ASSOCIATION OF CAROTID INTIMA-MEDIA THICKNESS WITH LEPTIN AND APOLIPOPROTEIN B/APOLIPOPROTEIN A-I RATIO REVEALS IMMINENT PREDICTORS OF SUBCLINICAL ATHEROSCLEROSIS IN PSORIASIS PATIENTS

Summary: Psoriasis patients are often susceptible to cardiovascular diseases (CVD), including atherosclerosis. Traditional markers (biochemical and inflammatory) and diagnostic tools could detect occlusive but not subclinical atherosclerosis. Carotid intima-media thickness (CIMT), has recently been recognised as a non invasive diagnostic tool for identification of premature atherosclerosis. Therefore we evaluated 80 psoriasis patients and 80 age sex matched healthy controls for serum leptin levels and apolipoprotein B/apolipoprotein A-I ratio (apoB/apoA-I ratio) in relation with CIMT of carotid artery. Carotid intima-media thickness and carotid plaques were simultaneously measured by carotid sonography. Serum concentration of leptin and apolipoprotein were measured using enzyme-linked immuno sorbent assay (ELISA) and nephelometry respectively. Raised CIMT correlated to age of onset of the disease, serum leptin and apoB/apoA-I ratio in psoriasis patients. Taking into account, values that were above the 75 percentile of the three markers (leptin, apoB/apoA-I ratio and CIMT) the odds ratio was 4.26 (2.06–8.80 CI). Leptin and apoB/apoA-I ratio showed significant cumulative association with CIMT. Results of predictive analysis supports measurement of CIMT along with estimation of serum leptin and apoB/apoA-I ratio for prediction of premature atherosclerosis in psoriasis patients.


Introduction
Psoriasis is an immune intervened inflammatory disorder of skin that affects nearly 2-3% of the total world's population. The disease is caused by unusual differentiation and hyper proliferation of the keratinocytes and can be identified by presence of red, scaly, sharp demarcated plaques mostly over extremities and scalp. Psoriasis is a systemic disease and studies suggest that patients with psoriasis are at increased risk of cardiovascular diseases, including subclinical atherosclerosis (1). The increased incidences of cardiovascular disease in these patients cannot be explained by conventional risk factors (older age, high blood pressure (BP), dyslipidemia, smoking, obesity, diabetes mellitus) alone (1,2). There is an urgent need to study role of other contributory factors common to both psoriasis and cardiovascular disease. Recently, American Association of Clinical Endocrinologists has approved routine measurement of apolipoprotein to assess risk of cardiovascular disease (3). Serum apoA-I (apolipoprotein A-I) is an index of concentration of HDL-C (high-density lipoprotein-cholesterol) in plasma and apoB (apolipoprotein B) is primary apolipoprotein associated with LDL-C (low-density lipoprotein-cholesterol). Thus the balance between the pro-atherogenic apoB and the anti-atherogenic apoA-I can help in the estimation of cardiovascular risk.
Previous studies have shown that both immune cell response and chronic inflammatory cells are involved in accelerating atherosclerotic risk in patients of psoriasis. Immune cell dysfunction in psoriasis produces systemic inflammation. Cytokines involved in inflammation further increases serum leptin, a 16 kDa obese protein, which in turn increases proinflammatory mediators, thus inducing vicious circle of inflammation (4). Thus, we evaluated patients for serum leptin which is known to regulate body weight, metabolism and immunity. Additionally, leptin induces proliferative and anti-apoptotic activities in T cell (5). Since psoriasis is an immune mediated disease, characterized by hyperproliferation of skin cells and infiltration of T lymphocytes, leptin may provide a link between T cell function and inflammation in psoriasis (6). Clinical studies done so far gave contradictory results, suggesting a multifaceted role of leptin, as an obese protein and/or immunomodulator (decreased immune cell response), in immune-mediated inflammatory conditions in humans.
CIMT measurement has been validated as a surrogate non invasive diagnostic tool to identify atherosclerosis at sub clinical stage (7,8). Thus, we measured CIMT and evaluated its potential relation with apoB/apoA-I ratio and serum leptin in patients of psoriasis.

