Advances in the diagnosis and treatment of Clostridium difficile infections

Clostridium difficile is a leading cause of antibiotic-associated diarrhea worldwide. The diagnosis of C. difficile infection (CDI) requires both clinical manifestations and a positive laboratory test for C. difficile and/or its toxins. While antibiotic therapy is the treatment of choice for CDI, there are relatively few classes of effective antibiotics currently available. Therefore, the development of novel antibiotics and/or alternative treatment strategies for CDI has received a great deal of attention in recent years. A number of emerging agents such as cadazolid, surotomycin, ridinilazole, and bezlotoxumab have demonstrated activity against C. difficile; some of these have been approved for limited clinical use and some are in clinical trials. In addition, other approaches such as early and accurate diagnosis of CDI as well as disease prevention are important for clinical management. While the toxigenic culture and the cell cytotoxicity neutralization assay are still recognized as the gold standard for the diagnosis of CDI, new diagnostic approaches such as nucleic acid amplification methods have become available. In this review, we will discuss both current and emerging diagnostic and therapeutic modalities for CDI.


Introduction
Clostridium difficile infection (CDI) has become a significant healthcare-associated infection with a considerable economic impact throughout the world and is particularly important in developed countries. In the United States alone, CDI is thought to cause approximately 453,000 infections and 29,000 deaths every year, with an annual economic burden ranging from $436 million to $3 billion dollars 1,2 . More worrisome is the increasing prevalence of fulminant C. difficile colitis in the past several decades 1,3 . This is due in part to newly recognized hypervirulent strains such as the C. difficile BI/ NAP1/027 clone that expresses a binary toxin (CDT) in addition to the two large-molecule toxins TcdA and TcdB that are recognized as the primary virulence factors of CDI 2,3 .
Early detection of this pathogen and its toxins is critical as this allows earlier treatment that can significantly reduce the morbidity, mortality, medical cost, and family burden of CDI. The Food and Drug Administration (FDA) has approved a number of laboratory tests for the diagnosis of CDI. However, the epidemiology of CDI has drastically changed, with increasingly virulent strains emerging during the past decade 4 . An optimized and accurate diagnostic modality that can accurately differentiate CDI vs. colonization is urgently needed.
Antibiotic therapy remains the treatment of choice for CDI. While recommended antibiotics such as metronidazole, vancomycin, and fidaxomicin are still effective for this disease 1,5-7 , recent reports of C. difficile isolates with significantly reduced susceptibility and even resistance to these antibiotics suggest a potentially serious problem with the continued use of these agents to treat CDI 8,9 . Therefore, the development of new antibiotics and/or alternative treatment strategies as well as novel diagnostic approaches for CDI have become increasingly important. In this review, we will discuss both current and emerging diagnostic and therapeutic modalities for CDI.

Current laboratory diagnosis
An effective diagnosis of CDI requires both clinical symptoms and a positive laboratory test 7,10 . A key clinical manifestation is diarrhea, which is defined as loose stools; this means taking the shape of the receptacle or corresponding to Bristol stool chart types 5-7, plus a stool frequency of at least three stools over 24 (or fewer) consecutive hours or more 7 . Diarrhea should be accompanied by abdominal pain as well as by systemic features such as fever, hypotension, and/or shock. Severe ileus in which diarrhea ceases, leukocytosis, and elevated serum creatinine levels are particularly important and should receive more attention 11 . Advanced age (≥65), marked leukocytosis (leukocyte count >15 × 10 9 /L), decreased blood albumin (<30 g/L), a rise in the serum creatinine level (≥133 µM or ≥1.5 times the premorbid level), and comorbidities (severe underlying disease and/or immunodeficiency) also should be regarded as poor prognostic markers for severe CDI 7 . In addition, antibiotic exposure should be assessed as most patients with CDI have had antibiotic exposure in the previous 3 months; any current antibiotic therapy should be discontinued if possible. When a patient exhibits loose stools that correspond to Bristol stool chart types 5-7 and has other risk factors for CDI in the absence of another obvious cause such as diarrheagenic medications or a different type of diarrheal illness, a fecal specimen should be collected for laboratory testing to assess the possibility of CDI. Except in case of paralytic ileus, formed stool samples should not be tested for CDI 10 .
Currently, there is no single stool test that can be relied upon as the reference standard for the diagnosis of CDI. Several methods are recommended for the diagnosis of CDI, including toxinogenic culture (TC), cell cytotoxicity neutralization assay (CCNA), enzyme immunoassays (EIA) for toxins A, B, and/or glutamate dehydrogenase (GDH), and nucleic acid amplification tests (NAATs) 7,10,[12][13][14][15][16] . Among these methods, either the TC or CCNA has been considered to be the gold standard for the diagnosis of CDI over the past 30 years 17,18 . These tests use the following principles: TC detects the presence of C. difficile strains that actively produce toxin(s) (e.g., organism detection), whereas CCNA detects fecal protein toxins that have been produced in the stool (e.g., fecal toxin detection) [19][20][21] . It is noteworthy that both the TC and CCNA assays have limitations. These tests are labor intensive and have a slow turnaround time; thus, these tests are infrequently used for routine clinical diagnosis. TC has low specificity for clinical disease (CDI). The CCNA is analytically sensitive for toxin B, which historically has been considered a good marker for clinical disease. However, the performance of CCNA is dependent on pre-analytic factors and user experience and thus occasional clinical cases of CDI may be missed. Interpretation of these test results is subjective and requires some expertise; this may lead to poor reproducibility among different laboratories.
