Lichen planus and dyslipidemia: a systematic review and meta-analysis of observational studies

Transcription

1 Report Lichen planus and dyslipidemia: a systematic review and meta-analysis of observational studies Yi C. Lai 1,2, *, MPH, Yik W. Yew 2,3, *, MBBS, MPH, and Robert A. Schwartz 4,5, MD, MPH, FRCP Edin 1 Rutgers New Jersey Medical School, Newark, NJ, 2 Harvard T.H. Chan School of Public Health, Boston, MA, USA, 3 National Skin Centre, Singapore, 4 Departments of Dermatology, Preventive Medicine, and Pathology, Rutgers New Jersey Medical School, and 5 Rutgers University School of Public Affairs and Administration, Newark, NJ, USA Correspondence Yik W. Yew, MBBS, MPH National Skin Centre 1 Mandalay Road Singapore yikweng.yew@mail.harvard.edu Conflicts of interest: None. *Joint first authors. Abstract Background Lichen planus (LP) is a chronic inflammatory disease that has been shown to be positively associated with dyslipidemia. However, the magnitude and types of the underlying lipid abnormalities have not been investigated. This study aims to conduct a systematic review and meta-analysis to investigate the qualitative and quantitative association between LP and dyslipidemia. Methods A systematic search of studies published from inception to April 1, 2015, was conducted using MEDLINE, EMBASE, Web of Science, and Cochrane library databases. Meta-analyses of observational studies with both categorical and continuous outcome were performed. DerSimonian and Lard random effects models were utilized to calculate the pooled odds ratio and weighted mean difference (WMD). Publication bias was evaluated by funnel plot and Egger s test. Results Seven studies with 5242 subjects were included in this meta-analysis. Patients with LP were significantly more likely to have dyslipidemia, with a pooled odds ratio of 1.74 (95% confidence interval [CI]: , P = 0.004). LP was associated with higher levels of triglycerides (WMD mg/dl, 95% CI , P = 0.048), low-density lipoprotein (18.75 mg/dl, 95% CI to 54.72, P = 0.307), total cholesterol (19.22 mg/ dl, 95% CI 8.80 to 47.25, P = 0.179), and lower levels of high-density lipoprotein cholesterol ( 8.96 mg/dl, 95% CI to 3.30, P = 0.152). Conclusions Despite considerable heterogeneity, this study demonstrated that LP was significantly associated with an increased risk of dyslipidemia and higher triglyceride levels. For patients presenting with LP, physicians should be cognizant of this association and consider screening them for dyslipidemia. Introduction Lichen planus (LP) is a chronic inflammatory disease that affects the skin, mucous membranes, scalp, and nails. It occurs in % of the population, 1 commonly affecting those who are over 45 years old. 2 The exact pathogenesis of LP remains unknown but has been postulated to be a T-cell-mediated autoimmune process, resulting in damage of keratinocytes. 3 5 The inflammatory cytokines involved in LP have been demonstrated to influence insulin signaling, lipid/carbohydrate metabolism, as well as adipogenesis. 6 Previous studies have suggested a positive association between LP and metabolic disorders such as diabetes mellitus 7,8 and dsylipidemia. 9,10 However, the magnitude of the underlying abnormalities of the different lipid components has not been extensively investigated. Further quantification of the apparent association between LP and dyslipidemia by analyzing continuous data is necessary to guide clinical management. We, therefore, aim to conduct a systematic review and metaanalysis to investigate the qualitative and quantitative association between LP and dyslipidemia by evaluating both categorical and continuous data. Materials and methods The meta-analysis of observational studies in epidemiology guidelines were followed in conducting this study. 11 A systematic and quantitative synthesis of all observational studies that evaluated the relationship between dyslipidemia and LP was planned a priori. Search strategy A comprehensive database search was performed independently by two reviewers (Y.C.L. and Y.W.Y.) using MEDLINE, EMBASE, Web of Science, and Cochrane library e295 ª 2016 The International Society of Dermatology International Journal of Dermatology 2016, 55, e295 e304

