Search strategy

We conducted searches on the PubMed, Web of Science, and Scopus databases up to November 2, 2023, with no limitations on language, date, or article type. We combined terms related to "Fecal microbiota transplantation" and "Viral diseases": ((("Feces" OR "fecal" OR "stool") AND ("microbiota" OR "Microbiome" OR "flora") AND ("Transplantation" OR "transfusion" OR "implant" OR "bacteriotherapy")) OR ("Donor Feces Infusion" OR "Intestinal Microbiome Transplant" OR "Intestinal Microbiota Transfer")) AND ("Viral Infection" OR "viral disease" OR "COVID-19" OR "HIV" OR "CMV" OR "Hepatitis" OR "HPV" OR "Herpes Simplex" OR "EBV" OR "VZ Virus") (S1 Table). We also investigated the Google Scholar search engine, as grey literature search following the completion of the full-text review [23]. Additionally, we carried out backward and forward citation searches for the included studies.

Study selection

The records were imported to EndNote software and any duplicated items were removed using that. We included the followings: 1) clinical trial (with or without control arm) in any phase; 2) participants with any types of viral diseases (e.g., HIV, COVID-19, viral hepatitis, and cytomegalovirus infection); 3) the intervention had to be FMT; 4) in case of any control arms, placebo, standard treatment, or other modalities except for FMT should be used; and 5) at least outcomes related to the efficacy or safety should be reported. We excluded studies enrolled healthy participants or patients with disorders other than viral diseases, studies that did not use FMT as an intervention, and studies other than clinical trials (e.g., animal studies, in vitro studies, editorials, letters, perspectives, opinions, reviews, news, and books). Also, studies that evaluated gut microbiota patterns without using FMT were excluded.

Two reviewers (MMM and AASAK) independently evaluated the titles/abstracts of each study. Then, the same two reviewers independently reviewed full-texts of articles included from the previous step. Any disagreements were brought to the deliberation by discussion or consultation with the senior investigator (SAN).

Data extraction

A data extraction sheet was designed by the senior investigator (SAN) in Microsoft Excel Office, version 2016. Then, two authors (RE and DRH) obtained the relevant data from the eligible studies independently. The principal investigator resolved discrepancies. The extracted variables were divided into four categories: 1) Study characteristics (first author’s name, year, country, and clinical trial phase and blinding); 2) Patient characteristics (age, sex, follow-up duration, type of viral disease, and underlying diseases); 3) Intervention characteristics (type, route of administration, schedule, and dose); and 4) Outcomes (effect sizes for each efficacy and safety outcome assessed in the intervention and control arms). In this systematic review, no missing data were encountered, and all relevant information from the included studies was fully available. Since this review did not involve a meta-analysis, there was no need for handling missing data or imputing values. All studies provided sufficient data to address the research questions and were included without the need for exclusions due to incomplete information.

Quality assessment

Two authors (RE and DRH) individually evaluated the quality of each included article, and any disagreements were resolved through discussion. The National Institutes of Health (NIH) quality assessment tool for case series and controlled intervention studies was employed in this process [24]. In brief, the nine items for uncontrolled studies are the clarity of the study question, description of the study population, consecutive case inclusion, comparability of subjects, clear intervention description, well-defined outcome measures, adequate follow-up duration, appropriate statistical methods, and a clear description of results. For controlled intervention studies, there are 14 domains which are study randomization, concealment of allocation of treatment, blinding status for participants, providers, and assessors of outcomes, baseline group similarity, drop-out rates, intervention adherence, avoidance of other interventions, the use of valid outcome measures, sample size sufficiency and pre-specification of outcomes and subgroups were considered. The overall quality was categorized as poor, fair, and good for scores 0–3, 4–6, and 7–9 for uncontrolled studies and for scores 0–6, 7–10, 11–14 for controlled studies, respectively.

