Abstract
Objective
To compare patients with SLE associated with either Evans syndrome (ES) or an isolated autoimmune cytopenia (immune thrombocytopenia (ITP) or autoimmune haemolytic anaemia (AIHA)).
Methods
Multicentre retrospective study including patients with SLE presenting with ITP, AIHA or ES of clinical significance (ie, requiring therapeutic intervention according to European Alliance of Associations for Rheumatology guidelines or clinician discretion). Clinical, laboratory and outcome data were compared between patients with ES and those with ITP or AIHA. Severe SLE flares were defined as flares with SLE Disease Activity Index ≥10.
Results
Among 95 patients with SLE included, 30 had ES, 43 ITP and 22 AIHA. The number of severe SLE flares per patient was higher in ES than in ITP (1.3 vs 0.4, p=0.0004) and patients with AIHA (0.4, p=0.006). Patients with ES had a higher incidence rate ratio (IRR) of severe SLE flares (IRR =3.03; 95% CI 1.50 to 6.41), which remained significant in inverse probability of treatment weighting analyses. At last follow-up, patients with ES presented higher rates of renal (36.7% vs 9.3%, p=0.007) and neurological (36.7% vs 11.6%, p=0.019) involvement than patients with ITP. Severe infections were more frequent in patients with ES than ITP (47% vs 23%, p=0.045), with a higher mean number of severe infections per patient (1.2 vs 0.5, p=0.04).
Conclusion
In patients with SLE, ES is associated with an increased risk of severe flares than isolated autoimmune cytopenia. These findings were consistent after adjustment for baseline imbalances, supporting ES as a high-risk SLE phenotype.
Keywords: Autoimmune Diseases; Hematology; Lupus Erythematosus, Systemic; Thrombocytopenia
WHAT IS ALREADY KNOWN ON THIS TOPIC
Patients with SLE with immune thrombocytopenia or autoimmune haemolytic anaemia are at higher risk of developing organ damage.
Evans syndrome (ES), regardless of its aetiology, is associated with a poor prognosis, but available data on SLE associated with ES is scarce.
WHAT THIS STUDY ADDS
In this cohort, ES was associated with an increased risk of severe SLE flares (SLE Disease Activity Index >10), while not being associated with a higher overall number of flares.
Inverse probability of treatment weighting analyses confirmed this association after adjustment for baseline disease activity and demographic factors.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Patients with SLE and ES represent a subgroup at increased risk of severe disease exacerbations and may require closer monitoring and tailored therapeutic strategies.
Future studies should investigate whether the increased rate of severe flares translates into long-term organ damage and evaluate optimal management strategies for this high-risk subset.
Introduction
SLE is a chronic autoimmune disease characterised by a wide range of clinical manifestations, including haematological disorders. Immune thrombocytopenia (ITP) and autoimmune haemolytic anaemia (AIHA) are relatively prevalent haematological conditions associated with SLE, occurring in approximately 20% and 5–10% of patients with SLE, respectively.1,3 It has been shown that patients with SLE associated with ITP or AIHA are at higher risk of developing organ damage.24,7 Moreover, a higher mortality has been reported in patients with SLE associated with ITP.8 Evans syndrome (ES) is characterised by the concomitant or subsequent association of at least two out of three autoimmune cytopenia: ITP, AIHA and autoimmune neutropenia (AIN); ES has been documented in 0.5% to 3% of SLE cases.9,11 ES, regardless of its aetiology, is associated with a poor prognosis, particularly due to the risk of infection. The available data on SLE associated with ES is scarce and inconsistent, particularly with regards to the potential impact of this association on renal impairment.10 11 Some data suggest that ES has a poorer prognosis than isolated autoimmune cytopenias.12,14 However, this comparison has not yet been conducted exclusively in patients with SLE.
The objective of the present study was to compare patients with SLE associated with either ES or an isolated autoimmune cytopenia (ie, ITP or AIHA) in terms of clinical and laboratory characteristics at SLE diagnosis, flares, morbidity and mortality.
Methods
Study design
This retrospective cohort study included patients with SLE admitted to three internal medicine units of three different centres (Hospices Civils de Lyon, Hôpital Saint Joseph Saint Luc and Centre Hospitalier Annecy Genevois) located in or near Lyon, France, between 1997 and 2024. Three principal investigators (MR, AF-B and NM) collected data from the medical files using standardised forms between December 2017 and April 2024.
Study population
To be included, patient had to be ≥18 years old, fulfil the Systemic Lupus International Collaborating Clinics 2012 and/or the American College of Rheumatology classification criteria for SLE, according to the time at which SLE diagnosis was made15,17 and present with ITP, AIHA or ES either before, at the time of or after the diagnosis of SLE. Only ITP of clinical significance (ITPCS), AIHA of clinical significance (AIHACS) and ES of clinical significance (ESCS) were considered.
ITPCS was defined as ITP with either attributable bleeding disorders (any clinical manifestation from purpura to life-threatening bleeding) and/or a platelet count ≤30×109/L.18 Other causes of secondary thrombocytopenia, such as sepsis, drug-induced thrombocytopenia, haematological disorders (haemophagocytic lymphohistiocytosis, thrombotic microangiopathy…) and portal hypertension were excluded.
AIHACS was defined as AIHA with a haemoglobin level <100 g/L.19 20 AIHA was defined as anaemia with positive direct antiglobulin test (DAT) and biological signs of haemolysis (decreased haptoglobin, elevated lactate dehydrogenase and high reticulocytosis). Other causes of anaemia, including blood loss from the gastrointestinal tract and other causes of haemolysis, including hereditary, drug-induced, microangiopathic haemolytic anaemia were excluded.
