Journal Pre-proof

Serum lipids and risk of atherosclerosis in xanthelasma palpebrarum: A systematic

review and meta-analysis

Hua-Ching Chang, MD, MS, Chih-Wei Sung, MD, Ming-Hsiu Lin, MD

PII: S0190-9622(19)32690-8
DOI: https://doi.org/10.1016/j.jaad.2019.08.082
Reference: YMJD 13806

To appear in: Journal of the American Academy of Dermatology

Received Date: 22 July 2019

Revised Date: 23 August 2019
Accepted Date: 28 August 2019

Please cite this article as: Chang H-C, Sung C-W, Lin M-H, Serum lipids and risk of atherosclerosis in
xanthelasma palpebrarum: A systematic review and meta-analysis, Journal of the American Academy of
Dermatology (2019), doi: https://doi.org/10.1016/j.jaad.2019.08.082.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier on behalf of the American Academy of Dermatology, Inc.

 1

1 Serum lipids and risk of atherosclerosis in xanthelasma palpebrarum: A

2 systematic review and meta-analysis

4 Hua-Ching Chang, MD, MSa, Chih-Wei Sung, MDb, Ming-Hsiu Lin, MDa,c,*

5 a. Department of Dermatology, Taipei Medical University Hospital, Taipei, Taiwan

6 b. Department of Emergency Medicine, National Taiwan University Hospital

7 Hsin-Chu Branch, Hsinchu, Taiwan

8 c. Department of Dermatology, School of Medicine, College of Medicine, Taipei

9 Medical University, Taipei, Taiwan

10

11 *Correspondence: Ming-Hsiu Lin

12 Department of Dermatology, Taipei Medical University Hospital, Taipei, Taiwan

13 Address: No. 252, Wuxing St, Xinyi District, Taipei City, 110, Taiwan

14 Tel: 886-2-27372181 ext 8285

15 E-mail: living-white@yahoo.com.tw

16

17

18

19
 2

20 Text word count: 2438

21 Table number: 1

22 Figure number: 4

23 Number of references: 35

24 Conflicts of interest: The authors have no conflict of interest to declare.

25 Funding sources: This article has no funding source.

26

27

28

29

30

31

32

33

34

35

36

37

38
 3

39 Abstract

40 Background: The association between dyslipidemia and xanthelasma palpebrarum

41 (XP) remains controversial, and no definite evidence has indicated atherosclerosis risk

42 in patients with XP.

43 Objective: The present study is a systematic review and meta-analysis to elucidate the

44 association of serum lipid profiles and risk of atherosclerotic diseases with XP.

45 Methods: We systematically searched for the eligible comparative studies published

46 before April 15, 2019, in the databases of Pubmed, Web of Science, Embase, and

47 Cochrane Library. Standard mean difference with 95% confidence interval was

48 calculated for each pooled estimate using a random-effects model.

49 Results: Fifteen case–control studies with 854 patients with XP were included for

50 quantitative analyses. The patients with XP had significantly higher serum levels of

51 total cholesterol and low-density lipoproteins (LDLs) than the controls. Similarly, the

52 patients had significantly higher apolipoprotein B levels and relatively lower

53 apolipoprotein A1 levels than the controls did. In addition, the carotid intima-media

54 thickness was significantly higher in the patients than in the controls.

55 Conclusion: Patients with XP had significantly higher serum levels of atherogenic

56 LDLs and bear significantly higher risk of atherosclerosis than the controls. Careful

57 monitoring and targeted intervention for prevention of cardiovascular diseases is

 4

58 essential for these patients.

59

60 Keywords: apolipoprotein; atherosclerosis; cholesterol; meta-analysis; serum lipids;

61 xanthelasma palpebrarum

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76
 5

77 INTRODUCTION

78 Lipids are essential substances required for maintaining normal physiological

79 functions. Cholesterol and triglycerides are major lipids that are transported within

80 lipoproteins in the systemic circulation. Plasma lipoproteins are divided into five

81 major classes based on their compositions and relative density, namely chylomicrons,

82 very-low-density lipoproteins (VLDLs), intermediate-density lipoproteins (IDLs),

83 low-density lipoproteins (LDLs), and high-density lipoproteins (HDLs).1 In addition

84 to lipid components, lipoproteins also contain specific proteins known as

85 apolipoproteins (apos), which serve numerous pivotal functions during lipid

86 metabolism. Different lipoproteins are associated with specific types of apos; for

87 example, apo B is associated with VLDLs and LDLs and apo A1 is associated with

