Copyright© &M all rights reserved. 128 to 2015 of NAFLD is as high as 42% according HongKong, theprevalence Chinese populationfrom (1).Inthe has becomeamajorpublic health problem overtimeand generalpopulation,has increased ofNAFLD,whichis5%-58%in worldwide prevalence consumption andanyother causeofliverdiseases.The in patients without significant alcohol and cirrhosis, to nonalcoholic steatohepatitis to advanced fibrosis simplehepaticsteatosis ofliverdamagefrom spectrum N INTRODUCTION Correspondence to: University, Guangzhou,China ofSun Memorial Hospital -sen and Gynecology, Sun Yat-sen 2 Guangzhou, China of Sun Yat-sen University, Yat-sen MemorialHospital 1 article original DOI: 10.20945/2359-3997000000119 Accepted onDec/12/2018 Received onOct/24/2018 [email protected] Guangzhou, China West Yanjiang Road 107, of Sun Yat-sen University Sun Yat-sen Memorial Hospital Gynecology, Department ofObstetricsand Dongzi Yang Department ofObstetrics Department ofUltrasound,Sun commonest diffuse liverdisease,comprisingthe commonest diffuse onalcoholic fattyliverdisease(NAFLD)isthe

Could semiquantitative analysisof Na Di polycystic ovary syndrome? liver diseaseinpatientswith alterations ofnonalcoholicfatty distinguish more liver parenchyma real-time ultrasound elastography metabolic disturbances syndrome;nonalcoholicfattyliverdisease;real-timeultrasoundelastography;semiquantitativeanalysis; Polycystic ovary Keywords disease needfurther research. syndrome moresensitively. The diagnosticvalueoftheproposedmethodfornonalcoholicfatty liver elastography parenchyma analysiscoulddistinguishliver alterationsinpatientswithpolycysticovary subjects withoutnonalcoholicfatty disease. liver associatedwithaveragestrainvalueamong metabolic syndromewerestand-aloneindependentfactors 0.721-0.895). Inmultiple linearregressionanalysis,polycysticovary syndrome,waist circumference,and value ofaveragestrainforelevatedaminotransferase,theareaundercurvewas 0.808(range differed significantlybetween thetwosubgroupswithloworhighaveragestrainvalue. For diagnostic and control(87.02 ±10.16 vs.96.31±11.44 vs.104.49 ±7.28, p<0.001). Clinicalandlaboratoryparameters nonalcoholic fatty disease,polycysticovary syndromewithoutnonalcoholicfatty liver disease, liver such as average strain value,differed significantly amonggroups polycystic ovary syndromewith of real-timeultrasoundelastography oftheliver. age wereincluded. All participants analysis underwentultrasonicexaminationandsemiquantitative syndrome patientswithout nonalcoholic fatty disease, and 70healthy liver femalecontrolsofreproductive Thirty-five polycystic ovary syndromepatientswithnonalcoholicfatty disease,70polycystic ovary liver nonalcoholic fatty diseaseinpatientswithpolycysticovary syndrome. liver ofreal-timeultrasoundelastography strainparameters for the diagnosticvalueofsemiquantitative with polycysticovary syndromeareatanincreasedrisk. The aimofthepresentstudywas toassess Objective: ABSTRACT Linlin Jiang 1,2 , Xinchuan Zhou Nonalcoholic fatty diseaseisthecommonestdiffuse disease,ofwhich liver liver women 2 , BaomingLuo 1 Arch EndocrinolMetab. 2019;63(2):128-36 , YaxiaoChen 1 , XiaoliChen from 23.8%to86.79% among womenwithPCOS from of NAFLDvaries theprevalence previously: as reported Women risk of NAFLD at an increased with PCOS are diseasesforpatientswithPCOS(4). of cardiovascular risk disturbances, whichmaypose ahighlong-term oftheaccompanyingmetabolic to highprevalence to 20%(1,3).Itisalsoametabolicdiseaseaccording 5% variesfrom age,andtheprevalence reproductive among women of common endocrinedisorder metabolic disorders. (2).NAFLDisalsousedasamarkerofsevere report Polycystic ovary syndrome (PCOS)isthemost syndrome Polycysticovary 2 , XiaomiaoZhao Results: Conclusion: 2 ,Dongzi Yang Main semi quantitative strain parameters, strainparameters, Main semiquantitative Semiquantitative real-time ultrasound Semiquantitative real-timeultrasound 2 Arch Metab. Endocrinol 2019;63/2 2 , LinLi Subjects andmethods: 2 , Arch Metab. Endocrinol 2019;63/2 calculated by PASScases andcontrols 11(NCSS LLC, of6.39inthevalue MEANbetween a difference hasa power of0.949at5%significancetodetect group hepatitis B(18),asamplesizeof5ineach in chronic aboutRTE