Unless otherwise indicated, values are number (%). TE = thrombotic events; IQR = interquartile range; SLEDAI-2K = Systemic Lupus Erythematosus Disease Activity Index 2000; AMS = adjusted mean SLEDAI score; BP = blood pressure; SLICC/ACR DI = Systemic Lupus International Collaborating clinics/American College of Rheumatology-Damage Index.
Contribution from the Field
Incidence rates of arterial and venous thrombosis after diagnosis of systemic lupus erythematosus
Article first published online: 4 AUG 2005
Copyright © 2005 by the American College of Rheumatology
Arthritis Care & Research
Volume 53, Issue 4, pages 609–612, 15 August 2005
How to Cite
Sarabi, Z. S., Chang, E., Bobba, R., Ibanez, D., Gladman, D., Urowitz, M. and Fortin, P. R. (2005), Incidence rates of arterial and venous thrombosis after diagnosis of systemic lupus erythematosus. Arthritis & Rheumatism, 53: 609–612. doi: 10.1002/art.21314
- Issue published online: 4 AUG 2005
- Article first published online: 4 AUG 2005
- Manuscript Accepted: 1 JAN 2005
- Manuscript Received: 24 SEP 2004
- Arthritis and Autoimmune Research Centre from the University Health Network
- 2003 Metro A. Ogryzlo Clinical/Research Fellowship in Adult Rheumatology from The Arthritis Society
- Arthritis Centre of Excellence, University Health Network
- Canadian Arthritis Network Award and Thrombosis Interest Group of Canada
- Joint Scientist Award from The Arthritis Society and the Institute of Musculoskeletal Health and Arthritis
Thrombosis has been reported in 7.2–12% of persons with systemic lupus erythematosus (SLE) (1–3). In a 10-year prospective cohort study of patients with SLE, the most frequent causes of death were active SLE (26.5%), thrombosis (26.5%), and infection (25%), with thrombosis dominating the second 5-year period of followup (4). The incidence rates of thrombosis after a diagnosis of SLE remain unknown. We report the yearly incidence of arterial and venous thrombosis in an inception cohort of patients with SLE.
Patients and Methods
We studied an inception cohort of patients with SLE from the University of Toronto Lupus Database; SLE was defined as either fulfillment of at least 4 1982 American College of Rheumatology classification criteria (5), or fulfillment of 3 criteria and a histologic lesion of SLE (such as a renal or skin biopsy specimen). Since 1970, all patients seen in our clinic have been asked to provide informed consent and to participate in our registry. Patients are followed prospectively on a quarterly basis according to a standard protocol. We selected all patients with a diagnosis of SLE that was made within 1 year of their first visit. Patients who had thrombosis more than 6 months prior to their diagnosis were excluded. The treating physician recorded thrombotic events (TE) as part of the protocol. The outcomes included TE, divided into arterial TE (ATE) and venous TE (VTE), since diagnosis of SLE. ATE was defined as angina, myocardial infarction, stroke, or peripheral arterial thrombosis, and VTE was defined as deep vein thrombosis (venous circulation in the extremities or an internal organ) or pulmonary emboli.
All statistical analyses were performed using SAS software, version 8.02 (SAS Institute, Cary, NC) and Microsoft Excel (Microsoft, Redmond, WA). Demographic and clinic characteristics were compared using the chi-square test for categorical data and either Student's t-test (parametric) or Wilcoxon's 2-sample rank sum test (nonparametric) for continuous data. Incidence rates of TE, ATE, and VTE were calculated for each year of followup, and 95% confidence intervals (95% CIs) were obtained based on a Poisson distribution. Kaplan-Meier survival curves were also estimated for each outcome.
Of the 1,127 patients seen in the Lupus Clinic, 583 were excluded either because they were not seen in the clinic within 1 year of their SLE diagnosis (n = 568), or they had a TE more than 6 months prior to their SLE diagnosis (n = 15). Therefore, a total of 544 patients were eligible for this study. These patients were followed for a median duration of 6.3 years (interquartile range 2–12 years) and contributed a total of 4,210.61 person-years of followup time. Eighty-eight patients (16%) had a TE after diagnosis of SLE, 58 patients (11%) had an ATE, and 30 patients (6%) had a VTE.
The characteristics of the patients who had a TE and those who did not are listed in Table 1. A total of 467 (86%) patients were women, with similar distributions in both groups. Those who had had a TE were less likely to be alive at the last followup than those patients who had not had a TE (73% versus 85%; P < 0.01), were older at the time of diagnosis with SLE (mean ± SD age 39.7 ± 14.9 years versus 34.4 ± 13.7 years; P < 0.001), and had a shorter median followup time during the first 20 years after diagnosis (4.2 versus 6.6 years; P = 0.001). The TE group had higher baseline disease activity and average disease activity over time, as measured by the adjusted mean Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K) (6), than patients in the non-TE group. There was no difference between the 2 groups in the percentage of patients with diabetes or hypertension, but the TE group had a higher proportion of use of antihypertensive medication at the time of event or last followup (50% versus 35%; P = 0.03) and a higher proportion of smokers (57% versus 35%; P < 0.001). Disease damage, as measured by the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (7), was higher in the TE group than in the non-TE group (mean ± SD 1.6 ± 1.7 versus 1.1 ± 1.5; P = 0.01).
