Hostname: page-component-78c5997874-lj6df Total loading time: 0 Render date: 2024-11-09T08:13:27.783Z Has data issue: false hasContentIssue false

Prevalence and trends of Clostridioides difficile infection among persons requiring maintenance hemodialysis: A systematic review and meta-analysis

Published online by Cambridge University Press:  23 September 2022

Yousef M. Elfanagely
Affiliation:
Department of Internal Medicine, Warren Alpert School of Medicine, Brown University, Providence, Rhode Island
Joshua Ray Tanzer
Affiliation:
Department of Biostatistics, Warren Alpert School of Medicine, Brown University, Providence, Rhode Island
Anuoluwapo Shobayo
Affiliation:
Division of Infectious Diseases, Warren Alpert School of Medicine, Brown University, Providence, Rhode Island
Mouhand F.H. Mohamed
Affiliation:
Department of Internal Medicine, Warren Alpert School of Medicine, Brown University, Providence, Rhode Island
Jonathan J.C. Ho
Affiliation:
Department of Internal Medicine, Warren Alpert School of Medicine, Brown University, Providence, Rhode Island
Douglas Shemin
Affiliation:
Division of Nephrology, Warren Alpert School of Medicine, Brown University, Providence, Rhode Island
Laura Pavlech
Affiliation:
Brown University Library, Providence, Rhode Island
Erika M.C. D’Agata*
Affiliation:
Department of Biostatistics, Warren Alpert School of Medicine, Brown University, Providence, Rhode Island
*
Author for correspondence: Dr. Erika M.C. D’Agata, E-mail: [email protected]
Rights & Permissions [Opens in a new window]

Abstract

Objective:

Clostridioides difficile infection (CDI) is among the most common cause of healthcare-associated infections. Persons requiring maintenance hemodialysis (MHD) are at increased risk of CDI and associated mortality compared to persons not requiring MHD. Given the clinical impact of CDI among persons requiring MHD, we aimed to quantify the burden of CDI and trends over time in this patient population.

Study design:

A systematic review and meta-analysis of studies reporting rates of CDI among persons requiring MHD in MEDLINE, Embase, Web of Science Core Collection, CINAHL, and Cochrane Central Register of Controlled Trials were performed. Searches were conducted on May 17, 2021, and March 4, 2022.

Results:

In total, 2,408 titles and abstracts were identified; 240 underwent full text review. Among them, 15 studies provided data on rates of CDI among persons requiring MHD, and 8 of these also provided rates among persons not requiring MHD. The pooled prevalence of CDI among persons requiring MHD was 19.14%, compared to 5.16% among persons not requiring MHD (odds ratio [OR], 4.35; 95% confidence interval [CI], 2.07–9.16; P = .47). The linear increase in CDI over time was significant, increasing an average of 31.97% annually between 1993 and 2017 (OR, 1.32; 95% CI, 1.1–1.58; P < .01). The linear annual increase was similar among persons requiring and not requiring MHD (OR, 1.28; 95% CI, 1.13–1.45; P = .11).

Conclusions:

Persons requiring MHD have a 4-fold higher risk of CDI compared to persons not requiring MHD, and rates of CDI are increasing over time in both groups.

Type
Original Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited
Copyright
© The Author(s), 2022. Published by Cambridge University Press on behalf of The Society for Healthcare Epidemiology of America

Clostridioides difficile infections (CDIs) are associated with considerable morbidity and mortality. In 2017, there were almost half a million CDIs and >20,000 associated deaths in the United States. Reference Guh, Mu and Winston1Reference Dubberke and Olsen3 CDI is the most common cause of gastroenteritis-related deaths and is among the most common cause of healthcare-associated infections. Reference Dubberke and Olsen3,Reference Miller, Chen, Sexton and Anderson4 Rates of CDI have been increasing both in the hospital and community settings. Reference Chitnis, Holzbauer and Belflower5

Persons with chronic kidney disease and those on maintenance hemodialysis (MHD) are at high-risk of CDI due to substantial antimicrobial exposure, frequent hospitalizations, and immune dysfunction. Reference Ramesh and Yee6 Rates of CDI and associated mortality are up to 2-fold higher among persons with chronic kidney disease compared to the general population, and the requirement for MHD increases these rates even further. Reference Keddis, Khanna, Noheria, Baddour, Pardi and Qian7Reference Thongprayoon, Cheungpasitporn, Phatharacharukul, Mahaparn and Bruminhent10

Given the clinical significance of CDI among persons requiring MHD and increasing CDI rates in healthcare settings, we evaluated the burden of CDI, potential risk factors for CDI, mortality rates, and mortality-associated risk factors among persons requiring MHD. We performed a systematic review of the literature that reported CDI prevalence among persons requiring MHD and a meta-analysis to quantify it. We compared the prevalence of CDI among persons requiring MHD with that of persons not requiring MHD. Trends of CDI over time among both groups were also evaluated.