forassessmentofliver fibrosis to report diagnose ofNAFLDin patients with PCOS. According of RTE in theefficacy studyhadreported No previous Sample sizeestimate SUBJECTS ANDMETHODS these parametersforNAFLDinpatientswithPCOS. andthento evaluate thediagnosticvalueof controls tohealthy with orwithoutNAFLDascompared strain parameters of RTE of the liver inPCOSpatients to analyzesemiquantitative The aimsofourstudywere age. neverinPCOSwomenofreproductive reported, semiquantitative parameters toNAFLD is rarely (12-18). fibrosis to useful for early diagnosis and staging of liver the averagestrainvalue(MEAN)ofRTE reported are analysis ofstrainwasdeveloped.Parametersincluding inRTE, theimprovements semiquantitative With liverdiseases(10,11). inpatientswithchronic fibrosis RTE is alsousefulforassessmentof reports, to recent diagnosis offocallesionsinvariousorgans. According been rapidlyevolvinginthelastdecade,beingusedfor one oftheultrasoundelastographytechnologies,has age.Real-timetissueelastography(RTE), reproductive monitoringofNAFLDinPCOS patientsof frequent It isimpracticaltousethesemethodsforassessmentand of the tissue. aboutmechanicalproperties information itcannotprovide operator’s experience.Furthermore, limited bysubcutaneousfatthicknessofpatientsandthe detection ofNAFLDindailypractice,butitsaccuracyis Conventional ultrasonographyhasbeenusedfor costly(9). ionizing radiation,andbothmethodsare imaging (MRI)canquantifyliverfat,butCTinvolves Computed tomography (CT) or magnetic resonance variability(8). andinterobserver pain, samplingerrors, suchasbleedingandsevere with possibleproblems isinvasiveandassociated thisprocedure fibrosis, fordetectionofaliverdiseasesuchas gold standard actively (7). forliverdiseasemore with PCOSshouldbescreened (5,6). Accordingly, ithasbeensuggestedthatwomen On theotherhand,applicationofRTE with the Although a liver biopsy is still considered Real-time ultrasound elastography for nonalcoholicfatty liver diseaseinPCOSpatients 3) polycystic ovaries (PCO) (presence in each ovary of in each ovary 3) polycystic ovaries (PCO) (presence 2) clinicaland/orbiochemicalhyperandrogenism, and/oranovulation, had tobemet:1)oligomenorrhea criteria 2003 criteria(19), two ofthefollowingthree (REDCap) databases. DataCapture Electronic deposited intheResearch were OOC-15006452. Detaileddataonallparticipates (http://www.chictr.org/en/) numberis ChiCTR- consent.TheChineseClinicalTrialinformed Registry written provided ofourHospital.Allparticipants Board bytheInstitutionalReview The studywasapproved Patients andcontrols May27,2015,toNovember31,2015. Hospital from inour recruited aged20to40yearswere controls PCOS patientswithoutNAFLD,and70healthyfemale Atotalof35PCOSpatientswithNAFLD,70 group. USA). rec COULTER, Inc.USA)duringthe earlyfollicular a automatic biochemistry quantified by chemiluminesce Immunoassayonan (TT)were estradiol (E2),andtotal testosterone (PRL), (LH),prolactin luteinizing hormone (DBP). (SBP), anddiastolicbloodpressure (mFG) forbodyhairevaluation,systolicbloodpressure score modifiedFerriman-Gallwey (BMI), acnescore, (HC), waistline/hiplineratio(WHR),bodymassindex (WC),hipcircumference including waistcircumference andrecorded measured datawere anthropometric All medications, and past diseases was recorded. regularity, includingmenstrual The medicalhistory Bioclinical tests orliverfunctionduringthelast3months. pressure, metabolism,blood taking medicationthatmayaffect diseaseorifthey were cardiocerebrovascular severe and alcoholdrinking,otherendocrinediseases,ora ovarian surgery, amalignanttumor, tobaccosmoking, thatincludedotherknownliverdiseases, medical history NAFLD. PCO,and women withoutendocrinedisorders, amenorrheic definedashealthynonpregnant, were criteria (conventionalultrasonography)(20). toconventional NAFLD wasdiagnosedaccording than10mL). ovarianvolume:more and/or increased folliclesmeasuring2to9mm indiameter 12 ormore PCOS was diagnosed according to the Rotterdam totheRotterdam PCOS wasdiagnosedaccording Basal follicle-stimulating hormone (FSH), Basal follicle-stimulating hormone Women thestudyiftheyhada excludedfrom were ruited more than 5 participants foreach than5participants more ruited nalyzer (DXI800,Beckman