ES was defined as the concomitant or sequential occurrence of ITP, AIHA and/or AIN. To be considered ESCS, at least one of the cytopenias had to be clinically significant. AIN of clinical significance was defined as absolute neutrophil count less than 0.8×109/L. Other causes of neutropenia, including drug-induced neutropenia, viral infections (Epstein-Barr virus, cytomegalovirus, HIV, parvovirus B19, influenza) and haematological malignancies were excluded. For clarity, in the remainder of the manuscript, we will refer to AIHACS as ‘AIHA’ and ITPCS as ‘ITP’.
Patient and public involvement
Patients or members of the public were not involved in the design, conduct, choice of outcome, recruitment or dissemination plans of this research.
Data collection
All available clinical, laboratory, treatment and outcome data were systematically collected from medical files and reviewed using a standardised form. Cytopenia flares were counted from the first cytopenia event, even if this occurred before the diagnosis of SLE. A cytopenia flare was defined by a platelet count <30 ×109/L or haemorrhage for ITP, and a haemoglobin (Hb) level <100 g/L with positive DAT and laboratory signs of haemolysis for AIHA. Bleeding severity was assessed using the Khellaf and WHO bleeding scores.21 22 Major haemorrhage was defined as a Khellaf score >8 and/or WHO score >2. SLE flares were counted from the time of SLE diagnosis. For each SLE flare, platelet count, Hb level, AIHA criteria, SLE Disease Activity Index (SLEDAI) score and haemorrhagic and thrombotic events were recorded. Severe SLE flares were defined as flares with a SLEDAI score ≥10. Renal impairment was defined as measured glomerular filtration rate <60% and/or proteinuria ≥0.5 g/24 hours. All thrombotic events were recorded and analysed. Infection was defined as an infection requiring antibiotics/antiviral treatment. Severe infection was defined as an infection requiring a hospitalisation and/or intravenous antibiotics.
All treatments to which patients had been exposed, whether or not they were initiated in the setting of cytopenia, were also collected. Corticosteroids (CSs), intravenous immunoglobulins and vinca alkaloids were considered as induction treatments. Maintenance therapies consisted of hydroxychloroquine (HCQ), immunosuppressive (IS) agents such as azathioprine, mycophenolate mofetil, ciclosporin A, methotrexate and cyclophosphamide, rituximab (RTX), belimumab, obinutuzumab, splenectomy, thrombopoietin receptor agonists (romiplostim and eltrombopag), erythropoietin and dapsone.
Statistical analysis
Patients with SLE and ES were compared with those with isolated ITP and isolated AIHA. In addition, for selected analyses, patients with SLE-associated cytopenias were compared with cytopenia-free SLE controls using age-matched and sex-matched cohorts (1:2 ratio, nearest-neighbour matching with a calliper of 0.2).
Categorical variables are presented as counts (percentages) and were compared using Fisher’s exact test or χ² test, as appropriate. Continuous variables are presented as mean±SD or median (IQR) and were compared using Student’s t-test, or Wilcoxon rank-sum test according to distribution normality and variance homogeneity.
Time-to-event analyses were performed using Kaplan-Meier survival curves and compared with the log-rank test.
Incidence rate ratios (IRRs) for relapses were estimated using Poisson regression models. Overdispersion was formally assessed using a dispersion test; when significant overdispersion was detected (> 1.5), negative binomial regression models were applied accordingly. An offset term corresponding to the logarithm of follow-up duration was included to account for differences in exposure time.
To reduce the impact of potential baseline imbalances between groups, inverse probability of treatment weighting (IPTW) analyses were conducted. Propensity scores were estimated using logistic regression models including baseline SLEDAI (imputed when missing with the median) and CS exposure. Weighted regression models were then used to estimate adjusted IRRs. Missing baseline SLEDAI values were handled using single imputation based on clinically relevant covariates to preserve sample size and avoid selection bias.
All tests were two-tailed, and a p value <0.05 was considered statistically significant. Statistical analyses were performed using R software V.4.2.3 (R Foundation for Statistical Computing, Vienna, Austria).
Results
Clinical and laboratory characteristics at SLE diagnosis
Among the 95 patients with SLE included in this study, 30 (32%) had ES, 43 (45%) ITP and 22 (23%) AIHA. After matching patients with SLE and cytopenias to cytopenia-free patients with SLE on age and sex (1:2 ratio, calliper=0.2), we obtained 94 patients with cytopenia matched to 187 controls. Matching successfully balanced age and sex between groups (mean age 33 years, 84% vs 83% female, standardised mean differences <0.05). Cutaneous, serositis and renal involvement were significantly less frequent in cytopenia-free SLE (skin: 42.6% vs 65.2%, p=0.002; serositis: 6.38% vs 23.5%, p=0.003; kidney: 11.7% vs 29.4%, p=0.005). Autoantibody titres (ANA) were comparable, but anti-double-stranded DNA (dsDNA) levels were higher in patients with cytopenia (median 17 vs 10.5, p=0.026). Other serologies (Sm [Smith], Sm/RNP [ribonucleoprotein], SSA [Sjögren's-syndrome-related antigen A], SSB [Sjögren's-syndrome-related antigen B], antiphospholipid profile, arterial antiphospholipid syndrome (APS), venous APS) showed no significant differences between groups. Comparative analysis showed that mortality was similar between groups (3.2% vs 2.7%, p=1.0) (online supplemental table S1).
Follow-up duration was comparable across cytopenia groups. A majority of cytopenias occurred prior to or concomitantly with SLE diagnosis in all groups (table 1). Patients with ITP were significantly younger at SLE diagnosis than patients with ES (29.7±13.5 years vs 38.4±19.1 years, p=0.03). There were significantly more European (77% vs 41%, p=0.01) patients in the ES group than in the AIHA group. Patients in the ES group exhibited a more active disease at baseline with significantly higher SLEDAI scores as compared with the ITP group (11.8±7.5 vs 5.9±3.5, p=0.001).
