Science & Technology

Ductus venosus blood flow velocity waveform in diabetic pregnancies

Ductus venosus blood flow velocity waveform in diabetic pregnancies
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  Ultrasound Obstet Gynecol   2010;  36 : 344–349Published online 5 August 2010 in Wiley Online Library (  DOI:  10.1002/uog.7573 Ductus venosus blood flow velocity waveform in diabeticpregnancies A. STUART*†, I. AMER-WÅHLIN‡, S. GUDMUNDSSON§, K. MAR ˇ S ´AL†, A. THURING†and K. K ¨ALLEN†¶ * Department of Obstetrics and Gynecology, Central Hospital, Helsingborg,  ‡ Department of Women’s and Children’s Health, KarolinskaInstitute, Stockholm, Departments of Obstetrics and Gynecology,  † Clinical Sciences Lund and   § Clinical Sciences Malm¨ o, Lund Universityand   ¶ Reproductive Epidemiology Center, Lund University, Lund, Sweden KEYWORDS : diabetes; ductus venosus; large-for-gestational age; pulsatility index ABSTRACT Objective  Maternal diabetes during pregnancy is asso-ciated with congenital cardiac malformations and hyper-trophiccardiomyopathy.Bloodflowintheductusvenosus(DV) has been postulated to reflect cardiac function. Theaim of our study was to investigate if diabetic pregnanciesexhibit abnormal DV hemodynamics, hence indicating changes in fetal cardiac function.  Methods  The pulsatility index of the DV (DV-PI) wasanalyzed retrospectively in 142 diabetic patients and compared to previously published DV-PI reference valuesfrom a non-diabetic low-risk population. DV values werethen correlated with maternal glycosylated hemoglobin(HbA1c).  Results  DV-PI was significantly higher in pregnanciescomplicated by either pre-existing insulin-dependent (DM) or gestational diabetes when compared withnormalreferencevalues.IncreasedDV-PIvalueswerestill evident in both diabetic groups when neonates that weresmall-for-gestational age and neonates with pathological umbilical blood flow pattern were excluded from theanalysis. In DM pregnancies a statistically significant correlation was found between DV-PI and maternal HbA1c. Conclusion  Diabetic pregnancies exhibit increased DV-PI values when compared to a normal low-risk pregnant  population, possibly indicating a fetal cardiac effect.Copyright   ©  2010 ISUOG. Published by John Wiley& Sons, Ltd. INTRODUCTION Maternal diabetes during pregnancy is often associatedwith increased neonatal morbidity and mortality 1 . Theincidence of congenital cardiac malformations in theoffspring of diabetic mothers (3–6%) is considerablyhigher than in the normal population 2–4 . Hypertrophiccardiomyopathy affects up to 50% of newborn infantsof Type 1 diabetic mothers and is characterized bythickening of the interventricular septum, producing leftventricular outflow obstruction 5 , 6 . The pathogenesis of diabetic cardiomyopathy is multifactorial and is not fullyunderstood; several mechanisms have been suggested,such as hyperinsulinemia, hyperglycemia 7 , 8 , increasedinsulin-like growth factor levels 9 , and an increasedexpressionandaffinity of insulinreceptors,whichleads tothe proliferation and hypertrophy of cardiac myocytes 10 .It has also been postulated that septal hypertrophydevelops secondarily as the response of the cardiac muscleto an increased resistance in the placental bed due tovillous thrombosis 11 . However, myocardial thickeninghas even been observed in fetuses of mothers with well-controlled diabetic pregnancies 12 .Analysis of fetal blood flow velocity waveform isa well-established tool in the surveillance of high-risk pregnancies complicated by intrauterine growthrestriction (IUGR) and/or hypertension 13 . Blood flow inthe ductus venosus (DV) has been postulated to reflectcardiac function 14 . Increased values of the pulsatilityindex (PI) for veins (DV-PI) and absent or reversed flowin the DV during atrial contraction have been correlatedwith adverse pregnancy outcomes 15 . Furthermore, anassociation between abnormal DV flow patterns andcongenital cardiac defects has been shown in the firsttrimesterof high-riskpregnancies withfetalchromosomalabnormalities 16 , 17 .The literature on DV and heart compromise indiabetic pregnancies is sparse. Zielinsky  et al  . 