| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
krha,
Tomá
Haas,
Gustav indelka1,
Martin Prázn
,
Ji
í Widimsk,
David Cibula and
t
pán Sva
ina
Third Department of Internal Medicine and Department of Obstetrics and Gynecology, First Faculty of Medicine, Charles University, 128 08 Prague 2, Czech Republic
Address all correspondence and requests for reprints to: Prof. Dr. Jan krha, Third Department of Internal Medicine, First Faculty of Medicine, Charles University, U Nemocnice 1, 128 08 Prague 2, Czech Republic. E-mail: jan.skrha{at}lf1.cuni.cz.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Several techniques have been used for the evaluation of insulin sensitivity in men (1, 2, 3, 4, 5); among them, hyperinsulinemic clamps are considered the gold standard (6). However, the clamps are time consuming and thus can hardly be used in routine clinical practice. The same is true of frequently sampled iv glucose tolerance tests, because a large number of samples have to be examined, and in addition, wide variations in the results have been observed (7, 8). The minimal model is therefore not reliable for the interpretation of data in an individual subject. The oral glucose tolerance test, exploring plasma glucose and insulin levels to evaluate the area under the curves of both analytes, has been used repeatedly in determining insulin sensitivity in addition to insulin secretion (9, 10).
The estimation of fasting plasma glucose and insulin concentrations has been recommended to calculate the homeostasis model assessment (HOMA) as well as the QUICKI indexes (11, 12). Only one blood sample is required, and no other equipment is necessary. Nevertheless, different experiences with the HOMA index have been published in the last few years (13, 14, 15). In addition, the analyses were mainly applied to diabetic patients and healthy subjects.
Insulin action is influenced by central obesity, serum lipid concentrations, arterial hypertension, hormonal changes, and other factors (16, 17, 18, 19). The molecular level of insulin insensitivity has not been fully explained in all of these cases.
The aim of this study was to compare the results of hyperinsulinemic clamps with the HOMA and QUICKI indexes in a large sample of the population involving different clinical states with impaired insulin action such as type 2 diabetes mellitus with and without obesity, insulinoma, essential high renin hypertension, secondary hypertension due to primary hyperaldosteronism, and polycystic ovary syndrome. In addition, the objective of the study was to analyze the influence of obesity, serum lipid concentration, and arterial hypertension on insulin action variables calculated from hyperinsulinemic clamps as well as from the HOMA and QUICKI indexes. We demonstrate for the first time that in very rare clinical situations, both indexes do not correspond with the results of clamps despite a good relationship found in patients with common clinical diagnoses.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A cohort of 168 subjects was enrolled in the study, consisting of 41 type 2 diabetic patients, 20 insulinoma patients, 16 patients with primary hyperaldosteronism, 12 patients with essential high renin hypertension, 32 women with polycystic ovary syndrome (PCOS), and 47 healthy control subjects. Their clinical characteristics are presented in Table 1
.
|
The diagnosis of insulinoma was made by findings of plasma glucose levels below 2.5 mmol/liter, hyperinsulinemia, and a high insulin/glucose ratio (>6.0) as well as by clinical signs of hypoglycemia during a prolonged fasting test. It was later confirmed by surgically removed nesidioma (20). Primary hyperaldosteronism was diagnosed by high plasma aldosterone and low renin concentrations in patients with arterial hypertension and was later confirmed by surgical findings of adenoma or hyperplasia (21). The clamps and other examinations were performed before surgery. Essential hypertension was diagnosed after exclusion of all other possible causes of arterial hypertension, and the presence of a high plasma renin concentration was confirmed in all of these subjects. Antihypertensive treatment was discontinued at least 2 wk before examination.
Women with PCOS fulfilled the diagnostic criteria of this syndrome (22). They had oligomenorrhea from menarche, increased serum concentration of at least one androgen above the upper limit of the reference range, and clinical manifestation of hyperandrogenism (acne, hirsutism, or both). Twenty-two had a body mass index (BMI) below 25 kg/m2; the remaining 10 had a BMI between 25 and 32 kg/m2. All women were normotensive, and their serum lipid concentration levels were within the normal limits.