Table 1. Baseline characteristics and laboratory findings.
| ES (N=30) | ITP (N=43) | AIHA (N=22) | Overall (N=95) | P value | |
|---|---|---|---|---|---|
| Age at SLE diagnosis, mean (SD) | 38.4 (19.1) | 29.7 (13.5) | 39.3 (14.7) | 34.4 (16.3) | p=0.03b* p=0.86* |
| Sex | p=0.001* p=0.16† |
||||
| Male | 8 (26.7) | 6 (14) | 2 (9.1) | 10 (16.8) | – |
| Female | 22 (73.3) | 37 (86.0) | 20 (90.9) | 79 (83.2) | – |
| Origin‡ | p>0.99* p=0.01† |
||||
| European | 23 (76.7) | 30 (69.7) | 9 (40.9) | 62 (65.3) | – |
| North African | 3 (10.0) | 5 (11.6) | 7 (31.8) | 15 (15.8) | – |
| Afro-Caribbean | 1 (3.3) | 2 (4.7) | 5 (22.7) | 8 (8.4) | – |
| Asian | 3 (10.0) | 4 (9.3) | 1 (4.5) | 8 (8.4) | – |
| Latin American | 0 (0) | 1 (2.3) | 0 (0) | 1 (1.1) | – |
| Missing | 0 (0) | 1 (2.3) | 0 (0) | 1 (1) | – |
| Follow-up duration, months, median (range) | 132 (37–206) | 131 (16–321) | 145 (9–203) | 132 (9–321) | – |
| Clinical manifestations | |||||
| Cutaneous | 13 (43.3) | 23 (53.5) | 5 (22.7) | 41 (43.2) | p=0.48 * p=0.15† |
| Arthritis | 16 (53.3) | 22 (51.2) | 12 (54.5) | 50 (52.6) | p>0.99* p>0.99† |
| Serositis | 2 (6.7) | 0 (0) | 4 (18.2) | 6 (6.3) | p=0.16* p=0.38† |
| Renal | 4 (13.3) | 4 (9.3) | 3 (13.6) | 11 (11.6) | p=0.71* p>0.99† |
| Central nervous system involvement | 4 (13.3) | 1 (2.3) | 3 (13.6) | 8 (8.4) | p=0.15* p>0.38† |
| Hepatitis | 2 (6.7) | 0 (0) | 1 (4.5) | 3 (3.2) | p=0.16* p=0.99† |
| Myositis | 1 (3.3) | 0 (0) | 0 (0) | 1 (1.1) | p=0.41* p>0.99† |
| Pulmonary | 2 (6.7) | 0 (0) | 0 (0) | 2 (2.1) | p=0.17* p=0.50† |
| SLEDAI, mean (SD) | 11.8 (7.5) | 5.9 (3.5) | 10.1 (6.4) | 9.1 (6.4) | p=0.001* p=0.44† |
| Haematological manifestations | |||||
| Thrombocytopenia at SLE onset | 16 (53) | 29 (67) | – | 45 (47) | p=0.55* |
| Autoimmune haemolytic anaemia | 14 (47) | – | 13 (59) | 27 (28) | p=0.62† |
| Neutropenia | 3 (10) | – | – | – | |
| Any cytopenia | 16 (53) | 29 (67) | 13 (59) | 58 (61) | – |
| Haemorrhage | |||||
| At ITP/ES diagnosis | 8 (27) | 13 (31) | – | 21 (22) | p=0.80* |
| Bleeding scores§ | |||||
| WHO score, mean (SD) | 1.75 (0.6) | 1.81 (0.6) | – | 1.78 (0.6) | – |
| Khellaf score, mean (SD) | 5.8 (4.2) | 5.8 (4.2) | – | 5.8 (4.2) | – |
| Laboratory findings | |||||
| Platelets (x109/L), mean (SD) | 70.6 (69.9) | 72.5 (83.8) | 226 (97.3) | 113 (107) | p=0.83 |
| Haemoglobin (g/L), mean (SD) | 92.8 (26) | 122 (19.1) | 81.3 (27.3) | 101 (29.4) | p=0.15 |
| Positive anti-double-stranded DNA antibody | 20 (66.7) | 25 (58.1) | 7 (31.8) | 52 (54.7) | p=0.21* p=0.02† |
| Hypocomplementaemia | 19 (36) | 15 (35) | 13 (59) | 45 (47) | p=0.21 p=0.07 |
| Anti-Sm | 3 (10.0) | 6 (14.0) | 9 (40.9) | 18 (18.9) | p>0.99* p=0.09† |
| Anti-RNP | 3 (10.0) | 5 (11.6) | 13 (59.1) | 21 (22.1) | p>0.99* p=0.009† |
| Anti-SSA/Ro | 9 (30.0) | 18 (41.9) | 8 (36.4) | 35 (36.8) | p>0.99* p=0.54† |
| Anti-SSB/La | 5 (16.7) | 5 (11.6) | 1 (4.5) | 11 (11.6) | p=0.28* p=0.07† |
| APL antibodies | p=0.79* p=0.23† |
||||
| Lupus anticoagulant | 12 (40.0) | 24 (55.8) | 8 (36.4) | 44 (46.3) | p>0.99* p>0.99† |
| Anticardiolipin | 10 (33.3) | 18 (41.9) | 7 (31.8) | 35 (36.8) | p>0.99* p=0.54† |
| Anti-β2GP1 | 11 (36.7) | 21 (48.8) | 7 (31.8) | 39 (41.1) | p=0.79* p=0.36† |
Data are presented as n (%) unless otherwise specified.