|Total group (n = 544)||TE (n = 88)||No TE (n = 456)||P|
|Alive||450 (82.9)||64 (72.7)||386 (84.8)||0.0058|
|Female||467 (85.9)||73 (83.0)||394 (86.4)||0.3954|
|Age at diagnosis, mean ± SD||35.2 ± 14.0||39.7 ± 14.9||34.4 ± 13.7||0.0009|
|Age at TE or last visit, mean ± SD||43.4 ± 15.1||45.5 ± 15.3||42.9 ± 15.0||0.1372|
|Disease duration at TE or last visit, median (IQR)||6.3 (2–12.2)||4.2 (1–9.8)||6.6 (2.2–13)||0.0014|
|SLEDAI-2K at presentation, mean ± SD||11.2 ± 8.6||13.2 ± 10.7||10.8 ± 8.1||0.0510|
|AMS at TE or last visit, mean ± SD||6.7 ± 6.0||9.3 ± 9.4||6.1 ± 5.0||0.0027|
|Diabetes (ever)||37 (6.9)||7 (8.1)||30 (6.6)||0.6341|
|Smoker (ever)||208 (38.3)||48 (54.6)||160 (35.2)||0.0006|
|High BP†||247 (46.3)||46 (53.5)||201 (44.9)||0.1419|
|Antihypertensive medication||152 (36.9)||29 (50.0)||123 (34.8)||0.0256|
|Hypertensive‡||278 (51.8)||53 (60.2)||225 (50.1)||0.0825|
|SLICC/ACR DI at TE or last visit, mean ± SD||1.2 ± 1.6||1.6 ± 1.7||1.1 ± 1.5||0.0161|
The annual incidence rate and 95% CI for all TE, ATE, and VTE for the first 20 years after diagnosis of SLE are listed in Table 2. Annual incidence rates for TE and ATE, in events per 100 person-years, ranged from 0 to 5.19 (95% CI 0.00–7.02 and 1.90–11.29, respectively). The incidence of ATE increased in the first year after SLE diagnosis with a rate of 2.42 (95% CI 1.25–4.22). Afterwards, the rates for ATE fluctuated but were within a constant range for the entire observation period. Wider ranges in the 95% CI were observed in the later periods due to a lower number of person-years of observation. The incidence rate of VTE, in events per 100 person-years, ranged from 0 to 2.01 (95% CI 0.00–7.02 and 0.97–3.70, respectively) and was lower than the incidence rate of ATE throughout the observation period. VTE occurred more frequently in the first 8 years of disease with rare occurrence after that time. In fact, for both ATE and VTE, most of the events occurred in the first 5 years after diagnosis of SLE.
|Years after SLE diagnosis||Person-years||Number of events||Incidence rate (95% CI)†|
|0 to < 1||496.46||22||12||10||4.43 (2.78–6.71)||2.42 (1.25–4.22)||2.01 (0.97–3.70)|
|1 to < 2||430.81||9||4||5||2.09 (0.96–3.96)||0.93 (0.25–2.38)||1.16 (0.38–2.71)|
|2 to < 3||386.17||7||7||0||1.81 (0.73–3.73)||1.81 (0.73–3.73)||0.00 (0.00–0.96)|
|3 to < 4||345.35||4||2||2||1.16 (0.32–2.97)||0.58 (0.07–2.09)||0.58 (0.07–2.09)|
|4 to < 5||315.75||5||2||3||1.58 (0.51–3.70)||0.63 (0.08–2.29)||0.95 (0.20–2.78)|
|5 to < 6||292.74||5||2||3||1.71 (0.55–3.99)||0.68 (0.08–2.47)||1.02 (0.21–3.00)|
|6 to < 7||263.80||7||4||3||2.65 (1.07–5.47)||1.52 (0.41–3.88)||1.14 (0.24–3.32)|
|7 to < 8||239.23||4||3||1||1.67 (0.46–4.28)||1.25 (0.26–3.67)||0.42 (0.01–2.33)|
|8 to < 9||222.45||2||2||0||0.90 (0.11–3.25)||0.90 (0.11–3.25)||0.00 (0.00–1.66)|
|9 to < 10||195.35||2||2||0||1.02 (0.12–3.70)||1.02 (0.12–3.70)||0.00 (0.00–1.89)|
|10 to < 11||170.42||2||1||1||1.17 (0.14–4.24)||0.59 (0.02–3.27)||0.59 (0.02–3.27)|
|11 to < 12||146.98||2||2||0||1.36 (0.16–4.91)||1.36 (0.16–4.91)||0.00 (0.00–2.51)|
|12 to < 13||132.16||3||3||0||2.27 (0.47–6.64)||2.27 (0.47–6.64)||0.00 (0.00–2.79)|
|13 to < 14||115.68||6||6||0||5.19 (1.90–11.29)||5.19 (1.90–11.29)||0.00 (0.00–3.19)|
|14 to < 15||98.70||1||0||1||1.01 (0.03–5.64)||0.00 (0.00–3.74)||1.01 (0.03–5.64)|
|15 to < 16||89.00||1||1||0||1.12 (0.03–6.26)||1.12 (0.03–6.26)||0.00 (0.00–4.15)|
|16 to < 17||81.67||1||1||0||1.22 (0.04–6.82)||1.22 (0.04–6.82)||0.00 (0.00–4.52)|
|17 to < 18||73.14||2||2||0||2.73 (0.33–9.87)||2.73 (0.33–9.87)||0.00 (0.00–5.05)|
|18 to < 19||62.20||3||2||1||4.82 (1.00–14.10)||3.22 (0.39–11.61)||1.61 (0.05–8.95)|
|19 to < 20||52.55||0||0||0||0.00 (0.00–7.02)||0.00 (0.00–7.02)||0.00 (0.00–7.02)|
The Kaplan-Meier event-free survival curve comparing total TE with ATE and VTE is shown in Figure 1. We can appreciate better from this figure that VTE occurs less frequently than ATE. We can also see that while VTE appears to occur early and to plateau after 8 years, ATE continues to occur throughout the period of observation.
