FETAL RENAL ANOMALIES: diagnosis, management and outcome.

Transcription

1 FETAL RENAL ANOMALIES: diagnosis, management and outcome.

2 cover: Albert Damen sr. print Budde-Elinkwijk, Nieuwegein

3 FETAL RENAL ANOMALIES: diagnosis, management and outcome Afwijkingen aan nieren en urinewegen bij de foetus: diagnose, behandeling en uitkomst. (met een samenvatting in het Nederlands) Proefschrift ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de Rector Magnificus, Prof. dr. W.H. Gispen, ingevolge het besluit van het College voor Promoties in het openbaar te verdedigen op dinsdag 14 december 2004 des middags te uur door Henrica Antonia Maria Damen Elias geboren te Tegelen, 27 mei 1941

4 Promotor: Copromotores: Prof. dr. G.H.A. Visser Department of Perinatology and Gynaecology University Medical Centre Utrecht Dr. T.P.V.M. de Jong Department of Paediatric Urology University Medical Centre Utrecht Dr. P.H. Stoutenbeek Department of Perinatology and Gynaecology University Medical Centre Utrecht ISBN: CIP-DATA KONINKLIJKE BIBLIOTHEEK DEN HAAG Damen-Elias, Henrica Antonia Maria FETAL RENAL ANOMALIES: diagnosis, management and outcome Utrecht, Universiteit Utrecht, Faculteit Geneeskunde Thesis Universiteit Utrecht Financial support for the publication of the this thesis is gratefully acknowledged: Stichting Kindernierziekten Easote Pie Medical Toshiba Medical Systems Europe Nierstichting Nederland

5 What does wisdom benefits us, if we do not possess love.

6 Referents: Prof. dr. F. van Bel Department of Neonatology University Medical Centre Utrecht Prof. dr. J.L.H.R. Bosch Department of Urology University Medical Centre Utrecht Prof. dr. H.W. Bruinse Department of Perinatology and Gynaecology University Medical Centre Utrecht Prof. dr. J.M. Nijman Department of Urology University Medical Centre Groningen Prof. dr. J.W. Wladimiroff Department of Obstetrics and Gynaecology Erasmus Medical Centre Rotterdam Paranimfen: Drs. P.J. Damen Mevr. M.J. Korenromp

7 Table of contents 7 Chapter 1 Introduction and aims of the thesis Chapter 2 Intra- and interobserver variability of fetal kidney and adrenal gland measurements revised version resubmitted to Ultrasound in Obstetrics and Gynaecology Chapter 3 Growth and size charts of the fetal kidney and the renal pelvis revised version resubmitted to Ultrasound in Obstetrics and Gynaecology Chapter 4 Growth and size charts of the fetal adrenal gland revised version resubmitted to Ultrasound in Obstetrics and Gynaecology Chapter 5 Congenital renal tract anomalies: outcome and follow-up of 402 cases detected antenatally over a period of 15 years. in press: Ultrasound in Obstetrics and Gynaecology Chapter 6 Concomitant anomalies in 100 children with unilateral multicystic kidney in press: Ultrasound in Obstetrics and Gynaecology Chapter 7 Mild pyelectasis diagnosed by prenatal ultrasound is not a predictor of urinary tract morbidity in childhood in press: Ultrasound in Obstetrics and Gynaecology Chapter 8 Variability in dilatation of the fetal renal pelvis during a bladder filling cycle in press: Ultrasound in Obstetrics and Gynaecology Chapter 9 Summary, discussion and conclusion Nederlandse samenvatting Dankwoord Curriculum Vitae

8

9 Chapter 1 Introduction, aims and outline of the thesis

10 10 Chapter 1 introduction Interest in fetal development has grown enormously in the past decades. Till then non-invasive methods were used to asses fetal health such as the feeling of movements by the pregnant woman, measuring growth by palpation of the fundal height and auscultation of the fetal heart tones with a wooden stethoscope. X-ray examination was used in search for fetal abnormalities and to diagnose multiple pregnancies but the questionable safety of such investigation in pregnancy made this method unsuitable for routine examinations. All changed when it became possible to observe the fetus in utero by ultrasound. As early as 1958, Donald et al 1 showed the outline of the fetal skull. The technical and methodological development of this technique made it possible to investigate the fetus directly and the consequence of this quickly expanding and improving new technique has been an increasing amount of information concerning fetal morphological and physiological development during pregnancy. Further improvement of the equipment with high-resolution real-time ultrasound has given the possibility to more detailed information of the fetus and later on transvaginal sonography has made it possible to obtain detailed information of early embryonic development. The introduction of routine prenatal ultrasound scanning since the early 1980 s in several countries in Europe has increased our knowledge on morphology and functional development of the fetus. Moreover it became possible to diagnose anomalies in utero. In population studies minor or major structural anomalies are detected in 2 to 3 % of cases 2-5. Abnormalities of the urinary tract account for 15-20% of these anomalies with a detection rate of approximately 90% 2, 4. The fetal kidneys can first be visualised by transabdominal ultrasound at 9 weeks gestational age and can be seen in all cases from 12 weeks onwards. In early pregnancy the echogenity of the fetal kidneys is high but this decreases in the course of gestation when they become hypoechoic 6. At about 28 weeks the renal pyramids can be detected and also the borders of the kidneys can be seen more clearly with progressing gestational age, since fat tissue is developing around the kidneys from that moment on.