18 studied56 fetuses of diabetic mothers, of which 20 had Correspondence to:  Dr A. Stuart, Department of Obstetrics and Gynecology, Central Hospital, Helsingborg, Sweden(e-mail: Accepted: 27 November 2009 Copyright © 2010 ISUOG. Published by John Wiley & Sons, Ltd. ORIGINAL PAPER  Ductus venosus in diabetic pregnancy  345 myocardial hypertrophy, and 53 healthy fetuses of non-diabetic mothers. The fetuses with septal hypertrophyhad significantly higher DV-PI than those without septalhypertrophy and control fetuses of non-diabetic mothers.Theaimofourstudywastoinvestigatewhetherdiabeticpregnancies exhibit abnormal DV hemodynamics, henceindicatingmodifiedfetalcardiacfunction,whencomparedto a reference group consisting of low-risk non-diabeticpregnancies. PATIENTS AND METHODS Fetal blood flow parameters of diabetic pregnancies werecollected retrospectively from the clinical databases of two university departments of obstetrics and gynecologyin Lund and Malm ¨o, for the 10-year period 1998–2007.Clinical data concerning maternal diabetes were obtainedfrom the regional perinatal register for health care qualityassessment (Perinatal Revision South (PRS)) 19 . Maternaldiabetes was divided into two types based on the Interna-tional Classification of Diseases (ICD) codes – womenwith pre-existing insulin-dependent diabetes (ICD-10code 0.24.0) (Group DM;  n = 86) and women with dia-betes mellitus arising in pregnancy (ICD-10 code 0.24.4;gestational diabetes) (Group GD;  n = 56). Informationwas also extracted from the PRS regarding neonatal out-come:Apgarscores at 1and5 min,umbilicalcordarterialblood pH and base excess, transfer to neonatal intensivecare unit (NICU), birth weight and perinatal death.Information on maternal glycosylated hemoglobin(HbA1c) was collected from the patients’ records in DM-pregnancies. Mean HbA1c values were calculated forthe first trimester (up to 11 + 6 weeks’ gestation), earlysecond trimester (12 + 0 to 23 + 6 weeks), late secondand third trimester (24 + 0 weeks to delivery) and for thelast 30 days before the last DV measurement. Ultrasonographic measurements Doppler velocimetry was performed with one of thefollowing ultrasound systems: Aspen Acuson System(Acuson, Mountain View, CA, USA), Acuson Sequoia512 (Acuson), Philips HDI 5000 System (Philips Med-ical Systems, Bothell, WA, USA) and Voluson Expert730 Ultrasound System (GE Kretztechnik, Zipf, Austria).The examinations were done transabdominally by experi-enced sonographers using transducers with a frequency of 2–7 MHz and with the lowest setting of high-pass filter(always  < 125 Hz). The recordings of DV blood velocitywere done in the transverse section of the fetal body; theangle between the ultrasound beam and the direction of the DV blood flow was kept below 45 ◦ . Umbilical arteryblood velocities were recorded from the mid-portion of the umbilical cord. At least five uniform heart cycles wereused for calculating the PI in the Doppler shift spectrumrecordsobtainedfromthe DVandumbilicalartery duringperiods without fetal breathing movements and generalmovements.The umbilical artery PI was calculated automaticallyby the ultrasound systems according to the method of Gosling  et al  . 20 . In addition, the umbilical artery bloodflowpatternswereclassifiedaccordingtosemiquantitativeblood flow classes (BFC) 21 : •  Normal BFC: normal umbilical artery PI, positivediastolic flow; •  BFC I: positive diastolic flow, PI  >  mean + 2 SD and  < mean + 3 SD; •  BFC II: PI  >  mean + 3 SD; •  BFC IIIA: absent diastolic flow; and •  BFC IIIB: reversed diastolic flow.Data on DV-PI as analyzed automatically on-line bythe ultrasound machine were available in 128 cases. Hardcopies of Doppler shift spectra recorded from the DVwere available for the remaining 14 patients. The spectrawere scanned and the DV-PI was calculated electronicallyafter manual tracing of the maximum velocity curve, i.e.the envelope of the spectra (software ImageJ, NationalInstitute of Health, Bethesda, MD, USA). The DV-PI wascalculated according to the formula:DV-PI = ( S − A )/ TAMX , where S  =  maximum systolic velocity, A  =  maximumvelocity during atrial contraction and TAMXV  =  timeaveraged maximum velocity of the entire cardiac cycle 22 .According to clinical routines, pregnant women withdiabetes