HomeCirculationVol. 145, No. 12Hemophagocytic Lymphohistiocytosis Associated With Endocarditis: A Case Years in the Making Free AccessCase ReportPDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessCase ReportPDF/EPUBHemophagocytic Lymphohistiocytosis Associated With Endocarditis: A Case Years in the Making Vanessa Blumer, MD, Joshua B. Parsons, MD, PhD, D. Ryan Anderson, MD, PhD, Gerald S. Bloomfield, MD, MPH and Cary Ward, MD Vanessa BlumerVanessa Blumer Correspondence to: Vanessa Blumer, MD, 2301 Erwin Rd, DUMC 3845, Room 7411A DN, Durham, NC 27701. Email E-mail Address: [email protected] https://orcid.org/0000-0001-5062-5015 Division of Cardiology, Department of Medicine (V.B., G.S.B., C.W.), Duke University Medical Center, Durham, NC. Duke Clinical Research Institute, Durham, NC (V.B., G.S.B.). Search for more papers by this author , Joshua B. ParsonsJoshua B. Parsons https://orcid.org/0000-0002-0589-3587 Division of Infectious Diseases, Department of Medicine (J.B.P.), Duke University Medical Center, Durham, NC. Search for more papers by this author , D. Ryan AndersonD. Ryan Anderson Division of Rheumatology, Department of Medicine (D.R.A.), Duke University Medical Center, Durham, NC. Search for more papers by this author , Gerald S. BloomfieldGerald S. Bloomfield Division of Cardiology, Department of Medicine (V.B., G.S.B., C.W.), Duke University Medical Center, Durham, NC. Duke Clinical Research Institute, Durham, NC (V.B., G.S.B.). Search for more papers by this author and Cary WardCary Ward https://orcid.org/0000-0002-5283-0490 Division of Cardiology, Department of Medicine (V.B., G.S.B., C.W.), Duke University Medical Center, Durham, NC. Search for more papers by this author Originally published21 Mar 2022https://doi.org/10.1161/CIRCULATIONAHA.121.057951Circulation. 2022;145:934–939Information about a real patient is presented in stages (boldface type) to expert clinicians (Drs Blumer, Parsons, Anderson, Bloomfield, and Ward), who respond to the information and share their reasoning with the reader (regular type). A discussion by the authors follows.Patient presentation: A 60-year-old man presented to a medicine service in February 2021 as a transfer from an outside hospital with nonremitting fatigue for at least 3 months. More recently, he was noted to have changes in mental status, mainly confusion and speech disturbances. At the time of admission, the patient denied fever, chills, night sweats, abdominal pain, and additional cardiac or neurological symptoms. He had a history of complex cardiac surgery in August 2019 that included (1) triple coronary artery bypass graft surgery, (2) mitral valve (MV) repair with neo chords to the posterior leaflet and annuloplasty with a complete ring, and (3) ligation of the left atrial appendage. Initial vital signs were temperature 98.7 °F, heart rate 90 bpm, blood pressure 128/80 mm Hg, and O2 saturation 97%. Cardiovascular physical examination was unremarkable, with no evidence of volume overload and normal cardiac auscultation. Admission laboratory tests were notable for pancytopenia (white blood cell count 3.1×109/L, hemoglobin 10.7 mg/dL, platelets 73×109/L) and elevated liver enzymes (aspartate aminotransferase 212 U/L, alanine aminotransferase 209 U/L, alkaline phosphatase 579 U/L, total bilirubin 1.7 mg/dL). A transthoracic echocardiogram demonstrated normal systolic function, mild mitral regurgitation, elevated gradients through the MV ring, and thickened MV leaflets with a mobile mass concerning for vegetation. A transesophageal echocardiogram redemonstrated a large vegetation encompassing and adherent to the anterior annulus and anterior leaflet that extended to the aortomitral continuity, and mild mitral stenosis (mean gradient 5 mm Hg and calculated pressure half-time MV area 1.8 cm2; Figure 1). Serial blood cultures were negative. Additional workup was concerning for embolic phenomena as an explanation for the altered mental status. Signs of systemic embolism included acute bihemispheric lacunar infarcts and splenic infarcts (Figure 2). The initial focused differential considered potential infectious and noninfectious causes for blood culture–negative endocarditis (BCNE).Download figureDownload PowerPointFigure 1. Transesophageal echocardiogram. A, Four-chamber view (0 degree) of the mitral valve showing thickened anterior mitral valve leaflet (blue arrow). B, Long-axis view (133 degrees) demonstrating thickened mitral valve leaflets (blue arrow) homogeneous echodensity around the sewing ring (green arrow) and extending into the aortomitral continuity (red arrow). C, Three-dimensional view mass-like thickening of the mitral valve (blue arrow). D and E, Three-dimensional en face view of the mitral valve