88 HDLs.2

89 Dyslipidemia, particularly hypercholesterolemia, is among crucial risk factors for

90 atherosclerotic vascular disease development. Although (high) LDLs and (low) HDLs

91 levels are the two most studied lipoproteins with regard to atherosclerotic disease,

92 VLDLs and triglyceride (TG) levels also play crucial roles in the progression of

93 atherosclerosis.3 Apos are important components of lipoprotein particles, and growing

94 evidence indicates that various forms of apos may serve as predictors of the risk of

95 cardiovascular diseases (CVDs).4 The concentration of apo B is highly correlated with

 6

96 the levels of potentially atherogenic lipoproteins (LDLs and VLDLs); conversely, apo

97 A1 is the major apolipoprotein associated with antiatherogenic lipoproteins, namely

98 HDLs.4 Hence, the ratio between apo B and apo A1 is proposed to be a reliable tool

99 for predicting cardiovascular risk.5

100 Xanthomas are pathological infiltrates of cholesterol-rich foam cells that can

101 appear in any body area, and xanthelasma or xanthelasma palpebrarum (XP) is the

102 most common cutaneous xanthoma.6 The prevalence of XP is higher in women than in

103 men, and the peaks of onset occur in the fourth and fifth decades of life.7 The typical

104 clinical manifestations of XP are symmetrically distributed yellowish plaques near the

105 inner canthus of the eyelid, and the upper eyelids are more predisposed than the lower

106 eyelids.6 Although XP is typically asymptomatic and benign, patients usually seek

107 treatment because of the cosmetic unpleasantness of XP.7

108 Hyperlipidemia was believed to be associated with XP; however, the association

109 between dyslipidemia and XP remains controversial. Furthermore, definite evidence

110 indicating the risk of atherosclerosis in patients with XP and the specific components

111 of serum lipid that should be checked in patients with newly diagnosed XP is not

112 currently available. Hence, we performed a systematic review and meta-analysis of

113 the currently available literature to elucidate associations between components of

114 serum lipids and risk of atherosclerosis in patients with XP.

 7

115

116 METHODS

117 A systematic review and meta-analysis were conducted in accordance with

118 guidelines for Meta-Analyses and Systematic Reviews of Observational Studies.8

119 Search strategy and study selection

120 A systematic search was performed by H.C. Chang for eligible studies published

121 before April 15, 2019 from the databases of PubMed, Web of Science, Embase, and

122 Cochrane Library. Keywords for search included “xanthelasma” combined with

123 “lipid,” “apolipoprotein,” or “atherosclerosis.” Manual searches were also performed

124 to identify from the reference lists of articles that were relevant, and studies published

125 in languages other than English were considered after appropriate translation. Articles

126 were included if they were case–control studies, involved patients with XP but not

127 other types of xanthomas, and had definite values of each major investigating target.

128 Studies were excluded if they did not meet inclusion criteria; were reviews, case

129 reports or case series, or conference abstracts; had poorly defined case or control

130 groups; or had study populations focusing on atypical ages or disease statuses.

131 Data extraction and quality assessment

132 Two authors (H.C. Chang and C.W. Sung) extracted data from eligible studies

133 individually by using a standardized data extraction protocol, and a senior author
 8

134 (M.H. Lin) finalized the data. The data extracted from eligible studies included

135 authors, publication year, country, sex, age, numbers of cases and controls, level of

136 serum lipoprotein/apo (TC, LDLs, HDLs, VLDLs, TG, apo B, and apo A1), and

137 carotid intima-media thickness (CIMT).

138 Two authors (H.C. Chang and C.W. Sung) assessed the qualities of each study

139 independently by using the Newcastle–Ottawa Scale (NOS) for case–control studies.

140 The total scores of NOS were from 0 to 9, and total scores were categorized into three

141 groups, namely very high risk of bias (0 to 3), high risk of bias (4 to 6), and low risk

142 of bias (7 to 9).9 Discrepancies were reassessed through objective discussion, and

143 decisions were finalized by the senior author (M.H. Lin).

144 Data synthesis and statistical analysis

145 The focus of current analysis was to compare the serum levels of various lipids

146 and markers for atherosclerosis, including serum levels of apo A1 and apo B, as well

147 as CIMT between the patients with XP (cases) and population without XP (controls).

148 We also conducted a subgroup analysis to explore the relationship between CIMT and

149 the patients with XP in different statuses of serum cholesterol (hyperlipidemia,

150 normolipidemia, or mixed) compared with the controls.

