [44, 65]. Metronidazole is not recommended due to the risk of neuropathy from prolonged use.

4.2.5 Fecal Microbiota Transplantation

Fecal microbiota transplantation (FMT) has continued to show remarkable clinical effectiveness, especially for recurrent and refractory CDI without serious adverse events. Overall response rates are approximately 90 % [66–70]. The rationale for this therapy is to re-establish normal gut microbiota and restore its colonization resistance by instillation of donor stool into the gastrointestinal tract of patients with CDI. FMT has been performed via nasojejunal tube, enema, gastroscope, colonoscope, or frozen fecal capsule preparation. A recent case series suggested that FMT was also effective for severe CDI [71]. In the ESCMID guideline, FMT in combination with oral antibiotic treatment is strongly recommended for multiple recurrent CDIs unresponsive to repeated antibiotic treatment [65]. However, caution is advised in treating immunocompromised hosts and those with toxic megacolon since severe outcomes and deaths have been reported [71].

4.2.6 Probiotics

Probiotics are generally well tolerated and have few adverse effects, with the exception of rare cases of fungemia with Saccharomyces boulardii [72, 73] or bacteremia [74] in immunocompromised hosts. Although there have been a number of studies published on probiotics in CDI prevention using various organisms and doses, the effectiveness as adjunctive therapy for CDI prevention or treatment is still unclear (Fig. 1). A recent systematic review concluded that probiotics, such as Saccharomyces boulardii and Lactobacillus GG, are effective for the primary prevention of CDI with moderate-quality evidence [75].

Fig. 1
figure 1

CDI prevention strategies for the elderly. CDI, Clostridium difficile infection

Full size image

5 Infection Control

5.1 Hand Hygiene

Spores of C. difficile are resistant to alcohol-based hand rubs and antibacterial soap. Healthcare providers’ hands are frequently contaminated with C. difficile after patient contact, and wearing gloves to provide a physical barrier is critically important to reduce the transmission of C. difficile [76].

5.2 Isolation and Contact Precaution

C. difficile can be transmitted directly and indirectly via contact with patients or their environment. Isolation precautions should be initiated at the first suspicion of CDI. Ideally, a patient with C. difficile should be isolated in a private room with a bathroom, and if private rooms are not available then the alternative is to cohort patients with C. difficile in the same room. If cohorting and separation are not feasible, strict contact precautions are necessary, including dedication of a bathroom to one patient and a bedside commode to the other. Although one study suggested that the use of gloves alone may be as effective as the use of gloves and gowns in preventing the transmission of C. difficile [77], current guidelines recommend the use of both gloves and gowns [78]. Although it is currently recommended that contact precaution may be discontinued when the patient no longer has diarrhea [79], an observational study has shown that skin contamination and environmental shedding of C. difficile often persist at the time of resolution of diarrhea, and persistent shedding is common 1–4 weeks after therapy [80]. Therefore, some experts recommend extending contact precautions for several days after diarrhea stops, or continuing them until hospital discharge if local rates of CDI are high.

5.3 Environmental Control

C. difficile spores can survive on environmental surfaces for as long as 5 months. Stethoscopes, thermometers, blood pressure cuffs, bedrails, call buttons, telephones, TV controls, light controls, bed sheets, scales, commodes, toilets, windowsills, tube feedings, and flow control devices for IVs can all be contaminated [81]. The patient’s room and all equipment used in the room should be disinfected with bleach [78, 82]. Ultraviolet irradiation has been shown to be effective in deactivating C. difficile endospores [83], and appears to be equivalent to bleach in decreasing environmental contamination with C. difficile spores [84, 85].

6 Conclusions

Advanced age is a major risk factor for developing CDI and it is associated with increased morbidity and mortality, as well as a higher risk of recurrence. Frequent exposure to healthcare environments, multiple comorbidities, increased use of antibiotics, age-related decrease in immune response, and altered intestinal microbiota are possible explanations for the disproportionate burden of CDI in elderly individuals. Timely diagnosis, appropriate treatment based on the severity of illness, and effective infection control are essential in managing CDI in the elderly.