Lipoprotein (a) levels are elevated in psoriasis: An updated systematic review and meta-analysis

Abstract

BACKGROUND

Psoriasis is a chronic inflammatory condition related to an increased atherosclerotic cardiovascular disease (ASCVD) risk.

AIM

To investigate whether lipoprotein (a) [Lp(a)] levels are increased in patients with psoriasis.

METHODS

A comprehensive literature search up to January 30, 2025 was conducted utilizing PubMed and Cochrane Library databases. A qualitative synthesis and a meta-analysis on Lp(a) mean differences (MD) between psoriasis cases and healthy controls (HC) was performed. The protocol of this meta-analysis has been registered to PROSPERO (No. CRD420250652465).

RESULTS

Eighteen studies with 1650 psoriasis patients and 1621 HC were eligible for qualitative synthesis. Pooled analysis from 16 studies (1401 psoriasis patients and 1320 HC) demonstrated that psoriasis patients had significantly higher Lp(a) levels compared with the HC group (MD: 6.72 mg/dL, 95%CI: 4.32-9.12, P < 0.00001, I² = 71%). Sensitivity analyses according to the region of origin was also performed. The pooled analysis of the European sub-population showed a pronounced increase in Lp(a) levels in 189 patients with psoriasis vs 178 HC (MD: 15.86 mg/dL, 95%CI: 5.79-25.92, P < 0.002, I² = 79%), while the pooled analysis on the Asian sub-population demonstrated a smaller but still significant difference in Lp(a) levels between 1177 psoriasis patients and 1127 HC (MD: 4.95 mg/dL, 95%CI: 2.99-6.92, P < 0.00001, I² = 58%).

CONCLUSION

Our findings suggest that Lp(a) levels are significantly elevated in psoriasis patients, further adding to their ASCVD risk.

Key Words: Psoriasis; Lipoprotein (a); Lipid profile; Atherosclerosis; Cardiovascular disease

Core Tip: Elevated lipoprotein (a) [Lp(a)] levels represent an atherosclerotic cardiovascular disease risk (ASCVD) factor. The findings of the present systematic review and meta-analysis show a significant increase of Lp(a) levels in psoriasis patients compared with healthy controls. As such, Lp(a) measurement in patients with psoriasis may be of use for a more precise determination of ASCVD risk and treatment goals.

INTRODUCTION

Psoriasis is a chronic systemic inflammatory skin disease, affecting more than 60 million people worldwide[1]. Central to its etiopathogenesis is an interplay between the innate and adaptive immune systems, while heritability is the major contributor in disease development[2]. Psoriasis-related persistent inflammation is associated with oxidative stress, vascular dysfunction, and lipid abnormalities, all of which are linked to atherosclerotic cardiovascular disease (ASCVD)[3]. Simultaneously, people with psoriasis commonly exhibit ASCVD risk factors including metabolic syndrome (i.e., central obesity, hypertension, glucose resistance, hypertriglyceridemia, and hypoalphalipoproteinemia), diabetes, metabolic-associated steatotic liver disease, and dyslipidemia[4-7].

Lipoprotein (a) [Lp(a)] is a low-density lipoprotein (LDL)-like particle synthesized in the liver, whose pathognomic constituent is apolipoprotein (apo) (a). Lp(a) is mostly genetically determined, following an autosomal dominant pattern, and is hardly influenced by lifestyle factors[8]. It possesses pro-inflammatory, pro-atherogenic and pro-thrombotic properties, and is associated with elevated ASCVD risk even in people with an otherwise normal lipid profile[9-11]. Lp(a) levels can be affected by chronic inflammatory conditions, including psoriasis, largely via the effect of the interleukin (IL)-6 axis on the expression of the LPA gene and other inflammation pathways. Consequently, biologic treatments have shown promise in lowering Lp(a) levels[8].

The aim of the present systematic review and meta-analysis is the updated synthesis of published data regarding Lp(a) levels in patients with psoriasis, incorporating all recent evidence.

MATERIALS AND METHODS

Study design

We performed a qualitative and quantitative synthesis of eligible studies to evaluate whether Lp(a) levels are increased in patients with psoriasis.

