Skip to main content
Download PDF

Risk prediction of new-onset thrombocytopenia in patients with systemic lupus erythematosus: a multicenter prospective cohort study based on Chinese SLE treatment and research group (CSTAR) registry

Arthritis Research & Therapy volume 26, Article number: 229 (2024) Cite this article

Abstract

Background

Thrombocytopenia (TP) is a hematological manifestation of systemic lupus erythematosus (SLE) and is associated with unfavorable prognostic outcomes. This study aimed to develop a risk prediction model for new-onset TP in SLE patients.

Methods

Based on the multicenter prospective Chinese SLE Treatment and Research Group (CSTAR) registry, newly diagnosed SLE patients without TP at registration were enrolled. The least absolute shrinkage and selection operator (LASSO) method was used for variable selection. The final model was developed using multivariate Cox regression and displayed as a nomogram. Internal validation was achieved using enhanced Bootstrap resampling.

Results

During follow-up, thrombocytopenia developed in 80 (3.52%) of 2270 lupus patients. The final risk prediction model incorporated six predictors: baseline SDI score ≥ 1 (HR 2.207, 95% CI 1.350–3.609, p = 0.002), hemolytic anemia (HR 1.953, 95% CI 1.017–3.750, p = 0.044), low complement level (HR 2.351, 95% CI 1.004–5.505, p = 0.049), positive anti-β2GPI antibody (HR 1.805, 95% CI 1.084–3.004, p = 0.024), positive Coombs test (HR 1.878, 95% CI 1.123–3.141, p = 0.017), and positive anti-histone antibody (HR 1.595, 95% CI 1.017–2.587, p = 0.059). The model’s performance was indicated by C-index values for risk prediction at one, two, and three years, which were 0.741 (0.660–0.823), 0.730 (0.655–0.805), and 0.710 (0.643–0.777), respectively; and Brier scores of 0.018 (0.012–0.024), 0.025 (0.017–0.032), and 0.037 (0.027–0.046), respectively. Calibration curves were drawn and situated near the diagonal line.

Conclusions

This study developed the first risk prediction model for TP onset in lupus patients. Patients with baseline organ damage, hemolytic anemia, low complement, positive anti-histone antibody, positive anti-β2GPI antibody, or positive Coombs test were identified as being at high risk for thrombocytopenia and require further clinical attention.

Introduction

Systemic Lupus Erythematosus (SLE) is an autoimmune disease characterized by various autoantibodies and multiorgan involvement, including hematological abnormalities [1]. Thrombocytopenia (TP) is one of the hematological manifestations, with a reported prevalence of 8–20% in SLE patients [1,2,3,4]. Lupus thrombocytopenia is mainly caused by excessive platelet destruction and impaired platelet production [5].

The clinical presentations of SLE-TP patients are in high heterogeneity, ranging from asymptomatic cases to life-threatening bleeding. Previous studies have indicated that thrombocytopenia is associated with unfavorable prognostic outcomes. The LUMINA cohort reported that early thrombocytopenia is associated with high disease activity and severe disease damage, as well as an independent risk factor for mortality in lupus patients [6]. This association has been confirmed in several cohort studies [2, 7, 8]. SLE patients with thrombocytopenia were found more prone to serious complications, including hemorrhages [7] and thrombotic microangiopathies [9]. A retrospective study compared the prognosis of SLE patients with and without thrombocytopenia, finding significantly reduced six-year survival in SLE-TP patients [8]. Moreover, it was demonstrated that SLE-TP patients who achieved complete remission after treatment had significantly longer survival periods [7].

There have been no reported prediction models for new-onset thrombocytopenia in SLE patients. The Chinese SLE Treatment and Research Group (CSTAR) is the largest multicenter registry in China. This study aimed to utilize the prospective CSTAR cohort to develop an onset risk prediction model for thrombocytopenia in SLE patients, aiding clinicians in early prediction and stratified management.

Methods

The methods in this study were reported following the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement(see Supplementary Material 1)[10].

Patient enrollment

This study was conducted from January 2009 to July 2023, based on the multicenter prospective CSTAR cohort in China. The inclusion criteria of the CSTAR cohort complied with the 1997 revised SLE classification criteria from the American College of Rheumatology (ACR) [11], the 2012 version of the Systemic Lupus International Collaborating Clinics (SLICC) [12], or the 2019 classification criteria updated by the European League Against Rheumatism (EULAR)/ACR [13]. Patients enrolled in this study were required to meet all of the following criteria: meeting any of the SLE classification criteria within the CSTAR cohort; disease duration < two years (for an inception cohort); without thrombocytopenia at registration; with baseline and at least one follow-up record available. Participants with new onset thrombocytopenia due to drugs, infections, or tumors; missing core data; and loss to follow-up were excluded before data analysis. This study was approved by the Medical Ethics Committee of Peking Union Medical College Hospital and the local Medical Ethics Committee of other medical centers. Written informed consent forms were obtained from all the patients before registration.

