An Unusual Case of Streptococcus mitis-Associated Infective Endocarditis Triggered by Upper Gastrointestinal Bleeding Secondary to Superficial Gastritis

Abstract

Background

Streptococcus (S). mitis, a viridans group streptococci (VGS) typically colonizes the oral cavity, is a recognized cause of infective endocarditis (IE) and is primarily associated with dental risk factors. A review of the literature revealed a few reported cases of IE resulting from invasion of S. mitis into the bloodstream. However, we report a case of S. mitis-associated IE originating from upper gastrointestinal bleeding (UGIB) secondary to superficial gastritis (SG) with erosion in a patient without prior cardiac or active oral pathology. To our knowledge, no such case has been reported to date.

Case Presentation

A 59-year-old man with hypertension presented to the hospital 14 days after the onset of melena. Gastroscopy revealed SG with erosions, leading to his admission for treatment. Upon admission, the patient exhibited signs of bloodstream infection and an unexplained fever. Abdominal computed tomography revealed a wedge-shaped splenic infarction. Blood cultures were positive for S. mitis. Transthoracic echocardiography revealed vegetations on the aortic valve and severe regurgitation, meeting modified Duke criteria for IE. The clinical findings included multiple Osler’s nodes, splinter hemorrhages under the nails, and Janeway lesions. The patient responded to empirical antibiotics and ultimately underwent aortic valve replacement.

Conclusion

We report a rare case of IE caused by S. mitis in a patient initially admitted for UGIB. This study suggests that the pathogenic invasion of S. mitis likely occurs through early mucosal erosions and ulcerations in the gastric lining, which facilitate bacterial entry into the bloodstream, resulting in IE. Clinicians should consider gastrointestinal mucosal injury in unexplained systemic infections and emphasize timely diagnosis and targeted therapy to mitigate IE risks.

Background

Streptococcus (S). mitis, belonging to the viridans group streptococci (VGS), is a member of the normal flora that mainly inhabits the oral cavity. However, an increasing body of evidence suggests that S. mitis plays a pathogenic role in various life-threatening infectious diseases, such as infective endocarditis (IE) Despite advances in diagnosis and treatment, IE remains associated with high morbidity and mortality, with in-hospital mortality rates ranging from 15% to 30%.¹ Traditionally, bloodstream invasion by S. mitis has been associated with dental conditions, oral procedures, and poor oral hygiene.² Recent studies, however, have indicated that the stomach can also serve as a habitat for streptococcal species, including S. mitis. This suggests that, in addition to oral sources, the gastrointestinal tract may be another potential source of S. mitis infection. Currently, there are limited reports of IE caused by S. mitis bacteremia, which have typically been linked to the aforementioned dental risk factors. Whether gastric mucosal erosion could serve as an alternative portal of entry for S. mitis remains unexplored. Here, we present a case of IE caused by S. mitis in a patient without active periodontal disease, gingival bleeding, or recent dental procedures, with evidence suggesting a possible association with UGIB and gastric erosions. We report this case along with a brief review of the relevant literature. The novelty of this presentation is discussed in detail in the Discussion section.

Case Presentation

A 59-year-old Chinese male with a two-year history of hypertension was treated with oral amlodipine besylate tablets (5 mg daily). The patient had no history of smoking or alcohol consumption. He reported having undergone dental scaling once approximately ten years ago and had maintained a routine of brushing his teeth once daily in the morning. Fourteen days prior to presentation, following a meal with friends, the patient began experiencing melena once daily, accompanied by loss of appetite. At that time, the patient did not exhibit chills, fever, nausea, vomiting, acid reflux, or heartburn. The patient self-administered Jianwei Xiaoshi Tablets (a traditional Chinese herbal medicine for strengthening the stomach and promoting digestion), but there was no improvement in the melena.

Upon presentation, the patient’s temperature was 37.6°C. Laboratory tests revealed an elevated white blood cell (WBC) count (11.5×10⁹/L, normal range 3.5 to 9.5×10⁹/L) and an increased neutrophil percentage (87.7%, normal range 40.0% to 75.0%). The fecal occult blood test results were positive. Gastroscopic examination revealed normal mucosal architecture in the gastric fundus (Figure 1A) and preserved anatomical folds on retroflexed view of the gastric body (Figure 1B). Diffuse hyperemic mucosa with focal edema (Figure 1C and D), scattered superficial erosions at the transition zone between antrum and body (Figure 1E), and multiple punctate erosions with fresh erythematous lesions along the greater curvature of the antrum (Figure 1F) were observed. These findings confirmed superficial gastritis with erosions. Consequently, the patient was admitted with UGIB. Prior to this episode, the patient had no history of gastrointestinal diseases.