Impaired glucose tolerance and diabetes mellitus were excluded in all subjects, except for those in the diabetic group, by normal oral glucose tolerance test results. Only subjects without previous kidney and liver disease and with normal biochemical tests were selected, and all active smokers were excluded from the study.
The control group consisted of 32 nonobese (BMI, 
25 kg/m2) and 15 obese (BMI, 
25 kg/m2) subjects. All controls were free of any drug treatment and did not have diabetes in their family histories.
Three blood pressure measurements were performed in all subjects on their nondominant arm using a manual sphygmomanometer. The mean value was used for calculations. Higher values were considered to be a pressure above 140/90 mm Hg.
Informed consent was obtained from all subjects, and the study was prepared in accordance with the Helsinki Declaration and was approved by the ethics committee of the First Faculty of Medicine.
Methods
All subjects were examined after an overnight fast. The patients taking oral antidiabetic drugs received their last dose 12 h before the examination.
The hyperinsulinemic isoglycemic clamp was performed as described previously (1, 16). This clamp was also euglycemic in all nondiabetic subjects. Briefly, a flexible cannula was inserted into the forearm vein to obtain blood samples for determination of basal insulin (I0), plasma glucose (G0), and potassium concentrations. The cannula was then connected to the Biostator infusion module (GCSII, Miles Corp., Elkhart, IN) to administer insulin solution [160 U Actrapid HM (Novo-Nordisk, Copenhagen, Denmark) in 500 ml 0.9% sodium saline solution], 40% glucose solution, and washout sodium saline solution (0.9%, wt/vol). At the same time, 7.5% potassium chloride solution diluted with physiological saline solution 1:4 was administered by perfusor (Infusor, Secura FT, B. Braun, Melsungen, Germany) into another channel of the cannula at a rate of 0.1 ± 0.05 ml/min to maintain basal potassium levels. The rate of this infusion was adjusted according to the results of repeatedly determined serum potassium concentrations. A double-lumen catheter was inserted into the contralateral forearm placed in a warm box to obtain arterialized blood for continuous glucose determination. A third cannula was inserted into a wrist vein for collecting blood samples for biochemical estimations. After a 30-min washout period and 60 min of stabilization of blood glucose levels, the hyperinsulinemic clamp was performed with a Biostator (mode 1:7) for 120 min using a constant insulin infusion rate (1 mU/kg·min) (20). The glucose solution (40%, wt/vol) was sampled to maintain blood glucose levels at a basal value. During the clamp, the blood glucose concentration was repeatedly measured by glucose analyzer (ESAT 6660–2, PWG, Medingen, Germany), and its coefficient of variation was less than 10%. Two blood samples for insulin (IC) determination were collected in the last 20 min of the clamp.
The following variables were used for the evaluation of insulin action from the clamps: plasma glucose, IC, glucose disposal rate (characterizing the sum of insulin-dependent and noninsulin-dependent transport of glucose), metabolic clearance rate of glucose (MCRG; expressed as the ratio of glucose disposal rate to blood glucose concentration) and the insulin sensitivity index (MRCG/IC; both describing insulin action).
The HOMA index was calculated from G0 and I0 concentrations using the formula: (G0 x I0)/22.5 (11). Logarithmic values of basal plasma glucose (log G0) and insulin (log I0) concentrations were used to express the QUICKI index: (l/(log G0 + log I0) (12).