Search strategy

An extensive search on PubMed and Cochrane Library databases was conducted until January 30, 2025 using the following search terms: [Lipoprotein (a) OR lp(a)] AND [(psoriasis) OR (psoriatic arthritis)]. The protocol of this systematic review was registered in PROSPERO (No. CRD420250652465) and adheres to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement (PRISMA Statement, Ottawa, ON, Canada)[12].

Eligibility criteria

Eligibility criteria for this systematic review included prospective and retrospective observational studies (case control or cohort) that compared adult patients with psoriasis with a control group. The eligibility criteria were defined using the population, intervention, comparators/controls, outcomes, and study design framework, as follows: (1) Population: Patients with psoriasis (including psoriatic arthritis); (2) Intervention: Measurement of Lp(a) levels in psoriasis patients vs controls; (3) Comparator: Healthy controls (HC); (4) Outcome: Lp(a) mean difference between psoriasis patients and HC; and (5) Study design: Observational studies, prospective and retrospective.

Data extraction

Upon deduplication, two independent authors (Fragkouli MR and Agouridis AP) screened and identified records by abstract, title, or both. Relevant full-texts were retrieved and assessed for eligibility based on the aforementioned criteria. Consensus was reached after discussion between reviewers. The following data were collected from eligible studies: (1) Authors; (2) Year of publication; (3) Country; (4) Study design; (5) Population characteristics; and (6) Lp(a) values of psoriasis patients and HC.

Methodological assessment of included studies

The Newcastle-Ottawa Scale (NOS) was applied for each study in order to perform a quality assessment of the included studies[13]. Three primary factors are evaluated by the NOS: (1) Selection; (2) Comparability; and (3) Exposure or outcome. For a more thorough evaluation, these categories are further divided, depending on study design. The highest possible score for each study is 9 points. Good quality studies are indicated by scores of 7-9, while studies scoring 4-6 and 0-3 are at moderate and high risk for bias, respectively.

Statistical analysis

A meta-analysis was also carried out to enable a quantitative assessment of the included studies. Continuous variables were used to represent outcome estimates. For continuous data, mean differences (MD) and 95%CI of Lp(a) levels among patients with psoriasis and HC were computed. In addition, a sensitivity analysis according to place of origin was performed among the European and Asian populations. Review Manager version 5.0 software (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark, 2008) was used for statistical analysis; a P value of less than 0.05 is deemed statistically significant.

Heterogeneity analysis

Regarding heterogeneity analysis, the I² test was used to determine whether statistical heterogeneity exists among the included studies. The heterogeneity was categorized as low, moderate, or high, if the I² values were 25%, 50%, or more than 75%, respectively. A random effects model was employed when the P value was less than 0.10. The Q test and the I² statistic were used to assess inter-trial heterogeneity.

RESULTS

Study selection

A PRISMA chart was created to summarize the literature search (Figure 1). Upon database search, 112 records were retrieved. Of these, 89 records were eliminated after title and abstract assessment. Following full-text review, 4 more studies were found ineligible for inclusion. Finally, 18 records[14-31] were included in the qualitative synthesis, while 16 were used for quantitative synthesis purposes.

Figure 1  Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart of the systematic review.

Study characteristics

Study characteristics are presented in Table 1[14-31]. All of the 18 included studies were observational (cohort and case control) and performed between 1994 and 2022. A total of 3271 participants were included, of whom 1650 were identified as cases and 1621 were identified as HC. The studies were conducted in the following countries: (1) United States (1 study); (2) India (2 studies); (3) United Kingdom (1 study); (4) Turkey (2 studies); (5) China (2 studies); (6) Portugal (2 studies); (7) Italy (2 studies); (8) Iran (3 studies); (9) Japan (1 study); (10) Poland (1 study); and (11) Greece (1 study). Included studies may be further classified into two distinct groups: European population (5 studies) vs Asian population (10 studies). Three studies[16,21,24] focused on patients with psoriatic arthritis. Reported comorbidities included smoking, alcohol consumption, dyslipidemia and obesity. The mean age of all participants was 42 years, and the male to female ratio was similar between groups, except for two studies[14,18] which enrolled only male participants. Lp(a) levels of both psoriasis cases and HC were available in all 18 studies (Table 2)[14-31].