11 introduction, aims and outline 11 The fetal bladder can be visualised from the onset of urine production, which occurs at about 10 weeks gestation 7. At 11 weeks of gestation the bladder can be visualised, both transvaginally and transabdominally, in 80% of fetuses and at 13 weeks almost in all fetuses 8, 9. It should always be possible to visualize the bladder when the crown-rump length is more than 67 mm (13 + weeks). The fetal adrenal glands are visible by ultrasonography as early as 9 weeks of gestation and in all cases from 12 weeks onwards. In the second trimester they appear as a disc-like structure in a transverse plane, cranially and medially to the kidney. Three layers can clearly be distinguished of which the outer layer is hypoechogenic and the central medulla hyperechoic. In the sagittal view they appear as heart-shaped figures of low echogenicity. During the last three decades numerous papers have dealed with the fetal renal system and its anomalies. Follow-up studies on (long term) outcome are, however, scarce and this hampers adequate counselling of parents and giving appropriate treatment advice. Moreover, up to date charts on normal fetal kidney size and growth are scarce. We therefore formulated the following aims of this thesis. aims of the thesis 1 To develop charts of size and growth of the fetal kidney, renal pelvis and adrenal gland. 2 To study long-term follow-up of a large cohort of infants with an antenatally diagnosed renal tract anomaly. 3 To answer the question whether mild pyelectasis (anteroposterior diameter of the fetal renal pelvis of 5 10 mm) as diagnosed around 18 to 20 weeks of gestation results in increased morbidity in childhood and therefore requires postnatal treatment. 4 To study the relationship between the size of the renal pelvis and the fetal bladder-filling cycle, to answer the question if fixed cut-off values regarding renal pelvis dilatation can be used or whether bladder filling has to be taken into account.

12 12 Chapter 1 outline of the thesis In Chapters 2 to 4 a prospective longitudinal study is described on size and growth of the fetal kidney, the fetal renal pelvis and the fetal adrenal gland in 111 fetuses from 16 weeks gestational age onwards till term. These studies were preceded by a study on intra- and inter-observer variation. In Chapter 5 we describe the findings and outcome of a large cohort of 402 fetuses in which urogenital anomalies were detected antenatally. We could use the database of the ultrasound unit of the department of obstetrics of the University Medical Centre, Utrecht, The Netherlands, which was established in At follow-up the youngest infant was 3 years and the oldest was 17 years (median 7 years 11 months). In Chapter 6 we present the outcome of 100 fetuses with an antenatally detected unilateral multicystic kidney. All additional urogenital and other anomalies are described. According to the advice of the Dutch Society of Paediatric Urology the non-functioning cystic kidney is removed at approximately 6 months of age to prevent life-time follow-up of these children because of an increased risk of hypertension 10, infection 11 or malignancy 12, 13. When the parents decided to do so, each child underwent a cystoscopy and girls also a colposopy prior to the operation. All these findings are included in the follow-up of these children (median 5 years and 4 months). For Chapter 7 we could use the data of the ultrasound department of the Amphia Hospital, Oosterhout, The Netherlands, which was established since Two hundred and eight children, who had had a mild pyelectasis of 5 10 mm at 18 to 20 weeks gestation, were compared with 416 matched controls regarding voiding and defecation patrons and urinary tract infections. A validated questionnaire, as used in the International Reflux Study in children, was used for this purpose. In Chapter 8 we describe a study in which 18 third trimester pregnant women are examined by ultrasound during several fetal bladder-filling cycles, to investigate if there is a correlation between the size of the renal pelvis and the extent of bladder filling. In Chapter 9 a summary and general discussion is described.

13 introduction, aims and outline 13 references 1 Donald I, Macvicar J, Brown TG. Investigation of abdominal masses by pulsed ultrasound. Lancet 1958;1(7032): Grandjean H, Larroque D, Levi S. Sensitivity of routine ultrasound screening of pregnancies in the Eurofetus database. The Eurofetus Team. Ann N Y Acad Sci 1998;847: Levi S. Ultrasound in prenatal diagnosis: polemics around routine ultrasound screening for second trimester fetal malformations. Prenat Diagn 2002;22(4): Levi S. Mass screening for fetal malformations: the Eurofetus study. Ultrasound Obstet Gynecol 2003;22(6): Stoll C, Clementi M. Prenatal diagnosis of dysmorphic syndromes by routine fetal ultrasound examination across Europe. Ultrasound Obstet Gynecol 2003;21(6): Green JJ, Hobbins JC. Abdominal ultrasound examination of the first-trimester fetus. Am J Obstet Gynecol 1988;159(1): McHugo J, Whittle M. Enlarged fetal bladders: aetiology, management and outcome.prenat Diagn 2001;21(11): Rosati P, Guariglia L. Transvaginal sonographic assessment of the fetal urinary tract in early pregnancy. Ultrasound Obstet Gynecol 1996;7(2): Braithwaite JM, Armstrong MA, Economides DL. Assessment of fetal anatomy at 12 to 13 weeks of gestation by transabdominal and transvaginal sonography. Br J Obstet Gynaecol 1996;103(1): Webb NJ, Lewis MA, Bruce J, Gough DC, Ladusans EJ, Thomson AP, et al. Unilateral multicystic dysplastic kidney: the case for nephrectomy. Arch Dis Child 1997;76(1): Wacksman J, Phipps L. Report of the Multicystic Kidney Registry: preliminary findings. J Urol 1993;150(6): Elder JS, Hladky D, Selzman AA. Outpatient nephrectomy for nonfunctioning kidneys. J Urol 1995;154(2 Pt 2):712-4; discussion LaSalle MD, Stock JA, Hanna MK. Insurability of children with congenital urological anomalies. J Urol 1997;158(3 Pt 2):

14

15 Chapter 2 Intra- and interobserver variability of fetal kidney and adrenal gland measurements H.A.M. Damen - Elias a, G.H.A. Visser a, P. Westers b, L. Pistorius a a Department of Perinatology and Gynaecology, University Hospital Utrecht, The Netherlands b Centre for Biostatistics, Utrecht University, The Netherlands

16 16 Chapter 2 ABSTRACT Objectives: To assess the intra- and interobserver variability of fetal kidney and fetal adrenal gland measurement by ultrasound. Methods: Data were obtained prospectively by experienced ultrasonographers. Thirty fetuses were measured for the intraobserver analysis and 20 fetuses for the interobserver analysis. Length, anteroposterior and transverse diameter from both the right and left kidney and from the length of the right and left adrenal gland were measured 3 times. Statistical analysis was performed by SPSS. Results: The reproducibility of the measurements of the kidney and of the adrenal gland were good with an intraclass correlation above 0,80 for all measurements for both the intraobserver and interobserver analysis. Conclusions: The high degree of intra- and interobserver reproducibility indicates that the three dimensions of the kidney and adrenal gland length are technically feasible to measure.