using light source–based illumination to demonstrate tissue characteristics and leaflet thickening.Download figureDownload PowerPointFigure 2. Radiological imaging of central nervous system and abdomen/pelvis. A, MRI of the brain demonstrates 2 punctate foci of elevated diffusion signal in the right centrum semiovale favored to reflect acute infarcts, likely embolic. Punctate focus of elevated diffusion in the left basal ganglia and punctate focus in the left occipital lobe were also found (not pictured). B, Contrasted computed tomography of the abdomen and pelvis showed wedge-shaped low-attenuation lesions within the periphery of the spleen measuring up to 4.2×2.5 cm. Given the history of known mitral valve vegetations, this likely represents splenic infarction in the setting of systemic embolism.BCNE can account for up to 70% of all endocarditis, depending on case series and geography, and remains a diagnostic and therapeutic challenge.1 In general, it is defined as endocarditis without cause after inoculation of at least 3 independent blood samples in a standard blood-culture system with negative cultures after 7 days of incubation and subculturing.2 BCNE includes infectious and noninfective causes.3 On this basis, a broad differential of infectious causes was considered in this case: (1) bacterial endocarditis with blood cultures sterilized by previous antibacterial treatment, (2) endocarditis related to fastidious microorganisms in which prolonged incubation is necessary, and (3) true BCNE attributable to intracellular bacteria that cannot be routinely isolated. In addition, noninfective causes of BCNE, such as nonbacterial thrombotic endocarditis and endocarditis related to systemic diseases (ie, systemic lupus erythematosus), were also considered (Table). Best practice requires a thorough systematic and multidisciplinary approach because BCNE is often diagnostically challenging and requires modern techniques of histology, molecular analysis, and essential epidemiological information.2Table 1. Common Causes of Blood Culture Negative EndocarditisCauseEpidemiological risk factorsDiagnostic testsInfectious Sterilized blood culturesBlood cultures drawn after antibiotic therapy– Coxiella burnetiiAbattoir workers, contaminated milk and farm animals (cattle, goats, sheep)Serology, PCR (serum or valve tissue) Bartonella speciesCat exposure, homelessness, liceSerology, PCR (serum or valve tissue) Brucella speciesUnpasteurized animal products, farm animals, travel to endemic area (Middle East, Central Asia, China, Central/South America, India)Serology, PCR (valve tissue) Tropheryma whippleiOccupational exposure to farm animals or soilPCR (valve tissue) Mycobacterial speciesNontuberculous mycobacteria: Cardiac surgery involving cardiopulmonary bypass, prosthetic valve.Mycobacterium tuberculosis: Born in endemic area, household contact with tuberculosis, homelessness and other M tuberculosis risk factorsMycobacterial blood culture, valve mycobacterial culture, valve histopathology, PCR (valve tissue) Atypical organisms (ie, Legionella species, Mycoplasma species)Cardiac surgeryLegionella species: urinary antigen, PCR (valve tissue)Mycoplasma species: culture and PCR (valve tissue) Fungal speciesProsthetic heart valve, intravenous drug use, solid-organ transplant, hematologic malignancy, human immunodeficiencyValve tissue histopathology, valve culture, fungal blood cultureNoninfectious Nonbacterial thrombotic (marantic)Underlying malignancy, autoimmune disorder, or antiphospholipid syndrome–* Secondary to systemic illnessSystemic lupus erythematosus, Behçet disease, rheumatoid arthritis, antiphospholipid syndrome–*PCR indicates polymerase chain reaction.* Clinicians are reliant on a constellation of epidemiological risk factors, clinical, echocardiographic, and absence of microbiological findings for the diagnosis.Patient presentation (continued): The infectious diseases consulting service was engaged, and the patient was started on broad spectrum antibiotic therapy with vancomycin and ceftriaxone. Extensive infectious workup was negative, and evaluation for the most common causes of noninfectious BCNE including rheumatologic, hypercoagulable, and malignant causes was initially unrevealing. The patient was transferred to the inpatient cardiology service for further management of presumptive endocarditis involving the MV. The morning after his transfer, he was noted to be febrile (103.1 °F) and developed acute worsening mental status. Repeat brain imaging was unchanged and cerebrospinal fluid analysis was unremarkable. The initial working diagnosis was challenged, and the differential diagnosis was refined to now account for neurological symptoms out of proportion to findings on