Assays
Plasma glucose concentrations were determined by glucose oxidase method by glucose analyzer (ESAT 6660–2, Medingen, Germany). Plasma insulin concentrations were measured by RIA kits (CIS Bio International, Gif-sur-Yvette, France) with 5.6% intraassay and 7.2% interassay coefficients of variation in our laboratory. The cross-reactivity with proinsulin declared by the manufacturer was less than 14%. Renin and aldosterone concentrations were measured by RIA kits (Immunotech, Prague, Czech Republic), plasma testosterone was determined by a chemiluminescent assay using an ACS:180 autoanalyzer (Bayer Diagnostics GmbH, Leverkusen, Germany), and SHBG was measured using immunoradiometric assay kits (Orion, Ospoo, Finland). Serum cholesterol and triglyceride concentrations were assessed in the Central Laboratory using a Hitachi analyzer (Miami, FL), and glycated hemoglobin A1c was determined by IMx kits using an Abbott analyzer (Chicago, IL).
Statistical analysis
The results were expressed as the mean ± SD, and the differences between the groups were tested using Wilcoxon rank-sum test. Spearman’s correlation was used to assess the relationship between insulin action (expressed by M, MCRG, or MCRG/IC) and BMI, serum cholesterol, and triglyceride concentrations as well as with systolic or diastolic blood pressure. The linear regression models were fitted in the separate groups to show the relationships between the variables (logarithmic transformation was used). Similarly, Spearman’s correlation was applied to demonstrate the association between MCRG and MCRG/IC vs. the HOMA and QUICKI indexes in the entire cohort of subjects as well as in the separate groups.
Using the data from the lean healthy subjects, the cut-off limit of insulin sensitivity (as measured by MCRG/IC) was determined as a value separating the lower 5% values. Similarly, the insulin-sensitive ranges for both HOMA and QUICKI indexes were determined in these insulin-sensitive subjects. Each subject could then be classified as either insulin sensitive or insulin resistant based on the insulin clamp (considered here as a standard) as well as on HOMA and QUICKI indexes. The sensitivity and specificity of the HOMA and QUICKI indexes as related to MCRG/IC values were calculated in separate groups of subjects.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. Although patients with type 2 diabetes or insulinoma and both groups of hypertensive patients were older compared with controls, in the control group there was a broad spectrum of ages. No significant influence of age on insulin action variables and other biochemical parameters was found in the separate groups of subjects.
Insulin action parameters calculated from clamps and also expressed by the HOMA and QUICKI indexes are shown in Table 2. In the control group the mean values of insulin action parameters (M, MCRG, MCRG/IC, HOMA, and QUICKI indexes) were influenced by the presence of a mixed population of both nonobese and obese individuals evaluated together. The effect of BMI on the above variables in healthy subjects is shown in Table 3. Therefore, we defined normal insulin sensitivity calculated from clamps in our lean healthy controls to be above 9.8 ml/kg·min per mU/liter x 100 for MCRG/IC (95% probability), corresponding to HOMA values below 3.20 and QUICKI values above 0.54.
|
|
The relationship between BMI and insulin sensitivity parameters was assessed using the Spearman correlation coefficient (Table 4). BMI showed the strongest association with insulin sensitivity in the entire cohort of examined subjects. A significant inverse relationship was found between BMI and glucose disposal rate, MCRG, and MCRG/LIC (P < 0.0001).
|
). The linear regression models were fitted in the separate groups to show the association between BMI and a logarithmically transformed MCRG/LIC. The lack of a relationship between BMI and insulin sensitivity was observed in the patients with insulinoma and primary hyperaldosteronism (Table 4).
|
. Serum cholesterol negatively influenced MCRG only in healthy subjects (r = -0.35; P < 0.02) and type 2 diabetic patients (r = -0.38; P < 0.02). No relationship was found between serum cholesterol and insulin sensitivity parameters in patients with insulinoma, polycystic ovary syndrome, primary hyperaldosteronism, or high renin hypertension. Serum triglycerides were inversely related to the MCRG/I in the healthy controls (r = -0.41; P < 0.005), in type 2 diabetes mellitus (r = -0.44; P < 0.005), and the most strongly in PCOS (r = -0.50; P < 0.005). No relationship was observed between serum triglycerides and insulin sensitivity in patients with insulinoma, primary hyperaldosteronism, or high renin hypertension. Systolic and diastolic blood pressures were inversely related to the MCRG/I only in healthy subjects (r = -0.37; P < 0.01 or r = -0.42; P < 0.005, respectively), whereas only diastolic blood pressure was inversely related to the MCRG/I in insulinoma patients (r = -0.67; P < 0.001). This effect disappeared in the other groups.
