Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Bone mineral density and vertebral fractures in patients with systemic lupus erythematosus: A systematic review and meta-regression

  • Claudia Mendoza-Pinto,

    Roles Conceptualization, Methodology, Visualization, Writing – original draft

    Affiliations Systemic Autoimmune Diseases Research Unit, Hospital de Especialidades, UMAE CMNMAC—CIBIOR, Instituto Mexicano del Seguro Social, Puebla, Puebla, México, Department of Immunology and Rheumatology, Medicine School, Benemérita Universidad Autónoma de Puebla, Puebla, Puebla, México

  • Adriana Rojas-Villarraga,

    Roles Conceptualization, Methodology, Validation, Writing – review & editing

    Affiliation Artmedica IPS, Bogotá, Colombia

  • Nicolás Molano-González,

    Roles Data curation, Formal analysis, Software

    Affiliation Center for Autoimmune Diseases Research (CREA), School of Medicine and Health Sciences, Universidad del Rosario, Bogotá, Colombia

  • Erick A. Jiménez-Herrera,

    Roles Conceptualization, Methodology, Writing – original draft

    Affiliation Systemic Autoimmune Diseases Research Unit, Hospital de Especialidades, UMAE CMNMAC—CIBIOR, Instituto Mexicano del Seguro Social, Puebla, Puebla, México

  • María de la Luz León-Vázquez,

    Roles Conceptualization, Data curation

    Affiliation Systemic Autoimmune Diseases Research Unit, Hospital de Especialidades, UMAE CMNMAC—CIBIOR, Instituto Mexicano del Seguro Social, Puebla, Puebla, México

  • Álvaro Montiel-Jarquín,

    Roles Conceptualization, Supervision

    Affiliation Research in Health Unit, UMAE, Instituto Mexicano del Seguro Social, México, Puebla, Puebla, México

  • Mario García-Carrasco ,

    Roles Conceptualization, Data curation, Resources

    mgc30591@yahoo.com

    Affiliations Systemic Autoimmune Diseases Research Unit, Hospital de Especialidades, UMAE CMNMAC—CIBIOR, Instituto Mexicano del Seguro Social, Puebla, Puebla, México, Department of Immunology and Rheumatology, Medicine School, Benemérita Universidad Autónoma de Puebla, Puebla, Puebla, México

  • Ricard Cervera

    Roles Conceptualization, Supervision, Validation, Writing – review & editing

    Affiliation Department of Autoimmune Diseases, Hospital Clinic, Barcelona, Catalonia, Spain

Abstract

Background

Observational studies have indicated a high but heterogeneous prevalence of low bone mineral density (BMD) and vertebral fractures (VF) in patients with systemic lupus erythematosus (SLE). Therefore, the objectives of this systematic review and meta-regression were: 1) to compare BMD between SLE patients and healthy controls and 2) to evaluate the relationship between BMD and glucocorticoid therapy and VF in SLE patients.

Methods and findings

Articles were identified from electronic databases (PubMed, Embase, VHL, SciELO and the Cochrane Library). Prospective longitudinal and cross-sectional studies were considered for review. We evaluated the quality of the evidence included using the Oxford Centre for evidence-based medicine (EBM) Levels of Evidence. In total, 38 articles were identified and analyzed (3442 SLE cases and 6198 controls) in the analysis of BMD (9232 women and 408 men). There were significant differences in mean BMD between SLE patients and controls. BMD mean difference in cases/controls: -0.0566 95% CI (-0.071, -0.0439; p = < 0.0001). When only SLE patients were analyzed, the BMD did not significantly differ between patients who had or had not received glucocorticoid (GCT) therapy. 694 SLE patients were included in the analysis of VF (189 with VF vs. 505 without VF). Patients with VF had lower BMD than patients without VF (BMD mean difference without VF/with VF: 0.033 (95%CI: 0.006–0.060); p-value: 0.0156).

Conclusions

Patients with SLE had lower BMD than healthy controls. Moreover, SLE patients with VF had lower BMD than patients without VF. However, our data did not show that GCT therapy had an impact on BMD.

Background

Patients with systemic lupus erythematosus (SLE) may have an increased risk of bone mineral density (BMD) loss and vertebral fractures (VF) according to cross-sectional studies [14]. Osteoporosis and fractures contribute to damage in the musculoskeletal system, which is frequently involved in patients with SLE [5].

BMD measurement by dual energy X-ray absorptiometry (DXA) is the gold standard to assess fracture risk in healthy men and women [6]. The utility of DXA in predicting fracture risk in SLE is unclear for two reasons. First, only small cross-sectional studies have reported on the use of DXA to discriminate the fracture status. Secondly, the relationship between low BMD and glucocorticoid therapy (GCT), which is extensively used for the treatment of SLE disease flares and complications, remains unclear [2,4,7,8].

Identifying prevalent VF is important, since prevalent vertebral deformities are associated with a reduced quality of life [9], and increased mortality and risk of future fractures in the general population [10]. There are only a few studies on prevalent VF (using a standardized method of scoring vertebral deformities) and these showed at least one VF in 20–26.1% of SLE patients [2,11,12]. We recently showed that 20% of 110 SLE patients [median follow-up 8 (IQR 8–9) years] had radiographic VF at baseline and 32% had a new VF. The reported annual incidence rate of new morphometric VF is 3.5 (95% CI 2.4–4.91) per 100 patient/years [13]. A recent meta-analysis, including studies of VF prevalence and low BMD, reported an almost three-fold higher risk of VF in SLE patients (RR 2.97, 95% CI 1.71–5.16, P < 0.001) compared with healthy controls [14]. Another recent systematic review without a meta-analysis also concluded that SLE patients are at risk of both reduced vitamin D plasma levels and low BMD [15].

