Navigating Complex Diagnostics During COVID-19: Repeated Testing Unveils Infective Endocarditis in a 61-Year-Old Woman

Background

Infective endocarditis (IE) is a systemic septic disease with various clinical manifestations, including bacteremia, vascular embolism, and cardiac injury caused by the formation of vegetation containing bacterial aggregation on the heart valve membrane, endocardium, and intima of the vessels [1]. Accurate diagnosis of IE is essential because IE frequently leads to severe complications and sometimes fatality if not treated appropriately. However, owing to the various IE symptoms, the diagnosis of IE may be delayed because it could be mistaken for other diseases that cause fever or the features of IE can be masked by coexisting diseases [2]. Here, we present a case of delayed diagnosis of IE during the coronavirus disease 2019 (COVID-19) pandemic owing to inattentive physical examination, cognitive errors, and systemic patient overload.

Case Report

A 61-year-old female patient presented to another hospital with a chief concern of a low-grade fever of around a range of 38°C, night sweats, and polyarthritis lasting more than 1 month. At the time, Japan was experiencing the COVID-19 pandemic. The result of a routine reverse transcription polymerase chain reaction (PCR) assay for acute severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was negative. The patient underwent a basic physical examination, laboratory test for inflammatory markers, and 2 sets of blood cultures and was placed on follow-up observation. The blood test revealed a high inflammatory response, and the blood cultures were negative. Computed tomography (CT) scans showed no significant findings leading to the cause of fever, including paravalvular lesions of the heart, and various blood tests, such as antinuclear antibodies, rheumatoid factor, and urine dipstick, were inconclusive; therefore, the patient was admitted to our hospital for further examination for fever of unknown origin on the 35th day from onset.

On admission, her vital signs showed a body temperature, pulse rate, blood pressure, and respiratory rate of 38.4°C (101.1°F), 99 beats/min, 122/72 mmHg, and 20 breaths/min, respectively. She denied experiencing any dyspnea and chest pain. After a detailed interview, it was found that the patient had a history of mitral valve regurgitation (MVR) (once she pointed it out, and she had no follow-up), which had not been ascertained by the previous physician. Her physical examination was unremarkable except for a holosystolic heart murmur (Levine 3/6), with the strongest point at the cardiac apex. Laboratory testing revealed a white blood cell count of 9.1×103/μL (neutrophils: 88%) and a C-reactive protein level of 3.6 mg/dL (normal range, ≤0.3 mg/dL). Liver and kidney functions were both within the normal range. Her urine test result was also within normal limits. The admitting team performed a transthoracic echocardiogram (TTE) that showed mitral valve prolapse with mild to moderate regurgitation but failed to reveal evidence of IE.

The diagnosis remained uncertain. A positron emission tomography/CT scan was considered to identify the cause of the fever (which would indicate vasculitis or a tumor). On the third day of admission, a meticulous clinical examination by another physician revealed conjunctival hemorrhage and painless nodules (Janeway lesions) on the soles of the feet (Figures 1, 2).

Considering IE, 3 new sets of blood cultures were performed, and Streptococcus oralis was detected. A transesophageal echocardiogram showed prolapse of both mitral valves with mild to moderate regurgitation; the A3-P3 area was thickened, and regurgitation was observed from the same area toward the posterior wall. The anterior cusp of the mitral valve was thickened and partially mobile on the surface. Vegetation was considered, and it was grouped together with the valve measuring 4.6×3.1 mm (Figure 3). The right atrium and ventricle sizes were normal. The left atrium was mildly dilated with normal left ventricular size and mild systolic dysfunction. The clinical course met the defined criteria for endocarditis using the Modified Duke’s Criteria shown by 1 major criterion (positive blood cultures comprising a typical organism for endocarditis from 2 separate blood cultures) and 3 minor criteria (fever >38°C, Janeway’s lesions, and conjunctival hemorrhage). Furthermore, this clinical course met the definite criteria for IE of the 2015 European Society of Cardiology (ESC) guidelines for managing IE [3] shown by 2 major criteria (positive for all 3 separate cultures of blood and echocardiogram positive for vegetation) and 3 minor criteria (predisposing heart condition, fever >38°C, conjunctival hemorrhage, and Janeway’s lesion). Following these criteria, a diagnosis of IE was formulated. After 6 weeks of 2 g ampicillin via intravenous drip every 4 h, the patient recovered and is currently doing well.