|
. A significant inverse relationship was found in the healthy controls, women with PCOS, and type 2 diabetic patients (P < 0.001). On the contrary, the HOMA indexes did not correlate with insulin sensitivity indexes in the patients with insulinoma and primary hyperaldosteronism (Fig. 2).
|
|
).
|
). The estimation of sensitivity for both indexes in subjects with diabetes, primary hyperaldosteronism, or high renin hypertension was limited by the absence of sufficient number of insulin-sensitive subjects.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In the present study we estimated insulin sensitivity in subjects with different clinical conditions using the clamp technique and comparing it with the HOMA and QUICKI indexes, which are suggested as simple and reliable parameters of insulin resistance. The correlation analysis demonstrated a strong association between BMI and insulin resistance in the entire population. However, a strong relationship between BMI and insulin sensitivity parameters or HOMA and QUICKI indexes was found only in the control subjects, type 2 diabetic patients, and less so in the women with PCOS. The relationship disappeared in the patients with insulinoma and primary hyperaldosteronism and was diminished in those with high renin hypertension. Our results partly correspond to those of Mari et al. (28), who found a significant relationship between BMI and clamp results in the mixed normal and diabetic population, but not in the separate groups. Emoto et al. (29) observed a highly significant correlation of both variables in type 2 diabetes mellitus. On the contrary, no similar relationship was found by others (14). We suppose that these differences are caused by the fact that BMI does not entirely reflect central obesity as contributing to this relationship. Different population samples with respect to adipose tissue distribution may be recruited in these studies, demonstrating less comparable relationships of BMI with insulin action.
We found that both serum cholesterol and triglyceride concentrations were inversely related to insulin sensitivity only in healthy controls and type 2 diabetic patients, whereas triglycerides alone related to insulin sensitivity in women with PCOS. The association of insulin resistance with dyslipidemia in type 2 diabetic patients has been repeatedly demonstrated (17, 30), and our data may support this finding. However, we could not confirm this observation in the other groups of subjects without diabetes. Systolic and diastolic blood pressures showed an inverse relationship in the controls. Different results comparing blood pressure and insulin sensitivity were reported in other studies (17, 18, 29, 30). This means that data obtained from correlation analysis are strongly dependent on the characteristics of the respective population.
The correlation analysis comparing insulin clamps with the HOMA and QUICKI indexes demonstrated that a relationship does exist among all parameters in the healthy controls, women with PCOS, and type 2 diabetic patients. In patients with high renin hypertension, both indexes correlated only with the insulin sensitivity index, whereas no relationship of the clamp parameters and HOMA and QUICKI indexes was found in patients with insulinoma or primary hyperaldosteronism. The results of other studies confirmed a good relationship between insulin clamp parameters and HOMA (13, 29, 31) or QUICKI (12) indexes in type 2 diabetic patients. However, a limited value of the HOMA index was demonstrated in older men with impaired glucose tolerance (14). Basal insulin and glucose concentrations, but not stimulated values, are used in the HOMA and QUICKI indexes. They reflect spontaneous homeostatic characteristics based on insulin affecting hepatic glucose production in the fasting state, but not its peripheral action in the postprandial state (13, 14). On the contrary, the hyperinsulinemic clamp may provide a measure of peripheral insulin action as a response to a higher insulin concentration (32, 33). A strong correlation may be expected in situations when both peripheral and hepatic resistances are present (24). They reflect most situations in both the physiological and disease states (i.e. hypertension, obesity, dyslipidemias, pregnancy, PCOS, etc.), where changes in hepatic insulin sensitivity parallel changes in peripheral insulin sensitivity.
