ORIGINAL ARTICLE

Postnatal outcome of prenatally diagnosed severe fetal renalpelvic dilatationSailesh Kumar1*, ShikhaWalia1, Obori Ikpeme2, Eko Zhang1, Gowrishankar Paramasivam1, Sanjiv Agarwal2, Karl Murphy1,3, Jane Deal4 andLorin Lakasing1,3

1Centre for Fetal Care, Queen Charlotte’s and Chelsea Hospital, Imperial College London, London, W12 0HS, UK2Department of Urology, Imperial College Healthcare NHS Trust, London, W12 0HS, UK3Department of Fetal Medicine, St Mary’s Hospital, Praed Street, London, W2 1NY, UK4Department of Paediatric Nephrology, St Mary’s Hospital, Praed Street, London, W2 1NY, UK*Correspondence to: Sailesh Kumar. E-mail: sailesh.kumar@imperial.ac.uk

ABSTRACTObjective We correlated the prenatal severity with the postnatal outcome of prenatally detected renal pelvic dilatation (RPD).

Methods Cases of prenatally detected RPD referred between January 2002 and December 2008 were included. SevereRPD was defined as an anterior–posterior diameter of 15mm, mild and moderate dilatation was defined as 6 to<10mmand 10 to<15mm, respectively. Postnatal diagnosis, the need for surgery and the correlation with the prenatalseverity was ascertained.

Results Of the 762 patients with RPD, 492 (64.5%) were mild, 167 (21.9%) were moderate, and 103 (13.5%) were severe.The male:female ratio for the severe cohort was 5:1. Of the sever cases, 68% had progressive dilatation. Of the mild/moderate cases, 5% progressed to severe dilatation. PUJ obstruction was confirmed in 48 cases (60.8%), severe VURin 11 cases (14%), VUJ obstruction in 5 cases (6%), PUV in 2 cases (2.5%), and a nonidentifiable cause in 13 cases(16.5%). Ten of the 48 (20.8%) babies with PUJ obstruction required surgery within the first year of life.

Conclusion An obstructive cause is usually present in severe cases, which are more likely to require surgery if there isPUJ obstruction. A high male:female ratio was present in this group. © 2012 John Wiley & Sons, Ltd.

Funding sources: SK was funded by the Imperial College Healthcare NHS Trust comprehensive Biomedical Research Centre (BRC) scheme

Conflicts of interest: None declared

INTRODUCTIONFetal renal pelvic dilatation (RPD) is one of the most commonanomalies detected in prenatal ultrasound with an estimatedprevalence of approximately 2 to 5.5%.1 It is frequentlydetected during routine midtrimester ultrasound assessmentof the fetal anatomy and represents a large cohort of referralsto a fetal medicine unit.

Dilatation of the fetal renal pelvicalyceal system may be anindicator of a variety of urinary tract abnormalities, ranging fromobstruction to vesicoureteric reflux. It may also be a marker foraneuploidy with increased incidence in fetuses with trisomy 21.2,3

However, in general, isolated urinary tract anomalies have a lowassociation with karyotypic abnormalities.4 What is increasinglyapparent however is that more severe dilatation of the renalpelvis is associated with postnatal pathology and that increasingseverity often requires surgical intervention in childhood.5–7

In recent years, prenatal grading of severity has beenstandardized with the anterior–posterior (AP) diameter ofthe renal pelvis being used to both define and grade theseverity of dilatation.6,8 This measurement provides aquantitative measure and allows correlation with postnataland longer term outcomes. A recent meta-analysis5

demonstrated that severe dilatation carried a significant riskof postnatal pathology (88.3%). These authors also showedthat mild and moderate dilatation was associated withpostnatal abnormalities in 11.9% and 45.1% of cases,respectively, suggesting that appropriate counselling andfollow up be organised when this abnormality is detectedprenatally.

The purpose of this study was to investigate the correlationbetween prenatal diagnosis of severe RPD and the need forpostnatal treatment.