Controls Controls 129

Copyright© AE&M all rights reserved. Copyright© AE&M all rights reserved. 130 linear transducer (L52)wasusedforRTE. A7-3MHz witha3.5 MHz probe. US wasperformed ultrasonography and Medical, Tokyo, Japan),whichhasbothconventional by meansofHIVISION PREIRUS(Hitachi clinical dataofalltheparticipants. for thelast5years.Theoperatorwasblindedto ultrasonography who hadspecialized in elastography with 10yearsofexperienceintransabdominal was recorded. irregular.if menseswere Antral follicle count(AFC) orrandomly regular early follicularphaseifmenseswere inthe Transvaginal ultrasonographywasperformed Ultrasound examination andRTE NCEP-ATP IIIcriteria(27). to (MS)wasdiagnosed according Metabolic syndrome defined asSBP≥130mmHgorDBP85mmHg. (EBP) was waistline ≥ 80 cm. Elevated blood pressure (AST) ≥40U/L.Centralobesitywasdefinedasthe transaminase alanine transaminase(ALT) oraspartate ≥ 3.6mM.Elevatedaminotransferasewasdefinedas (LDL-C) cholesterol mM, orlow-densitylipoprotein (CHOL)≥6mM,triglycerides (TG)≥2.3 as cholesterol to 2006WHOcriteria(26).Hyperlipidemiawasdefined diagnosedaccording type 2diabetesmellitus(DM)were glucosetolerance(IGT)and tolerance (FGT),impaired fastingglucose FIN ≥12.6mIU/L(24,25).Impaired (IR)wasdefinedasHOMA-IR ≥2.14and resistance glucose (FBG)×fastinginsulin(FIN)÷22.5.Insulin (HOMA-IR)wascalculated as fastingblood resistance (23).Homeostaticmodelassessmentforinsulin report toaprevious according wasmeasured The acnescore (22). reported was definedasmFG≥5previously (21).Hirsutism levelof2.39nMorgreater testosterone (bHA) was defined as a total hyperandrogenism criteria.Biochemical Rotterdam totherevised according (ADVIA Centaurxp,SIEMENS,Germen). an automaticchemiluminescenceimmunoassaysystem methods on using standard measured parameters were transaminases,plasmaglucose,insulin,andlipid Serum Inc.USA). (BiotekInstruments, ELx808 platereader enzyme-linked immunosorbent assay (ELISA) on a analyzedbyan binding globulin(SHBG)were (AMH) andsexhormone- Anti-Müllerian hormone pha Real-time ultrasound elastography for nonalcoholicfatty liver diseaseinPCOSpatients The ultrasonic examination (US) was performed The ultrasonicexamination (US)wasperformed byasingleoperator performed were All procedures defined and PCOwere Oligomenorrhea/amenorrhea se or first 3 days of progestin withdrawalbleeding. se orfirst3daysofprogestin RTE capabilities.Conventional or as a rate. The one-sample Kolmogorov-Smirnov or asarate.Theone-sampleKolmogorov-Smirnov deviation as mean±standard presented are The results Data analysis value (SD)wasalsodetermined. strain deviation of the relative Standard liver stiffness. (AREA, %) asthemaincharacteristicsof strain region ratio ofalow- units(a..)andthearea 0-255 arbitrary elastography imagingincludingMEANintherangeof value. We 11characteristicparametersof measured strain asarepresentative served measurements three foreachpatient.Themeanofthetotal performed acquisitionswere effective generate histograms.Three RTE (EZU-TESH1,Hitachi)wasusedto software The subjects’ownheartbeat. tissue straincamefrom causingliver bile ductsandcysts.Thepressure the Glisson’scapsuleinrightlobe,avoiding windowwassetto1cmunder ofinterest The region PCOS without NAFLDhadasignificantly higher betweenthetwoPCOS difference ( group in thecontrol than muchhigherinthePCOSgroups and AFCwere LDL-C, apoB,AMH,LH,TT, cycles, mFG,menstrual ( groups amongthethree an inversetrend ( group > control PCOSwithoutNAFLD PCOS withNAFLD>group group inthefollowingorder: significantly decreased (2hGLU), FIN,HOMA-IR,TG,ALT, andASTlevels BMI, WC,WHR,SBP, DBP, FBG,two-hourglucose RESULTS p USA), andstatisticalsignificancewasassumedat (SPSS,Inc.,Chicago,IL, version 13.0forWindows intheSPSS software, performed These analyseswere between MEANorAREAandclinicalparameters. analysiswasusedtoaccessthecorrelation regression of MEANforelevatedaminotransferase.Multiplelinear generatedtoanalyzethediagnosticvalue were curves test.Receiveroperatingcharacteristic(ROC) squared usingthe chi- among the groups compared data were Whitney by nonparametric tests (Mann- compared variants were (ANOVA). notassumed, Whenequalvarianceswere usingthe groups amongthe compared MEANswere variants wasnormal. whether the distributionof test was used to determine < 0.05. RTE was performed as reported previously (16). previously asreported RTE wasperformed U testorKruskal-Wallis test). Categorical t p test orone-wayanalysisofvariance <0.05).HDL-Clevelsshowed p <0.05),withoutsignificant Arch Metab. Endocrinol 2019;63/2 groups. Group Group groups. p < 0.05). <

Arch Metab. Endocrinol 2019;63/2 a P Values aremean±standard. Table 1. (42.49±9.41vs.33.58 group lowest inthecontrol PCOSwithoutNAFLD,and ingroup intermediate PCOS with NAFLD, value was the highest ingroup ± 11.44 vs. 104.49 ± 7.28, (87.02±10.16vs.96.31 group highest inthecontrol PCOSwithoutNAFLD,and ingroup intermediate PCOSwithNAFLD, value wasthelowestingroup ( group control PCOSwithNAFLDand level ofapoAthandidgroup Comparedby T testoroneway ANOVA; 1 AFC Longtest Menstrual cycle Shortest Menstrual cycle Acne score mFG LH FSH PRL TT AMH AST ALT apoB apoA LDLC HDLC TG CHOL HOMA-IR FIN 2hGLU FBG DBP SBP WHR WC BMI Age N : ComparisonbetweenPCOSwithandwithoutNAFLD Groups. P a b b (nmol/L) b (mIU//mL) (mmol/L) RTE for livers are shown in Figure 1.TheMEAN RTE showninFigure forliversare a b a a (mIU/L) a b a b a b (cm) a (U/L) (kg/m b (ng/mL) (mIU//mL) (U/L) b (mmHg) (mmol/L) a (mmHg) b b b (ng/dL) b b (g/L) (g/L) (mmol/L) (mmol/L) (mmol/L) b b b Baseline clinical characteristics and laboratory parametersofallsubjects Baseline clinicalcharacteristicsandlaboratory (day) b (day) 2 b ) p < 0.05)(Table 1). 1.9355 ±0.7381 0.8805 ±0.0528 0.1667 ±0.3791 108.07 ±79.06 122.68 ±12.79 35.32 ±12.47 51.52 ±35.63 27.44 ±16.25 25.39 ±13.95 79.89 ±9.51 86.23 ±7.94 25.41 ±3.40 15.11 ±7.67 With NAFLD 9.99 ±4.67 0.88 ±0.24 1.36 ±0.33 3.25 ±1.08 1.21 ±0.25 1.92 ±1.02 5.01 ±1.35 6.01 ±4.68 7.75 ±2.39 5.28 ±0.83 3.10 ±3.29 7.53 ±4.61 6.89 ±1.87 39 ±30.91 29 ±4.41 PCOS b 35 ComparedbyNonparametric Test (Mann Whitney UtestorKruskal-Wallis test). p < 0.001).The AREA 0.4706 ±0.7025 0.8337 ±0.0619

