About

Brendan J. Kelly, MD, MS, is an Assistant Professor of Medicine in the Department of Medicine (Infectious Diseases) at the Hospital of the University of Pennsylvania. He serves as an Attending Physician at the Hospital of the University of Pennsylvania and Penn Presbyterian Medical Center, and is the Director of the Fecal Microbiota Transplantation Program at the Hospital of the University of Pennsylvania. Dr. Kelly is also a Senior Scholar at the Center for Clinical Epidemiology and Biostatistics and the Director of the Antimicrobial Resistance and Healthcare Epidemiology Study collaborative (ARES) Scientific Laboratory at the University of Pennsylvania. His clinical expertise includes respiratory tract infections, Clostridium difficile infection, and HIV. His research focuses on microbial ecology, next-generation sequencing, the respiratory tract microbiome, ventilator-associated infections, antimicrobial resistance, and Clostridium difficile infection. Dr. Kelly holds degrees from Harvard College and Columbia University, and completed his MSCE at the University of Pennsylvania. His work involves advancing understanding and treatment of infectious diseases through microbiome research and epidemiological studies.

Research topics

• Medicine
• Biology
• Virology
• Emergency medicine
• Microbiology
• Intensive care medicine
• Zoology
• Genetics
• Evolutionary biology
• Internal medicine
• Immunology

Selected publications

• Fecal Microbiota Transplant for Multidrug Resistance—No Benefit Without Disruption?

  JAMA Internal Medicine · 2026-04-20

  article1st authorCorresponding
  PublisherDOI

• Screening and targeted prophylaxis for Clostridioides difficile infection

  Clinical Microbiology and Infection · 2025-08-13 · 2 citations

  articleOpen accessSenior author

  OBJECTIVE: Patients receiving immunosuppression for oncology treatment and solid organ transplantation are at high risk for developing hospital-onset Clostridioides difficile infection (HO-CDI). We studied the impact of a prophylactic enteral vancomycin intervention on the reduction of HO-CDI during high-risk inpatient admissions among immunocompromised patients. METHODS: The screening and targeted prophylaxis (STOP) intervention was implemented over a 2-year period. Patients admitted for solid organ transplant, autologous stem cell transplant, chimeric antigen receptor T-cell, or treatment for leukaemia were screened for colonization with C. difficile. Colonized patients were placed in contact isolation and were advised to start anti-C.difficile antibiotic prophylaxis. To assess the effect of this intervention on both C.difficile testing and incident HO-CDI, comparison was made to historical controls in the 2 years before implementation of the STOP intervention, both unweighted and weighted by treatment category and separately by treatment category and severity of illness. RESULTS: From November 2021 to December 2023, 696 patients were screened for C difficile, among whom 11.1% (77/696) were found to be colonized and received the prophylactic intervention. Compared with treatment-weighted controls, the odds of HO-CDI were significantly lower in the intervention group (6/696 [0.88%] vs. 81/1450 [5.6%]; odds ratio [OR], 0.15; 95% Credible Interval [CrI], 0.06-0.30). Significant reductions in 90-day CDI (OR 0.40, 95% CrI 0.25-0.64), stool output (incidence rate ratio, 0.84; 95% CrI, 0.77-0.92), and length of stay (-2.5 days; 95% CrI, -3.4 to -1.5) were also observed in the intervention group. There was no detected difference in vancomycin-resistant Enterococci infection (OR, 0.77; 95% CrI, 0.33-1.75) or mortality (OR, 0.44; 95% CrI, 0.11-1.49). DISCUSSION: The STOP intervention was effective at reducing HO-CDI, length of stay, and symptoms of C difficile in this high-risk cohort.

