microorganisms Article Impact of the Introduction of a Two-Step Laboratory Diagnostic Algorithm in the Incidence and Earlier Diagnosis of Clostridioides difficile Infection Nieves Sopena 1,2,3, *, Jun Hao Wang-Wang 4 , Irma Casas 2,5 , Lourdes Mateu 1,2,3 , Laia Castellà 6 , María José García-Quesada 6 , Sara Gutierrez 1 , Josep M. Llibre 1 , M. Luisa Pedro-Botet 1,2,3 and Gema Fernandez-Rivas 2,4 1 Infectious Diseases Department, Germans Trias i Pujol University Hospital, Carretera de Canyet s/n, 08916 Badalona, Spain; lmateu.germanstrias@gencat.cat (L.M.); saragserra@gmail.com (S.G.); jmllibre.fls.germanstrias@gencat.cat (J.M.L.); mlpbotet.germanstrias@gencat.cat (M.L.P.-B.) 2 Faculty of Medicine, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, 08193 Barcelona, Spain; icasas.germanstrias@gencat.cat (I.C.); gfernandezr.germanstrias@gencat.cat (G.F.-R.) 3 IGTP, Health Sciences Research Institute Germans Trias i Pujol, 08916 Badalona, Spain 4 Microbiology Department, Clinical Laboratory North Metropolitan Area, Germans Trias i Pujol University Hospital, 08916 Badalona, Spain; jhwang.germanstrias@gencat.cat 5 Preventive Medicine Department, Germans Trias i Pujol University Hospital, 08916 Badalona, Spain 6 Infection Control Nurse, Germans Trias i Pujol University Hospital, 08916 Badalona, Spain; lcastella.germanstrias@gencat.cat (L.C.); mjgarciaq.germanstrias@gencat.cat (M.J.G.-Q.) * Correspondence: nsopena.germanstrias@gencat.cat; Tel.: +34-934-978-510 Citation: Sopena, N.; Wang-Wang, J.H.; Casas, I.; Mateu, L.; Castellà, L.; Abstract: Our aim was to determine changes in the incidence of CD infection (CDI) following the García-Quesada, M.J.; Gutierrez, S.; introduction of a two-step diagnostic algorithm and to analyze CDI cases diagnosed in the study Llibre, J.M.; Pedro-Botet, M.L.; period. We retrospectively studied CDI (January 2009 to July 2018) in adults diagnosed by toxin Fernandez-Rivas, G. Impact of the enzyme immunoassay (EIA) (2009–2012) or toxin-EIA + polymerase chain reaction (PCR) algorithm Introduction of a Two-Step (2013 onwards). A total of 443 patients with a first episode of CDI were included, 297 (67.1%) toxin- Laboratory Diagnostic Algorithm in EIA-positive and 146 (32.9%) toxin-EIA-negative/PCR-positive were only identified through the two- the Incidence and Earlier Diagnosis of Clostridioides difficile Infection. step algorithm including the PCR test. The incidence of CDI increased from 0.9 to 4.7/10,000 patient- Microorganisms 2022, 10, 1075. days (p < 0.01) and 146 (32.9%) toxin-negative CDI were diagnosed. Testing rate increased from 24.4 https://doi.org/10.3390/ to 59.5/10,000 patient-days (p < 0.01) and the percentage of positive stools rose from 3.9% to 12.5% microorganisms10051075 (p < 0.01). CD toxin-positive patients had a higher frequency of severe presentation and a lower rate of immunosuppressive drugs and inflammatory bowel disease. Mortality (16.3%) was significantly Academic Editor: Lynne V. Mcfarland higher in patients with hematological neoplasm, intensive care unit admission and complicated disease. Recurrences (14.9%) were significantly higher with proton pump inhibitor exposure. The Received: 21 April 2022 two-step diagnostic algorithm facilitates earlier diagnosis, potentially impacting patient outcomes Accepted: 21 May 2022 and nosocomial spread. CD-toxin-positive patients had a more severe clinical presentation, probably Published: 23 May 2022 due to increased CD bacterial load with higher toxin concentration. This early and easy marker Publisher’s Note: MDPI stays neutral should alert clinicians of potentially more severe outcomes. with regard to jurisdictional claims in published maps and institutional affil- Keywords: Clostridioides difficile; Clostridioides difficile toxin; infection control and prevention; iations. healthcare-associated infection; immunosuppressed patients Copyright: © 2022 by the authors. 1. Introduction Licensee MDPI, Basel, Switzerland. This article is an open access article Clostridioides difficile (CD) is a common cause of healthcare-associated diarrhea, with distributed under the terms and increasing morbidity and mortality rates and concomitant hospitalization costs [1]. How- conditions of the Creative Commons ever, community-acquired Clostridioides difficile infection (CDI) is rising and currently Attribution (CC BY) license (https:// accounts for 25% to 40% of cases [1]. The incidence of CDI has increased in the last creativecommons.org/licenses/by/ decade, probably due to the growing numbers of patients at risk and improvements in 4.0/). diagnostic methods [2,3]. The incidence of CDI varies across countries, ranging from 2.8 Microorganisms 2022, 10, 1075. https://doi.org/10.3390/microorganisms10051075 https://www.mdpi.com/journal/microorganisms Microorganisms 2022, 10, 1075 2 of 12 to 15.8 cases per 10,000 patient-days [2–5]. These differences may be due in part to low suspicion awareness among clinicians and suboptimal diagnostic methods based only on toxin assessment [6]. Reported risk factors for CDI include