TC uses selective cycloserine-cefoxitin-fructose agar (CCFA) anaerobic culture and preliminary treatment with "heat shock" or "alcohol shock" in order to recover C. difficile from stool specimens; this method minimized the contaminating growth of other stool organisms. Suspect colonies are selected by presumptive tests, which include colony morphology, Gram stain, biochemical testing for indole (spot indole positive), and hydrolysis of L-proline-naphthylamide ("PRO Disk" positive). Presumptive identification of C. difficile is confirmed by RapID TM ANA (Remel Products, Thermo Fisher Scientific) or matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS). However, laboratory detection and identification of C. difficile alone does not diagnose CDI as 4% of healthy adults are colonized by C. difficile, and 20-25% of the C. difficile strains may be non-toxigenic 22 . Therefore, C. difficile isolates from positive TC tests should be evaluated for toxin production using CCNA and/or toxin EIA or evaluated by using NAAT to detect the presence of toxin A/B genes 10 .
Several commercial toxin-EIAs such as ProSpecT Toxin A/B (Remel Products, Thermo Fisher Scientific) and C. difficile Tox A/B II (TechLab, Inc.) as well as GDH-EIA tests such as C. Diff Chek-60 and C. Diff Quik Chek (TechLab, Inc.) have been introduced to the market. Overall, these products have a relatively low cost per test, but their sensitivity and specificity are not very good. The specificity of the toxin-EIAs varies widely, and sometimes their positive predictive values (PPVs) are inadequate for a diagnostic test 23 . While the GDH-EIA test methods are sensitive for screening C. difficile and also demonstrate a favorable negative predictive value (NPV) 23 , these methods are unable to differentiate toxigenic and nontoxigenic strains as both strains produce GDH 22 . Moreover, the GDH-EIA is not C. difficile-specific due to crossreaction with similar enzymes yielded by other clostridial species 24 . Therefore, EIAs combining GDH-EIA and toxin-EIA in one test have been developed. In 2009, TechLab, Inc. developed the C. Diff Quik Chek Complete to detect GDH and toxin A/B simultaneously. This combination test provided a rapid, cost-effective, and easy method for diagnosing CDI. With a 98% specificity and the results being available in 30 min, a negative result using the GDH-EIA can rule out CDI without the need for additional tests.
The use of NAATs for the detection of C. difficile from diarrheal stool specimens was documented in the early 1990s 23 . NAATs possess a series of advantages such as excellent sensitivity and specificity, low complexity, simplified reporting, reduced need for repeat testing, and improved turnaround time. It has been noted that the sensitivity of GDH screening tests for C. difficile is lower than that using NAATs, and NAATs for C. difficile toxin genes are superior to toxin-EIA testing as a standard diagnostic test for CDI 24 . Accordingly, NAATs are regarded as the most cost-effective method for the diagnosis of CDI 25 . Most NAATs target the encoding genes of TcdB, TcdA, and/or the binary toxin [26][27][28] . In particular, some NAATs such as multiplex NAATs can simultaneously detect C. difficile strains and toxin encoding genes from stool samples 29 . There are several commercially available NAATs, including a real-time PCR (RT-PCR) assay and loop-mediated isothermal amplification (LAMP) assay, both of which have an overall high analytical sensitivity (80-100%) and specificity (87-99%) 30 . In addition, several multiplex NAATs panels/platforms have been developed to detect stool pathogens including C. difficile from stool specimens. The first one is the Luminex xTAG GPP (Luminex Molecular Diagnostics Inc.), which targets 11 different stool pathogens (7 bacterial, 2 viruses, and 2 parasites); the second one is FilmArray™ Gastrointestinal Panel (BioFire Diagnostics), which is based on nested multiplex PCR and detects 23 stool pathogens (14 bacteria, 5 viruses, and 4 parasites) 30,31 . The sensitivities and specificities of these two assays for the detection of C. difficile were 91% and 100%, and 95% and 99%, respectively 32,33 . Moreover, multiplex NAATs can rule out diarrhea caused by other gastrointestinal pathogens even though C. difficile has been the most frequently detected pathogen in diarrheal disease. It is currently difficult to assess this novel syndromic approach to gastroenteritis due to its costs, insufficient data, and limited value in hospitalized patients. There are also some limitations with NAATs. These molecular methods do not differentiate between active toxin production in vivo and C. difficile colonization without toxin production 22 because NAATs only detect toxin encoding genes rather than directly detecting the toxin in stool. Thus, these molecular methods are unable to distinguish the CDI from colonization in patients with other reasons for a diarrheal illness, which may result in the overdiagnosis of CDI 34 . Therefore, additional tests such as an EIA test for C. difficile toxin A and/or toxin B may be needed to confirm the presence of toxin production in vivo and thus establish the likelihood of CDI and the need for treatment 11,22 . The final confirmation of CDI may require a combination of these tests along with the presence of clinical symptoms and signs.
Although NAAT methods are considered to be superior to other methods of diagnosing CDI, this testing strategy is unable to accurately distinguish between C. difficile colonization and disease 22 , which sometimes results in overdiagnosis of CDI 35 . Such overdiagnosis could result in overtreatment of CDI, delayed recognition of other causes of diarrheal illness/ outbreaks, unnecessary exposure of antibiotics used to treat CDI, and overestimation of hospital CDI rates 36 . Several test algorithms have been developed to improve the rapid and accurate diagnosis of CDI (Fig. 1). The European Society of Clinical Microbiology and Infectious Diseases (ESCMID) recommends the use of a twostep algorithm starting with either NAATs or GDH-EIA tests. Samples with a negative result from either NAATs or GDH-EIA tests can be reported as CDI negative, but those having a positive result should be confirmed by a toxin-EIA. Samples confirmed by this second toxin-EIA test can be reported as CDI positive 10 . This two-step algorithm may help reduce NAAT-related overdiagnosis of CDI.
Strategies such as intervention on the frequent use of PCR testing and/or using a modified NAAT cutoff to predict toxins also are options for reducing inappropriate   [36][37][38][39] . Because C. difficile PCR is a very sensitive method and frequently identifies colonized patients, the frequent use of PCR testing in low-risk patients may lead to the misdiagnosis of CDI. Therefore, highly sensitive PCR testing should not be ordered in patients with a low probability of infection (i.e., a patient without risk factors who has vomiting as a presenting complaint; a patient has a soft or formed stool; a patient with diarrhea who is using stool softeners or laxatives) 36 . In addition, quantitating C. difficile concentration in stool samples using RT-PCR can help to distinguish patients with CDI from the patients colonized with C. difficile who have diarrhea due to other causes, yet are still recognized as carriers of C. difficile requiring prevention of transmission 37 .