MATERIAL AND METHODSThis was a retrospective observational study of all patientsreferred to the Fetal Medicine Service (Queen Charlotte’s &Chelsea Hospital and St Mary’s Hospital) at Imperial CollegeLondon with a prenatal diagnosis of isolated severe RPD (atany gestation) between January 2002 and December 2008.Queen Charlotte’s and St Mary’s hospitals are tertiary unitsand receive a larger number of prenatal referrals for fetalassessment. As this was considered a clinical audit, ResearchEthics Committee approval was not required. Cases wereidentified from the department’s fetal medicine database

Prenatal Diagnosis 2012, 32, 1–4 © 2012 John Wiley & Sons, Ltd.

DOI: 10.1002/pd.2893

(Astraia, Hamburg, Germany), computerized medical recordsand supplemented by individual chart review, obstetric andneonatal discharge summaries, and from paediatric radiologyrecords to facilitate correlation between prenatal findings andpostnatal imaging.

Severe RPD was defined as an AP diameter (in the transverseplane) of 15mm, whereas mild and moderate dilatation wasdefined as AP measurements of 6 to <10mm and 10 to <15mm,respectively. When both kidneys were found to be affected, thedilatation of the one with the larger diameter was used tocategorise the patient. The presence of unilateral or bilateraldilatation was also recorded. Fetuses with extra renal structuralanomalies or who were aneuploid were excluded from the study.

All patients were given the opportunity to meet a paediatricurologist for counseling about the postnatal management oftheir baby. For severe cases, serial scans were performed every4weeks to assess fetal growth and to monitor for progression ofthe RPD. At each scan an assessment of calyceal extension,renal cortical thickness, amniotic fluid volume, and fetalgrowth was made. The final prenatal ultrasound generally tookplace at 36weeks’ gestation. Postdelivery, all babies receivedprophylactic antibiotics (Trimethoprim) from birth and hadultrasound assessment of the kidneys and urinary tractperformed within the first few weeks of life. Subsequent followup depended on the initial ultrasound findings but rangedfrom every 4 to 6months with ultrasound imaging at each visit.In addition to the initial ultrasound assessment of the neonatalkidneys, most babies (all in the severe group) also received amicturating cystourethrogram, MAG3 scan or further ultrasoundscans of the kidneys in infancy depending on the results of theearlier tests.

Significant uropathy was defined as the presence of a urinarytract abnormality such as pelvi-ureteric junction (PUJ)obstruction, vesico-ureteric reflux (VUR), posterior urethralvalves (PUV), or vesico-ureteric junction (VUJ) obstruction.PUJ ‘obstruction’ was defined as the presence of isolatedhydronephrosis associated with a delayed MAG3 scan drainagepattern. The decision of surgical intervention for obstructivelesions was either poor renal function (<40%) or deterioratingfunction (> 5% change) based on two consecutive renal scansor recurrent urinary tract infections in the presence of asurgically correctable lesion. PUJ obstruction was treated bythe Anderson–Hynes pyeloplasty.

RESULTS

Prenatal findings (Table 1)During the study period, 762 patients with fetal RPD, out of 39415 patients, were seen and assessed giving an incidence of1.93% for this study. Thirty-two cases (4.2%) were twinpregnancies with at least one affected fetus. The medianmaternal age at referral was 27years (range 17–39years) with amedian parity of 1 (range 0–5). The median gestational age atdiagnosis was 24weeks (range 19–36weeks). Of the 762 patientswith RPD, 492 (64.6%) were mild, 167 (21.9%) were moderate,and 103 (13.5%) were classified as severe dilatation. In themild/moderate cohort, 62% (409/659) of fetuses were male

(male:female ratio of 1.6:1) compared with 83.5% (86/103) (male:female ratio 5:1 ) in the severe group.

Of the 103 cases of severe RPD, 20 cases (19.4%) wereassociated with other structural malformations: mostcommonly cardiac abnormalities in 35% of cases followed byskeletal and gastrointestinal tract abnormalities in 15% and10% of cases, respectively. The remaining 83 cases (80.6%)were isolated RPD. Of the 659 mild/moderate cases, additionalabnormalities were present in 105 cases (16%). In total, therewere 637 cases (554 mild/moderate and 83 severe) where therenal abnormality was the only abnormality present (i.e.isolated).