older age, length of hospitalization, comorbidities and previous exposure to antibiotics and proton pump inhibitors (PPI) [7]. The clinical presentation of CDI varies from mild diarrhea to severe colitis [8]. CDI treatment depends on the severity of disease and the potential risk of recurrence [8]. The cornerstone of the diagnosis has been based on the detection of a toxigenic strain of CD in the stool. However, optimal diagnostic algorithms are still the subject of studies and not all centers offer the same panel of diagnostic tests with nucleic acid amplification test (NAAT) alone or a 2- or 3-step testing with CD glutamate dehydrogenase (GDH) + toxin and optional subsequent NAAT [9]. The existence of asymptomatic carriers with positive NAAT stool tests further complicates the diagnosis of CDI. A stepwise diagnosis protocol including toxins and NAAT could capture CDI cases not identified by CD toxin–enzyme immunoassay (EIA) and shed additional light on the severity of CDI outcomes through a surrogate estimation of CD bacterial load [10]. A surveillance study in Catalonia showed that the application of a multimodal training strategy that included an online course on CDI, two face-to-face workshops, and dissemina- tion of recommendations on its prevention and the use of an optimal diagnostic algorithm significantly increased the incidence of CDI from 2.20 cases per 10,000 patient-days in 2011 to 3.41 in 2016 [11]. An optimal diagnostic algorithm comprises a test with high sensitivity as a screening method, such as GDH, followed by a more specific test to detect toxins either by EIA in a first step or by molecular NAAT methods in a second step. The aim of the present study was to analyze the changes in the incidence of CDI in a university hospital over a decade with the introduction of a two-step laboratory diagnosis algorithm. Moreover, we analyzed epidemiological characteristics, risk factors, clinical presentation, diagnostic test, treatment and outcome of CDI patients diagnosed in this period. 2. Methods 2.1. Study Design and Patients This observational retrospective study was conducted at a 550-bed university-affiliated public hospital located in Badalona, Spain. All adult patients (≥18 years old) with a first episode of diarrhea diagnosed as CDI during the period 1 January 2009–31 July 2018 were included. Diarrhea was defined as ≥3 unformed stools in a 24 h period and symptoms lasting ≥ 24 h. A CDI case was a patient with diarrhea combined with a positive stool for CD toxin or for a toxigenic CD identified by polymerase chain reaction (PCR). Only unformed fecal samples following the Bristol stool scale were accepted for testing, after the patient’s medical history was reviewed. From 2009 to 2012, the diagnostic technique was a combined EIA, the C. DIFF QUIK CHEK COMPLETE® test (Abbott Laboratories, Chicago, IL, USA), which detects CD-specific GDH and CD toxin A/B in the fecal specimen. Starting in 2013, a real-time PCR (GenomEra® CD, Abacus Diagnostica, Turku, Finland) was added as the second step in a two-step screening protocol. Samples that were GDH-positive but toxin-negative were retested by PCR in order to detect the presence of the tcdB gene that encodes the tcdB protein in a toxigenic CD strain. Treatment of CDI followed hospital guidelines based on the European Society of Clini- cal Microbiology and Infectious Diseases (ESCMID) consensus and remained unchanged during the study period. Oral metronidazole was given to non-severe cases and oral vancomycin to severe cases [8]. For each case, we completed a database that included the following variables: demo- graphic characteristics, admission ward, place of acquisition of the infection, comorbidities, exposure to known factors associated with CDI in the last 30 days, symptoms, analytical parameters, diagnostic method, CDI treatment, outcomes and length of hospital stay (LOS). Data were obtained for all patients from their digital medical files. Microorganisms 2022, 10, 1075 3 of 12 Classification according to the site of acquisition of CDI: (1) Hospital-acquired CDI: infection identified >48 h after hospital admission and before discharge. (2) Non-nosocomial healthcare-related CDI: infection starting in the community or within 48 h of admission, in patients admitted to a health center (hospital, nursing home or community health center) in the 4 weeks prior to the onset of symptoms. (3) Community-acquired CDI: infection starting in the community or within 48 h of admission, with no admission to a health center in the last 4 weeks [12]. A case of severe CDI was defined when at least one of the following criteria was present: ≥15,000 white blood cells (WBC) per milliliter or >50% above the baseline increase in serum creatinine. As per the guidelines definition, a case of complicated CDI was defined when at least one of the following was present: circulatory shock, ileus, <2000 or ≥35,000 WBC per milliliter or >2.2 micromoles of serum lactate per milliliter [8]. A case was defined as recurrent CDI if a new episode of diarrhea with a positive stool test occurred within 8 weeks after the resolution of the previous episode with at least 10 days of treatment. Crude mortality was defined as death from any cause within 30 days of the diagnosis of CDI. 2.2. Patients Design Statement The design of the work has been approved by the Clinical Ethics Committee of the hospital (PI-18-150); the need for informed consent was waived due to the retrospective nature of the study. 