2 e296 Report Lichen planus and dyslipidemia meta-analysis Lai, Yew and Schwartz databases. The search strategy was formulated according to the population, intervention/exposure, comparator, and outcomes framework and further elaborated in the eligibility criteria section. The following search criteria were used: ( Lichen Planus [MeSH] OR Lichen Planus, Oral [MeSH] OR lichen planus*) AND ( Lipid Metabolism Disorders [MeSH] OR dyslipidemia* OR hyperlipidemia* OR lipid* OR cholesterol* OR LDL* OR HDL* OR triglyceride*). The MeSH terms for Lichen Planus and Lipid Metabolism Disorders included other synonyms or subheading terms such as lichenoid eruptions or papulosquamous skin diseases and dyslipidemias, hyperlipidemias, or hypolipoproteinemias, respectively. The search was limited to English language studies published from inception to April 1, All abstracts were evaluated based on the inclusion criteria to determine eligibility for the meta-analysis. Additional studies were identified from manual searches of references in retrieved articles. Web searches for abstracts/ proceedings of relevant studies presented at meetings or conferences were also done to identify any reports from the gray literature. Eligibility criteria The following inclusion criteria were used to select eligible studies for analysis: cohort, case control, or cross-sectional study that includes participants of any age with a previous diagnosis of LP (population); prevalence/incidence of dyslipidemia or laboratory measurements of lipid profile (total cholesterol, high-density lipoprotein cholesterol [HDL-C], lowdensity lipoprotein cholesterol [LDL-C], and triglycerides [TG]) (intervention/exposure); adequate control groups without a history of LP (comparator); analysis of the association between LP and dyslipidemia (outcomes). For meta-analysis of categorical data, studies would need to report sufficient information, such as odds ratio and 95% confidence interval (CI), so that the corresponding standard errors could be calculated. Where such information was not readily available, crude data with the number of cases with dyslipidemia in LP and control groups should be reported. For meta-analysis of studies with continuous data, studies would need to include mean values and corresponding standard deviations (SD) of serum lipid profile levels. If data were not presented in this format, they were converted using methods described in the Cochrane Handbook for Systematic Reviews of Interventions. 12 Studies were excluded if raw data were not available or could not be converted. The titles and abstracts of articles were reviewed by the reviewers. Based on the inclusion criteria and information from abstract, eligible articles were identified for full text review. Full text articles were then independently evaluated to determine their eligibility for inclusion in the meta-analysis. Any disagreements were resolved by consensus. Data extraction Reviewers independently extracted data from the nine reports with a standardized data extraction form. Relevant information extracted include the year of publication, country of study, study design, gender, number of cases and controls, inclusion based on LP or oral LP, mean values and SDs, and adjusted odds ratios and their respective 95% CIs when available, confounding factors adjusted, as well as other relevant characteristics of the studies. We also assessed the studies quality by a previously validated five-point scale, 13 with values 0 and 1 assigned to five categories: appropriate reporting of inclusion and exclusion criteria, diagnosis of LP, assessment of outcome (dyslipidemia), adjustment for confounders, and evidence of bias. The quality score of each study was derived from the summation of the scores of the five categories. Studies with quality scores 0 3 were considered lower quality, while those with a score 4 5 were considered higher quality. Any differences in scores were adjudicated by consensus. Meta-analysis Most selected studies with categorical outcomes reported odds ratios. Multivariate adjusted odds ratios were included for metaanalysis when available; otherwise, these ratios were estimated from the crude data. The confounding factors adjusted include age, sex, smoking status, socioeconomic status, as well as other comorbidities. Studies with mean values and corresponding SDs of lipid profile levels were included for calculation of weighted mean