Study selection

The systematic search initially identified a total of 1507 articles. After removing duplicated results, 1130 studies underwent screening. In this step 1122 did not meet the eligibility criteria and were excluded. No studies were excluded in the full-text review. No other eligible studies were found in backward/forward citation searching and searching of Google Scholar. Finally, eight studies included in this systematic review [2, 25–31] (Fig 1).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Study selection process.

https://doi.org/10.1371/journal.pone.0311731.g001

Characteristics of studies

Three studies were carried out in the United States [28–30], one study in each of the countries Poland [2], China [31], India [27], Romania [26], and Spain [25]. The studies were published between 2017 [29] and 2022 [26]. Among them, five studies were pilot trials [2, 25, 27, 29, 31]. The overall participant count was 196, and their ages ranged from 2 to 72 years. They were predominantly males (n = 136, 70%). Overall, there were 55 patients with HIV infection/acquired immunodeficiency syndrome (AIDS) [25, 28–30], 47 with chronic hepatitis B infection [27, 31], 86 with COVID-19 and C. difficile coinfection [26] and eight with cytomegalovirus colitis [2]. Out of these, 111 patients received FMT and the others were as control subjects. The duration of follow-up post-FMT ranged from eight [26] to 48 weeks [25] (Table 1).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Study and patient characteristics.

https://doi.org/10.1371/journal.pone.0311731.t001

Three studies administered FMT by colonoscopy [26, 29, 30], two via gastroscopy [27, 31], one by nasogastric tube [2], and two orally [25, 28]. FMT dosages and schedules varied. Some studies used 250 or 500 ml of stool suspensions introduced via colonoscopy [26, 30], while another study used capsules implying the delivery of 30 g of stool [25]. The frequency of administration varied from single [26] to multiple sessions [27], with intervals ranging from daily [2] to every four weeks [27, 31] (S2 Table).

Efficacy

The detailed outcome results of each study are shown in Table 2.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. The outcome results of the studies.

https://doi.org/10.1371/journal.pone.0311731.t002

Viral response

Two studies evaluated the HBeAg clearance [27, 31]. In the study by Chauhan et al., two (16.7%) patients in the FMT group had HBeAg clearance compared to none in the antiviral therapy (AVT) group (p = 0.19). Among these patients, HBsAg clearance was not observed in neither the FMT nor AVT arms [27]. Ren et al. reported HBeAg clearance in three participants in the FMT arm, while none of the participants in the control group achieved HBeAg clearance (p<0.01). Additionally, hepatitis B virus DNA levels were detected at four weeks after each FMT and remained under the lower limit of detection [31]. In the article by Utay et al., two participants maintained low HIV RNA levels throughout the study period, while others developed detectable HIV RNA levels [28]. Karolewska-Bochenek et al. reported that seven out of eight patients turned negative for cytomegalovirus (CMV) tissue DNA polymerase chain reaction (PCR) following a two-week FMT course and one patient who did not initially respond to FMT saw eradication of CMV after a repeated course of FMT [2].

Clinical outcomes

Karolewska-Bochenek and colleagues observed a clinical response, defined as a reduction of the Pediatric Ulcerative Colitis Activity Index (PUCAI) score by at least 20 points, in 37.5% (three out of eight). Similarly, an equivalent percentage attained clinical remission (PUCAI score of less than 10 points). Additionally, an improvement in the Mayo endoscopic score was evident in 75% of the subjects (six out of eight) [2]. Boicean et al. showed that there was only one case of C. difficile infection (CDI) relapse (2.17% relapse rate). Among the 46 patients co-infected with SARS-CoV-2 and C. difficile who underwent FMT, there were only 17 cases of recurrence in the control group, resulting in a relapse rate of 42.5% (p<0.05) [26].

Markers of immunocompetence

Utay et al. found no significant changes in CD4+ T cell counts between weeks 0 and 6 (p = 0.56), or from weeks 6 to 26 (p = 0.13). Likewise, there were no significant alterations in CD4+/CD8+ ratios during the periods of weeks 0 to 6 (p = 0.31) and weeks 6 to 26 (p = 0.22). Although CD8+ T cell counts exhibited a tendency to decrease from weeks 0 to 6 (p = 0.06), there was no significant change between weeks 6 and 26 (p = 0.56) [28]. Serrano-Villar and colleagues assessed circulating CD4+ and CD8+ T-cell counts, the CD4/CD8 ratio, immune activation markers (such as the percentage of HLADR+CD38 on CD8+ T cells), senescence markers (including CD28- and PD-1+ on CD8+ T cells), and plasma markers of inflammation and bacterial translocation. However, no discernible trends between groups over time were observed [25].