Materials and methods

This systematic review and meta-analysis were performed according to the PRISMA guidelines. Reference Hutton, Salanti and Caldwell11 The study did not require institutional review board nor ethics committee approval because all data were publicly available.

Search strategy

A medical librarian constructed comprehensive search strategies for each of the following databases: MEDLINE, Embase, Web of Science Core Collection, CINAHL, and Cochrane Central Register of Controlled Trials. The search strategies used a combination of controlled vocabulary terms and keywords to describe 2 concepts: Clostridioides difficile and hemodialysis or chronic or end stage kidney disease (Supplementary Material online). All databases were searched on May 17, 2021, and March 4, 2022, with the exception of Web of Science Core Collection, which was only searched on May 17, 2021, because access to this database was no longer available on March 4, 2022. The reference lists of and citations to key articles were reviewed to identify additional studies.

Results were exported to EndNote for Windows version X9.3.3 software (Clarivate Analytics, Philadelphia, PA), and duplicates were removed using a previously described method. Reference Bramer, Giustini, de Jonge, Holland and Bekhuis12 The deduplicated results were uploaded to Covidence (Veritas Health Innovations, Melbourne, Australia) for screening.

Selection criteria

Inclusion criteria. The following inclusion criteria were applied:

  1. (1) Studies published in English

  2. (2) Studies with a patient population that included patients with stage 5 chronic kidney disease, with end-stage renal disease, or on maintenance hemodialysis

  3. (3) Studies that reported symptomatic Clostridioides difficile infection as an outcome

  4. (4) Studies providing data on CDI rates among all persons on MHD

  5. (5) Randomized or nonrandomized controlled trials, cohort studies, and case–control studies.

Exclusion criteria. We applied the following exclusion criteria:

  1. (1) Meta-analyses, letters, case reports, commentaries, conference abstracts, and CDI outbreaks

  2. (2) Studies with duplicate results or those that continued work from previous publications

  3. (3) Studies with pediatric populations, peritoneal dialysis, or nursing home populations

  4. (4) Studies that addressed C. difficile colonization or used a presumptive diagnosis of CDI based on clinical suspicion or diarrhea, and not laboratory confirmation

  5. (5) Studies in which differentiation between acute hemodialysis, chronic kidney disease, and maintenance hemodialysis could not be confirmed

  6. (6) Studies that reported rates among persons requiring and not requiring MHD among persons with CDI, in contrast to rates of CDI among persons requiring and not requiring MHD.

Data extraction

Four researchers (Y.E., A.S., M.M., and J.H.) independently screened the literature and cross-checked the articles. Senior researchers (E.M.C.D. and J.R.T.) resolved discrepancies. After selecting which studies would be included, data were extracted, including first author’s name, year of publication, study population and location, study design, and number of persons requiring and not requiring MHD with and without CDI.

Quality assessment

Two reviewers (Y.E. and A.S.) appraised the quality of the studies, and these appraisals were confirmed by a third reviewer (E.M.C.D.). The Newcastle-Ottawa Scale, a star-based rating system of 9 domains, was used to assess the methodological quality of the included studies. Reference Wells, Shea and O’Connell13 The 2 comparability fields were not relevant to this analysis and were not included; thus, each study could receive a maximum of 7 stars. To consider the possibility of publication bias, funnel plots were visually examined and the Egger test was performed. Reference Rothstein, Sutton and Borenstein14

Data synthesis and statistical analyses

Effect sizes. For the primary research question, we compared the risks of CDI between persons on MHD and persons not requiring MHD. This focused, 2-group comparison was selected so that persons not requiring MHD within the study samples could act as a general control. According to the study inclusion criteria, all studies reported rates of CDI among persons on MHD; however, some studies exclusively described persons on MHD. For this reason, the effect size of interest was the log odds risk of infection, defined as ${\rm{log}}\left( {{{{{\hat p}_{{\rm{infect}}}}} \over {1 - {{\hat p}_{{\rm{infect}}}}}}} \right)$ where ${\hat p_{{\rm{infect}}}}$ is the estimated probability of CDI. This effect size was selected because estimated risks could be directly compared between persons on MHD and controls without having to combine estimates into a single metric (eg, risk ratio or risk difference), which would induce a missing-data problem based on how sampling was performed. Additionally, the log odds can be better approximated as normally distributed, making it more conductive to statistical modeling. Reference Norton and Dowd15

In each study, the log odds of infection was estimated from the reported summary data for persons on MHD and all other patients if data were available. Variances of estimates were approximated using the Monte Carlo simulation. Reference Rubinstein and Kroese16 For each log odds estimate, 1,000 random samples of the same size and estimated probability ${\hat p_{{\rm{infect}}}}$ were generated from a binomial distribution and were transformed into the log odds. The variance among the simulated log odds values was used as the variance of the estimate for the final meta-analysis model.