Of patients with VF, 29–35.8% had normal BMD [11,12], in line with results from studies in the general population reporting that the proportion of fractures attributable to osteoporosis is only 10–44%. This points to the limited value of BMD measurement in the assessment of future fracture risk. Moreover, these discrepancies between BMD and fracture risk may be because, in SLE, poor bone quality rather than decreased bone density plays the most important role in determining the risk of fractures, and VF occur at much high rates than expected on the basis of BMD, suggesting that the bone fragility of GCT users is not defined by the BMD. Indeed, cutoff values of BMD at the lumbar spine and femoral neck in women with VF treated with GCT were higher than those of controls [16].

Only one systematic meta-analysis has reported that SLE patients have significantly lower BMD levels than controls, and SLE is also significantly associated with increased fracture risk [14]. However, the relationship between BMD and GCT use and VF (using a standardized method of scoring vertebral deformities) has not been assessed in a meta-analysis or meta-regression. Therefore, we conducted a meta-analysis and meta-regression of studies evaluating: 1) BMD in SLE patients and controls, 2) BMD in SLE patients receiving GCT or not, and 3) BMD in SLE patients with or without VF.

Materials and methods

Search strategy

A systematic literature review was conducted using the following electronic databases: PubMed (1946-Week 2, January 2018), Cochrane library (1985-Week 2, Week 2, January 2018), EMBASE (1974-Week 2, January 2018), Virtual Health Library (VHL) (1998-Week 2, January 2018), and SciELO (1997-Week 2, January 2018), for published studies. We followed the PRISMA guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analysis, see S1 File) for meta-analysis of observational studies [17] in the data extraction, analysis, and reporting.

The following Medical Subject Heading (MeSH) terms were used: "Lupus Erythematosus, Systemic," "Osteoporosis," "Bone Density," "Densitometry” and "Spinal Fractures". Furthermore, we used ‘text words’ if there was no MeSH term, such as the cases of “SLE” abbreviation, “bone loss”, “BMD” abbreviation, “dual energy X ray absorptiometry”, “vertebral fractures” and “vertebral fragility”. DeCS terms (Health Sciences Descriptors) were also used to find records and sources of information through controlled concepts in order to search the SciElo and VHL databases. Studies were limited to those carried out in adult humans and published in English. References from the articles deemed relevant were hand-searched.

Study selection and eligibility criteria

Prospective longitudinal and cross-sectional studies in SLE patients (regardless of menopausal status) were considered for review. Studies that used accepted and validated classification criteria for SLE were included. Case reports, conference abstracts, letters to editors, and studies not reporting the prevalence of osteoporosis or low BMD or VF prevalence were excluded. Papers were included if BMD was evaluated by dual-energy X-ray absorptiometry (DXA) at any of the total hip, femoral neck and lumbar spine. We included studies where VF was identified by standardized vertebral morphometry. In addition, the reference lists of relevant reviews and articles were manually retrieved to find other possible studies.

When various reports from the same study were published, only the most recent or informative one was included. However, if more than one publication described a single study but each presented new and complementary data, both were included and analyzed in separate analyses.

We first reviewed the titles and abstracts of studies found in the literature search and decided whether they conformed to the following research topics: 1) mean difference in BMD levels between SLE patients and controls, 2) mean difference in BMD levels between women and men with SLE, 3) mean BMD differences in patients with and without GCT, and 4) mean BMD differences in SLE patients with and without VF. Articles selected were evaluated by two investigators independently using the same selection criteria. The two resulting databases were compared and disagreements resolved by consensus.

Data extraction

The following data were extracted from studies included: first author, publication year, study design, population studied, number of participants, number of cases and controls, mean age, ethnicity, use of GCT, cumulative GCT dose, BMD measurement sites and VF assessment method. Definitions of GCT therapy were taken from the articles included (S2 Table).

We were able to extract several effect sizes from a single study. For example, some studies reported mean BMD for cases and controls, disaggregated by region and gender. The nested nature of the data is taken into account in the multilevel linear (mixed-effects) model (see statistical methods).

Assessment of methodological quality

The Oxford Centre for Evidence-Based Medicine (EBM): 2011 Levels of Evidence criteria were used to assess the strength of the evidence for all studied included studies [18]. Clinical evidence was divided into 5 levels ranging from I to V as follows (diagnosis question): Level 1 Systematic review of cross-sectional studies with consistently applied reference standard and blinding; Level 2 Individual cross-sectional studies with consistently applied reference standard and blinding; Level 3 Non-consecutive studies, or studies without consistently applied reference standards; Level 4 Case-control studies, or “poor or non-independent reference standard; Level 5 Mechanism-based reasoning. Any disagreement was resolved through discussion and consensus of investigators.

Statistical methods

Due to the diversity of studies found, three meta-regression analyses were made: The first aimed to assess possible differences in BMD between SLE cases and controls. The second analyzed the effect of treatment (GCT) on BMD in SLE cases and the third evaluated differences in BMD between SLE cases with and without VF.

