Cardiovascular Events in Lupus Nephritis: A Systematic Review and Meta-Analysis Open Access

Systemic lupus erythematosus (SLE) is characterized by inflammation in various body organs, especially kidneys where it leads to the occurrence of lupus nephritis (LN) [1]. Kidney involvement in SLE may precede other organs, with an estimated prevalence of 40–50%, leading to progressive kidney failure and end-stage kidney disease (ESKD) [1‒5]. Although 5-year survival rates of patients with LN were approximately 0% in the 1950s, several therapeutic regimens in recent years have significantly improved the survival rate [1]. In 1995–1999, the mortality rate in LN was 11.1 per 100 patient-years. However, this rate decreased to 6.7 per 100 patient-years in the period between 2010 and 2014 [6]. However, significant morbidity and accompanying financial burden still persist. Indeed, the 1-year mean medical costs in patients with LN surpassed those without LN by USD 46,862 [7].

SLE mostly affects women of childbearing age, resulting in premature cardiovascular morbidity [8, 9]. Patients with SLE are susceptible to different cardiovascular diseases and previous studies have indicated that LN is associated with 2.8–8.5 times increased cardiovascular event (CVE) among the sufferers [10, 11]. Despite a reduction in cardiovascular death by 44% from 1995 to 1999 to 2010–2014 timeframes, the exact pathophysiological mechanisms for CVE development in patients with LN are currently unknown. More importantly, the true prevalence of CVE in these patients is yet to be elucidated, highlighting the need for a comprehensive study [6]. In this systematic review and meta-analysis, we aimed to evaluate the occurrence of overall CVE and myocardial infarction (MI), cerebrovascular accident (CVA), and cardiovascular or cerebrovascular deaths in patients with LN.

This systematic review and meta-analysis was registered in the International Prospective Register of Systematic Reviews (PROSPERO) with the identification number CRD42023450797. Throughout the study, no protocol deviation occurred, and we adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline (online suppl. Table S1; for all online suppl. material, see https://doi.org/10.1159/000546177) [12].

We conducted a comprehensive systematic review of peer-reviewed studies that investigated the CVE in patients with LN. We considered studies that reported the occurrence of CVE during follow-up as incidence, while those that reported the rate of CVE at baseline were defined as prevalence. Our research protocol included studies with various prospective or retrospective observational designs, such as cross-sectional, case-control, cohort studies, and randomized clinical trials (RCTs). In terms of exclusion criteria, we excluded meeting abstracts, animal studies, editorials, case reports, case series, non-English studies, and studies that did not report the desired outcomes.

We explored three electronic medical databases, including PubMed, Scopus, and Web of Science up to January 2025. In PubMed and Web of Science databases, titles and abstracts were evaluated. In the Scopus database, we searched titles, abstracts, and keywords. In our search strategy, we utilized both medical subject headings (MeSH) and non-MeSH terms to collect all relevant records through the search strategy protocol, provided in online supplementary Table S2.

Two authors conducted an independent screening of titles and abstracts in the databases and collected the full texts of relevant articles. Duplicate articles were eliminated, and any disagreements were resolved through consensus.

For each included record, the following data were extracted: the first author’s name along with the publication year, study design, sample size, frequency of female participants, age, duration of follow-up (if applicable), and occurrence of CVE, including MI, heart failure (HF), ischemic heart disease, angina pectoris, coronary artery disease, and cerebrovascular disease (stroke and/or transient ischemic attack).

To assess the quality and risk of bias in each included article, specific assessment tools were employed based on the study designs. For instance, cross-sectional studies were evaluated using the critical appraisal tool (AXIS), case-control studies with the National Institute of Health (NIH) quality assessment tool, cohort studies with the Joanna Briggs Institute (JBI) critical appraisal checklist for cohort studies, and RCTs with the JBI critical appraisal checklists for RCTs [13‒16].