Figure 1.

Endoscopic images of the stomach on Day 1. White circles in panels (A–F) indicate the representative focal areas of hyperemic and erosive mucosa: (A) Gastric fundus with normal mucosal architecture. (B) Retroflexed view of the gastric body demonstrating preserved anatomical folds. (C) Multiple linear erythema and erosions in the greater curvature and anterior wall of the upper gastric body. (D) Linear hyperemic erythema in the greater curvature of the middle and lower gastric body. (E) Transition zone between the gastric antrum and body, exhibiting scattered superficial erosions. (F) Antral mucosa with multiple punctate erosions and fresh erythematous lesions along the greater curvature.

Of note, gastric mucosal biopsy or culture for microbiological confirmation was not performed during gastroscopy, as the initial diagnostic focus was on the upper gastrointestinal bleeding, and the possibility of infective endocarditis was not yet recognized at that time.

On the second day of hospitalization, the patient’s temperature spiked to 38.9°C. Laboratory findings indicated an elevated high-sensitivity C-reactive protein (CRP) concentration (133.0 mg/L, normal range 0–6.0 mg/L). Coagulation function tests showed an increased International Normalized Ratio (INR) of 1.31 (normal range 0.84–1.10) and D-dimer levels of 4.00 mg/L (normal range 0.00–0.50 mg/L). Tumor marker assessments revealed elevated carbohydrate antigen 125 (CA125) concentrations of 44.48 U/mL (normal <24.00 U/mL) and ferritin levels of 742.66 ng/mL (normal range 23.90–336.20 ng/mL). An abdominal computed tomography (CT) scan provided radiological evidence of a wedge-shaped infarction in the spleen (Figure 2A), accompanied by positive percussion tenderness over the splenic flexure. Blood culture samples were collected and empirical antimicrobial treatment with levofloxacin injection (0.5 g once daily) was initiated.

Figure 2.

Abdominal CT scan obtained on Day 2 and Transthoracic echocardiogram on Day 7. (A) Axial abdominal CT scan demonstrates a wedge-shaped hypodense lesion (arrow) within the spleen, consistent with acute splenic infarction. (B) Transthoracic echocardiogram (parasternal long-axis view) reveals a large, irregular vegetation (14.1 mm × 8.75 mm, arrow) adherent to the aortic valve leaflets, accompanied by severe aortic regurgitation with possible valvular perforation.

On the fourth day of hospitalization, the patient’s temperature rose to 39.2°C. Laboratory tests revealed a procalcitonin (PCT) concentration of 1.036 ng/mL (normal < 0.065 ng/mL). On the fifth day, the blood cultures were positive for S. mitis. Infectious disease consultations have raised the possibility of IE development. An oral consultation was performed on admission, which documented poor oral hygiene but did not reveal active periodontal disease or gingival bleeding. Consequently, empirical antibiotic therapy was promptly escalated to intravenous piperacillin-tazobactam (TZP, 4.5 g every 8 hours). By the sixth day, the patient’s temperature had normalized, and vital signs were stable On the seventh day, an echocardiogram revealed left ventricular enlargement with a large vegetation on the aortic valve (14.1 mm × 8.75 mm), accompanied by severe regurgitation with possible valvular perforation. Mild regurgitation was also observed in the mitral, tricuspid, and pulmonary valves, with a small amount of pericardial effusion (Figure 2B). The patient exhibited diminished breath sounds in the lower lobes of both lungs and a murmur was audible over the aortic valve area. Upon thorough re-examination, dental wear was observed (Figure 3A–D). Multiple tender Osler’s nodes were identified on the fleshy parts of the distal phalanges of the index, middle, and ring fingers (Figure 4A and B), and splinter hemorrhages beneath the nail of the right ring finger were observed (Figure 4C). Additionally, painless reddish-brown bleeding macules were present on the toes and sole of the left foot, suggestive of Janeway lesions (Figures4D–G). According to the 2023 Duke-ISCVID criteria,^3,4 the patient met two major criteria (positive blood cultures for S. mitis and echocardiographic evidence of endocardial involvement) and three minor criteria (fever, vascular phenomena including splenic infarction and Janeway lesions, and immunologic phenomena including Osler’s nodes). Following multidisciplinary discussion, the patient was diagnosed with IE complicated by significant vegetation on the aortic valve.

Figure 3.