We were able to extract more than one effect measure from a single study; therefore, we fit a multilevel linear (mixed-effects) model [19] to the three scenarios. The first hierarchical level corresponds to a single effect measure, and in the second level several effect measures are reported.

The following strategy was employed in order to obtain a final, most parsimonious model in each case: first, the variance component structure was obtained by fitting several models with a saturated fixed-effects structure (all relevant covariates plus full interactions) and different variance models (fixed effects, mixed effects and multilevel mixed-effects). The most parsimonious variance structure was chosen according to the basis likelihood ratio test and AIC criteria. Once the variance structure was determined, the mean structure was assessed, subtracting interactions and covariates and, again, the most parsimonious mean structure was chosen on the basis likelihood ratio test and AIC criteria. Heterogeneity was calculated using Higgins’s (I2) tests. The I2 test showed the proportion of observed dispersion that was real rather than spurious and was expressed as a ratio ranging from 0% to 100%. I2 values of 25%, 50%, and 75% were qualitatively classified as low, moderate, and high respectively.

Using the selected models, the rank correlation test for funnel plot asymmetry was assessed to check for publication bias. The analysis was performed with the R 3.3.2 metaphor package [20].

Results

Identification of relevant studies

Fig 1 shows a flow diagram of how relevant studies were identified. A total of 12875 articles were identified through the PubMed database search and an additional 3050 articles were retrieved through other sources (additional databases and hand search of relevant bibliographies). After duplicates were removed there were 2185 potentially-relevant articles. In all, 2103 studies were excluded during the initial screening through review of titles and abstracts in addition to 14 foreign language studies. The full texts of the remaining 68 studies were thoroughly reviewed. Of these, 26 were excluded due to: randomized clinical trials (n = 7), incomplete data (n = 3), T-score report but not BMD (n = 4), overlapping reports from the same research group (n = 16). The remaining 38 studies were included in the final analysis. See supporting information file showing detailed search criteria for PubMed-MeSH database (S1 Table).

Characteristics of studies included in the final analysis

The 38 studies included were published between 1990 and 2017: 7 came from China [11,2126], 3 from Brazil [2729], 3 from Spain [3033], 2 from the United States [7,34], 2 from Germany [35,36], 2 from Hungary [37,38], 2 from Mexico [8,12], 2 from Italy [1], 2 from Sweden [4,39] and 1 each from the United Kingdom [40], South Africa [41], Australia [3], Denmark [42], Italy [1,43], Belgium [44], Norway [45], Austria [46], Bulgaria [47], Thailand [48], Singapore [49], The Netherlands [50] and Japan [51]. In total 3442 SLE cases and 6198 controls were included in the analysis of BMD (9232 women and 408 men). Eleven studies were suitable for BMD comparison in SLE patients receiving GCT (n = 358) or not (n = 354).

The study characteristics are described in Table 1 including the Oxford Centre for EBM 2011 Levels of Evidence. The most frequent level of evidence was level IV in 25 studies, followed by 8 and 3 studies corresponding to level II and III of evidence, respectively. The characteristics of studies evaluating VF are shown in Table 2. Only 5 articles evaluated VF using a standardized method, making it possible to compare BMD between SLE patients with and (n = 189) and without VF (n = 506).

thumbnail
Table 1. Characteristics of studies included comparing SLE cases and controls.

https://doi.org/10.1371/journal.pone.0196113.t001

thumbnail
Table 2. Characteristics of included studies comparing SLE with and without vertebral fractures.

https://doi.org/10.1371/journal.pone.0196113.t002

Results of meta-regressions

In the three groups analyzed, we found that the best variance structure was a two level model with random effects at the level of single effect measures and another random effect at the level of studies. In addition, we assessed the need for adjustment by gender and found no significant effect for this variable in any group. The only covariate that needed adjustment was the region of the measure, which appears systematically in the three final models. Table 3 shows the most relevant results from this analysis. High heterogeneity (see I2 in Table 3) between studies was observed.

thumbnail
Table 3. Results of the most parsimonious models for the groups of studies.

https://doi.org/10.1371/journal.pone.0196113.t003

Evidence of significant publication bias was identified by means of the rank correlation test for funnel plot in the case-control analysis group (p-value = 0.0446), but was not observed in the other two groups (S1 Fig.).

Comparison between SLE patients and controls

We identified 8986 participants (2899 patients with SLE vs. 6087 controls) in the analysis of BMD. There were significant differences in mean BMD between cases and controls, with controls having a higher BMD (mean difference cases/controls: -0.0566 95% CI (-0.071, -0.0439; p = < 0.0001). S2 Fig shows the forest plot corresponding to this meta-regression.

Comparison of BMD between SLE patients with (n = 348) and without (n = 169) GCT therapy showed no significant differences in BMD (S3 Fig) between treated and non-treated SLE patients (p-value: 0.1303).

Comparison of BMD between SLE patients with and without VF 694 SLE patients were included in the analysis of VF (189 with VF vs. 505 without VF) (Table 2). Patients with VF had a lower BMD than those without (mean difference without FV–with FV: 0.033; 95%CI: (0,006–0.06); p-value: 0.0156). S4 Fig shows the forest plot corresponding to this analysis.