difference (WMD). Different units used in different studies were converted to the same units to allow for comparison, e.g., from mmol/l to mg/dl. We performed a random effects analysis using the method of DerSimonian and Lard 14 to estimate the pooled odds ratio and calculate the pooled WMD. Heterogeneity between studies was assessed using the I 2 statistic. Publication bias was qualitatively assessed by constructing and visually inspecting the funnel plot for any asymmetry. Egger s test and plot, which evaluated small study effects, was utilized to assess quantitatively the publication bias. 15 To explore further the potential sources of study heterogeneity, meta-regression with pre-specified variables, including gender, study quality, and study design were performed. All analyses were performed using STATA Version 12.0 (StataCorp, College Station, TX, USA). Results Search results A search using MEDLINE, EMBASE, and Web of Science yielded 262 articles (Fig. 1). There were no identified studies from the Cochrane central register. Of the 150 articles initially identified (after removing 112 duplicate studies), 15 studies were selected based on the prescribed International Journal of Dermatology 2016, 55, e295 e304 ª 2016 The International Society of Dermatology

3 Lai, Yew and Schwartz Lichen planus and dyslipidemia meta-analysis Report e297 Figure 1 Study selection flow diagram inclusion criteria. After assessing the full-length articles, eight articles were excluded for the following reasons: review articles (n = 3), case reports (n = 2), insufficient information (n = 2), and duplicate study data (n = 1). After these exclusions, seven studies were included in the meta-analyses. However, two studies reported outcomes stratified according to gender, and these results were analyzed separately, yielding nine independent reports with 5242 subjects for meta-analyses. Three studies, yielding four reports, had categorical data with odd ratio estimates, while six studies, with eight independent reports, included continuous data on lipid profile levels. Description of included studies There were six case control studies and one cross-sectional study. Six studies were conducted in Europe, while the rest were in Egypt, India, and Israel. Table 1 summarizes the study population characteristics in each study. 6,9,10,16 19 All but one study identified cases of LP through histological confirmation. Three studies were deemed to be high quality based on the aforementioned validated five-point scale. Seven studies performed multivariate adjustments to control for potential confounding variables. The covariates adjusted variably included age, sex, smoking status, socioeconomic status, and comorbidities such as hypothyroidism, diabetes mellitus, hypertension, and obesity. The reported adjusted odds ratios among the four reports ranged from 1.34 to Table 2 summarizes the relationship between dyslipidemia and LP. Table 3 summarizes the mean values of lipid profile levels. Categorical outcome There was substantial heterogeneity among the four reports with categorical data (I 2 = 63.2%, P < 0.043). ª 2016 The International Society of Dermatology International Journal of Dermatology 2016, 55, e295 e304

4 e298 Report Lichen planus and dyslipidemia meta-analysis Lai, Yew and Schwartz Table 1 Study population characteristics: lichen planus and lipid profile Study Year Country of study Study design Total no. of study participants Gender Age (years) Dyslipidemia definition Categorical outcome Dreiher et al Israel Case control 4333 Both 53.0 (1) Diagnosis code of dyslipidemia OR (2) Prescribed statins or fibrates OR (3) Total cholesterol >250 mg/dl Arias-Santiago et al. 10 male 2011 Spain Case control 100 Male 46.9 (1) TG > 150 mg/dl Arias-Santiago et al. 10 female 2011 Spain Case control 100 Female 47.8 (2) TC > 200 mg/dl (3) LDL-C > 130 mg/dl OR (4) Treatment for dyslipidemia Lopez-Jornet et al Spain Cross-sectional 400 Both 57.6 (1) TG > 150 mg/dl (2) TC > 200 mg/dl (3) LDL-C > 130 mg/dl (4) HDL-C < 40 mg/dl AND (5) Total cholesterol/hdl-c of 5.1 for men and 4.5 for women Continuous outcome Arias-Santiago et al. 10 male 2011 Spain Case control 100 Male 46.9 As above Arias-Santiago et al. 10 female 2011 Spain Case control 100 Female 47.8 Lopez-Jornet et al. 16 male 2012 Spain Cross-sectional 200 Male 