Acute Mitral Regurgitation of Unknown Etiology Associated with Disseminated Intravascular Coagulation Eventually Diagnosed as Enterococcus Faecalis Infective Endocarditis by Mitral Valve Surgery

Background

Although infective endocarditis (IE) is prevalent worldwide, the yearly incidence of the disease in Japan is as low as 32.4 per 1 000 000 people [1]. Making the correct diagnosis of IE is rather difficult because the clinical presentations and findings of blood tests are non-specific [2]. Although the modified Duke criteria are usually helpful to make a diagnosis of IE, they could be invalid in atypical cases [3]. Certainly, the presence of acute mitral regurgitation (MR) could be a clue to make a diagnosis of IE. However, the incidence of acute MR complicated with IE itself is low [4]. Furthermore, the case of IE in the native valve complicated with disseminated intravascular coagulation (DIC) and acute MR occurring in patients who do not have any underlying systemic disease is rare. To our knowledge, there is only one previous case report, which was caused by methicillin-sensitive Staphylococcus aureus in a 24-year-old female patient [5].

We herein report a diagnostically challenging case of IE developing in the native mitral valve in a patient complicated with acute MR and DIC without underlying systemic disease. Since the patient failed to fulfill definitively the modified Duke criteria, only mitral annuloplasty made it possible to make the correct diagnosis.

Case Report

A 70-year-old Japanese female patient without smoking history or remarkable past history, had had shortness of breath while going up or down stairs for 5 days. Her condition gradually worsened to the point of feeling dyspnea even while standing still, and she subsequently visited her family doctor. Since chest x-ray showed pulmonary congestion, and transthoracic echocardiography revealed MR and mitral valve cusp prolapse, she was assumed to have acute MR and transferred to our hospital on the same day. On admission, her body temperature was 36.8°C, heart rate 94 beats/min, blood pressure 150/88 mmHg, respiratory rate 26/min and oxygen saturation 99% with 2 L/min of oxygen delivered via a nasal cannula. Physical findings showed jugular venous distention, apical systolic murmur, and coarse crackles in both lung fields. However, leg edema, conjunctival hemorrhage, or other cutaneous manifestations including purpura of the tip of digit were not revealed. Laboratory findings on admission are shown in Table 1. The patient had a white blood cell count of 9 200/µL, C-reactive protein of 1.95 mg/dL, high-sensitivity troponin T of 0.020 ng/mL, and N-terminal pro-brain natriuretic peptide of 1 464 pg/mL. In addition, she also had DIC, which was diagnosed based on the DIC scoring system of the Japanese Association for Acute Medicine (score=5) [6]. Arterial blood gas analysis showed type I respiratory failure. Urinalysis suggested the presence of glomerulonephritis without urinary tract infection. Electrocardiogram did not show ST-T changes or left ventricular high voltage. Echocardiograph revealed an ejection fraction of 70%, MR, prolapse of posterior mitral leaflet P3, expansion of the left atrium with left atrial volume index of 66 mL/m2 and left ventricular wall thickening with posterior wall thickness of 14 mm. However, no vegetation or wall motion abnormality was detected (Figure 1). Chest x-ray and contrast-enhanced thoracic computed tomography showed increased shadows in central parts of both lung fields, bilateral pleural effusion. Based on the aforementioned findings, her condition was diagnosed as congestive heart failure due to acute MR. Because acute myocardial infarction (AMI) was deniable due to the absence of chest pain and typical findings of AMI on electrocardiogram and echocardiograph, IE was the most likely cause of her acute MR and DIC. However, 2 sets of blood cultures on admission were negative and transesophageal echocardiography (TEE) failed to reveal vegetation or thickness of valves (Figure 2). Even with blood cultures performed 4 times during her admission and the second TEE, her condition did not definitively fulfill the modified Duke criteria [3]. Subsequently performed examinations to detect the cause of DIC, such as infectious diseases, malignancies, or collagen diseases, were all negative. Meanwhile, brain magnetic resonance imaging revealed 2 acute or sub-acute cerebral infarctions (Figure 3), suggesting the etiology of cardiogenic embolisms from mitral valves because of the involvements of both sides of the brain. Although surgical intervention on the mitral valve was required, control of heart failure and DIC was considered to be the priority. Her heart failure was temporally stabilized by the 39th hospital day only by conservative treatments. DIC had been improved by the 94th hospital day after transfusing human leukocyte antigen (HLA)-matched platelets because of the presence of HLA class I antibody. On the 170th hospital day, mitral annuloplasty was performed in another hospital following the patient’s wishes. Vegetation on a large portion of the mitral valve was found intraoperatively (Figure 4), and Enterococcus faecalis was detected by culture of the mitral valve. Because blood culture after surgery was also positive for E. faecalis, which denied the possibility of contamination, we made the definitive diagnosis of IE. A course of ampicillin was administered for 6 weeks after blood cultures turned negative. She had been followed at our hospital on an outpatient basis for 6 months without showing any symptoms and signs of recurrence of IE.

