Neck Circumference Is a Reliable Predictor for Hyperuricemia in Women With Polycystic Ovary Syndrome: a Case Control Study


 Background: The prevalence of hyperuricemia in the women with polycystic ovary syndrome (PCOS) is almost threefold higher than that in women without PCOS. Both PCOS and hyperuricemia are considered risk factors of metabolic disorders, including hypertension, diabetes and cardiovascular diseases. Previous study has indicated that neck circumference (NC) is a good predictor for insulin resistance in women with PCOS. The present analysis thus aims to assess the predictive value of NC for hyperuricemia in women with PCOS.Methods: This is a cross-sectional study that recruited one hundred ninety women with PCOS from January 2018 to December 2019. PCOS was diagnosed according to the Rotterdam definition. Hyperuricemia was defined as serum uric acid level of at least 357 μmol/L.Results: PCOS females with hyperuricemia had significantly greater values of NC, body mass index (BMI), waist circumference (WC) and hip circumference (HC). NC was positively associated with serum uric acid levels, with a standardized regression coefficient of 0.34 after adjusting for confounding factors. Furthermore, logistic regression analysis showed that NC was significantly associated with an increased risk of hyperuricemia, with an adjusted odds ratio of 1.58. The associations between NC and serum uric acid levels were more evident in those with medium BMI (22.02 kg/m2 ≤ BMI < 25.08 kg/m2), low WC (WC < 85 cm) or medium HC (90 cm ≤ HC < 96 cm). The optimal cutoff point of NC in predicting hyperuricemia was 32.0 cm (Youden index = 0.48), with the sensitivity and negative predictive value of 87.23% and 93.68%, respectively.Conclusions: Neck circumference is positively correlated with hyperuricemia and has a high negative predictive value for hyperuricemia in women with PCOS. Therefore, we suggest NC as a simple, novel, and reliable anthropometric measure to be used in the routine clinical assessment of women with PCOS to screen those at high risk of hyperuricemia.


Background
Polycystic ovary syndrome (PCOS) is one of the most common endocrine and metabolic disorders in reproductive-aged individuals, with an incidence of 3%-20% according to different diagnostic criteria (1)(2)(3). Women with PCOS are characterized by symptoms of hirsutism or oligo/amenorrhea or when resorting to infertility care in early adulthood (4). In addition to reproductive disorders, PCOS is also closely related to a variety of metabolic abnormalities, including insulin resistance, diabetes, dyslipidemia and cardiovascular disease, which exert negative and life-long impacts on quality of life (5).
Uric acid is a metabolite produced during purine metabolism, and elevated serum uric acid levels have been demonstrated as a risk factor of metabolic disorders, including hypertension, diabetes and cardiovascular diseases(6). In addition to the effect on metabolism, recent review suggests that there is a close relationship between serum levels of uric acid and female reproductive disorders (7). Previously, our study demonstrated that the prevalence of hyperuricemia in the PCOS population was almost threefold higher than that in women without PCOS(8). Moreover, serum uric acid can be used as a predictor of adverse pregnancy and foetal outcomes (9,10). Therefore, the early detection and intervention of hyperuricemia is of great signi cance for the health of women with PCOS. However, the detection of serum uric acid levels is time-consuming, and it requires professional personnel and speci c equipment, which is unapplicable to clinical practice on a large scale, especially in some rural places. In contrast, measuring anthropometric indices such as body mass index (BMI), neck circumference (NC), waist circumference (WC) and hip circumference (HC) using tape or a scale is relatively easy (11). It has been con rmed that BMI (12)(13)(14), waist circumference (15)(16)(17) and hip circumference are positively correlated with the risk of hyperuricemia in different populations. However, it is not always feasible and accurate to measure WC, HC and BMI in the winter with heavy clothes or postprandially. Therefore, there is a need for a reliable, simple and fast method to identify hyperuricemia early in clinical practice.
Neck circumference (NC) is a convenient anthropometric parameter that re ects the subcutaneous fat tissue of the upper body(18). Our previous study found that in PCOS women with obesity, the prevalence of hyperuricemia was approximately threefold higher than that in PCOS women with normal BMI(8). Therefore, we hypothesized that NC might be a good predictor for hyperuricemia in women with PCOS. It has been reported that NC is a potential predictor of hyperuricemia in the general population (19,20).
However, to the best of our knowledge, there are currently limited studies focused on assessing the association between NC and hyperuricemia in women with PCOS. Thus, we conducted a retrospective study to examine whether NC was a good predictor for hyperuricemia and to determine the optimal cutoff points of NC for hyperuricemia in PCOS.