57.6 As above Lopez-Jornet et al. 16 female 2012 Spain Cross-sectional 200 Female 57.6 Sahin et al Turkey Case control 95 Both 43.4 Not provided Krishnamoorthy et al India Case control 32 Both 33.1 (1) TG > 150 mg/dl (2) TC > 200 mg/dl (3) LDL-C > 130 mg/dl Polic et al Croatia Case control 102 Both 54.4 (1) Total cholesterol > 5.0 mm (2) LDL-C > 3.0 mm (3) TG > 1.7 mm (4) HDL-C < 1.0 mm for men and <1.2 mm for women Saleh et al Egypt Case control 80 Both 38.2 (1) TG > 150 mg/dl (2) HDL-C < 40 mg/dl HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; TG, triglyceride. Table 2 Study results: LP and dyslipidemia Study Year Outcome assessment Total LP cases Measure of association: odds ratio (95% CI) Adjustment Dreiher et al Medical records ( ) Age, sex, smoking, hypothyroidism, diabetes, hypertension, socioeconomic status, obesity Arias-Santiago et al. 10 male 2011 Histologically confirmed ( ) None Arias-Santiago et al. 10 female 2011 Histologically confirmed ( ) None Lopez-Jornet et al Histologically confirmed ( ) None LP, lichen planus. The random effects analysis indicated that patients with LP were significantly more likely to have dyslipidemia, with a pooled odds ratio of 1.74 (95% CI: , P = 0.004), as shown in Figure 2. The funnel plot, which illustrated the relationship between study size and strength of association, was visually inspected to assess for publication bias. The funnel plot demonstrated a lack of small size studies in the nega- International Journal of Dermatology 2016, 55, e295 e304 ª 2016 The International Society of Dermatology

5 Lai, Yew and Schwartz Lichen planus and dyslipidemia meta-analysis Report e299 Table 3 Study results: lichen planus and lipid profile (measured values) Study Lichen planus Controls Mean difference Total cholesterol (mg/dl) Arias-Santiago et al. 10 male Arias-Santiago et al. 10 female Lopez-Jornet et al. 16 male Lopez-Jornet et al. 16 female Krishnamoorthy et al Polic et al Saleh et al HDL-C (mg/dl) Arias-Santiago et al. 10 male Arias-Santiago et al. 10 female Lopez-Jornet et al. 16 male Lopez-Jornet et al. 16 female Krishnamoorthy et al Polic et al Saleh et al LDL-C (mg/dl) Arias-Santiago et al. 10 male Arias-Santiago et al. 10 female Lopez-Jornet et al. 16 male Lopez-Jornet et al. 16 female Sahin et al Krishnamoorthy et al Polic et al Saleh et al TG (mg/dl) Arias-Santiago et al. 10 male Arias-Santiago et al. 10 female Lopez-Jornet et al. 16 male Lopez-Jornet et al. 16 female Sahin et al Krishnamoorthy et al Polic et al Saleh et al HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; TG, triglyceride. tive association, suggesting a probable publication bias (Fig. 3). However, formal testing with Eggers test yielded an intercept term of 2.17 (P = 0.096), which provided no significant evidence for a small study effect. Continuous outcome LP was associated with a significantly higher level of TG, with a WMD of mg/dl (95% CI , P = 0.048). Although those with LP had higher levels of LDL-C and total cholesterol, with a WMD of mg/ dl (95% CI to 54.72, P = 0.307) and a WMD of mg/dl (95% CI 8.80 to 47.25, P = 0.179), respectively, these differences were not statistically significant. On the other hand, patients with LP had lower HDL levels, with a WMD of 8.96 mg/dl (95% CI to 3.30, P = 0.152); however, this difference also failed to reach statistical significance (Fig. 4). All meta-analyses of continuous outcomes had an I 2 > 75%, indicating considerable heterogeneity among studies. Visual inspection of funnel plots (data not shown) indicated that publication bias could not be ruled out. Formal testing with Eggers test, however, provided no evidence for small study effect. Meta-regression Meta-regression was performed to evaluate pre-specified sources of heterogeneity, such as gender, study quality, and study design; however, they did not contribute significantly to study heterogeneity. Discussion To our knowledge, this is the first meta-analysis that evaluates the association of dyslipidemia with LP and assesses the mean difference in lipid profile levels between patients with LP and controls. By analyzing data from the four ª 2016 The International Society of Dermatology International Journal of Dermatology 2016, 55, e295 e304

