Identifying phenotypes of patients with antiphospholipid antibodies: results from a cluster analysis in a large cohort of patients
TL;DR: While clusters 1, 2, 3 and 5 corresponded to well-known entities, cluster 4 might represent a bridging condition between purePrimary APS and defined SLE, with lower thrombotic risk when compared with primary APS but higher general features such as ANA and cytopenia (mainly throm bocy topenia).
Abstract: OBJECTIVES To identify the aggregation of patients with aPL into different subgroups sharing common features in terms of clinical and laboratory phenotypes. METHODS We applied a hierarchical cluster analysis from the multiple correspondence analysis to determine subgroups of patients according to clinical and laboratory characteristics in a cohort of subjects with confirmed aPL positivity who presented to our outpatient clinics from 2006 to 2018. RESULTS A total of 486 patients [403 women; age 41.7 years (26)] were included, resulting in five clusters. Cluster 1 (n= 150) presented with thrombotic events (65.3% with venous thrombosis), with triple aPL positivity found in 34.7% of them (the highest rate among the different clusters). All the patients from cluster 2 (n = 91) had a confirmed diagnosis of SLE and the highest rate of anti-dsDNA positivity (91.7%). Cluster 3 included 79 women with pregnancy morbidity. Triple positivity was present in 3.8%, significantly lower when compared with Cluster 1 (34.7% versus 3.8%, P <0.01). Cluster 4 included 67 patients, 28 (41.8%) of whom with APS. Thrombotic events were observed in 23.9% patients. Cluster 4 had the highest rate of cytopenia, with thrombocytopenia as high 41.8% with no anti-dsDNA antibodies. Cluster 5 included 94 asymptomatic aPL carriers. CONCLUSION While clusters 1, 2, 3 and 5 corresponded to well-known entities, cluster 4 might represent a bridging condition between pure primary APS and defined SLE, with lower thrombotic risk when compared with primary APS but higher general features such as ANA and cytopenia (mainly thrombocytopenia).
Citations
More filters
Journal Article•
TL;DR: Primary APS occurs when there is no evidence of associated diseases, and APS in the presence of an underlying disease, usually systemic lupus erythematosus, is called secondary APS.
Abstract: Antiphospholipid syndrome (APS) is a disorder characterised by recurrent arterial or venous thrombosis and/or pregnancy losses, in the presence of persistently elevated levels of anticardiolipin antibodies and/or evidence of circulating lupus anticoagulant (these abnormalities are detected by blood tests). Primary APS occurs when there is no evidence of associated diseases. APS in the presence of an underlying disease, usually systemic lupus erythematosus, is called secondary APS.
43 citations
TL;DR: Clusters of symptoms are evident in long COVID patients that are associated with functional impairments and may point to distinct underlying pathophysiologic mechanisms of disease.
Abstract: Abstract Background We aimed to describe the clinical presentation of individuals presenting with prolonged recovery from coronavirus disease 2019 (COVID-19), known as long COVID. Methods This was an analysis within a multicenter, prospective cohort study of individuals with a confirmed diagnosis of COVID-19 and persistent symptoms >4 weeks from onset of acute symptoms. We performed a multiple correspondence analysis (MCA) on the most common self-reported symptoms and hierarchical clustering on the results of the MCA to identify symptom clusters. Results Two hundred thirty-three individuals were included in the analysis; the median age of the cohort was 43 (interquartile range [IQR], 36–54) years, 74% were women, and 77.3% reported a mild initial illness. MCA and hierarchical clustering revealed 3 clusters. Cluster 1 had predominantly pain symptoms with a higher proportion of joint pain, myalgia, and headache; cluster 2 had a preponderance of cardiovascular symptoms with prominent chest pain, shortness of breath, and palpitations; and cluster 3 had significantly fewer symptoms than the other clusters (2 [IQR, 2–3] symptoms per individual in cluster 3 vs 6 [IQR, 5–7] and 4 [IQR, 3–5] in clusters 1 and 2, respectively; P < .001). Clusters 1 and 2 had greater functional impairment, demonstrated by significantly longer work absence, higher dyspnea scores, and lower scores in SF-36 domains of general health, physical functioning, and role limitation due to physical functioning and social functioning. Conclusions Clusters of symptoms are evident in long COVID patients that are associated with functional impairments and may point to distinct underlying pathophysiologic mechanisms of disease.