Random-effects model was used to investigate pooled prevalence as well as odds ratio (OR) with 95% confidence interval (CI). To convert median and interquartile range (IQR) or median and range to mean ± SD for continuous variables, the methods by Wan et al. [17] and Hozo et al. [18] were utilized, respectively. Heterogeneity was assessed using Cochran’s Q statistic, I², and tau squared (τ²). Forest plots were used to illustrate the prevalence and incidence rate of CVE according to studies reporting this event in subjects with LN and those with different stages of kidney diseases. We also provided forest plots for different CVE including MI, CVA, and cardiovascular and/or cerebrovascular death as well as the risk of CVE in patients with LN compared to individuals with SLE. Finally, the rate of CVE was compared according to the geographical locations of the studies. In order to address the potential heterogeneity, we performed meta-regression analysis. Publication bias was evaluated using funnel plots, Egger’s and Begg’s tests, as well as the Duval and Tweedie’s trim-and-fill method. In order to assess the robustness of the findings, sensitivity analysis using leave-one-out method was performed. Data entry was carried out in an Excel datasheet, and comprehensive meta-analysis software (version 2.0, Biostat Inc., Englewood NJ) was used for all analyses with consideration of p values <0.05 as statistically significant.

After including 1,366 studies in predefined databases and exclusion of 540 duplicates and other irrelevant papers, a total of 25 articles were found to be eligible for inclusion. Five studies used similar database and we included the one with the longest study duration [6, 19‒22]. Finally, we included 21 articles for further downstream analysis. Figure 1 displays the flow diagram of the current review. Of the included studies, four articles used a cross-sectional design [23‒26] and one study was performed with a case-control design [27]. The other studies were cohort studies [6, 11, 28‒41]. Study population ranged from 40 to 20,974. Articles included were published from 1999 to 2024 and reported CVE outcomes in different countries, including USA, Canada, China, Sri Lanka, Denmark, Italy, London, Spain, Sweden, Netherlands, Hungary and Morocco. The risk of bias assessment is provided in online supplementary material (online suppl. Tables S3–S5). A summary of all included studies is shown in Table 1.

Twenty-one studies, with a total of 29,489 participants, reported different CVE among patients with LN, with varied definitions. Female patients comprised 81.14% of the total population (reported in 19 studies) with the age of 39.5 ± 15.5 years. In three studies, MI and stroke were defined as CVE [36, 37]. In the study by Garg et al. [31], ischemic heart disease (including MI, abnormal stress test and/or angiogram, coronary artery revascularization, and cardiologist-documented events), stroke and/or transient ischemic attack (TIA), and peripheral vascular disease were included in the definition of CVE [31]. Jorge and colleagues defined overall cardiovascular disease and two subtypes (coronary heart disease and stroke) as CVE [21]. In two studies, MI, HF, stroke, and atherosclerosis were considered as CVE in the study by Hurst et al. [30]. Besides consideration of angina pectoris and MI as CVE in one study, another study added the following terminologies to diagnose CVE: congestive HF, coronary artery bypass graft, stroke and/or TIA, and percutaneous transluminal coronary angioplasty [27, 35]. Finally, coronary artery disease and other related disorders (including ST or non-ST segment elevation MI) were defined as CVE in the study by Sun et al. [23]. Chrysostomou et al. [40] defined CVE as a composite of MI, stroke, or CVD-related death, while Molnár et al. [38] included nonfatal MI, CVD death, or hospitalization due to stroke, HF, or coronary revascularization. van Schaik et al. [39] classified CVE as CVA, TIA, MI, peripheral arterial disease, or any coronary intervention, including percutaneous angioplasty or coronary artery bypass graft. Garg et al. [41] defined CVE as nonfatal or fatal atherosclerotic cardiovascular disease (ASCVD) events, encompassing stroke, TIA, ischemic heart disease, and peripheral arterial disease requiring intervention. While one study considered MI and CVA as CVE [26], another included acute coronary syndrome in its definition [25]. Other studies reported death in the context of cardiovascular or cerebrovascular diseases [6, 11, 29, 32‒34].