17 Intra- and interobserver variability 17 INTRODUCTION Many reference curves of the fetal renal kidney have been published 1-7. Knowledge of the normal range of the measurements of the fetus is essential when during screening an anomaly is identified. No intraobserver and interobserver variation analysis was done prior to any of these studies. Only Bertagnoli 2 in 1983 reported on the differences in measurements between 3 operators who each used a different ultrasound machine. Some studies describe the design, the patient selection and the methods of analysis 8-10 but do not involve differences in intraobserver and interobserver measurements. The discriminatory ability of a diagnostic test is in large extent depending on the repeatability of the test. The aim of this study was to assess the intra- and interobserver reproducibility of the length, the anteroposterior and the transverse measurements of the fetal kidney and of the length of the adrenal gland by ultrasound. MATERIALS AND METHODS Thirty pregnant women were asked to participate in the study for the intraobserver variation analysis and another twenty for the interobserver variation analysis. All women consented to participate in the study There were no exclusion criteria. Two experienced ultrasonographers, (HDE, LP) examined the fetuses. Each observer attempted to obtain three measurements with different time intervals between each measurement. The sonographers were not allowed to see their own measurements or to watch each other performing the measurements to avoid any possible influence. In both investigations the length, the anteroposterior and transverse diameter of the kidney and the length of the adrenal gland were measured transabdominally with the multifrequency transducer PVM 375 AT of the Toshiba Power Vision 6000, type SSA 370 A (manufacturer Toshiba, Tokyo, Japan). In a sagittal plane when the full length of the kidney with the renal pelvis was visualised the length of the kidney was measured and in the same sectional plane the length of the combination of the kidney and the adrenal gland. Subsequently the length of the adrenal gland was determined by subtraction of the kidney length from the combined length. Perpendicular to this plane in the largest sectional plane, the anteroposterior and transverse diameter of both kidneys were measured. The data

18 18 Chapter 2 were recorded on a photograph and stored in a database. Statistical analysis was performed by using SPSS, version 10.1 (Statistical Product and Service Solutions, Chicago). The mean and standard deviation were calculated to determine if there was a good consistency between the measurements. In addition the range, the Cronbachs alpha (α) and the intra class correlation (ICC) were calculated. The range is the distance between the highest and lowest value. Cronbachs α is a statistical index for internal consistency between the measurements. The index ranges from 0 = bad to 1 = excellent. The ICC is the measure of concordance and is a statistic that describes the reproducibility of repeated measures in the same subject and indicates true variance as a fraction of the total variance. Landis and Koch 11 have indicated the meaning of the different values of the ICC and a value of has a good agreement and larger > 0.81 an excellent one. The value of ICC of 1 for repeated measurements indicates perfect reproducibility while a value of 0 is interpreted as no better or worse than that expected by chance. RESULTS intraobserver variation analysis The kidney length, kidney anteroposterior diameter, kidney transverse diameter and adrenal gland length could be measured 3 times in all 30 women. Intraobserver agreement is given in Table 1. Statistic analysis shows a high alpha above the 0,9 and also a high IC above 0,8 for all different measurements. Table 1 Intraobserver variation of length, anteroposterior and transverse diameter of the kidney and of length of the adrenal gland. variable mean ± std.dev alpha intra class correlation (95% CI) RiKiL 22,706 ± ( ) LeKiL 22,413 ± ( ) RiKiAP 15,217 ± ( ) LeKiAP 15,190 ± ( ) RiKiTr 15,172 ± ( ) LeKiTr 15,412 ± ( ) RiGL 5,316 ± ( ) LeGL 5,275 ± ( ) RiKiL = right kidney length, LeKiL = left kidney length, RiKiAP = right kidney anterior/ posterior diameter, LeKiAP = left kidney anterior/ posterior diameter, RiKiTr = right kidney transverse diameter, LeKiTr = left kidney transverse diameter, RiGL = right adrenal gland length, LeGL = left adrenal gland length.

19 Intra- and interobserver variability 19 interobserver variation analysis Both investigators could take all measurements 3 times in all 20 women. The interobserver analysis is given in Table 2. There was a high alpha above 0,9 for all measurements as well as a high IC above 0,8. There was a high level of agreement between the two observers. Table 2 - Interobserver variation of length, anteroposterior diameter and transverse diameter of the kidney and of length of the adrenal gland. variable mean ± st.dev range alpha intraclass correlation observer 1 observer 2. 95% CI) HDE* LP# HDE LP RiKiL ± ± ( ) LeKiL ± ± ( ) RiKiAP ± ± ( ) LeKiAP ± ± ( ) RiKiTr ± ± ( ) LeKiTR ± ± ( ) RiGL ± ± ( ) LeGL ± ± ( ) HDE* = observer 1, LP# = observer 2 RiKiL = right kidney length, LeKiL = left kidney length, RiKiAP = right kidney anterior/ posterior diameter, LeKiAP = left kidney anterior/ posterior diameter, RiKiTr = right kidney transverse diameter, LeKiTr = left kidney transverse diameter, RiGL = right adrenal gland length, LeGL = left adrenal gland length. DISCUSSION Measurements of the kidney are of importance when an anomaly is identified. Measurements of the adrenal gland may be of importance in high-risk pregnancies when intra uterine growth retardation is suspected 12-14, when mothers use glucocosteriods 15 for a prolonged period of time or in case of congenital adrenal hyperplasia 16, 17. A valuable screenings test should be both feasible and repeatable. An intra- and interobserver analysis should be evaluated and measurements should only be introduced in routine setting if a good sensitivity and specificity is demonstrated. It was technically feasible to take three measurements of the fetal kidney and the length of the adrenal gland. We used various indices and coefficients to assess intra- and interobserver variability and found a good agreement of both for the intraobserver analysis and for the interobserver analysis. Comparison of our results with those of others is not possible because there is no study in literature on intra- and interobserver analysis of fetal kidney and adrenal gland measurements.