mellitus are offered sonographic evaluation of fetal growth and Doppler velocimetry at 28, 32 and36 weeks’ gestation, or more often, in cases with abnor-mal umbilical artery velocity index. For the clinicalmanagement of pregnancies, serial Doppler velocime-try examinations are often performed; in our study weincluded only the blood flow data from the last exami-nation before delivery. During the time period covered inthe present study, no DV-PI values were released to theclinicians managing the pregnancy and the DV-PI valueswere not used as criteria for delivery. The Ethics Commit-tee, University of Lund, approved the Doppler ultrasoundstudies in pregnancy.The birth-weight deviation from expected gesta-tional age-related birth weight was estimated accordingto the Swedish standard intrauterine growth curve 23 .Infants weighing less than mean − 2 SD or more thanmean + 2 SD were considered small-for-gestational age(SGA) or large-for-gestational age (LGA), respectively. Inorder to investigate the impact of IUGR on the results,a subanalysis of DV-PI was performed in which SGAinfants and infants with abnormal umbilical blood flow(BFC classes I–IIIB) were excluded. Statistical analysis The longitudinal reference ranges for DV flow velocitiespublished by Kessler  et al  . 14 were used as reference data.The DV-PI values in our study were transformed by themethod explained in the appendix to the article of Kessler Copyright © 2010 ISUOG. Published by John Wiley & Sons, Ltd.  Ultrasound Obstet Gynecol   2010;  36 : 344–349.  346  Stuart et al  . 14 , and expressed as SD-scores using the referenceranges of transformed DV-PI values according to gesta-tional age. For the DM and GD groups, 95% confidenceintervals (CI) for the SD-scores were computed under theassumption of normal distributions.The non-parametric Wilcoxon test was used for com-parison of continuous variables and Fisher’s exact testwas used for dichotomous outcome parameters. ThePearson rho was computed in order to investigate thecorrelation between HbA1c levels and DV-PI. All sta-tistical analyses were performed using Gauss software(Gauss TM , Aptech Systems Inc., Maple Valley, WA, USA, RESULTS Twenty-two (25.6%) mothers in the DM group ( n = 86)had hypertension or pre-eclampsia vs. nine (16.1%) of those in the GD group ( n = 56) ( P = 0 . 82). Forty-eightpercent of women in both groups were primiparous.There was no substantial difference between the mater-nal mean age of the DM group (32 years) and the GDgroup (31 years) ( P = 0 . 5). In the DM and GD groups,31 (36.0%) and 34 (60.7%) women, respectively, had avaginal non-instrumental delivery ( P = 0 . 005).Neonatal and maternal characteristics are shown inTable 1. The mean birth weight did not differ signifi-cantly between the DM and the GD groups ( P = 0 . 12).When birth weight for gestational age was evaluated, astatistically significant difference was found ( P = 0 . 002).Among the offspring of the DM and GD groups, 3.5 and14.3%, respectively, were SGA ( P = 0 . 02). The corre-sponding percentages for the LGA were 30.2 and 16.1%in the DM and GD groups, respectively ( P = 0 . 82).Among the 86 fetuses of mothers with DM, two werestillborn, and one died within 24 h after birth, giving atotal perinatal mortality rate of 3.5%. All three fetuseshad DV-PI within normal limits; the DV recordings wereperformed 6 weeks, 1 week and 2 days prior to delivery,respectively. No deaths occurred in the GD group.The percentages of infants with low Apgar score orlow umbilical artery cord pH were similar in the DM andGD groups. Seven neonates (8.1%) had an Apgar score at5 minof   < 7in the DM group andthree (5.4%) inthe GDgroup ( P = 0 . 74). Arterial cord pH  < 7 . 