Blood Culture-Negative Cardiovascular Syphilis in an Untreated HIV-Positive Man

Introduction

Syphilis is a systemic infectious disease caused by the bacterium Treponema pallidum [1]. The incidence of syphilis has increased in the United States despite widespread preventive and treatment protocols [2]. Syphilis can be divided into early- or late-stage syphilis [3]. Up to 10% of untreated tertiary syphilis patients develop cardiovascular syphilis, and 70% to 80% of these result in syphilitic aortitis [4,5]. T. pallidum-related aortic valve endocarditis is very rare [6].

Coinfection between human immunodeficiency virus (HIV)-infected individuals and syphilis is reported to be around 20% overall, but is higher in males [7,8]. Whether HIV infection alters the progression of syphilis remains controversial. It appears that the time needed to develop tertiary syphilis in untreated HIV-positive patients is shortened, resulting in earlier onset of cardiovascular and neurologic sequelae [8].

Infective endocarditis (IE) is an infection in the endothelium of the heart and is potentially life-threatening [9]. Blood culture-negative endocarditis (BCNE) occurs when typical blood culture protocols fail to grow a microbial pathogen. While the prevalence of BCNE varies widely in the literature, it makes up 20% to 30% of all infective endocarditis cases [10,11]. In a case of endocarditis caused by tertiary syphilis, the diagnosis is made with serological testing and molecular techniques since the fastidious spirochete does not grow on the routine blood culture protocol.

BCNE can often be insidious in nature, presenting with nonspecific symptoms, which can mislead practitioners attempting to identify the underlying cause of the patient’s illness. The broad range of culprits, including various fungal and bacterial species, adds complexity to isolating the correct organism. Additionally, recent exposure to antimicrobial therapy can further obscure the differential diagnosis.

Despite existing literature on BCNE and aortitis caused by syphilis, there is limited awareness of syphilis as a potential causative agent of BCNE, especially in the setting of untreated HIV, which demonstrates the value of the present case. Furthermore, our report highlights the diagnostic value of next-generation sequencing techniques such as 16S rRNA sequencing in identifying the causative agent when conventional methods fail, as illustrated by our patient with untreated HIV-with multifocal pneumonia and aortic valve malfunction secondary to cardiovascular syphilis. We present a case of T. pallidum-related aortic valve endocarditis without accompanying aortitis in an HIV-positive man.

Case Report

A 44-year-old man presented to the Emergency Department with a 48-hour history of shortness of breath, cough, congestion, fever, and chest discomfort. His past medical history was positive for HIV, diagnosed 4 years prior to admission and treated with bictegravir-emtricitabine-tenofovir alafenamide. He ran out of his medication approximately 8 months ago. He denied receiving a diagnosis of acquired immunodeficiency syndrome (AIDS), reported no history of sexual intercourse over the past year, and stated he had sexual contact with women only. According to the available records, 3 years before admission, he was diagnosed with late latent syphilis and was scheduled to receive 2.4 M units of benzathine penicillin G, administered as 3 intramuscular injections, 1 each week for 3 consecutive weeks, but he only received 1 injection. He had a remote history of intravenous drug use. On exam, there was no lymphadenopathy, no alopecia, no rash, and no mucosal lesions. There were no findings of secondary syphilis with absent papulosquamous non-itchy eruptions. The neurological examination was normal. The cardiovascular examination revealed a 4/6 diastolic murmur, best heard in the right second intercostal space, along with significant dental disease.

On admission, his complete blood count (CBC) indicated concerns for microcytic anemia, with a hemoglobin (Hgb) level of 11.7 gm/dL (normal range: 13.2–16.6 gm/dL), and a mean corpuscular volume (MCV) of 77.2 fL (normal range: 80–100 femtoliters [fL]). His white blood cell (WBC) count was 8800 cells/μL (normal range, 4500–11 000 cells/μL). His complete metabolic profile (CMP) showed a notable serum sodium level of 131 mEq/L (normal range: 135–145 milliequivalents per liter [mEq/L]) and a calculated osmolality of 270 mOsm/kg (normal range: 275–295 milliosmoles per kilogram [mOsm/kg]). He had a detectable HIV RNA viral load of 13 100 copies/mL, with a CD4 count of 509, at 15%. Additionally, hepatitis C virus (HCV) antibodies tested positive, and follow-up viral load revealed 2 600 000 copies.