The objective for this case report is to inform others that it could be very difficult to diagnose IE correctly if it fails to fulfill the modified Duke criteria. In such a case, only cardiac surgery might enable us to make an accurate diagnosis and save a patient’s life.

Discussion

Our patient had IE complicated with acute MR and DIC, which did not definitively fulfill the modified Duke criteria despite twice repeated TEE and 8 sets of blood cultures, and her condition was eventually diagnosed only by mitral valve surgery. Because coexistence of acute MR and DIC is an atypical and rare complication of IE, we had extreme difficulty in making an accurate diagnosis, despite multidisciplinary collaboration among the departments of general medicine, cardiology, cardiovascular surgery, and hematology in our hospital. Since we could not be convinced that the cause of her DIC was MR and IE, and her heart failure was temporally stabilized by medical treatment, we tried to find the cause of DIC first temporally pending mitral annuloplasty.

The unusual presentations of this patient made the diagnosis extremely difficult. The modified Duke criteria have been used universally to diagnose IE [3]. Positive blood cultures and confirmation of the presence of vegetation by echocardiography are the decision-making points of these criteria [3]. On one hand, 3 consecutive sets of blood cultures are useful bacteriological tools [2]. However, 10–30% of IE patients could have negative blood cultures [7]. The 2 major causes are antibiotic treatments prior to examinations and the involvement of fastidious organisms, which are difficult to culture [7]. On the other hand, among echocardiography examinations, TEE is the most useful imaging modality for detecting vegetation with sensitivity of 86%, specificity 97%, positive predictive value (PPV) of 89%, and negative predictive value of 96% [8]. Although repetition of TEE could improve sensitivity and specificity, the third TEE is reported to detect only 20% of IEs. Additionally, performing TEE more than 4 times is futile and would not provide additional useful findings [9]. Among cases posing a diagnostic challenge, 0.9% of patients who were diagnosed as having non-infectious endocarditis and underwent heart valve operations were reported to be diagnosed with IE only after cardiac operations [10]. In such cases, histological findings and bacterial detection on the valve using PCR were extremely useful [11,12]. In particular, histological examination could fulfill the strict version of the modified Duke criteria, which is evaluated by the presence of vegetation and bacteria on the valve, with both specificity and PPV of 100% [13]. Despite the lack of prior antibiotic therapy, our case did not definitively fulfill the modified Duke criteria on the basis of negative results of 4 blood cultures (8 sets) and 2 TEEs. Finally, we could make the correct diagnosis of IE only by performing a surgical operation on the mitral valve. This made it possible to detect vegetation and perform bacterial culture of the valve, which grew E. faecalis.

Another important feature of this case was the complication of DIC. DIC usually develops in the presence of underlying major disorders including sepsis, malignancies, trauma, major surgery, organ damages, obstetric complications, vascular abnormalities, severe hepatic failure, or immunological diseases [14]. Of all disorders, sepsis, trauma, and major surgeries account for about 75% of underlying disorders of DIC [6]. Our patient was diagnosed as having acute DIC on admission. Refractoriness to platelet transfusion and uncontrollable bleeding tendency, which were apparent by the presence of general subcutaneous bleeding, appeared in the clinical course. Her DIC was supposedly started by IE in the early disease phase, which deteriorated by a vicious circle of excessive consumption of platelets due to the production of HLA class I antibody, the presence of which was confirmed by laboratory examination. In short, uncontrolled bleeding tendency, which was improperly treated by HLA-unmatched platelet transfusion, worsened her DIC. Our hypotheses might be supported by the fact that the patient did not require additional platelet transfusions after administering 2 sets of HLA-matched platelet transfusions, which markedly improved bleeding tendency and DIC, promoting the disappearance of generalized subcutaneous bleeding.

Conclusions

IE on native mitral valve complicated with acute MR and DIC without underlying systemic disorders, which fails to fulfill definitively the modified Duke criteria, is rare and difficult to be diagnosed correctly. Only cardiac surgery might enable us to make an accurate diagnosis and save a patients’ life in such cases. Our case suggests that following guidelines or criteria should never replace clinical reasoning.