Comparison of ES versus ITP.
Comparison of ES versus AIHA.
Region of origin declared by the patient. French law does not allow patient ethnicity to be recorded.
ES and ITP groups only.
AIHA, autoimmune haemolytic anaemia; APL, antiphospholipid; ES, Evans syndrome; 2GP1, 2-glycoprotein-1; ITP, immune thrombocytopenia; RNP, Ribonucleoprotein; SLEDAI, SLE Disease Activity Index; Sm, Smith; SSA, Sjögren's-syndrome-related antigen A; SSB, Sjögren's-syndrome-related antigen B.
Cytopenia flares and management (haematological flares)
A total of 236 cytopenia flares were noted: 66 in the ES group, 132 in the ITP group and 36 in the AIHA group. The mean (SD) number of cytopenia flares did not differ significantly between the ITP or AIHA groups and the ES group (table 2). Rates of major bleeding events were comparable between ES and ITP groups. No statistical difference was found between the ES and ITP groups regarding major bleeding events per patient (table 2). Over the follow-up duration, the mean (SD) number of treatment lines received per patient was 4.23 (1.82). The mean (SD) number of cytopenia maintenance treatment lines was 1.79 (1.22). The overall treatment burden was substantial across groups. Patients with ITP received more maintenance therapies than those with ES (2.21 (1.30) vs 1.53 (1.14), p=0.02). HCQ and CSs were the main first-line treatments in all groups. The use of RTX and other IS agents was similar across groups (table 2).
Table 2. Cytopenia and bleeding events.
| ES (N=30) | ITP (N=43) | AIHA (N=22) | Overall (N=95) | P value | |
|---|---|---|---|---|---|
| Mean number of cytopenia flares (SD) | 2.6 (1.9) | 3.7 (3.3) | 1.7 (1.4) | 2.86 (1.17) | p=0.11* p=0.19† |
| Major Haemorrhages related to ITP | |||||
| Events,ठmean (SD) | 0.17 (0.37) | 0.19 (0.60) | Р| 0.18 (0.52) | p=0.67* |
| Patients, n (%) | 5 (17) | 5 (12) | – | 10 (11) | p=0.78* |
Comparison of ES versus ITP.
Comparison of ES versus AIHA.
Results are expressed per patient.
Patients not on anticoagulants or platelet aggregation inhibitors at the time of bleeding.
AIHA, autoimmune haemolytic anaemia; ES, Evans syndrome; ITP, immune thrombocytopenia.
SLE flares and organ involvement (extra-haematological flares)
Kaplan-Meier analyses were performed to evaluate flare-free (haematological, extra-haematological and mixed) survival according to organ involvement and serological profiles. Antiphospholipid antibody (aPL) positivity was consistently associated with significantly shorter flare-free survival across all definitions of relapse, including haematological, extra-haematological and overall SLE flares (onlinesupplemental figures S1S3). Among the different aPL profiles, lupus anticoagulant showed the strongest and most consistent association with relapse risk. Patients fulfilling criteria for definite APS also demonstrated markedly reduced flare-free survival compared with patients without APS.
As compared with the AIHA group, patients with ES experienced SLE flares more frequently (4.2 (3.4) vs 2.8 (1.9), p=0.03, table 3). Patients with ES experienced a higher number of severe SLE flares (SLEDAI >10) compared with both ITP and AIHA (respectively, 1.3 (1.4) vs 0.4 (0.8) p=0.0004 and 0.4 (0.4) p=0.006, table 3). Classical negative binomial and Poisson regression analyses showed that ES was not significantly associated with the total number of flares, nor with pure or mixed haematological flares after adjustment on prior CS use and SLEDAI at onset (figures1 2). However, severe flares defined by SLEDAI >10 were more frequent in patients with ES (IRR =3.03; 95% CI 1.50 to 6.41), whereas severe flares with SLEDAI >20 were too rare to allow robust estimation (IRR =2.04; 95% CI 0.55 to 8.34). To account for baseline imbalances, we performed IPTW models including imputed baseline SLEDAI, CS exposure, age and sex. IPTW analyses confirmed the trends observed in the primary models: for severe flares (SLEDAI >10): IRR = 2.64 (95% CI 1.61 to 4.34; p < 0.001), IPTW models indicate a significantly increased risk among patients with ES (table 4); for other flare types, no significant associations were found, although some trends suggest variable effects depending on flare type (figures1 2). These results are consistent with classical models, but provide more robust estimates by accounting for baseline differences between groups.
Table 3. SLE flares, morbidity and mortality.