Outcome and time

The outcome of this study was lupus-associated thrombocytopenia. Thrombocytopenia was defined as platelet counts < 100 × 109/L in at least two blood routine examinations, excluding thrombocytopenia caused by drugs, tumors, or infections. The study baseline was set at registration (the first visit) in the CSTAR cohort. Follow-up time was defined from the date of registration to the date of thrombocytopenia onset or the last follow-up record.

Data collection

Structured data collection and assessment protocols were uniformly implemented across CSTAR centers. Standardized quality control training for inter-laboratory comparison was conducted annually to ensure accuracy and comparability, particularly for autoantibody detections [14]. Detailed baseline records and follow-up updates were collected for all the participants, covering demographic characteristics, clinical manifestations, laboratory profiles, and treatment regimens. Demographic features included age at TMA onset, sex, and SLE duration. The Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI) [15] was utilized to evaluate the disease activity of lupus patients. The Physician Global Assessment (PGA) [16] was used for routine evaluation of disease flare. The SLICCC/ACR Damage Index (SDI) [17] was used to assess organ damage. Clinical characteristics included thrombocytopenia, hemolytic anemia, leukopenia, low complement, serositis, mucocutaneous, musculoskeletal, renal, neuropsychiatric, cardiovascular, and pulmonary involvement. Immunological profiles included the following antibodies: anti-double-stranded DNA (anti-dsDNA), anti-Smith (anti-Sm), anti-small nuclear ribonucleoprotein (anti-RNP), anti-ribosomal ribonucleoprotein (anti-rRNP), anti-Sjogren syndrome A (anti-SSA), anti-Sjogren syndrome B (anti-SSB), anti-nucleosome (anti-NuA), anti-histone (anti-Histone), anticardiolipin (ACL), anti-β2 glycoprotein I (anti-β2GPI), lupus anticoagulant (LA), and Coombs test. Notably, antiphospholipid antibody (aPL) positivity was defined as the presence of any of the three antiphospholipid antibodies. Treatment regimens included glucocorticoids (GC), hydroxychloroquine (HCQ), cyclophosphamide (CYC), mycophenolate mofetil (MMF), cyclosporine (CsA), methotrexate (MTX), tacrolimus, sirolimus, monoclonal antibodies, and intravenous immunoglobulin (IVIG).

Statistical analysis

Categorical data were presented as frequency (percentage, %). Quantitative data were reported as mean ± standard error, or median (interquartile range, [25th percentile, 75th percentile]). Quantitative variables were compared using either Student t-test or nonparametric Mann–Whitney U test, while categorical variables were analyzed using Pearson’s chi-square test or Fisher’s exact test. At least ten times the events per variable (EPV) were required for sample sizes. P < 0.05 was considered statistically significant, two-tailed. Data analyses were conducted using R software (V. 4.3.2).

Data preprocessing

To facilitate model application in clinical practice, all variables were appropriately categorized and transformed into dummy variables. For missing data, an exploration was conducted to reveal the missingness pattern, which was identified as missing at random (MAR). Multiple imputation techniques based on chained equations (MICE) using polytomous regression were employed to impute missing values [18]. All candidate predictors, estimates of the cumulative hazard function, and the outcome were included in the multiple imputation model. Five imputed datasets were generated. Rubin’s rule [19] was used to pool the estimates derived from these imputed datasets.

Variable selection and model development

Literature review and expert evaluations were combined to select preliminary candidate variables. The least absolute shrinkage and selection operator (LASSO) [20] method was used to select necessary variables from the candidates. The optimal variables were determined through cross-validation, satisfying sample size requirements and clinical significance checks. The final model was developed using the Cox proportional hazard regression model, with hazard ratios (HR) and 95% confidence intervals (CI) of predictors visualized in a forest plot. For each covariate, the Cox proportional hazard assumption was examined by the Schoenfeld residuals test [21]; cumulative risks were estimated using the Kaplan-Meier (KM) method, and differences were assessed using the log-rank test. The probability of new-onset thrombocytopenia in SLE patients can be calculated using the following formula: Pt = 1− S0(t) exp (onset index), where S0(t) represents the average occurrence probability at time t, and the onset index equals a sum of predictor values multiplied by their coefficients.

Model validation and model evaluation

Bootstrap resampling is a widely recognized method for internal validation, effectively conserving sample size [22]. The internal validation of this model was performed with 1000 bootstraps. As for the model performance, discrimination ability was evaluated using the receiver operating characteristic (ROC) curve and Harrell’s concordance index (C-index) [23]; predictive accuracy was assessed using the Brier score [24] and calibration curve [23].

Model presentation and risk stratification

To improve convenience for physicians and patients, a nomogram was constructed based on the multivariable Cox model. Meanwhile, the TP-onset risk score was calculated for each patient. According to individual risk scores, each participant was stratified into low-risk or high-risk groups. The optimal cutoff point was determined by the “survival_cutpoint” R package. The new-onset thrombocytopenia probabilities in different risk stratifications were described in the KM cumulative risk curve.