Analysis plan. A random-effects model was used to pool the effect sizes. This model was selected because many of the studies were conducted in different locations with different sample characteristics, so the assumed homogeneity of a fixed-effects model did not seem appropriate. Additionally, because many of the studies provided both estimates of risk of CDI for both persons on MHD and persons not requiring MHD, random-effects modeling is well equipped analytically to address heterogeneity of variances and correlations between different observations nested within studies. Reference Van Houwelingen, Arends and Stijnen17

The fixed effects in the model included year as a continuous value and MHD status as a binary indicator. For studies that were conducted for ≥1 year, the midpoint was used. We hypothesized that persons on MHD would be at greater risk of CDI and that risks of CDI would increase year after year. Trajectories over time between persons on MHD and controls were specifically tested within the model.

Results

Study identification and selection

The details of the selection process are summarized in the flow diagram (Fig. 1). The initial electronic search yielded 2,208 studies. After the removal of duplicates, 1,811 studies were screened by title and abstract. In total, 1,571 studies were excluded, and 240 studies were assessed based on full texts. Subsequently, 225 studies were excluded, a large proportion of which were excluded due to the inability to determine whether hemodialysis referred to acute or maintenance dialysis and/or to the number of persons on MHD among those with chronic kidney disease. Also, 15 studies were included in the quantitative analysis of trends in CDI rates over time among persons on MHD. Reference Guh, Mu and Winston1,Reference Keddis, Khanna, Noheria, Baddour, Pardi and Qian7,Reference Tirath, Tadros and Coffin8,Reference D’Agata, Mount, Thayer and Schaffner18Reference Wei, Wei and Huang29 Furthermore, 8 of these studies also provided CDI prevalence data among persons not requiring MHD and were included in the meta-analysis. Reference Keddis, Khanna, Noheria, Baddour, Pardi and Qian7,Reference D’Agata, Mount, Thayer and Schaffner18,Reference Demir, Cheng and Lee19,Reference Eui Oh, Lee and Lee22,Reference Huang, Wu, Chen, Xu, Fang, Weintraub and Nord23,Reference Morfin-Otero, Garza-Gonzalez and Garcia Garcia25,Reference Predrag27,Reference Wei, Wei and Huang29

Fig. 1. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow chart of the study search and review process examining CDI among persons on maintenance hemodialysis. Note. CDI, Clostridioides difficile infection; MHD, maintenance hemodialysis.

Study characteristics

Characteristics of the 15 studies are summarized in Table 1, including country, study design, and patient population. Of the 15 studies included in the trend analysis of CDI rates, 13 were conducted among hospitalized patients. Reference Keddis, Khanna, Noheria, Baddour, Pardi and Qian7,Reference Tirath, Tadros and Coffin8,Reference D’Agata, Mount, Thayer and Schaffner18Reference Predrag27,Reference Wei, Wei and Huang29 In 2 studies, the proportion of patients in whom CDI was diagnosed either in the outpatient or hospital setting could not be determined. Reference Guh, Mu and Winston1,Reference Sheth, Bernardini and Burr28 All 8 studies included in the meta-analysis were conducted in a hospital setting.

Table 1. Characteristics of Studies Included in the Systematic Review and Meta-analysis

Note. CDI, Clostridioides difficile infection; MHD, maintenance hemodialysis; CKD, chronic kidney disease; N/A, not available.

Risk of bias and quality assessment

The reviewers were in complete agreement that all studies were suitable for use in the systematic review. All studies included in the meta-analysis were deemed of high quality, with Newcastle Ottawa Scale scores of 7 of 7 stars for 8 studies and 6 of 7 stars for 2 studies (Supplementary Tables 1 and 2 online).

To assess publication bias, heterogeneity, and chance, a funnel plot was constructed to compare the risks of CDI to the precision of estimate (Supplementary Fig. 1 online). Most participants not on MHD demonstrated minimal change in risk regardless of precision; however, the risks also tended to be larger, making them more robust to chance findings. Among persons on MHD, we detected a visual trend such that more extreme risks tended to be observed in less precise samples. The Egger test did not indicate bias within controls (Z = 0.57; P = .5700); however, the Egger results were much closer to significance among persons on MHD (Z = 1.58; P = .1137). This trend was likely attributable to the high variation in estimate precision and modest sample size. After model estimation was complete, the more extreme observations and studies with disproportionately large sample sizes were removed from the analysis to assess the influence of these chance observations. This removal did not result in any changes to inference (data not shown).