20 20 Chapter 2 REFERENCES 1. Jeanty P, Dramaix-Wilmet M, Elkhazen N, Hubinont C, van Regemorter N. Measurements of fetal kidney growth on ultrasound. Radiology 1982;144(1): Bertagnoli L, Lalatta F, Gallicchio R, Fantuzzi M, Rusca M, Zorzoli A, et al. Quantitative characterization of the growth of the fetal kidney. J Clin Ultrasound 1983;11(7): Sagi J, Vagman I, David MP, Van Dongen LG, Goudie E, Butterworth A, et al. Fetal kidney size related to gestational age. Gynecol Obstet Invest 1987;23(1): Pruggmayer M, Terinde R. [Fetal kidney screening: growth curves and indices]. Geburtshilfe Frauenheilkd 1989;49(8): Cohen HL, Cooper J, Eisenberg P, Mandel FS, Gross BR, Goldman MA, et al. Normal length of fetal kidneys: sonographic study in 397 obstetric patients. AJR Am J Roentgenol 1991;157(3): Scott JE, Wright B, Wilson G, Pearson IA, Matthews JN, Rose PG. Measuring the fetal kidney with ultrasonography. Br J Urol 1995;76(6): Chitty LS, Altman DG. Charts of fetal size: kidney and renal pelvis measurements. Prenat Diagn 2003;23(11): Royston P, Wright EM. How to construct 'normal ranges' for fetal variables. Ultrasound Obstet Gynecol 1998;11(1): Altman DG, Chitty LS. Design and analysis of studies to derive charts of fetal size. Ultrasound Obstet Gynecol 1993;3(6): Altman DG, Chitty LS. Charts of fetal size: 1. Methodology. Br J Obstet Gynaecol 1994;101(1): Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33(1): Hata K, Hata T, Kitao M. Ultrasonographic identification and measurement of the human fetal adrenal gland in utero. Int J Gynaecol Obstet 1985;23(5): Hata K, Hata T, Kitao M. Ultrasonographic identification and measurement of the human fetal adrenal gland in utero: clinical application. Gynecol Obstet Invest 1988;25(1): Bronshtein M, Tzidony D, Dimant M, Hajos J, Jaeger M, Blumenfeld Z. Transvaginal ultrasonographic measurements of the fetal adrenal glands at 12 to 17 weeks of gestation. Am J Obstet Gynecol 1993;169(5):

21 Intra- and interobserver variability Esser T, Chaoui R. Enlarged adrenal glands as a prenatal marker of congenital adrenal hyperplasia: a report of two cases. Ultrasound Obstet Gynecol 2004;23(3): Saada J, Grebille AG, Aubry MC, Rafii A, Dumez Y, Benachi A. Sonography in prenatal diagnosis of congenital adrenal hyperplasia. Prenat Diagn 2004;24(8):

22

23 Chapter 3 Growth and size charts of the fetal kidney and the renal pelvis. H.A.M. Damen - Elias a, R.H. Stigter a,c, P. Westers b, G.H.A. Visser a a Department of Perinatology and Gynaecology, University Hospital Utrecht, The Netherlands b Centre for Biostatistics, Utrecht University, The Netherlands c Department of Obstetrics and Gynaecology, Deventer Hospital, Deventer, The Netherlands

24 24 Chapter 3 ABSTRACT Objectives The aim of this study was to develop a reference curve for the size and growth of the fetal kidney and of the fetal renal pelvis. Methods Prospective longitudinal study including 111 fetuses. Ultrasound examinations were made every four weeks beginning in one half of the cases at 16 weeks and in the other half at 18 weeks. Length, anteroposterior and transverse diameter from both the right and left kidney and the anteroposterior diameter and transverse diameter from the right and left renal pelvis were measured. Statistical analysis was performed by multilevel analysis. Results Fitted 5 th, 50 th and 95 th centile charts are presented with the raw data. Comparisons were made with previously published data. Conclusions The new charts for size and growth are of use for the three dimensions and for the perimeter of the fetal kidney and for the size of the renal pelvis.

25 Charts of fetal kidney and renal pelvis 25 INTRODUCTION Knowledge of the normal range of the dimensions of the fetal kidney and renal pelvis is essential. Various papers on the intrauterine growth of the renal size have been published but many failed to validate gestational age by measuring crown-rump length 1-5, do not cover the whole of the second and third trimester of pregnancy 1, 2, 6-8, did not measure the kidney in three dimensions 1, 2, 4-6 or used only pre-term born children or post-mortem specimens Several studies used a mixture of crosssectional and longitudinal data 1, 3-6, 8, 11. Cross-sectional data obtained by measurements of each fetus on a single occasion give information on size, whereas longitudinal data obtained by measurements of fetuses on a series of occasions may be used for a reference curve for size and growth The difficulty of any longitudinal study to construct nomograms is that there are some missing data. This problem has been overcome by the approach of multilevel analysis, which is able to correct for this problem 16. This statistical technique allows for the dependency of measurements in hierarchically structured data, whereas traditional regression analysis presupposes the independence of observations. Another advantage of this technique is that it can separately examine the effects of variables of different levels (repeated measures) and can also be used when measurements have been made at different times (e.g. recordings at 16 weeks, at 16 2/7 weeks or 16 6/7 weeks of gestation). An intraobserver and interobserver variation analysis forms no part of any publication of reference curves of the fetal renal kidney. Only Bertagnoli 6 in 1983 performed an interobserver analysis between 3 operators who each used a different ultrasound machine. Some studies describe the design, the patient selection and the methods of analysis but do not report on intra- or interobserver error 12, 13, 15. The aim of this study was to estimate a reference curve for the size and growth of the fetal kidney and fetal renal pelvis.