Four studies (N = 745, women: 82.47%) reported CVE prevalence [23‒26]. The overall pooled CVE prevalence among patients with LN was 16% (95% CI: 12–21%), as shown in Figure 2a. Heterogeneity indices are presented in online supplementary Table S6. Funnel plot (online suppl. Fig. S1) as well as other predefined tests did not show any evidence regarding publication bias (Begg’s test p value: 0.367, Egger’s test p value: 0.375) However, Duval and Tweedie’s trim-and-fill method showed one missing study (observed point estimate: 0.157, 95% CI: 0.116–0.209, adjusted point estimate: 0.168, 95% CI: 0.120–0.229). To assess the robustness of our findings, we performed a sensitivity analysis, showing consistency of our results (online suppl. Fig. S2).

We also assessed the studies that solely reported CVE incidence during their follow-up times. Seventeen studies (N = 28,744, women: 81.14%) [6, 11, 27‒41] reported this outcome as an incidence, with the incidence rate of 9% (95% CI: 6–12%) (Fig. 2b). There was no publication bias (funnel plot (online suppl. Fig. S3), Begg’s test p value: 0.133, Egger’s test p value: 0.356). However, Duval and Tweedie’s trim-and-fill method showed three missing studies (observed point estimate value: 0.085, 95% CI: 0.058–0.123 and adjusted point estimate value: 0.109, 95% CI: 0.076–0.154). Heterogeneity indices and sensitivity results are presented in online supplementary Table S6 and Figure S4, respectively.

Meta-regression analysis incorporating publication year, female proportions, study design, and geographical location as covariates identified significant contributors to heterogeneity in effect sizes. Geographical location emerged as a significant source of heterogeneity (p < 0.001), with studies conducted in Europe and North America showing higher effect sizes compared to those from Asia (Europe: β: 1.64, 95% CI: 0.70–2.57, p = 0.0006; North America: β: 1.20, 95% CI: 0.30–2.10, p = 0.008). This model led to a substantial reduction in heterogeneity, as indicated by a decrease in the Q statistic from 391.49 to 68.50, demonstrating the effectiveness of the covariates in explaining between-study variance (R²: 0.58).

Fifteen studies on 4,339 patients reported CVE incidence in LN with/without report of different CKD stages, with incidence rate of 9% (95% CI: 6–14%) (Fig. 3a) (heterogeneity indices (online suppl. Table S6), Begg’s test p value: 0.030, Egger’s test p value: 0.001). Duval and Tweedie’s trim-and-fill method results were in favor of 2 missing studies (observed point estimate value: 0.092, 95% CI: 0.061–0.138 and adjusted point estimate value: 0.113, 95% CI: 0.076–0.165). Funnel plot as well as sensitivity analysis are also provided in online supplementary Figures S5 and S6, respectively. Two studies on 24,405 individuals assessed CVE in only patients with ESKD due to LN. Data synthesis revealed CVE incidence rate of 5% (95% CI: 2–13%) (Fig. 3b). Heterogeneity indices are provided in online supplementary Table S6. Publication bias assessment was not done due to inclusion of two studies.

Eight studies on 5,735 patients reported MI incidence as their outcome variable [11, 27, 30, 35‒38, 40]. Figure 4 shows the forest plot for the incidence of MI (4%, 95% CI: 2–7%). Heterogeneity was considerable among included studies (online suppl. Table S6). However, no evidence of publication bias was found (funnel plot (online suppl. Fig. S7), Begg’s test p value: 0.355, Egger’s test p value: 0.170). Moreover, we used Duval and Tweedie’s trim-and-fill method and found observed and adjusted values were similar, suggesting no missing studies (0.038, 95% CI: 0.020–0.072). To test the robustness of our findings, we performed sensitivity analysis on the included articles in this review to assess the potential impact of each study on the results. As shown in online supplementary Figure S8, the results suggested consistency of the outcomes.

