The kinetics of gut microbial community composition in patients with irritable bowel syndrome following fecal microbiota transplantation

Background Gut microbiota alterations are important in irritable bowel syndrome (IBS). The aim was to investigate the effect of fecal microbiota transplantation (FMT) on gut microbiota and the symptoms in patients with IBS. Material and methods The study included 13 IBS patients according to Rome III criteria and 13 healthy donors. Freshly donated feces were administered to the descending part of the duodenum via a gastroscope. Feces were collected from donors and patients before FMT, and from the patients at 1, 3 and 12 weeks and donors and patients at 20/28 weeks after FMT. Microbiota analysis was performed using GA-map Dysbiosis test (Genetic Analysis AS, Oslo, Norway). The patients completed the following questionnaires before and at the aforementioned weeks after FMT: IBS Symptom Questionnaire (IBS-SQ), IBS-Symptom Severity Scoring system (IBS-SSS), Short Form of Nepean Dyspepsia Index (SF-NDI), Bristol stool form scale, the Eysenck Personality Questionnaire-Neuroticism and Hospital Anxiety and Depression. Results Donors and IBS patients had significantly different bacterial strain signals before FMT (Ruminococcus gnavus, Actinobacteria and Bifidobacteria) that became non-significant after 3 weeks following FMT. The changes in gut microbiota were similar between donors and patients at 20/28 weeks after FMT. Thus, patients’ microbiota profiles became more-or-less similar to donors. The scores of all the questionnaires were significantly improved at all time points following FMT. No reported adverse effects. Conclusions FMT was associated with a change in gut microbiota and improvement in IBS symptoms and quality of life lasting for up to 28 weeks. Trial registration ClinicalTrials.gov ID: NCT03333291

symptoms as defined by IBS-Symptom Severity Scoring system (IBS-SSS) score >175 [29] and were referred to the gastroenterology outpatient clinic, Haukeland University Hospital, Bergen, Norway. The exclusion criteria included history of inflammatory bowel diseases, GI malignancy, blood in stool, an immunocompromised state, a history of opportunistic infections within 1 year prior to FMT, oral thrush, or disseminated lymphadenopathy. Patients who were scheduled for abdominal surgery, pregnant or lactating women and patients taking probiotics or antibiotics within 4 weeks prior to fecal installation were also excluded.

Donors
A donor group of healthy family members, males and females who were over 18 years of age was included. The exclusion criteria of the donors were pregnancy, history of inflammatory bowel diseases, IBS, chronic abdominal pain, GI malignancy, diarrhea, blood in stool, antibiotic and probiotic use within 4 weeks prior to FMT, an immunocompromised state, history of opportunistic infections within 1 year prior to FMT, oral thrush and disseminated lymphadenopathy.
The study was performed in accordance with the Declaration of Helsinki [30] and was approved by the Regional Committee for Medical and Health Research Ethics in Western Norway (reference no.: 2013/1497). All participants provided written informed consent. According to the Norwegian legislation, clinical trials concerning fecal transplantation are not regarded as drug clinical trial. When the study started we were unfortunately not aware of the requirements of registering non-drug clinical trials. Hence this trial was registered retrospectively at ClinicalTrials.gov (ID: NCT03333291).

Study design
The FMT procedure was done only once and fecal samples were analyzed across several time points before FMT at screening (week -1) and FMT day (week 0), and then after FMT at weeks 1, 3, 12 and 28 weeks. The scheduled study visits are outlined in Table 1. The donors and patients completed several questionnaires and delivered fresh stool samples soon (a couple of hours) after defecation at screening and during visits after FMT as outlined in Table 1. The patients received special containers to preserve their stool in and were informed to place them in the refrigerator (4˚C) if it will take longer than a couple of hours before delivery. The patients were informed not to apply any changes to their diet or life style and to report any bout of new infections and/or use of new medications during the study.