6 e300 Report Lichen planus and dyslipidemia meta-analysis Lai, Yew and Schwartz Figure 2 Meta-analysis prevalence of dyslipidemia Figure 3 Funnel plot reports with categorical data involving 4733 LP cases, we found that patients with LP were approximately 75% more likely to have dyslipidemia. This finding was corroborated by sensitivity analyses. In terms of individual lipid profile components, a significantly higher level of TG was observed in patients with LP, while higher levels of total cholesterol and LDL- C and lower levels of HDL-C among these patients were not statistically significant. Despite this, the magnitude of these abnormalities in the lipid panel may have important clinical implications as risk factors of major cardiovascular events. Our analysis of eight reports with continuous data involving 909 LP cases showed a WMD of 8.96, 18.75, and mg/dl for HDL-C, LDL-C, and TG, respectively. Lowering LDL-C and TG while raising HDL-C can reduce the risks of cardiovascular diseases. The Justification for the Use of Statins in Prevention: An Intervention Trial Evaluating Rosuvastatin (JUPITER) trial, which investigated the impact on the rate of major cardiovascular events with statins use, had demonstrated a significant reduction in these events when HDL- C was increased by 1.97 mg/dl, LDL-C decreased by 54 mg/dl, and TG decreased by 21 mg/dl. 20 While the mean difference of LDL-C levels in our study is considerably lower than the LDL-C reduction needed to result in fewer cardiovascular events, the differences in HDL-C and TG levels are several times higher than those reported by the JUPITER trial. Moreover, higher TG and lower HDL-C levels have been demonstrated to promote the transition from atheroma to atherothrombosis, and management of these two lipid components may be beneficial in patients with subclinical dyslipidemia. 21 The abnormality in lipid levels among patients with LP, therefore, may be clinically significant, and patients with LP will likely benefit from screening and subsequent pharmaceutical management of dyslipidemia. An influence analysis of these eight reports showed a strong effect of the case control study by Saleh et al. 6 on the pooled estimate. It should be noted that the reported mean TG level of mg/dl among patients with LP was several times higher than those reported by other published studies, which ranged from to mg/dl. 6,9,10,16 18 The authors also reported a comparatively lower mean HDL- C level of mg/dl. 6 The HDL-C levels in patients with LP reported by other studies ranged from to mg/dl. 9,16,18 Mean total cholesterol and LDL-C levels in this study were also relatively higher than those from other studies. 6,9,10,16 18 The selection of healthy controls without obesity, diabetes mellitus, dyslipidemia, or metabolic syndrome in this study 6 may potentially skew the comparison in lipid profile between cases and controls. Sensitivity analyses excluding the study by Saleh et al. 6 revealed similar results, though the differences in lipid levels became smaller in magnitude. After exclusion of this study, 6 higher levels of TG, LDL-C, and total cholesterol, and lower levels of HDL were still observed in patients with LP, with smaller WMDs of 17.54, 10.75, 6.71, and 3.92 mg/dl, respectively. International Journal of Dermatology 2016, 55, e295 e304 ª 2016 The International Society of Dermatology

7 Lai, Yew and Schwartz Lichen planus and dyslipidemia meta-analysis Report e301 (A) (B) (C) (D) Figure 4 Forest plots of WMD of each component of lipid panel. (a) Low-density lipoprotein; (b) triglycerides; (c) high-density lipoprotein; (d) total cholesterol. WMD, weighted mean difference A recent meta-analysis of 18,666 patients with psoriasis showed that they tend to have a less favorable lipid profile, higher blood pressure and BMI, and poorer glucose control than do controls. 22 The chronic inflammation that underlies LP, a T-helper 1 dominant disease similar to psoriasis, has been postulated to be associated with metabolic syndrome, the hallmark of which is dyslipidemia. 10 Both LP and metabolic syndrome are characterized by increased activations of T-helper 1-mediated immune responses, the release of cytokines such as tumor necrosis factor alpha (TNF-a), interleukin (IL)-2, IL-4, IL-6, and IL-10, as well as the subsequent recruitment of inflammatory cells such as cytotoxic T lymphocytes and natural killer cells. These common pathways may contribute to the inflammatory processes involved in both diseases. 