In certain clinical settings, adjunct testing methods such as radiologic diagnostic imaging may be useful for diagnosing CDI. Diagnostic computed tomography (CT) imaging can assist with an early diagnosis and also may determine the severity of the disease in patients with CDI 2 . For example, patients with ileus can be identified by clinical symptoms of severely disturbed bowel function such as vomiting and absence of bowel movements along with radiological signs of bowel distension. Patients with toxic megacolon can be identified by clinical signs of a severe systemic inflammatory response along with radiological signs of colon distension (>6 cm for the transverse width of the colon) 7 . However, it should be noted that the radiological diagnostic methods generally provide little value for the diagnosis of CDI 40 .

Strain typing
Strain typing also may be an important component of the laboratory diagnosis of CDI. In particular, ESCMID recommends strain typing of isolates recovered from the CDI case in an outbreak situation 10 . Accurate methods for strain characterization also are required for distinguishing circulating strains, new strains, and emerging pathogenic strain variants. There are several molecular typing methods available for the laboratory typing of C. difficile. Pulsed-field gel electrophoresis (PFGE) was one of the first typing methods used for the molecular analysis of C. difficile strains in North America. However, this method cannot accurately separate very large molecules of DNA and thus can result in subjectivity for the interpretation of the results, especially when there are subtle differences between the target strains and the reference strains 23 . In Europe, ribotyping (RT) has become the most frequently used method for C. difficile typing. This method demonstrates a good discriminatory ability, despite some difficulty in differentiating the closely related strains such as RT 027, 106, and 017 41,42 . Multilocus sequence typing (MLST) was used for the first time for C. difficile typing in 2004 43,44 . The method is particularly useful for typing C. difficile strains from stool specimen owing to the genetically heterogeneous characteristics of such strains. Mutilocus variable-number tandem-repeat analysis (MLVA) is a discriminatory molecular subtyping method based on capillary electrophoresis that can evaluate seven or eight loci on the C. difficile genome that also has proven useful for C. difficile strain typing. MLVA has also been used in outbreaks to identify phylogenetic strains as well as for performing surveillance for transmission 23,41,45 .
The epidemiology of CDI has drastically changed over the past several decades due to the emergence of highly virulent stains, which demonstrates the capability of C. difficile strains in general to become more virulent 4 . Therefore, it is likely that additional new epidemic strains of C. difficile will emerge and will require typing. The assessment of hospital-wide transmission with the use of MLST or RT has been hampered by the large numbers of patients who share a genotype and hospital-based contact. Whole-genome sequencing (WGS) shows that substantial genetic diversity exists among these strains, even within isolates of the same genotype 46 . In 2012, WGS using an Illumina platform was used for investigating the dissemination of C. difficile in the United Kingdom 47 . A previous study that defined the global demographic structure of C. difficile 027/BI/NAP1 using WGS and phylogenetic analysis has noted two distinct lineages (FQR1, FQR2) for the epidemic strain in North America. This approach has been able to track the spread of this strain throughout the world 48 . WGS is a powerful genotyping tool and has been used to quantify the role of symptomatic patients in C. difficile transmission 49 , to track the transmission of individual clones in infected hospital patients 50 , and to identify varying rates of C. difficile transmission between institutions 51 . However, high cost and a slower turnaround time as well as the challenge of a large data file that requires a suitable bioinformatics system for analysis are several concerns regarding WGS.

Current antibiotics for the therapy of CDI
Antimicrobial therapy remains the first choice for CDI, and specific antimicrobial therapy guideline recommendations should be based on the severity of the disease. Therapy of CDI should include both cessation of the inciting antimicrobial agent as soon as possible as well as implementation of infection control measures. If continued antimicrobial therapy is required for the treatment of the primary infection, antimicrobial therapy with agents that are less frequently implicated in antibioticassociated CDI should be used; these include parenteral aminoglycosides, sulfonamides, macrolides, vancomycin, or tetracycline/tigecycline (Fig. 2). Supportive care such as correction of fluid losses and electrolyte balance also should be addressed. Finally, the severity of CDI should be assessed as the first step of therapy. There is no consensus definition for severe CDI, nor which clinical indicators should be considered when defining severity. Current guidelines suggest that a serum albumin <3 g/dL plus either a white blood cell count ≥15,000 cells/mm 3 or abdominal tenderness without additional criteria for complicated disease are important clinical indicators that suggest severe CDI 24 . In addition, age >60 years, temperature ≥38.5°C, endoscopic evidence of pseudomembranous colitis, and admission to the intensive care unit are also useful clinical indicators that suggest severe CDI. A fulminant or complicated course is suggested by shock, megacolon, perforation, need for colectomy, as well as older age, high leukocyte count, and acute renal failure 24 . Moreover, there are relatively few classes of antimicrobial agents that are currently recommended for the therapy of CDI; these include the imidazoles (e.g., metronidazole), glycopeptides (e.g., vancomycin), and macrolides (e.g., fidaxomicin) 5-7 . Both metronidazole and vancomycin have been used for more than 30 years and remain the drugs of first choice 1,2,11 .