None of the severe cases regressed and 68% of the casesdemonstrated progressive dilatation as the pregnancyadvanced. Of the moderate cases and the mild cases, 4% and1%, respectively, progressed to become severe dilatation(>15mm) during the pregnancy. In 3% of the cases in thesevere group, both kidneys were involved. The left kidney wasaffected in 56% of the severe cases. The median gestation atdelivery for all isolated cases was 39weeks (range 34–41weeks).The prenatal diagnosis of the renal abnormality did notinfluence the timing or mode of delivery.

Postnatal outcome (Table 2)Postnatal data were available from 79 of the 83 cases (95%) ofsevere RPD.

The median follow-up time was 18months (range 12–50months). Of the 79 cases where follow-up data wereavailable, PUJ obstruction was confirmed in 48 cases (60.8%),severe VUR (defined as Grade III–V) in 11 cases (14%), VUJobstruction in five cases (6%), PUV in two cases (2.5%), anda nonidentifiable cause (i.e. nonobstructive) for the severedilatation in 13 cases (16.5%). Ten out of the 48 (20.8%)babies with PUJ obstruction required surgery within the firstyear of life. In total, 27 of the 79 (34%) babies with severedilatation required some form of surgery postnatally. Theremaining 52 babies were managed conservatively.

When the postnatal diagnosis was correlated with theseverity of the prenatal RPD, for PUJ obstruction the mediandilatation was 21mm (range 18–36mm), for VUR it was16mm (range 15–18mm), for VUJ obstruction it was 15mm(range 15–19mm), for it was PUV 17mm (range 15–26mm),and for nonobstructive it was 14mm (range 13–20mm). Only12% of cases resolved (defined as RPD <7mm) withoutany treatment. There did not appear to be any correlationbetween the presence or absence of calyceal dilatation andpostnatal outcome.

Table 1 Characteristics of cases

Severity of RPD Mild/moderate Severe

Number of cases 659 103

Male : female ratio 1.6:1 5:1

Associated anomalies 105 (16%) 20 (19.4%)

Isolated RPD 554 (84%) 83 (80.6%)

Progression of severity 5% 68%

S. Kumar et al.

Prenatal Diagnosis 2012, 32, 1–4 © 2012 John Wiley & Sons, Ltd.

DISCUSSIONAs a proportion, severe RPD accounts for between 1.5 and13.4%9 of all cases of fetal hydronephrosis. It however isresponsible for almost 88% of renal pathology postnatally8

and the results from this study are consistent with otherpublished work. Some other authors1,10 however suggest thatlower cut-off thresholds (<15mm) for RPD are effectivepredictors for postnatal surgery albeit with a false positive rateof almost 60%. Although there has been much debate aboutthe prenatal threshold for RPD that should trigger the needfor postnatal investigations, the majority of mild to moderatecases of RPD are not associated with significant renalpathology and cause minimal long-term nephrologicalsequelae. However, there may be an increased incidence ofurinary tract infections.

To date, there is no consensus on the optimal thresholdfor determining the need for postnatal follow-up. Coplenet al.11 recommended a cut-off of 15mm should be used,giving a sensitivity of 73% and specificity of 82%. Lee et al.5

estimated that only 11.9% of postnatal abnormalitiespresented with late pregnancy measurements of less than9mm, while 39% of postnatal abnormalities were noted atlevels of less than 15mm. What is clear is that lower cut-offvalues will be more sensitive in detecting postnatal pathology;however, this would result in higher false positive rates andincrease in parental anxiety and postnatal investigation rates.

