Bacterial microbiome changes after fecal transplantation for recurrent Clostridioides difficile infection in the Brazilian center

Abstract

Clostridioides difficile infection (CDI) poses a significant global health threat owing to its substantial morbidity and associated healthcare costs. A key challenge in controlling CDI is the risk of multiple recurrences, which can affect up to 30% of patients. In such instances, fecal microbiota transplantation (FMT) is increasingly recognized as the optimal treatment. However, few related studies have been conducted in developing countries, and the microbiota composition of Brazilian patients and its dynamic modification post-FMT remain largely unexplored. This study aimed to evaluate the changes in the bacterial gut microbiome in Brazilian patients with recurrent CDI post-FMT. Ten patients underwent FMT, and the primary and overall CDI resolution rates were 80% and 90% after the first and second FMT, respectively. FMT was associated with an early increase in Shannon’s diversity, evident as soon as 1 week post-FMT and persisting for at least 25 days post-treatment. Post-treatment, the abundance of Firmicutes increased and that of Proteobacteria decreased. Specifically, the abundance of the genera Ruminococcus, Faecalibacterium, Lachnospira, and Roseburia of the Firmicutes phylum was significantly higher 1 week post-transplantation, with Ruminococcus and Faecalibacterium remaining enriched 25 days post-transplantation. This study is the first of its kind in Brazil to evaluate the microbiota of a donor and patients undergoing FMT. Our findings suggest that FMT can induce remarkable changes in the gut microbiota, characterized by an early and sustained increase in diversity lasting at least 25 days. FMT also promotes enrichment of genera such as Ruminococcus spp., Faecalibacterium spp., and Roseburia spp., essential for therapeutic success.

Supplementary Information

The online version contains supplementary material available at 10.1007/s42770-023-01227-4.

Short communication

Clostridioides difficile infection (CDI) poses a considerable health concern worldwide owing to its substantial morbidity, mortality, and associated healthcare costs [1]. Although data on CDI in emerging countries is limited, existing studies indicate a high incidence and prevalence [2]. In Brazil, for instance, an incidence density of 9.2 CDI cases per 10,000 patient-days has been reported in a referral center [3]. This figure is slightly higher than those reported in most studies conducted in European countries, Canada, and the USA [4–6]. The risk factors identified were consistent with those found in other regions, including antibiotic use, hospitalization, multiple comorbidities, and advanced age [2, 7].

The standard treatment for a non-severe initial episode of CDI could be vancomycin, metronidazole, or fidaxomicin administration, each boasting an 80% success rate [1]. In the case of a severe episode, vancomycin or fidaxomicin is recommended, albeit with a lower response rate of approximately 65% [1]. Additionally, more than 20% of patients encounter a recurrence after the initial treatment, with the recurrence rate escalating after each subsequent episode of CDI [1]. Fecal microbiota transplantation (FMT) has been recognized as an effective therapeutic option for achieving sustained clinical cure in patients with multiple relapses.

The microbiota of patients with recurrent CDI exhibits pronounced alterations in structure and composition compared to that in healthy individuals [8]. Metagenomic analyses have revealed a relative increase in opportunistic pathogens such as Proteobacteria (e.g., Escherichia spp., Shigella spp.) and a decrease in commensal bacteria from the phylum Firmicutes and Bacteroidetes (e.g., Prevotella spp., Bacteroides spp.) [8]. In terms of the metabolome, patients with CDI also exhibit reduced levels of secondary bile acids and short-chain fatty acids, both of which have been linked to recurrent CDI [9]. In this scenario, FMT is widely acknowledged as an effective treatment. Randomized controlled trials and meta-analyses have shown that FMT offers a high cure rate for recurrent CDI, ranging from 80 to 90% [1, 10]. Recently, the first Brazilian center for FMT reported a similar success rate using frozen samples from a local stool bank [11].