Severe, complicated, or fulminant cases of CDI
A combination of metronidazole (500 mg intravenously three times a day for 10-14 days) and vancomycin (125 mg orally four times a day for 10-14 days) is recommended for severe, complicated, or fulminant cases of CDI 1,11,24 . For patients with severe CDI who do not respond within 24-48 h or who develop complications such as renal failure or ileus, the oral dose of vancomycin should be increase to 500 mg four times a day. The use of additional vancomycin therapy via rectal retention enema, 500 mg in 100 mL normal saline every 6 h, is recommended if complete ileus is present. Fidaxomicin may be considered for patients who cannot tolerate vancomycin; the dose is 200 mg orally twice per day. Colectomy or ileostomy is still a useful adjunctive therapy for fulminant cases 1,11,24 . Close monitoring and supportive care is required for severe, complicated, or fulminant cases of CDI; surgery should be considered if the patient's clinical status is not improving and/or the serum lactate is above 2.2 mmol/L. Consider toxic megacolon if the patient develops abdominal distention with lessening of diarrhea, which strongly suggests paralytic ileus due to loss of colonic muscular tone.

Mild-to-moderate cases of CDI
Although oral metronidazole 500 mg three times per day for 10-14 days is still recommended for mild-tomoderate cases of CDI (defined by white blood cell count higher than 15,000 cells/mm 3 and blood urea nitrogen or creatinine levels above baseline) 1,11,24 , metronidazole is associated with more frequent side effects, and there has been a significant increase in treatment failures (especially in patients infected with the emergent 027/BI/ NAP1 strain).
Recent data have suggested an overall superiority of vancomycin to metronidazole for the treatment of patients with mild and moderate CDI 1 . However, vancomycin treatment is significantly more expensive than metronidazole. Oral vancomycin 125 mg four times per day for 10-14 days is recommended as an alternative for moderate cases 1,11 . For less severe cases, oral vancomycin 125 mg four times per day for 10-14 days is preferred, but oral fidaxomicin 200 mg twice a day for 10 days is also recommended if the risk of recurrence is high. Fidaxomicin is considered to have a similar therapeutic efficacy as oral vancomycin with a significantly lower recurrence rate 1,11 .

Recurrent cases of CDI
For recurrent cases of CDI, oral vancomycin 125 mg four times per day for 14 days or oral fidaxomicin 200 mg twice a day for 10 days is recommended for first recurrence. Vancomycin in a tapered and pulsed regimen, fecal microbial transplantation, or fidaxomicin 200 mg twice a day for 10 days is recommended for second or further recurrences 1,24 .
In addition to metronidazole, vancomycin, and fidaxomicin, other antibiotics such as rifaximin, nitazoxanide, ramoplanin, teicoplanin, and tigecycline have been used for cases where unacceptable adverse effects have been associated with standard therapy 1 . These antibiotics also have been used when there is need for salvage therapy due to fulminant disease, and surgery is not possible, as well as for the therapy of intractable recurrent infection 1 . However, these antibiotics are not recommended as drugs of first choice due to limited data, high cost, an unfavorable adverse-event profile, and/or resistance to C. difficile 1 .

Emerging antibiotics
There are a number of antibiotics that are currently being evaluated for the therapy of CDI. These agents are in different phases of clinical trials (Table 1). Among them, two unique types of antibiotics, cadazolid (developed by Actelion) and surotomycin (developed by Merck) have completed phase III clinical studies, while another two types of antibiotics, LFF571 (Novartis) and Ridinilazole (formerly SMT19969, Summit Pharmaceuticals), have completed phase II studies, and one antibiotic, CRS3123 (National Institute of Allergy and Infectious Diseases), is being evaluated in phase I studies 52 .
Cadazolid is a novel hybrid of oxazolidinone that displays a potent antimicrobial activity against C. difficile and has a lower propensity for inducing resistance 53 . In previous in vitro and in vivo antimicrobial evaluation studies, cadazolid was found to be more bactericidal than vancomycin. Moreover, cadazolid strongly inhibited de novo toxin A and B formation in stationary-phase cultures of toxigenic C. difficile and also inhibited C. difficile spore formation substantially at growth-inhibitory concentrations. In particular, cadazolid was active with a potency similar to that of vancomycin in conferring full protection of hamster and mouse models from diarrhea and death caused by C. difficile 54 . Another in vitro and in vivo study demonstrated that cadazolid had a stronger inhibitory effect on C. difficile in vivo than did moxifloxacin, linezolid, metronidazole, and vancomycin 55 . Finally, cadazolid more effectively treated CDI in a simulated gut model and had limited impact on the enumerated gut microflora and no signs of recurrence or emergence of resistance within the experimental timeframe 55 . Data from a multicentre, double-blind, phase 2 study of oral cadazolid in patients with CDI demonstrated a lower recurrence rate in patients that received 250, 500, or 1000 mg cadazolid twice daily than those receiving 125 mg vancomycin four times daily (18.2-25.0% vs. 50%) 56 . The patients treated with cadazolid had higher sustained clinical response rates than those treated with vancomycin (46.7-60.0% vs. 33.3%) 56 . A following study noted that cadazolid was more effective against the C. difficile strains isolated from the patients in this phase 2 study because the minimum inhibitory concentrations (MICs) for cadazolid were lower than those of vancomycin, linezolid, and moxifloxacin 57 . Cadazolid is now being evaluated in phase 3 clinical trials 52 ; the data from these studies are not yet available.