26 26 Chapter 3 MATERIAL AND METHODS One hundred and twelve low risk women with 116 fetuses (4 twin pregnancies) were asked during an ultrasound scan for validating gestational age by measuring the crown-rump length, to participate in the study. After written informed consent the women were divided in two groups in order of registration. One half was examined at around and 40 weeks of gestational age and the other half at around and 42 weeks of gestational age. All examinations were made by one ultrasonographer (HDE), using the multifrequency, abdominal transducer PVM 375 AT of the Toshiba Power Vision 6000, type SSA 370 A with (manufacturer Toshiba, Tokyo, Japan). An intra- and interobserver variation analysis preceded the study resulting in a good consistency in measurements. Exclusion criteria were: mono-chorionic twin pregnancies, chromosomal or congenital defects of the fetus, small for gestational age at birth (SGA = weight at birth <2.3% (17) percentile for gestational age) and maternal disease which might effect fetal growth (diabetes mellitus, hypertension requiring treatment). The length of the kidney was measured when in a sagittal plane the full length with the renal pelvis was visualised. Perpendicular to this, in the largest sectional plane, the anteroposterior (AP) and transverse diameter of the kidney were measured by placing the callipers from outer to outer border. In the same sectional plane the AP and transverse diameters of the renal pelvis were measured by placing the callipers on the inner borders of the renal tissue. All measurements were obtained three times and the data were averaged and stored in a database. Statistical analysis was performed by multilevel analysis with the software program Mln (Multilevel Model Project, London, UK) 16 and SPSS, version 10.1 (Statistical Product and Service Solutions, Chicago) to construct nomograms (medians and the 5 th and 95 th centiles). RESULTS Five fetuses were excluded: small for gestational age (n=2), congenital anomalies (n=3; triploidy, clubfoot and hydronephrosis >10mm anteroposterior diameter of the renal pelvis). Of the remaining 107 woman and 111 fetuses all measurements could be obtained from both kidneys and both renal pelves with a total of 628 data for each separate measurement for length, anteroposterior diameter and transverse diameter

27 Charts of fetal kidney and renal pelvis 27 of both kidneys and for the anteroposterior and transverse diameter of both renal pelves. Statistical analysis was performed on all separate measurement from the right and left fetal kidney. The charts of length, anteroposterior and transverse diameter and perimeter of the left and right kidney were about identical with a high correlation coefficient between the measurements from the right and left side: R = for length, R = for AP-diameter, R = for transverse diameter and R = for perimeter. There was no asymmetry between the right and left renal pelvic measurements (Table 1: Kappa 0,459). The charts made of the anteroposterior and transverse diameter of the right and left renal pelvis were similar with correlation coefficients between measurements of the right and left side of: R = for the AP-diameter and R = for the transverse diameter. In only 16 of 628 comparisons between right and left pelvic size there was a difference of more than 2 millimetres between the measurements (Table 1). Table 1 Partition of the dimensions of the right and left renal pelvis. Kappa 0,459 left renal pelvis measurements (mm) right renal pelvis measurements (mm) 0 <2 mm 2 <4 mm 4 <6 mm 6 <8 mm 8 <10 mm total 0 < 2 mm < 4 mm* < 6 mm < 8 mm <10 mm total The data of all right and left kidney and renal pelvis measurements were averaged to obtain the reference charts. The account of the nomograms of the 5 th, 50 th and 95 th fitted centiles of length, anteroposterior and transverse diameter and of the perimeter from the kidney are shown in Table 2 with the corresponding charts (Figures 1 to 4).

28 28 Chapter kidney length (mm) gestational age in weeks Figure 1 Fitted 5 th, 50 th and 95 th centiles for the kidney length and the raw data kidney AP-diameter (mm) gestational age in weeks Figure 2 Fitted 5 th, 50 th and 95 th centiles for the anteroposterior kidney diameter and the raw data.

29 Charts of fetal kidney and renal pelvis kidney transverse diameter (mm) gestational age in weeks Figure 3 Fitted 5 th, 50 th and 95 th centiles of the transverse kidney diameter and the raw data kidney perimeter gestational age in weeks Figure 4 Fitted 5 th, 50 th and 95 th centiles of the perimeter of the kidney and the raw data.

30 Table 2 Fitted centiles of fetal renal kidney length, anteroposterior diameter, transverse diameter and perimeter with the number of fetuses for exact weeks between 16 and 42 weeks of gestational age. weeks of gestation N fetuses fitted centiles kidney length fitted centiles kidney anteroposterior diameter fitted centiles kidney transverse diameter fitted centiles kidney perimeter 5 th 50 th 95 th SD 5 th 50 th 95 th SD 5 th 50 th 95 th SD 5 th 50 th 95 th SD ,9 14,3 15,7 0,70 7,4 8,6 9,8 0,60 7,3 8,6 9,9 0,66 23,5 27,0 30,6 1, ,3 15,7 17,2 0,74 8,2 9,5 10,7 0,64 8,3 9,7 11,1 0,70 26,2 29,9 33,7 1, ,6 17,1 18,7 0,79 9,0 10,4 11,7 0,68 9,3 10,8 12,2 0,74 29,2 33,2 37,2 2, ,9 18,5 20,2 0,84 9,8 11,2 12,6 0,71 10,3 11,8 13,2 0,78 31,8 36,0 40,2 2, ,2 19,9 21,6 0,88 10,6 12,1 13,5 0,75 11,2 12,8 14,4 0,82 34,6 39,1 43,5 2, ,0 21,8 23,7 0,94 11,3 12,8 14,4 0,79 12,1 13,9 15,4 0,86 37,9 42,3 46,9 2, ,7 22,6 24,5 0,96 12,0 13,6 15,3 0, ,8 16,5 0,90 39,8 44,6 49,5 2, ,0 24,0 25,9 1,00 12,7 14,4 16,1 0,86 13,9 15,7 17,6 0,94 42,3 47,3 52,4 2, ,2 25,3 27,3 1,05 13,4 15,1 16,9 0,90 14,7 16,6 18,6 0,99 44,6 49,9 55,2 2, ,5 26,6 28,7 1,10 14,0 15,8 17,7 0,95 15,5 17,5 19,5 1,03 46,8 52,4 57,8 2, ,6 27,9 30,1 1,14 14,6 16,5 18,4 0,98 16,3 18,4 20,5 1,07 49,1 54,8 60,6 2, ,8 29,0 31,3 1,19 15,2 17,2 19,2 1, ,3 21,4 1,10 50,9 57,2 62,8 3, ,0 30,4 32,8 1,23 15,7 17,8 19,9 1,05 17,8 20,1 22,3 1,15 53,3 59,5 65,7 3, ,2 31,7 34,2 1,27 16,3 18,4 20,6 1,10 18,5 20,8 23,2 1,19 55,3 61,7 68,1 3, ,3 32,9 35,5 1,31 16,8 19,0 21,2 1,13 19,2 21,6 24 1,23 57,2 63,8 70,5 3, ,5 34,2 36,8 1,35 17,3 19,6 21,8 1,17 19,9 22,4 24,9 1,27 59,0 65,9 72,7 3, ,6 35,3 38,1 1,40 17,7 20,1 22,5 1,20 20,5 23,1 25,7 1,31 60,8 67,9 74,9 3, ,3 36,5 40,1 1,43 18,4 20,6 23,4 1,23 21,2 23,7 26,5 1,35 63,1 69,5 77,5 3, ,0 37,7 40,8 1,49 18,6 21,1 23,6 1,28 21,8 24,5 27,2 1,40 64,1 71,6 79,1 3, ,0 38,9 41,9 1,52 19,0 21,6 24,1 1,31 22, ,9 1,43 65,7 73,4 81,2 3, ,0 40,0 43,1 1,57 19,3 22,0 24,7 1,35 22,9 25,8 28,7 1,48 67,1 75,0 83,0 4, ,9 41,0 44,2 1,61 19,7 22,4 25,1 1,39 23,5 26,4 29,5 1,52 68,5 76,7 84,8 4, ,8 42,0 45,2 1,66 20,0 22,8 25,6 1,43 23, ,56 69,8 78,2 86,6 4, ,6 42,9 46,2 1,71 20,3 23,2 26,0 1,47 24,4 27,5 30,7 1,60 71,0 79,6 88,3 4, ,4 43,8 47,1 1,75 20,6 23,5 26,5 1,50 24,8 28,1 31,3 1,64 72,2 81,0 89,9 4, ,1 44,8 48 1,79 20,8 23,8 26,8 1,54 25,2 28,5 31,8 1,68 73,3 82,3 91,4 4, ,8 45,6 48,8 1,81 21,1 24,1 27,2 1,57 25,6 28,9 32,2 1,72 74,4 83,6 92,9 4,68 total 111

