Antiphospholipid Autoantibodies and Brain Ischemic Lesions in Infective Endocarditis

Stroke is a frequent and severe complication of infective endocarditis (IE).[1] [2] Although different factors associated with a high risk of embolic complications are known,[3] no biomarker has been identified yet. Various autoantibodies have been described in the course of IE, including antiphospholipid (aPL) autoantibodies.[4] [5] [6] Their detection at IE diagnosis may be of particular interest considering the risk of ischemic events. We aimed to determine the frequency of anticardiolipin (aCL), anti-β2GPI (aβ2GPI), antiphosphatidylethanolamine (aPE), and antiphosphatidyl/prothrombin (aPS/PT) autoantibodies in patients with IE and their association with brain ischemic events detected by systematic brain magnetic resonance imaging (MRI).

The study is an ancillary study of the prospective, single-center IMAGE study on acute IE (see the Supplementary Material). Inclusion criteria have been previously described.[7] Briefly, patients were included consecutively between 2005 and 2008 if they had definite or possible acute IE according to the Duke modified criteria.[8] Patients systematically underwent a brain 1.5 tesla MRI within 7 days after admission, unless contraindicated. Brain MRI analyzed ischemic lesions defined on diffusion-weighted images as a hyperintense signal with restricted or normalized apparent diffusion coefficient. The radiologist (I.K.) was blinded to the patients' clinical data. All clinical, biological, microbiological, and imaging data of interest collected in the IMAGE study were retrieved via a case report form. The study was approved by the local ethics committee. Written informed consent was obtained from all patients. Serum samples were collected at the acute phase of IE during the 7 days after brain MRI and stored at −80°C. Serological tests were performed in a central laboratory using samples from all patients for whom serum was available for testing in April 2020. aCL and aβ2GPI immunoglobulin G (IgG)/immunoglobulin M (IgM) autoantibodies were determined by BIO-FLASH Chemiluminescent Immuno Assay technology (Inova, Diagnostics, Werfen, France) and expressed as chemoluminescent units (CUs). The cutoff level of positivity was 20 CU for aCL and 30 CU for aβ2GPI as recommended by the manufacturer. aPE and aPS/PT IgG/IgM autoantibodies were determined by commercial ELISA (Theradiag, France and Inova Diagnostics, Werfen, France) and expressed as arbitrary units. The cutoff was 20 U/mL for aPE and 30 U/mL for aPS/PT as recommended by the manufacturer.

Among the 130 patients included in the cohort, both serum samples and brain MRIs at IE diagnosis were available in 110 patients: 61 (46.5–68) years old, 80 (72.7%) males. The main characteristics of the 110 patients are shown in [Table 1].

Table 1 Characteristics of patients with IE

	N = 110 Median [IQR] or n (%)
General characteristics
Age at diagnosis (y)	61 [46.5–68]
Male gender	80 (72.7)
Intravenous drug use	10 (9.1)
Prosthetic valve endocarditis	34 (30.9)
Previous endocarditis	14 (12.7)
Definite IE	90 (81.8)
Clinical and biological features
Fever >38°C	88 (80.0)
C-reactive protein >50 mg/L	79 (71.8)
Serum creatinine level >100 µmol/L	56 (50.9)
Location of IE
Mitral	48 (43.6)
Aortic	25 (22.7)
Tricuspid	10 (9.1)
Pulmonary	2 (1.8)
Echocardiographic findings
Vegetation length ≥15 mm	25 (22.7)
Severe vegetation mobility	32 (29.0)
Regurgitation	89 (80.9)
Microorganisms
Staphylococcus aureus	29 (26.4)
Streptococcus viridans	23 (20.9)
Streptococcus bovis	10 (9.1)
Bartonella	2 (1.8)
Coxiella burnetii	1 (0.9)
Other	45 (40.9)
Brain ischemic lesions	77 (70.0)
Symptomatic events	9 (11.7)
In-hospital death	10 (9.1)
aPL autoantibodies
IgG/IgM aCL, n (%)	10 (9.1)
IgG titers	38.7 [30.9–554.2]
IgM titers	31.8 [24.0–37.2]
IgG/IgM aβ2GP1, n (%)	6 (5.5)
IgG titers	131.2 [82.7–3,095.8]
IgM titers	33.2 [32.6–64.6]
IgG/IgM aPE, n (%)	35 (31.8)
IgG titers	39.4 [27.2–49.4]
IgM titers	36.7 [23.6–56.3]
IgG/IgM aPS/PT, n (%)	5 (4.5)
IgG titers	46.1 [41.8–64.7]
IgM titers	92.5 [63.7–121.2]
≥1 aPL, n (%)	39 (35.5)
≥2 aPL, n (%)	12 (10.9)
≥3 aPL, n (%)	4 (3.6)

Abbreviations: aCL, anticardiolipin; aPE, antiphosphatidylethanolamine; aPL, antiphospholipid; aPS/PT, antiphosphatidyl/prothrombin; aβ2GPI, anti-β2GPI; IE, infective endocarditis; IQR, interquartile range.