Systematic review of CDI rates, risk factors and outcomes

Table 1 provides the percentage of CDI among persons requiring MHD and not requiring MHD. Among persons requiring MHD, CDI ranged from 0.05% to 77.8%. In a matched case–control study of 452 subjects, all 5 persons requiring MHD developed CDI. Reference Guh, Mu and Winston1 Among the 10 studies that compared CDI rates among persons on MHD to those not requiring MHD, 9 identified requirement for MHD as an independent risk factor for CDI, with adjusted odds ratios ranging from 1.33 (95% confidence interval [CI], 1.32–1.35) to 13.5 (95% CI, 2.85–63.8). Reference Keddis, Khanna, Noheria, Baddour, Pardi and Qian7,Reference Demir, Cheng and Lee19,Reference Eddi, Malik, Shakov, Baddoura, Chandran and Debari21Reference Morfin-Otero, Garza-Gonzalez and Garcia Garcia25,Reference Predrag27,Reference Wei, Wei and Huang29

Two studies analyzed risk factors for CDI among persons requiring MHD. Reference Tirath, Tadros and Coffin8,Reference Sheth, Bernardini and Burr28 In 2010, Sheth et al Reference Sheth, Bernardini and Burr28 identified a serum albumin ≤3 g/dL (adjusted hazard ratio [aHR], 3.03; 95% CI, 1.75–5.55) and a higher Charlson comorbidity index (aHR, 1.17; 95% CI, 1.00–1.36) as significantly associated with a higher risk of CDI. Reference Sheth, Bernardini and Burr28 In 2017, Tirath et al Reference Tirath, Tadros and Coffin8 identified 3 comorbidities associated with the greatest risk for CDI: age ≥65 years (adjusted risk ratio [aRR], 1.76; 95% CI, 1.7–1.82), human immunodeficiency virus (aRR, 2.68; 95% CI, 2.4–2.99), and bacteremia (aRR, 1.74; 95% CI, 1.68–1.80). Reference Tirath, Tadros and Coffin8 In this study, Hispanic ethnicity (aRR, 0.76; 95% CI, 0.72–0.70) and Black race (aRR, 0.75; 95% CI, 0.73–0.78) were associated with a decreased risk of CDI. Reference Tirath, Tadros and Coffin8

Three studies compared mortality among persons on MHD with and without CDI, which ranged from 13.2% to 68.8%. Reference Tirath, Tadros and Coffin8,Reference Pant, Deshpande, Anderson and Sferra26,Reference Sheth, Bernardini and Burr28 In 2 studies, mortality was significantly higher among persons requiring MHD with a 2-fold higher risk of death. Reference Tirath, Tadros and Coffin8,Reference Kim, Seo and Lee24 In the study by Tirath et al, Reference Tirath, Tadros and Coffin8 independent factors associated with mortality included age ≥65 years (aHR, 2.28) and presence of cirrhosis (aHR, 1.76); however, confidence intervals were not provided. Reference Tirath, Tadros and Coffin8

In 2012, Pant et al Reference Pant, Deshpande, Anderson and Sferra26 reported other outcomes associated with CDI among persons requiring MHD. Length of hospital stay was greater among persons on MHD with CDI (mean difference, 9.4 days; 95% CI, 9.2–9.5) and hospital costs were higher (mean difference, $62,824; 95% CI, 61,615–64,033). Reference Pant, Deshpande, Anderson and Sferra26

In contrast to the studies that compared rates among persons not requiring MHD, Keddis et al Reference Keddis, Khanna, Noheria, Baddour, Pardi and Qian7 compared mortality rates and other outcomes associated with CDI between persons requiring MHD and persons with chronic kidney disease (CKD) not requiring MHD. They detected no differences in length of hospital stays between the 2 groups, and lower rates of colectomy were detected among persons with CDI requiring MHD (adjusted odds ratio [aOR], 0.327; 95% CI, 0.26–0.40).

Meta-analysis and trends in CDI over time

Figure 2 shows the forest plot reporting individual study prevalence and odds ratios among the 8 studies providing estimates for both persons requiring and not requiring MHD. Reference Keddis, Khanna, Noheria, Baddour, Pardi and Qian7,Reference D’Agata, Mount, Thayer and Schaffner18,Reference Demir, Cheng and Lee19,Reference Eui Oh, Lee and Lee22,Reference Huang, Wu, Chen, Xu, Fang, Weintraub and Nord23,Reference Morfin-Otero, Garza-Gonzalez and Garcia Garcia25,Reference Predrag27,Reference Wei, Wei and Huang29 The pooled prevalence of CDI among persons requiring MHD was 19.14%, and for persons not requiring MHD, the pooled prevalence of CDI was 5.16%. Overall, persons requiring MHD had a 4-fold increased risk of CDI compared to persons not requiring MHD (OR, 4.35; 95% CI, 2.07–9.16; P = .47). The exclusion of the study by Keddis et al, Reference Keddis, Khanna, Noheria, Baddour, Pardi and Qian7 which included a large patient population that may have influenced the study results, had a minimal impact on the results of the meta-analysis (data not shown).