The composition of the human gut microbiota is influenced by several factors, including age, ethnicity, dietary patterns, and geographic location [12]. While certain taxonomic families have been identified as crucial for FMT success, knowledge about the microbiota composition of Brazilian patients and its dynamic alteration post-FMT is limited [13]. Characterizing the microbiota and its transformation post-FMT could enhance our understanding of disease pathogenesis and pave the way for novel biological therapeutic strategies. Hence, this study aimed to evaluate the alteration in the bacterial gut microbiome of Brazilian patients with recurrent CDI following FMT, using frozen samples obtained via colonoscopy.

This prospective study encompassed patients treated with FMT for recurrent CDI at Hospital das Clínicas, Federal University of Minas Gerais (HC/UFMG) from September 2017 to March 2020. We recruited adult outpatients with a minimum of two documented CDI episodes. CDI was identified by diarrhea persisting for over 48 h (exceeding three daily bowel movements with unformed stools) and subsequent microbiological confirmation. Recurrent CDI was characterized as a new CDI episode occurring within 8 weeks of a previously adequately treated infection [1]. We collected the following data from each patient: age, sex, comorbidities, number of CDI episodes, risk factors, and prior antimicrobial treatments. The study was approved by the Institutional Ethics Committee of the Federal University of Minas Gerais (CAAE: 13,552,719.3.0000.5149).

Patients with diarrhea of a different etiology or those who failed to meet the criteria for recurrent CDI were excluded. CDI resolution was characterized by the cessation of diarrhea following 8 weeks of treatment. This resolution could be primary, achieved with a single infusion, or overall, achieved with two or more FMTs. FMT failure was identified as the recurrence of CDI within 8 weeks post-fecal infusion, confirmed by laboratory testing.

Donors underwent screening in accordance with international guidelines and national regulatory resolutions [10, 14]. The processes of donor selection, fecal substrate processing, storage, and defrosting mirrored those detailed in our prior report [14]. Donor substrates were preserved at -80 °C for a maximum duration of six months within the institutional stool bank [14]. Baseline fecal samples from patients were gathered to verify the diagnosis and stored subsequently. Patients were administered vancomycin (125 mg orally, four times daily) for a period of 10 to 14 days, ceasing 48 h prior to FMT. On the eve of FMT, patients underwent bowel preparation using a polyethylene glycol (PEG) solution. The procedure was conducted via colonoscopy, with approximately 300 mL of fecal substrate infused into the cecum. Post-procedure, fecal samples from recipients were collected on the 7th and 25th days. Probiotics were not administered to any patients. Clinical follow-ups were conducted at 8 weeks, 3 months, 6 months, and 1 year to monitor the resolution of diarrhea and the occurrence of adverse events. One sample from the selected donor was also included in the present study for comparison porpoise.

Fecal microbiota from patients was examined both before and after FMT, specifically on days 7 and 25 post-transplantation. The samples were categorized into three groups: the “diarrheic group” (DA), which consisted of samples from patients with recurrent CDI prior to FMT; TP7, which included samples taken seven days post-transplantation; and TP25, which comprised samples collected 25 days post-transplantation. All samples were stored at -80 °C until DNA extraction was performed. Total DNA was isolated from 100 mg of each clinical fecal sample, following the QIAamp DNA Stool Mini Kit protocol (QIAGEN, Germany). Libraries were prepared using the TruSeq Library Prep Kit (Illumina, USA) to sequence bacterial amplicons, utilizing oligonucleotides 341F and 806R specific to the V3/V4 region of the 16S rRNA [15]. These libraries were adjusted to a final concentration of 17.5 pM and sequenced using the MiSeq system (Illumina, USA). Paired-end runs of 500 cycles were conducted with the V3 × 600 sequencing kit (Illumina, USA), achieving 100,000 coverage reads per sample in this study.

Fastq data were subjected to quality filtering, which included the elimination of truncated and low-quality reads (Phred score < 20), using Trimmomatic [16]. Subsequently, forward and reverse paired reads were consolidated into contigs. These sequences were then subjected to singleton removal and chimera filtering using Mothur [17]. Thereafter, the pre-processed sequences were grouped into operational taxonomic units (OTU) with a 97% identity, and taxonomically assigned using QIIME v.1.9.1 [18]. This required a 97% sequence similarity threshold against the Silva database [19].