2.3. Statistical Analysis Incidence density of CDI was calculated as the number of CDI diagnosed at the hospital per 10,000 hospital-days. The incidence density of tested stool samples for CD was calculated as the number of tested stool samples per 10,000 hospital-days. The rate of CD test positivity was calculated as the percentage of positive tests relative to the total number of tests performed. The incidence density of CDI and stool samples tested and positive samples for CD were calculated overall and per year to evaluate temporal trends. A Poisson regression model was used to assess longitudinal trends. Quantitative variables were described in terms of mean and standard deviation. Cat- egorical variables were expressed in absolute frequencies and percentages. A univariate analysis of the relationship between the variables and CD diagnostic test, mortality or recurrence was performed using the Student’s t-test for continuous data and the chi-square test for categorical variables. A logistic regression analysis adjusted for age and sex was per- formed, entering the variables that were significant on univariate analysis. SPSS Statistics 25 (IBM; Armonk, NY, USA) was used. 3. Results A total of 443 patients with a first episode of CDI were included, of which 297 (67.1%) were toxin-positive and the remaining 146 (32.9%) were identified thanks to the two-step algorithm including the PCR test in GDH positive/toxin negative cases. Seventy-five cases were diagnosed in the first time period before the introduction of the algorithm (2009–2012) while 368 cases were detected after its introduction. (2013–2018). 3.1. CDI Incidence along the Time Periods The overall mean incidence of CDI over the period 2009–2018 was 3.12 cases per 10,000 patient-days with a significant increase in incidence from 0.9 in 2009 to 5.5 in 2018 (p < 0.01) (Figure 1). This increase was observed regardless of CDI origin: hospital-acquired infections rose from 0.66 to 2.11, community-acquired from 0.12 to 0.83 and non-hospital healthcare- acquired from 0.24 to 0.60 (p < 0.01 for all). The overall testing rate for CDI was 46.1 tests per 10,000 patient-days, with a significant increase from 29.6 tests in 2009 to 65.7 per 10,000 patient-days in 2018 (p < 0.01). Moreover, Microorganisms 2022, 10, 1075 4 of 12 Microorganisms 2022, 10, 1075 4 of 12 the mean percentage of positive tests for CDI in the study period was 8.5%, with a significant increase from 3.9% in 2009 to 11.2% in 2018 (p < 0.01). Figure 1. Trends in the number of new cases and incidence incidence density density of of CDI CDI (2009–2018). (2009–2018).**PCR PCRwas was added as the second step in the two-step screening protocol. added as the second step in the two-step screening protocol. 3.2. Characteristics This increaseofwas the Patients observed regardless of CDI origin: hospital-acquired infections rose from 0.66 to 2.11, community-acquired The origin of CDI was hospital-acquired fromin 0.12 to 0.83 and 223 (50.3%) non-hospital cases, healthcare- healthcare-related in acquired 117 (26.4%)from and0.24 to 0.60 (p < 0.01 for in community-acquired all). 103 (23.3%). Demographics, risk factors, clinical The overall presentation, testing and treatment rate outcomes for CDI was are46.1 testsinper shown 10,000 Table 1. patient-days, with a signifi- cant increase A total offrom 29.6 tests 401 (90.5%) in 2009were patients to 65.7 per 10,000of hospitalized, patient-days in 2018were which 332 (82.7%) (p < 0.01). More- admitted to over, the the mean medical percentage department, of positive 41 (10.2%) tests to the for CDI surgical in the study department and period 36 (8.9%)was 8.5%, to the with a intensive significant care increase unit (ICU). Thefrom mean 3.9% in 2009 length to 11.2%stay of hospital in 2018 was(p28.6 ± 27.4 days and 76 patients < 0.01). (18.9%) were discharged to a long-term care facility (LTCF). The rate of discharge to an 3.2. Characteristics LTCF was higher of inthe Patientswith healthcare-associated CDI (non-hospital healthcare- patients related or hospital-acquired CDI) compared with The origin of CDI was hospital-acquired those in 223 withcases, (50.3%) community-acquired healthcare-related CDIin (30.3 vs. 10.6%, p = 0.02). 