brain imaging. The patient was found to meet 5 of 8 criteria for hemophagocytic lymphohistiocytosis (HLH; HLH 2004 Diagnostic Criteria), including fever, splenomegaly, pancytopenia, hypertriglyceridemia (394 mg/dL) and hypofibrinogenemia (76 mg/dL), and hyperferritinemia (4636 ng/mL).4 Furthermore, an H-Score of 268 supported this diagnosis with a >99% probability. Rheumatology was consulted and therapy was started with dexamethasone 5 mg/m2 twice a day and anakinra 100 mg every 6 hours (an interleukin-1 receptor antagonist). This resulted in marked improvement in his mental status and resolution of fevers within 24 to 48 hours. Although this presentation and response to therapy seemed to be consistent with secondary HLH, the underlying trigger for HLH and the cause of his MV pathology remained unclear. HLH is a severe hyperinflammatory syndrome induced by overactive but ineffective members of the innate immune system leading to aberrant activation of many immune cells, including cytotoxic T cells and macrophages. The primary (genetic) form is more common in children, and it is caused by mutations affecting lymphocyte cytotoxicity and immune regulation. Secondary (acquired) HLH is more frequent in adults and is commonly triggered by infections or malignancies but may also be induced by autoinflammatory/autoimmune disorders, in which case it is called macrophage activation syndrome (or macrophage activation syndrome-HLH).4 Most data on the diagnosis and treatment of HLH come from the pediatric literature, raising several challenges in the treatment of atypical HLH cases in adult population.Patient presentation (continued): Genetic evaluation for primary HLH was negative. To clarify the potential trigger for secondary HLH, our patient underwent workup for underlying malignancy and autoimmune disease, all of which was unrevealing. Although initial infectious workup was negative, several days into the hospital stay Mycobacterium chimaera was detected on next-generation sequencing of microbial cell-free DNA (Karius test). The Karius test is a commercially available blood test based on next-generation sequencing of microbial cell-free DNA that can identify and quantify >1000 clinically relevant pathogens, including bacteria, DNA viruses, fungi, and parasites.5 Proposed applications include complicated pneumonia, infections in immunocompromised patients, and endocarditis. There is still a scarcity of data about the real-life performance of culture-independent detection systems for bloodstream and deep-seated infections, and well-designed studies are awaited for assessing the impact of these emerging technologies on patient outcomes.6 Although blood cultures continue to be the gold standard, microbial cell-free DNA testing can become a valuable tool in patients with BCNE where the causative organism remains unclear despite thorough exposure and risk factor screening. In our case, microbial cell-free DNA testing was crucial in the early detection of M chimaera. M chimaera is a nontuberculous mycobacterium belonging to the Mycobacterium avium complex, described for the first time in 2004, and more recently associated with a global outbreak of disseminated and localized disease related to exposure to the Sorin Stockert 3T heater-cooler unit (HCU) used during cardiopulmonary bypass.7,8 During the past decade, M chimaera disseminated infections after cardiothoracic surgery, especially open-heart surgery, have been increasingly reported worldwide. From a pathogenic standpoint, it is acquired during cardiopulmonary bypass through bioaerosols emitted from contaminated HCU water systems. Patients with M chimaera infections after cardiac surgery have a range of clinical presentations, but the most common are cardiac manifestations including endocarditis, vascular graft infection, and mycotic aneurysm. Surgical site infections including mediastinitis and sternal wound infections have also commonly been described.7,8 Because of nonspecific symptoms and long latency, postoperative M chimaera infections may not be promptly diagnosed, and delays in treatment can become life-threatening. In cases of M chimaera, the indication and timing of surgical intervention appears to be critical for successful outcomes, and antibiotic therapy should ideally be based on drug susceptibility testing results.7 In the absence of other infections and a negative workup for malignancy and autoimmune disease triggering secondary HLH, endocarditis attributable to M chimaera acquired during cardiac surgery 2 years before the index admission, complicated by HLH, was considered the most likely unifying diagnosis in this complex clinical case.Patient presentation (continued): The patient and responsible institution were