The Shannon, Chao1 and Simpson index were employed to estimate alpha diversity as described elsewhere [20–22]. The RStudio packages Vegan, Fossil and Microbiome were utilized for this analysis. The relative abundance of OTU across the groups was determined using STAMP [23]. To investigate dissimilarities among the groups, PCoA analyses were conducted using Bray–Curtis, unweighted UniFrac and weighted UniFrac distances. (Supplementary file).

The Shapiro–Wilk Test was employed to confirm the normal distribution of OTU data. The one-way ANOVA, followed by the Tukey post hoc test, was used to estimate the variance in mean values across multiple group comparisons. For multivariate data, significant differences among groups were determined using ANOSIM. A p-value of less than 0.05 was considered statistically significant.

Ten patients with recurrent CDI underwent FMT via colonoscopy using frozen samples. The median age of the patients was 68 years, with a range of23 to 87 years, and 70% of them were women. All patients had used antibiotics, and half had been hospitalized prior to their first CDI episode. The median number of recurrent CDI episodes was 3, ranging from1 to 4 recurrences, and the median disease duration was 3.3 months, ranging from 1.7 to 7.1 months. All patients had previously been treated with vancomycin, eight of whom also received metronidazole, and one received additional fidaxomicin therapy.

The resolution of CDI with a single FMT was 80%, and the overall resolution of CDI after a second FMT increased to 90%. Among the non-responders, specifically patients P1 and P6, the median interval between FMT and CDI relapse was 9.5 days. Patient P6 underwent a second FMT, while patient P1 chose a new tapered course of vancomycin, which successfully prevented further recurrence. The median follow-up time post-transplantation was 14.4 months, with a range of 2.3 to 26.1 months.

The microbiota sequencing of patient P9 was unsuccessful and was therefore excluded from the analysis. The multivariate analysis using ANOSIM indicated that the dissimilarity among the DA, TP7, and TP25 groups exceeded the dissimilarity within each individual group (Fig. 1). These findings corroborate that the microbial community in patients with recurrent CDI, who underwent FMT, alters after 7 and 25 days (p = 0.0136). The differences in diversity of DA compared to TP7 and TP25 were also obvious on Chao1 and Simpsons analysis (Fig. 1). Furthermore, the PCoA analysis implied that the majority of the difference (79.4%) is linked to bacteria within the Firmicutes and Proteobacteria phyla (Fig. 2). A similar result was obtained both weighted and unweighted UniFrac (Supplementary File).

Fig. 1.

A) Multivariate analysis of dissimilarity among the groups with diarrhea (DA), seven days after transplantation (TP7), and 25 days after transplantation (TP25). The graph illustrates Bray–Curtis distance matrix with 9999 permutations. The dissimilarity among the groups is greater than that within the groups. ANOSIM, p < 0.05. B) Shannon index estimated for the groups DA, TP7, and TP25 and donor sample (DO). A significant increase can be seen in TP7 and TP25 samples compared to that in the DA group. ANOVA, p < 0.05. C) The Shannon alpha diversity for the groups DA, TP7, and TP25 and DO. D) Chao1 alpha diversity for the groups DA, TP7, and TP25 and DO

Fig. 2.

Multidimensional analysis of principal coordinates (PCoA). The graph illustrates Bray–Curtis distance matrix estimated for all the samples

Despite high proportions of Proteobacteria on day 7 post-FMT in non-responders (patient P1 [P1_DA] and patient PT6 [P6_DA]), no significant disparities were detected at the phylum level (Fig. 3). Figure 4 displays the relative abundances of taxa at the genus level from groups TP7 and TP25, respectively, revealing significant changes compared to those in the DA group. These findings suggest that FMT could enhance the presence of bacteria from the Ruminococcus, Faecalibacterium, Lachnospira, and Roseburia genera after 7 days (Fig. 4). The Ruminococcus and Faecalibacterium genera were also found to be enriched after 25 days, along with Oscillospira and Christensenella. Furthermore, a decrease in Streptococcus spp. and Anaerobacillus spp. was observed in the TP25 group.