117 (26.4%) and community-acquired in 103 (23.3%). Demographics, risk factors, clinical Characteristics presentation, treatmentandand outcomes outcomesregarding are showntheinCDI diagnostic Table 1. method are shown in Table 2. Severe presentation (OR 2.1; 95% CI: 1.26–3.55) was significantly associated with CD Tabletoxin-positive 1. Demographics,while inflammatory risk bowelcharacteristics factors and clinical disease (IBD) of (OR 7.2; included patients CI 95% CI: 1.90–27.26) in the study (N and = 443immunosuppressive cases). treatment (OR 2.5; 95% CI: 1.34–4.62) were significantly associated with toxin-negative/PCR-positive. Characteristic N (%) Age, mean (yr., SD) 68.2 ± 16.3 Male gender 241 (54.4) Risk factors Chronic pulmonary disease 95 (21.4) Chronic renal disease 119 (26.9) Microorganisms 2022, 10, 1075 5 of 12 Table 1. Demographics, risk factors and clinical characteristics of patients included in the study (N = 443 cases). Characteristic N (%) Age, mean (year, SD) 68.2 ± 16.3 Male gender 241 (54.4) Risk factors Chronic pulmonary disease 95 (21.4) Chronic renal disease 119 (26.9) Diabetes mellitus 127 (28.7) Heart failure 99 (22.3) Solid organ cancer 136 (30.7) Hematologic neoplasm 47 (10.6) Liver disease 53 (12) Inflammatory bowel disease 21 (4.7) HIV infection 14 (3.2) Charlson index ≥ 3 164 (37.1) Abdominal surgery 19 (4.3) Chemotherapy 50 (11.3) Enteral nutrition 53 (12) Immunosuppressive drugs 82 (18.5) Non-CDI antibiotic use 348 (78.6) PPI use 304 (68.6) Clinical and analytical presentation Abdominal pain 206 (46.5) Fever 108 (24.4) Shock 33 (7.4) Severe case 174 (39.7) Complicated case 66 (15) Leukocytes > 15,000 cells/mm3 110 (25.2) Albumin (g/L, SD) 27.8 ± 6.5 Creatinine (mg/dL, SD) 1.7 ± 1.7 Hemoglobin (g/dL, SD) 10.7 ± 2.1 SD: standard deviation; HIV: human immunodeficiency infection; CDI: Clostridioides difficile infection; PPI: proton pump inhibitor. 3.3. CDI Treatment A total of 437 (98.6%) patients received a specific antibiotic for CDI: metronidazole in monotherapy in 312 (70.4%) and vancomycin in 122 (27.5%) cases, in monotherapy in 86 cases or associated with metronidazole in 36 cases. Fidaxomicin was only administered in three cases. The mean duration of treatment was 13.3 ± 5.6 days. The use of vancomycin increased from 20% of cases in 2009 to 43.2% in 2018 (p = 0.09). Patients with severe (56.6% vs. 33.9%, p < 0.001) or complicated CDI (27% vs. 10.4%, p < 0.01) more frequently received vancomycin than metronidazole, in accordance with guideline recommendations. Non-CDI dispensable antibiotics were discontinued in 154 (34.8%) cases and proton-pump inhibitors (PPI) in 26 (5.9%). Microorganisms 2022, 10, 1075 6 of 12 Table 2. Univariate and multivariate predictors of toxin-positive CDI. Toxin + Toxin −/PCR + p Value p Value Characteristic N = 297 (67%) N = 146 (33%) (Unadjusted Analysis) (Adjusted Analysis) N (%) N (%) Age, mean (SD) 69.8 ± 16.1 64.9 ± 16.3 0.003 Male gender 155 (52.2) 86 (58.9) 0.21 Chronic pulmonary disease 69 (23.2) 26 (17.8) 0.23 Chronic renal disease 83 (27.9) 36 (24.7) 0.53 Diabetes mellitus 85 (28.6) 42 (28.8) 1 Heart failure 75 (25.3) 24 (16.4) 0.05 Solid organ cancer 83 (27.9) 53 (36.3) 0.09 Hematologic neoplasm 27 (9.1) 20 13.7) 0.18 Liver disease 28 (9.4) 25 (17.1) 0.03 Inflammatory bowel disease 3 (1) 18 (12.3) <0.001 0.004 HIV infection 7 (2.4) 7 (4.8) 0.27 Charlson index ≥ 3 107 (36.1) 57(39) 0.21 Abdominal surgery 11 (3.7) 8 (5.5) 0.53 Chemotherapy 28 (9.4) 22 (15.1) 0.07 Enteral nutrition 30 (10.1) 23 (15.8) 0.11 Immunosuppressive drugs 43 (14.5) 39 (26.7) 0.002 0.004 Non-CDI antibiotic use 244 (82.2) 104 (71.2) 0.01 Proton pump inhibitor use 217 (73.1) 87 (59.6) 0.006 Abdominal pain 150 (50.5) 56 (38.4) 0.01 Fever 73 (24.6) 35 (24) 0.98 Shock 28 (9.4) 5 (3.4) 0.03 Leukocytes > 91 (31.2) 19 (13.2) <0.001 15,000 cells/mm3 Albumin (g/L, SD) 27.6 ± 6.6 28.1 ± 6.2 0.48 Creatinine (mg/dL, SD) 1.7 ± 1.7 1.4 ± 1.4 0.07 Hemoglobin (g/dL, SD) 10.7 ± 1.9 10.5 ± 2.3 0.37 Severe presentation 133 (45.2) 41 (28.5) 0.001 0.004 Complicated case 48 (16.3) 18 (12.4) 0.35 Non-CDI 118 (40) 36 (25.4) 0.004 antibiotic suppression PPI suppression 19 (6.4) 7 (4.9) 0.66 Recurrence 52 (17.6) 14 (9.6) 0.04 Death (30d) 42 (14.2) 21 (14.4) 0.95 SD: standard deviation; HIV: human immunodeficiency infection; CDI: Clostridioides difficile infection; PCR: polymerase chain reaction; PPI: proton pump inhibitor. 3.4. Clinical Outcomes A total of 63 (14.2%) patients died within 30 days of the CDI diagnosis. On multivariate analysis, hematological neoplasm (OR 2.8; 95%CI: 1.19–6.59), intensive care unit (ICU) admission (OR 5.4; 95%CI: 2.35–12.61) and complicated CDI (OR 2.8; 95%CI: 1.42–5.71) were associated with increased mortality (Table 3). Microorganisms 2022, 10, 1075 7 of 12 Table 3. Univariate and multivariate predictors of death (30 d). Deaths (30 d) Cured p Value p Value Characteristics N = 63 N = 380 (Unadjusted Analysis) (Adjusted Analysis) N (%) N (%) Age, mean (SD) 69.8 ± 17.7 67.9 ± 16.1 0.39 Male gender 32 (50.8) 209 (55) 0.60 ICU admission 14 (22.6) 22(6.5) <0.001 <0.001 Chronic pulmonary disease 16 (25.4) 79 (20.8) 0.51 Chronic renal disease 20 (31.7) 99 (26.1) 0.42 Diabetes mellitus 23 (36.5) 104 (27.4) 0.18 Heart failure 16 (25.4) 83 (21.8) 0.64 Solid organ cancer 15 (23.8) 121 (31.8) 0.25 Hematologic neoplasm 12 (19) 35 (9.2) 0.03 0.01 Liver disease 10 (15.9) 43 (11.3) 0.41 Inflammatory bowel disease 3 (4.8) 18 (4.7) 1 HIV infection 4 (6.3) 10 (2.6) 0.12 Charlson index ≥ 3 24 (38.1) 140 (36.8) 0.96 Abdominal surgery 1 (1.6) 18 (4.7) 0.49 Chemotherapy 6 (9.5) 43 (11.3) 1 Enteral nutrition 9 (14.3) 44 (11.6) 0.68 Immunosuppressive drugs 17 (27) 65 (17.1) 0.09 Non-CDI antibiotic use 55 (87.3) 293 (77.1) 0.09 Proton pump inhibitor use 39 (61.9) 265 (69.7) 0.27 Abdominal pain 26 (41.3) 180 (47.4) 0.44 Fever 15 (23.8) 93 (24.5) 1 Shock 12 (19) 21 (5.5) <0.001 Ileus 2 (3.1) 2 (0.5) 0.09 Toxic megacolon 2 (3.2) 0 0.02 Leukocytes > 26 (42.6) 64 (22.5) 0.001 15,000 cells/mm3 Albumin (g/L, SD) 23.6 ± 6.5 28.5 ± 6.1 <0.001 Creatinine (mg/dL, SD) 1.7 ± 1.5 1.6 ± 1.6 0.53 Hemoglobin (g/dL, SD) 10.2 ± 2.1 10.7 ± 2.1 0.08 Severe presentation 33 (53.2) 146 (38.4) 0.05 Complicated case 20 (32.3) 46 (12.2) <0.001 0.003 CD Toxin-positive 42 (66.7) 255 (67.1) 1 Vancomycin ± metronidazole 24 (38.1) 98 (26.4) 0.79 Non-CDI 15 (24.6) 139 (37) 0.08 antibiotic suppression PPI suppression 2 (3.2) 24 (6.3) 0.55 SD: standard deviation; ICU: intensive care unit; HIV: human immunodeficiency infection; CDI: Clostridioides difficile infection; PCR: polymerase chain reaction; PPI: proton pump inhibitor; Clostridioides difficile. The rate of CDI recurrence was 14.9%, and exposure to PPI (OR 2.29; 95%CI: 1.1–4.7) and severe previous presentation (OR 1.79; 95%CI: 1.03–3.12, p = 0.03) were significantly associated with increased risks on multivariate analysis (Table 4). Microorganisms 2022, 10, 1075 8 of 12 Table 4. Univariate and multivariate predictors of recurrences. Recurrence Non-Recurrence p Value p Value Characteristic N = 66 N = 343 (Unadjusted Analysis) (Adjusted Analysis) N (%) N (%) Age, mean (SD) 71.9 ± 12.8 67.1 ± 16.6 0.05 Male gender 36 (54.5) 173 (55.1) 1 UCI admission 4 (6.7) 18 (6.4) 1 Chronic pulmonary disease 17 (25.8) 62 (19.7) 0.27 Chronic renal disease 23 (34.8) 76 (24.2) 0.07 Diabetes mellitus 22 (33.3) 82 (26.1) 0.23 Heart failure 20 (30.3) 63 (20.1) 0.06 Solid organ cancer 18 (27.3) 103 (32.8) 0.38 Hematologic neoplasm 6 (9.1) 29 (9.2) 0.97 Liver disease 9 (13.6) 34 (10.8) 0.51 Inflammatory bowel disease 2 (3) 16 (5.1) 0.75 VIH infection 3 (4.5) 7 (2.2) 0.38 Charlson index ≥ 3 28 (42.4) 112 (35.7) 0.30 Abdominal surgery 5 (7.6) 13 (4.1) 0.23 Chemotherapy 1 (1.5) 42 (13.1) 0.003 Enteral nutrition 8 (12.1) 36 (11.5) 0.88 Immunosuppressive drugs 10 (15.2) 55 (17.5) 0.64 Non-CDI antibiotic use 54 (81.8) 239 (76.1) 0.31 Proton pump inhibitor use 56 (84.8) 209 (66.6) 0.003 0.02 Abdominal pain 36 (54.5) 144 (45.9) 0.19 Fever 16 (24.2) 77 (24.5) 0.96 Leukocytes > 24 (36.9) 60 (19.4) 0.002 15,000 cells/mm3 Albumin (g/L, SD) 28.1 ± 4.8 28.6 ± 6.4 0.59 Creatinine (mg/dL, SD) 2.2 ± 2.3 1.4 ± 1.4 0.001 Hemoglobin (g/dL, SD) 10.5 ± 1.6 10.7 ± 2.1 0.52 Severe presentation 35 (53) 111 (35.4) 0.007 0.03 Complicated case 6 (9.1) 40 (12.7) 0.40 CD toxin-positive 52 (78.8) 203 (64.6) 0.02 Vancomycin ± metronidazole 19 (29.7) 79 (25.7) 0.51 Days of treatment (mean, SD) 14.8 ± 7.4 13.7 ± 4.6 0.27 Non-CDI 27 (42.2) 112 (35.9) 0.34 antibiotic suppression PPI suppression 6 (9.1) 18 (35.9) 0.34 SD: standard deviation; ICU: intensive care unit; HIV: human immunodeficiency infection; CDI: Clostridioides difficile infection; PCR: polymerase chain reaction; PPI: proton pump inhibitor; CD: Clostridioides difficile. 4. Discussion The results of this study show a sixfold increase in the incidence of CDI at our cen- ter from 2009 to 2018, coinciding with a significant augment in the testing rate and in the percentage of positive stools tested after the introduction of an optimized two-step diagnostic algorithm. Microorganisms 2022, 10, 1075 9 of 12 The incidence of patients diagnosed with a first episode of CDI in our center increased significantly from 0.9 to 5.5 cases per 10,000 patient-days and it is consistent with previous studies [4,7,10]. Several factors may explain this increase in the incidence of CDI in addition to the introduction of the more sensitive two-step diagnostic protocol in 2013 [7,10,11,13]. These factors include an augment in the population at risk and a higher index of clinical suspicion. Hospitalization is another important risk factor for CDI, since highly suscep- tible patients are exposed to a spore-contaminated environment when more cases are present. The incidence of CDI achieved in 2018 (5.5 per 10,000 patient-days) is similar to the rates reported in other Mediterranean countries [7,8,10], but lower than those reported from northern Europe [4,6], the United States [1] or Canada [10,12] (7.9 to 16.2 cases per 10,000 patient-days). However, we observed a plateau in the incidence during the last two years of the study (2016–2018). The reasons for this stabilization in the more recent years fall out of the scope of the present analysis but are consistent with a recent study showing a decline in CDI in the United States from 2011 to 2018 due to a fall in health-care and hospital-acquired infections [4]. Importantly, one-third of patients with CDI diagnosed using the two-step algorithm would have been missed or diagnosed later if only EIA-toxin