  PublisherDOI

• Comparative performance of sponge versus flocked swabs for culture-based and metagenomic detection of microbial contamination in the healthcare environment

  Infection Control and Hospital Epidemiology · 2025-05-27 · 1 citations

  articleOpen accessSenior author

  BACKGROUND: Identifying optimal methods for sampling surfaces in the healthcare environment is critical for future research requiring the identification of multidrug-resistant organisms (MDROs) on surfaces. METHODS: We compared 2 swabbing methods, use of a flocked swab versus a sponge-stick, for recovery of MDROs by both culture and recovery of bacterial DNA via quantitative 16S polymerase chain reaction (PCR). This comparison was conducted by assessing swab performance in a longitudinal survey of MDRO contamination in hospital rooms. Additionally, a laboratory-prepared surface was also used to compare the recovery of each swab type with a matching surface area. RESULTS: from laboratory-prepared surfaces, although the performance of flocked swabs improved when premoistened. In contrast, recovery of bacterial DNA via quantitative 16S PCR was greater with flocked swabs by an average of 3 log copies per specimen. CONCLUSIONS: The optimal swabbing method of environmental surfaces differs by method of analysis. Sponge-sticks were superior to flocked swabs for culture-based detection of bacteria but inferior for recovery of bacterial DNA.

  PublisherOA PDFDOI

• Amplicon-Based Sequencing is Able to Accurately Detect Clostridioides Difficile

  SSRN Electronic Journal · 2025-01-01

  preprintOpen accessSenior author
  PublisherDOI

• Microbiome Therapeutics for Clostridioides difficile Infection

  Infectious Disease Clinics of North America · 2025-08-11 · 2 citations

  reviewOpen access
  PublisherOA PDFDOI

• Escape Velocity—the Launch of Microbiome Therapies

  The Journal of Infectious Diseases · 2024-03-06 · 5 citations

  articleOpen access1st author

  Food and Drug Administration approval of the first microbiome therapies represents a true expansion the treatment paradigm for Clostridioides difficile but raises new questions about the future role of fecal microbiota transplantation. The authors outline the advances in microbiome therapeutic development that have addressed fecal microbiota transplantation's (FMT's) inherent limitations of safety and scalability. The authors also suggest that as microbiome therapeutic development continues for other indications, FMT will likely remain a necessary model of human microbiota dynamics for translational research.

  PublisherOA PDFDOI

• Resetting the environmental reservoir; evaluating the impact of a new hospital building on Clostridioides difficile infection

  Antimicrobial Stewardship & Healthcare Epidemiology · 2024-07-01

  articleOpen access

  Background: Prior research has implicated contaminated surfaces in the transmission of Clostridioides difficile within the hospital. To reduce the risk of transmission, enhanced environmental hygiene is performed in rooms of patients with known C.difficile infection (CDI). We wished to evaluate the residual impact of environmental surfaces on hospital-onset CDI (HO-CDI) by comparing HO-CDI rates before and after the opening of a new 504-bed hospital building, HUP Pavilion (PAV). We hypothesized that we would observe a reduction in HO-CDI after opening of PAV due to a reduced burden of C.difficile spores in the environment. Methods: We included NHSN reported HO-CDI rates for 28 months prior and 24 months after opening of PAV. Upon opening, patients were divided between the old building (HUP) and PAV. We included all patient units before and after opening. We created hierarchical models of HO-CDI rates using Stan Hamiltonian Monte Carlo (HMC) version 2.30.1, via the “cmdstanr” and “brms” packages with a GAM smooth function by month and intervention period with default, weakly-informative priors. Results: At baseline, there was an average of approximately 20,100 patient days per month, subsequently divided between HUP and PAV (mean 10,100 and 12,100 patient days per month). After opening of PAV, we observed a reduced HO-CDI rate (mean 0.21 vs 0.31 per 1000 patient days, P=0.01). When comparing the two specific buildings after opening of PAV, there was a greater reduction noticed in the old building (HUP) as compared to the new building (PAV) (0.12 vs 0.29 per 1000 patient days) (Figure 1). The predicted contrast in HO-CDI rate (Figure 2), shows no immediate change in HO-CDI after opening, however a sustained reduction estimated at 0.1 HO-CDI events per 1000 patient days for the duration of follow-up. Conclusions: We observed a reduction in HO-CDI rates after the opening of a new hospital building. The difference in HO-CDI rates between hospital buildings after the move is likely due to the concentration of high-risk patient cohorts within this building. Our findings suggests that there remains an opportunity to reduce HO-CDI through environmental hygiene. However, it is possible that other factors beyond surface environment contributed to an observed reduction in HO-CDI, including other concurrent infection control interventions that focused on smaller populations within the hospital. In future work we will investigate the durability of this observed effect with additional analyses including patient-level risk for HO-CDI.