Patients And Methods
Participants This was a cross-sectional study that enrolled 229 PCOS females aged between 20 and 40 years from January 2018 to December 2019 at the reproductive center of the First A liated Hospital of Wenzhou Medical University. The exclusion criteria were as follows: 1) patients with other causes of hyperandrogenemia, including congenital adrenal hyperplasia, androgen-secreting neoplasms, and Cushing's syndrome; 2) patients with any medical intervention or diseases that could alter the neck circumference, including neck surgery, neck malformation, thyroid dysfunction, thyromegaly, and regular oral glucocorticoid treatment; and 3) patients with incomplete information for anthropometric parameters or laboratory examination. Finally, 190 (83.0%) patients were included for further analysis. This study was approved by the Ethics Committee of the First A liated Hospital of Wenzhou Medical University.
Written informed consent for the whole procedure was obtained from each patient. The study protocol conforms to the ethical guidelines of the Declaration of Helsinki as re ected in the a prior approval by the institution's human research committee.

De nitions
Polycystic ovary syndrome was diagnosed according to the Rotterdam de nition, in which two of the following three criteria should be met: 1) oligomenorrhoea or amenorrhea (less than eight menstrual cycles in 12 months, or if the menstrual interval was more than 35 days); 2) biochemical or clinical hyperandrogenism (such as hirsutism and acne); and 3) characteristic image of polycystic ovaries (at least one ovary containing 12 or more peripheral follicles measuring 2-9 mm in diameter and/or ovarian volume of at least 10 mL) on transvaginal or abdominal ultrasound (21). Hyperuricemia was de ned as an SUA level of at least 357 μmol/L (22).

Anthropometric and laboratory measurements
The anthropometric measurements included BMI, NC, WC, HC and blood pressure. BMI was calculated as the body weight in kilograms divided by the body height in meters squared (23). Neck circumference was measured using a exible tape, with the subject remaining standing, head held erect, at the level of the thyroid cartilage (24). WC was measured at the midpoint between the lowest rib and the iliac crest, and HC was measured at the greater trochanter (25). Blood pressure was measured with an electronic sphygmomanometer in the sitting position after 10 min of rest. Fasting blood samples were collected after an overnight fast of at least 8 hours during the 2 nd to 5 th day of the menstrual cycle to measure hormonal and metabolic parameters. All biochemical measurements were tested in the central laboratory of the First A liated Hospital of Wenzhou Medical University. Serum LH, FSH, E2 and testosterone were measured using an autoimmunoassay analyzer [Unicel Dxl 800, Beckman Coulter, USA]. Fasting plasma glucose, total cholesterol (TC), serum triglycerides (TG), high-density lipoprotein (HDL) and low-density lipoprotein (LDL) were quanti ed by an autoanalyzer [AU 5800, Beckman, USA].

