A study to assess the prevalence and risk factors for Clostridioides difficile infection in patients with inflammatory bowel disease in a tertiary care hospital in Northern India

INTRODUCTION

Inflammatory bowel disease (IBD) is an entity that includes two distinct conditions, ulcerative colitis (UC) and Crohn’s disease (CD). The main characteristics of these two conditions are the chronic relapsing inflammation of the gut in individuals who are genetically predisposed, and exposed to defined environmental risk factors.^[1,2] According to the European Crohn’s and Colitis Foundation, all patients with IBD who are on corticosteroids, immunomodulators, and biological agents should be considered immunocompromised and at increased risk for opportunistic infections.^[3] In North America and Europe, over the past several decades, the prevalence of Clostridioides difficile infection (CDI) in IBD patients has increased rapidly.^[4] The clinical presentation of CDI varies widely and ranges from an asymptomatic carriage to fulminant colitis with toxic megacolon. However, the global epidemiology remains undetermined due to insufficient data from underdeveloped countries. In a retrospective study conducted by Vaishnavi et al., from a tertiary care hospital in India, 19% of patients with IBD were positive for C. difficile toxin.^[5] This study aims at enhancing awareness and improving the knowledge of CDI in IBD patients, which are important elements to optimize patient outcomes to implement preventive or early diagnostic and therapeutic strategies.

MATERIALS AND METHODS

This study was conducted in the Department of Microbiology in association with the Department of Gastroenterology from November 2017 to September 2019. Stool samples from 160 IBD cases, classified as UC and CD, were processed, diagnosed by clinical and endoscopic findings as per the World Gastroenterology Organization Global Guidelines on IBD.^[6] The severity of UC was classified into mild, moderate, and severe using the Truelove and Witt’s criteria.^[7,8] In this study, 112 patients without IBD but presenting with diarrhea were enrolled as controls. The demographic, clinical, and laboratory details of the patients were collected in a predesigned questionnaire. Anaerobic culture was put up on selective Cycloserine Cefoxitin Fructose Agar (CCFA) after alcohol shock and screening for glutamate dehydrogenase (GDH) was done by Meridian Bioscience Premier GDH kit while toxin detection was done by A/B by Meridian Bioscience Premier toxin A and B kit. For molecular detection, polymerase chain reaction (PCR) was done for TcdA and TcdB gene detection. For Toxin A (601bp) Forward Primer: YT28 - 5'CGATGATAAGGCAACTTCAGTGGTA 3' Reverse Primer: YT29 - 5' GAGTAAGATTCCTCCTGCTCCATTCAA 3'. For Toxin B (399 bp) Forward Primer: YT17 - 5' GGTGGAGCTGCTTCATTGGAGAG 3'Reverse Primer: YT18 - 5' GTGTAACCTACTTTCATAACACCA 3'.The reaction mixture contained 1x buffer (10 M Tris-HCL, 50 m MKCL, 1.5mM MgCl₂), 200 pmol of each deoxynucleoside triphosphate (MBI, ferments), 20 pmol of Toxin A and B primers, and 1.25 U of Taq polymerase (Bangalore gene, India). The template was denatured for 5 min at 94°C, and DNA was amplified for 30 cycles consisting of 1 min at 94°C, 1 min at 52°C, and 7 min at 72°C. The DNA fragments of C. difficile amplified by PCR were identified by agarose gel electrophoresis.^[9] C. difficile infection diagnosis was made as per the American College of Gastroenterology guidelines.^[10]

RESULTS

A total of 160 IBD patients were included in this study, out of which 24 had CD and 136 had UC. 45% patients were females, and 55% were males. The mean age of IBD patients was 37.19 ± 16 years. In this study, 7 out of 160 IBD patients and none of the non-IBD group had CDI confirmed by PCR (P = 0.04). The prevalence of CDI in IBD patients in this study was 4.4%. Predominant clinical symptoms present in IBD patients were diarrhea (98.75%), followed by blood in stool in 94.4%, fever in 8.12%, abdominal pain in 11.9%, and vomiting in 2.5%. Out of these, fever was significantly associated with CDI as compared to patients without CDI (P = 0.001).

In this study, CDI in IBD patients was detected by culture on selective CCFA media, GDH enzyme-linked immunoassay (ELISA), Toxin ELISA, and PCR. Out of 160 patients, only 1 (0.6%) culture was positive, 12 were positive for GDH by ELISA, 6 were positive for TcdA and TcdB by ELISA, and 7 were positive by PCR for toxin gene [Figure 1 a and b]. Among the control group, culture and PCR were negative in all, ELISA was positive in 9 (8%) patients. The sensitivity, specificity, positive predictive value, and negative predictive value of various tests are shown in Table 1.

Close

Figure 1:

(a) Gel doc picture showing toxin A gene; Lane 1- Ladder, Lane 2- sample, Lane 3- sample, Lane 4- sample, Lane 5- sample, Lane 6- sample, Lane 7- sample, Lane 8-sample, Lane 9- negative control, and Lane 10- positive control. (b) Gel doc picture showing toxin B gene; Lane1- Ladder, Lane 2- sample, Lane 3- sample, Lane 4- sample, Lane 5- sample, Lane 6- positive control, Lane 7- sample, Lane 8- positive control, and Lane 9- negative control.