151 Comprehensive Meta-Analysis Version 3.0 (Biostat, Englewood, NJ, USA) was

152 used for the meta-analysis. A random-effects model was used to generate pooled
 9

153 estimates. All data are expressed as the standard mean difference (SMD) with 95%

154 confidence interval (CI). The SMD is used as a summary statistic in meta-analysis

155 when the all studies assess the same outcome but measure it in a variety of scales or

156 standards. Heterogeneity across studies was evaluated using the Χ2 statistic and

157 quantified using the I2 statistic. Publication bias was evaluated using Egger’s test. P

158 values of <0.05 were regarded statistically significant in all analyses.

159

160 RESULTS

161 Studies for the meta-analysis

162 Initially, we identified 229 potentially relevant articles of which 214 articles were

163 excluded according to the four-step process of identification, screening, eligibility,

164 and inclusion to select studies for meta-analysis. Finally, 15 case–control studies

165 (published in 1981–2016) with a total 854 patients with XP and 907 healthy controls

166 were included in the quantitative analysis after sequential exclusion (Fig 1).10-24 Table

167 I shows the characteristics of these studies. The NOS scores of all included studies

168 were 7–8, which indicated that these studies had a low risk of bias. The mean age of

169 the patients with XP was 40 to 60 years, and the female-to-male ratio of the patients

170 with XP was approximately 3 to 1 within these included studies.

171 Association between serum lipid profiles and XP

 10

172 Fig 2 shows the comparison of serum levels of TC, LDLs, HDLs, VLDLs, and TG

173 between the patients with XP and the controls. Compared with the controls, pooled

174 analysis indicated that the patients with XP had significantly higher levels of TC

175 (SMD, 0.612; 95% CI, 0.376 to 0.848; P < 0.001) (Fig 2A) and LDLs (SMD, 0.587;

176 95% CI, 0.339 to 0.836; P < 0.001) (Fig 2B). Moreover, even in the normolipidemic

177 population, the patients with XP also showed significantly higher serum levels of TC

178 (SMD, 0.528; 95% CI, 0.058 to 0.997; P = 0.028) and LDLs (SMD, 0.415; 95% CI,

179 0.007 to 0.823; P = 0.046) than did the controls. Notably, trivial variations were

180 observed in the cutoff for the definition of hyperlipidemia in these studies.10, 11, 15, 17, 21

181 No significant differences were noted in the serum levels of HDLs (SMD, –0.135;

182 95% CI, –0.345 to 0.075; P = 0.207), VLDLs (SMD, 0.288; 95% CI, –0.010 to 0.585;

183 P = 0.058), and TG (SMD, 0.259; 95% CI, –0.019 to 0.537; P = 0.068) (Fig 2C–2E)

184 between the patients with XP and the controls. Notable heterogeneity across the

185 included studies was detected in each analysis (Fig 2).

186 Risk of atherosclerosis in the patients with XP

187 As mentioned, apo B and apo A1 play opposite roles in the pathophysiology of

188 atherosclerosis. Hence, we compared the serum levels of apo B and apo A1 as serum

189 risk markers of atherosclerosis between the patients with XP and controls. The serum

190 levels of apo B in the patients with XP were significantly higher than those in the
 11

191 controls (SMD, 1.036; 95% CI, 0.361 to 1.711; P = 0.003) (Fig 3A), and comparable

192 levels of apo A1 were observed in the two groups (SMD, –0.328; 95% CI, –0.704 to

193 0.048; P = 0.087) (Fig 3B). Similarly, significant heterogeneity across the included

194 studies was also noted in the analyses (Fig 3).

195 CIMT is considered a surrogate marker of atherosclerosis and CVDs risk.25 Fig 4

196 demonstrated the comparison of CIMT between the patients with XP (n = 138) and

197 healthy controls (n = 186). The patients exhibited XP had significantly higher CIMT

198 than did the controls (SMD, 1.848; 95% CI, 0.639 to 3.057; P = 0.003), but we could

199 not find similar significant results in the subgroup analysis with only the

200 hyperlipidemic and normolipidemic groups.

201 Sensitivity analysis and publication bias

202 A leave-one-out sensitivity analysis indicated the robustness of all of our

203 significant findings. For evaluation of publication bias, Egger’s test showed no

204 significant publication bias between the included studies in the analyses of the serum

205 levels of TC (P = 0.162) and Apo B (P = 0.258).

206

207 DISCUSSION

208 Our pooled analysis showed a significantly higher level of serum TC in the

209 patients with XP than in the controls, but this difference was not observed in the TG
 12

210 levels. Only the LDLs level and not the HDLs and VLDLs levels were significantly

211 higher in hyperlipidemic and normolipidemic patients with XP than in the controls.