10,23 The generation of reactive oxygen species from the destruction of keratinocytes by lymphocytic infiltrate results in cell membrane damage and lipid peroxidation. 24 Indirect evidence from studies showing a decrease in TG or an increase in HDL-C levels with the use of TNF-a inhibitors lends some support to the positive association between LP and dyslipidemia. 25,26 It is not known, however, if dyslipidemia is secondary to the inflammatory state as a result of LP or both LP and dyslipidemia are part of an underlying systemic inflammatory process. Although successful management of LP with TNF-a inhibitors such as etanercept and adalimumab have been reported, their effects on dyslipidemia in patients with LP have not been previously investigated. The use of TNF-a inhibitors in other chronic inflammatory disorders, however, had yielded beneficial effects on dyslipidemia and cardiovascular diseases. Treatment with TNF-a inhibitors such as infliximab in patients with rheumatoid arthritis had been demonstrated to be anti-atherogenic and improve lipid profile in the short term. 31 In addition, in patients with psoriatic arthritis, treatment with TNF-a inhibitors had been shown to decrease carotid intimamedia thickness and arterial stiffness, both of which are independent predictors of cardiovascular disease. 32,33 A retrospective cohort study by Wu et al. concluded that management of psoriasis with TNF-a inhibitors significantly reduced the risk and incidence of myocardial infarction. 34 While there is substantial evidence that supports the role of TNF-a inhibitors in modulating the risk of cardiovascular disease in patients with inflammatory diseases such as rheumatoid arthritis and psoriasis, ª 2016 The International Society of Dermatology International Journal of Dermatology 2016, 55, e295 e304

8 e302 Report Lichen planus and dyslipidemia meta-analysis Lai, Yew and Schwartz Table 4 Literature review of the association between dyslipidemia and various inflammatory skin conditions Skin conditions Type of study Total number of cases Results Study Psoriasis Meta-analysis Dyslipidemia (odds ratio 1.5, 95% CI: ) Miller et al. 35 Hidradenitis suppurativa Comparative crosssectional 3207 Dyslipidemia (odds ratio 1.14, 95% CI: ) Shalom et al. 36 study Granuloma annulare Case control study 140 Dyslipidemia (odds ratio 4.04, 95% CI: ) Wu et al. 37 Granulomatous variant of CPPD Case series 4 Three of four CPPD patients had dyslipidemia Lin et al. 38 Androgenic alopecia Comparative case control study 150 Men: Dyslipidemia (odds ratio 4.58, 95% CI: ) Women: Dyslipidemia (odds ratio 3.63, 95% CI: ) Arias-Santiago et al. 39 Systemic lupus erythematosus Meta-analysis Dyslipidemia (odds ratio 3.9, 95% CI: ) Ballocca et al. 40 CPPD, chronic pigmented purpuric dermatosis. more research is still needed to investigate the beneficial effect of TNF-a inhibitors in patients with LP and dyslipidemia. Other chronic skin conditions have also been suggested to be associated with dyslipidemia, including, but not limited to, psoriasis, hidradenitis suppurativa, granuloma annulare, granulomatous variant of chronic pigmented purpuric dermatosis, androgenic alopecia, as well as systemic lupus erythematosus. The underlying chronic inflammatory processes and associated proinflammatory cytokines shared by LP and the aforementioned conditions were considered paramount in the development of dyslipidemia. Various studies have provided epidemiological evidence for the association between these inflammatory skin diseases and abnormalities in lipid levels Table 4 summarizes some of the studies that evaluate the relationship between dyslipidemia and various inflammatory skin conditions. The major strength of this study was the large sample size and the quantification of the difference in lipid profile components between patients with LP and controls. One of the limitations of this study was the considerable heterogeneity between the included studies, although meta-regression of study characteristics failed to reveal any identified sources of heterogeneity. We have attempted to address heterogeneity by employing the random effects model; however, considerable heterogeneity remained. Hence, the results may be difficult to interpret and