Screening
Screening of the donors and the patients was scheduled one week before FMT. All of the donors and patients filled out symptom questionnaires (Table 1), received physical examinations and were screened (in blood and stool) for previous exposure to contagious infectious agents, inflammation and other organic diseases. Screening of the donors' blood included serologic testing for hepatitis A, B, C, human immunodeficiency virus (HIV), Epstein-Barr virus and cytomegalovirus. The blood from the patients was tested and included: Hemoglobin, leucocytes, platelets, creatinine, aspartate transaminase (AST), alanine transaminase (ALT), International Normalized Ratio (INR), electrolytes and chromogranin A. Stool samples from the donors and patients were examined for fecal calprotectin, cultured for enteric bacterial pathogens and screened for viruses and parasites.

The FMT procedure
On FMT day, the patients brought >60 g of fresh feces from their donors along with 60 g of their own feces before transplantation. Only 30 g of donor feces were used to prepare the fecal suspension by mixing them with 60 ml of normal saline. The remaining donor feces and feces from the patients were stored at -80˚C until they were analyzed for microbial analysis. The patients completed several questionnaires before FMT (Table 1). Gastroscopy was performed on the patients (after an overnight fast) to install 60 ml of fecal suspension followed by 60 ml of normal saline in the descending part of the duodenum distal to the papilla Vateri. All of the gastroscopies were performed by an endoscopist (T.M., G.A.L. or T.H.) at the gastrolab, Haukeland University Hospital, Bergen, Norway. The second visit was planned at week 3, instead of week 4 as outlined in the original protocol, due to practical reasons, and only 30 g of donor feces was used only once, as outlined in the original protocol (S1 File), in accordance with previous recommendations [10,31].

Gut microbiota analysis
Gut microbiota analysis was performed using the GA-map Dysbiosis test (Genetic Analysis AS, Oslo, Norway) by algorithmically assessing fecal bacterial abundance and profile (dysbiosis index, DI), and potential deviation in the microbiome from normobiosis. [32] Briefly, GAmap Dysbiosis test is based on fecal homogenization, mechanical bacterial cell disruption and automated total bacterial genomic DNA extraction using magnetic beads. DI is based on 54 DNA probes targeting more than 300 bacterial strains based on their 16S rRNA sequence in seven variable regions (V3-V9). Twenty-six bacteria probes are species specific, 19 detect bacteria on genus level, and 9 probes detect bacteria at higher taxonomic levels. Probe labeling is by single nucleotide extension and hybridization to complementary probes coupled to magnetic beads, and signal detection by using BioCode 1000A 128-Plex Analyzer (Applied Bio-Code, Santa Fe Springs, CA, USA). A DI above 2 shows a microbiota profile that differs from that of the normobiotic reference collection (DI 1-2: non-dysbiosis, DI 3: moderate, DI 4-5: severe dysbiosis) [32].

Questionnaires
Gastrointestinal symptoms and bowel habits were evaluated using IBS-SSS [29] in which a decrease of 50 points in IBS-SSS score using a visual assessment scale (VAS) from baseline (before FMT) correlated with improvement in clinical symptoms, and IBS symptom questionnaire (IBS-SQ) [33,34] that was completed on the day of screening and then daily for 20 days after FMT. Responders and late-responders were patients who achieved a reduction of >50 points in IBS-SSS score after 1 and 3 weeks following FMT, respectively [29]. Non-responders were those who achieved <50 points in IBS-SSS score following FMT at any time period compared to baseline. Stool consistency was evaluated using Bristol stool form scale [35], which ranges from 1 (constipation) to 7 (diarrhea). Quality of life (QoL) was assessed using Short Form of Nepean Dyspepsia Index (SF-NDI) questionnaire [36]. Psychometric evaluation was performed using the Eysenck Personality Questionnaire-Neuroticism (EPQ-N-12) with a cut-off value of 4 [37], and Hospital Anxiety and Depression (HAD) where scores >8 in either subscale were considered to indicate anxiety or depression, respectively [38,39].