Discussion

We present a case of delayed diagnosis of IE during the COVID-19 pandemic owing to inattentive physical examination, cognitive errors, and systemic patient overload, and present 3 major learning points in this case.

The first learning point is that during the COVID-19 pandemic there was a tendency for doctors to focus on eliminating the diagnosis of COVID-19 first (COVID-blindness), thereby delaying the treatment of other diseases (eg, bacteremia) [4]. Although conjunctival hemorrhage and Janeway lesions are only present in approximately 5–10% of patients with IE, it is important to be persistent in their detection when IE is a potential differential diagnosis, as was routine before and during the SARS-CoV-2 pandemic. However, there have been reported cases of false-positive SARS-CoV-2 PCR tests and misdiagnosis of COVID-19, resulting in fatal treatment delays [5]. In the case of IE, COVID-19 significantly impacts the physician’s diagnostic process.

The second learning point from this case is that physicians relied significantly on the initial negative blood culture results, which lowered the suspicion of IE. Normally, the blood culture positivity rate for native valve IE is as high as 90–95% [6]. However, in this case, only 2 sets of blood cultures were initially performed and not the 3 sets of blood cultures that should have been performed to detect IE [7]. IE reportedly is the cause of 4% of cases of fever of unknown origin [8], and 27.7% of cases of IE in Japan have a history of valvular disease [9]. Therefore, IE should have been considered in this patient based on her medical history of prolonged fever, MVR, and positive inflammatory markers.

This case demonstrates the importance of maintaining a high index of suspicion, particularly in patients at high risk of disease, and not ruling out a disease based on a single test result. Continuous clinical assessment is fundamental even during the pandemic to avoid misdiagnosis and its consequences.

The third learning point from this case is that the delayed diagnosis of IE was related to individual cognitive bias and system complexity, including burnout during the COVID-19 pandemic. Diagnostic error is considered a consequence of the interplay between cognitive and system-related vulnerabilities [10]. It is also assumed that errors are more frequently secondary to individual factors, whereas delays are more likely to be secondary to systemic and coordination issues [11]. Moreover, it has been reported that more than 20% of medical technologists experience burnout, particularly during the COVID-19 pandemic [12].

In this case, at the point of the TTE order, we informed the medical technologist that this patient’s former blood cultures were negative. It is possible that the IE findings were missed because the technologist heard that the blood culture was negative, which caused a cognitive bias, and the test was performed to rule out IE. Moreover, this technical error may be related to burnout and changes in the practice system surrounding COVID-19.

Therefore, to avoid such misdiagnosis, clinicians should constantly be aware of situations with a high risk of biased thinking. Especially in such situations, zero-based diagnostic thinking, such as a diagnostic thinking strategy, is considered useful. In zero-based diagnostic thinking, by imagining oneself back to a point before a certain decision is made, one can free oneself from biased information and make a clear decision [13].

The diagnostic process itself also comprises many processes. Thus, after a health problem occurs in a patient, a tentative diagnosis is made, treatment and follow-up are conducted, and treatment and subsequent explanation of the results are all considered part of the diagnostic process [14]. It is necessary to be flexible and return to each step of the process to examine the situation to make an accurate diagnosis. In this case, when the doctor in charge of the patient was changed, the patient returned to the physical examination step, which led to the diagnosis of IE.

Conclusions

The diagnosis of IE should not be neglected even if the criteria for its diagnosis were not fulfilled initially. Continuous clinical assessment is fundamental. It is also important to perform zero-based diagnostic thinking and to have the Anavar flexibility to return at any stage, even during the pandemic, to avoid misdiagnosis and its consequences.