Table 1 Characteristics of eligible studies.

Ref.	Year	Country	Study design	Population	Comorbidities	Psoriatic patients mean age (mean ± SD)	Controls mean age (mean ± SD)	Age and sex-matched controls
Seçkin et al[14]	1994	Turkey	Cohort	Psoriasis	Obesity, hypertension, smoking	36 ± 13	36 ± 15	Male only
Jones et al[16]	2000	United Kingdom	Case-control	Psoriatic arthritis	Gout	46 ± 13	46 ± 13	Yes
Rocha-Pereira et al[15]	2001	Portugal	Case-control	Psoriasis	-	47 ± 12	47 ± 13	Yes
Uyanik et al[17]	2002	Turkey	Case-control	Psoriasis	-	38 ± 5	-	Yes
Pietrzak et al[18]	2009	Poland	Cross-sectional	Psoriasis	-	37 ± 13	34 ± 12	Male only
Coimbra et al[19]	2010	Portugal	Cohort	Psoriasis	Smoking	43 ± 15	47 ± 15	Yes
Asefi et al[23]	2012	Iran	Case-control	Psoriasis	-	35 ± 11	36 ± 13	Yes
Ferretti et al[20]	2012	Italy	Case-control	Psoriasis	-	47 ± 13	41 ± 8	Yes
Oliviero et al[21]	2012	Italy	Case-control	Psoriatic arthritis	-	45 ± 15	-	Yes
Nemati et al[22]	2013	Iran	Cross-sectional	Psoriasis	-	35 ± 10	33 ± 11	Yes
Asefi et al[25]	2014	Iran	Case-control	Psoriasis	-	35 ± 11	36 ± 13	Yes
Papagoras et al[24]	2014	Greece	Cross-sectional	Psoriasis	Hypertension, smoking	48	49	Yes
Sunitha et al[26]	2015	India	Case-control	Psoriasis	-	45 ± 14	43 ± 11	Yes
Sorokin et al[27]	2018	Japan	Cohort	Psoriasis	Dyslipidemia	50 ± 13	40 ± 14	Yes
Wadhwa et al[28]	2019	India	Case-control	Psoriasis and psoriatic arthritis	Smoking and alcohol consumption, hypertension, diabetes, metabolic syndrome	41 ± 14	43 ± 5	Yes
Miao et al[29]	2019	China	Case-control	Psoriasis, psoriatic arthritis	Obesity, alcohol consumption, smoking, hypertension, hyperlipidemia, diabetes	52 ± 14	50 ± 14	Yes
Garshick et al[31]	2021	United States	Case-control	Psoriasis	Obesity	44 ± 14	-	Yes
Wang et al[30]	2022	China	Case-control	Psoriasis and psoriatic arthritis	Obesity, smoking and alcohol consumption, hypertension, diabetes, metabolic syndrome	NA	NA	No

NA: Not available.

Table 2 Lipoprotein (a) levels in psoriasis patients and healthy controls.

Ref.	Year	Psoriasis patients (n)	Psoriasis Lp(a) (mg/dL)	Controls (n)	Controls Lp(a) (mg/dL)	P value
Seçkin et al[14]	1994	32	13.8 ± 11.8	13	7.5 ± 8.9	NS
Jones et al[16]	2000	50	23.0 ± 20.1	50	21.8 ± 20.3	NS
Rocha-Pereira et al[15]	2001	48	63.8 ± 40.1	40	31.8 ± 18.1	< 0.001
Uyanik et al[17]	2002	72	27.4 ± 3.6	30	19.4 ± 4.5	< 0.01
Pietrzak et al[18]	2009	34	34.2 ± 23.6	26	21.9 ± 14.4	0.029
Coimbra et al[19]	2010	34	57.0 ± 49.7	37	34.9 ± 26.4	≤ 0.01
Asefi et al[23]	2012	100	19.2 ± 16.5	100	17.1 ± 17.0	NS
Ferretti et al[20]	2012	23	29.6 ± 18.3	25	12.8 ± 2.6	< 0.001
Oliviero et al[21]	2012	14	9.7	33	5.6	NS
Nemati et al[22]	2013	90	19.2 ± 16.5	90	17.1 ± 15.1	NS
Asefi et al[25]	2014	100	19.4 ± 16.9	100	17.1 ± 17.1	NS
Papagoras et al[24]	2014	56	12.5	71	13.5	NS
Sunitha et al[26]	2015	45	28.0 ± 9.1	45	20.8 ± 7.0	< 0.001
Sorokin et al[27]	2018	232	23.1 ± 19.4	20	14.7 ± 12.7	0.004
Wadhwa et al[28]	2019	132	12.6 ± 35.9	132	4.5 ± 4.0	< 0.001
Miao et al[29]	2019	222	19.0 ± 23.0	445	15.0 ± 19.0	0.028
Garshick et al[31]	2021	35	17.0 ± 14.0	15	9.0 ± 20.0	0.02
Wang et al[30]	2022	152	14.0 ± 17.5	152	12.5 ± 15.1	NS