Surotomycin is an orally dosed, non-absorbed cyclic lipopeptide analog of daptomycin 58 . Surotomycin works through a mechanism that disrupts C. difficile cellular membrane activity in both logarithmic and stationary phases. Surotomycin is minimally disturbing to the normal gastrointestinal microbiota because of its lack of activity against Gram-negative anaerobes and facultative anaerobes. This antibiotic has been successful in reducing C. difficile vegetative cell counts and toxin levels in chemostat gut models 59 . A phase 2 clinical trial has shown that oral surotomycin at 125 mg twice daily or surotomycin 250 mg twice daily for 10 days resulted in a similar clinical cure rate (92.4% and 86.6%) to that seen with vancomycin 125 mg four times daily for 10 days (89.4%), but the recurrence rates of the surotomycin treatment groups (27.9% for surotomycin 125 mg twice daily and 17.2% for surotomycin 250 mg twice daily) are lower than the vancomycin treatment group (35.6%) 60 . More importantly, the sustained clinical response rates seen in this trial for the surotomycin groups (66.7% for surotomycin 125 mg twice daily, 70.1% for surotomycin 250 mg twice daily) at the end of study were higher than the vancomycin group (56.1%) 60 . In addition, a phase 3 clinical trial of surotomycin in subjects with CDI has been completed 60 . In this study, subjects with CDI confirmed by a positive toxin result were randomized to receive surotomycin (250 mg twice daily) or vancomycin (125 mg 4 times daily) orally for 10 days. However, the initial data from this trial suggest that surotomycin failed to meet the criteria for non-inferiority compared with vancomycin for the primary and key secondary endpoints. In this trial, both the clinical cure rate and sustained clinical response rate of the subjects treated with surotomycin were lower than those who received vancomycin treatment (clinical cure rate: surotomycin 79.0% vs. vancomycin 83.6%; sustained clinical response rate: surotomycin 60.6% vs. vancomycin 61.4%) 60 .
LFF571 is a semisynthetic derivative of GE2270A and works through inhibiting bacterial protein synthesis by interacting with elongation factor Tu (EF-Tu) and interrupting complex formation between EF-Tu and aminoacyl-tRNA 61,62 . It has been noted that LFF571 has an excellent activity against C. difficile as well as a good activity against other Gram-positive anaerobes, but little activity against the anaerobic Gram-negative organisms 63 . This drug was found to be more efficacious in the hamster model of CDI than vancomycin; it had a lower effective dose and fewer recurrences 64 . The safety and tolerability of LFF571 in humans has been assessed in a randomized, double-blind, placebo-controlled phase I trial, which demonstrated that the drug was generally safe and well tolerated in single and multiple oral doses in healthy volunteers 65 . Now a phase II trial comparing the safety and efficacy of LFF571 and vancomycin for CDI has been completed; its results suggest that LFF571 treatment is non-inferior to vancomycin treatment. The incidence of adverse events related to LFF571 in this study was higher than that related to vancomycin (76.1% vs. 69.2%) 66 . However, the true incidence of adverse events related to LFF571 needs to be further investigated as the study was limited by a small enrollment. It should be noted that LFF571 has a role in reducing C. difficile toxins. A recent study found that an in vitro treatment with LFF571 led to a reduction in toxin A and B production from various C. difficile strains 62 . Ridinilazole, formerly known as SMT19969, is a novel, narrow-spectrum, non-absorbable small molecule antimicrobial with activity against C. difficile 67 . However, the detailed mechanism of action of this drug remains unclear. This agent does not appear to act through the usual mechanisms associated with antibiotics, such as inhibition of cell wall, protein, lipid, RNA or DNA synthesis 68 . A recent study found that ridinilazole had a robust killing effect on C. difficile, which significantly reduced toxin production and attenuated the inflammatory response 67 . Ridinilazole also elicited significant cell division effects 67 . Those findings suggest a potential mechanism of action. The antimicrobial activity of ridinilazole has been well studied. A comparative in vitro study of ridinilazole against C. difficile found that ridinilazole was more active against C. difficile isolates including the 027/BI/NAP1 strains than fidaxomicin, vancomycin, and metronidazole 69 . Ridinilazole also showed limited activity against other members of the Gram-positive anaerobic intestinal flora 69 . A phase 2, randomized, double-blind, active-controlled, noninferiority study of ridinilazole has been completed; results from this study demonstrated that ridinilazole was well tolerated and had an adverse event profile similar to that of vancomycin 70 . Moreover, data from this trial established the non-inferiority of ridinilazole to vancomycin, as patients in the ridinilazole group had a higher sustained clinical response than those in the vancomycin group. Finally, ridinilazole showed statistical superiority at the 10% level in comparison to vancomycin. CRS3123 (Crestone Inc., Johnston, CO) is a small molecule narrow-spectrum agent with Gram-positive coverage and limited oral bioavailability, whose use results in a high concentration of the drug in the gastrointestinal tract and low systemic exposure 71 . This agent is a novel aminoacyl-tRNA synthetase inhibitor. It has been proposed that this unique antibiotic has a role in inhibiting both the growth, spore formation, and toxin production of C. difficile 71,72 . CRS3123 has been shown to be active against all C. difficile clinical strains tested, including the epidemic 027/BI/NAP1 strains. Treatment of CDI in the hamster model using oral CRS3123 has long-lasting efficacy with no recurrence. In addition, this agent does not display cross-resistance to existing antibiotics and remains active to the tested C. difficile strains 72 . CRS3123 is currently investigated in early-stage clinical trials.

Alternative treatment strategies
Since the expression of clostridial toxins (TcdA and TcdB) is mandatory for the development of CDI, the development of agents such as monoclonal antibodies aimed at preventing the cytotoxic effect of these toxins is a sensible strategy for controlling the disease. In 2016, the US FDA approved Merck's ZINPLAVA™ (bezlotoxumab) to reduce the recurrence of CDI in adult patients receiving antimicrobial therapy for CDI who are at high risk of CDI recurrence (http://www.mercknewsroom. com/news). Bezlotoxumab (MK-6072) is a human monoclonal antibody which reduces recurrent CDI by blocking the binding of C. difficile toxin B to host cells and therefore limiting epithelial damage and facilitating microbiome recovery 73,74 . The data from two separate phase 3 trials comparing bezlotoxumab with placebo among participants who were receiving standard-of-care therapy with oral vancomycin, metronidazole, or fidaxomicin showed that bezlotoxumab achieved a significant benefit over placebo 74,75 . Both the bezlotoxumab regimen and the actoxumab-bezlotoxumab regimen had good safety profiles without substantial adverse reactions; diarrhea and nausea were the most common adverse events 74,75 . The rate of recurrent infection in one of the phase 3 clinical trials was 17% vs. 28%, favoring bezlotoxumab (P < 0.001); and the rate in the another phase 3 clinical trial was 16% vs. 26%, also favoring bezlotoxumab (P < 0.001) 74,75 . Besides bezlotoxumab, Merck also developed another human monoclonal antibody, actoxumab (MK-3415), which is designed to neutralize C. difficile toxin A 75 . Although the data from human clinical trials that evaluate actoxumab alone are still limited, a phase 3 study found that a significantly lower rate of recurrent CDI (15.4%) was observed in the group of combination use of actoxumab and bezlotoxumab vs. the group using placebo (26.6%, both 1-sided, P < 0.0001), and global cure was slightly higher for the combination of actoxumab and bezlotoxumab (58.1%, 1-sided P = 0.0426) vs. placebo (53.7%) 74,75 . In addition, a combination of actoxumab and bezlotoxumab was found to have an ability to facilitate normalization of the gut microbiota in CDI mice 76 .
