We provide a comprehensive assessment of the epidemiology and morbidity of infectious diarrhoea post-alloHSCT, with nearly 1 in 4 recipients experiencing an infection in the first year post-transplant (7, 21), and such patients experiencing inferior GRFS and being over four times more likely to develop severe lower GI GVHD. Our study is the first to show that GVHD risk varies according to the diarrhoeal pathogen responsible.
While rarely fatal in isolation, infectious diarrhoea affects one-quarter to one-third of patients' early post-alloHSCT and is associated with increased non-relapse mortality (8, 21). Since the 1980s, C. difficile has prevailed as the principle cause of infectious diarrhoea post-transplant, affecting 12–27% of patients (7, 8, 11, 12, 21). CDI primarily occurs in the pre-and early post-engraftment periods (8, 12, 14), a time when patients experience neutropenia, antimicrobial exposure and chemotherapy-related disruptions of the enteric mucosal barrier. Recalcitrant and protracted CDI are common in the transplant population who often receive antibiotics and corticosteroids (22, 23), with one-quarter of patients in our cohort experiencing a recurrence. Risk factors for CDI post-alloHSCT have been inconsistently reported with poor reproducibility but include; the type of conditioning used, the number of prior lines of chemotherapy and the receipt of broad-spectrum antibiotics (7, 8, 12–14, 24). None of these risk factors were identified in this review. However, data on the use of antibiotics was not collected.
Besides CDI, bacteria and viruses are responsible for a significant proportion of infectious diarrhoea post-transplant, accounting for a total of 8% of infections. While high, this is significantly less than reported in studies using multiplex PCR-based assays in which bacterial and viral pathogens each have a one-year post-alloHSCT cumulative incidence of ~ 15–20% (8, 25). While such assays can significantly augment the identification of infectious aetiologies, the pathogenic role of some pathogens, such as enteropathogenic Escherichia coli, remains to be confirmed (25). Compared to CDI, non-clostridial bacterial and viral causes typically occur late post-engraftment, potentially due to increased exposure to these pathogens in the ambulatory setting (8).
We demonstrate a strong association between early infectious diarrhoea, inferior GRFS and acute and chronic GVHD. In line with previous reports, the GI tract was the most common site of GVHD associated with infectious diarrhoea (7, 8, 12, 13). Infectious diarrhoea preceded a diagnosis of GI GVHD in most cases, supporting a formative role in GVHD pathogenesis (14). The function of damage and pathogen-associated molecular patterns, as well as a dysfunctional microbiome in the development of GVHD, provides a plausible biological explanation for this link (26–28).
Fifty years after the association between the gut microbiome and GVHD was first recognised (29), we now understand how the composition of the microbiome influences post-transplant outcomes and how it can be modulated therapeutically by faecal microbiota transplantation (FMT) (30–33). The interplay between the intestinal microbiota and the mucosal immune system is central to preventing infections (34, 35), maintaining mucosal integrity and producing immune-regulatory metabolites (36, 37). Increased bacterial diversity post-alloHSCT has been associated with reduced GVHD; specifically, an abundance of anaerobic commensals, including bacteria from the genus Blautia is associated with less frequent and severe GVHD (30, 38). AlloHSCT damages the microbiome through antibiotic and antacid exposure, intestinal inflammation from chemoradiotherapy and changes in diet (39). Infection by C. difficile and other enteric pathogens also detrimentally impacts microbial composition and function (15–18). Distinct from previous research, we confirm an association between non-clostridial bacterial forms of diarrhoea and GVHD but not viral forms of diarrhoea. The explanation for this is unclear, but it may be a consequence of the delayed onset of viral infections.
The management of infectious diarrhoea post-transplant encompasses infection control, supportive care measures, and, where appropriate, steps to preserve or restore gut microbial diversity. FMT, which refers to administering faecal matter from a donor to a recipient with the intent of modifying the composition of the recipient's microbiome, is an emerging therapy to restore microbiome diversity in alloHSCT recipients. FMT has been introduced for allograft recipients as a therapeutic intervention for GVHD and recurrent CDI, as well as to prevent the dysbiosis inherent to transplantation and thereby reduce microbiome-related complications such as CDI, GVHD or gut colonisation with antimicrobial-resistant bacteria (33, 40–43).
The role of FMT as a treatment for recurrent CDI is established, and reports of its use in alloHSCT recipients suggest it is safe and effective (43, 44). Encouraging data exists for FMT as a treatment for steroid-resistant gut GVHD with overall response rates of 43–75%. However, this needs to be confirmed in larger prospective studies (42, 45). Despite the above, practical issues such as the optimal delivery method, use of antibiotics, and donor choice remain to be decided. FMT entails risks from procedural complications and the hazard of transmitting infection to an immunocompromised host with impaired gut wall integrity. While data endorses FMT as safe, the concern for infection means efforts to screen donor stools to exclude potentially transmittable pathogens are crucial (41).
This research is subject to all the limitations inherent to a single-centre retrospective study. As a chart review, we could not confirm that all diarrhoeal episodes met standard criteria (i.e. ≥three unformed stools/day); thus, our definition was based on submitting a stool sample for diagnostic testing. We acknowledge the significant possibility of testing bias introduced by patients undergoing microbiological testing while being investigated for GI GVHD and the potential for false positive results given the high rates of diarrhoea and testing in the peri-transplant period. Another relevant limitation is the conventional microbiological stool assays used for the majority of patients in this study, which have less sensitivity for detecting many diarrhoeal pathogens (25).
In summary, this single-centre experience illustrates that infectious diarrhoea is a frequent, early complication post-transplant which is significantly associated with the development of GVHD and inferior GRFS. Research is still needed to examine whether the prevention of infectious diarrhoea may reduce the incidence of GVHD; however, in this vulnerable population, ensuring optimal compliance with infection prevention techniques, such as; hand hygiene, isolation precautions and adequate environmental cleaning is imperative. Investigation of innovative approaches, such as FMT, which modulate the gut microbiome to treat and prevent infectious diarrhoea and GVHD, are ongoing but hold promise to improve patient outcomes.