Dental attrition with visible wear and exposed dentin observed on Day 7 of hospitalization. (A) Left superior view. (B) Right superior view. (C) Left inferior view. (D) Right inferior view.

Figure 4.

Observation of Osler’s nodes and Janeway lesions in the extremities on Day 7 of hospitalization. White circles indicate the lesions of interest. (A) Multiple Osler’s nodes visible on the fleshy areas of the fingertips of the right hand (white circles). (B) Close-up view of the middle finger, highlighting additional Osler’s nodes (white circle). (C) Splinter hemorrhage observed under the nail of the right ring finger (white circle). (D) Janeway lesions observed on the tips of the right toes (white circle). (E) Janeway lesions observed on the tips of the left toes (white circle). (F) Overview showing lesions on the sole of the left foot (white circles; black arrow indicates the area magnified in G. (G) Magnified view of the black arrow-indicated area in F, highlighting the Janeway lesions in detail (white circle).

By the eighth day of hospitalization, the patient’s temperature had normalized, and vital signs were stable Subsequently, the patient was transferred to another hospital for aortic valve replacement surgery as a part of further treatment. At the 2-week postoperative follow-up, the patient reported that the surgical procedure had proceeded smoothly. Postoperatively, the patient adhered to the prescribed oral administration of linezolid tablets (0.6 g, every 12 hours). On re-examination, it was noted that the Osler’s nodes on the fingertips of the right hand had resolved. Repeat echocardiography did not reveal any significant vegetations.

Discussion

Over the past few decades, significant changes in the epidemiological characteristics of IE have been observed. Globally, the number of IE cases and related deaths has sharply increased over the past 30 years, from 478,000 cases in 1990 to 1,090,530 cases in 2019. During the same period, the number of deaths increased from 28,750 to 66,320.¹ By 2019, the incidence had gradually risen to 13.8 per 100,000 population, with older adults and males being disproportionately affected.⁵ Chamat-Hedemand et al investigated all patients with streptococcal bloodstream infections (BSIs), from 2008 to 2017, in the Capital Region of Denmark.⁶ Among 6506 cases with streptococcal BSIs (mean age 68.1 years [standard deviation (SD) 16.2], 52.8% men) the IE prevalence was 7.1% (95% confidence interval (CI), 6.5–7.8). The prevalence varied according to the species of Streptococcus, with the highest rates observed for S. mutans at 47.9% (95% CI: 33.3–62.8%), S. gordonii at 44.2% (95% CI: 34.0–54.8%), S. sanguinis at 34.6% (95% CI: 26.6–43.3%), S. gallolyticus (formerly S. bovis) at 30.2% (95% CI: 24.3–36.7%), and S. mitis/oralis at 19.4% (95% CI: 15.6–23.5%). In a multivariate analysis using S. pneumoniae as the reference, all species except S. pyogenes were associated with significantly higher risks of IE, with the following odds ratios (OR): S. gallolyticus OR 31.0 (95% CI: 18.8–51.1), S. mitis/oralis OR 31.6 (95% CI: 19.8–50.5), S. sanguinis OR 59.1 (95% CI: 32.6–107), S. gordonii OR 80.8 (95% CI: 43.9–149), and S. mutans OR 81.3 (95% CI: 37.6–176). As mentioned previously, streptococcal BSI is one of the most common causes of IE.

The human oral microbiome comprises over 700 microbial species, with oral streptococci being among the earliest colonizers of the oral surface.⁷ Almost all oral microorganisms are non-pathogenic opportunistic commensals to maintain oral health condition and defend against pathogenic microorganisms.⁸ Notably, many oral streptococci considered symbiotic have the potential to cause diseases such as bacteremia and IE, as observed in several oral species including S. mitis, S. parasanguinis, and S. salivarius.⁸ These organisms play roles in colonization and pathogenesis through mechanisms such as adhesin–receptor interactions, modulation by environmental cues, acid production, generation of alkaline substances, hydrogen peroxide production, and competitive inhibition.⁹ A study based on whole-genome sequencing (WGS) of 129 blood isolates from IE patients in the United Kingdom (UK) and Ireland revealed that S. mitis associated with IE exhibits high genetic diversity, with no specific hypervirulent or antibiotic-resistant lineages predominating.¹⁰ This suggests that the pathogenicity of S. mitis in IE patients may be incidental. Another cohort study involving 118 consecutive cases of VGS BSIs from July 1, 2011, to December 1, 2012, found that S. mitis was the most frequently implicated microorganism, exhibiting particularly potent virulence.¹¹ Additionally, a retrospective cohort study of all streptococcal BSIs from January 2010 to June 2020 identified Mitis group streptococci as the most common VGS causing IE.¹² The entry routes for bacteria and fungi into the human body are diverse, and the oral cavity serves as an important pathway for the introduction of blood-related symbiotic bacteria, including oral streptococci, into the bloodstream. In this case, the patient exhibited overall good health but had severely worn teeth. Tooth wear is associated with a variety of factors, including age, acid erosion, chewing hard substances, malocclusion, and gastroesophageal reflux disease, among others.¹³ This can lead to changes in the oral microbiome, providing more opportunities for colonization by microorganisms such as S. mitis. However, the patient had not undergone any recent dental procedures. Therefore, we assumed that there were other potential alternative routes for S. mitis to enter the bloodstream.