Results

Baseline characteristics

This study was conducted utilizing the multicenter prospective CSTAR cohort. An inception cohort of 2404 newly diagnosed lupus patients without thrombocytopenia at registration was formed. Among these patients, 41 cases were lost to follow-up, and 93 cases had missing core data. Consequently, 2270 patients were included in the final analysis. In our cohort, there existed missing data in 213 participants, visually presented in Figure S1 (see Supplementary Material 2). The missing proportions of predictive variables were all lower than 10%. The MICE method was employed for multiple imputations, generating five imputed datasets. Variable selection and model fitting were conducted separately on the five datasets, and the results were combined according to Rubin’s rules. Notably, 2057 patients without missing data were defined as the complete dataset and used for sensitivity analysis. Detailed information is available in the study flowchart (Fig. 1).

Fig. 1
figure 1

Study flowchart

Full size image

In the study, 2270 patients were included, of whom 80 developed TP during follow-up, resulting in an incidence of 3.52%. The baseline characteristics of this cohort are presented in Table 1. The median follow-up time was 629 (203–1305) days, approximately 1.72 years. The mean age at lupus onset was 33.79 ± 12.48 years, with 91.9% female patients. The median SLEDAI score at registration was 6 (2–10), with 287 cases (12.6%) showing baseline organ damage. Clinical manifestations included mucocutaneous (69.1%), musculoskeletal (54.8%), renal (40.6%), neuropsychiatric (8.6%), leukopenia (36.9%), and hemolytic anemia (7.4%). Approximately 76.7% of patients were discovered with anti-dsDNA antibody positivity, followed by anti-SSA (58.9%), anti-RNP (45.3%), and anti-Sm (43.4%). In our cohort, 87.5% of patients received glucocorticoids, with 110 cases (4.8%) receiving GC pulse therapy. HCQ (82.1%) was prescribed as a cornerstone treatment, and the five most common immunosuppressants were MMF (24.2%), CYC (17.2%), MTX (12.8%), tacrolimus (5.9%), and CsA (3.5%). Additionally, monoclonal antibodies and IVIG were used in 2.7% and 1.8% of patients, respectively.

Table 1 Baseline characteristics of participants
Full size table

Patients were divided into the SLE-TP group (n = 80) and the SLE-non-TP group (n = 2190). As shown in Table 1, the univariate analysis indicated that patients with baseline organ damage (SDI ≥ 1, 23.7% vs. 12.2%, p < 0.001), hemolytic anemia (16.2% vs. 7.1%, p = 0.004), hypocomplementemia (55.8% vs. 39.3%, p = 0.002), and serositis (26.2% vs. 16.2%, p = 0.026) were more likely to develop TP during follow-up. Higher positivity rates of anti-Histone (34.6% vs. 23.5%, p = 0.034), anti-β2GPI (29.9% vs. 18.6%, p = 0.020), and Coombs test (38.4% vs. 19.1%, p < 0.001) were also observed in the SLE-TP group. Additionally, more patients in the SLE-TP group received GC pulse, CYC, tacrolimus, and IVIG therapy. The characteristics of patients with different TP severity were displayed in Table S1 (see Supplementary Material 2). Higher positivity rates of anti-Sm and ACL were discovered in patients with moderate to severe TP.

Variable selection and model development

Incorporating literature review and expert opinions, 28 variables available in the CSTAR cohort were selected as preliminary candidate variables (see Table S2 in Supplementary Material 2). To enhance accuracy and reduce overfitting, the Lasso regularization method was employed for further variable selection (Figure S2, Supplementary Material 2). The β coefficients from the five imputed datasets are displayed in Table S3 (see Supplementary Material 2).

After confirming the clinical significance of the predictors, six variables were selected: baseline SDI score ≥ 1 (HR 2.207, 95% CI 1.350–3.609, p = 0.002), hemolytic anemia (HR 1.953, 95% CI 1.017–3.750, p = 0.044), low complement (HR 2.351, 95% CI 1.004–5.505, p = 0.049), anti-β2GPI antibody positivity (HR 1.805, 95% CI 1.084–3.004, p = 0.024), Coombs test positivity (HR 1.878, 95% CI 1.123–3.141, p = 0.017), and anti-histone antibody positivity (HR 1.595, 95% CI 1.017–2.587, p = 0.059). Definitions for each variable are detailed in Table S2 (see Supplementary Material 2). The final model was constructed using a multivariable Cox proportional hazards regression model. The β coefficients, estimated HRs, and 95% CIs of the risk predictors are presented in Table S4 (see Supplementary Material 2) and Fig. 2. For examination of the proportional hazard assumption, Schoenfeld individual tests of each variable are shown in Figure S3 (see Supplementary Material 2). Figure S4 (see Supplementary Material 2) plotted the cumulative risk curves for new-onset TP in lupus patients stratified by each predictor.