Export to PPT

The risk factors, including use of medications such as 5 amino salicylic acid (5-ASA), levofloxacin, cephalosporins, proton-pump inhibitors (PPIs), metronidazole, infliximab, and steroids were studied. Out of 7 toxin-positive cases by PCR, 5 were on azathioprine, 6 on mesalamine, 5 on levofloxacin, 2 on cephalosporin, 1 on infliximab, 5 on steroids, 5 on PPI, and none was on metronidazole. The factors found significantly associated with CDI were PPI use (P = 0.001), levofloxacin (P = 0.001), and azathioprine (P = 0.042), as shown in Table 2.

Clinical outcomes of patients with CDI: Out of 7 CDI patients, only 1 was outpatient, and 6 were inpatients. The maximum duration of hospitalization among patients with CDI was 14 days,with a mean duration of 7.83 days. In this study, among the total GDH-positive cases, 25% were UC in remission, 25% had UC relapse, 33% had severe UC, and 17% were CD patients [Table 3]. All toxin-positive cases presented as IBD flare, and none was in remission. All CDI patients received oral vancomycin as treatment and these patients subsequently recovered.

DISCUSSION

To the best of our knowledge, this is the largest prospective study on C. difficile infection in patients with IBD in India. There are many studies on C. difficile infection in hospitalized patients, but very few studies have been conducted emphasizing CDI in IBD patients. We found the prevalence of CDI in IBD patients 4.4%, which is less as compared to other studies. There are very few studies from India on C. difficile infections in IBD patients. Kochhar et al., found that C. difficile toxin was present in 6 of 25 (24%) of patients with active colitis. Balamurugan et al. found that C. difficile was detected in 34 of 37 patients with UC.^[11,12] These studies included a small number of patients. Another study by Vaishnavi et al., found a prevalence of 19% by toxin detection among patients with IBD, but it was a retrospective study, and only ELISA for toxins was done for diagnosis.^[5] The reason for the low prevalence rates observed in our study might be that we have used PCR as a confirmatory method, which clearly ruled out false positive test results.^[13]

The results of our study are in accordance with the study by Masclee et al., which reported toxigenic C. difficile in 3.6% IBD patients.^[14] In another study by Gillespie et al., from the USA, the incidence of CDI among IBD patients was 6.7% among 654 IBD patients, which is closer to our study.^[15] Another case–control study by Li et al. 2018 reported 7.41% of CDI cases among IBD patients.^[16]

In our study, all the CDI cases had UC, but none had CD. Risk factors that were found to be significantly associated with the development of CDI in IBD patients in this study were use of PPI (P = 0.001), fluoroquinolone (P = 0.001), and azathioprine (P = 0.042). However, there was no significant relationship between the use of infliximab, 5-ASA, and steroids. Janarthanan et al., showed a significant relationship between CDI and PPI therapy, which is consistent with our study; while in contrast, Bossuyt et al., did not find any association between PPI use and CDI in IBD patients.^[17,18] In our study, significant association between azathioprine use and CDI in IBD, which is in accordance with the study by Issa et al., which reported a greater than a two-fold increased risk of CDI with maintenance immunomodulator azathioprine.^[19] Marwick et al., showed fluoroquinolones, which are widely used in IBD as a risk factor for CDI similar to findings in our study.^[20] Schneeweiss et al. reported that the risk of CDI tripled with corticosteroid initiation among the IBD patients, while no such association was found with the initiation of immunomodulators or biologics (infliximab).^[21] Similarly, there is no study from India that had compared the sensitivity and specificity of different testing methods for CDI. In this study, we found that GDH has high sensitivity so it can be used for exclusion of C. difficile infection or carriage. This is in accordance with a study by Swindells et al.,^[22] which showed toxin detection by ELISA methods had lower sensitivity. Moreover, the most appropriate diagnostic results are obtained with PCR following combined antigen and toxin detection.^[22]

The strength of this study is a prospective design in which consecutive patients were enrolled. Most of the previous studies have used retrospectively collected cases. A multistep algorithm is used for diagnostic approaches of C. difficile infection in symptomatic patients. However, one of the limitations of the study is the toxigenic culture,which is considered as gold standard for toxin detection, was not done as it is not suitable for routine diagnosis since it is a time-consuming and labor-intensive method. In our study, PCR is taken as a reference method for detection of toxin. In our study, diagnosis of C. difficile infection was made by laboratory diagnosis of toxigenic strain confirmed by PCR along with clinical symptoms. All patients with C. difficile infection were relieved of the symptoms by oral vancomycin. In addition, we performed PCR in all cases, which is considered as the highly sensitive method for confirmation of C. difficile infection, whereas the previous studies have diagnosed CDI using toxin ELISA,which has a low sensitivity. Therefore, in this study, the overall prevalence of CDI in patients with IBD was relatively more precise.

CONCLUSIONS

An IBD flare and CDI have similar presentations; therefore, an early diagnosis is important. In this study, GDH ELISA had high sensitivity and low specificity. GDH ELISA has low cost, thus making it appropriate for a screening test for CDI. More studies are required to determine whether the patients with negative GDH and toxin enzyme immunoassay results and positive for toxin PCR results should be treated or not. However, this prospective study provides significant data on the prevalence of CDI in IBD in Indian patients, associated risk factors, and sensitivity and specificity of different testing methods.