Lp(a): Lipoprotein (a); NS: Not significant.

Study outcomes-meta-analysis

A quantitative analysis was performed using 16 of the 18 studies, due to insufficient reporting regarding standard deviations in the studies by Oliviero et al[21] and Papagoras et al[24]. The pooled analysis showed significantly higher Lp(a) levels in 1401 patients with psoriasis compared to 1320 HC, with the overall MD being 6.72 mg/dL (95%CI: 4.32-9.12, P < 0.00001, I² = 71%) (Figure 2A)[14-20,22-23,25-31]. Furthermore, asymmetry is evident in the funnel plot, although it is mainly limited to studies with a larger standard error (Figure 2A)[14-20,22-23,25-31].

Figure 2 Forest plot. A: Forest plot of comparison: Psoriasis vs controls, outcome: Lipoprotein (a) [Lp(a)] and funnel plot of comparison: Psoriasis vs controls, outcome: Lp(a); B: Forest plot of comparison: Psoriasis vs controls, outcome: Lp(a) in European sub-population; C: Forest plot of comparison: Psoriasis vs controls, outcome: Lp(a) in Asian sub-population. Lp(a): Lipoprotein (a).

Sensitivity analysis

Additional sensitivity analyses were performed to ascertain whether region of origin affected our results. Regarding the European sub-population, the pooled analysis showed a pronounced increase in Lp(a) levels in 189 patients with psoriasis vs 178 HC, as reflected by a MD of 15.86 mg/dL (95%CI: 5.79-25.92, P < 0.002, I² = 79%). (Figure 2B)[15,16,18-20]. On the other hand, the analysis on the Asian sub-population demonstrated a smaller difference in Lp(a) levels between 1177 patients with psoriasis and 1127 HC (MD: 4.95 mg/dL, 95%CI: 2.99-6.92, P < 0.00001, I² = 58%) (Figure 2C)[14,17,22,23,25-30].

Quality appraisal

Study quality was evaluated using the NOS for case-control or cohort studies as previously described. The overall mean score of the evaluated studies was 7.7 points, which indicates high quality studies with low risk for bias (Tables 3 and 4)[14-31].

Table 3 Quality appraisal of the included studies using the Newcastle-Ottawa Scale for case-control studies.

Ref.	Selection				Comparability	Exposure			Total score
	Case definition	Case representativeness	Controls selection	Controls definition		Ascertainment	Same method of ascertainment	Non- response rate
Jones et al[16]	Yes	Yes	Yes	Yes	Yes, yes	Yes	Yes	-	8/9
Rocha-Pereira et al[15]	Yes	Yes	Yes	Yes	Yes, yes	Yes	Yes	-	8/9
Uyanik et al[17]	Yes	Yes	Yes	Yes	Yes, yes	Yes	Yes	-	8/9
Pietrzak et al[18]	Yes	Yes	-	Yes	Yes, yes	Yes	Yes	-	7/9
Asefi et al[23]	Yes	Yes	Yes	Yes	Yes, yes	Yes	Yes	-	8/9
Ferretti et al[20]	Yes	Yes	Yes	Yes	Yes, yes	Yes	Yes	-	8/9
Oliviero et al[21]	Yes	Yes	Yes	Yes	Yes, yes	Yes	Yes	-	8/9
Nemati et al[22]	Yes	Yes	Yes	Yes	Yes, yes	Yes	Yes	-	8/9
Asefi et al[25]	Yes	Yes	Yes	Yes	Yes, yes	Yes	Yes	-	8/9
Papagoras et al[24]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	-	7/9
Sunitha et al[26]	Yes	Yes	Yes	Yes	Yes, yes	Yes	Yes	-	8/9
Wadhwa et al[28]	Yes	Yes	Yes	Yes	Yes, yes	Yes	Yes	-	8/9
Miao et al[29]	Yes	Yes	-	Yes	Yes, yes	Yes	Yes	-	7/9
Garshick et al[31]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	-	7/9
Wang et al[30]	Yes	Yes	Yes	Yes	Yes, yes	Yes	Yes	-	8/9