Computed tomography (CT) chest imaging revealed multifocal pneumonia and bilateral axillary lymphadenopathy. The patient began parenteral piperacillin-tazobactam therapy at 4.5 g over 4 hours every 8 hours along with azithromycin therapy (500 mg intravenous daily). Of note, the oxygen saturation on admission was 97%.

An echocardiogram was ordered for further evaluation of the patient’s chest discomfort, which revealed a large mobile mass on the right coronary cusp and non-coronary cusp of the aortic valve (Figures 1, 2), along with severe aortic regurgitation.

Concern shifted from multifocal pneumonia to infective endocarditis. At this point, piperacillin-tazobactam was discontinued, and the patient was started on empiric vancomycin at an initial dose of 1500 mg followed by 1 g every 12 hours and a daily intravenous dose of ceftriaxone of 2 g.

Blood cultures were unrevealing, and considering his immunocompromised state, a broad workup for BCNE was initiated. Serum Cryptococcus antigen, Tropheryma whipplei PCR, Coxiella burnetii, and Bartonella henselae serologies all returned negative results. Follow-up Microbial Cell-Free DNA testing also proved to be uninformative. Referring back to the patient’s significant dental disease, he underwent alveoloplasty and extraction of teeth in an effort to achieve source control prior to any decision on aortic valve repair/replacement.

His repeat RPR test was positive, with titer=1: 128 (normal <1: 16). Unfortunately, no prior titers could be found for comparison. Due to the patient’s prior history of syphilis, HIV, and lack of complete therapy he received a 2.4-M unit dose of benzathine penicillin G before undergoing aortic valve replacement surgery. A lumbar puncture was not performed.

Shortly thereafter, he underwent aortic valve replacement using an Inspiris 25-mm valve. There were no obvious signs of infection upon examination of the native aortic valve intraoperatively. Intraoperative Gram staining was negative for white blood cells (WBCs) and any microorganisms. Cultures showed no growth. Pathology of the native aortic valve was consistent with myxoid degeneration along with chronic and mildly acute inflammation, further confirming suspicions of BCNE. A spirochete stain was also negative. Aortic valve samples were sent for 16S ribosomal ribonucleic acid (rRNA) bacterial polymerase chain reaction (PCR) analysis (16 sRNA) and tested positive for Treponema pallidum, confirming the diagnosis of Treponema pallidum-induced infective endocarditis.

After 16 sRNA findings were revealed, antibiotic therapy was adjusted. Vancomycin was discontinued, and the patient remained on ceftriaxone therapy along with daily doses of aqueous penicillin G (24-M units) during the remainder of his admission. Due to penicillin supply chain shortages, he was discharged with 6 weeks of ceftriaxone therapy, with recommendations to initiate a 12-week course of sofosbuvir/ledipasvir for HCV treatment in the outpatient setting. His outpatient follow-up course with the Infectious Disease (ID) clinic has been complicated by a lack of follow-through by the patient.

Discussion

We describe a highly complex case involving 2 life-threatening diseases: BCNE secondary to tertiary syphilis and untreated HIV, along with serious comorbidities of HCV and generalized dental decay. It becomes even more challenging because of the patient’s choice to disregard health protocols and present only in a clinically deteriorated state; this unfortunately left little opportunity for an earlier diagnosis to be made.

The syphilitic BCNE was the reason for his presentation to our service, but his untreated HIV status and decision to stop his medication possibly influenced the rapid onset of syphilitic endocarditis symptoms [8,12].

There are other data suggesting that the presence of HIV does not impact IE [13]. However, these data do not mention whether the T. pallidum, HIV-positive patients were treated with antiretrovirals. Cardiac valvular involvement in tertiary syphilis is reported to be extremely rare [6] in previously healthy native valves, but there are no definitive data on the effect of HIV on the natural progression of cardiovascular syphilis. It may be that the immunocompromised state of untreated HIV, combined with untreated or partially-treated late syphilis, results in more valvular involvement, and because it is so difficult to diagnose, it has been overlooked and underreported in favor of aortitis [5].

