Eduardo Henrique de Souza Oliveira

Eduardo Henrique de Souza Oliveira

Auto-generated profile, based on publicly available information

Eduardo Henrique de Souza Oliveira

Department of Oral Medicine, Infection, and Immunity Harvard School of Dental Medicine Boston Massachusetts USA

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 April 2025 in Journal of Periodontology
Verified
Journal Article
Open access
BFBalazs Feher
Balazs Feher
EOEduardo H. de Souza Oliveira
Eduardo H. de Souza Oliveira
Poliana Mendes DuartePoliana Mendes Duarte
Poliana Mendes Duarte
AWAndreas A. Werdich
Andreas A. Werdich
WGWilliam V. Giannobile
William V. Giannobile
MFMagda Feres
Magda Feres
Background: Periodontitis is among the most prevalent chronic inflammatory conditions globally, and is associated with bone resorption, tooth loss, and systemic complications. While its treatment is largely standardized, individual outcomes vary, with some patients experiencing further disease progression despite adherence. Methods: We developed a machine learning (ML) approach to predict individual outcomes 1 year post‐treatment using retrospectively assessed baseline parameters. We trained a Random Forest model on 18 demographic, clinical, microbiological, and treatment‐related features of 414 patients from randomized clinical trials (RCTs) in South America. We subsequently performed internal testing, interpretability analysis, and external testing on a second dataset of 78 patients from previous RCTs in North America and Europe exhibiting less severe disease. Results: In internal testing, the ML model achieved an area under the receiver operator characteristics curve (AUROC) of 0.93, an area under the precision‐recall curve (AUPRC) of 0.90, an F 1 ‐score of 0.82, and an out‐of‐bag score of 0.71. Relative importances were 0.42 for clinical, 0.33 for treatment‐related, 0.21 for microbiological, and 0.04 for demographic features. In external testing, the ML model achieved an AUROC of 0.76, an AUPRC of 0.69, and an F 1 ‐score of 0.71. Conclusions: Our study indicates that an ML‐based approach can assist in predicting individual responses to periodontal treatment. Prospective validation is needed for clinical application. Plain language summary: Using comprehensive data from patients with periodontitis, an inflammatory condition of the tooth‐supporting tissues, a machine learning model was trained to predict how well patients might respond to different treatments after 1 year. The model was externally tested in patient populations from 2 continents different from the training dataset. The results suggest that with further research and refinement, this tool could eventually become a valuable asset in personalizing treatment plans for improved patient outcomes.

4 June 2023 in Journal of Oral Microbiology
Verified
Journal Article
Open access
FEFelix Ellett
Felix Ellett
NKNazli I. Kacamak
Nazli I. Kacamak
CACarla Alvarez
Carla Alvarez
EOEduardo H.S. Oliveira
Eduardo H.S. Oliveira
HHHatice Hasturk
Hatice Hasturk
DIDaniel Irimia
Daniel Irimia
Recent studies uncovered that Fusobacterium nucleatum (Fn), a common, opportunistic bacterium in the oral cavity, is associated with a growing number of systemic diseases, ranging from colon cancer to Alzheimer’s disease. However, the pathological mechanisms responsible for this association are still poorly understood. Here, we leverage recent technological advances to study the interactions between Fn and neutrophils. We show that Fn survives within human neutrophils after phagocytosis. Using in vitro microfluidic devices, we determine that human neutrophils can protect and transport Fn over large distances. Moreover, we validate these observations in vivo by showing that neutrophils disseminate Fn using a zebrafish model. Our data support the emerging hypothesis that bacterial dissemination by neutrophils is a mechanistic link between oral and systemic diseases. Furthermore, our results may ultimately lead to therapeutic approaches that target specific host-bacteria interactions, including the dissemination process.