117.18 ±10.38 1.80 ±1.1471 72.62 ±45.56 42.87 ±23.89 PCOS without 31.15 ±9.45 14.53 ±7.80 14.69 ±8.74 15.52 ±4.50 14.15 ±6.68 75.85 ±8.96 77.80 ±8.21 22.81 ±2.67 28.83 ±4.51 3.78 ±3.93 9.12 ±7.30 6.50 ±2.11 9.67 ±4.52 0.79 ±0.21 1.57 ±0.36 2.90 ±0.65 1.53 ±0.37 1.25 ±0.72 4.74 ±0.74 3.07 ±1.64 6.50 ±1.94 4.91 ±0.45 2 : ComparisonbetweenPCOSwithNAFLDandcontrol Groups. P NAFLD 70 Real-time ultrasound elastography for nonalcoholicfatty liver diseaseinPCOSpatients 0.4265 ±0.6063 1.2114 ±0.5210 0.8013 ±0.0735 111.01 ±9.48 72.91 ±8.53 21.15 ±2.61 28.97 ±4.71 14.49 ±3.28 31.68 ±2.84 27.70 ±2.51 15.09 ±6.81 15.17 ±3.19 11.03 ±4.89 69.66 ±7.14 0.01 ±0.12 4.23 ±1.53 7.84 ±2.55 4.52 ±3.16 0.70 ±0.12 1.40 ±0.23 2.55 ±0.67 1.44 ±0.24 0.81 ±0.28 4.59 ±0.72 1.96 ±1.34 9.20 ±5.51 5.53 ±0.93 4.72 ±0.31 Control 70 subjects were subdivided into group A(MEANvalue subdividedintogroup subjects were All175 strainvaluein70healthycontrols. relative oftheaverage(< 91.7)wassettothe5thpercentile valueofMEAN toMEANlevels.Thecutoff according subdividedintotwosubgroups subjects were 64.08 ±6.73vs.58.256.58, asforAREA(68.78±5.21vs. same descendingorder inthe groups amongthethree significantly different ± 10.91vs.24.558.14, To validatetheclinicalutilityofMEAN,all < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.039 0.001 0.002 0.005 0.007 0.033 0.021 0.001 NS NS NS NS NS NS NS NS NS NS NS NS NS P 1 3 : ComparisonbetweenPCOSwithoutNAFLDand control Groups. p < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.001 0.002 < 0.001).SD values were < 0.001).SDvalueswere NS NS NS NS NS P 2 p < 0.001) (Figure 2). < 0.001)(Figure < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.001 0.030 0.017 0.010 0.002 0.012 0.004 NS NS NS NS NS P 3 131

Copyright© AE&M all rights reserved. Copyright© AE&M all rights reserved. 132 between PCOS Without NAFLDandcontrolGroups. P1: ComparisonbetweenPCOS withandwithoutNAFLDGroups. P2: Comparison between PCOSwithNAFLDandcontrolGroups. P3: Comparison differences ofMEAN, SD, AREA betweengroups. There weresignificanttrendfromPCOSwithNAFLD group, PCOSwithoutNAFLDgrouptocontrolgroup. Figure 2. Figure 1. inage,2hGLU,CHOL,apoA,acne no differences B ( A than in group in group significantlyhigher cycles,andAFCwere menstrual IR, TG,LDL-C,apoB,ALT, AST, AMH,mFG, 126 cases). B(MEANvalue≥91.7,< 91.7,49cases)andgroup Real-time ultrasound elastography for nonalcoholicfatty liver diseaseinPCOSpatients BMI, WC,WHR,SBP, DBP, FIN,FBG,HOMA- 100.00 120.00 20.00 40.00 60.00 80.00 PCOS with NAFLD 0.00 A RTE forlivers. Comparison ofRTE parametersbetweengroupsPCOSwithNAFLD, PCOSwithoutNAFLD, andcontrol. Columnbargraphs showsignificant

PCOS without NAFLD MEAN

Controls P P P 3 2 1 <0.001 <0.001 <0.001 p < 0.05). There were were < 0.05). There 10.00 20.00 30.00 40.00 50.00 60.00 PCOS with NAFLD 0.00 C

