Poor Outcome of Rare Lactobacillus Bacteremia and Endocarditis in a Patient With Frequent Consumption of Live Culture Yogurts

Abstract

Lactobacillus rhamnosus is a gram-positive bacterium commonly found in dairy products and used as a probiotic due to its resistance to acid and bile. While generally considered safe, rare cases of L. rhamnosus bacteremia and endocarditis have been reported, primarily in individuals with significant risk factors. This report discusses an elderly male with a history of cardiovascular diseases, diabetes, and asthma, who developed L. rhamnosus bacteremia and endocarditis after consuming large quantities of Greek yogurt. The patient initially presented with fatigue, myalgias, and fever, and was treated empirically with cefepime and vancomycin. On that occasion, blood cultures recovered L. rhamnosus which was treated with a short course of oral antibiotics. Despite initial improvement, the patient was re-hospitalized with a transient cerebral ischemic attack. Subsequent investigations again revealed L. rhamnosus in blood cultures raising suspicion for subacute bacterial endocarditis. Antibiotic therapy with ampicillin resulted in the clearance of bacteremia; unfortunately, the patient experienced a non-ST elevation myocardial infarction leading to cardiac arrest and death. This case highlights the potential pathogenicity of L. rhamnosus in susceptible individuals and underscores the need for caution in consuming probiotic products among those with severe or multiple comorbidities.

1. Introduction

Lactobacillus rhamnosus (also known as Lacticaseibacillus rhamnosus) is a gram-positive facultative anaerobic non-spore-forming bacteria. L. rhamnosus strains are used in dairy products including cheeses and yogurts.1 They are also incorporated into various commercial products because of their natural resistance to acid and bile and superior ability to adhere to the intestinal epithelial layer compared to other Lactobacillus species.2 Strains of L. rhamnosus are naturally present in the healthy human gastrointestinal microbiome as well as in the normal flora of the female genitourinary tract.

Although probiotics containing Lactobacillus species are commonly used and generally considered safe, there are cases of bacteremia caused by these bacteria. Albeit extremely rare, Lactobacillus bacteremia has been reported in individuals with risk factors such as diabetes mellitus, heart disease, malignancy, kidney disease, steroid use, recent dental work, and prolonged antibiotic therapy.3 Several studies have linked the same strain of L. rhamnosus causing bacteremia to the probiotics that patients were consuming. In 2017, Aaron et al. described a case in which an elderly patient with previous history of atrial fibrillation underwent endoscopy then developed L. rhamnosus bacteremia and endocarditis. The genetic analysis found 99.6% similarity in the strain of L. rhamnosus from yogurt that the patient frequently consumed to the strain from the mitral valve vegetation.4 A review of several cases of Lactobacillus infective endocarditis by Ioannou et al. found probiotics to be a common identified source of the Lactobacillus, and reported good outcomes of the majority of cases.5,6 We present a case with L. rhamnosus bacteremia and endocarditis that had a poor outcome.

2. Case presentation

The patient was an elderly male (between 90 and 95 years of age) with a past medical history of aortic stenosis status post transcatheter bio-prosthetic aortic valve replacement, coronary artery disease status post coronary artery bypass surgery, congestive heart failure, paroxysmal atrial fibrillation on prophylactic anticoagulation, type 2 diabetes mellitus, and asthma, who presented initially for fatigue, myalgias, fevers, and chills that started three weeks prior. His vitals were unremarkable on arrival and respiratory panel PCR (including for SARS-CoV-2) were negative. He had an elevated white blood cell count of 13.0 × 10³/μL and elevated procalcitonin of 0.33 ng/mL. Blood cultures were collected prior to antibiotic initiation. The patient was then empirically started on cefepime 2 g every 12 h and vancomycin 1.0–1.5 g daily until discharge two days later. The 24-h blood cultures revealed no growth; however, after discharge, results from the 48-h blood cultures were available and revealed gram positive rods. The patient was notified and empirically prescribed doxycycline 100 mg twice daily for 14 days. The specific organism was identified to be L. rhamnosus a few days later, but the patient was lost to follow-up.

Four weeks after discharge, the patient was rehospitalized for sudden onset right-side weakness and dizziness. The patient took 325 mg of aspirin before arriving at the hospital, and the symptoms resolved within 30 min of onset. Computed tomography (CT) head and CT angiography head did not reveal any acute changes, only previously diagnosed chronic occlusion of the left internal carotid artery; the patient was diagnosed with transient cerebral ischemic attack. The patient was already taking apixaban for three months consistently when this incident occurred. Further investigation to determine the etiology and evaluate for any cerebral damage with magnetic resonance imaging (MRI) of the head and transesophageal echocardiography were recommended but were both ultimately declined by the patient.

On hospitalization day 2, apixaban 5 mg twice daily was continued and aspirin 81 mg daily was started. On the physical exam, the patient had normal vitals and no focal neurological deficits, but there was a notable diastolic decrescendo murmur heard at the left sternal border and splinter hemorrhages on the fingernails. The patient was agreeable to a transthoracic echocardiogram which revealed moderate to severe aortic regurgitation and mild aortic perivalvular leak. There was no vegetation noted on the transthoracic echocardiogram. Blood cultures were collected on the same day.