Human subjects
The study involved 80 adult patients (≥ 18 years) of either sex, diagnosed to have psoriasis, from Dermatology Outpatient Department, UCMS & GTB Hospital, Delhi and 80 age, sex and BMI matched healthy volunteers with no known systemic disease. Norms of National Cholesterol Education Program Expert Panel Adult Treatment Panel III for Asian definition was adapted for defining conventional cardiovascular risk factors (smoking, central obesity, dyslipidemia, diabetes, hypertension) in these patients (9).

Inclusion and exclusion criteria
Patients below the age of 18 years, pregnant and lactating female patients and subjects who had systemic treatment of retinoids in last six months for the management of their psoriasis were excluded from the study. Those who were on lipid-lowering medication were also not included in the study. The control group comprised healthy hospital staff and patients from dermatology outpatients clinics who were not undergoing any systemic therapy nor had been diagnosed with lipid metabolic disorders.

Physical and clinical evaluation of study groups
Since the present study involves investigations of various conventional and nonconventional risk factors so confounding factors which are known to have an effect on blood lipids value, were recorded. All patients were subjected to detailed personal interview and standardized clinical examination using a detailed questionnaire to obtain information including:

Clinical evaluation
Detailed medical and treatment history including the use of lipid-lowering medications, current cigarette smoking status and alcohol consumption. Clinical severity of disease was assessed using Psoriasis Area and Severity Index (PASI) (12).
The study was approved by institutional human ethical committee and a written consent was obtained from all participants of the study.

Sample collection
Following an overnight fasting, 5 ml of blood sample was drawn from participants using venipuncture, under aseptic conditions. Blood was centrifuged at 2000g for 10 minutes and serum sample thus obtained was preserved at −80 °C until assayed.

Assay of biochemical parameters
Fasting and post prandial glucose levels were measured spectrophotometrically at 500 nm using glucose oxidase-peroxidase method.
Serum concentration of apolipoprotein (apoA-I and apoB) and leptin were measured using kits based on nephelometry (Goldsite Diagnostics Inc, China) and ELISA (BioVendor) respectively as per manufacturer's instructions.

Measurement of Carotid intima media thickness (CIMT)
CIMT and plaques of both right and left carotid arteries were measured to assess subclinical atherosclerosis and structural changes in the vascular wall using a high resolution B mode ultrasound with a 10 MHZ linear vascular probe (13). All carotid ultrasound measurements were performed by an expert cardiologist. Values above 0.7 mm were defined as elevated CIMT (14). The mean CIMT was calculated as the arithmetical mean of two measurements of the left and right carotid arteries.

Statistical analysis
Statistical analysis was carried out using standard statistical software (SPSS software version 20). Quantitative data were expressed as mean ± standard deviation . Comparison between patients and controls for different parameters was done using t-test. Pearson correlation analysis was used to determine the relation of CIMT with other risk variables. The difference between patients with psoriasis and control group under study has been explained using test of equality of correlations coefficients. We generated receiver operator characteristic (ROC) curve for serum leptin, apoB/ apoA-I ratio and mean CIMT. Specificity, sensitivity and predictive values were calculated using MedCalc statistical software. Association of PASI with CIMT was investigated using adjusted logistic regression model. Multivariate regression analysis was used to study the effect of inde-

Clinical characteristics
A total of 80 patients of psoriasis and an equal number of age-sex matched controls were included in the study, to rule out any possible role of confounding factors. Most of the patients (30% each) were either from the 31-40 years or 51-60 years age group. The demographic and clinical data of study subjects has been summarized in the Tab. 1. Eighty percent of patients were suffering with psoriasis vulgaris, the most common variant of psoriasis.

Association of atherogenic risk factors with disease in patients
The main atherogenic characteristics of patient and control groups are shown in Tab. 2.