212 These results provide more specific items for survey of dyslipidemia in the patients

213 with XP during clinical practice. In the aspect of predictors for atherosclerotic CVDs,

214 our meta-analyses revealed that the patients with XP had significantly higher serum

215 levels of apo B and greater CIMT than those in control groups. Although no statistical

216 significance was detected in the pooled analyses of TG, HDLs, VLDLs, and apo A1,

217 trends toward the pro-atherosclerotic changes of these parameters were consistent.

218 The borderline non-significant results in certain analyses may result from relative

219 lower numbers of included studies compared with those with significant findings.

220 Clinically, the cutaneous xanthomas are divided into eruptive, tuberous,

221 tuberoeruptive, tendinous, and planar forms. Plane xanthomas can be further

222 classified based on their anatomical distribution into intertriginous plane xanthomas,

223 xanthoma striatum palmare, xanthoma diffusum planum, and XP.26 Different

224 xanthomas share some pathological features with abnormally excessive lipid deposits

225 in the skin or tendons; however, they do not belong to a single disease entity

226 themselves and are clinical signs from various lipoprotein disorders or monoclonal

227 gammopathy.27 The presence of familial hyperlipoproteinemia is greater in eruptive,

228 tuberous, and tendinous xanthoma than in XP,26, 27

and the coexisting tendinous
 13

229 xanthomas and familial hyperlipoproteinemia were associated with a 3.20-fold higher

230 risk of CVDs.28 Generally, patients who develop XP before the age of 25 years should

231 be aware of suspicious familial hypercholesterolemia.29 Among the 15 included

232 studies in current pooled analyses, only 3 studies definitely showed the enrollment of

233 population with positive family history of hyperlipidemia,10, 21, 23 but we found no

234 major impact on conclusion of each pooled analysis after exclusion of these studies.

235 The mean age of most our included patients was approximately 40 to 60 years; hence,

236 secondary etiology of hyperlipidemia should be surveyed in patients with XP.

237 Ultrasound CIMT is one of the most widely used and validated imaging

238 techniques for atherosclerosis. It has advantages of noninvasiveness and visualizing of

239 the final consequence of atherosclerosis in the vessel wall directly.30 The CIMT is

240 measured using B-mode ultrasound between the intimal-luminal and the

241 medial-adventitial interfaces of the carotid artery.31 All of the included studies in our

242 pooled analysis utilized the values of maximal CIMT in the common carotid artery,

243 which was at least 1 cm proximal to the bifurcation.16, 17, 20, 21 Other studies have

244 indicate that the level of apo B is significantly positively associated with values of

245 CIMT.32 Consistent with this finding, we also found patients with XP have

246 significantly higher apo B and CIMT compared with controls.

247 One prospective cohort study of 35-year observation by Christoffersen et al.33

 14

248 enrolling 12745 people including 563 patients with XP also showed higher incidence

249 of atherosclerotic diseases in patient with XP compared with control. Similarly, this

250 study revealed significant higher levels of TC, LDLs and apo B in patients with XP

251 then those without the traits. Nonetheless, they indicated the XP is the independent

252 risk factor of atherosclerotic CVDs after multifactorial adjustments. Even though the

253 XP itself was considered an independent risk factor of CVDs prediction, dyslipidemia

254 is still the major controllable risk factor for prevention of CVDs. It is noteworthy that

255 this cohort study enrolled only Danes population, but our studies provide results of

256 pooled analysis of patients with XP from different ethnic groups.

257 In addition to dyslipidemia, both hypertension and diabetes mellitus are

258 well-established risk factors of CVDs.34 Four studies included in our final

259 meta-analysis also showed the prevalence of hypertension and diabetes mellitus in the

260 XP patients and controls, and the prevalence of hypertension and diabetes mellitus in

261 both the groups were not significantly different in each study.13, 17, 18, 35 Furthermore,

262 we also found no significant difference of comparison of prevalence of hypertension

263 (P = 0.086) and diabetes mellitus (P = 0.704) when we used pooled analysis between

264 the patients with XP and controls. On the basis of our findings, dyslipidemia would be

265 one significant modifiable CVDs risk factor in patients with XP.