Different results were observed in patients with insulinoma and primary hyperaldosteronism, both representing rare conditions influencing insulin action. A significantly higher insulin resistance was expressed by clamp variables compared with the HOMA index in primary hyperaldosteronism, whereas in the insulinoma patients a significantly higher insulin resistance was calculated by the HOMA index. Primary hyperaldosteronism represents a case where peripheral and hepatic insulin sensitivities are uncoupled. High fasting plasma insulin levels accompanied by low plasma glucose concentrations reflect spontaneous conditions in patients with insulinoma. We did not use exogenous glucose before examination, and therefore the HOMA and QUICKI indexes could characterize steady state conditions. This may cause the relatively higher values of the HOMA index and the lower QUICKI index in comparison with clamp results due to high glucose clearance. We speculate that the main defect of insulin action in the insulinoma patients could be localized in the liver, whereas primary hyperaldosteronism could be associated with more advanced insulin resistance in the peripheral tissue, but further studies confirming this hypothesis will be necessary. These rare clinical states demonstrate differences between the HOMA index and clamp parameters expressing insulin action. The results of the QUICKI index were inversely related to HOMA values in all groups of subjects.
Calculation of the sensitivity and specificity of the HOMA and QUICKI indexes as related to the clamp results demonstrated differences in their abilities to predict insulin sensitivity or insulin resistance in different clinical diagnoses. However, the results are limited by the low number of insulin-sensitive subjects in separate groups of patients, which may underscore sensitivity for both indexes.
We used isoglycemic, instead of euglycemic, clamps in diabetic patients because morning plasma glucose levels were considered a result of the homeostatic regulations reached after an overnight fast. However, in these patients the glucose disposal rate involves both insulin-dependent and noninsulin-dependent glucose transportation performed by a mass effect of glucose (34). We therefore evaluated the MCRG and MCRG/IC to more properly express insulin action rather than by glucose disposal itself. We did not find any relationship between the HOMA or QUICKI indexes and the glucose disposal rate in type 2 diabetic patients. In addition, the single insulin infusion (1 mU/kg·min) used in this study may not be appropriate for all subjects, especially for patients with high insulin resistance (some of the type 2 diabetic patients) or highly hormonally active insulinoma patients (with a high insulin/glucose ratio) compared with more insulin-sensitive subjects. A constant amount of insulin infused may cause a different response in subjects with significantly different insulin action. This potential limitation has to be considered when evaluating the results.
The diabetic subjects were not withdrawn from their chronic therapy, because our aim was to analyze different parameters evaluating insulin action in subjects receiving chronically balanced antidiabetic treatment. This may offer results on actual insulin sensitivity in these individuals.
In summary, our results demonstrate that the HOMA and QUICKI indexes reflect insulin action expressed by clamps in most of the subjects. However, they do not offer the same information as the clamps in rare cases with a predominance of peripheral insulin resistance. The comparison of clamps with HOMA and QUICKI indexes may therefore be valuable because they express different aspects of insulin action. Our findings also demonstrate a dominant role of BMI in impaired insulin action in different clinical conditions.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: BMI, Body mass index; G0, basal plasma glucose; HOMA, homeostasis model assessment; IC, plasma insulin concentration; I0, basal insulin; MCRG, metabolic clearance rate of glucose; MCRG/I, insulin sensitivity index; PCOS, polycystic ovary syndrome.
1 G.S. died tragically before preparation of the manuscript. 
Received December 31, 2002.