Effective vaccines against CDI are not yet available, but several promising vaccine candidates are being studied. According to information on the National Institutes of Health website ClinicalTrials.gov, there are three C. difficile vaccines currently in different stages of clinical trials: Cdiffense in phase III (NCT01887912), which is a vaccine containing toxoids of TcdA and TcdB from Sanofi Pasteur; IC84 in phase II (NCT02316470), which is a vaccine consisting of recombinant protein of the two truncated toxins TcdA and TcdB from Valneva, and a bivalent toxin vaccine from Pfizer in phase II (NCT02561195, NCT02117570). All three vaccine candidates target TcdA and TcdB. In addition, there are other vaccines being developed; these include recombinant vaccines based on the polysaccharide glycans, glycoconjugate vaccines, and DNA-based vaccines 52,71,77 . Many of these vaccines have displayed good efficacy for CDI under laboratory conditions or in clinic trials.
Fecal microbiota transplantation (FMT) is designed to restore the normal gut microflora for patients with CDI. This method has been used for more than 1000 years, but has only been recognized as a valid and effective method for the therapy of multiple recurrent CDI since 2010 78 . The use of FMT to treat recurrent/relapsing cases of CDI has proven to be safe and effective; its advantage over antimicrobial therapy is now being widely lauded 72 . Previous reviews of FMT have documented that oral or rectal transplantation of feces from a healthy, pretested donor combined with the simultaneous cessation of all antimicrobial use in the recipient is successful in treating >90% of patients with recurrent CDI 1,79 . Moreover, these reviews have found that the administration of vancomycin followed by an infusion of donor feces delivered by a nasoduodenal tube was safe and superior to vancomycin alone for recurrent CDI 1 . Particularly, FMT is now recommended as the therapeutic option of choice if there is a third recurrence after a pulsed vancomycin regimen 24 . However, the potential role of FMT in primary CDI is still not well understood.
Donor screening should be considered as the most important factor for FMT. When a proper donor is decided, his/her feces are handled via a series of processes including dilution of the specimen with normal saline (not the only choice of diluent but that which is generally used), homogenization, and filtration for direct transplantation or frozen for future use 80 . Routes of FMT administration include the upper gastrointestinal tract (with endoscopy, nasointestinal tubes, or pill ingestion), the proximal part of the colon by colonoscopy, or the distal part of the colon by enema, rectal tube, or sigmoidoscopy. A method that combines several of these methods of administration may be preferable in complex cases 78 . A recent study noted that fecal samples delivered via the lower gastrointestinal tract is a safe and effective treatment for refractory and recurrent CDI, and yields quicker results than delivering through upper gastrointestinal tract 81 . It should be pointed out that there is no FDA indication for FMT. Therefore, FMT for CDI can be performed only with an Investigational New Drug (IND) approval if the providers follow general ethical guidelines 1,78 . Moreover, it has been suggested that practitioners not having an experience in FMT first should consult FDA guidelines before performing FMT 78 . It also should be pointed out that the safety profile of FMT has not been well studied. Common postprocedural symptoms after FMT include abdominal pain, bloating, flatulence with borborygmus, diarrhea, constipation, vomiting, transient fever, and belching; these symptoms are usually transient and resolve within a few hours 78 .
Although FMT has an 80-95% success rate with longterm durability 81 , a number of disadvantages still exist. In particular, the manipulation of feces and the classical enteral administration methods are not only laborious, but tend to make the procedure rather unattractive for physicians and patients alike 52 . With regard to these disadvantages, a number of efforts have been made to enhance the feasibility and social acceptance of microbiota transplantation. Oral administration of capsulized intestinal flora is a good example of such efforts. Another example is the commercial application of intestinal microbiota restoration using live encapsulated microorganisms that are taken orally by the patient. On July 29, 2013, the US FDA has approved Rebiotix Inc.'s (Roseville, Minnesota) IND application to begin the phase 2 clinical study of RBX2660 for the treatment of recurrent CDI. RBX2660 is a commercially prepared, standardized, nextgeneration FMT live biotherapeutic product that initially was administered by enema for recurrent CDI. The data from the first of two phase II studies showed an overall 87.1% success rate of using RBX2660 to treat CDI 82 . SER-109 (Seres) is another live biotherapeutic that comprises an encapsulated mixture of purified Firmicutes spores, derived from human feces 83 . Two phase II studies of SER-109 have already been completed, and their results indicated that SER-109 was not effective overall at reducing CDI recurrence, but was efficacious in patients aged at least 65 years old 83 .
Probiotic therapy is another strategy for restoring the colonic microbiota that has been investigated for many years as a therapy to treat or prevent CDI. A study performed at Xinhua/Yuyao Hospital in Shanghai, China, has demonstrated that the use of a probiotic combination containing Lactobacillus acidophilus CL1285 and Lactobacillus casei LBC80R (Bio-K+) was well tolerated and effective in reducing the risk of antibiotic-associated diarrhea (AAD), including C. difficile-associated diarrhea (CDAD), in hospitalized patients on antibiotics 84 . Another observational study performed at a community hospital in Quebec, Canada, provides additional support. In this study, all adult inpatients on antibiotics in the community hospital Pierre-Le Gardeur (PLGH) between 30 April 2004 and 31 March 2014 were given a probiotic mixture containing L. acidophilus CL1285, L. casei LBC80R, and Lactobacillus rhamnosus CLR2 (Bio-K+) within 12 h of receiving an antibiotic prescription. During the 10 years of observation, 44,835 inpatients received the probiotic, and the CDI rate declined from 18.0 cases per 10,000 patientdays to an average of 2.3 cases per 10,000 patient-days 85 .