266 Our study has some limitations. First, we could not perform in-depth analyses
 15

267 based on patient characteristics, such as sex and ethnicity, because of limited data of

268 subclasses from original studies. Second, data of included studies were insufficient for

269 estimating the true incidence or prevalence of stroke or ischemic heart diseases in the

270 patients with XP. Conversely, we could only use indirect indexes, such as serum apo

271 B and CIMT, for evaluating the risk of atherosclerotic CVDs in the patients with XP.

272 Third, additional studies are warranted to compensate for the low numbers of

273 available studies in meta-analysis of apos and each subgroup analysis for evaluating

274 the relationship between CIMT and the patients with XP. Finally, substantial

275 heterogeneity across the included studies was observed. The heterogeneity is

276 speculated from differences in experimental methods, varing disease status of XP,

277 diverse ethnicity, and other uncontrolled confoundings. We attempted to resolve this

278 issue by using a random-effects model with SMD rather than mean difference to

279 calculate the main analyses. We also performed the sensitivity test to prove the

280 robustness of our pooled effect estimates.

281 In conclusion, our results revealed that significantly higher serum TC levels,

282 particularly LDLs levels, were observed in the patients with XP than in the controls,

283 and the XP patients also exhibit a significantly high risk of atherosclerosis because of

284 higher levels of serum apo B and CIMT. These findings offer more definite evidence

285 that patients with XP are highly associated with dyslipidemia and risk of
 16

286 atherosclerosis. Additional careful monitoring and targeted intervention is required for

287 patients with XP beyond cosmetic concerns.

288 Abbreviation and acronym list

289 apos, apolipoproteins

290 CI, confidence interval

291 CIMT, carotid intima-media thickness

292 CVDs, cardiovascular diseases

293 HDLs, high-density lipoproteins

294 IDLs, intermediate-density lipoproteins

295 LDLs, low-density lipoproteins

296 NOS, Newcastle–Ottawa Scale

297 SMD, standard mean difference

298 TG, triglyceride

299 VLDLs, very-low-density lipoproteins

300 XP, xanthelasma palpebrarum

301

302

303

304
 17

305

306

307 REFERENCES

308 1. Verges B. Pathophysiology of diabetic dyslipidaemia: where are we? Diabetologia

309 2015;58:886-899.

310 2. Toth PP. Triglyceride-rich lipoproteins as a causal factor for cardiovascular disease.

311 Vasc Health Risk Manag 2016;12:171-183.

312 3. Helkin A, Stein JJ, Lin S, Siddiqui S, Maier KG , Gahtan V. Dyslipidemia Part

313 1--Review of Lipid Metabolism and Vascular Cell Physiology. Vasc Endovascular

314 Surg 2016;50:107-118.

315 4. Walldius G , Jungner I. Apolipoprotein B and apolipoprotein A-I: risk indicators of

316 coronary heart disease and targets for lipid-modifying therapy. J Intern Med

317 2004;255:188-205.

318 5. Walldius G , Jungner I. The apoB/apoA-I ratio: a strong, new risk factor for

319 cardiovascular disease and a target for lipid-lowering therapy--a review of the

320 evidence. J Intern Med 2006;259:493-519.

321 6. Nair PA , Singhal R. Xanthelasma palpebrarum - a brief review. Clin Cosmet

322 Investig Dermatol 2018;11:1-5.

323 7. Laftah Z , Al-Niaimi F. Xanthelasma: An Update on Treatment Modalities. J Cutan

 18

324 Aesthet Surg 2018;11:1-6.

325 8. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D et al.

326 Meta-analysis of observational studies in epidemiology: a proposal for reporting.

327 Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. Jama

328 2000;283:2008-2012.

329 9. Lo CK, Mertz D , Loeb M. Newcastle-Ottawa Scale: comparing reviewers' to

330 authors' assessments. BMC Med Res Methodol 2014;14:45.

331 10. Watanabe A, Yoshimura A, Wakasugi T, Tatami R, Ueda K, Ueda R et al. Serum

332 lipids, lipoprotein lipids and coronary heart disease in patients with xanthelasma

333 palpebrarum. Atherosclerosis 1981;38:283-290.

334 11. Gomez JA, Gonzalez MJ, de Moragas JM, Serrat J, Gonzalez-Sastre F , Perez M.

335 Apolipoprotein E phenotypes, lipoprotein composition, and xanthelasmas. Arch

336 Dermatol 1988;124:1230-1234.

337 12. Pinto X, Ribera M , Fiol C. Dyslipoproteinemia in patients with xanthelasma.

338 Arch Dermatol 1989;125:1281-1282.

339 13. Ribera M, Pinto X, Argimon JM, Fiol C, Pujol R , Ferrandiz C. Lipid metabolism

340 and apolipoprotein E phenotypes in patients with xanthelasma. Am J Med

341 1995;99:485-490.