05 was found inthree neonates of 65 examined (4.6%) in the DM groupand in two neonates of 45 examined (4.4%) in the GDgroup ( P  close to 1).One complex cardiac malformation (pulmonary valveatresia, ventricular septum defect and dextrocardia) wasdiagnosed postnatally in the GD group. The neonate wasborn at 37 + 3 weeks’ gestation, and had had a normalDV-PI (0.52) at 33 + 3 weeks. No other cardiac morbidi-ties were recorded. Ductus venosus velocimetry The mean interval between the last DV-PI measurementand delivery was 19.6 (median 9 (range, 1–109)) days. Table 1  Maternal characteristics, pregnancy complications andperinatal and obstetrical outcome according to type of diabetes(pre-existing diabetes (DM group) and gestational diabetes (GDgroup)) CharacteristicDM group(  n = 86  )GD group(  n = 56  )Maternal characteristics Primiparous 41 (47.7) 27 (48.2)Maternal age (years) 32 31 Pregnancy and delivery Hypertension of pregnancy 12 (14.0) 5 (8.9)Pre-eclampsia 10 (11.6) 4 (7.1)Induction of labor 12 (14.0) 1 (1.8)Mode of deliverySpontaneous vaginal 31 (36.0) 34 (60.7)Elective Cesarean section 14 (16.3) 7 (12.5)Emergency Cesarean section 36 (41.9) 13 (23.2)Instrumental vaginal 5 (5.8) 2 (3.6) Fetal characteristics Birth weight (g) 3472 * ± 982 3317 ± 839Birth weight according to GA ‡ Small for gestational age 3 (3.5) 8 (14.3)Appropriate for gestational age 56 (65.1) 39 (69.6)Large for gestational age 26 (30.2) 9 (16.1)Macrosomia (birth weight  > 4500  g) 9 (10.5) 4 (7.1)Female gender 46 (53.5) 27 (48.2)Umbilical cord arterypH  < 7 . 05†  3/65 (4.6) 2/45 (4.4)Base excess  < − 12  mmol/L 1/65 (1.5) 0 (0.0)1-min Apgar score < 4  4 (4.7) 2 (3.6)4–6 13 (15.1) 5 (8.9)7–10 69 (80.2) 49 (87.5)5-min Apgar score < 4  2 (2.3) 0 (0.0)4–6 5 (5.8) 3 (5.4)7–10 79 (91.9) 53 (94.6)GA at delivery < 28  weeks 2 (2.3) 1 (1.8)28–31 weeks 7 (8.1) 2 (3.6)32–36 weeks 18 (20.9) 6 (10.7)37–41 weeks 59 (68.6) 46 (82.1) ≥ 42  weeks 0 (0.0) 1 (1.8)Admission to NICU 41 (47.7) 21 (37.5)Ventilatory support 18 (20.9) 5 (8.9)Neonatal morbidityRespiratory distress syndrome 5 (5.8) 3 (5.4)Delayed closure of ductusarteriosus2 (2.3) 1 (1.8)Intracranial (non-traumatic) 1 (1.2) 0 (0.0)hemorrhageData are given as  n  (%), mean or mean ± SD.  * One value missing. † Twenty-one values missing in DM group and 11 values missing inGD group.  ‡ Mean and SD according to Swedish standard forintrauterine growth 23 . GA, gestational age; NICU, neonatalintensive care unit. Sixty-two (43.7%) and 84 (59.2%) patients weredelivered within 1 and 2 weeks from the last DV-PImeasurement, respectively.Figure 1 shows a scatter diagram of DV-PI values plot-ted against the gestational age and reference values. Inthe DM and GD groups, nine and four individual values,respectively, were above the 95th centile. The mean and Copyright © 2010 ISUOG. Published by John Wiley & Sons, Ltd.  Ultrasound Obstet Gynecol   2010;  36 : 344–349.  Ductus venosus in diabetic pregnancy  347 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41Gestational age at examination (weeks)    D  u  c  t  u  s  v  e  n  o  s  u  s  p  u   l  s  a  t   i   l   i  t  y   i  n   d  e  x Figure 1  Ductus venosus pulsatility index in relation to gestationalage at examination and according to type of diabetes: pre-existingdiabetes ( ) or gestational diabetes ( × ). Reference values showingthe  50 th ( ),  25 th and  75 th ( ) and  5 th and  95 th ( )percentiles obtained from Kessler  et al  . 14 are shown. Table 2  Descriptive statistics for ductus venosus pulsatility index(DV-PI) values and for DV-PI SD-scores for the group withpre-existing insulin-dependent diabetes (DM group) and the groupwith gestational diabetes (GD group) Parameter DM group (  n = 86  ) GD group (  n = 56  ) DV-PIMean (95% CI) 0.56 (0.52 to 0.59) 0.55 (0.51 to 0.60)Median (range) 0.55 (0.25 to 1.15) 0.53 (0.23 to 1.07)DV-PI SD-score * Mean (95% CI) 0.37 (0.14 to 0.59) 0.38 (0.11 to 0.65)Median (range) 0.24 ( − 1.50 to 4.08) 0.27 ( − 1.8 to 3.8) * SD-scores of transformed values according to gestational age(after Kessler  et al  . 14 ). median DV-PI values are displayed in Table 2. No signif-icant difference in absolute DV-PI was detected betweenthe DM and GD groups ( P = 0 . 86).The mean, median and range of the transformed SD-scores are also shown in Table 2. Fetuses in both GD( P = 0 . 006)andDMpregnancies( P = 0 . 0015)hadsignif-icantly higher DV-PI SD-scores than expected accordingto the reference values used. There was no significantdifference in the transformed DV-PI values between DMand GM pregnancies ( P = 0 . 88).Fourteen fetuses in the DM group had abnormal UABFC, of which three were SGA, and 10 fetuses in theGD group had abnormal UA BFC, of which eight wereSGA. After excluding all SGA infants and all fetuseswith abnormal BFC, the mean DV-PI SD-scores werestill statistically significantly increased (above zero SD)in both the DM ( P = 0 . 