A case series of Corynebacterium striatum native valve infective endocarditis

Abstract

Corynebacterium species isolated in blood cultures are commonly dismissed as a contaminant. They are also recognized as an uncommon pathogen in infective endocarditis. We report two cases of native valve endocarditis due to Corynebacterium striatum. The first patient, a 36-year-old female with hemolytic anemia, whose risk factor for endocarditis was a Port-a-Cath (Smiths Medical, Los Angeles, California) used for routine blood transfusions. She was diagnosed with triple valve endocarditis via transthoracic echocardiogram. Her multiple comorbidities made her a poor surgical candidate for valve replacement and she elected to go on hospice care after antibiotic treatment completion. The second patient, a 46-year-old, was found to have coronavirus disease 2019 (COVID-19) pneumonia in addition to persistent Corynebacterium striatum bacteremia. A transthoracic echocardiogram was highly suggestive of aortic valve endocarditis. A confirmatory transesophageal echocardiogram was unable to be obtained given his clinical instability and COVID-19 status. Unfortunately, this patient expired due to complications of severe COVID-19 pneumonia. We highlight the need for prompt recognition of risk factors of infective endocarditis due to uncommon pathogens that may aid in the diagnosis and treatment, while utilizing a multidisciplinary approach.

Learning objective

The aim of this case series is to emphasize the importance of Corynebacterium species as a cause of native valve infectious endocarditis and to illustrate the challenges it poses in diagnosis and management.

Introduction

Corynebacterium species are aerobic, non-sporulating, Gram-positive bacilli comprising more than 100 species of which a substantial proportion has been associated with human infections [1]. Often, when Corynebacterium species are isolated from blood cultures, they are dismissed as a contaminant as they are part of the normal skin flora and mucosal membranes. Of the non-diphtheriae Corynebacterium species, Corynebacterium striatum and Corynebacterium jeikeium have often been reported as pathogens [[2], [3], [4]]. However, there are insufficient data on how species determination can help in assessing the risk for true infection or even infective endocarditis (IE) in patients with Corynebacterium bacteremia [5].

Available studies describe a limited number of cases of IE due to Corynebacterium species. It has mostly been documented in patients with prosthetic heart valves and those with indwelling cardiovascular medical devices, with only 0.2–0.4% of cases of native valve endocarditis [4]. Our series aims to describe two cases of C. striatum native valve IE, reinforcing the potential virulence of this pathogen. It also emphasizes the importance of a multidisciplinary team approach to ensure a correct diagnosis as well as appropriate management.

Case report

Case 1

A 36-year-old female with Gorham-Stout syndrome, Bell's palsy, heart failure with preserved ejection fraction, chronic severe lower extremity lymphedema, and transfusion-dependent hemolytic anemia presented for a routine clinic visit for a scheduled blood transfusion. Upon arrival, she reported chills, shortness of breath for 24 h, chest tightness along with sharp chest pain upon inspiration, and worsened generalized weakness compared to baseline. She was found to be hypoxic, requiring 2 l of oxygen, febrile to 38.2 °C, and hypotensive. The patient was referred to the emergency department (ED) for further work up. In the ED, she denied known coronavirus disease 2019 (COVID-19) exposure, environmental exposures, recent travel, or sick contacts. She reported having dogs at home but denied routine interactions. She denied any pain surrounding her Port-a-Cath (Smiths Medical, Los Angeles, California).

Laboratory studies were remarkable for leukocytosis of neutrophilic predominance, in addition to chronic anemia and acute thrombocytopenia. COVID-19 nasopharyngeal polymerase chain reaction (PCR) was negative. The patient was admitted with sepsis of unknown source, and started on broad-spectrum antibiotics with vancomycin and cefepime. Routine urine and blood cultures were obtained. A computed tomography of the chest showed small bilateral pleural effusions with compressive atelectasis. Soon thereafter, admission blood cultures revealed C. striatum growth in four sets. Repeat blood cultures the following day continued to grow C. striatum in two sets despite treatment with vancomycin. The infectious diseases team was consulted and recommended a transthoracic echocardiogram (TTE) be obtained due to sustained bacteremia in the setting of an existing indwelling catheter. The TTE revealed diffuse thickening of the aortic valve leaflet tips consistent with valve vegetations, with a large vegetation noted on the aortic valve. There was also a smaller vegetation seen on the posterior leaflet of the mitral valve, and an associated vegetation on the pulmonic valve (Fig. 1, Fig. 2). Given these findings, the Port-a-Cath (Smiths Medical, Los Angeles, California) was removed as it was perceived to be the source of bacteremia; cefepime was also discontinued. Bacteremia clearance was documented after four days of vancomycin therapy.

Fig. 1.

Parasternal long-axis view demonstrating large aortic valve vegetation and small mitral valve vegetation on the posterior leaflet.

Fig. 2.

Parasternal short-axis magnification of thickened pulmonic valve leaflets consistent with endocarditis.

The cardiology department was again consulted one month later due to worsened chest pain at which time a repeat TTE was performed. The patient was found to have diffusely thickened aortic and pulmonic valves with apparent resolution of the previously noted large aortic vegetation. There was, however, new severe aortic and pulmonic regurgitation. The cardiothoracic surgery service was consulted due to findings of multiple valve endocarditis with severe pulmonic and aortic insufficiency. The patient was deemed a poor surgical candidate for triple valve replacement due to severe deconditioning and severe thrombocytopenia in addition to other debilitating comorbidities. She completed six weeks of treatment with vancomycin monotherapy in the hospital, after which point the patient elected to be discharged on hospice care.