testing had been performed [9]. An earlier diagnosis allows for quick onset of targeted treatment, prompt correction of predisposing factors and could improve the prognosis of the disease. Moreover, considering that patients who are PCR-positive only present a risk of transmission similar to those who test toxin-positive, an earlier detection of CDI cases is important to prevent nosocomial transmission [14]. Concerns about the detection of colonized patients with a positive PCR were unlikely a problem in our study because only patients with diarrhea were tested by the microbiology laboratory after the patient’s medical history was reviewed. Therefore, all PCR-positive cases were symptomatic. Regarding patient characteristics, there was a high prevalence of known risk factors for CDI, such as older age, health-care origin and exposure to antimicrobials and PPI [2,7]. It should be noted that nearly half of the patients had some kind of immunosuppression such as solid or hematological neoplasm, HIV infection, IBD, immunosuppressive ther- apy or chemotherapy [15,16]. In general, immunocompromised patients have a higher incidence of CDI and recurrence, probably due to their own immunosuppression, higher antimicrobial use, increased exposure to healthcare settings and higher prevalence of CD colonization [11,13]. The origin of the CDI in a quarter of our series was community acquired, a finding that is comparable to other studies [17,18]. Though this incidence was still lower than that of healthcare-associated CDI, it increased over the period of study. This has been attributed in other countries to the greater awareness among physicians of the possibility of CDI with no hospital exposure, and probably also by the increased use of antibiotics and proton-pump inhibitors in the community [5]. We identified that CD toxin-positive patients have a more severe clinical presentation. This finding is in agreement with previous reports [19–22]. The most likely explanation is an increased CD bacterial load and higher toxin concentration than those toxin-negative diagnosed only by more sensitive methods such as a PCR [14]. The attending physicians should therefore be particularly aware when the patient is diagnosed with CDI by means of a positive toxin EIA in order to optimize treatment [21]. An unexpected finding was that PCR-diagnosed patients more frequently were re- ceiving immunosuppressive drugs or had an IBD, perhaps due to the higher suspicion among gastroenterologists and the common practice of proactively testing for CDI in these fields [15,16,20]. Contrary to other studies, there were no differences in mortality and recurrences between the two groups [15–19]. Regarding antibiotic treatment of the first episode of CDI, metronidazole was the treatment most frequently prescribed; although, vancomycin was preferred for severe and complicated cases in accordance with guideline recommendations [12]. Coinciding with previous studies, there was a progressive increasing use of vancomycin along the study Microorganisms 2022, 10, 1075 10 of 12 time period, probably due to the increase seen in the severity of cases [20]. Fidaxomicin was only used in a few cases due to its restricted use due to its cost. Metronidazole could still be an alternative for non-severe CDI with low risk of recurrence when the above agents are unavailable [23]. Crude 30-day all-cause mortality (14.2%) was similar to that reported in other stud- ies [19,21,22]. As previously described, mortality was significantly higher in patients admitted to ICU and in those with complicated diseases [4,24–26]. Patients with hemato- logical neoplasms also presented a higher mortality, making them a particular target for improved treatment and prevention of CDI [23]. Other factors previously reported such as older age, non-CDI antibiotic use and the presence of underlying comorbid conditions were not associated with increased mortality in multivariate analysis in our study [8,27]. In addition, as previously described, patients hospitalized with CDI had a higher probability of being discharged to an LTCF, mainly those with nosocomial and healthcare- associated infections [27]. Several risk factors for a non-home discharge include comorbidity and aging. The rate of recurrence (14.9%) was comparable to other studies, with rates ranging from 12% to 30% [18,27,28]. Consistent with other investigations, recurrence was more common in patients with previous exposure to PPI and severe presentation [1]. However, other known risk factors associated with recurrence in other studies such as advanced age, chronic renal insufficiency, elevated white blood cell counts, low serum albumin levels and continued use of systemic antimicrobials during the initial CDI episode were not significantly associated with a higher recurrence rate in our multivariate analysis [28]. Our study has several limitations. A weakness is the retrospective design, which may have introduced selection bias and could imply information loss relative to the variables studied. The increase in CDI diagnosis could be associated with other uncontrolled factors in addition to the two-step lab diagnostic algorithm, such as increased physician awareness or increased rates of immunosuppressed individuals or aged patients. 