342 14. Sharma SB, Dwivedi S, Prabhu KM, Kumar N , Baruah MC. Preliminary studies
 19

343 on serum lipids, apolipoprotein-B and oxidative stress in xanthelasma. Indian J Clin

344 Biochem 1999;14:245-248.

345 15. Tursen U, Eskandari G, Kaya TI, Tamer L, Ikizoglu G , Atik U. Apolipoprotein E

346 polymorphism and lipoprotein compositions in normolipidaemic xanthelasma patients.

347 J Eur Acad Dermatol Venereol 2006;20:260-263.

348 16. Noel B. Premature atherosclerosis in patients with xanthelasma. J Eur Acad

349 Dermatol Venereol 2007;21:1244-1248.

350 17. Chan CC, Lin SJ, Hwang JJ, Sun CC, Jeng JS, Hwang BS et al. Xanthelasma is

351 not associated with increased risk of carotid atherosclerosis in normolipidaemia. Int J

352 Clin Pract 2008;62:221-227.

353 18. Ozdol S, Sahin S , Tokgozoglu L. Xanthelasma palpebrarum and its relation to

354 atherosclerotic risk factors and lipoprotein (a). Int J Dermatol 2008;47:785-789.

355 19. Nakazone MA, De Marchi MA, Pinhel MA, Barros CF, Julio MA, Pinheiro A et al.

356 Effects of APOE, APOB and LDLR variants on serum lipids and lack of association

357 with xanthelasma in individuals from Southeastern Brazil. Genet Mol Biol

358 2009;32:227-233.

359 20. Pandhi D, Gupta P, Singal A, Tondon A, Sharma S , Madhu SV. Xanthelasma

360 palpebrarum: a marker of premature atherosclerosis (risk of atherosclerosis in

361 xanthelasma). Postgrad Med J 2012;88:198-204.

 20

362 21. Esmat S, Abdel-Halim MR, Fawzy MM, Nassef S, Esmat S, Ramzy T et al. Are

363 normolipidaemic patients with xanthelasma prone to atherosclerosis? Clin Exp

364 Dermatol 2015;40:373-378.

365 22. Akyuz AR, Agac MT, Turan T, Sahin S, Kul S, Korkmaz L et al. Xanthelasma Is

366 Associated with an Increased Amount of Epicardial Adipose Tissue. Medical

367 Principles and Practice 2016;25:187-190.

368 23. Kavoussi H, Ebrahimi A, Rezaei M, Ramezani M, Najafi B , Kavoussi R. Serum

369 lipid profile and clinical characteristics of patients with xanthelasma palpebrarum. An

370 Bras Dermatol 2016;91:468-471.

371 24. Nair PA, Patel CR, Ganjiwale JD, Diwan NG , Jivani NB. Xanthelasma

372 Palpebrarum with Arcus Cornea: A Clinical and Biochemical Study. Indian J

373 Dermatol 2016;61:295-300.

374 25. Ravani A, Werba JP, Frigerio B, Sansaro D, Amato M, Tremoli E et al.

375 Assessment and relevance of carotid intima-media thickness (C-IMT) in primary and

376 secondary cardiovascular prevention. Curr Pharm Des 2015;21:1164-1171.

377 26. Zak A, Zeman M, Slaby A , Vecka M. Xanthomas: clinical and pathophysiological

378 relations. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub

379 2014;158:181-188.

380 27. Bansal M, Manchanda K , Pandey S. Familial type IIa hyperlipoproteinemia

 21

381 presenting with different types of xanthomas. Indian Journal of Paediatric

382 Dermatology 2014;15:39-41.

383 28. Oosterveer DM, Versmissen J, Yazdanpanah M, Hamza TH , Sijbrands EJ.

384 Differences in characteristics and risk of cardiovascular disease in familial

385 hypercholesterolemia patients with and without tendon xanthomas: a systematic

386 review and meta-analysis. Atherosclerosis 2009;207:311-317.

387 29. Mytilinaiou M, Kyrou I, Khan M, Grammatopoulos DK , Randeva HS. Familial

388 Hypercholesterolemia: New Horizons for Diagnosis and Effective Management.

389 Front Pharmacol 2018;9:707.

390 30. Centurion OA. Carotid Intima-Media Thickness as a Cardiovascular Risk Factor

391 and Imaging Pathway of Atherosclerosis. Crit Pathw Cardiol 2016;15:152-160.