Fig. 2
figure 2

Forest plot of thrombocytopenia onset prediction model in SLE patients

Full size image

The cumulative thrombocytopenia risks of lupus individuals can be calculated using the following formulas: P (1-year) = 1 − 0.9955971exp(onset index), P (2-years) = 1 − 0.9940235exp(onset index), P (3-years) = 1 − 0.9905769exp(onset index). The onset index is defined as 0.7917 × SDI ≥ 1 + 0.6694 × hemolytic anemia + 0.8549 × low complement + 0.5905 × anti-β2GPI + 0.6304 × Coombs + 0.4666 × anti-Histone. The predictive model algorithm is illustrated in a nomogram (Fig. 3).

Fig. 3
figure 3

Nomogram for thrombocytopenia onset prediction model in SLE patients

Full size image

Nomogram aids in predicting the probability of developing thrombocytopenia for a lupus patient. Points for baseline SDI score ≥ 1, hemolytic anemia, low complement, Coombs test positivity, anti-β2GPI antibody positivity, and anti-histone antibody positivity can be obtained using a point caliper and summed for total points. The total points can be matched to the 1-year, 2-year, and 3-year cumulative incidence scales.

Model performance and internal validation

Model performance was evaluated among 2270 patients with 80 events. The apparent total C-index of the prediction model was 0.714 (0.653–0.775). The apparent C-index values for risk prediction at one, two, and three years were 0.741 (0.660–0.823), 0.730 (0.655–0.805), and 0.710 (0.643–0.777); while the Brier scores were 0.018 (0.012–0.024), 0.025 (0.017–0.032), and 0.037 (0.027–0.046), respectively. After enhanced Bootstrap resampling, the optimism-corrected C-index values for total, 1-year, 2-year, and 3-year predictions were 0.702, 0.726, 0.715 and 0.695. The optimism of the Brier score was lower than 0.001, resulting in minimal differences between apparent and optimism-corrected values. Detailed information is provided in Table S5 (see Supplementary Material 2). The calibration curve (1000 Bootstraps) depicted the comparison between predicted and observed risk probabilities (Fig. 4). Sensitivity analysis was conducted using the complete dataset (2057 patients, 73 events), as shown in Table S6 (see Supplementary Material 2).

Fig. 4
figure 4

The receiver operating characteristic (ROC) curves of the model and calibration curves in internal validation. A. Time-dependent ROC curves of the predictive model. B. 1-year calibration curve. C. 2-year calibration curve. D. 3-year calibration curve

Full size image

Risk stratification

The nomogram scores for SDI ≥ 1 (baseline organ damage), hemolytic anemia, low complement, anti-β2GPI positivity, anti-histone positivity, and Coombs test positivity were 91, 80, 100, 60, 59, and 71 points, respectively. The total point was calculated for every patient. A total risk score of 211 was determined as the optimal cut-off point. According to the nomogram-derived risk scores, participants were classified into a high-risk group (> 211) and a low-risk group (0–211), with TP-onset risks of 7.59% and 2.52%, respectively. The risk stratifications for new-onset TP are shown in Table S7 (see Supplementary Material 2). The cumulative risk curves of two different risk groups were drawn in Fig. 5.

Fig. 5
figure 5

Cumulative risk curve of thrombocytopenia onset in SLE patients after risk stratification. A. Imputed data. B. Complete data

Full size image

Discussion

This is the first clinical prediction model focused on new-onset TP risk in SLE patients based on a large prospective multicenter cohort. Among the 2270 included patients, TP occurred in 80 cases (3.52%) during follow-up. The final model integrated six independent risk factors, including baseline organ damage, hemolytic anemia, hypocomplementemia, positive anti-β2GPI antibody, positive anti-histone antibody, and positive Coombs test. A nomogram was well constructed and appropriately calibrated to predict TP risk at one, two, and three years for lupus patients.

SDI is an effective score system for irreversible lupus organ damage assessment [17]. It has been clarified that baseline organ damage is associated with exacerbation of existing damage or accumulation of new damage [25]. Our results indicated that baseline organ damage (SDI ≥ 1) was identified as a predictive factor for TP onset in lupus patients, as supported by other clinical studies [4, 26] in which SLE-TP patients exhibited increased SDI scores. A meta-analysis [27] suggested that SLE-TP patients were discovered with higher risks of mortality and end-organ damage.

Clinical practice and literature reviews have demonstrated that lupus thrombocytopenia often coexists with other hematological manifestations [28, 29]. An independent association between hemolytic anemia and thrombocytopenia was revealed in our patients, consistent with previous research [4, 29,30,31,32]. Additionally, several researchers reported that patients with SLE-TP presented a higher proportion of concomitant leukopenia [8, 28], which did not reach statistical significance in our analysis, warranting further investigation or systematic review.

Low complement levels have been identified as a risk factor for developing lupus among immune thrombocytopenic purpura patients, according to hematologists [33]. Interestingly, hypocomplementemia was also found to be relevant in lupus thrombocytopenia onset [4, 8, 28, 30]. The probable explanation could be that the low complement level reflects an active disease phase. Furthermore, a higher prevalence of serositis was presented in our patients with SLE-TP, in accordance with previous findings [8, 28, 30]. Although serositis failed to be included in our final model, it’s worth noting that this topic still holds significant potential for future research.