| ES (N=30) | ITP (N=43) | AIHA (N=22) | Overall (N=95) | P value | |
|---|---|---|---|---|---|
| Flares | |||||
| Mean number of SLE flares (SD)*† | 4.0 (2.2) | 5.2 (4.2) | 2.8 (1.9) | 4.2 (3.4) | p=0.28‡ p=0.03§ |
| Renal involvement, n (%) | 11 (36.7) | 4 (9.3) | 6 (27.3) | 21 (22.1) | p=0.007‡ p=0.56§ |
| Central nervous system involvement, n (%) | 11 (36.7) | 5 (11.6) | 4 (18.2) | 20 (21.1) | p=0.02‡ p=0.22§ |
| Myocarditis, n (%) | 2 (6.7) | 1 (2.3) | 0 (0) | 3 (3.2) | p=0.41‡ p=0.50§ |
| Mean (SD) SLEDAI >10 | 1.3 (1.4) | 0.4 (0.8) | 0.4 (0.4) | 0.7 (1.1) | p=0.0004‡ p=0.006§ |
| Severe infections | |||||
| Mean events (SD)* | 1.2 (1.9) | 0.5 (0.9) | 0.9 (0.9) | 0.8 (1.36) | p=0.04‡ p=0.78§ |
| Patients, n (%) | 14 (47) | 10 (23) | 12 (54) | 36 (38) | p=0.045‡ p=0.97§ |
| Total events | 29 | 15 | 19 | 63 | – |
| During AIN, n (%) | 1 (3) | 1 | – | ||
| On CS, n (%) | 26 (87) | 14 | 13 (68) | 53 (84) | – |
| On RTX, n (%) | 4 (14) | 4 (27) | 4 (14) | 12 (19) | – |
| On IS, n (%) | 16 (55) | 7 (45) | 7 (24) | 30 (48) | – |
| On CMP, n(%) | 7 (24) | 1 (7) | 3 (16) | 11 (17) | – |
| After Spl, n(%) | 1 (3) | 5 (33) | 6 (10) | – | |
| Thrombosis | |||||
| Patients, n (%) | 8 (27) | 10 (23) | 7 (32) | 25 (26.3) | p=0.78‡ p>0.99§ |
| Total events | 8 | 15 | 8 | 31 | – |
| APS, patients, n (%) | 4 (13.3) | 8 (18.6) | 5 (22.7) | 17 (17.9) | p>0.99‡ p>0.99§ |
| Arterial | 2 (6.7) | 1 (2.3) | 1 (4.5) | 4 (4.2) | p=0.03‡ p=0.08§ |
| Venous | 2 (6.7) | 4 (9.3) | 5 (22.7) | 11 (11.6) | p=0.13‡ p=0.58§ |
| Obstetrical | 1 (3.3) | 4 (9.3) | 0 (0) | 5 (5.5) | – |
| Death | 3 (10) | 1 (2.3) | 0 | 4 | – |
Results are expressed per patient.
Including haematological and extra haematological flares.
Comparison of ES versus ITP.
Comparison of ES versus AIHA.
AIHA, autoimmune haemolytic anaemia; AIN, autoimmune neutropenia; APS, antiphospholipid syndrome; CMP, cyclophosphamide; CS, corticosteroid; ES, Evans syndrome; IS, immunosuppressive agent; ITP, immune thrombocytopenia; RTX, rituximab; SLEDAI, SLE Disease Activity Index; Spl, splenectomy.
Figure 1. Association between Evans syndrome and SLE flare outcomes including baseline SLEDAI and CTC. (A) Total flares, (B) extra-haematological flares, (C) mixed flares, (D) pure haematological flares, (E) total haematological flares, (F) severe flares (SLEDAI ≥10) and (G) Severe flares (SLEDAI ≥20). CTC, corticosteroid exposure; IRR, incidence rate ratio; SLEDAI, SLE Disease Activity Index. P-value significance: <0.05: *; <0.01: **; <0.001: ***.
Figure 2. Association between Evans syndrome and SLE flare outcomes excluding baseline SLEDAI and CTC. (A) Total flares, (B) extra-haematological flares, (C) mixed flares, (D) pure haematological flares, (E) total haematological flares, (F) severe flares (SLEDAI ≥10) and (G) severe flares (SLEDAI ≥20). CTC, corticosteroid exposure; IRR, incidence rate ratio; SLEDAI, SLE Disease Activity Index. p-value significance : <0.05: *; <0.01: **; <0.001: ***.
Table 4. Association between ES and SLE flare outcomes after inverse probability of treatment weighting.
| Outcome | IRR ES (95% CI) | P value | IRR Initial SLEDAI (95% CI) | IRR CTC (95% CI) |
|---|---|---|---|---|
| Total flares | 1.08 (0.79 to 1.47) | 0.64 | 0.96 (0.94 to 0.99) | 0.92 (0.78 to 1.08) |
| Extra-haematological flares | 1.34 (0.81 to 2.24) | 0.26 | 0.98 (0.93 to 1.03) | 0.77 (0.58 to 1.01) |
| Mixed flares | 1.16 (0.74 to 1.78) | 0.50 | 1.03 (1.00 to 1.07) | 0.76 (0.59 to 0.98) |
| Haematological flares (pure) | 0.90 (0.52 to 1.56) | 0.70 | 0.89 (0.83 to 0.94) | 1.11 (0.85 to 1.45) |
| Haematological flares (total) | 0.95 (0.66 to 1.35) | 0.77 | 0.96 (0.93 to 0.99) | 1.02 (0.85 to 1.23) |
| Severe flares (SLEDAI ≥10) | 2.64 (1.61 to 4.34) | 0.0001 | 1.08 (1.04 to 1.12) | 0.89 (0.66 to 1.21) |
| Severe flares (SLEDAI ≥20) | 1.00 (0.27 to 3.00) | 0.999 | 1.14 (1.07 to 1.23) | 1.14 (0.53 to 2.44) |
Significant results appear as bold entries in this table.
CTC, corticosteroid exposure; ES, Evans syndrome; IPTW, inverse probability of treatment weighting; IRR, incidence rate ratio; SLEDAI, SLE Disease Activity Index.
At the end of the follow-up, patients in the ES group had higher rates of renal involvement without (36.7% vs 9.3%, p=0.007). Patients in the ES group also presented higher rates of neurological (36.7% vs 11.6%, p=0.019) involvement than those in the ITP group (table 5).