392 31. Nezu T, Hosomi N, Aoki S , Matsumoto M. Carotid Intima-Media Thickness for

393 Atherosclerosis. J Atheroscler Thromb 2016;23:18-31.

394 32. Qu B , Qu T. Causes of changes in carotid intima-media thickness: a literature

395 review. Cardiovasc Ultrasound 2015;13:46.

396 33. Christoffersen M, Frikke-Schmidt R, Schnohr P, Jensen GB, Nordestgaard BG ,

397 Tybjaerg-Hansen A. Xanthelasmata, arcus corneae, and ischaemic vascular disease

398 and death in general population: prospective cohort study. Bmj 2011;343:d5497.

399 34. Arnett DK, Blumenthal RS, Albert MA, Buroker AB, Goldberger ZD, Hahn EJ et
 22

400 al. 2019 ACC/AHA Guideline on the Primary Prevention of Cardiovascular Disease:

401 Executive Summary. Circulation 2019:Cir0000000000000677.

402 35. Akyuz AR, Agac MT, Turan T, Sahin S, Kul S, Korkmaz L et al. Xanthelasma Is

403 Associated with an Increased Amount of Epicardial Adipose Tissue. Med Princ Pract

404 2016;25:187-190.

405

406

407

408

409

410

411

412

413

414

415

416

417

418
 23

419

420

421 Figure legends

422 Fig 1. PRISMA flowchart of study selection.

423 Fig 2. Comparison of serum levels of (A) total cholesterol, (B) low-density

424 lipoproteins, (C) high-density lipoproteins, (D) very low-density lipoproteins, and (E)

425 triglycerides between the patients with xanthelasma palpebrarum and healthy controls

426 Fig 3. Comparison of serum levels of (A) apolipoprotein B and (B) apolipoprotein A1

427 between the patients with xanthelasma palpebrarum and healthy controls

428 Fig 4. Meta-analysis with subgroup analysis of carotid intima-media thickness

429 between the patients with xanthelasma palpebrarum and healthy controls

430

431

432

433

434

435

436

437
 24

438
 25

Table I Basic characteristics of studies included in the meta-analysis

Study Country Group (n) Age, mean ± Sex TCa. LDLsa. HDLsa. VLDLsa. TGa. apo Ba. apo A1a. NOSc
.
SD or range, yr (M:F)
Watanabe et al., Japan XP (53) 56.8 ± 10.1 21:32 237.94 ± 65.99 168.89 ± 62.73 42.15 ± 12.70 22.92 ± 19.67 132.85 ± 81.73 NA NA 7

198110 Control (40) 55.5 ± 9.4 20:20 172.00 ± 21.50 96.50 ± 23.83 58.50 ± 11.90 9.50 ± 4.50 83.50 ± 24.49

Gómez et al., Spain XP (50) 48.2 ± 10.6 15:35 232.79 ± 44.86 153.52 ± 41.65 58.00 ± 18.95 14.69 ± 10.44 126.66 ± 57.57 88.50 ± 19.44 140.38 ± 25.20 8

198811 Control (50) matched matched 230.86 ± 40.99 155.07 ± 38.67 54.14 ± 11.60 18.95 ± 16.24 132.86 ± 57.57 102.64 ± 29.21 139.78 ± 22.50

Pinto et al., Spain XP (96) 52.3 ± 11.4 25:71 252.13 ± 52.20 182.91 ± 50.27 54.14 ± 13.53 16.63 ± 9.28 111.60 ± 70.86 120.30 ± 32.90 149.49 ± 29.27 7
12
1989 Control (104) 50.7 ± 10.6 56:48 218.10 ± 46.40 97.06 ± 67.67 55.68 ± 14.69 14.69 ± 11.60 102.75 ± 53.14 102.00 ± 24.82 149.77 ± 36.00

Ribera et al., Spain XP (115) 51.7 ± 10.8 31:84 253.29 ± 49.50 182.52 ± 49.50 52.98 ± 13.92 18.56 ± 23.20 109.83 ± 59.34 121.60 ± 31.99 148.97 ± 31.64 7
13
1995 Control (105) 50.3 ± 11.0 56:49 219.64 ± 47.56 147.72 ± 43.70 54.14 ± 15.47 15.47 ±13.92 118.69 ± 152.35 103.81 ± 25.53 151.24 ± 36.67

Sharma et al., India XP (12) NA NA 220.07 ± 37.43 142.40 ± 36.66 42.73 ± 7.55 33.73 ± 8.71 168.20 ± 40.22 118.33 ± 12.92 NA 8
14
1999 Control (6) matched matched 172.00 ± 36.00 112.00 ± 35.60 43.00 ± 8.20 16.90 ± 6.40 84.6 ± 32.0 79.00 ± 9.40

