Congenital bladder neck obstruction Is there a role for prenatal the - PDF document

1 1 Congenital bladder neck obstruction :
1 Congenital bladder neck obstruction : Is there a role for prenatal therapy :the PLUTO Study Mark. D Kilby. High morbidity and mortality associated with: Fetal lower urinary tract obstruction (LUTO)30% of renal tract anomalies detected at autopsy (Brand et al 1994)•Heterogeneous number of pathologies:Posterior urethral valves (PUV): 57%. Commonest, usually males.Urethral atresia 39%Cloacal plate anomalies, including megacystis microcolon syndrome-Prune Belly syndrome’ 4% Percentage of subjects in ESRF with age, according to underlying diagnosis (UK Renal Registry, 1996-2005)(Lewis MA. Semn.Fetal& Neonatol.2008; 13, 118-124) Causes of Congenital bladder neck obstructionPosterior urethral valvesUrethral atresia DilatedProximal urethra 2 Epidemiology of LUTOTotal prevalence of LUTO is 3.34 per 10,000 births, 2.24 per 10,000 live births-Prevalence significantly higher in Black and minority ethnic groupscompared to British European (OR 2.38;-Association of prevalence with area based deprivationmeasures LUTO subtypes:posterior urethral valves-48 cases (16.9%) 22% complex (1in 5) hromosomal anomalies)Prenatal detection:-66.2% abnormality detected antenatally -26.9% false positive diagnoses –24.5% reflux Postnatally detected cases have greater survival to one year 91% v 50% p0001 (isolated) (Morris RK et al BJOG. 2012;119(12):1455-64) Management of LUTO Diagnosis:Exclude other structural and chromosomal anomalies Prognostic indicatorsFetal urinalysisUltrasound features -Diagnostic.-Prognostic (R Morris. In: Progress in O&G Elsevier. 2006 (17)78-97.) Detection of L.U.T.O

2 in human pregnancyOligohydramnios•Fetal
in human pregnancyOligohydramnios•Fetal hydronephosisEnlarged fetal bladderDilated proximal urethraRenal parenchymal disease Prenatal ultrasound detecting � 60 –70% cases BladderUrethra 3 Fetal hydronephrosis -US grading system (Society for Fetal Urology) Grade II Dilated ureter (RobyrR. et al, Ultrasound ObstetGynecol. 2005;25(5):478-82) Urinary tract obstruction :effect on kidneys•Time of onset•Severity of obstruction Obstructive cystic dysplasia Postnatal renal damage SensitivitySpecificity 44% 100%73% 80%(Mahony BS. Radiology. 1984;152:143) 4 Not all dysplastic kidneys can be detected by US“Echogenic kidney” : microcysticchange Role of vesicocentesisand urinary analyteanalysis Fetal renal maturation effect on urinary analytes Tubular reabsorbtionNamicroglobulin•Urine becomes more hypotonic Tubular reabsorbtionCa•Mustuse gestation specific cut offs Gestation (wk) 122436(Nicolini U. Prenat Diagn 1991;11:159-66). LUTO -Good prognosisOsmolality 210 mOsm/L Sodium 100 mg/dL Chloride 90 mg/dL Calcium 8 mg/dL Protein 200 mg/dL microglobulin 6 mg/dL Normal Renal echogenicityNo oligohydramnios in early T (Crombleholme TM et al. AmJOG 1990. 162: 1239–1244).(Nicolini U. Prenat Diagn 1991;11:159-66). 5 Muller F. Clin Chem 1996;42:1855Muller F. Obstet Gynecol. 1993;82:813microglobulin Serial fetal urinalysis Good PrognosisFall in Na+, TP, Osm& B2-microglobulin with time.Poor prognosis:No Change in urinary analytes over time. Negative LR (95% CI)Urinary Sodium to predict postnatal renal function in fetuses with obstructive uropat

