Overweight and erythrocyte polyunsaturated fatty acid changes in menopause

37 menopausal women without cardiovascular, ipertensive, thyroid, diabete, and cancer disease were included in the present study authorized from the Local Health Unit Company Umbria (Cod. PG263274). All patients signed the informed consent. A specific questionnaire was carried out with the patients to find out: BMI, pregnancies, menarche age, pregnancies, physical activity, smoker, age of onset of menopause, years since onset of menopause. BMI was assessed according to the World Health Organization criteria and the study above reported (6).

Blood samples were collected between October 1, 2018, and February 28, 2019 by using standardized procedures.

The blood sample was taken by puncturing and dripping blood from the finger until collecting 0,5 ml in vacutainer tubes containing ethylenediaminetetraacetic acid (EDTA) in the fasting state. Samples were stored at 4 °C until analysis. For mature erythrocyte selection, blood samples were centrifuged and entered a specific automatized procedure in which cell fraction was isolated on the basis of high density of the aged cells (9) The robotics performs all the subsequent steps for the cell lysis, isolation of the membrane pellets, phospholipid extraction from pellets and treatments (9). Erythrocyte cell membrane lipid analysis was performed by Fatpharmacy test that permits a lipidomic analysis with molecular check up of membrane lipids. The analysis includes capillary column gas chromatography (GC). Gas chromatography is the gold standard method for the determination of fatty acids, and it is performed under optimal separation condition, to identify a group of 10 cis fatty acids and 2 trans fatty acids (called lipidomic cluster). This cluster is always present in biological membranes because the fatty acids that make it up are part of the fundamental structure of the hydrophobic layer of the membrane. The lipidomic cluster of healthy subjects and the optimal intervals, to be used as a reference for the object of study samples, were obtained from an examination of the literature together with the study of the values obtained by Italian subjects present in the Lipinutragen database (10). The amount of each FA was calculated as a percentage of the total FA content (relative %), being > 97% of the GC peaks recognized with appropriate standards (9).

We consider two samples S1${{\mathcal{S}}_{1}}$(control sample with BMI<25) and S2${{\mathcal{S}}_{2}}$(case samples with BMI>25) of n1 (15) and n₂ (22) patients, respectively, chosen among menopause women. One asks, whether the differences between the obtained experimental data are accidental, or the second sample is significant with respect to the first.

For i = 1,2,..., r and j = 1, 2, let yi,j be the mean of the data concerning the concentration of the acid Xt under the treatment Jj, ${{\mathcal{J}}_{j}},$and set

Moreover, let .μi, bj e the mean corresponding to the world menopause women populations associated with, the conentration of the acid X under the treatment and put

LeX Sj be the covariance matrix associated with j = 1,2, and define a pooled-sample covariance matrix by S

We compute S by means of the Excel functions COVAR or COVARIANZA. Since these functions deal with population covariance rather than sample covarianc, in (1) we take the product n1S1 + n2S2 rather than the product (n1–1)S1 + (n2 – 1)S2 in ([1, §6.4 (3)]). We test the null hypothesis μ1= μ2 (H0), without specifying the common value, versus the alternative hypothesis μ1≠μ2 (H1). Note that, in general, it would not be advisable to test H0 by taking each acid separately (see also [1, Exercise 6.4.7). In our samples n1=22, n2=15, r=12, and hence n1+n2 > r+1. This condition guarantees that the matrix S is non-singular, and so the matrix S–1 is well-defined (see also [1]).

We consider the statistic ϕ defined by

Without loss of generality, we can assume that the total world population of menopause women has a normal distribution, and that the samples and are independent. The studied corresponding test statistic is

which has a Fisher-type distribution with r and n1+n2^–r^–1 degrees of freedom (see also Theorem 6.4.1) (11). We compute the value of the statistics ϕ in (2). The Excel functions FDIST and DISTRIB.F compute the p-value related to F and ϕ, that is the quantity

where the symbol in (3) denotes the probability that the test statistic F is greater than or equal to ϕ, when the null hypothesis H0 is satisfied. The p-value is the smallest significance level at which the obtained data lead to reject the null hypothesis. As we consider the significance levels α1 = 0.05 or α2 = 0.01, the null hypothesis is rejected if and only if the p-value is less than or equal to α1 (or α2). Thus, a p-value p < α1 (or p < α2) indicates that the sample S2${{\mathcal{S}}_{2}}$is significant (resp. highly significant) with respect to S1,${{\mathcal{S}}_{1}},$or equivalently that S1${{\mathcal{S}}_{1}}$is significant (resp. highly significant) with respect to S2.${{\mathcal{S}}_{2}}.$

The analysis of BMI revealed that of 37 menopausal women object of sutdy, 15 had a BMI value <25 and 22 >25 (Tab.1). According to the World Health Organization criteria and the study showing the value of reference of BMI in menopause (6), the control samples were considered women with BMI <25 and case samples women with BMI >25. The media ±SD of BMI showed significant differences between the two groups (*p<0.001). There were no significant differences in age. Also percentage values of smokers, physical activity, and pregnancies were similar.

We then analysed the time from menopause and divided samples between those who were within and those who were over 5 years after menopause (Fig.1). 60% (n°9) of control samples were within 5 years and 40% (n°6) over 5 years (Fig.1a); 45% (n°10) of control samples were within 5 years and 55% (n°12) over 5 years (Fig.1b)

We then analyzed the percentage of the population with an age of appearance of the menarche greater than 13 years and the onset of menopause at the age less than 38 years (Fig.2). Interestingly, the results showed that among women with menarche age >13 years, 75% were case samples (Fig.2a) and among women with the age of onset the menopause <38, 83% were case samples (Fig.2b).

Figure 1

Figure 2

SFA, MUFA, PUFA, and n-6 PUFA/n-3 PUFA levels in erytrocyte membrane were considered (Table 2). The results showed that the percentage of women who had physiological levels of SFA was lower in case samples than control samples, and percentage of physiological levels MUFA and PUFA was higher. All samples that did not have physiological levels, they always had higher values and never lower. Therefore, erythrocyte membrane of woman with BMI>25 were enriched in SFA and poorer in MUFA and PUFA than woman with BMI<25. Physiological level of n-6/n-3 PUFA ratio was lower in woman with BMI>25 than in woman with BMI<25, by indicating a high level of n-6 PUFA (Table 2).

Then, we analyzed the percentages of patients who had physiological values by dividing them between women who were within 5 years of menopause and those who were over 5 years of menopause (Table 3). As you can see, the percentages of physiological values increased over time, with the exception of the SFA of women belonging to the control group, indicating an improvement of the parameters over time regardless of the BMI.

However, if you consider the absolute values of SFA, MUFA, PUFA there are no significant variations between the case and the group control both in reference to the total of women and between women who are within 5 years of menopause and those who are beyond 5 years after menopause

Notably, the value of n-6/n-3PUFA ratio of control group, that was within the physiological range (Table 3), was similar if you consider the time within 5 years and beyond 5 years after menopause (Fig.4a). Differently, medium values (5.5-7 range) was reached by woman who were in menopause within 5 years and high value (>7) by woman who were in menopause beyond 5 years (Fig.4a). As above reported, no woman with BMI >25 who were in menopause within 5 years had physiological values of n-6/n-3PUFA ratio but this was achieved after 5 years, indicating a possible normalization over time (Table 3, Fig. 4b). Medium and high values were similar within and beyond 5 years after menopause (Fig.4b)

Figure 3

Figure 4