- Top of page
- Material and Methods
Aim: The apolipoprotein B/apolipoprotein A1 (ApoB/ApoA1) ratio is well known to be related to metabolic syndrome (MS) and its components in adults of different races. There is low prevalence of MS but high occurrence of various metabolic disorders in Chinese adolescent women with polycystic ovary syndrome (PCOS). We sought to assess if the ApoB/ApoA1 ratio can be used as a predictive marker of MS and pre-MS in Chinese adolescent women with PCOS.
Material and Methods: This cross-sectional study included 160 Chinese adolescent women. Based on International Diabetes Federation criteria for MS, patients who had no less than two components of MS but did not meet the criteria for the diagnosis of MS were considered as having pre-MS.
Results: The ApoB/ApoA1 ratio was higher in obese subjects with high free androgen index (FAI). The ApoB/ApoA1 ratio increased significantly as the number of MS components increased and provided 87.5% of sensitivity and 78.9% of specificity with a threshold value of 0.63 for MS, 86.2% of sensitivity and 79.4% of specificity with a threshold value of 0.58 for pre-MS in Chinese adolescent women with PCOS.
Conclusion: The ApoB/ApoA1 ratio was a good predictive marker of MS and pre-MS in Chinese adolescent women with PCOS. FAI could be involved in obesity-related metabolic abnormalities.
- Top of page
- Material and Methods
Patients with polycystic ovary syndrome (PCOS) often have complications associated with dyslipidemia, obesity, hypertension and insulin resistance (IR), and are therefore at an increased risk of having metabolic syndrome (MS) and cardiovascular diseases (CVD).1,2 In addition, metabolic aberrations in PCOS patients start during the adolescent period, and extend throughout their lifetime.
In our previous study, we reported a low prevalence of MS but a high occurrence of various metabolic disorders in adolescent women with PCOS in South China.3 That is to say, there were lots of adolescent women with PCOS who did not meet the criteria for the diagnosis of MS but had no less than two components of MS (we defined this stage as pre-MS). The patients with pre-MS are at high risk of suffering from MS, CVD and type 2 diabetes (T2DM), and thus early prediction and treatment of pre-MS are very important.
Apolipoprotein B (ApoB) represents the total amount of potentially atherogenic circulating compounds, including small dense low-density lipoproteins (LDLs). It is easily oxidized and induces an inflammatory reaction and the formation of plaques in the arterial wall. Apolipoprotein A1 (ApoA1) is the major component in high-density lipoprotein (HDL) particles and plays a prominent role in the reverse transportation of cholesterol, as well as anti-inflammatory and antioxidant processes. Hence, the ApoB/ApoA1 ratio reflects the status of pro- and anti-atherogenic lipoproteins, and is a marker of MS and CVD.4,5
In adults, the ApoB/ApoA1 ratio is an independent risk factor for IR and is related to MS and its components. Moreover, it is superior to any of the cholesterol ratios as a summary index of the risk of CVD in populations of different races, sexes and ages.6–9 However, reports in children and adolescents are limited. Furthermore, no data is available on adolescent women in China. The aim of this study was to investigate whether the ApoB/ApoA1 ratio can be used as a predictive marker of MS and pre-MS in Chinese adolescent women with PCOS.
- Top of page
- Material and Methods
Of 160 subjects, there were 94 subjects with oligomenorrhea and 25 with amenorrhea. There was no significant difference in BMI between the subjects with oligomenorrhea and those with amenorrhea. (19.88 [18.86–23.14] vs 21.93 [18.64–25.67]; P > 0.05)
As shown in Table 1, BMI, SBP, DBP, WC, FIN, TG, FAI, HOMA-IR and the ApoB/ApoA1 ratio were higher in subjects with MS as compared to those without MS, while HDL-c and ApoA1 were lower in the same subjects. There were no significant differences in age, FPG, CHOL, LDL-c and ApoB levels between the patients with MS and those without.