Conventional risk factors
Of the total 80 psoriatic patients studied, 4 (5.0 %) had diabetes, 4 (5 %) had hypertension, 54 (60 %) had dyslipidemia and 38 patients (47.5%) were still actively smoking with mean smoking index (number of cigarettes or beedi/ day × total duration in years) of 19.28 ± 8.29. The mean of the severity of disease (PASI) (15.60 ± 10.79) in patient group was found to be significantly associated with duration of disease (p = 0.001) (r = 0.638) and waist: hip ratio (r = 0.412, p < 0.01). Smoking Index was significantly correlated (r = 0.483), (p < 0.05) to severity (PASI) of disease. The data shows that disease gets severe with time and a change in lifestyle may help in curbing the disease.

Nonconventional risk factors
Serum apoB levels (P < 0.05) and apoB/apoA-I (P < 0.001) ratio in patients were significantly higher as compared to control subjects while the difference in serum apoA-I concentration of two groups were statistically insignificant. Serum leptin levels in patients of psoriasis (16.57 ± 7.68) was significantly higher (p < 0.001) than those of healthy controls (8.43 ± 3.64) (Tab. 2). The study shows considerable correlation (r = 0.603, p < 0.05) between higher serum leptin level and raised waist/hip ratio in patients. However no correlation was observed between serum leptin levels and lipid profile.

CIMT measurement and its correlation with atherogenic parameters
The mean CIMT values of the right and left carotid artery of the patients were found to be significantly higher (p < 0.001) compared with the controls (0.658 ± 0.13 mm vs 0.489 ± 0.09 mm and 0.656 ± 0.13 mm vs 0.482 ± 0.08 mm respectively). Among patients, 26 (32.5 %) reported an abnormal increase in CIMT i.e. above 0.7 mm. Data in Tab. 3 shows intra group comparison in patients of psoriasis, based on whether or not the CIMT was more than 0.7 mm. There was significant difference between two subgroups (based on CIMT cut off of 0.7mm) of patients for CIMT (p < 0.001), apoB/ apoA-I ratio (p < 0.001) and serum leptin (p < 0.05) levels.
We analysed the association between CIMT and other variables studied in the patient and in the control group (Tab. 4). All correlations were standardized for age and BMI. Significantly different trends were found for the correlation of CIMT with leptin in patient group than those in the control group.
Data in Fig. 1 suggests that greater CIMT values were significantly associated with age of patient, age of onset of disease (p < 0.001), smoking index, systolic blood pressure, diastolic blood pressure, serum leptin, apoB concentration (p < 0.05) and apoB/apoA-I (p < 0.001) in patient group. However, CIMT was not found to be correlated to either duration, severity of the disease or serum apoA-I concentration in the patients of psoriasis. Even after adjustment for other confounding factors (smoking index and disease duration) no significant association (partial R = 0.067, p = 0.07) was observed between CIMT and severity of psoriasis (PASI).  Results from regression analysis (F = 8.945, P < 0.001) suggested that increased CIMT in patients may be associated to cumulative chronic inflammation and impaired apolipoprotein and thus it can serve as a reliable indicator of subclinical disease and may help in early diagnosis of CVD risk factors associated with psoriasis.

Tab. 3: Anthropometric and laboratory values classified by CIMT in psoriasis patients
The 75th percentile values used to determine high risk values for apoB/apoA-I ratio, leptin levels and mean CIMT were 0.73, 23.67 ng/ml and 0.70 mm respectively. We calculated predictive values (sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) for each biomarker. The results are as follows: 30%, 92.5%, 80% and 56.92% for apoB/apoA-I ratio; 25%, 90%, 71.43% and 54.55% for leptin and 30%, 92.5%, 80% and 56.92% for mean CIMT. The predictive values increased (47.50%, 82.50%, 73.08% and 61.11%) when we took three biomarkers into account (Tab. 5). A fairly significant kappa value of 0.3 depicted that the three biomarkers can be used to assess early risk of cardiovascular disease in psoriasis patients. Assuming 75th percentile value as cut off for leptin, apoB/apoA-I ratio and CIMT, the odds ratio calculated was 4.26 (2.06-8.80 CI). According to results of ROC curves (Fig. 2), the variable with highest AUC (area under curve) was mean CIMT followed by serum leptin and apoB/apoA-I ratio respectively.  however cardiovascular risk factors associated with the disease, at times, make it fatal (15,16,17). Recently few studies from western countries have reported that psoriasis may be independently associated with myocardial infarction (18,19). Anthropometric and serum lipid measurement has been recommended in psoriasis patients, but their role is not certain till date (2,19,20,21). Thus, we evaluated psoriasis patients for both conventional and non conventional cardiovascular risk factors. Recently, CIMT has been recognised as reliable non invasive diagnostic tool to identify premature atherosclerosis. We studied potential role of apoB/apoA-I ratio and serum leptin levels as predictor for subclinical atherosclerosis, in association with CIMT. We matched age, sex and BMI in patients and controls to rule out their possible role in evaluation of atherogenic parameters studied.