3 hy.Positive LR 100.0 1.25 (0.09 -17.0
hy.Positive LR 100.0 1.25 (0.09 -17.02) 7.00 (0.42 -116.40) 3.26 (1.43 -7.45) 9.44 (0.68 -131.60) 2.46 (1.00 -6.07) 3.03 (1.22 -7.51) 3.63 (0.28 -46.47) 9.90 (0.61 -161.74) 2.20 (1.27 -3.84) 34.45 (7.10 -167.10) 10.07 (2.13 -47.64) 3.71 (1.22 -11.26) 1.57 (0.48 -5.20)Positive LR (95% CI) Negative LR 0.01 Anumba et al 2005 Na�95th0.92 (0.36 -2.32) Bunduki et al 1998 Na�100mEq/l0.54 (0.22 -1.32) Bussieres et al 1995 Na�50mmol/l0.19 (0.04 -0.84) Grannum et al 1989 Na�100mmol/l0.06 (0.00 -0.93) Johnson et al 1994 Na�100mg/dl0.59 (0.31 -1.12) Johnson et al 1995 Na�100mg/dl0.41 (0.17 -1.01) Lipitz et al 1993 Na�100mg/dl0.48 (0.23 -0.96) Miguelez et al 2006 Na�2sd0.58 (0.32 -1.02) Muller et al 1993 Na�50mmol/l0.31 (0.11 -0.82) Muller et al 1999 Na�75mmol/l0.03 (0.00 -0.40) Nicolaides et al 1992 Na�95centile 0.30 (0.18 -0.50) Nicolini et al 1992 Na�95centile0.20 (0.06 -0.66) Reuss et al 1987 Na�100mEq/l0.43 (0.07 -2.63) Authora Second urine sampleB severe renal disease onlyPoor sensitivity and specificity for predicting renal dysfunction Morris RK et al. Prenat Diagn. 2007;27(10):900-11. Best:% for GA % for GA Worst:2-microglobulin less Fetal urine analysis to predict POOR : postnatal renal function in LUTO(Morris K. Prenat Diagn 2007; 27: 900–911) 6 Prognostic sonographic findings in fetuses with LUTO Diagnostic measureSensitivity (95% Confidence intervals

4 )Specificity (95% Confidence intervals)t
)Specificity (95% Confidence intervals)test and p valueArea under receiver characteristic curveOligohydramnios(0.510.74)0.76(0.650.85)19.67p=0.020.74Renal cortical appearance(0.370.76)0.84(0.710.94)10.29p=0.040.78Gestation at diagnosis 24 weeks(0.260.70)0.82(0.660.92)3.88p=0.140.68Morris et al, BJOG 2009;116:1290-1299•Systematic review 13 articles, 205 women Management of LUTO Exclude other structural and chromosomal anomalies Prognostic indicators:Fetal urinalysisUltrasound features•Intervention:Vesico-amniotic shunting (VAS)Fetal cystoscopy Therapeutic options (1) -CystoscopyCystoscopy with endoscopic ablation of PUV:-Allows restoration of normal fetal bladder dynamics and option for diagnosis-Systematic review n=4 papers:* Improved diagnosis 25-36.4%Search updated Jan 2012 no new papers (Morris et al Ultrasound ObstetGynecol. 2011;37(6):629-3) Therapeutic options (1) -CystoscopyCystoscopy:-perinatalsurvival Morris et al Ultrasound Obstet Gynecol 2011 0.65 (0.12, 3.55)0.23 (0.01, 3.83)1.18 (0.14, 9.95)6.49 (1.24, 33.85)6.31 (1.02, 39.01)7.39 (0.15, 372.38) 0.01 0.2 0.5 100 1000 Petoodds ratio (95% confidence interval)Meta-analysis n=2 Cochrane Q 0.82 p=0.364Holmes 200111Cystoscopyversus vesico-amniotic shuntingMeta-analysis n=2 Cochrane Q 0.005 p=0.94Ruano201023Quintero 19955Cystoscopyversus no treatment/ vesicocentesisonlyQuintero 19957AuthorNo of patientsPetoOdds ratio (95% CI)Favours cystoscopyFavours other treatmentCystoscopyimproved perinatalsurvival OR 6.49 (1.24 –33.85)Cystoscopyreduced perinatalsurvival OR (Morris et al Ultrasound ObstetGynecol. 2011;37(6)