41 citations
TL;DR: Using cluster analysis, the characteristics of the APS patients with the poorest prognosis were clarified and it was found that risk factors for cardiovascular disease may further increase events in patients with APS.
Abstract: Objective Using cluster analysis, to identify the subgroup of patients with APS with the poorest prognosis and clarify the characteristics of that subgroup. Methods This is a longitudinal retrospective cohort study of APS patients. Using clinical data and the profile of aPL, cluster analysis was performed to classify the patients into subgroups. Events were defined as thrombosis, severe bleeding, and mortality. Results A total of 168 patients with APS were included. Cluster analysis classified the patients into three subgroups; Cluster A (n = 61): secondary APS, Cluster B (n = 56): accumulation of cardiovascular risks and arterial thrombosis, Cluster C (n = 61): triple positivity of aPL and venous thrombosis. Cluster B showed significantly higher frequency of the events and higher mortality compared with the other clusters (P = 0.0112 for B vs A and P = 0.0471 for B vs C). Conclusion Using cluster analysis, we clarified the characteristics of the APS patients with the poorest prognosis. Risk factors for cardiovascular disease may further increase events in patients with APS.
12 citations
TL;DR: Elucidation of the B cell tolerance breakdown mechanisms in APS patients may help to develop and guide the use of novel therapeutic agents that target B cells or specific immune pathway.
7 citations
TL;DR: SLE+APS patients with extra-criteria manifestations and older APS patients with arterial thrombosis and increased cardiovascular risk have higher cumulative damage, and effective treatment of SLE disease activity and control of cardiovascular risk may help to reduce cumulative damage in these patients.
Abstract: Objective To identify the different clinical phenotypes of antiphospholipid syndrome (APS) by using cluster analysis and describe cumulative damage of disease clusters. Methods This retrospective study includes patients with APS (±systemic lupus erythematosus (SLE)). Two-step cluster analysis was applied by considering clinical data. Damage was calculated for all patients by applying damage index for APS (DIAPS). Results A total of 237 patients (198 females; median age of 43 years; median follow-up of 9.5 years) were classified into four clusters. Cluster 1 (n = 74) consisted of older patients with arterial-predominant thrombosis, livedo reticularis, and increased cardiovascular risk; cluster 2 (n = 70) of SLE+APS patients with thrombocytopenia and heart valve disease; cluster 3 (n = 59) of patients with venous-predominant thrombosis, less extra-criteria manifestations; and cluster 4 (n = 34) of patients with only pregnancy morbidity with lower frequency of extra-criteria features and cardiovascular risk. Patients with SLE+APS (n = 123) had the highest mean DIAPS. Regarding clusters, 1 and 2 had high cumulative damage. While cumulative survival rates of clusters did not differ, cluster 2 and 3 had lower survival rates at further years. There was no correlation between DIAPS and mortality. Conclusion SLE+APS patients with extra-criteria manifestations and older APS patients with arterial thrombosis and increased cardiovascular risk have higher cumulative damage. Effective treatment of SLE disease activity and control of cardiovascular risk may help to reduce cumulative damage in these patients.
3 citations
References
More filters
TL;DR: In 1992, Piette and colleagues suggested that the ACR revised criteria be reevaluated in light of the above discoveries, and the presence and clinical associations or antiphospholipid antibodies in patients with SLE was suggested.