Table 1. General characteristics of PCOS adolescents without and with MS (median [range])
| ||Without MS (n = 151)||With MS (n = 8)|| P-value|
|Age (years)||19.0 (17.0–19.0)||19.0 (17.5–19.0)||0.364|
|BMI (kg/m2)||19.6 (18.5–26.2)||29.4 (25.5–33.6)||<0.001|
|WC (cm)||70.0 (65.0–76.0)||95.0 (82.3–104.3)||<0.001|
|SBP (mmHg)||110 (100–116)||130 (121–142)||<0.001|
|DBP (mmHg)||70 (63–76)||84 (77–87)||<0.001|
|FPG (mmol/L)||4.8 (4.6–5.1)||5.1 (4.6–5.3)||0.326|
|FIN (mU/L)||8.2 (4.7–13.1)||20.7 (12.4–27.0)||0.001|
|CHOL (mmol/L)||4.60 (4.27–5.16)||4.77 (4.06–5.61)||0.597|
|TG (mmol/L)||0.96 (0.65–1.33)||2.13 (1.25–2.46)||0.001|
|HDL-c (mmol/L)||1.58 (1.36–1.93)||1.19 (1.01–1.29)||0.001|
|LDL-c (mmol/L)||2.56 (2.13–3.15)||2.99 (2.37–3.52)||0.140|
|HOMA-IR||1.67 (0.87–2.71)||4.49 (2.65–6.45)||0.001|
|FAI||4.05 (2.46–6.60)||13.44 (7.35–20.21)||0.003|
|ApoB||0.68 (0.57–0.77)||0.73 (0.66–0.94)||0.105|
|ApoA1||1.35 (1.14–1.54)||1.09 (0.95–1.21)||0.006|
|ApoB/ApoA1 ratio||0.50 (0.41–0.59)||0.69 (0.64–0.84)||0.001|
The correlation coefficients between the ApoB/ApoA1 ratio and individual components of MS, HOMA-IR and FAI are shown in Table 2. There was a significant positive correlation between the ApoB/ApoA1 ratio and BMI, WC, TG, HOMA-IR, FAI (P < 0.05) and a significant negative correlation of the ApoB/ApoA1 ratio with HDL-c (P < 0.05). The strongest correlation was found between the ApoB/ApoA1 ratio and HDL-c (r = −0.632).
Table 2. Spearman rank correlations between the ApoB/ApoA1 ratio and individual components of the MS, HOMA-IR, FAI
We divided the patients into four groups according to the values of WC (or BMI) and FAI (WC = 80 cm, BMI = 25 kg/m2, FAI = 7.4796). This analysis showed that the ApoB/ApoA1 ratio was significantly higher in the group with central obesity (WC ≥ 80 cm) and high FAI when compared to the group with no central obesity and low FAI (0.485 vs 0.699; P < 0.001). The result was similar when obesity was taken into account instead of central obesity (BMI ≥ 25 kg/m2) (0.484 vs 0.699; P < 0.001) (Fig. 1).
Figure 1. (a) Differences in apoB/apoA1 ratio according to the combination of central obesity and free androgen index (FAI) percentile. Non-central obesity and FAI < 75th group versus central obesity and FAI > 75th group, P < 0.001. (b) Differences in apoB/apoA1 ratio according to the combination of obesity and FAI percentile. Non-obesity and FAI < 75th group versus obesity and FAI > 75th group, P < 0.001.
Download figure to PowerPoint
Of all the patients, only 56.3% (n = 90) had no component of MS; 25.6% (n = 41) had one component of MS; 11.9% (n = 19) had two components; 4.4% (n = 7) had three components; 1.9% (n = 3) had four components; and no patient had all five components. The median values of the ApoB/ApoA1 ratios rose significantly along with the increasing number of MS components (Table 3).
Table 3. Median ApoB/ApoA1 ratio of subjects with different numbers of MS components
|No. components||Case (%)||ApoB/ApoA1 ratio|
|0||90 (56.3)||0.48 (0.39–0.55)|
|1||41 (25.6)||0.50 (0.43–0.68)|
|2||19 (11.9)||0.66 (0.59–0.95)|
|3||7 (4.4)||0.69 (0.69–0.85)|
In our study, the ApoB/ApoA1 ratio provided a sensitivity of 87.5% and a specificity of 78.9% with a threshold value of 0.63 for MS; a sensitivity of 86.2% and a specificity of 79.4% with a threshold value of 0.58 for pre-MS. The area under the ROC was 0.850 (95% CI = 0.768–0.931) for MS, 0.843 (95% CI = 0.762–0.923) for pre-MS, and the Youden index was 0.664 for MS and 0.656 for pre-MS. (Fig. 2, Table 4).
Figure 2. (a) Receiver operating characteristic (ROC) curves for detecting metabolic syndrome (MS) in Chinese adolescent women with polycystic ovary syndrome (PCOS). (b) ROC curves for detecting pre-MS in Chinese adolescent women with PCOS.