Statistical analysis
Graphpad Prism 6 (GraphPad Software Inc., La Jolla, CA, USA) was used for all analysis. Kruskal-Wallis non-parametric test with Dunn's post test and Mann-Whitney U test were used to analyse the data between the donors and patients before and after FMT. One-way ANOVA with repeated measures and Paired t-test was used to analyze the data for the patients before FMT and each visit after FMT. Multiple t-test corrected using Holm-Sidak method, was used to compare between the bacterial signals of responders and non-responders. P < 0.05 is considered to indicate a statistically significant difference. Cluster analysis and principal component analyses (PCA) were used to visualize the microbiota data, showing the extent to which microbial communities share branch length.

Participants
The recipients group included 16 patients with IBS and the donors group included 16 healthy subjects. Participants of both groups were recruited after fulfilling the inclusion's criteria and, most importantly, none has used antibiotics during the past 6 months prior to inclusion in the study (Fig 1). Three, originally recruited, patients were excluded after withdrawing their consent to participate for practical reasons (n = 1), being diagnosed with functional dyspepsia (n = 1) and finding Clostridium difficile in stool culture (n = 1). Hence, 13 patients (9 males and 4 females, mean age of 32 years and age range of 20-44 years) and 13 donors (6 males and 7 females, mean age of 33 years and age range of 20-42 years) completed the whole study, and filled out the questionnaires and delivered fecal samples as previously explained. All of the patients had IBS mostly diarrhea-predominant, in which six patients had PI-IBS (after a local Giardia outbreak in Bergen in 2004 [40]) and seven patients had idiopathic IBS. The last visit was originally scheduled at 28 weeks after FMT, but 4 patients and also their respective donors were scheduled for a last visit at 20 weeks instead of 28 weeks after FMT due to practical reasons. The blood tests and stool cultures of both donors and patients were normal prior FMT and control blood tests for the patients were also normal at the end of the study. Detailed health and symptom questionnaires were provided to both groups at screening day and only to the patients following FMT throughout the study. We asked both the patients and donors to report any changes in their diet, life style, medications or health history during the whole study. Neither group reported any such changes during study participation.

Gut microbiota
At baseline, the patients had significantly higher DI than the donors (4±0.5 and 2.6±0.2, respectively, P = 0.046, Fig 1). Following FMT, the DI for the patients gradually decreased to 3.9±0.4 at week 1, then 3.3±0.3 at week 3 and 2.9±0.2 at week 12 but then increased again to 3.5±0.3 at week 20/28. The changes in the DI comparing between the patients following FMT and the donors were not statistically significant (Fig 2).
Donors and IBS patients had significantly different bacterial signals before FMT, namely, signals for Ruminococcus gnavus, Actinobacteria and Bifidobacteria, which became non-significantly different after 3 weeks following FMT (Table 2). At weeks 12 and 20/28, new bacterial strains in the patients feces; namely Bacteroides/Prevotella, Alistipes, Actinobacteria and Bifidobacteria became significantly different from that of the donor at the beginning of the study, as shown in Table 2, but not statistically different from that of the donors at the end of the study at week 20/28 (P = 0.09, 0.08, 0.6, 0.14 and 0.9, respectively). The signal levels of Actinobacteria and Bifidobacteria increased significantly towards the levels measured for the donors and lasted for 12 weeks after FMT but then significantly decreased at week 20/28. The PCA scores plot of the gut microbiota profiles corrected for sample differences showed a gradual shift of the gut microbiota profile over time (Fig 3).
Comparing the microbiota between the groups, responders (n = 8) and non-responders (n = 5), which were defined based on achieving >50 or <50 IBS-SSS points, respectively, at week 20/28 compared to baseline, showed significant differences (adjusted P-values) in the Bacteroides signals between the donors at the beginning and the end of the study (P<0.0001 and 0.23, respectively) and between the patients at weeks 0, 1, 3, 12 and 20/28 (P<0.0001, <0.0001, 0.23, 0.08 and 0.08, respectively), Fig 4A. The respective values for Desulfitispora signals between the donors at the beginning and the end of the study (P = 0.0005 and 0.98, respectively) and between the patients at weeks 0, 1, 3, 12 and 20/28 (P = 0.0002, <0.0001, 0.0005, 0.0025 and <0.0001, respectively), Fig 4B, and for Megasphaera/Dialister signals between the donors at the beginning and the end of the study (P = 0.54 and 0.53, respectively) and between the patients at weeks 0, 1, 3, 12 and 20/28 (P = 0.13, 0.029, 0.24, 0.013 and 0.029, respectively), Fig 4C. As for Bifidobacteria, differences in the bacterial signals between responders and nonresponders were noted, however, they were statistically not significant, Fig 4D.