Abstract: In 1982, the Diagnostic and Therapeutic Criteria Committee of the American College of Rheumatology (ACR)published revised criteria for the classification of systemiclupus erythematosus (SLE) (1). During the ensuing decade several investigators, including Drs. Graham Hughes and Donato Alarcon-Segovia, among others, have described the presence and clinical associations or antiphospholipid antibodies in patients with SLE, as well as the occurrence of theprimary antiphospholipid syndrome (2-5). In 1992, Piette and colleagues suggested that the ACR revised criteria be reevaluated in light of the above discoveries (6).
9,999 citations
University of New South Wales1, Hospital for Special Surgery2, Hokkaido University3, University of Utah4, University of Texas Health Science Center at San Antonio5, Utrecht University6, University of Milan7, Geneva College8, Sheba Medical Center9, University of Brescia10, National and Kapodistrian University of Athens11
TL;DR: This document appraise the existing evidence on clinical and laboratory features of APS addressed during the forum and proposes amendments to the Sapporo criteria, including definitions on features ofAPS that were not included in the updated criteria.
5,699 citations
TL;DR: The problems of determining the number of clusters and the clustering method are solved simultaneously by choosing the best model, and the EM result provides a measure of uncertainty about the associated classification of each data point.
Abstract: We consider the problem of determining the structure of clustered data, without prior knowledge of the number of clusters or any other information about their composition. Data are represented by a mixture model in which each component corresponds to a different cluster. Models with varying geometric properties are obtained through Gaussian components with different parametrizations and cross-cluster constraints. Noise and outliers can be modelled by adding a Poisson process component. Partitions are determined by the expectation-maximization (EM) algorithm for maximum likelihood, with initial values from agglomerative hierarchical clustering. Models are compared using an approximation to the Bayes factor based on the Bayesian information criterion (BIC); unlike significance tests, this allows comparison of more than two models at the same time, and removes the restriction that the models compared be nested. The problems of determining the number of clusters and the clustering method are solved simultaneously by choosing the best model. Moreover, the EM result provides a measure of uncertainty about the associated classification of each data point. Examples are given, showing that this approach can give performance that is much better than standard procedures, which often fail to identify groups that are either overlapping or of varying sizes and shapes.
2,576 citations
TL;DR: An association with SLE, the patient's sex, and the patient’s age at disease onset can modify the disease expression and define specific subsets of APS.
Abstract: Objective. To analyze the clinical and immunologic manifestations of antiphospholipid syndrome (APS) in a large cohort of patients and to define patterns of disease expression. Methods. The clinical and serologic features of APS (Sapporo preliminary criteria) in 1,000 patients from 13 European countries were analyzed using a computerized database. Results. The cohort consisted of 820 female patients (82.0%) and 180 male patients (18.0%) with a mean +/- SD age of 42 +/- 14 years at study entry. "Primary" APS was present in 53.1% of the patients; APS was associated with systemic lupus erythematosus (SLE) in 36.2%, with lupus-like syndrome in 5.0%, and with other diseases in 5.9%. A variety of thrombotic manifestations affecting the majority of organs were recorded. A catastrophic APS occurred in 0.8% of the patients. Patients with APS associated with SLE had more episodes of arthritis and livedo reticularis, and more frequently exhibited thrombocytopenia and leukopenia. Female patients had a higher frequency of arthritis, livedo reticularis, and migraine. Male patients had a higher frequency of myocardial infarction, epilepsy, and arterial thrombosis in the lower legs and feet. In 28 patients (2.8%), disease onset occurred before age 15; these patients had more episodes of chorea and jugular vein thrombosis than the remaining patients. In 127 patients (12.7%), disease onset occurred after age 50; most of these patients were men. These patients had a higher frequency of stroke and angina pectoris, but a lower frequency of livedo reticularis, than the remaining patients. Conclusion. APS may affect any organ of the body and display a broad spectrum of manifestations. An association with SLE, the patient's sex, and the patient's age at disease onset can modify the disease expression and define specific subsets of APS.
1,803 citations
Journal Article•
998 citations