With the advent of 16S rRNA sequencing technology, the genus S. has been further classified into eight distinct groups based on the phylogenetic relationships among its members. These groups are Mitis, Sanguinis, Anginosus, Salivarius, Dolosigranulum (formerly Downei), Mutans, Pyogenic, and Bovis.¹⁴ Streptococci, especially members of the Mitis group, are the first organisms detected in the mouth of newborn infants and are considered pioneer species (primary colonizers) that allow for the assembly of a complex microbiota. The Mitis group is the largest of the groups found in the oral cavity, with 20 species.⁹ Li et al profiled the bacterial microbiota in paired gastric biopsies (antrum and body) from normal and antral gastritis patients.¹⁵ The results indicated that S. and Prevotella were the two most abundant genera. In patients with gastritis, the relative abundance of S. genus was significantly increased, with no significant difference observed between the antrum and body regions at the genus level. Furthermore, through 16S rRNA sequencing and bacterial culture experiments, researchers have found that most streptococcal species in gastric biopsy samples are viable. Among the streptococcal species isolated, S. mitis had the highest proportion. Khosravi et al utilized tissue biopsy techniques to analyze the cultivable gastric microbiota of 131 Helicobacter pylori (H. pylori)-positive patients and 84 H. pylori negative patients.¹⁶ The study revealed a significant positive correlation between S. parasanguinis, S. mitis, S. salivarius, Neisseria (N). flavescens, N. perflava, and Rothia (R). mucilaginosa within the genus Streptococcus and peptic ulcer disease. Furthermore, these bacteria were found to be capable of colonizing the gastric environment rather than existing transiently.

Recently, Xia et al presented the concept of the oral-gastric microbial axis, which refers to the dynamic interconnection of oral and gastric health status and local microecology.¹⁷ The representative oral microbiota in healthy individuals contains the major gastric microbiota in healthy individuals, although most oral microbes are blocked by the mucus barrier secreted by the healthy gastric mucosa. However, under certain circumstances, such as alkaline-producing oral microorganisms,¹⁸ persistent H. pylori infection,¹⁹ and long-term use of proton pump inhibitors (PPIs),²⁰ all can neutralize stomach acid, alter the gastric microenvironment, and weaken the gastric mucus barrier function. This can subsequently lead to the ectopic colonization of oral-related microorganisms. Microorganisms that are ectopically colonized can promote the development of gastric mucosal lesions through direct or indirect means. As gastric lesions develop from inflammation to cancer (including chronic gastritis, atrophic gastritis, intestinal metaplasia, dysplasia, and ultimately carcinogenesis), the abundance of the dominant colonizer, H. pylori decreases, whereas that of oral-associated microbes increases. Oral-associated microbes co-occur with each other and have complex interactions with other microbes, such as H. pylori, which may further affect the biological processes of gastric lesions, such as gastric cell apoptosis and epithelial-mesenchymal transition. Oral microbes are the source of microbiota in downstream organs, and continuously seed the gastrointestinal tract during eating and swallowing. Vomiting and gastroesophageal reflux allow bacteria to travel retrogradely from the gastrointestinal tract to the oral cavity. Therefore, there is a significant similarity between the oral and gastric microbiomes. S. mitis is a crucial member of the oral-gastric microbiome axis and plays a pivotal role in maintaining the ecological balance between the oral cavity and stomach. In our case, the patient was hospitalized for UGIB after a group meal, and gastroscopy suggested SG with erosion, which further supported the hypothesis that bacteria may enter the bloodstream through the damaged gastrointestinal tract. Therefore, we speculate that oral microbes such as S. mitis may break through the barrier under specific pathological conditions of impaired gastric mucosal barrier function, colonize ectopically in the stomach, participate in the development of gastric mucosal lesions, and enter the bloodstream through early gastric mucosal erosion and ulceration pathways, ultimately causing IE It should be acknowledged that a reverse causal pathway has been reported, in which IE leads to gastric ischemia through septic embolization to the splanchnic arteries.²¹ In the present case, this possibility is less likely, as melena preceded fever and bacteremia by 14 days, and the endoscopic findings of diffuse hyperemic mucosa with scattered punctate erosions are more consistent with superficial gastritis than with the well-demarcated ischemic lesions typical of embolic events. However, the presence of splenic infarction on admission CT suggests that subclinical embolic events may have occurred before overt fever, so the reverse pathway cannot be completely excluded. On balance, the morphological and temporal evidence favors gastric mucosal erosion as the primary portal of entry.