5 :629-3) 7 Therapeutic options (1) -Cysto
:629-3) 7 Therapeutic options (1) -Cystoscopy(Morris et al Ultrasound ObstetGynecol. 2011;37(6):629-3) TableOutcomedataforfetuseswithconfirmedposteriorurethralvalves(voluntaryterminationpregnancyexcluded)Therapeuticcystoscopy(n=10)Vesicoshunting(n=10)intervention(n=5)CystoscopyvVASCystoscopyvinterventionSurvival(10/10)60(6/10)20(1/5)0.50.19Survivedrenalimpairment*(3/10)83(5/6)0(0/1)0.290.79Perinatalmortality(0/10)40(4/10)80(4/5)0.090.03renalimpairmentwasdefinedtwopapersserumcreatinine�50mg/dlmonthsagedefinedtwopapers Therapeutic options (2) -Vesico-amniotic shunting (VAS) Intervention v no treatment -Perinatal Morris et al BJOG 2010;117:382-390 13.85 (1.25, 153.03)10.31 (0.20, 541.25)28.03 (1.07, 735.05)20.09 (0.31, 1283.97)8.52 (0.62, 116.98)9.65 (0.55, 169.75)11.52 (0.94, 140.68) 26.19 (4.39, 156.25) 57.55 (3.98, 832.00) 2.58 (0.79, 8.45)2.06 (0.49, 8.70)2.93 (0.25, 33.94)3.82 (2.14, 6.84)0.14 (0.01, 6.82)0.54 (0.02, 11.82)1.91 (0.38, 9.60)0.21 (0.01, 6.08)5.62 (1.93, 16.37)3.99 (1.14, 13.95)2.37 (0.29, 19.39) 0.01 0.1 0.2 0.5 1.00E+05 Peto odds ratio (95% confidence interval)Pooled peto odds ratio (n=2)Freedman et al 1996Crombleholme et al 1990Poor prognosis based on fetal urinalysisPooled petoodds ratio (n=3)Quintero et al 1995Freedman et al 1996Crombleholme et al 1990Good prognosis based on fetal urinalysisPooled peto odds ratio (n=12)Wisser et al 1997Wilkins et al 1987Szaflick et al 1998Quintero et al 1995Nicolini et al 1991McLorie et al 2001McFadyen et al 1984Lipitz et al 1993Hutton et al 1997Freedman et al 1996Crombleholme et al 1990Anumba et al 2005A

6 uthorandyearFavourstreatmentFavourstreat
uthorandyearFavourstreatmentFavourstreatmentPetooddsratioandconfidenceintervalCochrane Q=12.15 p=0.35, I=0%, Z=4.52 p0.0001Cochrane Q=0.57 p=0.75, I=0%, Z=1.57 p=0.12Cochrane Q=0.60 p=0.44, I=0%, Z=3.58 p=0.0003 Shunting overall perinatal survival OR 3.86 (2.00-7.45) Poor prognosisOR 26.19 (4.39-156.25) Intervention v no treatment – Survival with normal renal functiMorris et al BJOG 2010;117:382-390 Peto odds ratio (95% confidence interval)2.98 (0.45, 19.62)0.92 (0.08, 10.55)17.05 (0.87, 334.20)Pooled peto odds ratio (n=2)Freedman et al 1996Crombleholme et al 1990Good prognosis based on fetal urinalysis0.67 (0.22, 2.00)12.18 (0.22, 665.65)0.42 (0.02, 9.83)0.22 (0.01, 5.77)0.12 (0.01, 5.84)0.64 (0.07, 5.67)2.98 (0.28, 31.65)0.15 (0.01, 2.77)0.01 0.2 100 1000 Pooled peto odds ratio (n=7)McFadyen et al 1984Lipitz et al 1993Johnson et al 1994Hutton et al 1997Freedman et al 1996Crombleholme et al 1990Anumbaet al 2005Cochrane Q=5.87 p=0.75, I=0%, Z=0.72 p=0.47Cochrane Q=2.21 p=0.53, I=0%, Z=1.13 p=0.26 1.00E+05 AuthorandyearFavourstreatmentFavoursinterventionPetooddsratioand95%confidenceinterval Shunting Postnatal survival with normal renal functionOR 0.50 (0.13-1.90) 8 (November 2006) Percutaneous shunt in L.U.T.O (PLUTO) -To determine whether vesico-amniotic shunting for LUTO, compared with conservative care, improves prenatal and perinatal mortality and renal function.-Perinatal morbidity, prognostic risk assessment at diagnosis, long-term outcomes for infants to 5 years. -Multicentre randomised controlled trial including a prospective registry Shunt Conservativemana