not generalizable to typical patients with LP. Another limitation was the inclusion of only English-language articles, thus limiting the number of studies available for analysis. The small number of eligible studies included in this meta-analysis may result in a lack of statistical power to detect modest associations between LP and lipid levels. In addition, due to insufficient data, the analysis did not take into account the concurrent use of medications such as lipid-lowering or oral retinoid agents, which can potentially confound the association. More research, particularly long-term follow-up cohort studies, would be needed to evaluate the incidence of dyslipidemia in patients with LP. This meta-analysis demonstrated that dyslipidemia was significantly more common among patients with LP, in particular TG levels. Though not statistically significant, they also tend to have higher levels of total cholesterol and LDL-C, and lower levels of HDL-C, compared to the general population. However, given the considerable heterogeneity and limited number of studies available for meta-analysis, more prospective cohort studies are required to elucidate this relationship further. Nonetheless, from a preventive medicine perspective, understanding the mean difference in lipid levels may have clinical relevance in managing patients with LP. It is also important to recognize that, similar to psoriatic patients, patients with LP have significant cardiovascular risk factors. For patients presenting with LP, physicians should be cognizant of this association, recommend screening for dyslipidemia, and manage lipid abnormalities accordingly. References 1 Seyhan M, Ozcan H, Sahin I, et al. High prevalence of glucose metabolism disturbance in patients with lichen planus. Diabetes Res Clin Pract 2007; 77: Manolsche L, Seceleanu-Petrescu D, Benea V, et al. Lichen planus patients and stressful events. J Eur Acad Dermatol Venereol 2008; 22: Gorouhi F, Davari P, Fazel N. Cutaneous and mucosal lichen planus: a comprehensive review of clinical subtypes, risk factors, diagnosis, and prognosis. ScientificWorldJournal 2014; 2014: Usatine RP, Tinitigan M. Diagnosis and treatment of lichen planus. Am Fam Physician 2011; 84: International Journal of Dermatology 2016, 55, e295 e304 ª 2016 The International Society of Dermatology

9 Lai, Yew and Schwartz Lichen planus and dyslipidemia meta-analysis Report e303 5 Iijima W, Ohtani H, Nakayama T, et al. Infiltrating CD8 + T cells in oral lichen planus predominantly express CCR5 and CXCR3 and carry respective chemokine ligands RANTES/CCL5 and IP-10/CXCL10 in their cytolytic granules: a potential self-recruiting mechanism. Am J Pathol 2003; 163: Saleh N, Samir N, Megahed H, et al. Homocysteine and other cardiovascular risk factors in patients with lichen planus. J Eur Acad Dermatol Venereol 2014; 28: Atefi N, Majedi M, Peyghambari S, et al. Prevalence of diabetes mellitus and impaired fasting blood glucose in patients with lichen planus. Med J Islam Repub Iran 2012; 26: Muammer S, Hamdi O, Ibrahim S, et al. High prevalence of glucose metabolism disturbance in patients with lichen planus. Diabetes Res Clin Pract 2007; 77: Dreiher J, Shapiro J, Cohen AD. Lichen planus and dyslipidaemia: a case control study. Br J Dermatol 2009; 161: Arias-Santiago S, Buendia-Eisman A, Aneiro-Fernandez J, et al. Cardiovascular risk factors in patients with lichen planus. Am J Med 2011; 124: Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000; 283: Deeks JJ, Higgins JPT, Altman DG, eds. Chapter 9: analysing data and undertaking meta-analyses. In: Higgins JPT, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions Version (updated March 2011). The Cochrane Collaboration, (accessed on 1 April 2015). 