Nine studies including 6,053 subjects assessed incidence of stroke (ischemic or hemorrhagic) or TIA [11, 27, 30‒32, 36‒38, 40]. We defined and pooled all the previously mentioned outcomes as CVA. The pooled occurrence rate of CVA in patients with LN was 4% (95% CI: 2–7%) (Fig. 5). We presented heterogeneity indices and funnel plot in online supplementary Table S6 and Figure S9, respectively. Begg’s (p value: 0.125) and Egger’s (p value: 0.307) tests suggested no publication bias. However, in Duval and Tweedie’s trim-and-fill method, observed and adjusted point estimate values were different, indicating the presence of two missing studies (0.037, 95% CI: 0.019–0.068 and 0.047, 95% CI: 0.026–0.084, respectively). We then performed sensitivity analysis and found the results were notably robust and reliable (online suppl. Fig. S10).

Ten studies (N = 26,511) reported death in context of either cardiovascular or cerebrovascular events [11, 21, 28, 32‒34, 36‒38, 40]. Figure 6 presents the forest plot of mortality rates according to the studies that reported cardiovascular and/or cerebrovascular death in patients with LN. The pooled incidence rate was 5% (95% CI: 3–7%). Heterogeneity among enrolled studies was considerable (online suppl. Table S6). Begg’s test (p value: 0.429), Egger’s test (p value: 0.116), and funnel plot (online suppl. Fig. S11) did not show any publication bias. However, Duval and Tweedie’s trim-and-fill method was indicative of two missing studies (observed point estimate value: 0.045, 95% CI: 0.027–0.073 and adjusted point estimate value: 0.052, 95% CI: 0.033–0.081). We also provided the results of sensitivity analysis, suggesting no impact of any studies on overall outcomes and reliable and consistent findings (online suppl. Fig. S12).

Five studies examined CVE in patients with and without LN [11, 23, 24, 29, 30]. After excluding one study that focused on patients with LN versus healthy controls [11], the remaining studies assessed CVE between SLE patients with and without LN. The pooled data indicated that SLE patients with LN had 1.18 times higher risks of experiencing CVE (OR: 1.18, 95% CI: 1.03–1.34, p = 0.014) (Fig. 7a). In terms of MI, although patients with LN exhibited an increased risk compared to those without, the result was not statistically significant (OR: 1.15, 95% CI: 0.83–1.59, p = 0.389) (Fig. 7b). Due to the limited number of studies on other outcomes, further data analysis could not be conducted.

Due to presence of geographical location as the main sources of heterogeneity, we performed subgroup analysis among 17 studies reported CVE incidence in patients with LN. Six studies were performed in North America (including USA and Canada) [6, 27, 30, 31, 36, 41]. In European and Asian countries, seven [11, 29, 35, 37‒40] and three [28, 32, 34] studies were done, respectively. Figure 8 illustrates the CVE incidence according to continent distribution. Although the incidence of CVE was quite similar in North America (10.1%, 95% CI: 5.7–17.2%) and Europe (13.3%, 95% CI: 7.6–22.4%), they were higher when compared to Asian countries (2.3%, 95% CI: 1.6–3.3%).

The primary objective of this systematic review and meta-analysis was to evaluate CVE in patients with LN. Results revealed that 9% (95% CI: 6–12%) of patients with LN experienced some forms of cardiovascular or cerebrovascular events. LN significantly increases the risk of CVE compared to the SLE patients (OR: 1.18, 95% CI: 1.03–1.34, p = 0.014). Our findings were comparable with a recent meta-analysis indicated that patients with LN have a 1.75 times higher risk of CVD compared to individuals with systemic SLE without kidney involvement [42]. These findings underscore the need for appropriate management strategies and considering kidney involvement in patients with LN as a risk factor for CVE. In terms of prevalence, a recently published study reported MI and CVA prevalence of 8% (95% CI: 4–14%) and 10% (95% CI: 5–17%), respectively. However, the authors included studies from 1947 to 2022, resulted in a total of five included studies [43]. While the exact association between CVE in patients with LN is yet to be fully understood, several underlying risk factors have been suggested, including proteinuria, long-standing corticosteroid usage, dyslipidemia, endothelial dysfunction, hypertension, chronic inflammation, and the development of ESKD [44, 45].