In the same study, 10-year data revealed that the CDI rates at PLGH were consistently and continuously lower compared with those at similar hospitals in Quebec 85 . Moreover, a number of systematic reviews of probiotics have suggested that probiotics are safe and may be effective in treating and preventing CDAD 86,87 . Finally, a recent study has found that a combination of probiotic and prebiotic products were more effective in preventing the germination of C. difficile spores and thus preventing CDI 88 . However, it should be noted that the evidence for the use of adjunct probiotics to decrease recurrences in patients with recurrent CDI is still limited. For example, another series of studies found no benefit with probiotic administration in the prevention of CDI 89,90,91,92,93 ; currently probiotic therapy is not recommended for the therapy of CDI due to the limited data and potential risk of bloodstream infection 2 .
In addition to FMT and probiotic therapy, minimizing the disruption of the normal gut microflora should be also considered. Some bacteriocins such as thuricin CD, nisin and lacticin 3147, NVB302, and GE2270 derivative have demonstrated selective and potent inhibition of C. difficile and other Gram-positive bacteria, but have little or no impact on other commensal gut microbes 77 . Other antimicrobial agents such as MGB-BP-3, OPS-2071, Av-CD291.2, lauric acid (derived from virgin coconut oil), berberine, and bovine lactoferrin also have been described in a recent review that systematically summarizes the experimental and off-label therapies for CDI 52 . While further studies are still required, many of those emerging agents might have a potential application for the prevention or treatment of CDI.
The administration of non-toxigenic C. difficile strains is another approach to prevent the colonization of the intestinal microbiota by toxigenic C. difficile. Much like FMT, this strategy uses non-toxin-producing C. difficile to colonize the gastrointestinal tract and thus prevent colonization by toxin-producing strains; this strategy has been shown to alleviate CDI in both patients and animal models 94,95,96,97 . Moreover, studies using animal models have demonstrated that the administration of nontoxigenic C. difficile strains provided good protection to hamsters and/or mice against a challenge with the hypervirulent BI/NAP1/027 C. difficile 95,96 . In a more recent phase II trial, colonization with the non-toxigenic C. difficile was found to correlate with reduced recurrence of CDI; recurrence of CDI in patients purposefully colonized with non-toxigenic C. difficile was 2% compared with 31% in patients who did not receive non-toxigenic C. difficile in order to colonize their intestinal microbiota 97 . Another commercial product, VP20621 developed by Shire, is composed of a non-toxigenic strain M3 and has finished a phase II clinical trial in 2013 (NCT01259726) 94 . This trial has demonstrated that the oral administration of non-toxigenic C. difficile strain M3 spores was well tolerated and safe; the non-toxigenic strain colonized the gastrointestinal tract and significantly reduced CDI recurrence 94 . It should be noted that although the results of this study are positive, the use of non-toxigenic C. difficile for the therapy of CDI is still thought to be questionable as the non-toxigenic C. difficile strains might acquire the known pathogenicity locus that encodes the C. difficile toxins from toxigenic strains via horizontal gene transfer and thus become toxin producers 98 .
Surgical consultation is recommended for all patients with complicated CDI. Such consultation should be considered for patients having any of the following symptoms/signs: hypotension with or without required use of vasopressors, fever ≥38.5°C, ileus or significant abdominal distention, mental status changes, WBC ≥35,000 cells/mm 3 or <2000 cells/mm 3 , serum lactate levels >2.2 mmol/L, and/or end organ failure (mechanical ventilation, renal failure, etc.) 32 . Complicated CDI with failure to improve on medical therapy after 5 days also requires surgical consultation 32 . For cases of CDI, the more negative prognostic signs a patient has, the earlier operative management should be considered as surgery is more likely to improve the survival when done in a timely manner.

Emerging strategies for the therapy of CDI
Due to the continued development of science and technology, a number of non-antibiotic therapeutics are emerging that may be useful for the therapy of CDI. While most of these therapeutics are still in the laboratory phase, their emergence might provide future therapies for the treatment of CDI. The potential use of bacteriophage lysin proteins is one of such emerging strategies. A bacteriophage lysin protein and its catalytic domain (PlyCD 1-174 ) cloned from the prophage sequence harbored in C. difficile CD630 genome has demonstrated an excellent capability for lysing C. difficile 99 . Indeed, the catalytic domain (PlyCD 1-174 ) was found to have a broader lytic spectrum against this pathogen 99 . In addition, subinhibitory doses of vancomycin combined with the PlyCD catalytic domain in vitro were significantly more bactericidal against C. difficile than was vancomycin alone 99 .
The human alpha-defensins HNP-1, HNP-3, and HD-5 also have been reported to have a potential role in preventing the cytotoxic effects of C. difficile toxin B in intestinal epithelial cells and in a large array of other cells 100 . This suggests a defense mechanism for human defensins against some types of clostridial glucosylating cytotoxins 100 . A similar study found that both HNP-1 and HD-5 also displayed good killing effects against C. difficile, with HD-5 having a significant bactericidal activity against hypervirulent C. difficile strains 101 . It is suggested that HD-5 used in combination with FMT therapies would be useful for treating recurrent forms of CID due to the antitoxin, bactericidal, and immunostimulatory effects of such a treatment 101 .