Markers of inflammation and gut damage

Utay et al. noted only zonulin, a marker of gut permeability, increased significantly between weeks 0 and 6 (p = 0.03) [28]. Karolewska-Bochenek et al. reported fecal calprotectin significantly reduced in three patients and C-reactive protein (CRP) normalized in two patients six weeks after FMT [2]. Serrano-Villar et al. found no significant trends in plasma markers of inflammation and bacterial translocation between groups, although an early 0.5-fold-decline in the intestinal fatty-acid binding protein (IFABP) was noted in FMT group than controls (p = 0.06) which was statistically significant at week 4 (p = 0.04) and until week 48 (p = 0.01) [25]. Boicean et al. recorded a statistically significant improvement in CRP and white blood cell (WBC) count after FMT (p<0.05) [26]. While, Vujkovic-Cvijin et al. observed no significant changes in markers of HIV-associated inflammation following FMT (p = 0.07) [29].

Microbiota profile and engraftment

SahBandar and colleagues emphasized variations in the relative abundances of all bacterial species from pre-fecal microbiota transplantation (FMT) to 24 weeks post-FMT. Principle component analysis did not show distinct clustering between microbial abundances before and after FMT. Interestingly, Fusobacteria, absent in the fecal bacteria preparation from FMT HIV-uninfected donors, was present in the majority of HIV-infected recipients. Additionally, some recipients of FMT who underwent post-treatment antibiotic therapy retained detectable Fusobacteria [30]. In the study by Utay et al., mean alpha diversity showed no significant change over weeks 0–6 (p = 0.29) and 6–26 (p = 0.21). Four participants (67%) saw an increase in alpha diversity initially, especially those with the lowest at week 0. By week 26, alpha diversity in three of these participants resembled week 0 more than week 6. At week 6, three participants (50%) showed a shift in their microbiome distribution towards the donors’ distribution but it was reversed by week 26. No significant changes were observed in specific microbial patterns or phage distribution. One long-term HIV patient experienced temporary improvement in constipation with notable microbiome shifts, but symptoms and microbial composition reverted by week 26 [28]. In the study by Vujkovic-Cvijin et al., FMT recipients showed a significant increase in similarity between their fecal microbiota and that of the donor, whereas control subjects did not show any significant change in their microbial compositional similarity to that of the donor microbial community. The compositional relatedness between donors and recipients were significant at weeks 2, 4 (p<0.01) and 8 (p = 0.04) following FMT. After FMT, there was increased proportions of shared operational taxonomic units between donor and recipient microbiota (p = 0.002), without any significant changes in specific microbial genera after adjusting for false discovery rates. However, the recipients showed an increasing presence of Faecalibacterium, Rikenellaceae, and a decrease in Erysipelotrichaceae post-FMT [29]. Serrano-Villar et al. discovered that FMT led to significant enhancements in alpha diversity when compared to individuals in the placebo arm up to week 6 (p = 0.01) and from weeks 7 to 48 (p = 0.04). Moreover, FMT induced a mild and temporary integration of the donor’s microbiota, with the potential for greater integration in cases with recent antibiotic use prior to FMT. The families Lachnospiraceae and Ruminococcaceae, which are typically diminished in HIV-infected patients, displayed more pronounced integration over time [25].

Safety

Seven studies reported data regarding the adverse events [2, 25–29, 31]. None of the studies reported serious adverse events developed by the patients receiving FMT. However, the reported adverse events included mild abdominal pain [2, 28] and distension [25], nausea [2, 28], diarrhea and flatulence [25], and bloating [28]. According to Chauhan et al., only one patient experienced significant abdominal pain, requiring hospitalization [27]. In the study by Boicean et al., over 91% (42 out of 46) of FMT patients reported no abdominal pain post-treatment, versus 82.5% (33 out of 40) in the control group who experienced persistent moderate pain (p<0.05) [26] (Table 3).