We found that patients with LN have a higher risk of CVEs compared to SLE patients without kidney impairment. The involvement of renal vasculature in LN has been associated with an increased risk of CVE, with renal arteriosclerosis emerging as a significant factor. Garg et al. [41] conducted a study involving 209 patients with biopsy-proven LN to evaluate the impact of renal arteriosclerosis on long-term ASCVD, defined as fatal or nonfatal ischemic heart disease, including MI, cardiologist-documented ischemic events, coronary artery revascularization, stroke, peripheral arterial disease requiring angioplasty or bypass, or peripheral arterial disease, leading to critical limb ischemia. After a median follow-up of 11 years, the study found that renal arteriosclerosis involving >25% of the renal vasculature was associated with a threefold increase in 10-year ASCVD risk (95% CI: 1.3–7.0, p = 0.014).

We found that patients with LN living in Asian countries had a lower occurrence of CVEs compared to those residing in North America or Europe. One possible explanation could be the influence of race and ethnicity. Gómez-Puerta et al. [20] conducted a study involving 12,533 patients with ESKD secondary to LN (mean age: 40.7 ± 14.9 years) to investigate the impact of racial and ethnic differences on all-cause mortality and CVE. They found that African Americans had a higher risk of mortality (hazard ratio [HR]: 1.27, 95% CI: 1.18–1.36) compared to whites. A similar trend was observed among patients with LN under the age of 40, with African Americans having a 1.67 times higher mortality rate (95% CI: 1.44–1.93) compared to whites of the same age group. On the other hand, Asian and Hispanic patients had a lower hazard of all-cause mortality compared to whites (HR: 0.70, 95% CI: 0.58–0.84 and HR: 0.79, 95% CI: 0.71–0.88, respectively). Regarding CVE, there was no significant association observed in Asian and Native American patients compared to whites individuals in terms of HF, MI, ischemic stroke, and hemorrhagic stroke. However, African American patients had a higher risk of hospitalization due to HF and hemorrhagic stroke in the fully adjusted model (HR: 1.38, 95% CI: 1.26–1.52 and HR: 1.66, 95% CI: 1.19–2.31, respectively) [20]. Social determinants of health (SDoH) represent another potential mechanism that may contribute to differences in CVE occurrence across populations. Garg et al. [46] conducted a systematic review and meta-analysis of 13 studies to assess the impact of adverse SDoH on poor outcomes – defined as ESKD, CVD, or death – among patients with LN. Adverse SDoH were defined as social barriers affecting health outcomes, including poverty, smoking, homelessness, depression, limited access to healthcare facilities, food insecurity, lack of transportation or social support, discrimination, lack of insurance, and unemployment. The analysis revealed that patients with LN experiencing adverse SDoH had a 1.47 times higher risk of poor outcomes (95% CI: 1.11–1.93, p = 0.006). This relationship remained significant when examining both individual-related SDoH (OR: 1.64, 95% CI: 1.13–2.39, p = 0.009) and healthcare-related SDoH (OR: 1.77, 95% CI: 1.39–2.25, p < 0.0001).

Suboptimal care for ESKD has also been identified as a contributing factor to higher odds of CVE. A study examining the quality of ESKD care in patients with LN-ESKD found significant disparities. Black patients had lower odds of receiving pre-ESKD care compared to white patients (OR: 0.73, 95% CI: 0.63–0.85) and were less likely to be on the kidney transplantation waitlist (OR: 0.78, 95% CI: 0.68–0.91).

Insurance status also played a critical role, with uninsured patients or those covered by Medicaid having significantly lower odds of being placed on the waitlist compared to patients with private insurance (OR: 0.36, 95% CI: 0.29–0.44 and OR: 0.51, 95% CI: 0.44–0.58, respectively). Furthermore, LN-ESKD patients residing in the Northwest and Northeast regions of the USA were reported to receive higher levels of care [47]. However, data on CVE differences among diverse LN populations remain limited, and further investigation is warranted to better understand the underlying mechanisms driving these disparities.