Statistical analysis
Statistical analyses were performed using SPSS version 23.0 software (IBM Corporation), and receiver operating characteristic (ROC) analyses were conducted using MedCalc Application version 19.7.2 software. Data are presented as the median (interquartile range) or as the mean ± standard deviation for continuous variables. Skewness and kurtosis tests for normality were performed and found that the level of basal LH, basal FSH, LH/FSH ratio, basal E2, AMH, FINS, HOMA-IR, HOMA-β, TG, and uric acid did not follow normal distributions. Variables with skewed distributions were logarithmically transformed before statistical analysis. Differences between the two groups were analyzed by using Student's t test for normally distributed continuous variables and the Kruskal-Wallis test for those with skewed distributions. Multivariable linear regression was used to explore the association of NC with serum uric acid level (logtransformed) in different models with adjustment for potential confounders. Binary logistic regression analysis was used to calculate the odds ratios (OR) and 95% con dence interval (CI) of NC for hyperuricemia. For both logistic regression analyses and multivariable linear regression, no variables were adjusted in model 1. Adjusted variables in model 2 included age, SBP, and DBP. In model 3, TG (logtransformed), HDL, FINS (log-transformed), HOMA-IR (log-transformed), HOMA-β (log-transformed) and eGRF (log-transformed) were further adjusted. The interactions of NC with BMI, WC and HC were tested using binary logistic regression. Patients enrolled were strati ed into quartiles according to their NC and strati ed into low, medium and high groups according to the tertiles of BMI, WC and HC. Receiver operating characteristic (ROC) curves were used to compare the predictive ability of NC, BMI, WC and HC for hyperuricemia by calculating the area under the curve (AUC). The Youden index, de ned as sensitivity + speci city -1, was calculated to identify the optimal cutoff points. The sensitivity and speci city of NC, BMI, WC and HC as well as positive and negative predictive values were calculated for each cutoff point in the sample. All statistical tests were two-sided, and P< 0.05 was considered statistically signi cant.

Results
Baseline characteristics in women with PCOS The baseline characteristics categorized by the existence of hyperuricemia in women with PCOS are presented in Table 1. The prevalence of hyperuricemia in women with PCOS was 24.74%. Age was matched between the two groups. Compared with women with PCOS without hyperuricemia, women with PCOS with hyperuricemia had signi cantly greater BMI, NC, WC, HC, SBP, DBP, fasting insulin, HOMA-IR, HOMA-β, TG and serum uric acid levels and lower HDL levels (all P < 0.05). There were no signi cant differences between the two groups in basal hormone levels (LH, FSH, E2, T, AMH), LH/FSH ratio, FBG, TC, LDL or eGFR. Associations of NC with serum uric acid and hyperuricemia Multivariable linear regression analysis was performed to explore the associations between NC and serum uric acid levels ( Table 2). In unadjusted model 1, NC was signi cantly associated with the level of uric acid, and the standardized coe cient was 0.51 (P < .001). In model 2 (adjustment for age, SBP, and DBP) and model 3 (further adjustment for log TG, HDL, log FINS, log HOMA-IR, log HOMA-β and log eGFR), NC was still signi cantly associated with the level of uric acid, and the standardized coe cients were 0.45 (P < 0.001) and 0.34 (P = 0.004), respectively. In addition, binary logistic regression analysis was also conducted to further identify the correlation between NC and hyperuricemia (