5. Conclusions The incidence of CDI at our center increased significantly over the decade comprising this study period after the introduction of a two-step laboratory diagnosis algorithm including a PCR. This occurred simultaneously with increased rates of testing and positivity in the samples. A third of the cases diagnosed with the new algorithm would have been missed using only a toxin-based screening strategy. This earlier diagnosis can have an impact on patient outcomes and hospital transmission. CD toxin-positive patients had a more severe clinical presentation, probably due to an increased CD bacterial load with higher toxin concentration. Therefore, a positive CD toxin must alert clinicians about the worse prognosis of the case. Author Contributions: Conceptualization, N.S., L.M., J.H.W.-W., G.F.-R., I.C.; Investigation, N.S., S.G., J.H.W.-W., G.F.-R., M.J.G.-Q., L.C.; Methodology, N.S., I.C., G.F.-R., L.M.; Writing—Original Draft Preparation, N.S., J.H.W.-W., G.F.-R., M.L.P.-B., I.C.; Writing—Review and Editing, N.S., J.M.L., I.C., G.F.-R., J.H.W.-W. All authors have read and agreed to the published version of the manuscript. Funding: The research received no external funding. Institutional Review Board Statement: The design of the work has been approved by the Clinical Ethics Committee of the hospital (PI-18-150). Informed Consent Statement: The need for informed consent was waived due to the retrospective nature of the study. Data Availability Statement: The data collected for this study can be shared upon a justified re- quest and following the approval of the Ethics Committee of the University Hospital Germans Trias i Pujol, Badalona, Spain. Those interested should address the request to Nieves Sopena: nsopena.germanstrias@gencat.cat. Conflicts of Interest: The authors declare no conflict of interest. Microorganisms 2022, 10, 1075 11 of 12 References 1. Evans, C.T.; Safdar, N. Current Trends in the Epidemiology and Outcomes of Clostridium Difficile Infection. Clin. Infect. Dis. 2015, 60, S66–S71. [CrossRef] [PubMed] 2. Bogaty, C.; Lévesque, S.; Garenc, C.; Frenette, C.; Bolduc, D.; Galarneau, L.-A.; Lalancette, C.; Loo, V.; Tremblay, C.; Trudeau, M.; et al. Trends in the Use of Laboratory Tests for the Diagnosis of Clostridium Difficile Infection and Association with Incidence Rates in Quebec, Canada, 2010–2014. Am. J. Infect. Control 2017, 45, 964–968. [CrossRef] [PubMed] 3. Longtin, Y.; Trottier, S.; Brochu, G.; Paquet-Bolduc, B.; Garenc, C.; Loungnarath, V.; Beaulieu, C.; Goulet, D.; Longtin, J. Impact of the Type of Diagnostic Assay on Clostridium Difficile Infection and Complication Rates in a Mandatory Reporting Program. Clin. Infect. Dis. 2013, 56, 67–73. [CrossRef] [PubMed] 4. Guh, A.Y.; Mu, Y.; Winston, L.G.; Johnston, H.; Olson, D.; Farley, M.M.; Wilson, L.E.; Holzbauer, S.M.; Phipps, E.C.; Dumyati, G.K.; et al. Trends in U.S. Burden of Clostridioides Difficile Infection and Outcomes. N. Engl. J. Med. 2020, 382, 1320–1330. [CrossRef] 5. Bauer, M.P.; Notermans, D.W.; van Benthem, B.H.B.; Brazier, J.S.; Wilcox, M.H.; Rupnik, M.; Monnet, D.L.; van Dissel, J.T.; Kuijper, E.J. Clostridium Difficile Infection in Europe: A Hospital-Based Survey. Lancet 2011, 377, 63–73. [CrossRef] 6. Davies, K.A.; Longshaw, C.M.; Davis, G.L.; Bouza, E.; Barbut, F.; Barna, Z.; Delmée, M.; Fitzpatrick, F.; Ivanova, K.; Kuijper, E.; et al. Underdiagnosis of Clostridium Difficile across Europe: The European, Multicentre, Prospective, Biannual, Point-Prevalence Study of Clostridium Difficile Infection in Hospitalised Patients with Diarrhoea (EUCLID). Lancet Infect. Dis. 2014, 14, 1208–1219. [CrossRef] 7. Roncarati, G.; Dallolio, L.; Leoni, E.; Panico, M.; Zanni, A.; Farruggia, P. Surveillance of Clostridium Difficile Infections: Results from a Six-Year Retrospective Study in Nine Hospitals of a North Italian Local Health Authority. Int. J. Environ. Res. Public Health 2017, 14, 61. [CrossRef] 8. Rodríguez-Pardo, D.; Almirante, B.; Bartolomé, R.M.; Pomar, V.; Mirelis, B.; Navarro, F.; Soriano, A.; Sorlí, L.; Martínez-Montauti, J.; Molins, M.T.; et al. Epidemiology of Clostridium Difficile Infection and Risk Factors for Unfavorable Clinical Outcomes: Results of a Hospital-Based Study in Barcelona, Spain. J. Clin. Microbiol. 2013, 51, 1465–1473. [CrossRef] 9. Crobach, M.J.T.; Planche, T.; Eckert, C.; Barbut, F.; Terveer, E.M.; Dekkers, O.M.; Wilcox, M.H.; Kuijper, E.J. European Society of Clinical Microbiology and Infectious Diseases: Update of the Diagnostic Guidance Document for Clostridium Difficile Infection. Clin. Microbiol. Infect. 2016, 22, S63–S81. [CrossRef] 10. Khanafer, N.; Oltra, L.; Hulin, M.; Dauwalder, O.; Vandenesch, F.; Vanhems, P. Clostridium Difficile Infection in a French University Hospital: Eight Years of Prospective Surveillance Study. Medicine (Baltimore) 2016, 95, e3874. [CrossRef] 11. Sopena, N.; Freixas, N.; Bella, F.; Pérez, J.; Hornero, A.; Limon, E.; Gudiol, F.; Pujol, M. Impact of a Training Program on the Surveillance of Clostridioides Difficile Infection. Epidemiol. Infect. 