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Reported adverse events among the included studies.

https://doi.org/10.1371/journal.pone.0311731.t003

Quality assessment

Uncontrolled studies met most of the quality criteria such as clear definition of study question, population, intervention, outcome measures, and results. With eight scores, they had good qualities. However, the consecutive case selection in the studies by Utay et al. and SahBandar et al. was not appropriate and in the study by Karolewska-Bochenek et al. the statistical methods were not described [2, 28, 30] (S3 Table).

The total scores of controlled interventional studies ranged from 7 to 13 (out of 14) with a mean score as 8.4. Studies by Boicean et al., Chauhan et al., Vujkovic-Cvijin et al. and Ren et al. had 7, 7, 7, and 8 scores respectively, indicating a fair quality [26, 27, 29, 31]. None of these studies received a score for randomization, treatment allocation concealment, and blinding of study participants and providers. The study by Serrano-Villar et al. received a score of 13 and met all of the assessed criteria except for reporting the sufficiency of the sample size [25] (S4 Table).

Main findings

Our findings indicated that FMT had the potential to trigger clinical response, achieve remission, and eliminate CMV in individuals with CMV colitis. Furthermore, FMT led to a significantly lower relapse rate of CDI in patients co-infected with COVID-19 and C. difficile. HBeAg clearance was also achieved following FMT, even in patients with persistent HBeAg positivity after long-term AVT. FMT was found to increase zonulin levels and contribute to an early decline in IFABP, along with normalization of inflammatory markers in co-infected patients with COVID-19 and C. difficile. Additionally, partial engraftment of the donor microbiome and increases in alpha diversity were observed in participants after FMT. In terms of safety, there were no serious adverse effects.

Clinical outcomes

FMT appears to be a viable and safe therapeutic choice for CMV colitis in pediatric patients with ulcerative colitis (UC). Similarly, Hsu et al. showed that clinical response one month after FMT was observed in 58.8% of pediatric IBD patients, with 64.7% achieving clinical remission and 44.1% attaining both clinical response and remission. The findings suggested that FMT exhibits enhanced safety and efficacy in the pediatric population when contrasted with the adult population [32]. In accordance with our results, Tun et al. found that FMT is an effective and safe therapy for pediatric and adolescent patients with CDI, demonstrating a pooled success rate of 86% (95% confidence interval (CI): 77–95%; p<0.001) across the overall cohort [33].

Additionally, FMT demonstrated potential in safely and efficiently treating CDI in patients with co-existing COVID-19, resulting a significantly lower CDI relapse rate. Similarly, another study noted that relapse occurred in only one individual five weeks post-FMT, while 98% of 54 participants exhibited clearance of C. difficile from their feces at 4–8 weeks after FMT, without any recurrences. There was a notable decrease in symptoms such as abdominal pain, rectal bleeding, and diarrhea. The outcomes support the idea of considering FMT as a preferable initial treatment for C. difficile, rather than delaying until multiple recurrences occur [34]. Furthermore, it has been demonstrated that FMT is more effective than vancomycin regimen for the treatment of recurrent CDI [35]. In this context, a clinical trial demonstrated that 81% of patients experiencing recurrent CDI achieved resolution of C. difficile-associated disease after a single session of FMT, in contrast to 31% of patients treated with vancomycin alone (p<0.001) [36]. In the same way with our findings, no recurrent C. difficile was observed during 16 months follow-up in all successfully treated patients, demonstrating an improvement of dysbiosis through FMT that prevents a repeated overgrowth of C. difficile [37]. Also, El-Salhy et al. evaluated the long-term effects of FMT in patients with irritable bowel syndrome (IBS). They included 77 patients with IBS who had responded to FMT and found that the patients had experienced improvement in abdominal symptoms and quality of life up to one year, as compared with three months following FMT. The patients had also comprehensive changes in the fecal bacterial profile and short-chain fatty acids [38].