Table 5. Morbidity and mortality.
| ES (N=30) |
AIHA (N=22) |
ITP (N=43) |
Overall (N=95) |
P value | |
|---|---|---|---|---|---|
| Flares | |||||
| Mean number of SLE flares*† | 4 (1–10) | 2.8 (1–7) | 5.2 (1–24) | 4.26 (1–24) | p=0.03‡ |
| Mean number of cytopenia flare | 2.6 (1–7) | 1.7 (1–7) | 3.7 (1–17) | 2.86 (1–17) | |
| Lupus nephritis, n (%) | 11 (36.7) | 6 (27.3) | 4 (9.3) | 21 (22.1) | p=0.007§ |
| Central nervous system involvement | 11 (36.7%) | 4 (18.2%) | 5 (11.6%) | 20 (21.1%) | p=0.019§ |
| Myocarditis | 2 (6.7%) | 0 (0%) | 1 (2.3%) | 3 (3.2%) | – |
| SLEDAI >10 | 1.30 (0–5) | 0.364 (0–1) | 0.395 (0–4) | 0.674 (0–5) | p=0.006† p=0.0004§ |
| Severe infections | |||||
| Events* | 1.23 (0–8) | 0.9 (0–3) | 0.46 (0–3) | 0.8 (0–8) | p=0.038 |
| Patients, n (%) | 14 (47) | 12 (54) | 10 (23) | 36 (38) | p=0.045 |
| Total events | 29 | 19 | 15 | 63 | |
| During AIN (%) | 1 (3) | 1 | |||
| On CS (%) | 26 (87) | 13 (68) | 14 (93) | 53 (84) | |
| On RTX (%) | 4 (14) | 4 (14) | 4 (27) | 12 (19) | |
| On IS (%) | 16 (55) | 7 (24) | 7 (45) | 30 (48) | |
| On CYC (%) | 7 (24) | 3 (16) | 1 (7) | 11 (17) | |
| After Spl (%) | 1 (3) | 5 (33) | 6 (10) | ||
| Major haemorrhages related to ITP | |||||
| Events*¶ | 0.167 (0–1) | – | 0.195 (0–3) | 0.183 (0–3) | |
| Patients, n (%) | 5 (17) | – | 5 (12) | 10 (11) | |
| Thrombosis | |||||
| Patients, n (%) | 8 (27) | 7 (32) | 10 (23) | 25 (26.3) | |
| Total events | 8 | 8 | 15 | 31 | |
| APLS, patients, n(%) | 4 (13.3) | 5 (22.7) | 8 (18.6) | 17 (17.9) | |
| Arterial | 2 (6.7) | 1 (4.5) | 1 (2.3) | 4 (4.2) | p=0.03 |
| Venous | 2 (6.7) | 5 (22.7) | 4 (9.3) | 11 (11.6) | |
| Obstetrical | 1 (3.3) | 0 (0) | 4 (9.3) | 5 (5.5) | |
| Deaths | 3 (10) | 0 | 1 (2.3) | 4 | |
Results are expressed per patient.
Including haematological and extra haematological flares.
Comparison of ES versus AIHA.
Comparison of ES versus ITP.
Patients not on anticoagulants or platelet aggregation inhibitors at the time of bleeding.
AIHA, autoimmune haemolytic anaemia; AIN, autoimmune neutropenia; APLS, Anti phospholipid syndrome; CS, corticosteroid; CYC, cyclophosphamide; ES, Evans syndrome; IS, immunosuppressive agent; ITP, immune thrombocytopenia; RTX, rituximab; SLEDAI, SLE Disease Activity Index; Spl, splenectomy.
Infections and thrombosis
The proportion of patients with severe infections was significantly higher in the ES group than in the ITP group (47% vs 23%, p=0.045). In the total cohort, 31 thrombotic events were reported. A total of 17 patients presented with APS: 4 in the ES group, 8 in the ITP group and 5 in the AIHA group. There were more arterial APS thrombotic events in the ES group (n=2, 6.7%) than in the ITP group (n=1, 2.3%; p=0.03, table 3). Two thromboses occurred during an AIHA flare in the ES group and one in the AIHA group. There were three deaths in the ES group (10%) and one in the ITP group (2.3%), without any significant difference.
Discussion
In this multicentre cohort, patients with SLE and ES exhibited a distinct and more severe disease phenotype, characterised by increased severe flares, higher neurological involvement and a greater burden of severe infections compared with patients with isolated autoimmune cytopenias (ITP and AIHA).
The observed prevalence of ES in this retrospective cohort was slightly higher than that of AIHA, which contrasts with previously reported rates of 0.5–3%10 11 for ES versus 5–10% for AIHA in patients with SLE.2 4 This discrepancy may reflect a selection bias inherent to recruitment from tertiary care centres, as well as differences in ethnicity and referral patterns. Because our study was partly based on case reports and observational submissions from multiple centres, including non-tertiary centres, we did not have access to systematic denominator data for the overall SLE population. Therefore, the frequencies of ES and other autoimmune cytopenias reported here should not be interpreted as prevalence estimates in the general SLE population.
In our cohort, 65% of participants were of European origin, with a smaller proportion of North African patients, limiting direct comparison with prior Brazilian and Chinese cohorts. Notably, patients with ES were older at SLE onset than patients with ITP, consistent with previous observations that ITP and AIHA occur more frequently in younger-onset SLE.23 24
Disease activity at diagnosis was higher in patients with ES, as reflected by SLEDAI scores and a greater prevalence of anti-dsDNA positivity, suggesting a more aggressive systemic disease from onset. These findings align with the Brazilian cohort of Costallat et al, which reported life-threatening multisystem involvement in ES at SLE onset, including nephritis, serositis and neuropsychiatric manifestations.10 Although the Chinese cohort found lower rates of lupus nephritis during follow-up in patients with ES, these differences likely reflect population heterogeneity, study design and sample size.11 In the Chinese study, SLE disease activity was also high at SLE onset, with a median initial SLEDAI of 10. A recent cohort study also reported that patients with SLE and cytopenia had higher SLEDAI scores and higher anti-DNA antibody levels.25 In adjusted analyses using propensity score matching accounting for baseline SLEDAI and CS treatment, the association between ES and severe flares was conserved, suggesting that ES and higher disease activity at diagnosis partly explain the increased risk of severe flares in patients with SLE. However, these analyses are limited by a substantial proportion of missing data in the present study.