Fig. 3.

Relative abundance at the Phylum taxonomy level. Stacked bars represent the proportion of reads classified within a taxon

Fig. 4.

Changes in abundance at the genus level between DA and TP7 (A) and DA and TP25 groups (B). Only taxa with significant differences in the mean proportion are shown. ANOVA, p < 0.05

In this study, we observed a 90% resolution of CDI and a significant enhancement in microbiota diversity following FMT, counteracting the typical pronounced decrease in alpha diversity seen in patients with CDI. This shift was marked by an augmentation in Firmicutes abundance and a reduction in Proteobacteria post-treatment. Our findings align with prior reports from other FMT centers, underscoring the potential importance of increased diversity in suppressing CDI [24, 25].

Firmicutes is a phylum encompassing several beneficial bacterial families and genera, including Ruminococcus, Lactobacillus, Lachnospiraccae, and Faecalibacterium. These bacteria metabolize primary bile acids and produce short-chain fatty acids (SCFAs), such as butyrate, which are crucial for maintaining intestinal homeostasis [26]. Proteobacteria is a phylum that includes a diverse array of gram-negative bacteria, such as Enterobacteriaceae. A high abundance of Proteobacteria is linked to various gastrointestinal diseases, including recurrent CDI [26]. A decrease in Proteobacteria, particularly the Enterobacteriaceae family, following FMT is a key indicator of responders, aligning with the demonstrated high therapeutic success [24]. Further, Shannon's diversity is known to significantly increases in responders 1 week post-FMT [24].

The current study demonstrates that changes in fecal microbiota following FMT via colonoscopy manifest early and endure for a minimum of 25 days post-treatment. Our observations of notable early differences post-FMT align with prior American and European studies, regardless of the administration method employed [24, 25, 27]. These results underscore the notion that these alterations arise early and are not contingent on the route of administration (upper or lower).

The genera Ruminococcus, Faecalibacterium, Lachnospira, and Roseburia demonstrated a significant increase in abundance 1 week post-transplantation, with Ruminococcus and Faecalibacterium remaining enriched after 25 days. Both Ruminococcus and Faecalibacterium species are linked to a decreased risk of CDI recurrence, and their abundance typically escalates following FMT [27]. Importantly, a direct correlation between the success of FMT and the abundance of Ruminococcus species in the donor fecal sample has been proposed previously [27], a hypothesis that requires further validation. However, it is evident that an increase in Ruminococcus species abundance post-FMT is a promising sign of the recipient's gut health restoration [13, 26]. Besides Ruminococcaceae, a high abundance of Lachnospiraceae has also been suggested as a characteristic of “super-donors” and is indicative of a restored recipient gut environment [13]. FMT was also linked to the enrichment of Firmicutes families, supporting previous reports [24, 25, 27]. Roseburia, a Firmicute short-chain-fatty acid producing bacterium previously associated with recurrence prevention [28], was significantly more abundant in responders in this study.

The primary constraint of this study lies in the limited number of subjects included in the sample. The small sample size precluded robust inferences regarding the differences between responders and non-responders. Nevertheless, this study represents the first Brazilian investigation into the microbiota of donors and patients undergoing FMT in Brazil. Our results indicate that FMT can induce significant alterations in gut microbiota, characterized by an immediate and sustained increase in diversity for a minimum of 25 days. FMT facilitates the enrichment of taxa crucial for therapeutic success, including Ruminococcus spp., Faecalibacterium spp., and Roseburia spp.

Supplementary Information

Below is the link to the electronic supplementary material.

Funding

This study was funded by Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG RED-00132–16 and APQ-00524–17), Coordination for the Improvement of Higher Education Personnel (CAPES – Prêmio CAPES 2015—0774/2017), and National Council for Scientific and Technological Development (CNPq—406402/2018–3).

Data availability

The authors confirm that most data supporting the findings of this study are available within the article or its supplementary materials.

Declarations

Competing interests

The authors declare no competing interests.