Discussion

This study provides insights into the inconsistently reported relationship between SLE and BMD through a meta-analysis implementing a meta-regression tool. As suggested by the majority of studies reviewed studies our meta-analysis concluded that, overall, individuals with SLE have a lower BMD than non-SLE controls. Similarly, a recent meta-analysis showed that SLE patients had significantly lower BMD levels than controls in the whole body, femoral neck, lumbar spine and total hip [14]. The strength of this meta-analysis using a meta-regression lies in evaluating the anatomical regions of BMD measure as a whole, acting as covariates in the regression, without the need for subgroup analyses that might have a greater bias as the previous meta-analysis did [14].

No gender-specific differences were found between SLE cases and controls, even though low BMD is a gender-related condition. Both female and male SLE patients had a lower BMD at any region (lumbar spine, total hip and femoral neck) than healthy controls. To date, no gender-specific differences have been reported in these pathogenic mechanisms, although an already-fragile bone, such as that observed in postmenopausal women due to hormonal loss, may play a role in accelerating bone structure disruption [52]. Most studies included involved a higher number of premenopausal (n = 1913) than postmenopausal women (n = 823), which could be a reason for the lack of impact of gender on BMD.

We found no significant differences in the effect of GCT therapy on BMD at different regions between SLE patients with and without GCT. GCT are widely used to treat SLE disease flares and complications and might have beneficial effects by reducing the adverse effects of systemic inflammation on bone. The beneficial effects produced by suppressing the impact of inflammation on bone turnover might outweigh the harmful effects of GCT. Cross-sectional studies on the relationship between glucocorticoid use and BMD in SLE show conflicting results [53]. There is a wide disparity in the GCT doses, range, mean and cumulative dose and time of exposure in the studies evaluated that could partly explain the lack of between-group differences and result in probable bias [54].

The pathophysiologic mechanisms of impaired bone quality in GCT users remain unclear. Bone quality is determined by architecture, turnover, microdamage accumulation, mineralization and bone matrix protein such as collagen. In vivo studies have shown that GCT administration causes low bone turnover due to the suppression of osteoblast function, the induction of apoptosis in osteoblasts [55], and a major loss of trabecular connectivity [56]. GCT also affect bone geometry by reducing bone formation in periosteal surfaces [57]. This suggests that GCT administration leads to deterioration in bone structure. Since very few studies have assessed the trabecular microstructure either by computed tomography (CT) or magnetic resonance techniques in SLE [5861], these studies were not included in our meta-analysis.

VF are the most common type of osteoporotic fracture and are associated with substantial morbidity and decreased survival. They are diagnosed using the Genant semi-quantitative method, which requires a ≥20% decrease in vertebral height (anterior, mid or posterior dimensions), estimated visually, to diagnose a vertebral fracture. A recent meta-analysis including two studies [62,63] reported an almost three-fold higher risk of VF in SLE patients (RR 2.97, 95% CI 1.71–5.16, P < 0.001) compared with healthy controls [14]. However, in the studies included, VF were not assessed using a standardized method such as the Genant semi-quantitative method and, in addition, the impact of BMD measurements was not evaluated. In our meta-analysis, including more studies (five)[11,12,28,29,51] using a standardized method for VF detection, SLE patients with VF had lower BMD measurements either at the lumbar spine, total hip or femoral neck, compared to patients without VF, even though VF in GCT-induced osteoporosis may occur at higher BMD measurements than those associated with postmenopausal osteoporosis, according to a large study [64]. As mentioned, we found no significant effect for gender in any group, including VF. We were not able to evaluate the direct impact of GCT use on the VF risk, because not all studies included measured this relationship and also because differences in reporting steroid use limited the ability to extract this information.

Meta-regression is a tool used in meta-analysis to examine the impact of moderator variables on study effect sizes using regression-based techniques. This type of approach was different than that used by a previous meta-analysis [14] which used subgroup analyses techniques. Meta-regression is more effective at this task than standard meta-analytic techniques [65].

Our study has several limitations and the results should be interpreted with caution. First, the review is preliminary, partly due to the scarcity of reports, making definitive conclusions difficult. We included only studies in adults and did not include children or adolescents because bone density differs considerably between them. Secondly, there was high heterogeneity (see Table 3) between studies. A meta-analysis using individual patient data is recommended to provide good evidence for bone loss and fracture risk in SLE patients. Additionally, there were variations in the definition of GCT therapy between studies. Several methods were identified to define the risk of low BMD due to GCT treatment in SLE patients. However, we used “current use” or “ever use” binary response definitions to compare SLE patients with or without GCT therapy. The “current use” definition examines the association between BMD or fracture and whether the patient was exposed to GCT on the day of the measurements: important assumptions for this definition are that any prior GCT exposure does not affect the risk of low BMD or fracture, and the dose of GCT on the day of the measurements is not important. The “ever use” definition, conversely, assumes that all historical therapy affects the risk of fracture, but is regardless of how recently the therapy was taken. Thirdly, several covariates of clinical importance were not included in our analysis, such as ethnicity and post-menopausal status when comparing SLE patients and controls. Fourthly, we were not able to calculate odd ratios or relative risk for VF in SLE patients, since most studies in this subgroup analysis did not include healthy controls. Fifthly, most studies included had cross-sectional designs with a small sample size; there is no doubt that large prospective cohort studies adjusting for cofounders are more appropriate in assessing the fracture risk in SLE patients, in order to establish a real temporal cause-effect relationship. In fact, our review has disclosed the scarcity of longitudinal studies on this topic. We believe this meta-regression should encourage the research community to initiate and design future cohort studies.