Discussion
In the current study, we demonstrated that neck circumference was positively associated with serum uric acid levels and was also signi cantly correlated with hyperuricemia in women with PCOS. We also observed signi cant interactions between BMI, WC and HC with NC in relation to serum uric acid concentrations in women with PCOS. Additionally, ROC analysis demonstrated that NC had a higher negative predictive value for hyperuricemia in women with PCOS than BMI, WC, and HC. To the best of our knowledge, this is the rst study to demonstrate such a correlation between neck circumference and hyperuricemia in PCOS women.
Hyperuricemia is characterized by an abnormal increase in serum uric acid levels in the human body due to aberrant purine metabolism, abnormal renal secretion and reabsorption (26,27). Studies have demonstrated that elevated serum uric acid levels are associated with increased risks of hypertension, type 2 diabetes and cardiovascular diseases(28-30), which has received increasing attention as a major public health problem (31). Our previous study showed that the prevalence of hyperuricemia in the PCOS population was 25.48%(8), which is almost threefold higher than that of women in the general population (32). In the present study, the results are in line with our previous ndings that showed that 24.74% of PCOS patients were diagnosed with hyperuricemia. Therefore, it is of great importance to nd a simple detection method for early recognition of high-risk populations during symptomless periods. Our previous study also found that 58.75% of women with PCOS and obesity had hyperuricemia, which was nearly threefold higher than that in women with PCOS and a normal BMI, which indicated that accumulated fat was positively associated with a high risk of hyperuricemia(8).
As a reliable anthropometric index of upper-body subcutaneous fat, neck circumference has the advantages of convenience and standardized measurement and has been widely applied in the screening of abnormal fat distribution (33). It has been well acknowledged that NC is positively associated with the risk of obstructive sleep apnea (34), insulin resistance (35), type 2 diabetes and metabolic syndrome(36) in the general population. For women with PCOS, studies have shown that NC was positively correlated with visceral fat and could be adopted as an innovative tool for assessing body adiposity distribution (37,38).
Several studies have suggested that body fat accumulated in the upper body segment may contribute to hyperuricemia (12,39). Each additional 5 mm increase in NC was associated with a 17% higher likelihood of having hyperuricemia in women (20). In the current study, we also found a positive association between NC and the level of uric acid. Further adjustment for age and other confounders still found a positive association between NC and hyperuricemia, suggesting that neck circumference was an independent predictive marker of hyperuricemia in PCOS patients. We also observed signi cant interactions between BMI, WC and HC with NC in relation to serum uric acid concentrations in women with PCOS. Interestingly, the associations between NC and serum uric acid level were more evident in those with medium BMI (22.02 kg/m 2 ≤ BMI < 25.08 kg/m 2 ), low WC (WC < 85 cm) or medium HC (90 cm ≤ HC < 96 cm), which suggests that NC as an indicator for upper-body subcutaneous fat is a pathogenic and independent fat depot that confers additional risks for hyperuricemia. In addition, our results showed that the area under the curve of NC in predicting hyperuricemia appeared comparatively larger than that of BMI, WC and HC. NC has good sensitivity and negative predictive value for the identi cation of hyperuricemia, which could avoid unnecessary medical intervention in some ways.
Several potential mechanisms account for the high prevalence of hyperuricemia in women with PCOS and larger NCs. First, recent compelling evidence indicates that in women with PCOS, NC is a good predictor for insulin resistance (IR) (40), which consequently increases the risk of hyperuricemia by reducing the excretion of uric acid through increased proximal tubular sodium reabsorption(26, 41).
Second, since an increased prevalence of obstructive sleep apnea has been observed in women with PCOS(42), xanthine oxidase, an enzyme that plays a key role in uric acid synthesis, could be activated under hypoxic conditions in those patients (43). Third, increasing evidence has shown that more than 60% of free fatty acids (FFAs) are released from upper-body subcutaneous adipose tissue (44). High levels of FFAs exert pathogenic effects on the glomerulus [42] and tubulointerstitium (45), leading to abnormal renal secretion and reabsorption of uric acid (27,46). In summary, the underlying mechanism of higher uric acid levels in PCOS women with larger NCs lies in more free fatty acid release and higher airway pressure from upper body adiposity tissue, both of which could result in oxidative stress and insulin resistance (20).
This study is the rst to assess the effects of NC as a parameter in the prediction of hyperuricemia in women with PCOS by using ROC curves. The strengths of our study lie in the simple and standardized measuring method of neck circumference, as well as the validated and complete metabolic data, which make our ndings highly applicable to clinical practice. Moreover, the high Youden index, high sensitivity and high negative predictive value of NC in predicting hyperuricemia enable NC to be a more practical predictor in clinical practice. However, several limitations should be taken into consideration. First, the single-center retrospective design has some limitations with regard to interpreting the causality of associations. Second, due to the limited sample size, we did not classify the phenotypes of PCOS and did not calculate the prevalence of hyperuricemia in each phenotype. Thus, prospectively designed studies and follow-up of reproductive outcomes await further evaluation.

Conclusions
In summary, we found a positive correlation between NC and serum uric acid, and NC has a good predictive ability for hyperuricemia in PCOS women. Neck circumference could be recommended as a comparatively simple, fast, and reliable measuring method in the routine clinical assessment of women with PCOS to screen those at high risk of hyperuricemia.