Traditionally, it has been believed that hematogenous S. mitis infections are often associated with dental diseases, oral procedures, or poor oral hygiene. In this study, we searched the PubMed database using the keywords “Streptococcus mitis”, “Mitis”, “Infective endocarditis”, and “Endocarditis”, and reviewed all cases of endocarditis caused by S. mitis infection from the establishment of the database up to December 31, 2024. Our primary objective was to identify the risk factors contributing to disease onset and the specific pathways through which S. mitis translocates into the bloodstream. This study aimed to enhance our understanding of the pathogenic mechanisms underlying S. mitis-associated IE.

Using a PubMed literature search, we identified 40 cases of IE caused by S. mitis (Table 1). Among these patients, had cardiac risk factors for IE, including history of IE, valvular heart disease (VHD), valve replacement surgery (VRS), hypertrophic obstructive cardiomyopathy (HOCM), and congenital heart disease (CHD). Twelve cases were linked to dental factors: six involved poor dental health such as caries or malocclusion; two lacked antibiotic prophylaxis during dental procedures; one case used amoxicillin for prophylaxis, but the S. mitis isolate showed intermediate resistance to penicillin, potentially reducing treatment efficacy; one case involved infected molars with poor medication adherence leading to IE; one case resulted from mucosal trauma caused by orthodontic appliances, leading to S. mitis infection and IE; and one case involved a lactating mother who developed IE, with bacteria likely originating from the infant’s oral flora. Six patients reported recent intravenous drug or substance abuse. Four patients were in a clinically immunosuppressed state. Two cases suggested that S. mitis originated from the nasopharynx. One case was linked to a highly inflammatory and hypercoagulable state induced by Coronavirus Disease 2019 (COVID-19), leading to IE. Seven patients lacked a detailed medical history or had missing information regarding their medical history and medication use. Notably, among the S. mitis-associated IE cases, purely cardiac risk factors were identified in 6 of 40 cases (15%), whereas purely dental factors accounted for 5 of 40 cases (12.5%). Nine of the 40 cases (22.5%) had both cardiac and dental risk factors. Only one out of 40 cases (2.5%) involved cardiac risk factors combined with intravenous drug use, and only two out of 40 cases (5%) involved cardiac risk factors combined with clinical immunosuppression.

Table 1.