Fig. 2. Forest plot of the 10 included studies providing Clostridioides difficile prevalence among persons requiring and not requiring maintenance hemodialysis.

The trend analysis included 15 studies providing data on CDI prevalence among persons requiring MHD from 1993 to 2017 Reference Guh, Mu and Winston1,Reference Keddis, Khanna, Noheria, Baddour, Pardi and Qian7,Reference Tirath, Tadros and Coffin8,Reference D’Agata, Mount, Thayer and Schaffner18Reference Wei, Wei and Huang29 and 8 studies Reference Keddis, Khanna, Noheria, Baddour, Pardi and Qian7,Reference D’Agata, Mount, Thayer and Schaffner18,Reference Demir, Cheng and Lee19,Reference Eui Oh, Lee and Lee22,Reference Huang, Wu, Chen, Xu, Fang, Weintraub and Nord23,Reference Morfin-Otero, Garza-Gonzalez and Garcia Garcia25,Reference Predrag27,Reference Wei, Wei and Huang29 providing data on CDI prevalence among persons not requiring MHD from 1995 to 2016 (Fig. 3). For both groups, the linear increase in CDI risk over time was significant, increasing 31.97% on average annually during the study period (OR, 1.32; 95% CI, 1.1–1.58; P < .01). The linear annual increase in risk among persons requiring MHD was similar to that of persons not requiring MHD (OR, 1.28; 95% CI, 1.13–1.45; P = 0.11).

Fig. 3. Trends in Clostridioides difficile infection over time comparing persons requiring and not requiring maintenance hemodialysis. The line and shaded regions represent the modeled average and 95% confidence intervals.

Discussion

We performed a systematic review and meta-analysis to determine the rates and trends over time of CDI among persons requiring MHD compared to persons not requiring MHD and to provide a qualitative analysis of risk factors and mortality associated with CDI among persons on MHD.

Rates of CDI were 4-fold higher among persons requiring MHD compared to persons not requiring MHD, with pooled prevalences of 19.14% and 5.16%, respectively. The trend analysis, from 1993 to 2017, demonstrated that there was a significant increase in rates of CDI over time in both the groups of persons who did and did not require MHD, increasing by an average of 31.97%, annually. The rise in trajectories was parallel between the 2 groups. Notably, CDI rates among persons not requiring MHD were very high in some studies, but this finding reflects the fact that studies with this population included persons with diarrhea or chronic kidney disease who were at higher risk of CDI.

Antimicrobial exposure and hospitalizations are among the main risk factors for an increased risk of CDI among all patients. Reference Lessa, Mu and Bamberg2 In this systematic review, risk factors specific to the MHD population included age ≥65 years, serum albumin ≤3 g/dL, higher Charlson comorbidity index, human immunodeficiency virus (HIV), and bloodstream infections. Reference Tirath, Tadros and Coffin8,Reference Sheth, Bernardini and Burr28 Mortality associated with CDI was 2-fold higher among persons requiring MHD compared to persons not requiring MHD, and ranged from 13.2% to 68.8%. Reference Tirath, Tadros and Coffin8,Reference Pant, Deshpande, Anderson and Sferra26,Reference Sheth, Bernardini and Burr28 Risk factors for CDI-associated mortality included age ≥65 years and presence of cirrhosis. Reference Tirath, Tadros and Coffin8 Lastly, a diagnosis of CDI increased the length of hospital stay by 9 days, with increased hospital costs among persons requiring MHD. Reference Pant, Deshpande, Anderson and Sferra26

The substantially higher and rising rates of CDI among persons requiring MHD and associated higher morbidity and mortality compared to persons not requiring MHD both emphasize the importance of preventing C. difficile spread and CDI in this patient population.

The Nephrologists Transforming Dialysis Safety work group, an initiative funded by the Centers for Disease Control and Prevention in collaboration with the American Society of Nephrology, recently published recommendations for the prevention of C. difficile spread in outpatient dialysis facilities. Reference D’Agata, Apata and Booth30 The strategies outlined parallel those for multidrug-resistant organisms; the transmission dynamics of C. difficile are very similar. One major difference, however, is that C. difficile produces spores that survive on inanimate surfaces for many months. Reference Fawley, Underwood and Freeman31,Reference Kim, Fekety and Batts32 These spores are resistant to a variety of routinely used environmental disinfectants and require specific C. difficile sporicidal agents registered by the Environmental Protective Agency. 33 Hand hygiene measures also differ. Soap and water substantially reduces C. difficile spores, in contrast to alcohol-based sanitizers. Reference Fawley, Underwood and Freeman31,Reference Kim, Fekety and Batts32,Reference Edmonds, Zapka and Kasper34 Thus, soap and water is the preferred hand hygiene measure to prevent C. difficile spread in outpatient dialysis units. Reference D’Agata, Apata and Booth30 However, given the greater compliance with alcohol-based sanitizers, this hand hygiene measure is an alternative, except when there is visible hand soiling, during CDI outbreaks, or when there is concern for spread of CDI within the dialysis facility. Reference D’Agata, Apata and Booth30