Finally, significant publication bias was found for the first analysis (comparison of SLE patients and controls). However, this may probably be due to the small study effect rather than true publication bias, especially in the presence of significant heterogeneity between studies [66]. The majority of meta-analyses are based on a series of studies to produce a point estimate of an effect and measures of the precision of that estimate. In addition, a meta-regression model seeks to determine whether a study-level covariate is a plausible source of heterogeneity in a set of treatments or an output variable effect. Upon doing such analyses, as we have shown in the present study, the bias of the small study sample size of the independent studies is overcome. As studies become less precise, such as in smaller trials, the results of the studies can be expected to be more variable than the more precise larger studies; this aspect was cancelled out in the present study through the publication bias analyses and, as a result, allowed an objective assessment[67,68].

In conclusion, this meta-analysis, using meta-regression analytic techniques, showed that SLE patients have lower BMD levels than healthy controls independently of the skeletal site of measurement and gender. In addition, BMD levels at any region did not differ between SLE patients with and without GCT therapy. SLE patients with VF have lower BMD levels than those without. Larger prospective cohort studies are needed to provide a more accurate assessment of the relationship between SLE and fracture risk. Future studies evaluating effective osteoporosis screening and prevention in SLE patients are also required.

Supporting information

S1 Fig. Funnel plots for the three models supplementary.

A. Funnel plot of the model for case-controls. B. Funnel plot for GCT therapy. C. Funnel plot for vertebral fractures.

https://doi.org/10.1371/journal.pone.0196113.s001

(TIF)

S2 Fig. Forest plot for case- control study.

Regions codified are as follows: 1) lumbar spine 2) femoral neck and 3) total hip.

https://doi.org/10.1371/journal.pone.0196113.s002

(TIF)

S3 Fig. Forest plot for therapy effect study.

Regions codified are as follows: 1) lumbar spine 2) femoral neck and 3) total hip.

https://doi.org/10.1371/journal.pone.0196113.s003

(TIF)

S4 Fig. Forest plot for vertebral fractures study.

Regions codified are as follows: 1) lumbar spine 2) femoral neck and 3) total hip.

https://doi.org/10.1371/journal.pone.0196113.s004

(TIF)

S1 Table. Detailed search criteria for PubMed-MeSH database.

https://doi.org/10.1371/journal.pone.0196113.s006

(PDF)

S2 Table. Glucocorticoid use definitions for the analysis comparing patients with and without GCT therapy.

https://doi.org/10.1371/journal.pone.0196113.s007

(PDF)

Acknowledgments

This research was supported, in part, by the Institutional Planning Department/Benemerita Universidad Autonoma de Puebla. We thank David Buss for technical assistance.