Note: Continuous variables were expressed as median and interquartile range (IQR; 25th–75th percentiles). Categorical variables were expressed as numbers and percentages.

Brain MRI performed at IE diagnosis revealed at least one ischemic lesion in 77 of 110 patients (70.0%) including, either singly or in combination, territorial ischemic strokes and ischemic spots in 27 (24.5%) and 54 (49.1%) patients, respectively. Ischemic lesions were asymptomatic in all but 9 (88.3%) cases.[7] Moderate to strong titers of aPL autoantibodies were detected in 39 (35.4%) of the 110 patients analyzed ([Table 1]). aPL autoantibodies were IgG/IgM aPE in most cases (n = 35/39, 89.7%). aCL or aβ2GP1 IgG/IgM autoantibodies were found in 13 (11.8%) patients. aCL IgG (median titer, 38.7 CU) and IgM (median titer: 31.8 CU) were found in 4 (3.6%) and 7 (6.4%) patients, respectively. aβ2GP1 IgG (median titer, 131.2 CU) and IgM (median titer, 33.2 CU) were detected in 3 (2.7%) patients for each isotype. Overall, 12 patients (10.9%) were positive for ≥2 aPL specificities. Four patients (3.6%) had three or more of the four tested aPL. Patient characteristics, clinical and biological features at IE diagnosis, location and size of IE, microorganisms, and outcomes did not differ between patients with and without aPL autoantibodies ([Supplementary Table S1], available in the online version). No association was found between moderate to strong titers of aPL autoantibodies, either singly or in combination, and brain ischemic lesions ([Table 2]).

Table 2 aPL and brain ischemic lesions

	Ischemic lesions, N = 77 n (%)	No ischemic lesions, N = 33 n (%)	p
IgG/IgM aCL	7 (9.1)	3 (9.1)	0.717
IgG	4 (5.2)	0 (0)	0.314
IgM	4 (5.2)	3 (9.1)	0.733
IgG/IgM aβ2GP1	5 (6.5)	1 (3.0)	0.783
IgG	3 (3.9)	0 (0)	0.553
IgM	2 (2.6)	1 (3.0)	0.609
IgG/IgM aPE	26 (33.8)	9 (27.3)	0.655
IgG	20 (26.0)	6 (18.2)	0.524
IgM	10 (13.0)	4 (12.1)	0.851
IgG/IgM aPS/PT	3 (3.9)	2 (6.1)	0.999
IgG	1 (1.3)	2 (6.1)	0.443
IgM	2 (2.6)	0 (0)	0.999
≥1 aPL	29 (37.7)	10 (30.3)	0.602
≥2 aPL	9 (11.7)	3 (9.1)	0.688
≥3 aPL	3 (3.9)	1 (3.0)	0.824
High titer[a]	12 (15.6)	2 (6.1)	0.222

Abbreviations: aCL, anticardiolipin; aPL, antiphospholipid; aβ2GPI, anti-β2GPI; aPE, antiphosphatidylethanolamine; aPS/PT, antiphosphatidyl/prothrombin.

Note: Univariate analysis of aPL autoantibodies associated with brain ischemic lesions used the Wilcoxon test for continuous variables and the χ ² test or Fisher's test for categorical variables. Two-sided p-values less than 0.05 were considered statistically significant.

^a Defined as aPL autoantibody titer >3 times the cutoff for positivity for at least one aPL tested (i.e., >60 CU for aCL, and/or >90 CU for aβ2GPI, and/or >60 U/mL for aPE, and/or >90 U/ml for aPS/PT).

Our prospective single-center IMAGE ancillary study investigates aCL, aβ2GPI, aPE and aPS/PT autoantibodies, of both IgG and IgM isotypes, in the setting of IE by using homogeneous laboratory testing with the same ELISA kits. Our study showed that (1) a high frequency of patients with IE has aPL autoantibodies at diagnosis and (2) the presence of infection-related aPL autoantibodies was not associated with brain ischemic events detected by systematic brain MRI.