Regarding cytopenia flares, patients with ES required fewer maintenance therapies compared with patients with ITP, supporting the notion that targeted SLE therapy, including HCQ and IS agents, may mitigate haematological relapses in this subgroup. However, patients with ES experienced more severe infections, likely related to higher exposure to IS therapy, with neutropenia contributing minimally to infectious risk. This was concordant with Costallat et al findings.10 These findings underscore the need for vigilant infection monitoring and preventive strategies in patients with ES.
ES is known to be associated with a high rate of deaths and infections.12,14 In the present cohort, despite a standard follow-up for all patients, those with ES were more likely to have severe infections compared with patients with ITP or AIHA. This difference may be due to a higher rate of immunosuppression, since a large proportion of the patients with ES were on CS or IS at the time of infection. The incidence of thrombotic events was similar across groups, but a higher proportion of arterial APS events occurred in the ES group, highlighting a potential signal that warrants further investigation. These findings align with the prevailing clinical practice of exercising caution regarding thrombotic risk in patients with SLE and ES.26,28 Mortality was numerically higher in patients with ES, although this difference did not reach statistical significance, consistent with previous studies indicating elevated risk in ES but limited by small sample sizes and heterogeneity. In a Danish nationwide prospective cohort study comparing mortality between 242 patients with ES with that of patients with AIHA and ITP, a significantly higher mortality was found in patients with both primary and secondary ES compared with those with ITP.14
This study has several limitations. Its retrospective design may have resulted in incomplete data, and the relatively small sample size limits statistical power for several outcomes such as renal involvement, thrombosis or mortality. Inclusion from three centres within the same geographical area may introduce selection bias, and temporal heterogeneity (patients diagnosed between 1997–2024) may affect management and follow-up consistency. Residual confounding cannot be entirely excluded despite multivariable adjustment and propensity score analyses.
In conclusion, ES identifies a subset of patients with SLE with a more aggressive systemic phenotype, characterised by severe flares, renal and neurological involvement, and increased infection risk with a different immune profile (ie, higher titres of autoantibodies). There is also a higher risk of relapse in patients with APS and autoimmune cytopenias, suggesting a distinct phenotypic profile. Recognition of this phenotype may facilitate closer monitoring and inform therapeutic decisions, aiming to reduce morbidity in this high-risk population.
Supplementary material
Acknowledgements
The authors thank all patients for their participation and the physicians involved in patient recruitment. The authors also thank Verena Landel for her careful revision of the manuscript.
Footnotes
Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: This study was approved by the local ethics committee (CEREES-920064) and by the national data protection agency (Commission Nationale de l’Informatique et des Libertés, #DR-2020-219). According to French legislation, informed consent was not required for this retrospective observational study.
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Data availability statement
Data are available upon reasonable request.
References
- 1.Velo-García A, Castro SG, Isenberg DA. The diagnosis and management of the haematologic manifestations of lupus. J Autoimmun. 2016;74:139–60. doi: 10.1016/j.jaut.2016.07.001. [DOI] [PubMed] [Google Scholar]
- 2.Kokori SI, Ioannidis JP, Voulgarelis M, et al. Autoimmune hemolytic anemia in patients with systemic lupus erythematosus. Am J Med. 2000;108:198–204. doi: 10.1016/s0002-9343(99)00413-1. [DOI] [PubMed] [Google Scholar]
- 3.Durán S, Apte M, Alarcón GS, et al. Features associated with, and the impact of, hemolytic anemia in patients with systemic lupus erythematosus: LX, results from a multiethnic cohort. Arthritis Rheum. 2008;59:1332–40. doi: 10.1002/art.24020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Sultan SM, Begum S, Isenberg DA. Prevalence, patterns of disease and outcome in patients with systemic lupus erythematosus who develop severe haematological problems. Rheumatology (Oxford) 2003;42:230–4. doi: 10.1093/rheumatology/keg069. [DOI] [PubMed] [Google Scholar]
- 5.Jeffries M, Hamadeh F, Aberle T, et al. Haemolytic anaemia in a multi-ethnic cohort of lupus patients: a clinical and serological perspective. Lupus (Los Angel) 2008;17:739–43. doi: 10.1177/0961203308090990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Beyan E, Beyan C, Turan M. Hematological presentation in systemic lupus erythematosus and its relationship with disease activity. Hematology . 2007;12:257–61. doi: 10.1080/10245330701214145. [DOI] [PubMed] [Google Scholar]
- 7.Zhao H, Li S, Yang R. Thrombocytopenia in patients with systemic lupus erythematosus: significant in the clinical implication and prognosis. Platelets. 2010;21:380–5. doi: 10.3109/09537101003735564. [DOI] [PubMed] [Google Scholar]