Viral response

FMT appears to have potential benefits for HBeAg clearance in chronic hepatitis B patients. Studies indicated that both spontaneous and treatment-induced HBeAg seroconversion were associated with decreased rates of disease progression towards cirrhosis and hepatocellular carcinoma, as well as improved survival rates [39, 40]. For patients with HBeAg-positive chronic hepatitis B, HBeAg seroconversion is a prerequisite for a definite course of AVT [41]. Despite many years of AVT with entecavir and tenofovir, only a small group of patients achieve HBeAg clearance or seroconversion. Accordingly, entecavir resulted in HBeAg loss in 53% of patients [42] and tenofovir treatment led to HBeAg loss in 49% and HBeAg seroconversion in 40% of patients five years after treatment [43]. Xie et al. found a gradual decline in serum HBsAg after each time of FMT in HBeAg negative chronic hepatitis B patients who remained HBsAg positive after more than one year of entecavir antiviral therapy, indicating that gut microbiota may be a new target for the treatment of HBeAg negative chronic hepatitis B [44].

FMT also seems to be an effective and safe treatment option for CMV colitis in patients with UC, as indicated by the eradication of CMV in the patients. During the last decade, the involvement of CMV in IBD flare-ups has been highly debated. While some authors advocate for an active role of CMV in inflammatory flares, others support the concept of the innocent bystander, suggesting that the virus does not contribute to the disease’s progression [45]. Although some studies indicated that AVT with ganciclovir or foscarnet improved outcomes in in UC flare with CMV infection [46–53], others found no benefits from these treatments [54, 55]. Given that both C. difficile and CMV infections can potentially impact the advancement of IBD, it is pertinent to examine studies investigating the application of FMT in C. difficile-infected patients with IBD. In the research conducted by Hourigan et al., all five participants (aged 10–17 years, four diagnosed with Crohn’s disease and one with UC) experiencing recurrent CDI witnessed the resolution of symptoms within three days following a single FMT session. Furthermore, 12–20 weeks after FMT, all patients tested negative for C. difficile toxin B PCR [56]. In the research conducted by Russel et al., FMT administration resulted in the clearance of CDI in two out of three children with both CDI and IBD [57]. The outcomes of these studies are encouraging, indicating that FMT, by restoring gut microbiota, may represent a highly promising treatment for infections arising due to gut dysbiosis. However, CMV and C. difficile are distinct pathogens and therefore, findings related to C. difficile cannot be directly applied to CMV.

Markers of immunocompetence

We found that FMT did not modify CD4+ and CD8+ levels and CD4+/CD8+ ratio and there was also no change in the immune activation and senescence markers. While Serrano-Villar et al. chose three donors with comparable microbiota profiles to investigate diverse donor effects, the study by Utay et al. involved each recipient receiving FMT from a single donor. Additionally, the potential benefits might have been missed due to the small sample sizes, since Serrano-Villar et al. and Utay et al. performed FMT in 14 and six patients, respectively. Studies focusing on CD4+ T cell depletion and its direct impact on immune deficiency in HIV-infected patients mostly used peripheral blood samples [58–60]. Yet, it is in the gastrointestinal mucosa where the most significant and earliest reduction of CD4+ T cells is often observed, especially during the acute phase of HIV infection [61]. Given that an estimated 60% of CD4+ T cells are found in gut-associated lymphoid tissue and the observed incomplete restoration of these cell populations and intestinal microbial composition even with highly active antiretroviral therapy, the human intestine warrants significant focus in HIV research [62].

Markers of inflammation and gut damage

Even with the use of antiretroviral therapy, individuals with HIV still face an elevated risk health complications, likely attributable to persistent inflammation [63]. An influential factor contributing to this inflammation is the heightened movement of microbial products through an permeable intestinal barrier [64]. Despite prolonged treatment, the levels of biomarkers indicating intestinal permeability, microbial translocation, and systemic inflammation continue to be elevated. Importantly, these biomarkers serve as predictive indicators for morbidity and mortality [65–67]. The elevated levels of zonulin in HIV infected patients may indicate more damage or turnover of tight junctions, potentially aligning with localized inflammation. On the other side, this rise could also suggest improved integrity of the gut barrier and reduced inhibition of zonulin production.