References

  1. 1. Sinigaglia L, Varenna M, Binelli L, Zucchi F, Ghiringhella D, Gallazzi M, et al. Determinants of bone mass in systemic lupus erythematosus: a cross sectional study on premenopausal women. J Rheumatol. 1999;26:1280–4. pmid:10381043
  2. 2. Bultink IEM, Lems WF, Kostense PJ, Dijkmans B, Voskuyl AE. Prevalence of and risk factors for low bone mineral density and vertebral fractures in patients with systemic lupus erythematosus. Arthritis Rheum. 2005;52:2044–50. pmid:15986345
  3. 3. Kipen Y, Buchbinder R, Forbes A, Strauss B, Littlejohn G, Morand E. Prevalence of reduced bone mineral density in systemic lupus erythematosus and the role of steroids. J Rheumatol. 1997;24:1922–9. pmid:9330933
  4. 4. Almehed K, Hetenyi S, Ohlsson C, Carlsten H, Forsblad-d’Elia H. Prevalence and risk factors of vertebral compression fractures in female SLE patients. Arthritis Res Ther. 2010;12:R153. pmid:20678217
  5. 5. Gladman D, Ginzler E, Goldsmith C, Fortin P, Liang M, Urowitz M, et al. The development and initial validation of the Systemic Lupus International Collaborating Clinics/American College of Rheumatology damage index for systemic lupus erythematosus. Arthritis Rheum. 1996;39:363–9. pmid:8607884
  6. 6. Cummings SR, Black D. Bone mass measurements and risk of fracture in Caucasian women: a review of findings from prospective studies. Am J Med. 1995;98:24S–28S. pmid:7709929
  7. 7. Lakshminarayanan S, Walsh S, Mohanraj M, Rothfield N. Factors associated with low bone mineral density in female patients with systemic lupus erythematosus. J Rheumatol. 2001;28:102–8. pmid:11196509
  8. 8. Mendoza-Pinto C, García-Carrasco M, Sandoval-Cruz H, Escárcega RO, Jiménez-Hernández M, Etchegaray-Morales I, et al. Risks factors for low bone mineral density in pre-menopausal Mexican women with systemic lupus erythematosus. Clin Rheumatol. 2009;28:65–70. pmid:18670734
  9. 9. Oleksik A, Lips P, Dawson A, Minshall ME, Shen W, Cooper C, et al. Health-related quality of life in postmenopausal women with low BMD with or without prevalent vertebral fractures. J Bone Miner Res. 2000;15:1384–92. pmid:10893688
  10. 10. Hasserius R, Karlsson MK, Nilsson BE, Redlund-Johnell I, Johnell O. Prevalent vertebral deformities predict increased mortality and increased fracture rate in both men and women: a 10-year population-based study of 598 individuals from the Swedish cohort in the European Vertebral Osteoporosis Study. Osteoporos Int. 2003;14:61–8. pmid:12577186
  11. 11. Li EK, Tam LS, Griffith JF, Zhu TY, Li TK, Li M, et al. High prevalence of asymptomatic vertebral fractures in Chinese women with systemic lupus erythematosus. J Rheumatol. 2009;36:1646–52. pmid:19605677
  12. 12. Mendoza-Pinto C, García-Carrasco M, Sandoval-Cruz H, Muñoz-Guarneros M, Escárcega RO, Jiménez-Hernández M, et al. Risk factors of vertebral fractures in women with systemic lupus erythematosus. Clin Rheumatol. 2009;28:579–85. pmid:19224131
  13. 13. Garcia-Carrasco M, Mendoza-Pinto C, Leon-Vazquez M de la L, Mendez-Martinez S, Etchegaray-Morales I, Montiel-Jarquin A, et al. Incidence of Vertebral Fractures in Women with Systemic Lupus Erythematosus After 8 Years of Follow-Up. Calcif Tissue Int. 2017;
  14. 14. Wang X, Yan S, Liu C, Xu Y, Wan L, Wang Y, et al. Fracture risk and bone mineral density levels in patients with systemic lupus erythematosus: a systematic review and meta-analysis. Osteoporos Int. 2016;1413–23. pmid:26753541
  15. 15. Salman-Monte TC, Torrente-Segarra V, Vega-Vidal AL, Corzo P, Castro-Dominguez F, Ojeda F, et al. Bone mineral density and vitamin D status in systemic lupus erythematosus (SLE): A systematic review. Autoimmun Rev. Netherlands; 2017;16:1155–9. pmid:28899800
  16. 16. Kaji H, Yamauchi M, Chihara K, Sugimoto T. The threshold of bone mineral density for vertebral fracture in female patients with glucocorticoid-induced osteoporosis. Endocr J. 2006;53:27–34. pmid:16543669
  17. 17. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ. 2009;339:b2535. pmid:19622551
  18. 18. Howick J, Chalmers I, Glasziou P, Greenhalgh T, C H, Liberati A, et al. “The 2011 Oxford CEBM Levels of Evidence (Introductory Document)”. Oxford Cent. Evidence-Based Med.
  19. 19. Konstantopoulos S. Fixed effects and variance components estimation in three-level meta-analysis. Res Synth Methods. 2011;2:61–76. pmid:26061600
  20. 20. V W. Conducting meta-analyses in R with the metafor package. J Stat Softw. 2010;36:1–48.
  21. 21. Chen CJ, Yen JH, Tsai WC, Lin MB, Hsu SC, Tsai JJ, et al. Decreased bone mineral density in premenopausal patients with systemic lupus erythematosus. Kaohsiung J Med Sci. 1996;12:567–72. pmid:8918077
  22. 22. Mok CC, Ying SKY, To CH, Ma KM. Bone mineral density and body composition in men with systemic lupus erythematosus: a case control study. Bone [Internet]. 2008;43:327–31. Available from: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=18515206 pmid:18515206
  23. 23. Mok CC, Wong SN, Ma KM. Childhood-onset disease carries a higher risk of low bone mineral density in an adult population of systemic lupus erythematosus. Rheumatology. 2012;51:468–75. pmid:22096013