Discussion
It was observed that 47.5% patients were active smoker and nearly 60% of patients were either overweight (BMI > 25 kg/m 2 ), hypertensive or had abnormal serum lipid profile. Obesity could be sign of inactiveness in these patients. In addition, waist/hip ratio was significantly higher in patients than controls (p < 0.001). In 1995, Henseler and Christophers also reported that patients with severe psoriasis were mostly overweight (22). In a similar study Naldi et al., observed that a moderate increase in BMI was associated with increased risk of psoriasis and clinically obesity doubled the risk of psoriasis (23).
Serum leptin (OB protein) in patient group was significantly related to raised waist hip ratio, PASI score and increased CIMT, thus providing its association with psoriasis. These findings were in agreement with previous results of Enany et al. (24), Cerman et al. (25) and Hamminga et al. (26) who suggested that leptin may serve a role of marker in psoriasis. Thus we may infer that increased serum leptin in obese patients may be correlated to events of metabolic dysfunction.
In the study, apoB levels and apoB/apoA-I ratio were elevated in psoriasis and abnormal apolipoprotein metabolism may be related to the high incidence of atherosclerosis.
Among patients, 32.5% had high CIMT (above 0.7 mm) with the mean CIMT (0.65 ± 0.02 mm). Recent study by Kimhi et al. (27) showed that psoriatic arthritis patients had significantly higher CIMT compared with controls (0.76 ± 0.11 versus 0.64 ± 0.27, P < 0.00001) which was similar to results given by Gonzalez et al. and Mongi et al. (28,29). In our current study, we have observed a similar trend with significantly higher CIMT (P < 0.001). Grobbee et al. (30) and Tam et al. (31,32) have shown that presence of carotid lesions and thickened intima can be considered as a surrogate marker for the detection of early atherosclerosis in subclinical stage of disease and can be correlated to cardiovascular risks among patients and this increase is significantly associated with clinical (age of onset of the disease) and atherogenic (apoB/apoA-I ratio) parameters as also revealed in our study. We did not find any correlation of CIMT with disease duration or severity which was similar to reports given by Youssef et al. (33). This was perhaps due to quite young age group of our patients and fluctuating course of the disease.
Our finding was concordant to those of Huang et al. who found that both apoB levels and apoB/apoA-I ratio were associated to increased CIMT (34). As reported by Dahlen et al. we also observed that high apoB/apoA-I ratio, and not serum lipid profile, were associated with increased CIMT (35). Previous studies from various countries have also demonstrated that an increased CIMT value often correlates to atherogenic risks in various chronic diseases (36). By regression analysis, we found a significant cumulative effect of apoB/apoA-I ratio and serum leptin (as predictors) on mean value of carotid artery records of psoriasis patients studied (as dependent variables) (F = 8.945, P < 0.001). Thus a significant correlation of cardiovascular risk factors and inflammatory biomarkers with increased CIMT in present population may be linked to increased risk of subclinical atherosclerosis in the diseased group. The study further shows that CIMT along with leptin and apoB/apoA-I ratio can be used as biomarkers to estimate cardiac risk in patients of psoriasis.

Conclusions
We may conclude that the measurement of apoB/apoA-I ratio rather than serum apolipoprotein estimation, can help in diagnosis of premature atherosclerosis in patients of psoriasis. Chronic inflammation plays a major contributing role in both psoriasis and atherosclerosis. This study demonstrates the need to screen patients for serum leptin, apoB/apoA-I ratio and CIMT along with known conventional risk factors, to predict subclinical atherosclerosis and hence the future risk of CVD in patients of psoriasis.