PCOS without NAFLD AREA and MS was significantly higher in group Athan and MSwassignificantly higheringroup PCOS, IR,IFG,hyperlipidemia,centralobesity, EBP, PCO, ofoligomenorrhea, B.Theprevalence group Athanin disturbances wasobviouslyhigheringroup (Tablegroups 2). andotherendocrineparametersbetweenthetwo score, Prevalence of endocrine disorders andmetabolic ofendocrinedisorders Prevalence

Controls P P P 3 2 1 <0.001 <0.001 <0.001 20.00 30.00 50.00 60.00 70.00 80.00 10.00 40.00 PCOS with NAFLD 0.00 B

Arch Metab. Endocrinol 2019;63/2 PCOS without NAFLD SD

Controls P P P 3 2 1 <0.001 <0.001 <0.001 b a Values aremean±standard. Arch Metab. Endocrinol 2019;63/2 the MEAN level of 0.808 (range 0.721-0.895). With was underthecurve elevated aminotransferase:the area generated toanalyzethediagnostic valueofMEANfor (Tablebetween thetwogroups 3). ofHAandhirsutism intheprevalence no differences A( higher ingroup ofIGTandDMwasslightly 0.05). Theprevalence A( aminotransferase wasalsohigheringroup B( in group formed onthebasisofcutoffvalueMEAN Table 2. Comparedby T testoroneway ANOVA. ComparedbyNonparametric Test (Mann Whitney UtestorKruskal-Wallis test). AFC Longtest Menstrual cycle Shortest Menstrual cycle Acne score mFG LH FSH PRL TT AST ALT apoB apoA AMH LDLC HDLC TG CHOL HOMA-IR F IN 2hGLU FBG DBP SBP WHR WC BMI Age N a a b Receiver operating characteristic (ROC) curves were were Receiver operatingcharacteristic (ROC)curves a b a a b a b a a a b a a a a a a a a a b b b (day) b (day) Comparison ofclinicalcharacteristicsbetweenthegroups a p 0.4091 ±0.7256 0.8683 ±0.0644 < 0.05). The prevalence ofelevated <0.05).Theprevalence 22.904 ±14.23 120.90 ±10.20 (MEAN <91.7) 96.86 ±69.07 48.93 ±26.01 28.07 ±27.04 22.66 ±13.12 31.9 ±12.68 13.75 ±7.96 78.73 ±7.53 84.73 ±8.70 24.84 ±2.73 29.31 ±4.47 3.41 ±3.53 7.89 ±7.42 6.87 ±2.14 1.74 ±0.62 9.79 ±4.82 0.84 ±0.21 1.43 ±0.36 3.07 ±0.92 1.30 ±0.35 1.72 ±0.99 4.83 ±1.06 5.36 ±4.27 7.51 ±2.20 5.20 ±0.76 Group A p 49 > 0.05). Nevertheless, there were were there >0.05).Nevertheless,

0.3934 ±0.6106 0.8147 ±0.0682 113.68 ±11.05 (MEAN ≥91.7) 21.86 ±10.73 50.12 ±37.50 35.69 ±22.04 15.23 ±7.11 15.82 ±8.79 13.04 ±5.61 1.457 ±0.30 11.26 ±6.32 4.801 ±0.39 72.28 ±9.12 74.65 ±8.35 21.80 ±2.93 28.77 ±4.59 1.64 ±3.18 6.43 ±4.78 7.22 ±2.40 1.53 ±1.00 6.85 ±4.53 0.75 ±0.19 2.73 ±0.73 1.47 ±0.30 1.01 ±0.58 4.70 ±0.82 2.43 ±1.53 7.31 ±2.49 Group B 126

< 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 Real-time ultrasound elastography for nonalcoholicfatty liver diseaseinPCOSpatients 0.002 0.001 0.017 0.025 0.013 0.002 NS NS NS NS NS NS NS NS NS p p <

Sensitivity of MEAN disturbances betweenthegroupsformedonbasisofcutoffvalue Table 3. 3). of 0.562(Figure 0.762and0.8 respectively,were withtheYouden index value,thesensitivityandspecificity 91.7 asthecutoff aminotransferase. performance todiscriminatepatients withandwithoutelevated aminotransferase. forMEANvalueexhibitreasonableoverall ROCcurves Figure 3. Values arenumber(%). 0.0 0.2 0.4 0.6 0.8 1.0 Hyperlipidemia DM IGT IFG IR PCOS Hirsutism bHA PCO Oligomenorrhea N aminotransferase Elevated MS EBP Central obesity 0.0 Comparison offrequency of endocrine disorders and metabolic ROC curves for the diagnostic value of MEAN for elevated ROC curves 0.2 Diagonal segmentsareproducedbyties (MEAN <91.7) 11 (22.45%) 19 (38.77) 41 (83.67) 46 (93.88) 11 (22.45) 17 (34.69) 46 (93.88) 44 (89.80) 17 (34.69) 14 (28.57) 39 (79.59) 9 (18.37) Group A 2 (4.08) 3 (6.12) 49 1 -Speci city 0.4

ROC Curve (MEAN ≥91.7) 22 (17.46) 26 (20.63) 39 (30.95) 55 (43.65) 33 (26.19) 40 (31.75) 57 (45.24) 53 (42.06) 17 (13.49) 33 (26.19) Group B 4 (3.17) 1 (0.79) 2 (1.59) 6 (4.76) 0.6 126