The diagnosis of T. pallidum endocarditis is not possible by routine serology, culture, or microscopy. Until his 16S rRNA PCR result returned, the exact etiology of his valvular pathogen was unknown. Advanced molecular testing using 16S rRNA PCR in explanted tissue is now an important method to determine the etiology of BCNE, and in this case, the T. pallidum [14]. It is now a routine step in diagnosing BCNE, and the information it provides is critical to designing an antibiotic treatment plan.

The preferred and alternative treatments for tertiary cardiovascular syphilis are intramuscular (IM) benzathine penicillin G, 2.4 million units once weekly for 3 weeks, and ceftriaxone 2 g daily IM or IV for 10 to 14 days, respectively. Oral doxycycline is also an alternative [15]. While expert opinions vary, the accepted protocols state that the antibiotic management of syphilis is unchanged by HIV status. However, neurosyphilis treatment is increased in cases of tertiary syphilis with high RPRs (>1: 32) and absent retroviral therapy [16]. Our patient was thus at increased risk for neurosyphilis. The preferred and alternative treatments for neurosyphilis are aqueous penicillin G 3 to 4 million units IV every 4 hours (or 18 to 24 million units continuous IV infusion) for 10 to 14 days and ceftriaxone 2 g IM daily for 14 days [16].

Despite its rarity, syphilitic endocarditis warrants careful consideration due to the risk of being overlooked, as the classic visual characteristics of syphilis can often be absent in the tertiary form of this disease [17]. The pathophysiology of syphilitic endocarditis typically involves infection in the media of the aorta, with subsequent weakening and potentially forming an aneurysm. When the aortic valve is involved, the differential diagnosis is wide and includes other causes of valvulitis, including atherosclerosis. As noted, our patient had an extensive workup for valvular disease on admission.

Given the lack of documentation and information volunteered by the patient, it is not possible to say with certainty the length of time he was living with this infection prior to its initial diagnosis. An opportunity to prevent a progression of syphilis was clearly missed. Early detection and completed therapy can potentially reduce the severity of cardiovascular syphilis and the necessity for complex interventions [18].

In an ideal setting, such a timeline could be used to help determine the duration of time needed for syphilitic endocarditis to manifest. This point has been illustrated by a previous report of a patient who developed syphilitic endocarditis approximately 4 years after initial infection, suggesting that an early onset is possible [17]. It is important to note that the patient in the previous report had a bicuspid aortic valve, raising the question of valvular anatomy and its significance in the timeline of developing this disease.

Conclusions

The unique element to this case is the presence of aortic valve involvement from syphilis without evidence of involvement of either the ascending or descending aorta in an HIV-positive patient. Whether this reflects a change in the natural history of cardiovascular syphilis because of his comorbidities is not known, but it would seem prudent that in patients with untreated HIV and evidence of structural valvular disease in a clinical setting of late syphilis, T. pallidum should be considered as a possible causative agent and raise awareness of the possibility of syphilis.

This case has significant social and healthcare implications, as syphilis can be viewed as a marker of social disparities, disproportionately affecting disadvantaged individuals within our socio-economic system. It is our hope that this case demonstrates the desperate need to increase health literacy and thus empower individuals to take an active role in managing their health. Additionally, this case serves as a potent reminder regarding the need to adhere to screening guidelines (such as those recommended by the US Preventive Services Task Force [19]); which have shown a net benefit in early detection and prevention within our population, particularly among populations at increased risk for syphilis infection [18].

Lastly, this case demonstrates the emerging role of next-generation sequencing in the diagnosis of culture-negative endocarditis, offering a powerful diagnostic tool when conventional methods fail to reveal a causative agent. It is our hope that this case adds to the clinical awareness and diagnostic arsenal of physicians, enabling earlier recognition and intervention in similarly challenging presentations.