02) and GD group ( P = 0 . 035).The mean DV-PI SD-score in the DM group was 0.27(95% CI, 0.04–0.50) ( n = 72) and the correspondingvalue in the GD group was 0.34 (95% CI, 0.02–0.65)( n = 46). 43210 − 1 − 2    D  u  c  t  u  s  v  e  n  o  s  u  s  p  u   l  s  a  t   i   l   i  t  y   i  n   d  e  x   S   D -  s  c  o  r  e 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0Mean glycosylated hemoglobin ( % ) Figure 2  Ductus venosus pulsatility index SD-score in relation tomean glycosylated hemoglobin levels in pregnancies in women withpre-existing diabetes in the month before the last ductus venosusmeasurement before delivery. Table 3  Maternal glycosylated hemoglobin (HbA1c) valuesaccording to ductus venosus pulsatility index (DV-PI) SD-scores inpregnancies with pre-existing diabetes mellitus (DM group,  n = 86 ) DV-PI SD-scoreHbA1c  ≤ 0  > 0  P Third trimesterMean (95% CI) 4.7 (4.5–4.9) 5.2 (5.0–5.5) 0.003Median (range) 4.7 (4.1–6.0) 5.1 (3.6–8.1)Month before last DV measurementMean (95% CI) 4.7 (4.5–4.9) 5.4 (5.0–5.7 ) 0.003Median (range) 4.7 (4.0–6.0) 5.2 (3.9–7.6)Data are given as %. AstatisticallysignificantcorrelationwasfoundbetweenDV-PI SD-scores and mean HbA1c values in DM-pregnancies during the last month before the last DVmeasurement (Pearson rho = 0.31 (95% CI, 0.04–0.54); P = 0 . 02) (Figure 2). No statistically significant correla-tion was found between the DV-PI SD-scores from thelast measurement before delivery and the mean HbA1c inthe first ( P = 0 . 71) and the second ( P = 0 . 31) trimesters,respectively. The HbA1c values were also comparedbetween two subgroups – a group with DV-PI SD-scores ≤ 0 and a group with DV-PI SD-scores  > 0. A statisticallysignificant difference was found between the two groupsduring the third trimester and during the month beforethe last DV measurement. HbA1c values are displayed inTable 3. DISCUSSION The present study shows increased DV-PI values in preg-nancies complicated by pre-existing insulin-dependentdiabetes mellitus or gestational diabetes, when comparedwith reference values obtained from a healthy low-risk Copyright © 2010 ISUOG. Published by John Wiley & Sons, Ltd.  Ultrasound Obstet Gynecol   2010;  36 : 344–349.  348  Stuart et al. pregnant population. Furthermore, in DM-pregnancies acorrelation exists between maternal mean HbA1c levelsand increased DV-PI SD-scores.The DV plays a central role in the return of venousblood from the placenta, by shunting oxygenated bloodfrom the umbilical vein through the inferior atrial inletto the foramen ovale. Studies performed on sheep haveshown that 50% of the umbilical vein blood enters theDV and accounts for 98% of the blood flow throughit 24 . In the human fetus a higher proportion of umbilicalblood is directed to the liver and at 31 weeks’ gestationonly 18% is shunted through the DV 25 . As the DV bloodvelocity depends on the pressure gradient between theumbilical vein and the right atrium, pathological bloodflow velocities can provide information on fetal cardiaccompromise 26 .Baez  et al  . 27 found significantly higher DV-PI in fetuseswith congenital heart disease that died than in thosethat survived. Furthermore, increased pulmonary vein PIhas been shown in fetuses of diabetic mothers when com-paredtofetusesofnon-diabeticwomen,indicatingaltereddiastolic cardiac function 28 . Girsen  et al  . 29 showed thatfetuses with an abnormal DV blood velocity waveformpattern had increased levels of the heart failure markerNT-proBNP(amino-terminalfragmentofpro-brainnatri-uretic peptide) in the umbilical artery blood, indicatingfetal cardiac compromise. In a later study, the same groupalso demonstrated increased fetal cardiac natriuretic pep-tide in Type-1 diabetic pregnancies when compared toa control group, even in the presence of good glycemiccontrol and normal placental hemodynamics 30 . Fetuseswith myocardial cell damage, indicated by increasedplasma concentrations of cardiac troponin T, showedincreased pulsatility in the DV 31 . In a study on fetalelectrocardiography(ECG)duringlaborwehaveobservedST depression in the ECG waveform