  PublisherOA PDFDOI

• Metagenomic Profiling Reveals Post-Kidney Transplant Urinary Microbiome Change

  SSRN Electronic Journal · 2024-01-01

  preprintOpen accessSenior author
  PublisherDOI

• Microbiome-Based Diagnostics for Disease: Where Are We Now and Where Are We Headed?

  Clinical Chemistry · 2024-06-01 · 2 citations

  articleOpen access
  PublisherOA PDFDOI

• Comparison of SARS-CoV-2 variants of concern in primary human nasal cultures demonstrates Delta as most cytopathic and Omicron as fastest replicating

  mBio · 2024-03-13 · 42 citations

  articleOpen access

  The SARS-CoV-2 pandemic was marked with emerging viral variants, some of which were designated as variants of concern (VOCs) due to selection and rapid circulation in the human population. Here, we elucidate functional features of each VOC linked to variations in replication rate. Patient-derived primary nasal cultures grown at air-liquid interface were used to model upper respiratory infection and compared to cell lines derived from human lung epithelia. All VOCs replicated to higher titers than the ancestral virus, suggesting a selection for replication efficiency. In primary nasal cultures, Omicron replicated to the highest titers at early time points, followed by Delta, paralleling comparative studies of population sampling. All SARS-CoV-2 viruses entered the cell primarily via a transmembrane serine protease 2 (TMPRSS2)-dependent pathway, and Omicron was more likely to use an endosomal route of entry. All VOCs activated and overcame dsRNA-induced cellular responses, including interferon (IFN) signaling, oligoadenylate ribonuclease L degradation, and protein kinase R activation. Among the VOCs, Omicron infection induced expression of the most IFN and IFN-stimulated genes. Infections in nasal cultures resulted in cellular damage, including a compromise of cell barrier integrity and loss of nasal cilia and ciliary beating function, especially during Delta infection. Overall, Omicron was optimized for replication in the upper respiratory tract and least favorable in the lower respiratory cell line, and Delta was the most cytopathic for both upper and lower respiratory cells. Our findings highlight the functional differences among VOCs at the cellular level and imply distinct mechanisms of pathogenesis in infected individuals. IMPORTANCE: Comparative analysis of infections by SARS-CoV-2 ancestral virus and variants of concern, including Alpha, Beta, Delta, and Omicron, indicated that variants were selected for efficiency in replication. In infections of patient-derived primary nasal cultures grown at air-liquid interface to model upper respiratory infection, Omicron reached the highest titers at early time points, a finding that was confirmed by parallel population sampling studies. While all infections overcame dsRNA-mediated host responses, infections with Omicron induced the strongest interferon and interferon-stimulated gene response. In both primary nasal cultures and lower respiratory cell line, infections by Delta were most damaging to the cells as indicated by syncytia formation, loss of cell barrier integrity, and nasal ciliary function.

  PublisherDOI

Recent grants

• Longitudinal analysis of respiratory tract microbiome change and sequence-based infection diagnosis during mechanical ventilation

  NIH · $965k · 2016–2022

Frequent coauthors

• Ronald G. Collman

  Pulmonary and Allergy Associates

  49 shared

• Frederic D. Bushman

  35 shared

• Kyle G. Rodino

  University of Pennsylvania

  22 shared

• Carolyn C. Cannuscio

  University of Pennsylvania

  20 shared

• Katherine M. Strelau

  University of Pennsylvania

  20 shared

• Matthew J. Ziegler

  19 shared

• Andrew D. Marques

  University of Pennsylvania

  19 shared

• Ebbing Lautenbach

  19 shared

Labs

• Brendan J Kelly LaboratoryPI

Awards & honors

• Screening and targeted prophylaxis for Clostridioides diffic…
• Escape Velocity-the Launch of Microbiome Therapies. J Infect…
• Amplicon-Based Sequencing is Able to Accurately Detect Clost…
• Microbiome Therapeutics for Clostridioides difficile Infecti…
• Screening and targeted prophylaxis for Clostridioides diffic…