39.515 36.469 32.960 32.536 16.864 27.689 41.547 8.935 0.088 0.070 4.485 0.262 0.140 5.504 0.8 X2 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.05 < 0.05 < 0.05 NS NS NS NS P 1.0 133

Copyright© AE&M all rights reserved. Copyright© AE&M all rights reserved. 134 histogram ( onthestrain In addition,target sizehasnoeffect to visualscoringforassessment oftarget strain. elasticity. to be superior is reported Thisapproach abouttexture information more of RTE mayoffer (28). variabilityremarkably interobserver reduce (16). These technologies accuracy of the measurements shows beneaththeRTE image,andthiscurve curve theoperator. from any pressure isalsoasinus There without isinducedbytheheartbeat liver parenchyma patients withobesity. Automaticdisplacement ofthe indeeperlocations,.g., allows formeasurements Owing tothelowerworkingfrequency, thenewprobe than theold devices did. system has better resolution ThenewRTE organs thanbefore. been usedinmore tohealthycontrols. as compared RTE for assessment of NAFLD in patients with PCOS mention thepopulationswithPCOS. notto of RTE reported, withNAFLDhasbeenrarely methods such as transient elastography (), the use tootherelastographic Compared stiffness. of texture daily practice.InRTE, indicator elasticstrainisarelative accuracyin diagnosis offocallesionswithimpressive ultrasonography. Nonetheless, it is mainly used for gynecologists. from attention receiving more (5,6), andthisfindingis are atahigherriskofNAFLD thancontrols with PCOSare thatpatients individuals.Studiesconfirmed normal among patients with PCOS than among frequent more are Thesemetabolicdisorders metabolic disorders. toobesity,NAFLD isrelated MS,IR,DM,andother DISCUSSION stand-alone independentfactors( alsointhelistof MEAN), PCOS,WC,andMSwere subjects withoutNAFLD( factors associatedwith the MEANvalueamong140 stand-aloneindependent only PCOS,WC,andMSwere AST, SBP, DBP, showedthat MS,andPCOS.Theresults AMH, CHOL,TG,LDL-C,HDL-C,apoA,apoB,ALT, AGE, BMI,WC,WHR,mFG,FIN,FBP, HOMA-IR, was adependentvariable.Independentvariablesincluded Real-time ultrasound elastography for nonalcoholicfatty liver diseaseinPCOSpatients Semiquantitative parameters of strain histograms rapiddevelopmentoftechnologies,RTEWith has So far, to our knowledge, this is the first study on RTE isanexcellentsupplementtoconventional When AREAwasadependentvariable(insteadof analyses,theMEANvalue In multiplelinearregression

29). There is intrinsic relevance among isintrinsic relevance 29). There p < 0.05). p < 0.05). with hepatic stiffness could beassessedbyRTE (30). with hepaticstiffness associated thatfibrosis reported major researchers Histologicallyspeaking, ofinterest. region and harder relatively uneven a values.HigherAREAreflects region tissue.AREAistheratiooflow-strain(stiffer) the stiffer ultrasoundechoes(MEAN)issmallerin of thereflected tissue; thus, the displacement easily than a stiffer more asfollows:asoftertissuecanbecompressed theoretically studies (12,16,17,29).Thisfindingmaybeexplained withother inagreement are theseresults controls; toPCOSwithoutNAFLDandhealthy as compared PCOS withNAFLD,whileAREAandSDincrease MEAN, AREA,andSD,especiallyMEAN. study, weanalyzedonly themainparametersincluding isnoconsensus(17,18).Inthis there and currently formulas, on theparameterscalculatedwithdifferent scoringsystemsbased authors developeddifferent all the 11 semiquantitative strain parameters. Several limitation of this study. Nonetheless,the PCOS thisnotion. confirm is needed to research ultrasonic examination. Further visual detection by conventional sensitively: before more distinguish alterationsinthe liverparenchyma NAFLD. ItseemsthatstrainparametersinRTE may MEAN orAREAvaluesamongthesubjectswithout is alsoastand-aloneindependentfactorassociatedwith analyses,besidesWCandMS,PCOS linear regression tomultiple According NAFLD andhealthycontrols. betweenthoseinPCOSpatientswith intermediate AREA inPCOSpatientswithoutNAFLDwere ThelevelsofMEANand tocontrols. as compared also hadsignificantlylowerMEANandhigherAREA in additiontoconventionalultrasonography. semiquantitative strainparametersofRTE forNAFLD thediagnosticvalueof corroborated These results good diagnostic value for elevated aminotransferase. thatMEANhasa analysisconfirmed ROC curve endocrine and metabolic disturbances. with greater WC, BMI,andlipidtransaminaselevelsaswell a liverelastographicparameter(29,31-33). that the steatosis grade is independently associated with alsofound consistently withsteatosis.Otherresearchers and cols.showedthattissuemeanelasticityvaries byOrlacchio Research liverstiffness. could alsoaffect believethatsteatosis someresearchers Furthermore, Our research confirmed that MEAN decreases in that MEAN decreases confirmed Our research The lack of histological data from livers is a The lack of histological data from We foundthatPCOSpatientswithoutNAFLD In ourstudy, lowMEANwasassociatedwithhigher Arch Metab. Endocrinol 2019;63/2 Arch Metab. Endocrinol 2019;63/2 7. 6. 5. 4. 3. 2. 1. REFERENCES was reported. relevant tothisarticle nopotentialconflictofinterest Disclosure: (grant#2018YJ034). Province 00512), andtheScienceTechnology ofGuangdong Project Technology ofGuangzhoucity(grant#2014Y2- project research Yat-sen University(grant#2014005),theScienceand atSun Fundforclinicalmedicineresearch the 5010Programs Natural ScienceYouth FundofChina(grant#81703784), Science FoundationofChina(grant#81402168),theNational bythe National Natural in part Funding: the study wassupported Jinglinag Ruan,andJiayiWu. of HuiZhi,Yabo Yang, LingLi,Wenchang , XinhuaMa, Acknowledgment: wegratefullyacknowledgethecontributions populations includingpatientswithPCOS. liverdiseasesinvarious of NAFLDandotherdiffuse needed onsemiquantitativeRTE analysisfordiagnosis sensitively.patients withPCOSmore studiesare More alterationsin liverparenchyma may beabletoreveal method Theproposed without NAFLDandcontrols. is useful for distinguishing PCOS patients with or information. more mayprovide with PCOSandhealthycontrols during assessmentofliverchangesbetweenpatients biopsy. comparisonofMRIandRTE Nevertheless, forthemtoundergo aninvasiveliver inappropriate itwas forahealthcheck-up.Therefore, the controls and problems, ormenstrual fertility mostly regarding agecametoourdepartment patients ofreproductive