Questionnaires
The score of IBS-SSS (mean±SEM) for the donors was 18±8.9 and for IBS-SQ is 0.7±0.3, which indicated asymptomatic status. The score of IBS-SSS of 11 patients were considered severe (IBS-SSS score >300) and only two patients had moderate severity (IBS-SSS score = 176-300) before FMT. The score of IBS-SSS (mean±SEM) for the patients at screening day was 333.6±20. IBS-SSS scores at FMT-day and at weeks 1, 3, 12 and 20/28 after FMT and comparisons between FMT-day and at each week are presented in Table 3. Using paired t test showed no significant difference in the aforementioned IBS-SSS scores for the patients between screening and FMT-day (P = 0.45), however, a significant reduction was noted in IBS-SSS scores of the patients between screening and weeks 1, 3, 12 and 20/28 (P = 0.003, 0.0004, 0.0095 and 0.012, respectively). No significant differences were observed by comparing the scores in weeks 1, 3, 12 and 20/28 interchangeably between each other, Fig 5. Four out of the 13 patients did not achieve >50 points reduction in IBS-SSS scores from baseline at week 1. However, two out of these four patients were late responders and achieved >50 points reduction in IBS-SSS scores from baseline at week 3. Therefore, a total of 9, 11, 9 and 8 out of the 13 patients achieved >50 points reduction in IBS-SSS scores from baseline at weeks 1, 3, 12 and 20/28 following FMT, respectively, and were considered as responders.
The scores for the following questionnaires are presented in Table 3: Bristol stool form scale, total SF-NDI scores, EPQ-N-12 and HAD. Bristol stool form scale showed significant changes in the stool form from diarrhea type before FMT to normal following FMT (weeks 1 and 3, P = 0.07 and 0.04, respectively), however, no significant differences were noted between weeks 1, 3, 12 and 20/28 when comparing them interchangeably with each other. The total SF-NDI scores showed a significant improvement in the QoL following FMT that lasted to the end of the study and no significant difference was found between weeks 3 and 20/28. The scores for EPQ-N-12 and HAD showed only significant improvement in HAD scores (anxiety and depression) 3 weeks after FMT (P = 0.016 and 0.038, respectively). The scores for the different domains of IBS-SQ during the first 3 weeks (20 days) showed significant improvements (except for anorexia) after receiving FMT, as presented in Table 4.

Post-FMT complications
No complications were reported during and following FMT until the end of the study.