On hospitalization day 4, the blood cultures grew L. rhamnosus; however, susceptibility testing was unable to be completed by our in-house laboratory and was sent to a specialized outside laboratory. The patient was empirically started on ampicillin 2 g every 6 h. From the clinical, laboratory, and imaging data gathered, the patient was diagnosed with subacute bacterial endocarditis. The patient’s transient cerebral ischemic attack was presumed to be caused by septic emboli from endocarditis.

Blood cultures collected on hospitalization day 4 and hospitalization day 6 continued to grow L. rhamnosus. An in-depth interview of the patient was completed to investigate the source of the L. rhamnosus. His work, living environment, and travel history were unremarkable. However, patient’s dietary habits revealed that he consumed large amounts of Greek yogurt on a regular basis, approximately 1–2 L per week, under the presumption that the probiotics from the yogurt were promoting health. He did not frequently eat cheese and denied taking any other probiotic supplements.

The blood cultures grew no organisms by the 8^th hospitalization day. The patient continued ampicillin therapy. After discussion with the patient and shared decision making, it was decided that benefits of surgical intervention to remove and replace the bioprosthetic aortic valve, which likely had vegetations, did not outweigh the risks considering the patient’s age and frailty. Thus, a salvage plan was created to have the patient continue intravenous ampicillin 2 g every 6 h at home for a total of 6 weeks following his last negative blood culture, followed by lifelong suppressive therapy with oral amoxicillin 500 mg twice daily.

Unfortunately, on hospitalization day 11, the patient had a non-ST elevation myocardial infarction that led to cardiac arrest, requiring advanced cardiovascular life support. The patient had return of spontaneous circulation, but then changed his code status to “do not resuscitate” and passed away after a second cardiac arrest a few hours later.

During this entire hospitalization the patient was afebrile and white blood cell counts were within normal limits. Procalcitonin levels were initially mildly elevated at the beginning of this hospitalization but normalized by hospitalization day 10. Table 1 presents a summary of the laboratory investigations conducted during this hospitalization period. The antibiotic susceptibility data of the L. rhamnosus strain isolated from our patient is presented in Table 2. A summary report of the transthoracic echocardiogram is available in the Supplementary Materials.

Table 1.

Laboratory investigations from the first admission (day –32) to the second admission (day 1) and discharge (day 11).

Laboratory Test	Reference Values	Day –32	Day 1	Day 2	Day 3	Day 4	Day 6	Day 8	Day 9	Day 10	Day 11
White blood count	4.2–10.8 × 103/μL	13.0a	7.2	7.0	6.1	–	–	6.7	7.5	9.9	10.5
Hemoglobin	14.0–18.0 g/dL	13.0a	12.0a	11.6a	11.0a	–	–	10.6a	10.1a	10.8a	10.5a
Platelet count	130–450 × 103/μL	267	248	235	235	–	–	284	257	305	287
Absolute neutrophil count	1.80–7.00 × 103/μL	11.63a	5.03	5.38	4.36	–	–	4.57	5.31	7.88a	9.49a
Absolute lymphocyte count	1.00–4.80 × 103/μL	0.63a	1.46	0.91a	1.01	–	–	1.31	1.31	1.03	0.48a
Absolute eosinophil count	0.05–0.35 × 103/μL	0.03a	0.17	0.13	0.22	–	–	0.14	0.21	0.17	0.03a
Sodium	136–145 mmol/L	134a	135a	134a	137	139	–	143	139	140	136
Potassium	3.3–5.1 mmol/L	3.6	3.7	3.8	3.8	4.0	–	4.0	3.9	4.4	4.4
Blood urea nitrogen	7–18 mg/dL	20a	42a	30a	24a	16	–	13	12	13	15
Creatinine	0.60–1.30 mg/dL	1.30	1.50a	1.20	1.10	1.10	–	1.00	1.10	1.30	1.30
Glucose	70–99 mg/dL	154a	227a	112a	117a	144a	–	164a	139a	175a	235a
Total bilirubin	0.2–1.0 mg/dL	0.60	–	0.30	0.30	0.30	–	0.20	0.30	0.30	0.40
Alkaline phosphatase	45–117 IU/L	127	–	107	101	113	–	99	86	101	110
Aspartate aminotransferase	15–37 IU/L	21	–	24	21	26	–	18	15	16	23
Alanine aminotransferase	13–61 IU/L	36	–	34	29	34	–	24	20	22	24
Blood culture, L. rhamnosus	Negative	Positivea	–	Positivea	–	Positivea	Positivea	Negative	–	–	–

^aRepresents abnormal values.

Table 2.

Antibiotic susceptibility of Lactobacillus rhamnosus strain isolated from patient during the second admission.