Another more recent study has described a synthesized, bioactive, low molecular weight organoselenium compound, ebselen, that directly targets the glucosyltransferase domain (GDT) of C. difficile toxins. This agent was found to have a good activity against both TcdA and TcdB 102 , and in treatment tests in a mouse model that closely resembles human infection, treatment with ebselen reduced the disease pathology in murine tissues by inhibiting the release of the toxic GDT 102 .
Tolevamer is a non-absorbable high molecular weight anionic polymer that could potentially absorb the toxins (TcsA and TcdB) involved in CDI. Clinical trials evaluating tolevamer for the therapy of CDI have been initiated. Even though a lower CDI recurrence rate (4.5%) was achieved with tolevamer than those of metronidazole (23%) and vancomycin (21%), tolevamer did not show promising results in terms of time to resolution of diarrhea and also had a lower rate of clinical success than the therapy with metronidazole and vancomycin 103 . However, anion-binding resins might still be useful as a substitute for antibiotics as an emerging alternative for the treatment of CDI.

Future perspectives
The current status of CDI management is still worrisome. Available therapeutic agents and effective vaccines for CDI remain limited, and the prevalence of CDI is increasing. Therefore, active prevention will play an increasingly important role in managing this disease more effectively. It has been documented that previous hospitalization, underlying disease, advanced age (≥65 years), and prior use of antibiotics are important risk factors for CDI 1,29 . Based on these risk factors, the Society for Healthcare Epidemiology of America (SHEA) and the Infectious Diseases Society of America (IDSA) have published a series of guidelines for the infection prevention and control of CDI 5 . As most CDIs are hospitalacquired, and C. difficile is commonly present in healthcare facilities 1 , measures aiming at preventing the spread of the disease in healthcare facilities should be implemented. Healthcare workers and visitors must use gloves and gowns on entry to a room of a patient with CDI, and they are required to wash hands with soap (or antimicrobial soap) and water after caring for or otherwise having contact with these patients; patients with CDI should be accommodated in a single room with contact precautions, or at least, patients should be cohorted and provided with a dedicated commode for each patient when private rooms are not available 5,24 . Moreover, environmental cleaning and disinfection also are important for reducing C. difficile transmission and lowering the incidence of CDI. Any potential environmental sources of C. difficile should be determined and removed as far as possible; environmental contamination in areas associated with increased rates of CDI can be addressed by using chlorine-containing cleaning agents or other sporicidal agents; but routine environmental screening for C. difficile is not recommended 5 .
Antibiotic stewardship is another important aspect for the prevention and control of CDI, as antibiotic use is regarded as the most important risk factor for CDI 1,29 . Almost all antibiotics have been associated with CDI, but ampicillin, amoxicillin, cephalosporins, clindamycin, and fluoroquinolones are the most frequently cited 1 . Therefore, implementation of an antimicrobial stewardship program is recommended; this program should minimize the frequency and duration of antibiotic therapy in general and specifically restrict the use of those antibiotics most frequently associated with CDI (i.e., cephalosporins and clindamycin) in order to reduce the risk of CDI 5,24 .
The effect of using probiotics to help prevent and control CDI is uncertain at present. While a number of investigators have documented that the use of probiotics composed of various microbial strains had an effect on preventing AAD as well as CDAD 84,85,86,87,88 , other researchers have found no benefit with the administration of probiotics for the prevention of CDI 89,90,91,92,93 . Therefore, the routine use of probiotics for preventing CDI is not recommended at present 1,5 .
In addition to the above recommendations, rapid diagnostic testing of the patients with diarrheal illness acquired in the hospital or associated with antimicrobial therapy is recommended 29 . A hospital-based infection prevention program combined with active surveillance also has been proposed to decrease the incidence of CDI 29 .
It is widely known that C. difficile also is able to colonize the intestinal tract of various animals, including dogs, cats, swine, cattle, poultries, and goats 104 . Although there is still a lack of direct evidence that CDI can be transmitted between animals and humans through contact, a large number of studies have demonstrated a close phylogenetic relationship between C. difficile originated from humans and animals 44,105,106 . In particular, the hypervirulent 078 isolate has been commonly detected in farm animals as well as in human cases of CDI in North America and Europe. Moreover, this hypervirulent strain also has been frequently detected in meat products 107 . It has been postulated that direct transmission of ribotype 078 strains has occurred between pigs and humans in the Netherlands 108 . Cases of human CDI caused by a livestock-associated ribotype 237 strain recently have been reported in Western Australia 109 . These findings would seem to confirm the possibility that C. difficile transmission from animals to humans can occur. Therefore, proper managements of animals, especially those that are in close contact with humans on a daily basis such as household pets, and/or animals that humans consume as food is also important for the prevention of CDI. Measurements such as frequently cleaning of farm animals/pets and their environment, careful handling of animal feces and dead animals, use of antibiotic stewardship in animal husbandry, and required vaccination of both pets and farm animals (when veterinary vaccines become available) are potentially effective and practical methods to prevent zoonotic CDI. In addition, because C. difficile and its spores also are commonly found in meats, fish, fruits, and vegetables 104 , good food habits such as washing meats, fish, fruits, and vegetables as well as thoroughly cooking meats and fish before eating may also contribute to the prevention of CDI.
In conclusion, proper and specific therapeutic management of CDI is important for the prevention and control of CDI. A rapid and accurate diagnostic approach for CDI also is a key step for the prevention and control of CDI. Although antimicrobial therapy remains the first choice for CDI, alternate treatment strategies such as FMT and surgical intervention should be considered. Because of the reduction of susceptibility of C. difficile to metronidazole, vancomycin, and fidaxocimin, as well as the potential for future problems with resistance, novel antibiotics and therapeutics should continue to be developed. While many guidelines have been suggested for the control and prevention of CDI, proper diagnostic testing and antimicrobial stewardship are the most important factors because antibiotic use remains the most important risk factor for CDI in the absence of effective vaccines. However, some non-medical factors such as proper management of pets and food-producing animals as well as developing good food practices should not be ignored.
Author details 1