Compliance with the Nephrologists Transforming Dialysis Safety strategies aimed at preventing the spread of C. difficile should be monitored in outpatient dialysis facilities to prevent de novo acquisition and C. difficile outbreaks. Several outbreaks in outpatient dialysis facilities have been reported; they have affected a substantial number of patients within the dialysis unit in addition to dialysis healthcare workers. Reference Bagchi and Booth35,Reference Lachowicz, Szulencka, Obuch-Woszczatyński, van Belkum and Pituch36 Implementation of C. difficile–specific infection prevention strategies has led to the end of CDI outbreaks. Reference Bagchi and Booth35

Many of the risk factors for CDI are modifiable, especially antimicrobial exposure. Up to 30% of antimicrobial doses administered in the outpatient dialysis facility are not indicated according to national guidelines. Reference Snyder, Patel, Kallen, Strom, Tucker and D’Agata37 Implementation of antimicrobial stewardship programs in dialysis facilities has been shown to significantly reduce antimicrobial prescribing. These programs have yielded substantial decreases in infections and mortality caused by C. difficile and multidrug-resistant organisms without negative outcomes such as increased hospitalizations or bloodstream infections. Reference D’Agata, Lindberg and Lindberg38,Reference D’Agata, Tran, Bautista, Shemin and Grima39

This review had several limitations. First, most studies were based in the hospital setting and did not distinguish between acquisition of CDI during a hospitalization versus the presence of CDI at hospital admission. Thus, we were unable to quantify the prevalence of CDI in the hospital versus the dialysis unit setting. Second, hospitals or outpatient dialysis settings with high CDI prevalence could be more likely to publish studies of CDI rates, which would introduce a selection bias toward higher rates. Third, only 3 studies meeting the inclusion and exclusion criteria reported rates of mortality associated with CDI among persons requiring MHD; therefore, a meta-analysis could not be performed. Lastly, changes in diagnostic testing for CDI over the study period, such as nucleic acid amplification tests, may have identified colonization in the presence of diarrhea from other causes, leading to higher reported rates of CDI in studies published after 2005.

In summary, C. difficile causes a substantial burden on persons requiring MHD, with higher mortality rates compared to persons not requiring MHD. Adherence to C. difficile–specific infection prevention recommendations and improving antimicrobial prescribing patterns are some of the important strategies to limit C. difficile spread in the population of patients requiring MHD.

Supplementary material

To view supplementary material for this article, please visit https://doi.org/10.1017/ice.2022.217

Acknowledgments

Financial support

This work was supported by the National Institute of Allergy and Infectious Diseases (grant no. K24 AI119158 to E.M.C.D.) and the Agency of Healthcare Quality and Research (grant no. R01HS026895 to E.M.C.D. and D.S.). The funding sources for this study played no role in the design or conduct of the study; collection, management, analysis, or interpretation of the data; or preparation, review, or approval of the manuscript.

Conflicts of interest

All authors report no conflicts of interest relevant to this article.