13 Micha R, Wallace SK, Mozaffarian D. Red and processed meat consumption and risk of incident coronary heart disease, stroke, and diabetes mellitus: a systematic review and meta-analysis. Circulation 2010; 121: DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986; 7: Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997; 315: Lopez-Jornet P, Camacho-Alonso F, Rodrˇguez-Martˇnes MA. Alterations in serum lipid profile patterns in oral lichen planus: a cross-sectional study. Am J Clin Dermatol 2012; 13: Sahin M, Bilgili SG, Simsek H, et al. Increased P-wave dispersion in patients with newly diagnosed lichen planus. Clinics (Sao Paulo) 2013; 68: Krishnamoorthy B, Suma GN, Mamatha NS, et al. Lipid profile and metabolic syndrome status in patients with oral lichen planus, oral lichenoid reaction and healthy individuals attending a dental college in northern India a descriptive study. J Clin Diagn Res 2014; 8: ZC92 ZC Polic MV, Miskulin M, Solic K, et al. Imbalanced concentrations of serum lipids and lichen planus. Coll Antropol 2014; 38: Ridker PM, Danielson E, Fonseca FA, et al. Rosuvastatin to prevent vascular events in men and women with elevated C-reactive protein. N Engl J Med 2008; 359: Sharrett AR, Sorlie PD, Chambless LE, et al. Relative importance of various risk factors for asymptomatic carotid atherosclerosis versus coronary heart disease incidence: the Atherosclerosis Risk in Communities Study. Am J Epidemiol 1999; 149: Miller IM, Skaaby T, Ellervik C, et al. Quantifying cardiovascular disease risk factors in patients with psoriasis: a meta-analysis. Br J Dermatol 2013; 169: Simark-Mattssona C, Bergenholtza G, Jontell M, et al. Distribution of interleukin-2, -4, -10, tumour necrosis factor-a and transforming growth factor-b mrnas in oral lichen planus. Arch Oral Biol 1999; 44: Meller S, Gilliet M, Homey B. Chemokines in the pathogenesis of lichenoid tissue reactions. J Invest Dermatol 2009; 129: Wakkee M, Thio HB, Prens EP, et al. Unfavorable cardiovascular risk profiles in untreated and treated psoriasis patients. Atherosclerosis 2007; 190: Haroon M, Devlin J. Marked hypertriglyceridemia upon treatment with etanercept. Joint Bone Spine 2009; 76: Yarom N. Etanercept for the management of oral lichen planus. Am J Clin Dermatol 2007; 8: Chao TJ. Adalimumab in the management of cutaneous and oral lichen planus. Cutis 2009; 84: Irla N, Schneiter T, Haneke E, et al. Nail lichen planus: successful treatment with etanercept. Case Rep Dermatol 2010; 2: Hollo P, Szakonyi J, Kiss D, et al. Successful treatment of lichen planus with adalimumab. Acta Derm Venereol 2012; 92: Popa C, van den Hoogen FH, Radstake TR, et al. Modulation of lipoprotein plasma concentrations during long-term antietnf therapy in patients with active rheumatoid arthritis. Ann Rheum Dis 2007; 66: Tam LS, Li EK, Shang Q, et al. Tumor necrosis factor alpha blockade is associated with sustained regression of carotid intima-media thickness for patients with active psoriatic arthritis: a 2-year pilot study. Ann Rheum Dis 2011; 70: Angel K, Provan SA, Gulseth HL, et al. Tumor necrosis factor-alpha antagonists improve aortic stiffness in patients with inflammatory arthropathies: a controlled study. Hypertension 2010; 55: Wu JJ, Poon KY, Channual JC, et al. Association between tumor necrosis factor inhibitor therapy and myocardial infarction risk in patients with psoriasis. Arch Dermatol 2012; 148: ª 2016 The International Society of Dermatology International Journal of Dermatology 2016, 55, e295 e304

10 e304 Report Lichen planus and dyslipidemia meta-analysis Lai, Yew and Schwartz 35 Miller IM, Ellervik C, Yazdanyar S, et al. Meta-analysis of psoriasis, cardiovascular disease, and associated risk factors. J Am Acad Dermatol 2013; 69: Shalom G, Freud T, Harman-Boehm I, et al. Hidradenitis suppurativa and the metabolic syndrome: a comparative cross-sectional study of 3207 patients. Br J Dermatol 2015; 173: Wu W, Robinson-Bostom L, Kokkotou E, et al. Dyslipidemia in granuloma annulare: a case-control study. Arch Dermatol 2012; 148: Lin WL, Kuo TT, Shih PY, et al. Granulomatous variant of chronic pigmented purpuric dermatoses: report of four new cases and an association with hyperlipidaemia. Clin Exp Dermatol 2007; 32: Arias-Santiago S, Gutierrez-Salmeron MT, Buend ˇa- Eisman A, et al. A comparative study of dyslipidaemia in men and woman with androgenic alopecia. Acta Derm Venereol 2010; 90: Ballocca F, DAscenzo F, Moretti C, et al. Predictors of cardiovascular events in patients with systemic lupus erythematosus (SLE): a systematic review and metaanalysis. Eur J Prev Cardiol 2015; 22: International Journal of Dermatology 2016, 55, e295 e304 ª 2016 The International Society of Dermatology

11 本文献由 学霸图书馆 - 文献云下载 收集自网络, 仅供学习交流使用 学霸图书馆 ( 是一个 整合众多图书馆数据库资源, 提供一站式文献检索和下载服务 的 24 小时在线不限 IP 图书馆 图书馆致力于便利 促进学习与科研, 提供最强文献下载服务 图书馆导航 : 图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具