31 Charts of fetal kidney and renal pelvis 31 The account of the nomograms of the 5 th, 50 th and 95 th fitted centiles of the anteroposterior and transverse diameter from the renal pelvis are given in Table 3 with the corresponding charts (Figures 5 and 6). Table 3 - Fitted centiles of the anteroposterior and transverse diameter of the fetal renal pelvis with the number of fetuses for exact weeks between 16 and 42 weeks of gestational age weeks of gestation N fetuses fitted centiles pelvis anteroposterior diameter fitted centiles pelvis transverse diameter 5 th 50 th 95 th SD 5 th 50 th 95 th SD ,2 1,1 1,9 0,4-0,8 1,4 3,6 1,1 17 0,5 1,3 2,2 0,5-0,6 1,7 4,0 1, ,7 1,6 2,5 0,5-0,1 2,3 4,7 1,2 19 0,9 1,8 2,8 0,5 0,1 2,6 5,1 1, ,2 2,1 3,1 0,5 0,5 3,1 5,6 1,3 21 1,3 2,3 3,3 0,5 0,9 3,4 6,0 1, ,6 2,6 3,6 0,5 1,0 3,7 6,5 1,4 23 1,7 2,7 3,8 0,5 1,2 4,0 6,8 1, ,9 3,0 4,0 0,5 1,4 4,3 7,2 1,5 25 2,0 3,1 4,2 0,6 1,6 4,5 7,5 1, ,2 3,3 4,4 0,6 1,7 4,8 7,9 1,6 27 2,3 3,4 4,6 0,6 1,9 5,0 8,2 1, ,4 3,6 4,7 0,6 2,0 5,2 8,5 1,7 29 2,5 3,7 4,9 0,6 2,0 5,4 8,8 1, ,5 3,8 5,0 0,6 2,1 5,5 9,0 1,8 31 2,6 3,8 5,1 0,6 2,1 5,7 9,2 1, ,6 3,9 5,2 0,7 2,1 5,8 9,4 1,9 33 2,7 4,0 5,3 0,7 2,1 5,8 9,6 1, ,7 4,0 5,4 0,7 2,0 5,9 9,7 2,0 35 2,6 4,0 5,4 0,7 2,0 5,9 9,8 2, ,6 4,0 5,5 0,7 1,9 6,0 10,0 2,1 37 2,6 4,0 5,5 0,7 1,8 5,9 10,0 2, ,5 4,0 5,5 0,8 1,6 5,9 10,1 2,2 39 2,5 4,0 5,5 0,8 1,5 5,8 10,1 2, ,4 4,0 5,5 0,8 1,3 5,7 10,1 2,3 41 2,3 3,9 5,4 0,8 1,1 5,6 10,1 2, ,3 3,9 5,4 0,8 1,0 5,5 10,0 2,3 total 111

32 32 Chapter pyelum AP-diameter (mm) gestational age in weeks Figure 5 Fitted 5 th, 50 th and 95 th centiles of the A-P diameter of the renal pelvis and the raw data transverse pyelum diameter (mm) gestational age in weeks Figure 6 Fitted 5 th, 50 th and 95 th centiles of the transverse diameter of the renal pelvis and the raw data.

33 Charts of fetal kidney and renal pelvis 33 The centile charts were compared with previously published charts from Chitty and Altman 18 for length, AP-diameter, transverse diameter and renal pelvis measurements and from Pruggmayer and Terinde 3 for length and transverse diameter. See figures 7, 8, 9 and kidney length (mm) gestational age in weeks Figure 7 Comparison of 5 th, 50 th and 95 th centiles for kidney length measurements obtained in this study (solid lines) and the 10 th, 50 th and 90 th centiles of Chitty (dashed lines ) and the 5 th, 50 th and 95 th centiles of Pruggmayer (dashed lines ). 35 kidney A-P diameter (mm) gestational age in weeks Figure 8 Comparison of 5 th, 50 th and 95 th centiles for kidney anteroposterior diameter obtained in this study (solid lines) and the 10 th, 50 th and 90 th centiles of Chitty (dashed lines ).