- 8.Chen Z, Zhong H, Dong G. Thrombocytopenia as a Prognostic Marker for Systemic Lupus Erythematosus: A Systematic Review and Meta-analysis. Am J Med Sci. 2019;357:461–7. doi: 10.1016/j.amjms.2019.03.002. [DOI] [PubMed] [Google Scholar]
- 9.Fattizzo B, Marchetti M, Michel M, et al. Diagnosis and management of Evans syndrome in adults: first consensus recommendations. Lancet Haematol. 2024;11:e617–28. doi: 10.1016/S2352-3026(24)00144-3. [DOI] [PubMed] [Google Scholar]
- 10.Costallat GL, Appenzeller S, Costallat LTL. Evans syndrome and systemic lupus erythematosus: clinical presentation and outcome. Joint Bone Spine. 2012;79:362–4. doi: 10.1016/j.jbspin.2011.07.004. [DOI] [PubMed] [Google Scholar]
- 11.Zhang L, Wu X, Wang L, et al. Clinical Features of Systemic Lupus Erythematosus Patients Complicated With Evans Syndrome. Medicine (Abingdon) 2016;95:e3279. doi: 10.1097/MD.0000000000003279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Michel M, Chanet V, Dechartres A, et al. The spectrum of Evans syndrome in adults: new insight into the disease based on the analysis of 68 cases. Blood. 2009;114:3167–72. doi: 10.1182/blood-2009-04-215368. [DOI] [PubMed] [Google Scholar]
- 13.Fattizzo B, Michel M, Giannotta JA, et al. Evans syndrome in adults: an observational multicenter study. Blood Adv. 2021;5:5468–78. doi: 10.1182/bloodadvances.2021005610. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hansen DL, Möller S, Andersen K, et al. Evans syndrome in adults - incidence, prevalence, and survival in a nationwide cohort. Am J Hematol . 2019;94:1081–90. doi: 10.1002/ajh.25574. [DOI] [PubMed] [Google Scholar]
- 15.Hochberg MC. Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum . 1997;40:1725. doi: 10.1002/art.1780400928. [DOI] [PubMed] [Google Scholar]
- 16.Aringer M, Costenbader KH, Daikh DI, et al. 2019 EULAR/ACR Classification Criteria for Systemic Lupus Erythematosus. Arthritis Rheumatol. 2019;71:1400–12. doi: 10.1002/art.40930. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Petri M, Orbai A-M, Alarcón GS, et al. Derivation and validation of the Systemic Lupus International Collaborating Clinics classification criteria for systemic lupus erythematosus. Arthritis Rheum . 2012;64:2677–86. doi: 10.1002/art.34473. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Roussotte M, Gerfaud-Valentin M, Hot A, et al. Immune thrombocytopenia with clinical significance in systemic lupus erythematosus: a retrospective cohort study of 90 patients. Rheumatology (Oxford) 2022;61:3627–39. doi: 10.1093/rheumatology/keab925. [DOI] [PubMed] [Google Scholar]
- 19.Barcellini W, Fattizzo B, Zaninoni A, et al. Clinical heterogeneity and predictors of outcome in primary autoimmune hemolytic anemia: a GIMEMA study of 308 patients. Blood. 2014;124:2930–6. doi: 10.1182/blood-2014-06-583021. [DOI] [PubMed] [Google Scholar]
- 20.Jäger U, Barcellini W, Broome CM, et al. Diagnosis and treatment of autoimmune hemolytic anemia in adults: Recommendations from the First International Consensus Meeting. Blood Rev. 2020;41:100648. doi: 10.1016/j.blre.2019.100648. [DOI] [PubMed] [Google Scholar]
- 21.Khellaf M, Michel M, Schaeffer A, et al. Assessment of a therapeutic strategy for adults with severe autoimmune thrombocytopenic purpura based on a bleeding score rather than platelet count. Haematologica . 2005;90:829–32. [PubMed] [Google Scholar]
- 22.Fogarty PF, Tarantino MD, Brainsky A, et al. Selective validation of the WHO Bleeding Scale in patients with chronic immune thrombocytopenia. Curr Med Res Opin . 2012;28:79–87. doi: 10.1185/03007995.2011.644849. [DOI] [PubMed] [Google Scholar]
- 23.Nikolopoulos D, Kostopoulou M, Pieta A, et al. Evolving phenotype of systemic lupus erythematosus in Caucasians: low incidence of lupus nephritis, high burden of neuropsychiatric disease and increased rates of late-onset lupus in the “Attikon” cohort. Lupus (Los Angel) 2020;29:514–22. doi: 10.1177/0961203320908932. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ambrose N, Morgan TA, Galloway J, et al. Differences in disease phenotype and severity in SLE across age groups. Lupus (Los Angel) 2016;25:1542–50. doi: 10.1177/0961203316644333. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Govea-Peláez S, Campos-Guzmán J, Demichelis-Gómez R, et al. A Novel Scoring System for Diagnosing Systemic Lupus Erythematosus-Related Cytopenias: Insights From a Cohort Study. J Clin Rheumatol. 2025;31:249–55. doi: 10.1097/RHU.0000000000002239. [DOI] [PubMed] [Google Scholar]
- 26.Fattizzo B, Cecchi N, Bortolotti M, et al. Thrombopoietin receptor agonists in adult Evans syndrome: an international multicenter experience. Blood . 2022;140:789–92. doi: 10.1182/blood.2022016818. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Guitton Z, Terriou L, Lega JC, et al. Risk of thrombosis with anti-phospholipid syndrome in systemic lupus erythematosus treated with thrombopoietin-receptor agonists. Rheumatology (Oxford) 2018;57:1432–8. doi: 10.1093/rheumatology/key119. [DOI] [PubMed] [Google Scholar]
- 28.Marques C, Moulis G, Roussotte M, et al. Efficacy and Thrombotic Risk of Thrombopoietin Receptor Agonists for Immune Thrombocytopenia Secondary to Systemic Lupus and Antiphospholipid Syndrome: French Experience With 80 Patients. Am J Hematol. 2025;100:1972–82. doi: 10.1002/ajh.70052. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available upon reasonable request.