Our study incorporated anti-β2GPI antibody positivity into the predictive model. Indeed, a strong correlation is well-established between lupus thrombocytopenia and antiphospholipid antibodies. Positive autoantibodies, including ACL, anti-β2GPI, and LA, have been more frequently tested in patients with TP during lupus follow-up [4, 26, 31, 32, 34]. A high-quality meta-analysis [34] summarized that aPL-positive lupus patients held a 2.48-fold increased risk of TP occurrence. Conflicting reports exist regarding other autoantibodies, such as anti-Sm, anti-RNP, and anti-SSA antibodies [35, 36]. Aligning with our findings, another clinical research [8] based on the CSTAR cohort indicated no significant associations with these antibodies. Notably, the higher proportion of anti-histone positivity in our SLE-TP patients requires further support from additional studies. These discrepancies may stem from differences in racial composition and disease stages among various cohorts.

Moreover, the positive Coombs test was considered an important serological predictor in our cohort, supported by Thelma Skare et al. [37]. Coombs tests provide evidence for autoimmune hemolytic anemia diagnosis, through detecting antibodies bound to red blood cell surfaces. However, patients with positive Coombs tests may be asymptomatic, while those with negative results could suffer severe hemolysis, indicating no one-to-one relationship between them. Consequently, a positive direct Coombs test without hemolytic anemia has been included in the 2012 SLE classification criteria [12]. Additionally, a correlation between positive aPLs and positive Coombs tests has been proposed, possibly due to the cross-reactivity of aPLs with red blood cell membrane phospholipid epitopes [38]. In general, the underlying reasons are likely intricate, necessitating further research for exploration.

Excessive platelet destruction and impaired platelet production are recognized as the two main causes of lupus thrombocytopenia [5]. Long-lived plasma cells can produce anti-platelet antibodies, contributing to macrophage-mediated platelet phagocytosis. CD8+ T cells and the complement pathway are also involved in platelet destruction. Additionally, thrombopoietin (TPO) antibodies may inhibit the formation, maturation, and production of platelets within the bone marrow. If feasible, it is believed that incorporating variables derived from lymphocyte subsets or TPO antibodies may enhance model performance.

Overall, the predictive model could aid clinicians in accurately assessing the risk of thrombocytopenia in SLE patients, effectively identifying high-risk individuals. This would allow for closer monitoring and earlier intervention, ultimately improving patient outcomes. Presented as an easy-to-use nomogram, the model can stratify risk levels based on the predicted probability of TP occurrence, offering a valuable tool for clinical practice by providing a practical reference for risk management and decision-making.

Our predictive model possesses some advantages. As part of a longitudinal prospective cohort study, baseline characteristics analyses of lupus patients without TP were conducted to predict future onset risk rather than identify potential patients. Moreover, easily changeable variables were excluded to avoid potentially contradictory outcomes at different time points. The included variables are accessible and suitable in clinical settings.

However, limitations also exist in this study. Firstly, all samples were utilized for model development to maximize performance, resulting in an insufficient number of patients for external validation. Model generalizability needs to be taken into account, especially before application to other regions or ethnicities. Further studies with larger sample sizes are required to validate these findings. Secondly, our cohort had missing data, which was compensated using the multiple imputation method. In the real world, not all patients receive every clinical assessment or laboratory examination, often opting for individualized and cost-effective approaches, which may contribute to the data missingness. Nevertheless, we acknowledge the complexity of missing patterns and the possible bias introduced into model predictions. Thirdly, the model was developed in an inception cohort, potentially restricting its applicability in specific scenarios.

Conclusions

In conclusion, the first risk prediction model for new-onset thrombocytopenia in lupus patients has been developed and internally validated in the multicenter prospective CSTAR cohort. Baseline organ damage, hemolytic anemia, low complement, positive anti-β2GPI antibodies, anti-histone antibodies, and Coombs tests were identified as independent risk factors. Closer attention should be paid to patients with these specific traits. Further prospective research and longer follow-ups are necessary to provide more strategies for early recognition and risk stratification.

Data availability

No datasets were generated or analysed during the current study.

Abbreviations

ACL:

Anticardiolipin antibody

ACR:

American college of rheumatology

Anti-β2GPI:

Anti-β2 glycoprotein I antibody

Anti-dsDNA:

Anti-double-stranded DNA

Anti-NuA:

Anti-nucleosome antibody

Anti-RNP:

Anti-small nuclear ribonucleoprotein antibody

Anti-rRNP:

Anti-ribosomal ribonucleoprotein antibody

Anti-Sm:

Anti-Smith antibody

Anti-SSA:

Anti-Sjogren syndrome A antibody

Anti-SSB:

Anti-Sjogren syndrome B antibody

aPL:

Antiphospholipid antibody

CI:

Confidence interval

C-index:

Harrell’s concordance index

CsA:

Cyclosporine

CSTAR:

Chinese systemic lupus erythematosus treatment and research group

CYC:

Cyclophosphamide

EPV:

Events per variable

EULAR:

European league against rheumatism

HCQ:

Hydroxychloroquine

HR:

Hazard ratio

IVIG:

Intravenous immunoglobulin

GC:

Glucocorticoids

KM:

Kaplan-Meier

LA:

Lupus anticoagulant

LASSO:

Least absolute shrinkage and selection operator

MAR:

Missing at random

MICE:

Multiple imputation techniques based on chained equations

MMF:

Mycophenolate mofetil

MTX:

Methotrexate

PGA:

Physician global assessment

ROC:

Receiver operating characteristic

SDI:

Systemic lupus international collaborating clinics/American college of rheumatology damage index

SLE:

Systemic lupus erythematosus

SLEDAI:

Systemic lupus erythematosus disease activity index

SLICC:

Systemic lupus international collaborating clinics

TP:

Thrombocytopenia

TPO:

Thrombopoietin

TRIPOD:

Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis

References

  1. Fanouriakis A, Tziolos N, Bertsias G, Boumpas DT. Update on the diagnosis and management of systemic lupus erythematosus. Ann Rheum Dis. 2021;80(1):14–25.

    Article  PubMed  Google Scholar 

  2. Costa Pires T, Caparrós-Ruiz R, Gaspar P, Isenberg DA. Prevalence and outcome of thrombocytopenia in systemic lupus erythematous: single-centre cohort analysis. Clin Exp Rheumatol. 2021;39(3):601–5.

    Article  PubMed  Google Scholar 

  3. Fanouriakis A, Bertsias G, Boumpas DT. Population-based studies in systemic lupus erythematosus: immune thrombocytopenic purpura or ‘blood-dominant’ lupus? Ann Rheum Dis. 2020;79(6):683–4.

    Article  PubMed  Google Scholar 

  4. Cornudella Lema J, Sánchez-González B, Carrión-Barberà I, Vázquez Montes de Oca S, García Pallarols F, Salman-Monte TC. Immune thrombocytopenia in systemic lupus erythematosus: Prevalence, risk factors, and a novel predictive model for risk assessment. Med Clin (Barc). 2024;162(10):461–9.

  5. Jiang Y, Cheng Y, Ma S, Li T, Chen Z, Zuo X, et al. Systemic lupus erythematosus-complicating immune thrombocytopenia: From pathogenesis to treatment. J Autoimmun. 2022;132:102887.

    Article  CAS  PubMed  Google Scholar 

  6. Fernandez M, Alarcon GS, Apte M, Andrade RM, Vila LM, Reveille JD, et al. Systemic lupus erythematosus in a multiethnic US cohort: XLIII. The significance of thrombocytopenia as a prognostic factor. Arthritis Rheum. 2007;56(2):614–21.

    Article  PubMed  Google Scholar 

  7. Jung JH, Soh MS, Ahn YH, Um YJ, Jung JY, Suh CH, et al. Thrombocytopenia in Systemic Lupus Erythematosus: Clinical Manifestations, Treatment, and Prognosis in 230 Patients. Med (Baltim). 2016;95(6):e2818.

    Article  CAS  Google Scholar 

  8. Jiang N, Li M, Zhang M, Xu J, Jiang L, Gong L, et al. Chinese SLE Treatment and Research group (CSTAR) registry: Clinical significance of thrombocytopenia in Chinese patients with systemic lupus erythematosus. PLoS ONE. 2019;14(11):e0225516.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Yang F, Tian J, Peng L, Zhang L, Liu J, Tian X, et al. Thrombocytopenia Is an Independent Risk Factor for the Prognosis of Thrombotic Microangiopathy in Chinese Patients With Systemic Lupus Erythematosus. Front Med (Lausanne). 2021;8:772607.

    Article  PubMed  Google Scholar 

  10. Collins GS, Reitsma JB, Altman DG, Moons KG. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement. BMJ. 2015;350:g7594.

    Article  PubMed  Google Scholar 

  11. Hochberg MC. Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum. 1997;40(9):1725.

    Article  CAS  PubMed  Google Scholar 

  12. Petri M, Orbai AM, Alarcón GS, Gordon C, Merrill JT, Fortin PR, et al. Derivation and validation of the Systemic Lupus International Collaborating Clinics classification criteria for systemic lupus erythematosus. Arthritis Rheum. 2012;64(8):2677–86.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Aringer M, Costenbader K, Daikh D, Brinks R, Mosca M, Ramsey-Goldman R, et al. 2019 European League Against Rheumatism/American College of Rheumatology classification criteria for systemic lupus erythematosus. Ann Rheum Dis. 2019;78(9):1151–9.

    Article  PubMed  Google Scholar 

  14. Li M, Wang Y, Zhao J, Wang Q, Wang Z, Tian X, et al. Chinese SLE Treatment and Research Group (CSTAR) Registry 2009–2019: Major Clinical Characteristics of Chinese Patients with Systemic Lupus Erythematosus. Rheumatol Immunol Res. 2021;2(1):43–7.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Gladman DD, Ibañez D, Urowitz MB. Systemic lupus erythematosus disease activity index 2000. J Rheumatol. 2002;29(2):288–91.