Table 4 Quality appraisal of the included studies using the Newcastle-Ottawa Scale for Cohort studies.

Ref.	Selection				Comparability	Outcome			Total score
	Exposed cohort representativeness	Non exposed cohort selection	Exposure ascertainment	Outcome demonstration of interest not present at start of study		Assessment	Follow up long enough	Cohorts follow up adequacy
Seçkin et al[14]	Yes	Yes	Yes	Yes	Yes, yes	Yes	-	-	7/9
Coimbra et al[19]	Yes	Yes	Yes	Yes	Yes, yes	Yes	-	Yes	8/9
Sorokin et al[27]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	-	7/9

DISCUSSION

Our results support that patients with psoriasis have significantly higher Lp(a) levels compared with HC, suggesting an additional risk factor for ASCVD in these individuals. Our sensitivity analysis indicates that psoriasis patients of European origin have considerably elevated Lp(a) levels compared to their Asian counterparts. Such a racial/ethnic difference regarding Lp(a) concentrations has been previously demonstrated in several studies[32,33].

According to international guidelines and expert consensus statements, Lp(a) measurement should be performed at least once during an individual's lifetime, since the finding of an elevated Lp(a) concentration could redefine their cardiovascular risk assessment[34,35]. As mentioned above, Lp(a) concentration is predominantly determined by genetics (90%) in more than any other lipoprotein[9]. Indeed, the Kringle IV repeat polymorphism determines the variability in Lp(a) concentrations. Several frequent and rare functional single-nucleotide polymorphisms (SNPs) profoundly modify the inverse correlation of isoform size and eventually, the Lp(a) concentrations, explaining the variations among individuals of different races and ethnicities[9]. The above could interpret the findings of our sensitivity analysis where psoriasis patients of European origin demonstrated higher alterations in Lp(a) levels compared to the Asian population.

Our findings are in line with the results of a 2019 systematic review and meta-analysis of 49 studies by Ramezani et al[36]. In their meta-analysis focused on Lp(a) levels, 8 studies were included with 446 psoriasis patients and 369 HC, and showed that Lp(a) levels were higher by 8.51 mg/dL in patients vs HC (95%CI: 4.86-12.17, P < 0.00001)[36]. Comparably, our updated meta-analysis, which included 16 studies with 3271 psoriasis patients, corroborated these results, although the MD was found to be lower. Regarding classic lipid profile parameters, Ramezani et al[36] showed significant elevations in total cholesterol (TCHOL), triglycerides, LDL cholesterol (LDL-C), apo B, and a decrease in high-density lipoprotein cholesterol (HDL-C) but not apoA1 levels in psoriasis patients compared with HC.

An increased incidence of major adverse cardiovascular events (MACE) including myocardial infarction and stroke has been demonstrated in patients with psoriasis[37]. Postulated mechanisms include a synergy between traditional ASCVD risk factors and inflammation-related mechanisms[38]. Importantly, six studies have shown that psoriasis is associated with an increased risk of atherosclerosis and MACE in an independent, dose-response relation to inflammatory activity[37,39-43]. Therefore, traditional cardiovascular disease scores tend to underestimate the effect of disease-related factors on ASCVD risk[44] and, as such, proposals for a 1.5 risk multiplier have been made when using these scores[45,46].