Download figure to PowerPoint
Table 4. ROC curves data for detecting MS and pre-MS using ApoB/ApoA1 ratio
| ||AUC||Cut-off point|| P-value||Sensitivity (%)||Specificity (%)||Youden index|
- Top of page
- Material and Methods
In this study, the level of ApoA1 in adolescent women with MS was lower, and the ApoB/ApoA1 ratio was higher when compared with those without MS, but no difference was noted in the level of ApoB. These results suggested that the ApoB/ApoA1 ratio is associated with MS, and the increased ApoB/ApoA1 ratio resulted primarily from a decrease in anti-atherogenic ApoA1 levels. Our study also demonstrated that the levels of HDL-c were lower but LDL-c shows no differences in the patients with MS compared with those without. However, these findings are of importance as they may give us some guidance in preventing MS and CVD in the adolescent women with PCOS.
The ApoB/ApoA1 ratio correlated with WC, TG, HDL-c and HOMA-IR, but not with FPG and blood pressure. Moreover, it was noted that there was significant difference in HOMA-IR but no difference in FPG between the groups. These results indicate that the ApoB/ApoA1 ratio may be a risk factor for IR, and HOMA-IR may be a stronger indicator of MS than FPG in our study population. This may be because IR/hyperinsulinemia is the major underlying pathological process of MS, and hyperinsulinemia precedes abnormal FPG stage in the development of T2DM. Sharma et al. reported that HOMA-IR was more significantly interrelated with the other four MS components than FPG, and using HOMA-IR was preferred to FPG to identify MS in African American children, which were in accordance with our results.16 Another study of obese children from Turkey showed that the ApoB/ApoA1 ratio correlated with TG and HDL-c.17 However, no significant correlation of the ApoB/ApoA1 ratio with WC or BMI was found. This was not surprising because the subjects in their study were all obese.
As we all know, hyperandrogenism is a common characteristic in adolescents and PCOS patients. As FAI reflects the status of biologically active androgens, we investigated the relationship between FAI and the ApoB/ApoA1 ratio. Our study showed that the ApoB/ApoA1 ratio was significantly higher in obese subjects with high FAI than that in non-obese subjects with low FAI, which was in accordance with a study from Korea.18 This can be explained by two factors. First, androgen may worsen the obesity-related metabolic abnormalities. FAI, SHBG and BMI were all independent determinants for serum ApoA1 levels in women with PCOS.19 Treatment with flutamide, a nonsteroidal androgen receptor antagonist, has been shown to decrease central fat mass in obese women with PCOS.20 Second, low SHBG might also have some effects on increasing the ApoB/ApoA1 ratio. Two prospective studies indicated that obesity, per se, in the absence of the metabolic characteristics of IR, does not increase the incidence of either CVD or T2DM.21–23 However, there were some studies which demonstrated opposite views.24,25
The ApoB/ApoA1 ratio increased as the number of MS components increased; this was in line with the study of Canadian Aboriginal adolescents (aged 10–19 years).26 However, the gap of the median ApoB/ApoA1 ratio between one MS component group and two MS components group was the largest among all the gaps of the ratios between two groups consecutively. This result may indicate that the risk of the adverse events (CVD or T2DM) was most significantly increased as the number of the abnormal components of MS elevated from one to two. It was also the main reason why we defined pre-MS as above.
Lower prevalence of MS (5%) but higher prevalence of pre-MS (13.1%) were found in our subjects. Concerning the long-term health risk, earlier identification should be done and treatment should also be given to these patients. Using the ROC curves and Youden index, the ApoB/ApoA1 ratio provided good sensitivity and specificity for evaluating MS and pre-MS in our study, the ApoB/ApoA1 ratio 0.58 (with sensitivity 86.2%, specificity 79.4%) and 0.63 (with sensitivity 87.5%, specificity 78.9%) were found to be the cut-off points for pre-MS and MS, respectively. Our study confirmed that the ratio may be a good indicator of pre-MS and MS in Chinese adolescent women with PCOS. Moreover, apolipoproteins have significant practical advantages for clinical use. Their testing do not require patient to be on a fasting status or immediate laboratory work-up but use standardized assays that are accurate and automated.27
In conclusion, the ApoB/ApoA1 ratio was higher in obese adolescents with high FAI, and it is a good predictive marker for both MS and pre-MS in Chinese adolescent women with PCOS. In light of the practical advantages for clinical use, measurement of the ApoB/ApoA1 ratio might be a useful and convenient tool to predict metabolic abnormality and potential cardiometabolic risks in this cohort of patients.
Despite these relevant findings, it is important to point out the limitations of our study. Due to the difficulties in inviting adolescent controls to participate in the study, the clinical significance of the ApoB/ApoA1 ratio in adolescent women with PCOS could not be compared with that in adolescent controls from the same region. In addition, all the subjects were recruited from the gynecological outpatient department of Memorial Hospital of Sun Yat-Sen University and in this present study, we only included those with complete medical records, which can not represent all the adolescent women with PCOS in China and there could be a possible bias regarding the results.