    PubMed  Google Scholar 

  16. Petri M, Genovese M, Engle E, Hochberg M. Definition, incidence, and clinical description of flare in systemic lupus erythematosus. A prospective cohort study. Arthritis Rheum. 1991;34(8):937–44.

    Article  CAS  PubMed  Google Scholar 

  17. Gladman DD, Goldsmith CH, Urowitz MB, Bacon P, Fortin P, Ginzler E, et al. The Systemic Lupus International Collaborating Clinics/American College of Rheumatology (SLICC/ACR) Damage Index for Systemic Lupus Erythematosus International Comparison. J Rheumatol. 2000;27(2):373–6.

    CAS  PubMed  Google Scholar 

  18. Lang KM, Wu W. A Comparison of Methods for Creating Multiple Imputations of Nominal Variables. Multivar Behav Res. 2017;52(3):290–304.

    Article  Google Scholar 

  19. Rubin DB. Multiple imputation. In: Flexible Imputation of Missing Data, Second Edition. Chapman and Hall/CRC; 2018: 29–62.

  20. Tibshirani R. The lasso method for variable selection in the Cox model. Stat Med. 1997;16(4):385–95.

    Article  CAS  PubMed  Google Scholar 

  21. SCHOENFELD D. Partial residuals for the proportional hazards regression model. Biometrika. 1982;69(1):239–41.

    Article  Google Scholar 

  22. Steyerberg EW, Harrell FE Jr., Borsboom GJ, Eijkemans MJ, Vergouwe Y, Habbema JD. Internal validation of predictive models: efficiency of some procedures for logistic regression analysis. J Clin Epidemiol. 2001;54(8):774–81.

    Article  CAS  PubMed  Google Scholar 

  23. Alba AC, Agoritsas T, Walsh M, Hanna S, Iorio A, Devereaux PJ, et al. Discrimination and Calibration of Clinical Prediction Models: Users’ Guides to the Medical Literature. JAMA. 2017;318(14):1377–84.

    Article  PubMed  Google Scholar 

  24. BRIER GW. VERIFICATION OF FORECASTS EXPRESSED IN TERMS OF PROBABILITY. Mon Weather Rev. 1950;78(1):1–3.

    Article  Google Scholar 

  25. Sutton EJ, Davidson JE, Bruce IN. The systemic lupus international collaborating clinics (SLICC) damage index: a systematic literature review. Semin Arthritis Rheum. 2013;43(3):352–61.

    Article  PubMed  Google Scholar 

  26. Mohamed SS, Gamal SM, Mokbel A, Alkamary AK, Siam I, Soliman A, et al. Thrombocytopenia and disease outcomes in a cohort of patients with systemic lupus erythematosus. A post hoc analysis of the COMOSLE-EGYPT study. Int J Rheum Dis. 2024;27(1):e15016.

    Article  CAS  PubMed  Google Scholar 

  27. 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(6):461–7.

    Article  PubMed  Google Scholar 

  28. Mao YM, Shi PL, Wu L, Hu YQ, He YS, Xiang K, et al. Prevalence and influential factors of thrombocytopaenia in systemic lupus erythematosus patients: a retrospective analysis of 3140 cases in a Chinese population. Lupus. 2020;29(7):743–50.

    Article  CAS  PubMed  Google Scholar 

  29. Artım-Esen B, Çene E, Şahinkaya Y, Erdugan M, Oğuz E, Gül A, et al. Autoimmune haemolytic anaemia and thrombocytopaenia in a single-centre cohort of patients with systemic lupus erythematosus from Turkey: clinical associations and effect on disease damage and survival. Lupus. 2019;28(12):1480–7.

    Article  PubMed  Google Scholar 

  30. Alger M, Alarcon-Segovia D, Rivero SJ. Hemolytic anemia and thrombocytopenic purpura: two related subsets of systemic lupus erythematosus. J Rheumatol. 1977;4(4):351–7.

    CAS  PubMed  Google Scholar 

  31. Su K, Cheng H, Jia Z, Yuan Y, Yang H, Gao Q et al. Predictors of refractory risk in systemic lupus erythematosus-related thrombocytopenia: a dual-centre retrospective study. Lupus Sci Med. 2022;9(1).

  32. Gonzalez-Naranjo LA, Betancur OM, Alarcon GS, Ugarte-Gil MF, Jaramillo-Arroyave D, Wojdyla D, et al. Features associated with hematologic abnormalities and their impact in patients with systemic lupus erythematosus: Data from a multiethnic Latin American cohort. Semin Arthritis Rheum. 2016;45(6):675–83.

    Article  PubMed  Google Scholar 

  33. Ahn SM, Choi EJ, Oh JS, Kim YG, Lee CK, Yoo B, et al. Prognostic factors for the development of systemic lupus erythematosus in patients with immune thrombocytopenia. Arthritis Res Ther. 2022;24(1):213.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Chock YP, Moulinet T, Dufrost V, Erkan D, Wahl D, Zuily S. Antiphospholipid antibodies and the risk of thrombocytopenia in patients with systemic lupus erythematosus: A systematic review and meta-analysis. Autoimmun Rev. 2019;18(11):102395.