2019, 147, e231. [CrossRef] 12. Thabit, A.K.; Alsolami, M.H.; Baghlaf, N.A.; Alsharekh, R.M.; Almazmumi, H.A.; Alselami, A.S.; Alsubhi, F.A. Comparison of Three Current Clostridioides Difficile Infection Guidelines: IDSA/SHEA, ESCMID, and ACG Guidelines. Infection 2019, 47, 899–909. [CrossRef] 13. Fong, K.S.; Fatica, C.; Hall, G.; Procop, G.; Schindler, S.; Gordon, S.M.; Fraser, T.G. Impact of PCR Testing for Clostridium Difficile on Incident Rates and Potential on Public Reporting: Is the Playing Field Level? Infect. Control Hosp. Epidemiol. 2011, 32, 932–933. [CrossRef] 14. Guerrero, D.M.; Chou, C.; Jury, L.A.; Nerandzic, M.M.; Cadnum, J.C.; Donskey, C.J. Clinical and Infection Control Implications of Clostridium Difficile Infection with Negative Enzyme Immunoassay for Toxin. Clin. Infect. Dis. 2011, 53, 287–290. [CrossRef] 15. Maharshak, N.; Barzilay, I.; Zinger, H.; Hod, K.; Dotan, I. Clostridium Difficile Infection in Hospitalized Patients with Inflammatory Bowel Disease. Medicine 2018, 97, e9772. [CrossRef] 16. Revolinski, S.L.; Silvia Munoz-Price, L. Clostridium Difficile in Immunocompromised Hosts: A Review of Epidemiology, Risk Factors, Treatment, and Prevention HEALTHCARE EPIDEMIOLOGY: Robert Weinstein, Section Editor. Healthc. Epidemiol. Clin. Infect. Dis.® 2019, 2019, 2144–2153. [CrossRef] 17. Zanichelli, V.; Garenc, C.; Villeneuve, J.; Moisan, D.; Frenette, C.; Loo, V.; Longtin, Y. Increased Community-Associated Clostridioides Difficile Infections in Quebec, Canada, 2008–2015. Emerg. Infect. Dis. 2020, 26, 1291–1294. [CrossRef] 18. Larrainzar-Coghen, T.; Rodríguez-Pardo, D.; Fernández-Hidalgo, N.; Puig-Asensio, M.; Pigrau, C.; Ferrer, C.; Rodríguez, V.; Bartolomé, R.; Campany, D.; Almirante, B. Secular Trends in the Epidemiology of Clostridium Difficile Infection (CDI) at a Tertiary Care Hospital in Barcelona, 2006–2015: A Prospective Observational Study. Anaerobe 2018, 51, 54–60. [CrossRef] 19. Beaulieu, C.; Dionne, L.L.; Julien, A.S.; Longtin, Y. Clinical Characteristics and Outcome of Patients with Clostridium Difficile Infection Diagnosed by PCR versus a Three-Step Algorithm. Clin. Microbiol. Infect. 2014, 20, 1067–1073. [CrossRef] 20. Avni, T.; Babich, T.; Ben-Zvi, H.; Atamna, A.; Yahav, D.; Shepshelovich, D.; Leibovici-Weissman, Y.; Bishara, J. Molecular-Based Diagnosis of Clostridium Difficile Infection Is Associated with Reduced Mortality. Eur. J. Clin. Microbiol. Infect. Dis. 2018, 37, 1137–1142. [CrossRef] 21. Origüen, J.; Corbella, L.; Orellana, M.; Fernández-Ruiz, M.; López-Medrano, F.; San Juan, R.; Lizasoain, M.; Ruiz-Merlo, T.; Morales-Cartagena, A.; Maestro, G.; et al. Comparison of the Clinical Course of Clostridium Difficile Infection in Glutamate Dehydrogenase-Positive Toxin-Negative Patients Diagnosed by PCR to Those with a Positive Toxin Test. Clin. Microbiol. Infect. 2018, 24, 414–421. [CrossRef] [PubMed] Microorganisms 2022, 10, 1075 12 of 12 22. Guh, A.Y.; Hatfield, K.M.; Winston, L.G.; Martin, B.; Johnston, H.; Brousseau, G.; Farley, M.M.; Wilson, L.; Perlmutter, R.; Phipps, E.C.; et al. Toxin Enzyme Immunoassays Detect Clostridioides Difficile Infection with Greater Severity and Higher Recurrence Rates. Clin. Infect. Dis. 2019, 69, 1667–1674. [CrossRef] [PubMed] 23. Gentry, C.A.; Campbell, D.L.; Williams, R.J. Outcomes Associated with Recent Guideline Recommendations Removing Metron- idazole for Treatment of Non-Severe Clostridioides Difficile Infection: A Retrospective, Observational, Nationwide Cohort Study. Int. J. Antimicrob. Agents 2021, 57, 106282. [CrossRef] [PubMed] 24. Barbut, F.; Day, N.; Bouée, S.; Youssouf, A.; Grandvoinnet, L.; Lalande, V.; Couturier, J.; Eckert, C. Toxigenic Clostridium Difficile Carriage in General Practice: Results of a Laboratory-Based Cohort Study. Clin. Microbiol. Infect. 2019, 25, 588–594. [CrossRef] 25. Reveles, K.R.; Dotson, K.M.; Gonzales-Luna, A.; Surati, D.; Endres, B.T.; Alam, M.J.; Garey, K.W. Clostridioides (Formerly Clostridium) Difficile Infection during Hospitalization Increases the Likelihood of Nonhome Patient Discharge. Clin. Infect. Dis. 2019, 68, 1887–1893. [CrossRef] 26. Na’amnih, W.; Adler, A.; Miller-Roll, T.; Cohen, D.; Carmeli, Y. Risk Factors for Recurrent Clostridium Difficile Infection in a Tertiary Hospital in Israel. Eur. J. Clin. Microbiol. Infect. Dis. 2018, 37, 1281–1288. [CrossRef] 27. Appaneal, H.J.; Caffrey, A.R.; Beganovic, M.; Avramovic, S.; LaPlante, K.L. Predictors of Mortality among a National Cohort of Veterans with Recurrent Clostridium Difficile Infection. Open Forum Infect. Dis. 2018, 5, ofy175. [CrossRef] 28. Mora Pinzon, M.C.; Buie, R.; Liou, J.I.; Shirley, D.K.; Evans, C.T.; Ramanathan, S.; Poggensee, L.; Safdar, N. Outcomes of Community and Healthcare-Onset Clostridium Difficile Infections. Clin. Infect. Dis. 2019, 68, 1343–1350. [CrossRef]