IFABP, a biomarker signifying intestinal damage, autonomously forecasts mortality in individuals undergoing HIV treatment [65, 67]. The early decline of IFABP among HIV infected patients points to FMT orally for acute HIV infection due to the disruption in gut immune system [68]. However, the absence of noticeable improvements in other biomarkers that represent various chronic inflammation pathways (sCD14, sCD164, sTNFR-II, IP-10, and D-dimer) suggests that the microbiome might not be directly connected with these inflammation pathways, or alternatively, more significant changes may be required to affect these biomarkers.

We found significant reduction of CRP, WBC count and fecal calprotectin following FMT. In the study by Wang et al., the serum concentrations of IL-1Ra, IL-6, Interferon gamma-induced protein 10 (IP-10), ENA-78, vascular cell adhesion molecule 1 (VCAM-1), and granulocyte colony-stimulating factor significantly decreased after FMT in patients with active UC (P<0.05). Furthermore, the serum concentrations of IL-6, IL-1Ra, IP-10, and VCAM-1 demonstrated significant positive correlations with both CRP and erythrocyte sedimentation rate [69]. Considering the pronounced inflammatory response in COVID-19 and CDI infections, with elevated levels of IL-6, TNF-α, IL-1β, and IP-10, as well as the reported dysbiosis [70–72], FMT could play a crucial role in the treatment of CDI or more severe conditions like COVID-19 and C. difficile co-infections.

Microbiota profile and engraftment

In patients with HIV, there is greater abundance of Fusobacteria in subgingival plaques [73] and anal swabs [74] compared to HIV-uninfected group.

Previous studies proposed that incorporating dietary supplements containing probiotics [75] or undergoing FMT [76] could serve as a beneficial complementary approach to restore the gut microbial environment preceding immune checkpoint blockade. However, their data showed that FMT may be insufficient. We found that the gut colonization of Fusobacteria in HIV infection is persistent and may influence anti-HIV T-cell immunity to PD-1 or TIGIT blockade. Blocking these pathways may synergistically enhance the functions of HIV-specific CD8+ T cells [77].

Full engraftment of FMT was not observed, aligning with similar observations in conditions such as CDI and IBD. In these cases, FMT does not result in a complete substitution of the recipient’s microbiome with that of the donor. However, notable clinical improvements and alterations in the microbiome have been noted with FMT [78–80].

We found a modest shift towards donor microbiome profiles in treated HIV-infected patients whereas it was significantly smaller in magnitude than in patients with CDI, which may be due to the vastly decreased alpha diversity observed in patients with CDI. The concept of "resilience in diversity" suggests that a diverse microbial community has greater capacity to restore its composition after stress compared to a community with less diversity [81]. Therefore, the uniform and phylogenetically constrained community found in cases of recurrent CDI might be more prone to successful colonization by a diverse donor microbiome. Considering the effective engraftment observed in CDI, adopting a protocol that mimics C. difficile treatment, which include antibiotic conditioning prior to FMT is recommended.