References

Guh, AY, Mu, Y, Winston, LG, et al. Trends in US burden of Clostridioides difficile infection and outcomes. N Engl J Med 2020;382:13201330.CrossRefGoogle ScholarPubMed
Lessa, FC, Mu, Y, Bamberg, WM, et al. Burden of Clostridium difficile infection in the United States. N Engl J Med 2015;372:825834.CrossRefGoogle ScholarPubMed
Dubberke, ER, Olsen, MA. Burden of Clostridium difficile on the healthcare system. Clin Infect Dis 2012;55 suppl 2:S88S92.CrossRefGoogle Scholar
Miller, BA, Chen, LF, Sexton, DJ, Anderson, DJ. Comparison of the burdens of hospital-onset, healthcare facility–associated Clostridium difficile infection and of healthcare-associated infection due to methicillin-resistant Staphylococcus aureus in community hospitals. Infect Control Hosp Epidemiol 2011;32:387390.CrossRefGoogle ScholarPubMed
Chitnis, AS, Holzbauer, SM, Belflower, RM, et al. Epidemiology of community-associated Clostridium difficile infection, 2009 through 2011. JAMA Intern Med 2013;173:13591367.CrossRefGoogle ScholarPubMed
Ramesh, MS, Yee, J. Clostridioides difficile infection in chronic kidney disease/end-stage renal disease. Adv Chronic Kidney Dis 2019;26:3034.CrossRefGoogle ScholarPubMed
Keddis, MT, Khanna, S, Noheria, A, Baddour, LM, Pardi, DS, Qian, Q. Clostridium difficile infection in patients with chronic kidney disease. Mayo Clin Proc 2012;87:10461053.CrossRefGoogle ScholarPubMed
Tirath, A, Tadros, S, Coffin, SL, et al. Clostridium difficile infection in dialysis patients. J Investig Med 2017;65:353357.CrossRefGoogle ScholarPubMed
Phatharacharukul, P, Thongprayoon, C, Cheungpasitporn, W, Edmonds, PJ, Mahaparn, P, Bruminhent, J. The risks of incident and recurrent Clostridium difficile–associated diarrhea in chronic kidney disease and end-stage kidney disease patients: a systematic review and meta-analysis. Dig Dis Sci 2015;60:29132922.CrossRefGoogle ScholarPubMed
Thongprayoon, C, Cheungpasitporn, W, Phatharacharukul, P, Mahaparn, P, Bruminhent, J. High mortality risk in chronic kidney disease and end stage kidney disease patients with Clostridium difficile infection: a systematic review and meta-analysis. J Nat Sci 2015;1:e85.Google ScholarPubMed
Hutton, B, Salanti, G, Caldwell, DM, et al. The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of healthcare interventions: checklist and explanations. Ann Intern Med 2015;162:777784.CrossRefGoogle Scholar
Bramer, WM, Giustini, D, de Jonge, GB, Holland, L, Bekhuis, T. Deduplication of database search results for systematic reviews in EndNote. J Med Libr Assoc 2016;104:240243.CrossRefGoogle ScholarPubMed
Wells, G, Shea, B, O’Connell, D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. The Ottawa Hospital Research Institute website. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Published 2009. Accessed March 13, 2022.Google Scholar
Rothstein, HR, Sutton, AJ, Borenstein, M. Publication Bias in Meta-analysis: Prevention, Assessment and Adjustments. Hoboken, NJ: John Wiley & Sons; 2006.Google Scholar
Norton, EC, Dowd, BE. Log odds and the interpretation of logit models. Health Serv Res 2018;53:859878.CrossRefGoogle ScholarPubMed
Rubinstein, RY, Kroese, DP. Simulation and the Monte Carlo method, Vol. 10. Hoboken, NJ: John Wiley & Sons; 2016.CrossRefGoogle Scholar
Van Houwelingen, HC, Arends, LR, Stijnen, T. Advanced methods in meta-analysis: multivariate approach and meta-regression. Stat Med 2002;21:589624.CrossRefGoogle ScholarPubMed
D’Agata, EM, Mount, DB, Thayer, V, Schaffner, W. Hospital-acquired infections among chronic hemodialysis patients. Am J Kidney Dis 2000;35:1083–188.CrossRefGoogle ScholarPubMed
Demir, KK, Cheng, MP, Lee, TC. Predictive factors of Clostridioides difficile infection in hospitalized patients with new diarrhea: a retrospective cohort study. PLoS One 2018;13:e0207128.CrossRefGoogle ScholarPubMed
Do, AN, Fridkin, SK, Yechouron, A, et al. Risk factors for early recurrent Clostridium difficile–associated diarrhea. Clin Infect Dis 1998;26:954959.CrossRefGoogle ScholarPubMed
Eddi, R, Malik, MN, Shakov, R, Baddoura, WJ, Chandran, C, Debari, VA. Chronic kidney disease as a risk factor for Clostridium difficile infection. Nephrology 2010;15:471475.CrossRefGoogle ScholarPubMed
Eui Oh, S, Lee, SM, Lee, YK, et al. Clostridium difficile–associated diarrhea in dialysis patients. Kidney Res Clin Pract 2013;32:2731.CrossRefGoogle ScholarPubMed
Huang, H, Wu, S, Chen, R, Xu, S, Fang, H, Weintraub, A, Nord, CE. Risk factors of Clostridium difficile infections among patients in a university hospital in Shanghai, China. Anaerobe 2014;30:6569.CrossRefGoogle Scholar
Kim, SC, Seo, MY, Lee, JY, et al. Advanced chronic kidney disease: a strong risk factor for Clostridium difficile infection. Korean J Intern Med 2016;31:125133.CrossRefGoogle Scholar
Morfin-Otero, R, Garza-Gonzalez, E, Garcia Garcia, G, et al. Clostridium difficile infection in patients with chronic kidney disease in Mexico. Clin Nephrol 2018;90:350356.CrossRefGoogle ScholarPubMed
Pant, C, Deshpande, A, Anderson, MP, Sferra, TJ. Clostridium difficile infection is associated with poor outcomes in end-stage renal disease. J Investig Med 2012;60:529532.CrossRefGoogle ScholarPubMed
Predrag, S. Analysis of risk factors and clinical manifestations associated with Clostridium difficile disease in Serbian hospitalized patients. Braz J Microbiol 2016;47:902910.CrossRefGoogle ScholarPubMed
Sheth, H, Bernardini, J, Burr, R, et al. Clostridium difficile infections in outpatient dialysis cohort. Infect Control Hosp Epidemiol 2010;31:8991.CrossRefGoogle ScholarPubMed
Wei, HL, Wei, SH, Huang, CW, et al. Molecular typing and epidemiology of Clostridium difficile in respiratory care wards of central Taiwan. J Microbiol Immunol Infect 2015;48:6571.CrossRefGoogle ScholarPubMed
D’Agata, EMC, Apata, IW, Booth, S, et al. Suggestions for the prevention of Clostridioides difficile spread within outpatient hemodialysis facilities. Kidney Int 2021;99:10451053.CrossRefGoogle ScholarPubMed
Fawley, WN, Underwood, S, Freeman, J, et al. Efficacy of hospital cleaning agents and germicides against epidemic Clostridium difficile strains. Infect Control Hosp Epidemiol 2007;28:920925.CrossRefGoogle ScholarPubMed
Kim, KH, Fekety, R, Batts, DH, et al. Isolation of Clostridium difficile from the environment and contacts of patients with antibiotic-associated colitis. J Infect Dis 1981;143:4250.CrossRefGoogle ScholarPubMed
LIST K: EPA’s registered antimicrobial products effective against Clostridium difficile Spores. United States Environmental Protection Agency website. https://www.epa.gov/pesticide-registration/list-k-epas-registered-antimicrobial-products-effective-against-clostridium. Published 2021. Accessed March 17, 2022.Google Scholar
Edmonds, SL, Zapka, C, Kasper, D, et al. Effectiveness of hand hygiene for removal of Clostridium difficile spores from hands. Infect Control Hosp Epidemiol 2013;34:302305.CrossRefGoogle ScholarPubMed
See I, Bagchi, S, Booth, S, et al. Outbreak of Clostridium difficile infections at an outpatient hemodialysis facility—Michigan, 2012–2013. Infect Control Hosp Epidemiol 2015;36:972974.Google Scholar
Lachowicz, D, Szulencka, G, Obuch-Woszczatyński, P, van Belkum, A, Pituch, H. First Polish outbreak of Clostridium difficile ribotype 027 infections among dialysis patients. Eur J Clin Microbiol Infect Dis 2015;34:6367.CrossRefGoogle ScholarPubMed
Snyder, GM, Patel, PR, Kallen, AJ, Strom, JA, Tucker, JK, D’Agata, EMC. Antimicrobial use in outpatient dialysis units. Infect Control Hosp Epidemiol 2013;34:349357.CrossRefGoogle Scholar
D’Agata, EMC, Lindberg, CC, Lindberg, CM, et al. The positive effects of an antimicrobial stewardship program targeting outpatient hemodialysis facilities. Infect Control Hosp Epidemiol 2018;39:14001405.CrossRefGoogle ScholarPubMed
D’Agata, EMC, Tran, D, Bautista, J, Shemin, D, Grima, D. Clinical and economic benefits of antimicrobial stewardship programs in hemodialysis facilities: a decision analytic model. Clin J Am Soc Nephrol 2018;13:13891397.CrossRefGoogle ScholarPubMed
Figure 0