In conclusion,semiquantitativeRTE analysis fatty disease and polycystic ovary syndrome are associated liver Targher G,Rossini M,Lonardo A. Evidencethatnon-alcoholic metabolic syndrome. Arq Gastroenterol. 2015;52(2):117-23. women withpolycysticovary syndromeanditscorrelation with Urbanetz AA. Prevalence ofnon-alcoholicfatty diseasein liver Romanowski MD,Parolin MB,Freitas AC, PiazzaMJ, BassoJ, Hepat Mon.2014;14(11):e23235. at ahighriskofnon-alcoholicFatty disease;ameta-analysis. liver MS, Alavian SM. Are womenwithpolycysticovarian syndrome Ramezani-Binabaj M,MotalebiKarimi-Sari H,Rezaee-Zavareh Consensus Workshop Group.Fertil Steril.2012;97(1):28-38.e25. drome (PCOS):the Amsterdam ESHRE/ASRM-Sponsored 3rdPCOS al. Consensusonwomen’s healthaspectsofpolycysticovary syn Fauser BC, Tarlatzis BC,Rebar RW, Legro RS,Balen AH, Lobo R,et cystic ovary syndrome.NatRev Endocrinol.2014;10(10):624-36. Jayasena CN,Franks S. The managementofpatientswithpoly Int.2015;35(2):542-9.Hong Kong healthcensus.Liver liver of non-alcoholicfatty diseaseintheChinese –resultsfromthe liver Fung J, Lee CK,ChanM,Seto WK, LaiCL, Yuen MF. Highprevalence drome (PCOS).JClinDiagnRes. 2014;8(2):31-3. berg U, etal.Statusofhomocysteineinpolycysticovary syn Mathiesen UL,Franzen LE, Aselius H,Resjo M,Jacobsson L,Fo Real-time ultrasound elastography for nonalcoholicfatty liver diseaseinPCOSpatients - - - -

13. 12. 11. 10. 9. 8. 24. 23. 22. 21. 20. 19. 18. 7. 1 16. 15. 14.