notified of possible M chimaera infection following the guidelines by the International Society for Cardiovascular Infectious Diseases and in accordance with guidance from the US Centers for Disease Control and Prevention and the Food and Drug Administration. The patient was empirically treated for disseminated nontuberculous mycobacterium infection with amikacin, azithromycin, ethambutol, and rifampin. MV surgery was initially deferred to allow for treatment of systemic infection and nutritional support, yet eventually the patient underwent successful bioprosthetic MV replacement with a 25-mm Carpentier-Edwards valve without complication. MV histopathologic tissue examination demonstrated active endocarditis (Figure 3) and MV culture confirmed the diagnosis of M chimaera infection. Timing of surgery was debated, but early surgery was favored in this case for source control, given concern for systemic embolism. He was continued on combination antibiotic therapy with plans to treat for at least 12 months from MV replacement. Steroids and anakinra were slowly tapered on an outpatient basis.Download figureDownload PowerPointFigure 3. Histopathologic slides. A and B, Low-power view (2×) hematoxylin and eosin stain of valvular tissue with evidence of brisk active endocarditis and fibrin deposition. C, Higher-power view (10×) of an organizing lymphohistiocytic inflammatory infiltrate with scattered neutrophils involving the valve tissue.DiscussionAcquired HLH can pose a clinical conundrum for clinicians, and rigorous investigation into past medical and surgical history is critical in the diagnosis. Furthermore, a high level of awareness is essential to suspect Mchimaera infection. Although the worldwide, clonal outbreak of Mchimaera infections related to contaminated HCUs was initially reported by the European Centers for Disease Control and Prevention in April 2015, invasive infections caused by Mchimaera and other species of nontuberculous mycobacterium continue to occur, posing ongoing clinical and infection prevention dilemmas. The primary clonal Mchimaera outbreak was linked to contamination that occurred at a Sorin HCU manufacturing center in Germany; however, the worldwide investigation that followed found that other HCU models produced by different manufacturers, and hospital water sources, as well, were also contaminated with Mchimaera.9 Therefore, despite mitigation measures implemented at most hospitals to decrease the risk of Mchimaera infection after cardiac bypass surgery, these infections can continue to occur, and ahigh index of suspicion is necessary to make a prompt diagnosis. In our reported case, we confirmed that the center performing the patient’s cardiac surgery had replaced their Sorin HCUs 2 years before the patient’s procedure and this was the first possible case of Mchimaera infection. However, since the clinical syndrome and incubation period of this patient are consistent with most reported cases of HCU-related nontuberculous mycobacterium infections and because we do not have a plausible alternative mechanism for endocarditis with Mchimaera, we are confident that inoculation was related to HCU contamination and occurred during cardiac surgery 2 years before his index admission. To our knowledge, this is only the second case documented in the literature of HLH with Mchimaera precipitated by an infected valve but the first to have a successful outcome, including postdischarge patient survival.10Significant diagnostic and therapeutic challenges existed in our case, and there were tremendous learning opportunities across different medicine subspecialties. This case illustrates (1) HLH as an underappreciated entity in adults and infectious endocarditis as a possible trigger, (2) the broad differential diagnosis of BCNE and microbial cell-free DNA testing as a means of more rapid identification of a causative pathogen, and (3) the need for improved awareness of M chimaera infection and its early consideration as a potential cause of infections after cardiopulmonary bypass surgery.Article InformationAcknowledgmentsThis case truly illustrates effective multidisciplinary care at its finest. The authors acknowledge the instrumental contributions of Drs Omar Martinez-Uribe, Michael Sketch, Carolyn Glass, Jordan Pomeroy, Brian Andonian, Sana Arif, Joseph W. Barwatt, Manuela Carugati, and Eileen Maziarz, although they were not able to be listed as authors in this case report.Sources of FundingNone.Disclosures None.FootnotesFor Sources of Funding and Disclosures, see page 939.Circulation is available at www.ahajournals.org/journal/circCorrespondence to: Vanessa Blumer, MD, 2301 Erwin Rd, DUMC 3845, Room 7411A DN, Durham, NC 27701. Email vanessa.[email protected]edu