Another major culprit is endothelial dysfunction and consequent atherosclerosis. One of the main factors in this pathogenesis is inflammatory cytokines. Type I interferons have been suggested to interfere with the vascular healing process and suppress neovascularization and reduce endothelial progenitor cells [48]. Moreover, they can raise the pace of foam cell production and platelet activation, leading to impairment in smooth muscle cell maturation, consequently resulting in plaque rupture and CVE [49, 50]. Other inflammatory cytokines include tumor necrosis factor-α, interleukin 6 and 17, which may augment atherosclerosis in patients with SLE by stimulating the differentiation of monocytes to macrophages and foam cells [51, 52]. For example, Asanuma et al. [52] reported that IL-6 and monocyte chemoattractant protein-1 were present at significantly higher levels in SLE patients compared to controls. Additionally, IL-6 was identified as an independent risk factor for coronary calcification (OR: 1.07, p = 0.035).

The primary cells responsible for vascular repair following damage are myelomonocytic circulating angiogenic cells (CACs) and endothelial progenitor cells (EPCs). Denny et al. [48] demonstrated that EPC levels are significantly lower in SLE patients compared to controls, potentially impairing their maturation into endothelial cells. This impairment may result in reduced secretion of pro-angiogenic factors such as vascular endothelial growth factor and hepatocyte growth factor. Inflammation-induced apoptosis of EPCs and CACs, driven by type I interferons, disrupts the balance between endothelial cell damage and repair.

Another potential theory is linked to the presence of elevated levels of oxidized low-density lipoprotein cholesterol (LDL-C). In individuals with SLE, antibodies against high-density lipoprotein cholesterol (HDL-C) or apolipoprotein A1 can hinder the normal antioxidant activity of HDL-C [53]. Additionally, there have been reports of increased occurrence of pro-inflammatory HDL-C in patients with SLE. This particular HDL-C is unable to prevent the oxidation of LDL-C, resulting in a higher quantity of oxidized LDL-C and an elevated risk of CVE [54].

Endothelial cell malfunction in the context of SLE and LN leads to increased CVE risk [44]. Some abundant endothelial cell dysfunction biomarkers in patients with LN include platelet endothelial cell adhesion molecules, vascular endothelial growth factor receptor-2, vascular endothelial cadherin, and Annexin V [55‒57]. For instance, the latter has been indicated as an independent marker of atherosclerosis in patients with SLE [58]. Activated leukocyte cell adhesion molecule (ALCAM) has emerged as a novel biomarker for assessing kidney involvement in SLE patients. ALCAM is a glycoprotein expressed on the surface of endothelial cells. Studies in animal models have demonstrated its upregulation in the renal tissues of mice with lupus-like conditions [59]. Human studies also support its potential as a urinary biomarker. Din et al. [60] measured urinary ALCAM levels in 228 patients with SLE and LN and 28 healthy controls, reporting significantly higher levels in patients with active LN compared to those with inactive LN, active or inactive SLE without kidney involvement (p < 0.001), and healthy controls (p < 0.001). Furthermore, ALCAM demonstrated a good ability to differentiate between active and inactive LN, with an area under the curve of 0.83 (95% CI: 0.73–0.93, p < 0.001). Although further research is necessary, ALCAM shows promise as a clinically useful biomarker for assessing kidney involvement and monitoring therapeutic responses in SLE patients.

Another major factor in normal endothelial cell activity is endothelial nitric oxide synthase. SLE results in the reduction of this molecule, leading to endothelial dysfunction and atherosclerosis [44].

Another critical factor to consider when assessing CVE in patients with SLE and LN is the presence of antiphospholipid antibodies (aPLs), which include lupus anticoagulant, anticardiolipin antibodies, and anti-β2-glycoprotein I antibodies [61]. Although anti-β2-glycoprotein I antibody is classified as one of the most common antibodies in patients with antiphospholipid syndrome, it has been reported that 30–40% of SLE patients would become positive for this antibody which resulted in enhancing foam cell formation through complex of oxidized LDL-C, β2-glycoprotein I, and anti-β2-glycoprotein I antibody. Simultaneous presentation of SLE as a thrombotic risk factor and anti-β2-glycoprotein I antibody result in endothelial dysfunction and increase prothrombotic status [62‒64].