Accepted September 15, 2003.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
indelka G, krha J, Prázn M, Haas T 2002 Association of obesity, diabetes, serum lipids and blood pressure regulates insulin action. Physiol Res 51:85–91[Medline]
krha J, indelka G, Haas T, Hilgertová J, Justová V 1996 Comparison of insulin sensitivity in patients with insulinoma and obese type 2 diabetes mellitus. Horm Metab Res 28:595–598[Medline]
J, indelka G, Haas T, Prázn M, Hilgertová J, krha J 2000 Impaired insulin action in primary hyperaldosteronism. Physiol Res 49:241–244[Medline]
krha J, Hill M, Fanta M, Haaková L, Vrbíková J, 
ivn J 2002 Prediction of insulin resistance in non-obese women with polycystic ovary syndrome. J Clin Endocrinol Metab 87:5821–5825This article has been cited by other articles:
 |
|
 |
|
  A. W. Krug and M. Ehrhart-Bornstein Aldosterone and Metabolic Syndrome: Is Increased Aldosterone in Metabolic Syndrome Patients an Additional Risk Factor? Hypertension, May 1, 2008; 51(5): 1252 - 1258. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Lee, R. Muniyappa, X. Yan, H. Chen, L. Q. Yue, E.-G. Hong, J. K. Kim, and M. J. Quon Comparison between surrogate indexes of insulin sensitivity and resistance and hyperinsulinemic euglycemic clamp estimates in mice Am J Physiol Endocrinol Metab, February 1, 2008; 294(2): E261 - E270. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Muniyappa, S. Lee, H. Chen, and M. J. Quon Current approaches for assessing insulin sensitivity and resistance in vivo: advantages, limitations, and appropriate usage Am J Physiol Endocrinol Metab, January 1, 2008; 294(1): E15 - E26. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Hitomi, H. Kiyomoto, A. Nishiyama, T. Hara, K. Moriwaki, K. Kaifu, G. Ihara, Y. Fujita, T. Ugawa, and M. Kohno Aldosterone Suppresses Insulin Signaling Via the Downregulation of Insulin Receptor Substrate-1 in Vascular Smooth Muscle Cells Hypertension, October 1, 2007; 50(4): 750 - 755. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Beysen, E. J. Murphy, T. McLaughlin, T. Riiff, C. Lamendola, H. C. Turner, M. Awada, S. M. Turner, G. Reaven, and M. K. Hellerstein Whole-Body Glycolysis Measured by the Deuterated-Glucose Disposal Test Correlates Highly With Insulin Resistance In Vivo Diabetes Care, May 1, 2007; 30(5): 1143 - 1149. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Bowers, D. Albanes, P. Limburg, P. Pietinen, P. R. Taylor, J. Virtamo, and R. Stolzenberg-Solomon A Prospective Study of Anthropometric and Clinical Measurements Associated with Insulin Resistance Syndrome and Colorectal Cancer in Male Smokers Am. J. Epidemiol., October 1, 2006; 164(7): 652 - 664. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Catena, R. Lapenna, S. Baroselli, E. Nadalini, G. Colussi, M. Novello, G. Favret, A. Melis, A. Cavarape, and L. A. Sechi Insulin Sensitivity in Patients with Primary Aldosteronism: A Follow-Up Study J. Clin. Endocrinol. Metab., September 1, 2006; 91(9): 3457 - 3463. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Bridger, S. MacDonald, F. Baltzer, and C. Rodd Randomized Placebo-Controlled Trial of Metformin for Adolescents With Polycystic Ovary Syndrome Arch Pediatr Adolesc Med, March 1, 2006; 160(3): 241 - 246. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Chen, G. Sullivan, and M. J. Quon Assessing the Predictive Accuracy of QUICKI as a Surrogate Index for Insulin Sensitivity Using a Calibration Model Diabetes, July 1, 2005; 54(7): 1914 - 1925. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Harbis, S. Perdreau, S. Vincent-Baudry, M. Charbonnier, M.-C. Bernard, D. Raccah, M. Senft, A.-M. Lorec, C. Defoort, H. Portugal, et al. Glycemic and insulinemic meal responses modulate postprandial hepatic and intestinal lipoprotein accumulation in obese, insulin-resistant subjects Am. J. Clinical Nutrition, October 1, 2004; 80(4): 896 - 902. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
), insulinoma (INS; 
), primary hyperaldosteronism (PHA; 
), essential high renin hypertension (HRH; •), women with PCOS (
), and the control group (C; 
).