  24. 24. Tang XL, Griffith JF, Qin L, Hung VW, Kwok AW, Zhu TY, et al. SLE disease per se contributes to deterioration in bone mineral density, microstructure and bone strength. Lupus. 2013;22:1162–8. pmid:23884986
  25. 25. Sun YN, Feng XY, He L, Zeng LX, Hao ZM, Lv XH, et al. Prevalence and possible risk factors of low bone mineral density in untreated female patients with systemic lupus erythematosus. Biomed Res Int. United States; 2015;2015:510514. pmid:25738154
  26. 26. Guo Q, Fan P, Luo J, Wu S, Sun H, He L, et al. Assessment of bone mineral density and bone metabolism in young male adults recently diagnosed with systemic lupus erythematosus in China. Lupus. England; 2017;26:289–93. pmid:27522093
  27. 27. Coimbra IB, Costallat LTL. Bone mineral density in systemic lupus erythematosus and its relation to age at disease onset, plasmatic estradiol and immunosuppressive therapy. Joint Bone Spine. 2003;70:40–5. pmid:12639616
  28. 28. Borba VZC, Matos PG, da Silva Viana PR, Fernandes A, Sato EI, Lazaretti-Castro M. High prevalence of vertebral deformity in premenopausal systemic lupus erythematosus patients. Lupus [Internet]. 2005;14:529–33. Available from: http://www.ncbi.nlm.nih.gov/pubmed/16130509 pmid:16130509
  29. 29. Bonfa AC, Seguro LPC, Caparbo V, Bonfa E, Pereira RMR. RANKL and OPG gene polymorphisms: associations with vertebral fractures and bone mineral density in premenopausal systemic lupus erythematosus. Osteoporos Int. 2015;26:1563–71. pmid:25609157
  30. 30. Formiga F, Moga I, Nolla JM, Pac M, Mitjavila F, Roig-Escofet D. Loss of bone mineral density in premenopausal women with systemic lupus erythematosus. Ann Rheum Dis. 1995;54:274–6. pmid:7763104
  31. 31. Pons F, Peris P, Guanabens N, Font J, Huguet M, Espinosa G, et al. The effect of systemic lupus erythematosus and long-term steroid therapy on bone mass in pre-menopausal women. Br J Rheumatol. 1995;34:742–6. pmid:7551659
  32. 32. Formiga F, Nolla JM, Mitjavila F, Bonnin R, Navarro MA, Moga I. Bone mineral density and hormonal status in men with systemic lupus erythematosus. Lupus. 1996;5:623–6. pmid:9116708
  33. 33. Salman-Monte TC, Torrente-Segarra V, Munoz-Ortego J, Mojal S, Carbonell-Abello J. Prevalence and predictors of low bone density and fragility fractures in women with systemic lupus erythematosus in a Mediterranean region. Rheumatol Int. Germany; 2015;35:509–15. pmid:25030324
  34. 34. Alele JD, Kamen DL, Hunt KJ, Ramsey-Goldman R. Bone geometry profiles in women with and without SLE. J Bone Miner Res. 2011;26:2719–26. pmid:21755534
  35. 35. Teichmann J, Lange U, Stracke H, Federlin K, Bretzel RG. Bone metabolism and bone mineral density of systemic lupus erythematosus at the time of diagnosis. Rheumatol Int. 1999;18:137–40. pmid:10220833
  36. 36. Becker A, Fischer R, Scherbaum WA, Schneider M. Osteoporosis screening in systemic lupus erythematosus: impact of disease duration and organ damage. Lupus. 2001;10:809–14. pmid:11789491
  37. 37. Bhattoa HP, Kiss E, Bettembuk P, Balogh A. Bone mineral density, biochemical markers of bone turnover, and hormonal status in men with systemic lupus erythematosus. Rheumatol Int. 2001;21:97–102. pmid:11765229
  38. 38. Bhattoa HP, Bettembuk P, Balogh A, Szegedi G, Kiss E. Bone mineral density in women with systemic lupus erythematosus. Clin Rheumatol. 2002;21:135–41. pmid:12086164
  39. 39. Ajeganova S, Gustafsson T, Jogestrand T, Frostegard J, Hafstrom I. Bone mineral density and carotid atherosclerosis in systemic lupus erythematosus: a controlled cross-sectional study. Arthritis Res Ther. England; 2015;17:84. pmid:25885788
  40. 40. Dhillon VB, Davies MC, Hall ML, Round JM, Ell PJ, Jacobs HS, et al. Assessment of the effect of oral corticosteroids on bone mineral density in systemic lupus erythematosus: a preliminary study with dual energy x ray absorptiometry. Ann Rheum Dis. 1990;49:624–6. pmid:2396869
  41. 41. Kalla AA, Fataar AB, Jessop SJ, Bewerunge L. Loss of trabecular bone mineral density in systemic lupus erythematosus. Arthritis Rheum. 1993;36:1726–34. pmid:8250992
  42. 42. Hansen M, Halberg P, Kollerup G, Pedersen-Zbinden B, Horslev-Petersen K, Hyldstrup L, et al. Bone metabolism in patients with systemic lupus erythematosus. Effect of disease activity and glucocorticoid treatment. Scand J Rheumatol. 1998;27:197–206. pmid:9645415
  43. 43. Carli L, Tani C, Spera V, Vagelli R, Vagnani S, Mazzantini M, et al. Risk factors for osteoporosis and fragility fractures in patients with systemic lupus erythematosus. Lupus Sci Med. England; 2016;3:e000098. pmid:26848397
  44. 44. Jardinet D, Lefèbvre C, Depresseux G, Lambert M, Devogelaer JP, Houssiau F. Longitudinal analysis of bone mineral density in pre-menopausal female systemic lupus erythematosus patients: deleterious role of glucocorticoid therapy at the lumbar spine. Rheumatology (Oxford). 2000;39:389–92.
  45. 45. Gilboe IM, Kvien TK, Haugeberg G, Husby G. Bone mineral density in systemic lupus erythematosus: comparison with rheumatoid arthritis and healthy controls. Ann Rheum Dis. 2000;59:110–5. pmid:10666165
  46. 46. Redlich K, Ziegler S, Kiener HP, Spitzauer S, Stohlawetz P, Bernecker P, et al. Bone mineral density and biochemical parameters of bone metabolism in female patients with systemic lupus erythematosus. Ann Rheum Dis. 2000;59:308–10. pmid:10733481