Several pathophysiological mechanisms are common between psoriasis and atherosclerosis. The two diseases are characterized by an activation of effector T cells, specifically T helper (Th) 1 and Th17 cells. On the one hand, psoriasis is related to dendritic cell-derived secretion of IL-12 and IL-23, which lead to differentiation of T cells into Th1 and Th17 types, respectively. Within the psoriatic plaque, Th1 cells produce tumor necrosis factor-alpha (TNF-α) and interferon-gamma (IFN-γ), inducing keratinocyte proliferation, upregulation of adhesion molecules, and angiogenesis. Th17 cells release IL-17 and IL-22, producing similar results. Further, T regulatory cell inhibition coexists with Th1 and Th17 pathway overactivation. On the other hand, atheromatic plaque-related endothelial activation is also associated with the production of IL-12 and IL-23 by macrophages and dendritic cells. In this context, Th1 cells promote plaque buildup, and Th17 cells induce plaque angiogenesis and instability[47,48]. A study by Mehta et al[49] pointed towards a dominant role of the Th1 cascade, particularly IFN-γ, in inflammation-induced atherogenesis. Moreover, systemic inflammation participates in insulin resistance and causes lipoprotein oxidation, foam cell formation and, hence, atherogenesis propagation[50]. Given the shared mechanisms between psoriasis and atherosclerosis, treatments may aim in both directions. Efficacious psoriasis treatments with anti-atherogenic properties include methotrexate, anti-TNF, anti-IL-12/23 and anti-IL-17 agents[50-52]. However, the role of such therapies on ASCVD is yet to be determined[53].

The elevated ASCVD risk in patients with psoriasis is further delineated by the higher prevalence of co-morbid risk factors including metabolic syndrome, type 2 diabetes, dyslipidemia, and smoking compared to the general population[4-6]. In fact, incidence of metabolic syndrome and other risk factors is analogous to psoriasis severity, while disease activity and treatment response are associated with the occurrence of such comorbidities[53-56]. Of note, the link between psoriasis and traditional cardiovascular risk factors is independent of confounders like age, sex or smoking, and is determined by aberrations in common pathophysiological pathways and genetic factors[57,58].

In regard to dyslipidemia, the aforementioned lipoprotein level alterations [i.e., elevated Lp(a), TCHOL, triglycerides, LDL-C, apo B and low HDL-C] are explained by comorbid metabolic diseases (e.g. obesity), systemic inflammation, as well as type of received treatment[29]. Specifically, Lp(a) levels have been shown to correlate with Psoriasis Area Severity Index score, an indicator of disease severity, while Lp(a) decreases have been demonstrated secondary to administration of anti-inflammatory agents such as methotrexate and anti-TNF, pointing toward the involvement of a pathway other than the IL-6 axis[14,18,30,59]. This provides further evidence in favor of the inflammation hypothesis connecting psoriasis and ASCVD[18]. Disease severity is also associated with Lp(a) peroxidation and oxidative stress[20], while high Lp(a) levels affect chemotaxis and expression of adhesion molecules, and, therefore, T-cell migration[22]. Nevertheless, a Mendelian randomization analysis by Ti et al[60] showed no causal genetic association between psoriasis and Lp(a) levels.

The present systematic review has certain limitations. It relies on observational studies, which, albeit of high quality as per the quality assessment, may suffer from confounding and bias. Furthermore, our meta-analysis has a high heterogeneity, which may be attributed to non-standardized Lp(a) measurement methods. Finally, the funnel plot presents relative asymmetry, which may possibly suggest publication bias. However, the asymmetry is limited to studies with a larger standard error, pointing toward random inter-study differences.

CONCLUSION

The present systematic review and meta-analysis suggest a significant increase of Lp(a) levels in psoriasis patients compared with HC. Elevated Lp(a) represents an ASCVD risk factor additional to the metabolic derangement frequently encountered in psoriasis. As such, Lp(a) measurement in psoriatic patients may be of use for a more precise determination of cardiovascular risk and treatment goals. Finally, given the pro-inflammatory effects of Lp(a) and its association with the severity of systemic inflammation, our results highlight the possible cardiovascular benefit of disease-modifying anti-inflammatory treatments in such patients, yet further research is required in this direction.