This is the first study to date that shows the yearly incidence rate of thrombosis after diagnosis of SLE and builds survival models covering up to 20 years of disease. Overall, a high incidence rate of TE in patients with SLE was observed. ATE occurred in 11%, and VTE in 5% of patients, and the estimated 20-year risks were 10% for VTE, 26% for ATE, and 33% for TE (Figure 1). These results duplicate the findings in other studies (1, 4, 8). Somers et al reported a projected cumulative VTE incidence rate of 9% over a 20-year period, derived from the prospective observation of 678 patients with SLE from the John Hopkins Lupus Cohort (1). Similarly, Brouwer et al reported incidence rates of 10% for VTE, 11% for ATE, and 1% for combined VTE and ATE in a retrospective analysis of 144 patients with SLE with a median followup of 12.7 years (8). In Brouwer et al's study, 50% of the cases of VTE occurred within the first 2.5 years and 50% of the cases of ATE within the first 8.5 years after diagnosis. These results were similar to our findings of 50% of the cases of VTE occurring in the first 2 years, and 50% of the cases of ATE in the first 6 years after diagnosis (Table 2).
Cervera et al reported a 9.2% cumulative incidence rate of TE in 1,000 European patients with SLE over a period of 10 years (4). The probability of remaining event-free at 15 years was 79% and compares similarly with our finding of 74% event-free survival at 15 years. These 3 studies had different methodology from our study in that the analyses were not restricted to an inception cohort. Inception cohorts such as the one in our current study allow for the better analysis of the occurrence of TE in a more homogeneous population from the time of first diagnosis of SLE. Our finding of a higher incidence rate of TE in the first year following diagnosis is therefore of even greater interest.
The incidence of both ATE and VTE increased in the first year. Possible reasons for this early higher incidence of thrombosis could be the untreated high levels of disease activity and circulating immune complexes, cytotoxic antibodies, or a higher inflammatory state (9). Our finding that a higher SLEDAI-2K score at the patient's initial presentation is associated with thrombosis supports the possible contributing factor of active disease in the thrombogenic process. Vascular inflammation as a response to injury and lipid peroxidation plays a role in all stages of atherothrombosis (10).
Petri described the risk factors for coronary artery disease in patients with SLE as age, sex, pregnancy or puerperium, high blood pressure, hyperlipidemia, diabetes mellitus, nephrotic syndrome, smoking, and the use of corticosteroids (11). Among the established risk factors for atherosclerosis, only dyslipoproteinemia has been demonstrated in patients with SLE (12). Elevated levels of lipoprotein(a) have also been reported in patients with SLE (13). Additional studies on the pathophysiology of thrombosis may explain the increased risk of thrombosis in SLE.
Specific antibodies such as anticardiolipin antibodies, lupus anticoagulant, or antiendothelial antibodies may also be more pathogenic early in the disease process. Antiphospholipid antibodies (aPL) are detected in one-third of patients with SLE (14). A positive association between aPL and thrombosis has been documented in multiple, predominantly retrospective studies (15).
Thrombosis is common in patients with SLE and was seen in 16% of our inception cohort. Higher rates of all thrombotic events were observed in the first year after diagnosis, which may be due to disease activity, inflammation, and endothelial injury. Because thrombosis occurs frequently in early SLE and is associated with a higher mortality, it is important to identify the modifiable risk factors that contribute to the development of thrombosis and to determine means of prevention or treatment.
- 4European Working Party on Systemic Lupus Erythematosus. Morbidity and mortality in systemic lupus erythematosus during a 10-year period: a comparison of early and late manifestations in a cohort of 1,000 patients. Medicine (Baltimore) 2003; 82: 299–308., , , , , , et al, and the
- 10Inflammation as a cardiovascular risk factor. Circulation 2004; 109 Suppl 1: II2–10., .