Discussion
In this study of the kinetics of gut microbial community composition after FMT in IBS patients, the gut microbiota profile of the patients, which differed significantly from the donors Table 2. Characteristics of the fecal bacterial signals between the donors at the beginning and end of the study, and patients in weeks 0, 1, 3, 12 and 20/28. The left part of the table shows the bacterial signals given for donors and recipients at different time points before and after FMT and the directionality towards or away from that of the donors at baseline. The right part of the table shows the P-values when comparing the bacterial signals for the recipients at different time points to that for the donors at baseline. before FMT, have shown some dynamic changes during the 28 weeks of follow up period post FMT. In addition, the patients complained of severe IBS symptoms and low QoL before FMT. The change in the gut microbiota in IBS patients during the course of the study parallels a rapid improvement in the patients' symptoms and QoL that lasts up to 28 weeks. The DI of the patients changed from severe dysbiosis before FMT to moderate dysbiosis after 12 weeks following FMT and maintained its new status throughout the course of the study. This showed that administrating the FMT via gastroscope in to the duodenum did not cause/worsen dysbiosis, on the contrary, it helped change the gut microbiota towards normobiosis. In general, dysbiosis in IBS is characterized by a decrease in Actinobacteria, Bifidobacteria and Lactobacillus [41], and an increase in Bacteroidetes, Firmicutes and Proteobacteria in the feces [41,42]. Increased Proteobacteria in diarrhea-predominant IBS (IBS-D) including E. coli is associated with increased inflammation [41]. Bifidobacteria count is either decreased [41,43] or increased in IBS patients [44]. In the current study, the bacterial signals for Actinobacteria in general and especially in Bifidobacteria in the total IBS group and the subgroups of IBS patients are significantly reduced compared to the donors' group at the beginning of the study. The gut microbiota profile in patients with IBS-D has a significant increase in bacteria belonging to the Bacteroidetes phylum and in Ruminococcus species [41,45]. Increased Ruminococcus species may cause degradation of the mucus layer that allows infiltration of Streptococcus species causing low-grade inflammation [41]. In contrast to a previous study [24] that found no significant difference in the gut microbiota in IBS patients compared to donors other than a significant increase in Streptococcus counts in donors, we found significant differences in several bacterial signals in our IBS patients compared to their donors including a significant increase in bacterial signals for Actinobacteria and Bifidobacteria in the donors' group. Actinobacteria and Bifidobacteria are important for gut mucosal barrier to keep pathogens from crossing over [41]. These alterations in the gut microbiota profile, especially that of Ruminococcus gnavus, Proteobacteria and Shigella/Escherichia might have contributed to the mechanism of low-grade inflammation in PI-IBS and IBS-D. Proteobacteria and Shigella/Escherichia signals in the recipients were significantly higher than that of the donors before FMT. Proteobacteria and Shigella/Escherichia signals further increased during the first week following FMT but then changed (decreased) towards values of the donors. No inflammatory changes were clinically noted among the recipients but one cannot exclude that a low-grade inflammation occurs in patients with IBS even from before FMT. The PCA scores show that FMT seems to have an effect on the gut microbiota as systematic change occurred from baseline before FMT and over a period of 28 weeks. In a comparison of the gut microbiota profiles between the responders and non-responders groups, the changes in the bacterial signals of Bacteroides, Desulfitispora and Megasphaera/ Dialister between the patients during the study are similar to or tend to change toward those  measured for the donors at the beginning of the study. This may be explained by the fact that the donors and the patients are relatives and/or living in the same environment and also by the impact the donors' gut microbiota may have on the patients' gut microbiota. Another interesting observation is that the signals for Bifidobacteria in the responders group are lower than those in the non-responders group. Low Bifidobacteria signals have also been observed in IBS patients following a low-FODMAP diet as shown in a previous study [46]. The changes in the patients' gut microbiota following FMT may have contributed to the subsequent improvement in their symptoms and thus QoL [25]. Similar observation has been recently described after using FMT to restore the bacterial diversity and resolve the dysbiosis in patients with recurrent Clostridium difficile infection [47]. Some of the bacterial strains changed towards the end of the study (weeks 12 and 20/28) without significantly worsening the symptoms of IBS and/or QoL during the same period. This may suggest that the changes in the symptoms and QoL may be related to the collective changes in the gut microbiota rather than individual bacterial change. The change in the gut microbiota profile of the patients at weeks 12 and 20/28 following FMT compared to the donors before FMT ( Table 2) resembled that of the donors at the end of the study compared to the donors at the beginning of the study, which may have contributed to the increase in the DI towards the end of the study (Fig  2). The changes towards the end of the study are quite interesting as they raise a question to whether other factors may have influenced these changes such as the participants' milieu, dietary (which were not changed according to the metadata) or hereditary factors; since the donors and the patients are relatives either living in the same environment (for example: spouse) or sharing the same genes (for example: a parent or a sibling) or both. Another explanation may have been due to the imposed changes to the patients' gut microbiota following FMT rendering them susceptible to the same changes occurring to the donors' gut microbiota.