Huang et al. [65] conducted a study involving 1,573 SLE patients (mean age: 34.9 ± 11.0 years; females: 96.1%) and found that 33.4% tested positive for aPLs. Over a mean follow-up period of 4.51 ± 2.32 years, 116 (7.37%) of the participants experienced ASCVD, defined as stroke, MI, CVD death, or peripheral or coronary artery revascularization. Multivariate Cox regression analysis identified LA, aCL IgG, and aCL IgM as independent predictors of ASCVD in SLE patients, with HRs of 5.13 (95% CI: 3.23–8.20, p < 0.001), 1.95 (95% CI: 1.25–3.00, p = 0.003), and 1.83 (95% CI: 1.03–3.20, p = 0.039), respectively. These findings underscore the importance of regular monitoring and management of aPLs in patients with SLE and LN to mitigate CVE risk effectively.

In addition to aforementioned factors, traditional cardiovascular risk factors have been reported to be higher in patients with SLE and may partly explain the higher likelihood of CVE [66]. Bruce et al. reported that hypertension and diabetes mellitus were frequently observed in women with SLE [67]. Among traditional cardiovascular risk factors, a systematic review reported smoking, dyslipidemia, and hypertension as CVE predictors among patients with SLE [66]. Other than SLE, renal involvement is also associated with higher CVE [22]. Consequently, patients suffering from renal involvement, especially ESKD, in context of SLE might be more susceptible to CVE. Thus, implementation of more aggressive cardiovascular therapy is advisable for these patients. However, further studies are still required regarding patients with LN and preserved renal function.

Moreover, some of the SLE therapies may play a role in increasing the rates of CVE in this population. One of the main drugs administered is corticosteroids. Increased reactive oxygen species production in the context of corticosteroids therapy causes downregulation of endothelial nitric oxide synthase mRNA levels [68]. Another possibility is related to HDL-C. Yin and colleagues performed a prospective cohort study to investigate the impact of HDL-C on CVD mortality in 775 patients with LN. CVD mortality was defined as any death in the context of MI, congestive HF, cardiac arrhythmia or arrest, cardiomyopathy, atherosclerotic cardiac disorder, peripheral vascular disease, CVA, and anoxic or ischemic brain damage. They followed enrolled participants for a median of 56 months and found that individuals with lower HDL-C levels were more vulnerable to CVD death in comparison to the intermediate group (HR: 4.77, 95% CI: 1.35–16.94, p = 0.016). They suggested that low HDL-C might be indicative of severe inflammation and higher oxidative stress that is usually observed in patients with SLE [28].

Regarding current study strengths, we used different keywords with no time limitation to gather all relevant studies in well-known databases. Another strength of our study is the absence of publication bias in most outcomes. However, there are still some limitations that should be considered. Despite excluding articles not written in English, mainly due to infeasibility of risk of bias assessment [69], we included studies across diverse geographical regions to identify notable differences in CVE incidence across these regions. Additionally, conducting complementary analyses based on patients’ risk factors or administered therapy was not feasible due to inconsistencies in reporting across studies.

In conclusion, this study highlights the ongoing significant occurrence of CVE in individuals with LN, despite a decreasing trend in CVD rates, and increased CVR risk in patients with LN. These findings underscore the need for optimized healthcare management plans to implement appropriate preventive and therapeutic strategies, aiming to reduce the burden of LN-related CVE.

A statement of ethics is not applicable because this study is based exclusively on published literature.

None of the authors had any personal or financial conflicts of interest.

No funding has been attributed to this study.

Study concept and design: M.V., M.G., G.H., A.A., and S.N. Acquisition of data: M.V., M.G., and N.E. Analysis and interpretation of data: M.V., R.M., A.A., N.D., and S.N. Drafting of the manuscript: M.V., M.G., N.E., and S.N. Critical revision of the manuscript for valuable intellectual content: M.V., G.H., R.M., A.A., N.D., S.H., M.H., S.N., and A.K.S. Statistical analysis: M.V. and N.D. Administrative, technical, and material support and supervision: A.A. and S.N.

The datasets generated during and/or analyzed during the current study are not publicly available due to confidentiality issues but are available from the corresponding author on reasonable request.