Antibiotic	Minimum Inhibitory Concentration	Susceptibility
Ampicillin	1	Susceptible
Clindamycin	≤0.500	Susceptible
Daptomycin	≤0.500	Susceptible
Erythromycin	≤0.500	Susceptible
Linezolid	≤1	Susceptible
Penicillin	0.500	Susceptible
Vancomycin	>16	Resistant

3. Discussion

L. rhamnosus bacteremia remains a rare phenomenon, and even rarer are the few reported cases of L. rhamnosus causing endocarditis.4,5,7–12 At our institution, this was the first ever case of bacteremia due to any Lactobacillus species. This case is unique in that it involved the oldest patient with L. rhamnosus endocarditis reported, and was one of very few that involved septic emboli to the brain causing focal neurological deficits and only one of few that resulted in death.11

Our patient had multiple cardiovascular comorbidities, all of which ultimately contributed to his death. However, the presence of subacute bio-prosthetic valve infective endocarditis, which we believe was developing for at least several months, resulted in additional stress from aortic regurgitation and septic emboli, and likely played a major role in the patient’s myocardial infarction and cardiogenic shock. The metabolic stresses of persistent bacteremia also likely contributed to the patient’s death.

The brand of Greek yogurt which our patient consumed explicitly lists L. rhamnosus as one of the live and active cultures in their yogurts, along with S. thermophilus, L. bulgaricus, L. acidophilus, Bifidus, and L. casei.13 Although there were six bacterial species, including four from the Lactobacillus genus, in the Greek yogurt consumed by our patient, only L. rhamnosus was ever isolated from the five separate sets of blood cultures that were collected. This suggests that L. rhamnosus is potentially more pathogenic, or at least,more likely to cause endocarditis compared to other Lactobacillus species. This theory is also supported by a review of 50 cases of Lactobacillus endocarditis by Campagne et al. which found that the most common species involved was L. rhamnosus.11

Probiotics and dairy products with live cultures are generally thought to be harmless or at least very low risk for the general population. L. rhamnosus, specifically, remains a bacterial species with qualified presumption of safety in the European Food Safety Authority despite awareness of L. rhamnosus bacteremia and endocarditis cases.14,15 However, populations with severe or multiple co-morbidities, including valve replacement and immunosuppression/immunodeficiency, should avoid products that contain live bacteria in order to decrease unnecessary risk of severe infection or death.

4. Conclusion

Although consumption of products with live Lactobacillus species is common and generally safe for healthy individuals, in people with severe or multiple co-morbidities such as valve replacement, Lactobacillus bacteremia and endocarditis is possible, and may be fatal.

Supplementary Material. Transthoracic Echocardiogram Report on Hospitalization Day 2

Summary:

The study is of diagnostic quality.

1. Left ventricular ejection fraction, by visual estimation, is 55%.

2. Global left ventricular systolic function was normal.

3. Elevated left ventricular end-diastolic pressure.

4. Pseudonormal pattern of diastolic filling, consistent with grade 2 out of 4 LV diastolic dysfunction.

5. Severely dilated left atrium.

6. Moderately dilated right atrium.

7. There is no evidence of pericardial effusion.

8. Mitral annular calcification and degenerative mitral valve.

9. Moderate to severe aortic regurgitation.

10. The interatrial septum is not well seen.

11. Mild perivalvular leak.

12. DVI on prior 0.36 progression in AI, some views appears severe although PHT ~365 ms with multiple regurgitant jets noted.

13. The aortic valve peak pressure gradient is 30.5 mmHg and the mean pressure gradient is 15.0 mmHg.

14. The dimensionless index by Doppler interrogation is: 0.46 (<0.25 is Severe Prosthetic Valve stenosis).

Physician Interpretation:

Left Ventricle: Left ventricular ejection fraction, by visual estimation, is 55%. The left ventricular internal cavity size was normal. No left ventricular hypertrophy. Global LV systolic function was normal. Spectral Doppler shows pseudonormal pattern of diastolic filling, consistent with grade 2 out of 4 LV diastolic dysfunction. Elevated left ventricular end-diastolic pressure.

Right Ventricle: The right ventricular size is normal. Global RV systolic function is normal.

Left Atrium: The left atrium is severely dilated.

Right Atrium: The right atrium is moderately dilated.

Pericardium: There is no evidence of pericardial effusion.

Mitral Valve: The mitral valve has annular calcification and is degenerative in appearance.

Aortic Valve: The aortic valve is a normally functioning bioprosthetic valve. The dimensionless index by Doppler interrogation is: 0.46 (<0.25 is Severe Prosthetic Valve stenosis).Mild perivalvular leak. The aortic valve peak pressure gradient is 30.5mmHg and the mean pressure gradient is 15.0 mmHg. Moderate to severe aortic valve regurgitation is seen.

Tricuspid Valve: The tricuspid valve is normal in structure. Trace tricuspid regurgitation is visualized.

Pulmonic Valve: The pulmonic valve is not well visualized.

Aorta: The aortic root is normal in size.

Venous: The inferior vena cava is not well visualized.

Additional Comments: The interatrial septum is not well seen. In comparison to the previous echocardiogram( s): Prior examinations are available and were reviewed for comparison purposes. DVI on prior 0.36 progression in AI, some views appears severe although PHT ~365 ms with multiple regurgitant jets noted.