Bacteria strain
FMT was associated in time with rapid improvement in IBS-SSS score (>50 points reduction from baseline) [29]. In the current study, 70% of the patients has improved IBS-SSS scores during the first week, 85% by 3 weeks, 70% by 12 weeks and 62% over 20 weeks towards the Fecal microbiota transplantation in irritable bowel syndrome end of the study. The long lasting effect of FMT on IBS symptoms is in line with previous studies [23,24]. In addition to an immediate improvement (during the first 3 days after FMT) in IBS-SQ-total and specific-symptom scores, namely; bloating, abdominal pain and diarrhea, a gradual but statistically significant improvement in all of the IBS symptoms as assessed by IBS-SQ (except for anorexia/loss of appetite) was observed on daily basis during the first 3 weeks following FMT. The improvements in IBS related QoL, abdominal pain and bloating during the course of the study is consistent with a previous report [24]. Currently, dietary manipulation is one of the methods for the management of IBS symptoms. [48][49][50] The response rate of an elimination diet ranges between 15 and 71% [51], and of low FODMAP diet is up to 86% [46,48], with a high placebo response rate reaching to 40% [48]. One can suggest that FMT may serve as an alternative method for managing IBS instead of the dietary manipulation due to similar high response rate, easy to apply, and long lasting improvements in the symptoms and QoL up to one year [23,24]. However, head to head comparison studies must be further conducted.
The route of administration of feces, either via gastroscopy [23] or colonoscopy [24] in to the upper or lower GI tract, respectively, have reported similar effects on IBS symptoms [23,24]. In a systematic review of FMT in recurrent Clostridium difficile infection, the use of gastroscopy as an administration route had a response rate of 76% vs. 88% for colonoscopy [52]. In this study, we have chosen to use gastroscopy because it is an easy and a fast route of administration and because patients with IBS often have dilation of small bowel segments giving small intestinal bacterial overgrowth, SIBO, [53], which would have escaped the suspension had we chosen to use colonoscopy for FMT. The procedure is considered to be safe [31]. No complications were reported during and following FMT until the end of the study. In other studies, short-term adverse events occurred after FMT such as abdominal cramps, belching and nausea but they were self-limited and transient [23,54]. Long-term, follow-up studies (3-68 months after FMT) have found FMT to be relatively free of adverse effects [55].
The strength of the study is the usage of validated methods to study the kinetics in the gut microbiota and validated questionnaires to assess the changes in the stool form, symptoms and QoL. The main limitations of the study are its design as an open-label trial, not placeboor sham-controlled, and the small sample size. However, our main focus was to study the changes in the gut microbiota in the patients compared to the gut micobiota of the donors, which are supposed to have more-or-less stabile profiles during the study. Nevertheless, significant results were obtained despite the small sample size. Larger double-blinded, placebo-controlled studies are necessary to address applicability of FMT in IBS and are currently running else where in Norway (NCT02154867).

Conclusions
This is the first study to investigate the kinetics of microbial community composition in IBS patients following FMT. FMT was associated with a rapid change in the alterations in the signals for several strains of the gut microbiota making it statistically not significantly different from the donors after 3 weeks following FMT. The gut microbiota profile at the end of the study was similar to the profile of the donors taken at the same time. The symptoms and QoL have improved significantly quite soon after FMT and lasted up to 28 weeks.