Prevalence of cervical HPV infection in women with systemic lupus erythematosus: a systematic review and meta-analysis

Methods

We conducted a systematic literature review (PubMed, Cochrane library, Embase, Virtual Health Library and SciELO databases) following PRISMA guidelines and meta-regression to investigate the pooled prevalence of cervical HPV infection in adult women with SLE. The included articles were independently evaluated by two investigators who extracted information on study characteristics, defined outcomes, risk of bias and summarized strength of evidence [Quality of evidence using the Oxford Centre for evidence-based medicine (EBM) Levels of Evidence] Using meta-regression, we further analyzed whether factors such as multiple sexual partners and immunosuppressive therapy were associated with HPV prevalence.

We evaluated the quality of evidence included using the Oxford Centre for evidence-based medicine (EBM) levels of evidence. Pooled odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for studies providing data on HPV prevalence in women with SLE and in healthy controls. Results: A total of 687 articles were identified; 9 full-text articles examining the prevalence of cervical HPV infection in SLE women were included, comprising 751 SLE women.

Eight studies employed PCR using general primers. The HPV prevalence varied from 3.1% to 80.7%. From the random effects meta-analysis, the pooled prevalence of cervical HPV infection in SLE vs. controls was 34.15% (95% CI: 19.6%-52.5%) vs. 15.3% (95% CI 0.79-27.8%), OR = 2.87 (95% CI: 2.20-3.76) p< 0.0001, with large between-study heterogeneity (I2= 95.4%).

When only SLE women were evaluated, meta-regression showed no significant differences between patients with and without a background of multiple sexual partners and any immunosuppressive therapy. In addition, the prevalence of cervical HPV infection did not significantly differ between SLE patients on azathioprine or cyclophosphamide. A detailed dataset presenting the totally of the articles included is attached.

The variables included are the result of two datasets constructed by two independently reviewers (CMP and MGC) after resolving disagreements by consensus. The variables included in the attached file are: first author, year of publication, country where the study was carried out, study design, the number of SLE patients and/or controls, their mean age, glucocorticoid use and dose in SLE participants as well as immunosuppressant, increased number of sexual partners, method sample collection, HPV detection method, HPV genotype and HPV high risk prevalence"