Most mild cases tend to be transient or physiologic andhave no clinical significance and are likely to resolve.1

Obstructive lesions, however, are predisposed to deterioratingrenal function and/or recurrent urinary tract infections. Themost common cause of a severe unilateral hydronephrosisis usually PUJ obstruction. VUJ obstruction or severeunilateral VUR can also cause unilateral hydronephrosis,although there is usually associated ipsilateral uretericdilatation as well. Prenatally, PUV tends to cause bilateralhydronephrosis.

Lee et al.5 showed that specific prenatal ultrasoundfeatures, such as the AP renal pelvis diameter, could helpto predict the risk for postnatal abnormalities to stratifybabies that would require postnatal evaluation. The resultsfrom our study agree with Lee et al.5 and clearly show thatfor cases of severe RPD, obstructive causes predominate,particularly PUJ obstruction. Our results are consistent with

other published series.11–13 All of which indicate that themore severe the prenatal dilatation, the more likely it is tobe associated with significant postnatal pathology. Becauseof this, we and other authors recommend that patients withsevere fetal RPD should have comprehensive postnataldiagnostic investigations and be carefully followed up asthere is a significant risk of obstructive pathology thatwould predispose to longer term morbidity.

Other investigators14 have suggested that mild RPD(<10mm) is rarely linked to significant abnormality andnot predictive of urinary tract morbidity in childhood. Whatis interesting from our study is that despite extensiveinvestigations, no obvious cause for the RPD was identifiedin 16.5% of cases. Clearly, much longer follow up will berequired to confirm if there is any longer term morbidity,such as recurrent urinary tract infections or renal scarring.The median dilatation for this group of babies was 15mm,lower than for the other identifiable causes. It is not clearwhy such severe dilatation develops in the absence ofobstruction. It is possible that ureteric peristalsis istransiently impaired in this cohort, resulting in delayedemptying of the intrarenal collecting system. Nevertheless,this cohort of babies did not require any postnatal surgicalintervention. The incidence of transient RPD ranged from41 to 88%,1,15 although the incidence declined considerablythe more severe the dilatation was.16

Although our overall male:female ratio of 1.6:1 wasconsistent with other published series,17 what was strikingfrom our results was the much higher ratio (5:1) in the severegroup. To our knowledge, such a high ratio has not previouslybeen reported.

The findings from our study should improve our ability tocounsel parents about the likelihood of significant postnatalpathology and the need for further investigations andtreatment and should help guide clinicians in postnatalmanagement. Parents should be advised that whenprenatally diagnosed severe RPD is present, it is likely thatsurgery will be required in up to 34% of cases. They alsoneed to be advised that their child is more likely to needmore intensive postnatal investigations, including renalscintigraphy. It is also reasonable to counsel parents thatif severe RPD is identified midtrimester following the fetalanomaly scan, it is unlikely to resolve by delivery andtherefore they should be prepared for further investigationsand possible treatment postdelivery. Mild cases, in contrast,probably can be reassured and less intensively monitoredduring pregnancy.

In conclusion, we have demonstrated that severe RPD isassociated with significant postnatal pathology that requiresfurther treatment. Follow up in many cases can be quitelong and parents need to be advised about this. We havealso identified a high male:female ratio in this cohort.

ACKNOWLEDGEMENTSThe authors wish to thank Mrs Sheila Jacques and Ms MarciaWilson for their help in retrieving medical notes.

Table 2 Outcome of severe RPD and correlation with prenatalseverity

Post natal diagnosis Number of casesPrenatal dilatation (mm)(median and range)

PUJ obstruction 48 (60.8%) 21 (18–36)

Severe VUR 11 (14%) 16 (15–18)

PUV 2 (2.5%) 17 (15–26)

VUJ obstruction 5 (6%) 15 (15–19)

Nonobstructive 13 (16.5%) 14 (13–20)

Surgery required 27 (34%) 22 (15–36)

Fetal renal pelvic dilatation and outcome

Prenatal Diagnosis 2012, 32, 1–4 © 2012 John Wiley & Sons, Ltd.

WHAT’S ALREADY KNOWN ABOUT THIS TOPIC?

• Severe renal pelvic dilatation is more likely to be associated with anobstructive lesion.