of 30–40% of thefetuses of diabetic mothers as compared to 10–20% of fetuses of healthy mothers 32 . These findings might indi-cate altered activity of the myocardium, as could also thepresent findings of increased DV-PI.Our study population was a high-risk one. Thosepatients with pre-existing insulin-dependent diabetesexhibited expected obstetric and fetal outcome. In accor-dance with other studies 33 , the offspring of patients withpre-existing insulin-dependent diabetes had a high mor-tality rate and a high proportion of operative deliveries,and 52% had birth weight above the mean + 1 SD.The group of women with GD constitutes a morediverse group. The gestational diabetics of our study hada high prevalence (16%) of associated conditions suchas essential hypertension and pre-eclampsia. In a Swedishstudy of almost 3500 women with GD, a 6.1% incidenceof pre-eclampsia was found 34 . In the GD group of thepresent study 14% of infants were SGA and 16% wereLGA.Our gestationaldiabetics seemedto exhibit a higherprevalence of pregnancy complications (suspected IUGR( n = 8) and pre-eclampsia/hypertension ( n = 9)), suggest-ing that umbilical artery and DV Doppler velocimetrywas indicated in GD pregnancies with other complicatingfactors.A drawback of our study is that we did not have a low-risk reference group. Instead, we used the longitudinalreferencerangespublishedbyKessler et al  . 14 thatincluded160 low-risk Norwegian pregnancies. The population inthat study had a mean maternal age of 29 years, meangestationalageatdeliveryof40 + 3 weeksandmeanbirthweight of 3550g; there were 9.4% Cesarean sections, and4.4% of the neonates were admitted to the NICU. TheNorwegianpopulationoflow-riskpregnanciesisregardedas equivalent to a corresponding population in Sweden 35 .In addition, the protocol for ultrasonographic measure-ment of DV velocities by Kessler  et al  . is the same as theone used in Sweden.As mentioned above, our diabetic pregnancies werea high-risk population with a number of other com-plicating conditions. One cannot therefore exclude thepossibility that a reason for the increased DV-PI val-ues might not be diabetes  per se , but that it might bea result of hemodynamic compromise caused by pre-eclampsia and/or IUGR, factors known to influence DVblood flow. However, when SGA neonates and neonateswithabnormalumbilicalarterybloodflowwereexcluded,significantly increased DV-PI values for gestational agewere still evident in both the DM and GD groups.In DM pregnancies a statistically significant correlationwas found between DV-PI SD-scores and mean HbA1cvalues during the last 30 days before the last DV measure-mentbeforedelivery.WhenDV-PISD-scoresweredividedinto two groups – DV-PI SD-scores  < 0 and DV-PI SD-scores  > 0 – HbA1c values were significantly increased inthe latter group during both the last trimester and duringthe month before the last DV measurement. A possibleexplanation for the significant correlation with HbA1conly during the later part of pregnancy could be thatdysregulated maternal glucose during the first trimestermainlyaffectstheprevalenceofcongenitalmalformations,but poor glycemic control later in the gestation can leadto fetal overgrowth with cardiac septal hypertrophy anda secondary influence on cardiac function.In conclusion, we found increased DV-PI values bothin fetuses of pregnancies complicated by pre-existinginsulin-dependent diabetes and in pregnancies with gesta-tional diabetes, when compared to a reference group of healthy low-risk Scandinavian pregnant women. Mater-nal glycemic control seems to be of importance, as wefound a correlation between increased HbA1c values andDV-PI SD-scores in DM pregnancies, regardless of thefact that most women had well regulated diabetes. Theseresults, similarly to other studies, imply that intrauterineexposure to a diabetic environment might have an impacton the fetal heart. ACKNOWLEDGMENTS The authors would like to thank Anders Fridh MD, PhD,Department of Endocrinology, University Hospital MAS,Malm ¨o, RN Margareta Larsson, midwife, Maternal Care Copyright © 2010 ISUOG. Published by John Wiley & Sons, Ltd.  Ultrasound Obstet Gynecol   2010;  36 : 344–349.
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!