Overview of 40 Cases of Infectious Endocarditis with S. mitis in Humans

Author	Sex, Age	Cardiac RF	Odontogenic RF	GI RF	IV Drug Use-Related RF	Immunosuppression-Related RF	Other RF	Cause (Author)	Valve	Treatment (Med/Surg)	Outcome
Masuda et al22	F, 44	•VSD						•VSD	•RVW	•PAPM/BP	Recovered
Shinonaga et al23	F, 52							N.r.	•MV	•IPM, CLDM •MVR, TAP	Recovered
le Bayon et al24	F, 36	•Barlow’s disease •Streptococcal endocarditis with SI	•Caries					N.r.	N.r.	•PEN G •Extractions with prophylactic antibiotics	Recovered
Hall et al25	M, 66	•MVP	•Several dental procedures (AMX prophylaxis each time)					•Periodontal surgery •Moderate resistance to PEN G in S.mitis isolates	•Anterior leaflet of the MV	•PEN G, GEN	Recovered
Ng et al26	M, 37	•RHD •History of IE with CVA						N.r.	•MV	•LZD	Recovered
Aoyagi et al27	F, 23						•G1P0, 29w2d	N.r.	•MV	•CS at 29w3d •PEN G, GEN •MV repair	Recovered
Franz et al28	M, 37	•HOCM					•Frequent nosebleeds for years	•Chronic nosebleeds	•AV •MV	•Flap replacement, Septal myectomy •PEN G, GEN	Recovered
Cone et al29	M, 87	•MR	•Infected molar treated with only 1 g of AMX.					•Infected molar treated with only 1 g of AMX.	•MV	•AMP, CTRX	Recovered
Taketani et al30	Boy, 2							N.r.	•MV	•LZD	Recovered
Dinani et al31	M, 85	•Old heart murmur				•Diabetes mellitus		•With underlying conditions including diabetes	•Posterior leaflet of the MV	•CTRX	Recovered
Takayasu et al32	F, 73	•MV disease •CHF				•Anorexia and weight loss	•Hypertensive cardiomyopathy	•MV disease •An elderly individual •Consumptive syndrome	•Left atrium •MV	•CTRX, CAM	Died
Matsui et al33	M, 23					•Cerebral palsy •After splenectomy •AIHA •Long-term steroid treatment •Congenital IgA deficiency		•Immunosuppression	•Anterior commissure of the MV	•AZM •MINO, CLDM	Died
Deo et al34	N.r.					•Multiple myeloma		•Immunosuppression	•MV leaflet	•MVR, AVr	Recovered
Geisenberger et al35	M, 38	•Affection of the left cardiac valves			•IV drug users			•Left valve involvement •IV drug users	•AV	•Emergency craniotomy	Died
Ezure et al36	M, 44							N.r.	•MV	•AMP, GEN, AMP/Cloxa, LVFX •MPL	Recovered
Caliskan et al37	F, 24	•MVP, MR					•G3P2	•MVP	•MV posterior leaflet	•PEN G	Recovered, vaginal delivery
Leuenhagen et al38	F, 23				•10-year daily THC abuse •5-year pernasal amphetamine abuse			•Long-term substance abuse leads to immunosuppression	•MV	•VAN, CTRX, PEN G, GEN	Died
Hosokawa et al39	M, 52							•No predisposing RF	•Left atrial mural	•PEN G, GEN	Recovered
Byrd et al40	M, 57		•Poor dentition	•Antral ulcers, duodenitis, and esophagitis	•Polysubstance abuse (alcohol, cocaine, tobacco, prior IV drugs)			•Poor dentition •Alcohol causes subclinical immunosuppression.	•AV	•CTRX	Recovered
Bamford et al41	M, 74							N.r.	•PV	•PEN G, GEN, AMX/Clav, CLDM •PVR	Recovered
Hakimah et al42	F, 32						•Breastfeeding	•Possibly from the baby’s oral commensal	•Posterior MV leaflet	•PEN G, GEN	Recovered
El Barzouhi et al43	M, 38	•CHD •History of Ross surgery						•CHD	•PV	•PEN G, CLDM	Recovered
Pericàs et al44	M, 37	•MVR, AVR	•Dental cleaning procedure		•History of IV drug use			N.r.	•MV	•VAN, GEN	Died
Degheim et al45	M, 38	•BAV	•Poor dentition					•Poor dentition •BAV	•MV	•Antibiotics •MVR, AVR	Recovered
Kumanayaka et al46	M, 38						•COVID-19	•COVID-19	•Anterior MV leaflet	•CTRX	Recovered
Demetriades et al47	F, late 30s							•No predisposing risk factors	•AV	•PEN G •AVR	Recovered
Amorgianos et al48	F, 17	•Perimembranous VSD	•Orthodontic appliances					•Oral mucosa trauma from orthodontic appliances	•Tricuspid	•PEN G, GEN, CTRX •Inpatient surgery	Recovered
Casey et al49	F, 49						•Chronic Allergic rhinitis	•Nasal irrigation	•MV	•CTRX, VAN •AVR	Recovered
Zubair et al50	M, 82	•History of AVR, CABG, CAD		•GERD •Non-bilious emesis				•History of AVR, CABG, CAD •Actinomyces bacteremia may have originated from the esophagus	•Aortic prosthetic •MV	•CTRX, AMP and SUL AMX	Recovered
Moysidis et al51	F, 33	•VSD •CHD						•CHD •Intracardiac high-velocity shunts	•Tricuspid	•GEN, CIP, VAN, LZD, CTRX •TVr, Annuloplasty, VSD closure	Recovered
Karmali et al52	F, 52	•TA, D-TGA	•Multiple carious root tips					•Poor dentition	•MV	•CTRX	Recovered
		•ASD, VSD
Liu-An et al53	M, 40	•History of recurrent S. mitis IE	•Poor dentition					•Poor dentition •Prior episodes of IE •Male	•Aortic root	•CTRX, VAN, RIF, GEN	Recovered
Xiao et al54	M, 43				•Heroin use disorder			•History of IV drug abuse	•Tricuspid	•Zosyn, VAN •AngioVac procedure	Recovered
Aoyagi et al55	M, 71	•AR •HF		•GC •AG				•GC	•AV	•SUL/AMP, TEC, PEN G •AVR	Recovered
Bah et al56	F, 54		•Moderate gingivitis, tooth extraction					•Tooth extraction	•MV	•CTRX	Recovered
Luo et al57	F, 33							N.r.	•MV	•VAN, LZD	Died
Khan et al58	M, 40s		•Dental treatment 3 months pre-admission.					•Dental treatment	•MV	•PEN G	Recovered
Matsuhashi et al59	F, 31							•Unknown	•MV	•PEN G	Recovered
Paula et al60	M, 53	•BAV	•Poor dentition					•GI cancer cannot be ruled out •BAV	•MV leaflet	•PEN G •MPL	Recovered
Rafique et al61	F, 22	•BAV		•GERD	•Cannabis use disorder			•BAV	•BAV	•PEN G	Recovered
Present study	M, 59			•SG with erosions •UGIB				•SG with erosions •UGIB	•AV	•TZP •AVR	Recovered