34 34 Chapter 3 kidney transverse diameter (mm) gestational age in weeks Figure 9 Comparison of 5 th, 50 th and 95 th centiles for kidney transverse diameter obtained in this study (solid lines) and the 10 th, 50 th and 90 th centiles of Chitty (dashed lines ) and the 5 th, 50 th and 95 th centiles of Pruggmayer (dashed lines ). renal pelvis (mm) gestational age in weeks Figure 10 Comparison of 5 th, 50 th and 95 th centiles for anteroposterior renal pelvis measurements obtained in this study (solid lines) and the 10 th, 50 th and 90 th centiles of Chitty (dashed lines). DISCUSSION Several charts of fetal kidney sizes have been published before, but some with shortcomings in data collection or with methodological weaknesses. Our charts of the fetal kidney and fetal renal pelvis were obtained from longitudinal data derived from prospective investigations that were done specifically for the development of the centile charts. The data were obtained from a large longitudinal sample and therefore the charts are suitable for size and growth measurements i.e. for comparing renal size at a known gestational age between 16 and 42 weeks of gestation and for

35 Charts of fetal kidney and renal pelvis 35 following the growth of the fetal kidney by comparing the measured data of the fetus between two separate occasions with the fitted data. Until now there had not been a chart published, which combines both possibilities, when taken into account methodological pitfalls and incorrect methods in design such as not validating gestational age 1-5, only partly covering the second and third trimester 1, 2, 6-8, measuring only one or two dimensions of the kidney 1, 2, 4-6 or averaging both crosssectional and longitudinal data 1, 3-6, 8, 11. Moreover this study has the added advantage that the statistical analysis used is able to correct for missing data. We were limited in comparing our findings with previously published charts of kidney size because several studies gave no raw data or when they did so they gave evidence of methodological weakness or did not cover the whole second and third trimester of pregnancy. None of the previous studies had carried out an intra- and interobserver variation analysis before data collecting, as we did. Chitty and Altman 18 obtained their data in a cross-sectional study, including approximately 15 to 20 cases per week. Measurements were only done once. They published the 3 rd, 10 th, 50 th, 90 th and 97 th centile. Pruggmayer and Terinde 3 also performed a cross-sectional study including 612 fetuses. They did not give data on the number of fetuses measured each week and excluded 18.6% of infants because they were either large or small for dates. The most obvious differences between our kidney charts and those of the other two groups relate to the smaller ranges that we found. The 50 th centile of measurements of Chitty and Altman was higher than ours at around 30 weeks of gestation, but almost the same near term. The 50 th centile for the transverse kidney diameter was about the same in the three studies, apart from term age, when our data were in between those of the two other groups. It is difficult to explain the differences between the findings of the three studies. The fact that we have repeated the measurements three times may have resulted in narrowed ranges. Regarding the fetal renal pelvic dimensions only two charts have been published before. Chitty et al 18 published a chart based on cross-sectional data from fetuses measured only once for the purpose of the study but examined only a low number of fetuses at each week of gestation (3 to 11: mean 7). Scott 5 published a chart with a mixture of cross-sectional and longitudinal data obtained during routine scanning and

36 36 Chapter 3 as a consequence some fetuses were only included once whereas others were included at many ages; the resulting scatter diagrams of kidney and pelvis sizes show many data at around weeks and at around weeks and a lack data beyond 36 weeks of gestational age. They did not publish the raw data. We therefore only compared our chart with the one published by Chitty and Altman. Their chart suggests a linear growth of the renal pelvis size, with a wide range, whereas ours shows a curved line, with no increase size from 32 weeks onwards, an a considerable narrower range (especially, when taken into account the fact that we gave the 5 th, 50 th and 95 th centile and Chitty and Altman the 10 th, 50 th and 90 th centiles) (Figure 8). Renal pathology often presents itself late in pregnancy. A chart for size and growth chart of the kidney may be useful in case of diagnostic problems. The same holds for renal pelvic dilatation, which is a common sonographic finding in pregnancy.

37 Charts of fetal kidney and renal pelvis 37 REFERENCES 1. Cohen HL, Cooper J, Eisenberg P, Mandel FS, Gross BR, Goldman MA, et al. Normal length of fetal kidneys: sonographic study in 397 obstetric patients. AJR Am J Roentgenol 1991;157(3): Gloor JM, Breckle RJ, Gehrking WC, Rosenquist RG, Mulholland TA, Bergstralh EJ, et al. Fetal renal growth evaluated by prenatal ultrasound examination. Mayo Clin Proc 1997;72(2): Pruggmayer M, Terinde R. [Fetal kidney screening: growth curves and indices]. Geburtshilfe Frauenheilkd 1989;49(8): Sagi J, Vagman I, David MP, Van Dongen LG, Goudie E, Butterworth A, et al. Fetal kidney size related to gestational age. Gynecol Obstet Invest 1987;23(1): Scott JE, Wright B, Wilson G, Pearson IA, Matthews JN, Rose PG. Measuring the fetal kidney with ultrasonography. Br J Urol 1995;76(6): Bertagnoli L, Lalatta F, Gallicchio R, Fantuzzi M, Rusca M, Zorzoli A, et al. Quantitative characterization of the growth of the fetal kidney. J Clin Ultrasound 1983;11(7): Bernaschek G, Kratochwil A. [Ultra-sound study on the growth of the fetal kidney in the second half of pregnancy (author's transl)]. Geburtshilfe Frauenheilkd 1980;40(12): Jeanty P, Dramaix-Wilmet M, Elkhazen N, Hubinont C, van Regemorter N. Measurements of fetal kidney growth on ultrasound. Radiology 1982;144(1): Chiara A, Chirico G, Barbarini M, De Vecchi E, Rondini G. Ultrasonic evaluation of kidney length in term and preterm infants. Eur J Pediatr 1989;149(2): Gonzales J. [Anatomical measurements during fetal growth of the kidney. Its value for the ultrasonographer and the anatomo-pathologist (author's transl)]. J Gynecol Obstet Biol Reprod (Paris) 1981;10(2): Vries de L, Levene MI. Measurement of renal size in preterm and term infants by real-time ultrasound. Arch Dis Child 1983;58(2): Altman DG, Chitty LS. Design and analysis of studies to derive charts of fetal size. Ultrasound Obstet Gynecol 1993;3(6): Altman DG, Chitty LS.Charts of fetal size: 1. Methodology. Br J Obstet Gynaecol 1994;101(1): Royston P, Altman DG. Design and analysis of longitudinal studies of fetal size. Ultrasound Obstet Gynecol 1995;6(5): Royston P, Wright EM. How to construct 'normal ranges' for fetal variables. Ultrasound Obstet Gynecol 1998;11(1): Goldstein H. Multilevel statistical models. 2nd ed. London: University of London. ed; Kloosterman GJ. On intrauterine growth. Int J Gynaecol Obstet 1970;8: Chitty LS, Altman DG. Charts of fetal size: kidney and renal pelvis measurements. Prenat Diagn 2003;23(11):891-7.