    Article  CAS  PubMed  Google Scholar 

  35. To CH, Petri M. Is antibody clustering predictive of clinical subsets and damage in systemic lupus erythematosus? Arthritis Rheum. 2005;52(12):4003–10.

    Article  CAS  PubMed  Google Scholar 

  36. Rodsaward P, Chottawornsak N, Suwanchote S, Rachayon M, Deekajorndech T, Wright HL et al. The clinical significance of antinuclear antibodies and specific autoantibodies in juvenile and adult systemic lupus erythematosus patients. Asian Pac J Allergy Immunol. 2021;39(4).

  37. Skare T, Picelli L, Dos Santos TAG, Nisihara R. Direct antiglobulin (Coombs) test in systemic lupus erythematosus patients. Clin Rheumatol. 2017;36(9):2141–4.

    Article  PubMed  Google Scholar 

  38. Win N, Islam SI, Peterkin MA, Walker ID. Positive direct antiglobulin test due to antiphospholipid antibodies in normal healthy blood donors. Vox Sang. 1997;72(3):182–4.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

We would like to express our sincere thanks to all the medical centers and researchers who participated in data collecting, updating and sharing.

Funding

This study was supported by the Chinese National Key Technology R&D Program, Ministry of Science and Technology (2021YFC2501300, 2021YFC2501304), Beijing Municipal Science & Technology Commission (No. Z201100005520022, 23, 25–27), CAMS Innovation Fund for Medical Sciences (CIFMS) (2021-I2M-1-005, 2022-I2M-1-004, 2023-I2M-2-005), National High-Level Hospital Clinical Research Funding (2022-PUMCH-B-013, C-002, D-009).

Author information

Author notes

  1. Yupei Zhang and Nan Jiang contributed equally to this study.

Authors and Affiliations

  1. Department of Rheumatology and Clinical Immunology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, Peking Union Medical College, National Clinical Research Center for Dermatologic and Immunologic Diseases (NCRC-DID), Ministry of Science & Technology, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Key Laboratory of Rheumatology and Clinical Immunology, Ministry of Education, Beijing, China

    Yupei Zhang, Nan Jiang, Jiuliang Zhao, Qian Wang, Xinping Tian, Mengtao Li & Xiaofeng Zeng

  2. Department of Rheumatology, The Second Affiliated Hospital of Nanchang University, Nanchang, China

    Xinwang Duan

  3. Department of Rheumatology, First Affiliated Hospital of Kunming Medical University, Kunming, China

    Jian Xu

  4. Department of Rheumatology, People’s Hospital of Xinjiang Uygur Autonomous Region, Urumqi, China

    Lijun Wu

  5. Department of Rheumatology, General Hospital of Tianjin Medical University, Tianjin, China

    Wei Wei

  6. Department of Rheumatology, The First Affiliated Hospital of China Medical University, Shenyang, China

    Weiguo Xiao

  7. Department of Rheumatology, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, China

    Li Luo

  8. Department of Rheumatology, The First Affiliated Hospital of Jilin University, Changchun, China

    Zhenyu Jiang

  9. Department of Epidemiology and Biostatistics, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences & School of Basic Medicine Peking Union Medical College, Beijing, China

    Yanhong Wang

Authors
  1. Yupei Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Nan Jiang
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Xinwang Duan
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Jian Xu
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Lijun Wu
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Wei Wei
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Weiguo Xiao
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Li Luo
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Zhenyu Jiang
    View author publications

    You can also search for this author inPubMed Google Scholar

  10. Yanhong Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  11. Jiuliang Zhao
    View author publications

    You can also search for this author inPubMed Google Scholar

  12. Qian Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  13. Xinping Tian
    View author publications

    You can also search for this author inPubMed Google Scholar

  14. Mengtao Li
    View author publications

    You can also search for this author inPubMed Google Scholar

  15. Xiaofeng Zeng
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

YPZ, NJ, and MTL contributed to the conception and design of the study. YPZ accomplished the data analysis and drafted the manuscript. NJ, YHW, JLZ, QW, XPT, MTL, and XFZ critically revised the intellectual content. YPZ, NJ, XWD, JX, LJW, WW, WGX, LL, and ZYJ contributed to the acquisition and documentation of data. MTL is responsible for reading the proofs and communicating with the other authors.

Corresponding author

Correspondence to Mengtao Li.

Ethics declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Peking Union Medical College Hospital and conducted in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki). All the patients were provided with written informed consent for participation.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, Y., Jiang, N., Duan, X. et al. Risk prediction of new-onset thrombocytopenia in patients with systemic lupus erythematosus: a multicenter prospective cohort study based on Chinese SLE treatment and research group (CSTAR) registry. Arthritis Res Ther 26, 229 (2024). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13075-024-03460-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13075-024-03460-0

Keywords

Arthritis Research & Therapy

ISSN: 1478-6362

Contact us