  47. 47. Boyanov M, Robeva R, Popivanov P. Bone mineral density changes in women with systemic lupus erythematosus. Clin Rheumatol. 2003;22:318–23. pmid:14579164
  48. 48. Uaratanawong S, Deesomchoke U, Lertmaharit S, Uaratanawong S. Bone mineral density in premenopausal women with systemic lupus erythematosus. J Rheumatol [Internet]. 2003;30:2365–8. Available from: http://www.ncbi.nlm.nih.gov/pubmed/14677178 pmid:14677178
  49. 49. Mak A, Lim JQ, Liu Y, Cheak AAC, Ho RCM. Significantly higher estimated 10-year probability of fracture in lupus patients with bone mineral density comparable to that of healthy individuals. Rheumatol Int. 2013;33:299–307. pmid:22441963
  50. 50. Jacobs J, Korswagen L-A, Schilder AM, van Tuyl LH, Dijkmans BAC, Lems WF, et al. Six-year follow-up study of bone mineral density in patients with systemic lupus erythematosus. Osteoporos Int. 2013;24:1827–33. pmid:23052940
  51. 51. Furukawa M, Kiyohara C, Horiuchi T, Tsukamoto H, Mitoma H, Kimoto Y, et al. Prevalence and risk factors of vertebral fracture in female Japanese patients with systemic lupus erythematosus. Mod Rheumatol. 2013;23:765–73. pmid:22903260
  52. 52. Riggs BL, Melton LJ 3rd Iii, Robb RA, Camp JJ, Atkinson EJ, Peterson JM, et al. Population-based study of age and sex differences in bone volumetric density, size, geometry, and structure at different skeletal sites. J Bone Miner Res. 2004;19:1945–54. pmid:15537436
  53. 53. Bultink IEM, Lems WF. Systemic lupus erythematosus and fractures. RMD open [Internet]. 2015;1:e000069. Available from: http://www.ncbi.nlm.nih.gov/pubmed/26557383%5Cnhttp://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC4632145 pmid:26557383
  54. 54. Velentgas P, Dreyer NA, Nourjah P, Smith SR, Torchia MM, editors. Developing a Protocol for Observational Comparative Effectiveness Research: A User’s Guide. Rockville (MD); 2013.
  55. 55. Weinstein RS, Jilka RL, Parfitt AM, Manolagas SC. Inhibition of osteoblastogenesis and promotion of apoptosis of osteoblasts and osteocytes by glucocorticoids. Potential mechanisms of their deleterious effects on bone. J Clin Invest. 1998;102:274–82. pmid:9664068
  56. 56. Dalle Carbonare L, Arlot ME, Chavassieux PM, Roux JP, Portero NR, Meunier PJ. Comparison of trabecular bone microarchitecture and remodeling in glucocorticoid-induced and postmenopausal osteoporosis. J Bone Miner Res. 2001;16:97–103. pmid:11149495
  57. 57. Akahoshi S, Sakai A, Arita S, Ikeda S, Morishita Y, Tsutsumi H, et al. Modulation of bone turnover by alfacalcidol and/or alendronate does not prevent glucocorticoid-induced osteoporosis in growing minipigs. J Bone Miner Metab. 2005;23:341–50. pmid:16133683
  58. 58. Hansen S, Gudex C, Ahrberg F, Brixen K, Voss A. Bone geometry, volumetric bone mineral density, microarchitecture and estimated bone strength in Caucasian females with systemic lupus erythematosus. A cross-sectional study using HR-pQCT. Calcif Tissue Int. 2014;95:530–9. pmid:25326144
  59. 59. Tang XL, Qin L, Kwok AW, Zhu TY, Kun EW, Hung VW, et al. Alterations of bone geometry, density, microarchitecture, and biomechanical properties in systemic lupus erythematosus on long-term glucocorticoid: A case-control study using HR-pQCT. Osteoporos Int. 2013;24:1817–26. pmid:23104200
  60. 60. Li EK, Zhu TY, Tam LS, Hung VW, Griffith JF, Li TK, et al. Bone microarchitecture assessment by high-resolution peripheral quantitative computed tomography in patients with systemic lupus erythematosus taking corticosteroids. J Rheumatol. 2010;37:1473–9. pmid:20472932
  61. 61. Paupitz JA, Lima GL, Alvarenga JC, Oliveira RM, Bonfa E, Pereira RMR. Bone impairment assessed by HR-pQCT in juvenile-onset systemic lupus erythematosus. Osteoporos Int. 2016;27:1839–48. pmid:26694597
  62. 62. Weiss RJ, Wick MC, Ackermann PW, Montgomery SM. Increased fracture risk in patients with rheumatic disorders and other inflammatory diseases—a case-control study with 53,108 patients with fracture. J Rheumatol. 2010;37:2247–50. pmid:20889599
  63. 63. Ekblom-Kullberg S, Kautiainen H, Alha P, Leirisalo-Repo M, Julkunen H. Frequency of and risk factors for symptomatic bone fractures in patients with systemic lupus erythematosus. Scand J Rheumatol. 2013;42:390–3. pmid:23721483
  64. 64. Van Staa TP, Laan RF, Barton IP, Cohen S, Reid DM, Cooper C. Bone Density Threshold and Other Predictors of Vertebral Fracture in Patients Receiving Oral Glucocorticoid Therapy. Arthritis Rheum. 2003;48:3224–9. pmid:14613287
  65. 65. Higgins JPT GS (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. Meta-regression chapter. The Cochrane Collaboration, 2011. Available from www.handbook.cochrane.org. 2011.
  66. 66. Terrin N, Schmid CH, Lau J, Olkin I. Adjusting for publication bias in the presence of heterogeneity. Stat Med. England; 2003;22:2113–26. pmid:12820277
  67. 67. Rotondi MA, Donner A, Koval JJ. Evidence-based sample size estimation based upon an updated meta-regression analysis. Res Synth Methods. England; 2012;3:269–84. pmid:26053421
  68. 68. Haidich AB. Meta-analysis in medical research. Hippokratia. Greece; 2010;14:29–37. pmid:21487488