Abbreviations: F, Female; M, Male; N.r., Not reported; RF, Risk factors; VSD, Ventricular septal defect; SI, Splenic infarction; MVP, Mitral valve prolapse; RHD, Rheumatic heart disease; IE, Infective endocarditis; CVA, Cerebrovascular accident; HOCM, Hypertrophic obstructive cardiomyopathy; MR, Mitral Regurgitation; MV, Mitral valve; CHF, Congestive Heart Failure; MVR, Mitral valve replacement; CHD, congenital heart disease; AVR, Aortic valve replacement; BAV, Bicuspid aortic valve; CABG, Coronary artery bypass grafting; CAD, Coronary artery disease; TA, Tricuspid atresia; D-TGA, D-Transposition of the Great Arteries; ASD, Atrial Septal Defect; S., Streptococcus; AR, Aortic regurgitation; HF, Heart failure; AMX, Amoxicillin; GI, Gastrointestinal; GI RF, Gastrointestinal risk factors; GERD, Gastroesophageal reflux disease; GC, Gastric cancer; AG, Atrophic gastritis; SG, Superficial gastritis; UGIB, Upper gastrointestinal bleeding; IV, Intravenous; THC, Tetrahydrocannabinol; AIHA, Autoimmune hemolytic anemia; IgA, Immunoglobulin A; G1P0, Gravida 1, Para 0; 29w2d, 29 weeks 2 days; G3P2, Gravida 3, Para 2; COVID-19, Coronavirus Disease 2019; PEN, Penicillin; RVW, Right ventricular wall; AV, Aortic valve; PV, Pulmonary valve; Med, Medications; Surg, Surgery; PAPM/BP, Panipenem/betamipron; IPM, Imipenem; CLDM, Clindamycin; TAP, Tricuspid annuloplasty; GEN, Gentamicin; LZD, Linezolid; CS, Caesarean section; 29w3d, 29 weeks 3 days; AMP, Ampicillin; CTRX, Ceftriaxone; CAM, Clarithromycin; AZM, Azithromycin; MINO, Minocycline; AVr, Aortic valve repair; LVFX, Levofloxacin; Cloxa, Cloxacillin; MPL, Mitral valve plasty; VAN, Vancomycin; Clav, Clavulanic Acid; PVR, Pulmonary valve replacement; SUL, Sulbactam; CIP, Ciprofloxacin; TVr, Tricuspid valve repair; RIF, Rifampin; TEC, Teicoplanin; TZP, Piperacillin/tazobactam

The 40-case review underscores the importance of recognizing diverse sources of S. mitis bacteremia. Notably, among these 40 cases, none explicitly identified UGIB or gastric erosive disease as the primary portal of entry, except for Byrd et al⁴⁰ who described upper gastrointestinal lesions (antral ulcers, duodenitis, and esophagitis) in a patient with polysubstance abuse, but did not attribute the bacteremia to the gastric pathology. Similarly, Aoyagi et al reported a case of S. mitis IE in a patient with early gastric carcinoma.⁵⁵ However, that case did not present with UGIB as the triggering event, and the patient had pre-existing aortic regurgitation. Therefore, to our knowledge, the present case is the first to suggest that superficial gastritis with erosions, presenting as UGIB, may serve as the primary portal of entry for S. mitis bacteremia leading to IE, in the absence of significant cardiac or dental risk factors.