Our study did not confirm the previously reported association between brain ischemic events and aPL autoantibodies.[2] [6] Such discrepancies probably reflect the poor standardization of the laboratory methods and the different thresholds for positivity. In the first study,[6] aPL autoantibodies (i.e., aCL IgG and lupus anticoagulant) were found in 14.3% of patients with a positivity threshold of 12 U/mL for aCL IgG. In the most recent study,[2] aCL and/or aβ2GPI IgM/IgG antibodies were detected in 17% of patients with a positivity threshold of 14 UGP and 8 UGP for IgG and IgM, respectively. Of note, the frequency of brain ischemic lesion was far higher in our study (70% of patients vs. less than 30%[2] [6]) where most ischemic events were asymptomatic and detected by the systematic use of brain MRI performed shortly after admission in all patients. Eventually, whereas the clinical significance of isolated aPL IgM in aPL syndrome remains debated,[9] the association of cerebral embolisms with aPL was shown for aCL/β2GPI IgM only.[2] Alternatively, our findings of a high frequency of IgG/IgM autoantibodies directed against various aPL in IE patients, along with the prior demonstration of a high prevalence of ANCAs (antineutrophil cytoplasmic antibodies) and rheumatoid factor in such setting,[10] suggest the existence of a wide array of unconnected autoantibodies which may result from a nonspecific humoral response boosted by the persistent infection. Interestingly, PE is a major lipid component of bacterial membrane.[11] Persistent infection—such as IE—may thus lead to a high amount exposure of bacterial PE, promoting the formation of anti-PE antibodies.

Our study has some limitations. Analyses according to ischemic events located in other territories than brain, infected valve, or pathogens were underpowered. aPL autoantibodies related to infection are usually transient but testing was not performed overtime. Testing for lupus anticoagulant activity could not be performed on stored blood specimen.

Our study has several strengths: the prospective design with robust collection of the clinical data and storage of blood specimens in a biobank, the homogeneous patient management, the use of four different coatings and two isotypes' detection for aPL ELISAs and the stringent cutoffs for positivity.

In conclusion, aPL—mostly aPE—autoantibodies are frequently detected in patients with IE but are not associated with brain ischemic lesions.

Authors' Contribution

K.S. directed the project. X.D., P.R.-N., B.I., and K.S. designed the study. S.N., P.R.-N., I.K., and S.T. conducted analysis. S.N. and K.S. wrote the manuscript. T.P. was involved in the project development and edited the manuscript. All authors reviewed and approved of the final manuscript.

Patient Consent for Publication

Obtained.

Data Sharing Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request

Patient and Public Involvement

Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.

Data Availability

Data supporting the findings of this study are available within the article and its supplementary materials.

^* Equal senior authors.

Publication History

Article published online:
24 January 2023

• References

• 1 Selton-Suty C, Delahaye F, Tattevin P. et al; AEPEI (Association pour l'Etude et la Prévention de l'Endocardite Infectieuse). Symptomatic and asymptomatic neurological complications of infective endocarditis: impact on surgical management and prognosis. PLoS One 2016; 11 (07) e0158522
• 2 Selton-Suty C, Maigrat CH, Devignes J. et al; Endocarditis Team of the University Hospital of Nancy, France. Possible relationship between antiphospholipid antibodies and embolic events in infective endocarditis. Heart 2018; 104 (06) 509-516
• 3 Yang A, Tan C, Daneman N. et al. Clinical and echocardiographic predictors of embolism in infective endocarditis: systematic review and meta-analysis. Clin Microbiol Infect 2019; 25 (02) 178-187
• 4 Bojalil R, Mazón-González B, Carrillo-Córdova JR, Springall R, Amezcua-Guerra LM. Frequency and clinical significance of a variety of autoantibodies in patients with definite infective endocarditis. J Clin Rheumatol 2012; 18 (02) 67-70
• 5 Zaratzian C, Gouriet F, Tissot-Dupont H. et al. Antiphospholipid antibodies proposed in the diagnosis of infective endocarditis. Eur J Clin Microbiol Infect Dis 2017; 36 (07) 1159-1162
• 6 Kupferwasser LI, Hafner G, Mohr-Kahaly S, Erbel R, Meyer J, Darius H. The presence of infection-related antiphospholipid antibodies in infective endocarditis determines a major risk factor for embolic events. J Am Coll Cardiol 1999; 33 (05) 1365-1371
• 7 Duval X, Iung B, Klein I. et al; IMAGE (Resonance Magnetic Imaging at the Acute Phase of Endocarditis) Study Group. Effect of early cerebral magnetic resonance imaging on clinical decisions in infective endocarditis: a prospective study. Ann Intern Med 2010; 152 (08) 497-504 , W175
• 8 Li JS, Sexton DJ, Mick N. et al. Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis. Clin Infect Dis 2000; 30 (04) 633-638
• 9 Kelchtermans H, Pelkmans L, de Laat B, Devreese KM. IgG/IgM antiphospholipid antibodies present in the classification criteria for the antiphospholipid syndrome: a critical review of their association with thrombosis. J Thromb Haemost 2016; 14 (08) 1530-1548
• 10 Mahr A, Batteux F, Tubiana S. et al; IMAGE Study Group. Brief report: prevalence of antineutrophil cytoplasmic antibodies in infective endocarditis. Arthritis Rheumatol 2014; 66 (06) 1672-1677
• 11 Sohlenkamp C, Geiger O. Bacterial membrane lipids: diversity in structures and pathways. FEMS Microbiol Rev 2016; 40 (01) 133-159