7 February 2026 in Journal of Periodontology
Verified
Journal Article
Open access
DFDaiane Fermiano
Daiane Fermiano
EOEduardo de Souza Oliveira
Eduardo de Souza Oliveira
MMMaria Josefa Mestnik
Maria Josefa Mestnik
LFLuciene C. Figueiredo
Luciene C. Figueiredo
TSTakahiko Shiba
Takahiko Shiba
MFMagda Feres
Magda Feres
Background: The aim of this study is to determine whether the timing of metronidazole (MTZ) and amoxicillin (AMX) administration, relative to scaling and root planing (SRP), influences the clinical and microbiological outcomes of periodontal treatment. Methods: In this double‐blind, placebo‐controlled, randomized clinical trial, subjects with stage III/IV periodontitis were assigned to receive SRP over 14 days with either (1) MTZ+AMX three times a day (TID) during SRP followed by placebo ( early antibiotic group) or (2) placebo during SRP followed by MTZ + AMX TID immediately after SRP ( late antibiotic group). Clinical parameters (probing depth [PD], clinical attachment level [CAL], visible plaque, bleeding on probing, suppuration) were recorded at baseline, 3 months and 1 year. Subgingival biofilm samples from nine sites per subject were individually analyzed using checkerboard DNA–DNA hybridization targeting 40 bacterial species. Bacterial correlation networks were constructed using SparCC and visualized in Cytoscape based on the bacterial abundance. Results: Sixty‐eight subjects were included in the study, with 34 per group. Both groups achieved comparable clinical improvements. Approximately 60% of participants in each group reached the primary treatment endpoint at 1 year (≤ 4 sites with PD ≥ 5 mm at 1 year; p > 0.05), and the reductions in PD and CAL did not differ significantly between groups ( p > 0.05). Microbiologically, the early antibiotic group demonstrated faster and more pronounced ecological reorganization, with reduced centrality of red and orange complex species and earlier integration of health‐associated taxa such as Actinomyces spp. In contrast, the late group showed slower microbial restructuring, with Actinomyces taxa remaining entangled with orange‐complex organisms and persistent dysbiotic clusters dominated by Fusobacterium subspecies. Conclusions: While the timing of antibiotic administration relative to SRP did not influence the overall clinical efficacy of periodontal treatment, initiating antibiotics on the first day of mechanical debridement promoted a more rapid and beneficial restructuring of the subgingival microbial community. Plain Language Summary: When treating severe gum disease, dentists often combine deep cleaning with antibiotics to help eliminate harmful bacteria. However, the best time to give these antibiotics—at the start or the end of the cleaning process—has never been tested. This study compared two approaches in 68 patients with advanced gum disease: one group received antibiotics during their cleaning sessions over 2 weeks, while the other received them immediately after cleaning finished. Both groups showed similar improvements in gum health after 1 year, with about over half achieving successful treatment outcomes. However, analyzing the bacterial communities revealed an interesting difference: patients who started antibiotics early experienced faster and more thorough changes in their mouth bacteria. Their harmful bacteria disappeared more quickly, and beneficial bacteria formed a stable, protective community sooner. In contrast, the late‐antibiotic group showed slower bacterial changes, with beneficial bacteria remaining mixed with harmful ones for longer. While both timing strategies work clinically, starting antibiotics at the beginning of treatment may help to establish a healthier, more stable bacterial community in the gums, potentially offering better long‐term protection against disease recurrence.

13 November 2026 in European Journal of Dental Education
Verified
Journal Article
Closed access
VPViviane de Oliveira Prado
Viviane de Oliveira Prado
HSHelena Justiniano Svícero
Helena Justiniano Svícero
LRLarissa Ruiz Rodrigues
Larissa Ruiz Rodrigues
JNJosé Rino Neto
José Rino Neto
EOEduardo H. de Souza Oliveira
Eduardo H. de Souza Oliveira
MFMurilo Fernando Neuppmann Feres
Murilo Fernando Neuppmann Feres

11 October 2026 in Journal of Clinical Periodontology
Verified
Journal Article
Closed access
SPSeung‐Hyun Park
Seung‐Hyun Park
DKDa‐Mi Kim
Da‐Mi Kim
SISeulbin Im
Seulbin Im
EOEduardo H. de Souza Oliveira
Eduardo H. de Souza Oliveira
ISIgnacio Sanz‐Martin
Ignacio Sanz‐Martin
JCJae-Kook Cha
Jae-Kook Cha

1 January 2020 in Revista de Odontologia da UNESP
Verified
Journal Article
Open access
PDF file
EOEduardo Henrique de Souza Oliveira
Eduardo Henrique de Souza Oliveira
MSMarcelo Yudi Sakamoto
Marcelo Yudi Sakamoto
GCGiovana Carneiro
Giovana Carneiro
VRViviane Maria Rankel
Viviane Maria Rankel
GSGeisla Mary Silva Soares
Geisla Mary Silva Soares
HSHumberto Osvaldo Schwartz-Filho
Humberto Osvaldo Schwartz-Filho
Resumo Introdução Tratamentos têm sido propostos para a peri-implantite com o objetivo de descontaminar a superfície dos implantes, removendo microrganismos que podem estar associados à doença. Objetivo O objetivo deste estudo foi avaliar a ação in vitro de diferentes métodos de aplicação de digluconato de clorexidina (CLX) na descontaminação de discos de titânio (Ti) com microtopografia e seu efeito físico-químico sobre a superfície. Material e método Vinte discos de Ti foram expostos a inóculo de Escherichia coli por 24 horas. Foram distribuídos em quatro grupos de descontaminação (n=5): 1 - um minuto de exposição à solução de CLX 0,12%; 2 - dois minutos de exposição à solução de CLX 0,12%; 3 - esfregaço durante um minuto com gel de CLX 1%; 4 - esfregaço durante um minuto com gel de CLX 2%. O produto de cada disco foi diluído e plaqueado individualmente. Após 24 horas, realizou-se contagem das unidades de colônias formadas (UFC). Resultado O grupo com o menor número de crescimento de UFC foi o grupo 4 (0,20±0,37), com apenas UFC em uma das amostras. Seguido do grupo 2 (0,40±0,73), grupo 1 (18,60±33,96). O grupo 3 apresentou as maiores quantidades de UFC (36,07±41,39). Em todas as amostras, foi possível observar uma diminuição estatisticamente significante da concentração superficial de Ti, assim como um aumento de Oxigênio. Conclusão Pode-se concluir que o uso de CLX gel a 2% em superfícies de Ti com microtopografia contaminadas com E. coli propicia a eliminação das UFC e que sua aplicação resulta em diminuição do percentual de Ti e aumento do teor de O.