Tursen et al., Turkey XP (25) 54 ± 7 10:15 206 ± 45 126 ± 42 48 ± 13 NA 149 ± 88 135 ± 18 113 ± 29b. 7

200615 Control (27) 55 ± 6 NA 182 ± 28 113 ± 26 47 ± 15 150 ± 45 96 ± 22 153 ± 29b.

Noël, Switzerland XP (17) 45.7 ± 9.7 5:12 212.68 ± 50.27 139.21 ± 50.27 46.4 ± 11.6 NA 115.15 ± 70.86 NA NA 8

200716 Control (21) 49.1 ± 15.9 9:12 189.48 ± 30.94 116.01 ± 3.87 46.4 ± 15.47 150.58 ± 79.72

Chan et al., Taiwan XP (63) 50.78 ± 11.51 9:54 232.79 ± 70.38 147.33 ± 54.14 51.82 ± 13.92 NA NA NA NA 8

200817 Control (120) 50.06 ± 10.41 22:98 204.56 ±29.78 116.01 ± 27.07 54.91 ± 13.15

Ozdöl et al., Turkey XP (100) 49 ± 11 34:66 210 ± 40 134 ± 58 56 ± 15 NA 131 ± 99 NA NA 8

200818 Control (100) 48 ± 11 40:60 194 ± 42 112 ± 35 53 ± 13 141 ± 95

Nakazone et al., Brazil XP (100) 46.2 ± 13.6 37:63 203.6 ± 26.9 126.3 ± 26.3 46.9 ± 8.3 30.0 ± 8.4 149.8 ± 41.8 NA NA 8

200919 Control (100) 44.0 ± 12.6 36:64 206.1 ± 32.7 126.3 ± 35.6 46.4 ± 7.6 32.7 ± 6.6 163.4 ± 33.0

Pandhi et al., India XP (40) 42.9 ± 10.99 9:31 180.95 ± 39.30 110.75 ± 29.33 40.18 ± 5.92 21.73 ± 7.25 115.45 ± 50.78 126.55 ± 35.40 117.83 ± 20.61 8
20
2012 Control (40) matched matched 179.15 ± 26.63 112.23 ± 19.69 39.55 ± 4.42 20.28 ± 3.38 98.25 ± 15.48 88.45 ± 12.39 127.48 ± 14.31

Esmat et al., Egypt XP (40) 40.73 ± 5.10 14:26 236.90 ± 47.48 120.13 ± 26.65 37.43 ± 10.37 NA 158.52 ± 67.01 NA NA 8
 26

201521 Control (40) 40.43 ± 5.19 18:22 160.67 ± 24.04 99.79 ± 26.24 45.39 ± 7.15 79.52 ± 22.42

Akyüz et al., Turkey XP (52) 51 ± 10 12:40 216 ± 54 142 ± 45 49 ± 13 NA NA NA NA 8

201622 Control (52) 50 ± 18 12:40 181 ± 42 115 ± 36 53 ± 15

Kavoussi et al., Iran XP (42) 47.4 ± 9.97 6:36 221.51 ± 60.40 120.30 ± 43.60 NA 37.70 ± 17.60 185.98 ± 71.70 NA NA 8

201623 Control (42) 48.1 ± 10.05 8:34 198.82 ± 34.80 110.60 ± 26.70 30.10 ± 12.46 149.39 ± 94.20

Nair et al., India XP (49) Most in 30-70s 9:40 210.57 ± 62.92 142.79 ± 80.99 53.20 ± 18.25 30.95 ± 20.52 123.06 ± 50.64 NA NA 7

201624 Control (60) matched 21:39 186.45 ± 38.51 110.57 ± 35.16 49.51 ± 10.61 23.81 ± 11.65 116.07 ± 58.12

apo, apolipoprotein; HDLs, high-density lipoproteins; LDLs, low-density lipoproteins; NA, not available; TC, total cholesterol; TG, triglyceride; VLDLs,
very-low-density lipoproteins; XP, xanthelasma palpebrarum
a.
Values (mg/dL) of TC, LDLs, HDLs, TG, apo B, and apo A1 are expressed as mean ± standard deviation (SD)
b.
Data of apo A
c.
Newcastle–Ottawa Scale (NOS): total scores (0–9)

439
Capsule summary

Patient with xanthelasma palpebrarum are associated with significantly higher

levels of serum atherogenic lipoproteins and risk of atherosclerosis

Clinician should monitor and manage the dyslipidemia in patients with

xanthelasma palpebrarum beyond cosmetic concerns