Safety

Overall, FMT was safe and well-tolerated in viral diseases that evaluated in our study. No serious adverse events were reported by the patients receiving FMT. However mild symptoms like abdominal pain, nausea, diarrhea, and flatulence were observed. FMT is also a safe, effective, and well-tolerated therapy for patients with IBD, as the safety analysis in the study by Tan et al. revealed that most adverse events following FMT were mild and self-resolving [82]. Furthermore, a one-year monitoring of individuals diagnosed with IBS who underwent FMT revealed that adverse effects such as abdominal pain, diarrhea, and constipation were mild and resolved on their own [83]. Lee et al. evaluated the long-term efficacy and safety of FMT for recurrent CDI. The study involved twenty-three patients who received FMT via a retention enema for recurrent CDI within the period from 2008 to 2012, and a follow-up questionnaire was completed by the patients four to eight years after the last FMT. They found that FMT produced no significant long-term adverse events, and about 30% of participants experienced improvements in pre-existing medical conditions, including IBD [84]. Also, Fang et al. conducted a study to assess the long-term safety and efficacy of monotherapy with a single fresh FMT for recurrent ulcerative colitis. The study sample was composed of 20 patients, with 10 participants in each the FMT group and the control group and the mean length of follow-up was 19.1 ± 10.1 months. They found that participants tolerated FMT treatment, and no adverse events were reported during the follow-up, except for one treatment-related Epstein–Barr virus infection which occurred within two weeks following FMT [85]. Marcella et al. conducted a review of 129 studies, including 4241 individuals, among whom the majority of patients had recurrent or refractory CDI, followed by UC, Crohn’s disease, and IBS. The commonly noted adverse effects associated with FMT included diarrhea (10%) and abdominal discomfort/pain/cramping (7%). Serious adverse events related to FMT, such as infections and fatalities, were documented in 1.4% of individuals who underwent the procedure [86]. Additionally, FMT has been documented as a safe procedure, with minimal reported adverse effects, even among patients with potential immunocompromise [87, 88].

Methods of FMT administration

FMT can be administered via both the upper and lower gastrointestinal tract. Some research indicated that employing upper gastrointestinal administration necessitates careful consideration of potential adverse events, such as the occurrence of vomiting [89], and aspiration pneumonia [90]. Moreover, there are also risks when utilizing lower gastrointestinal tract administration methods. The advantages of colonoscopic administration include the ability to visually assess the colon, directly target specific areas, and deliver larger volumes of microbiota. Nevertheless, this approach carries risks associated with the use of anesthesia and the potential for bowel perforation [91]. Capsule-based FMT, on the other hand, has shown comparable effectiveness to colonoscopy in treating recurrent CDI and provides the benefits of convenience and patient satisfaction [92]. Besides efficacy, key factors such as patient comfort and compliance, cost-effectiveness, invasiveness, risk of aspiration and infection, the necessity for multiple drugs, and relapse rate inform physicians’ selection of the administration route [93].

Strengths and limitations

To best of our knowledge, this is the first systematic review evaluating the efficacy and safety of FMT across various viral diseases. However, we had several limitations. The studies included in the review showed considerable variability in FMT protocols and dosages, reflecting diverse approaches across different research settings. Such variations could potentially influence the outcomes and comparability of the results. In addition, the number of participants involved in the studies was generally small which can significantly impact the generalizability and reliability of the findings, lead to reduced statistical power and make it difficult to detect differences between treatment and control groups. Furthermore, such limited samples may not represent the full range of patient diversity and heterogeneity, which may make the study findings less generalizable and applicable to patients with varying characteristics, comorbidities, or demographics. Moreover, a small sample size might not be capable of detecting subtle treatment differences or rare adverse events, which can make it challenging to evaluate the true efficacy and safety of the intervention. Additionally, the number of studies and patients involved in evaluating the effectiveness of FMT for a particular viral disease was also relatively low. Therefore, the results of our study should be interpreted with caution when it comes to drawing conclusions on the efficacy of FMT in treating specific viral infections. In addition, it is important to note that out of the 196 participants included in the study, 86 of them were diagnosed with COVID-19, which is a dominant viral infection. This means that the clinical course of COVID-19 may differ significantly from chronic virus infections such as chronic hepatitis B or CMV colitis. Therefore, findings of the study may not be generalizable across all types of viruses and viral infections. Also some studies lacked a control group, making it challenging to attribute the observed improvements solely to FMT. The age range of participants also varied, including both pediatric and adult populations, which adds another layer of complexity when interpreting the results and their applicability to broader patient demographics. These restrictions limited us about conducting meta-analysis to quantitatively report the pooled effect size. Moreover, reported adverse events can differ across studies, mainly due to differences in definitions and reporting methodologies. These disparities can make it challenging to compare results consistently across studies, leading to possible inconsistency in safety assessments. Also, the included studies mainly assessed the short-to-mid-term outcomes of FMT and the long-term effects of FMT on viral diseases and gut microbiota composition remain unclear.