38

39 Chapter 4 Growth and size charts of the fetal adrenal gland H.A.M. Damen - Elias a, R.H. Stigter a,c, P. Westers b, G.H.A. Visser a a Department of Perinatology and Gynaecology, University Hospital Utrecht, The Netherlands b Centre for Biostatistics, Utrecht University, The Netherlands c Department of Obstetrics and Gynaecology, Deventer Hospital, Deventer, The Netherlands

40 40 Chapter 4 ABSTRACT Objectives It was the aim of this study was to develop a reference curve for size and growth of the length of the fetal adrenal gland. Methods Longitudinal prospective study of one hundred eleven fetuses that were scanned every four weeks. The ultrasound measurements started in one half of the cases at sixteen weeks of gestation and in the other half at eighteen weeks. Statistical analysis was performed by multilevel analysis. Results Fitted 5 th, 50 th and 95 th centile charts for the length of the adrenal gland are presented with the raw data. There was a high correlation between adrenal and kidney length (R = 0,932) with a ratio of 2 to 7. Conclusions The chart for size and growth is of use for measurements of the length and growth of the adrenal gland.

41 Charts of fetal adrenal gland 41 INTRODUCTION Knowledge of the normal range of the growth and size of the fetal adrenal gland is of importance for the identification of morphological changes once an anomaly is suspected. A number of reference values of adrenal gland measurements have been published 1-7, but none covers the whole second and third trimester of the pregnancy. Moreover in none of these studies data were used with an ultrasound validated gestational age and in some publications only post-mortem specimens were used 1, 2. All studies used cross-sectional data. Since 1990 there has been no study using abdominal investigations. In 1993 a study has been published using transvaginal ultrasound between 12 and 17 weeks of gestation. It was the aim of this prospective study to develop a reference curve for the size and growth of fetal adrenal gland length. MATERIAL AND METHODS Data of the length of the adrenal gland were collected in a longitudinal, prospective study. Measurements of the fetal kidney and fetal renal pelvis were collected simultaneously and will be published in a separate paper. One hundred and twelve low risk women with 116 fetuses (4 twin-pregnancies) were included after written informed consent had been obtained. They were divided in two groups who were scanned at four weeks interval starting at 16 weeks or at 18 weeks gestational age, respectively. All women had had a dating scan before 13 weeks of gestation. One experienced ultrasonographer (HDE) made all the examinations using the multifrequency transducer PVM 375 AT of the Toshiba Power Vision 6000, type SSA 370 A (manufacturer Toshiba, Tokyo, Japan). An intra- and interobserver variation analysis was performed before carrying out this study. The results of which will be published elsewhere. Women were excluded when they had a mono-chorionic twin pregnancy, a chromosomal or congenital defect of the fetus, a small for gestational age infant (SGA = birth weight according to growth charts <2.3% percentile 8 ) or a maternal disease which might effect fetal growth (diabetes mellitus, hypertension requiring treatment).

42 42 Chapter 4 The suprarenal adrenal glands are heart-shaped structures located cranially to the kidney like little helmets placed askew on top of the kidneys. They are imaged as relatively anechoic pyramidal structures but sometimes the echogenicity of the adrenal glands is similar to that of the adjacent kidney. In a transverse scan three layers can usually be distinguished; the cortex is hypoechoic and the central medulla hyperechoic. When visualising the full length of the kidney in a sagittal plane the length of the kidney was measured and in the same plane the length of the kidney including the adrenal gland was measured by placing the callipers from the outer to outer border. The adrenal gland length was determined by subtraction of the kidney length from the total length (Figure 1). Statistical analysis was performed by using SPSS, version 10.1 (Statistical Product and Service Solutions, Chicago) and by multilevel analysis using the software program Mln (Multilevel Model Project, London, UK 9 ) to construct nomograms (medians and the 5 th and 95 th centiles). The latter method is able to rectify the problem of missing data inherent to any study with longitudinal data. A adrenal gland B D kidney ureter C Figure 1 The length of the fetal adrenal gland is AC BC.

43 Charts of fetal adrenal gland 43 RESULTS Five fetuses were excluded, SGA (n=2), congenital anomalies (n=3; triploidy, clubfoot and hydronephrosis >10mm anteroposterior diameter of the renal pelvis), leaving data from 111 fetuses for analysis. At each gestational age all measurements could be obtained from both the right and the left adrenal gland with a total of 628 data for each side. Statistical analysis was performed on the separate measurements of the length of the right and left adrenal gland. The charts were virtually identical with a high correlation of R = (Figure 2) left adrenal gland length (mm) rigth adrenal gland length (mm) Figure 2 Scatter from the data of the right and left adrenal gland length with the fitted correlation line (correlation R = 0,979). The data of the length of right and left adrenal gland were averaged to construct the reference chart. The account of the nomograms of the 5 th, 50 th and 95 th fitted centiles are given in Table 1 with the corresponding chart in Figure 3.