Fig. 1. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow chart of the study search and review process examining CDI among persons on maintenance hemodialysis. Note. CDI, Clostridioides difficile infection; MHD, maintenance hemodialysis.

Figure 1

Table 1. Characteristics of Studies Included in the Systematic Review and Meta-analysis

Figure 2

Fig. 2. Forest plot of the 10 included studies providing Clostridioides difficile prevalence among persons requiring and not requiring maintenance hemodialysis.

Figure 3

Fig. 3. Trends in Clostridioides difficile infection over time comparing persons requiring and not requiring maintenance hemodialysis. The line and shaded regions represent the modeled average and 95% confidence intervals.

Supplementary material: File

Elfanagely et al. supplementary material

Elfanagely et al. supplementary material 1

Download Elfanagely et al. supplementary material(File)
File 28 KB
Supplementary material: File

Elfanagely et al. supplementary material

Elfanagely et al. supplementary material 2

Download Elfanagely et al. supplementary material(File)
File 14.7 KB
Supplementary material: File

Elfanagely et al. supplementary material

Elfanagely et al. supplementary material 3

Download Elfanagely et al. supplementary material(File)
File 14.1 KB
Supplementary material: Image

Elfanagely et al. supplementary material

Elfanagely et al. supplementary material 4

Download Elfanagely et al. supplementary material(Image)
Image 47.3 KB