elastography, andaspartate-to-platelet ratioindexintheassess et al. Performance of real-time strain elastography, transient Ferraioli G, C,MalfitanoTinelli A,DalBelloB,FiliceG,C, J Gastroenterol.2011;46(3):350-8. stiffnessassessment ofliver in patientswithchronic hepatitisC. et al.Real-time tissueelastography asatoolforthenoninvasive Morikawa H,Fukuda K,Kobayashi S,Fujii H,Iwai S,EnomotoM, troenterol. 2012;47(4):461-9. fibrosis. JGas impulse (ARFI) imaging in the diagnosis ofliver real-time tissueelastography (RTE), andacousticradiationforce M, etal.Head-to-headcomparisonoftransientelastography (TE), Colombo S,BuonocoreM,DelPA, Jamoletti C,EliaS, Mattiello 2014;2014:210407. sis inchronic viralhepatitisBandC.GastroenterolRes Pract. Diagnostic usefulnessofreal-timeelastography fibro forliver Kim YW, Kwon JH,Jang JW, KimMJ, OhBS,ChungKW, etal. Hepatol. 2015;7(4):638-48. ods forthediagnosisofnonalcoholicfatty disease. liver World J Papagianni M,Sofogianni A, Tziomalos K. Non-invasivemeth 2015;41(5):1161-79. of ultrasoundelastography: Part 3:liver. UltrasoundMedBio. al. WFUMB guidelinesandrecommendationsforclinicaluse Ferraioli G,FiliceC,CasteraL,ChoiBI,Sporea Wilson SR,et docrine by necessityratherthanchance: anovel hepato-ovarian axis? 2006;21(8):2027-32. in Chinese women with polycystic ovary syndrome. Hum Reprod. Chen X, Yang D, Li L, Feng S, Wang L. Abnormal glucose tolerance 1991;24(3):495-500. Dermatol. J AmAcad et al.Report oftheConsensusConferenceon Acne Classification. Pochi , Shalita AR, Strauss JS, Webster SB, Cunliffe WJ, Katz HI, 2011;96(3):792-6. sutism inChinesewomen:across-sectionalstudy. Fertil Steril. Zhao X,NiR,LiL,Mo Y, HuangJ, HuangM,etal. 2011;154(2):187-91. age.EurJObstet Gynecol Reproden ofreproductive Biol. normal cut-off levelsforhyperandrogenemia inChinesewom Zhao X,HeZ,Mo Y, Chen X, Y, Yang D.Determiningthe Dis.2002;34(7):516-22. transaminases. DigLiver ymptomatic patientswithmild/moderateabnormalitiesofliver amination reflectsdegreeofsteatosisbutnotfibrosisinas Foberg U, echogenicity etal.Increasedliver atultrasoundex Mathiesen UL,Franzen LE, Aselius H,Resjo M,Jacobsson L, Reprod. 2004;19(1):41-7. health risksrelatedtopolycysticovary syndrome(PCOS).Hum consensusondiagnosticcriteriaandlong-term Revised 2003 brosis inchronic hepatitisB.EurJRadiol. 2012;81(1):31-6. technologywith anovel quantitative fi forassessmentofliver Wang J, GuoL,ShiX,Pan W, Bai Y, Ai H.Real-time elastography fibrosisandcirrhosis.of liver Exp TherMed.2015;10(3):983-90. inevaluatingthestage parameters time elastography quantitative L, Shi B, Song , Li Y, Wang S, Wang X. Clinical valueof real- diseases.DiagnIntervRadio.liver 2014;20(1):90-9. comparison withtransientelastography inpediatricpatientswith ment ofreal-timetissueelastography inaphantommodeland Schenk JP, Alzen G,Klingmuller V, Teufel U, ElSS,etal.Measure chronic viralhepatitis.Oncology. 2013;84 Suppl1:13-20. fibrosiswithreal-timetissueelastographyAssessment ofliver in Yada N,Kudo M,Morikawa H,Fujimoto K,Kato M,Kawada N. hepatitis C.Oncology. 2013;84 Suppl1:3-12. noninvasive evaluationofhepaticfibrosisinpatientswith chronic Novel imageanalysismethodusingultrasoundelastography for Fujimoto K,Kato M,Kudo M, Yada N, Shiina T, UeshimaK,etal. 2012;199(1):19-25. ment offibrosisin chronic hepatitisC. AJR AmJ Roentgenol. . 2016;51(2):211-21. Defining hir 135 ------

Copyright© AE&M all rights reserved. Copyright© AE&M all rights reserved. 136 29. 28. 27. 26. 25. Real-time ultrasound elastography for nonalcoholicfatty liver diseaseinPCOSpatients elastography (RTE). UltrasoundMedBiol.2012;38(4):537-44. elasticityinNASHet al.Liver patientsevaluatedwithreal-time Orlacchio A, Bolacchi F, Antonicoli M,CocoI,Costanzo E, Tosti D, elastography –aphantomstudy. PLoS One.2014;9(2):e88699. of visualscoringandsemi-quantificationultrasoundstrain Carlsen JF, Ewertsen C, Saftoiu A, Lonn L, Nielsen MB. Accuracy Artery Disease?IntJPrev Med.2012;3(8):552-8. Which Oneis Associated withDiabetesMellitusandCoronary national DiabeticFederation DefinitionofMetabolic Syndrome, terol EducationProgram Adult Treatment Panel III Versus Inter Rezaianzadeh A, NamayandehSM,Sadr SM.NationalCholes www.who.int/diabetes/publications/diagnosis_diabetes2006/en. hyperglycaemia Report ofa WHO/IDF consultation. Available at:http:// WHO. Definition and diagnosis of diabetes mellitus andintermediate PCOS patientsinSouthern China.HumReprod. 2010;25(5):1295-302. Chen X, Yang D,LiL,Feng S, Wang L. Appropriate BMIlevelsfor - - 33. 32. 31. 30. 2014;20(16):4702-11. hepatic steatosisinChinesepatients. World JGastroenterol. trolled attenuation parameterfornon-invasiveassessmentof Shen F, ZhengRD,Mi YQ, Wang XY, Pan Q, Chen GY, etal.Con Korean Int.2014;34(1):102-9.of anative population.Liver atosis inpatientswithchronic study diseases:aprospective liver trolled attenuation parameter(CAP)fordetectionofhepaticste Chon Y E,Jung KS,KimSU, Park JY, Park YN, KimDY, etal.Con enterol. 2014;49(5):611-6. biopsy. studywithliver rameter: acomparative Scand JGastro tection ofhepaticsteatosisusingthecontrolledattenuation pa Yilmaz Y, Yesil A, GerinF, ErgelenR, CA,etal. Akin H,Celikel atology. 2012;56(4): 1271-1278. and portal hypertension innonalcoholicfatty diseases.Hep liver Real-time tissue elastography for evaluation of hepatic fibrosis Ochi M, Koizumi H,Hirooka Y, Miyake T, Tokumoto Y, Soga Y, et al. Arch Metab. Endocrinol 2019;63/2 ------