In a series of 40 cases of S. mitis endocarditis, the primary hematogenous routes of infection included dental diseases, oral procedures, and poor oral hygiene. However, specific hematogenous bacterial routes were also observed. For instance, Zubair et al reported a case of polymicrobial IE involving Actinomyces naeslundii, S. mitis, S. oralis, and Abiotrophia defectiva in a patient with dysphagia.⁵⁰ The patient had a history of prosthetic aortic valve replacement, coronary artery bypass grafting, coronary artery disease, and gastrointestinal symptoms including gastroesophageal reflux disease, loss of appetite, nausea, non-bilious vomiting, and dyspepsia. A barium swallow study revealed an irregular appearance of the distal esophagus with spasms, reflux, and multiple protrusions suggestive of diverticula. The authors speculated that actinomycosis bacteremia might have originated from the esophagus. Byrd et al also reported upper gastrointestinal lesions in a patient with S. mitis IE, but did not consider the potential link to gastrointestinal bleeding.⁴⁰ However, they did not consider the potential link to gastrointestinal bleeding. This highlights the importance of recognizing diverse sources of S. mitis bacteremia and the need for a comprehensive evaluation of both oral and gastrointestinal health in patients presenting with IE.

A limitation of this case analysis is the lack of bacterial culture from the patient’s stomach, which prevented us from definitively determining whether S. mitis originated from the bleeding gastric site. Additionally, gastric mucosal biopsy or culture was not performed during gastroscopy, as the initial diagnostic workup was focused on the upper gastrointestinal bleeding, and the possibility of IE was not suspected at that time. This precludes microbiological confirmation of S. mitis colonization at the gastric mucosal level and limits our ability to definitively establish the stomach as the primary portal of entry. Furthermore, postoperative valvular material culture was not obtained, as the patient was transferred to another hospital for surgical intervention. The initial empirical use of levofloxacin may also be considered suboptimal for suspected streptococcal bacteremia, as fluoroquinolone resistance has been reported in viridans group streptococci. However, levofloxacin was initiated before blood culture results were available, and the antibiotic regimen was promptly escalated to piperacillin-tazobactam once S. mitis was identified and IE was suspected. The patient’s fever resolved rapidly following this adjustment, suggesting that the subsequent targeted therapy was effective. Finally, the literature search was conducted exclusively in the PubMed database; Embase, Cochrane Library, and Scopus were not searched, which may have resulted in the omission of relevant case reports and limits the comprehensiveness of our literature review. Despite these limitations, the temporal relationship between UGIB and the development of IE, together with the exclusion of other common portals of entry, supports our hypothesis that gastric mucosal erosion may have served as the primary portal of entry for S. mitis bacteremia in this case.

It is also important to consider whether gastroscopy itself could have served as the portal of entry for S. mitis bacteremia. However, this possibility is effectively ruled out by the temporal sequence. The patient developed melena 14 days before admission. Fever (38.9°C) developed on the second day of hospitalization, whereas gastroscopy was performed on Day 1. The bacteremia was therefore already present before the procedure, making gastroscopy an unlikely source of the infection.

Conclusion

This report presents a case of IE associated with S. mitis, which was triggered by UGIB secondary to SG. The temporal relationship suggests a temporal correlation between UGIB and aortic valve vegetation formation. Black stool symptoms appeared 14 days before admission, followed by fever and positive blood cultures. Based on this, we hypothesize that S. mitis invasion may occur through early gastric mucosal erosions and ulcers, leading to bacteremia and subsequently IE. Therefore, we recommend that clinicians remain vigilant about the risk of bacterial dissemination through compromised gastrointestinal mucosa in cases of unexplained infections. Prompt and appropriate diagnostic testing and treatment are crucial for preventing the occurrence and recurrence of IE. Additionally, although current guidelines primarily recommend antibiotic prophylaxis for certain dental procedures, our case raises the question of whether patients with known cardiac risk factors and gastrointestinal mucosal injury might also benefit from such consideration, warranting further investigation.

Funding Statement

The authors received no specific funding for this work.

Ethics Approval and Informed Consent

This case report was reviewed and approved by the Ethics Committee of Tongde Hospital of Zhejiang Province (Protocol No. 2025-025), which required institutional approval for the publication of case details. Written informed consent for participation in this study was obtained from the patient’s legal guardian.

Consent for Publication

Written informed consent for publication of this case report and any accompanying images was obtained from the patient’s legal guardian (the patient’s son), as the patient has limited literacy and is unable to provide written consent. The patient was fully informed about the publication of his case details and images and expressed his willingness to publish. The guardian was actively involved in the clinical decision-making process and provided written consent on the patient’s behalf.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors report no conflicts of interest in this work.