Case 2

A 46-year-old male with uncontrolled type 2 diabetes, rheumatoid arthritis, anti-neutrophil cytoplasmic antibodies + vasculitis, benign prostate hyperplasia with urinary retention, paraplegia with a chronic stage four coccygeal pressure injury, who presented with shortness of breath and chest pain of sudden onset, worsened with deep inspiration, located on the center of the chest. He reported generalized malaise and subjective fevers for the previous three days; said that he did not have an associated cough or any known COVID-19 exposures. He was noted to be febrile to 39.2 °C, tachycardic, tachypneic, with elevated C-reactive protein, lactic acidosis, and without leukocytosis. Intravenous fluids were initiated, as well as empiric vancomycin and piperacillin-tazobactam.

A COVID-19 nasopharyngeal PCR test returned positive, which was consistent with his presentation. However, admission blood cultures later grew C. striatum in four bottles. The infectious diseases team was consulted. Upon chart review, it was noted that the patient had multiple recent blood cultures from previous admissions, dating up to one month previously, with C. striatum deemed to be contaminants during those encounters. It was recommended to obtain further imaging and vancomycin monotherapy be continued. Bacteremia clearance was documented approximately 24 h after antibiotic initiation.

Magnetic resonance imaging of the pelvis revealed bilateral trochanteric bursitis as the presumed source of infection. Given the findings of persistent and prolonged bacteremia of approximately one month, a TTE was performed. It showed at least two jets of aortic regurgitation consistent with possible aortic valve endocarditis (Fig. 3). There were no signs of paravalvular abscess on TTE. Although a confirmatory transesophageal echocardiogram was recommended, it was not performed due to the patient's positive COVID-19 status. Had further imaging been obtained, there may have been a more definite demonstration of aortic valve destruction and/or paravalvular abscess. Therefore, the decision was made to treat bilateral trochanteric bursitis with presumptive native aortic valve endocarditis with six weeks of vancomycin monotherapy. Unfortunately, his hospital course culminated in death after suddenly developing asystole with unsuccessful cardiopulmonary resuscitation attempts and an acute ischemic stroke.

Fig. 3.

Magnified apical five-chamber view demonstrating two new separate aortic regurgitation jets with color Doppler.

Discussion

IE remains a deadly disease with in-hospital mortality rates reported as high as 24% [6]. Given the often-vague presenting symptoms, rapid recognition and diagnosis remain at the forefront of these cases. This case series offers several valuable learning opportunities, the first being the patient characteristics predisposing them to IE prior to diagnosis. While underlying structural heart disease and prosthetic valves remain the two largest predisposing factors to development of IE, other factors include injection drug use, presence of an intravascular catheter, large pre-existing wounds, and recurrent healthcare visits (most often in the form of hemodialysis) [7]. Multiple factors were present in both of our patients; the first having an intravascular catheter, the second having a large coccygeal wound, and both having frequent healthcare visits. Identifying predisposing factors, in addition to presenting symptoms, can help guide appropriate management more rapidly.

Fortunately, in the case of IE, the modified Duke criteria remain at the forefront of diagnosis for endocarditis [8]. The drawbacks in both the Duke criteria and our perception of Corynebacterium species, however, were clearly demonstrated in Case 2. The timing between the first positive blood cultures for Corynebacterium and TTE was almost two weeks and across two separate hospitalizations. The Duke criteria play a role in this as Corynebacterium is not one of the pathogens typically thought to cause IE, which would satisfy one of the major criteria for diagnosis [9]. Due to the fact that Corynebacterium is often considered a contaminant and not a common pathogen for IE, there were a total of 7 blood culture bottles positive for Corynebacterium prior to TTE assessment of cardiac involvement. Although the Duke criteria do not account for bacteria not thought to be “typical” for endocarditis, the persistent positive blood cultures raised concern for a cardiogenic source of infection.

An additional consideration, and often challenge, taken in cases of IE is if the patient would meet criteria for a surgical intervention rather than the sole use of prolonged antibiotics. Different classification of recommendations exists for surgical intervention. However, broadly speaking, the indications for surgery include: infecting organism, vegetation size (>10 mm), presence of perivalvular infection, presence of embolism or heart failure, non-cardiac comorbidities, and available surgical expertise [10]. The first presented case clearly demonstrates a patient who meets surgical criteria, however for multiple reasons was a poor surgical candidate. Unfortunately, it is often the case that the patients who are in most need of surgical intervention, are often the ones with the most pre-existing conditions that may preclude them from surgery. Additionally, the timing of surgical intervention, be it early or late, has been a discussed topic, however complete evidence for one versus the other has yet to emerge.

In conclusion, these cases bring to light multiple crucial factors to consider in regard to the diagnosis and overall treatment of IE. These include the rapid suspicion of those who are most at risk, in addition to identifying those patients who may qualify for more invasive measures, and utilizing a multidisciplinary approach for the management of these patients. Most importantly is the realization that bacteria, such as Corynebacterium species, which may often be considered contaminants can in fact be the pathogen causing IE and lead to devastating consequences if not quickly identified.

Declaration of competing interest

The authors certify that they have no affiliations with or involvement with any organization or entity with any financial interest, or non-financial interest in the subject matter or materials discussed in this article.