• The majority of fetuses with mild renal pelvic dilatation will notrequire any significant treatment after birth.

• The risk of a structural renal anomaly is very low with mild renalpelvic dilatation.

WHAT DOES THIS STUDY ADD?

• Severe renal pelvic dilatation is more common in male fetuses.• Pelviureteric junction obstruction is the most common cause

diagnosed postnatally.• A non identifiable cause was present in almost 16% of cases of

severe isolated renal pelvic dilatation.

REFERENCES1. Sairam S, Al-Habib A, et al. Natural history of fetal hydronephrosis

diagnosed on mid-trimester ultrasound. Ultrasound Obstet Gynecol2001;17(3):191–6.

2. Yagel S, Anteby EY, et al. The role of midtrimester targeted fetal organscreening combined with the “triple test” and maternal age in thediagnosis of trisomy 21: a retrospective study. Am J Obstet Gynecol 1998;178(1 Pt 1):40–4.

3. Dagklis T, Plasencia W, et al. Choroid plexus cyst, intracardiac echogenicfocus, hyperechogenic bowel and hydronephrosis in screening for trisomy 21at 11 + 0 to 13 + 6 weeks. Ultrasound Obstet Gynecol 2008;31(2):132–5.

4. Staebler M, Donner C, et al. Should determination of the karyotype besystematic for all malformations detected by obstetrical ultrasound?Prenat Diagn 2005;25(7):567–73.

5. Lee RS, Cendron M, et al. Antenatal hydronephrosis as a predictor ofpostnatal outcome: a meta-analysis. Pediatrics 2006;118(2):586–93.

6. Nguyen HT, Herndon CD, et al. The Society for Fetal Urology consensusstatement on the evaluation and management of antenatalhydronephrosis. J Pediatr Urol 2010;6(3):212–31.

7. Yang Y, Hou Y, et al. Long-term follow-up and management of prenatallydetected, isolated hydronephrosis. J Pediatr Surg 2010;45(8):1701–6.

8. Kitchens DM, Herndon CD. Antenatal hydronephrosis. Curr Urol Rep2009;10(2):126–33.

9. Ahmad G, Green P. Outcome of fetal pyelectasis diagnosed antenatally.J Obstet Gynaecol 2005;25(2):119–22.

10. Gramellini D, Fieni S, et al. Diagnostic accuracy of fetal renal pelvisanteroposterior diameter as a predictor of significant postnatalnephrouropathy: second versus third trimester of pregnancy. Am JObstet Gynecol 2006;194(1):167–73.

11. Coplen DE, Austin PF, et al. The magnitude of fetal renal pelvicdilatation can identify obstructive postnatal hydronephrosis, and directpostnatal evaluation and management. J Urol 2006;176(2):724–7.

12. Wollenberg A, Neuhaus TJ, et al. Outcome of fetal renal pelvic dilatationdiagnosed during the third trimester. Ultrasound Obstet Gynecol2005;25(5):483–8.

13. Sidhu G, Beyene J, et al. Outcome of isolated antenatal hydronephrosis:a systematic review and meta-analysis. Pediatr Nephrol 2006;21(2):218–24.

14. Damen-Elias HA, Luijnenburg SE, et al. Mild pyelectasis diagnosed byprenatal ultrasound is not a predictor of urinary tract morbidity inchildhood. Prenat Diagn 2005;25(13):1239–47.

15. Mallik M, Watson AR. Antenatally detected urinary tract abnormalities: moredetection but less action. Pediatr Nephrol 2008;23(6):897–904.

16. Ransley PG, Dhillon HK, et al. The postnatal management ofhydronephrosis diagnosed by prenatal ultrasound. J Urol 1990;144(2 Pt 2):584–7.

17. Ismaili K, Hall M, et al. Results of systematic screening for minor degreesof fetal renal pelvis dilatation in an unselected population. Am J ObstetGynecol 2003;188(1):242–6.

S. Kumar et al.

Prenatal Diagnosis 2012, 32, 1–4 © 2012 John Wiley & Sons, Ltd.