7 gement Shunt Conservativemanagement Foll
gement Shunt Conservativemanagement Followuntilyrs Followuntilyrs FetusRANDOMISED FetusentersREGISTRY FetalMedicinespecialistUNCERTAINshunt YESNO Percutaneous shunt in L.U.T.O (PLUTO) •Population:-Pregnant women with singleton, male fetus with isolated lower urinary tract obstruction•Setting:-21 fetal medicine centres within UK, National Maternity Hospital Dublin and University Medical Centre, Leiden, Netherlands. 9 Percutaneous shunt in L.U.T.O (PLUTO) Outcome measures:Pregnancy outcome: -Perinatal mortality, admission to SCBU, length of stay, oxygen dependency at discharge, birth weight, serum creatinine/ renal At 4-6 weeks: -Renal USS findings, weight, , surgery, MCUG At 1 year: -as 4-6 weeks, need for dialysis/transplantation, UTIs, BP, weight At 2 years: -renal function, developmental questionnaire (PARCA)At 5 years: -as 2 years with micturition questionnaire and Quality of life assessment (PedsQL) Intention to treat analysisRisk ratios and confidence intervals for survival Multi-level modelling for continuous repeatedvariables e.g. Creatinine clearanceSub-group analysis: gestational age at diagnosis, liquor volume, maternal age, learning curve effectsSecondary research:-Bayesian priors-Qualitative research-Epidemiology study-Economic analysis Results (1) –baseline characteristicsRandomised to shunt:N=16 [received n=15 (13 allocation, 2 crossovers)]Median GA 142 days (IQR 112-154) (range 98-224) p=0.71Liquor volume MPD median 1.6cm (IQR 0-2.93) (range 0-4.4); 62.5% 43.8% urinalysis25% karyotypeRandomised to conservative::(allocation, 3 crossovers)]Median GA 15

8 1 days (IQR 133-160) (range 112-296)Liqu
1 days (IQR 133-160) (range 112-296)Liquor volume MPD median 1.0cm (range 0-4.9); 60% centileNone had urinalysis26.7% karyotype •Recruitment Sep 2005 –Dec 2010•Cohort n= 145-Randomised n=31-Registered n=46-TOP n= 68 3,1,4,8, Terminationpregnancy Intrauterinedeath4weeks Neonataldeath Alive�28Allocated to shunt n=162 TOPs treatment relatedAll NND were within 24 hours of birthMedian interval randomisation to delivery 88.5 days (IQR 30.5-115) Median GA at delivery live births 254 days (IQR 238.5-263) 10 1,8,4, Terminationpregnancy Intrauterinedeath4weeks Neonataldeath Alive�28Allocated to conservative n=15 7/8 NND all within 24 hoursMedian interval randomisation to delivery 105 days (IQR 64-115) Median GA at delivery live births 255.5 days (IQR 241.5-264) (range Results (4)-Complications ComplicationsinterventionComplicationShuntinserted(n=15)OutcomeSROMwithindaysprocedure1/15(6.7%)IUDSROMwithindaysprocedure3/15OneIUD,pregnanicesaffectedchorioamnionitisterminationpregnancyShuntdislodgment3/15OnepregnancySROMTOPforchorioamnionitis+6weekspregnancyasciteshypoplasia,pregnancyshuntreinsertedalivedaysShuntblocked1/15(6.7%)TOPweeksascites3/15Onepregnancyalivedays,TOPweeks,hypoplasiaShuntreinserted1/15(6.7%Alivedaysterminationpregnancy;IUDintrauterinedeath;SROMspontaneousrupturemembranes;neonataldeath Preterm delivery eeks (n=4) , 2 in each arm, 1 in shunt arm eeks No significant difference mode of deliveryVaginal delivery shunt v cons 66.7% v 58.3% No significant difference between arms in birth weight median or shunt v cons 41.7% v 33.3% Results (5) –Live birt

9 h rate NNT to prevent a pregnancy loss =
h rate NNT to prevent a pregnancy loss = 20 harm As treated analysis RR 0.90 (0.61-1.33) Results (6) –Survival to 28 days Perinatal mortality NNT=5 As treated analysis RR 3.20 (1.06-9.62) p=0.03 11 Results (7) –Survival to 1 year Infant death NNT=5 As treated analysis RR 4.27 (1.07-17.0 ) p=0.02 Results (8)-Long-term follow-up Shuntallocation (n=8)DiagnosisOutcome (final follow-up)Renal function at final follow-up (Nadir Cr mol/l)DialysisTransplantSurgeryMorbidityYesPUVAlive (3 years)Moderate impairment (88)NoNoResection of valves, orchiopexy50th centile height and weight; 2 UTIS; DMSA 50% functionYesPUVAlive (3 years)No impairment (34)NoNoResection of valves, orchiopexy2 years significant motor and cognitive impairmentYesUrethral atresiaAlive (2 years)No impairment (37)NoNoTesticular surgeryYesPUVAlive (2 years)Moderate impairment (96)NoNoVesicostomy, nephrostomy.102 days in hospital in first 12 monthsYesUrethral atresiaAlive (2 years)Moderate impairment (227) NoNoVesicostomy, right nephrectomyFirst 6 weeks 43 days in hospital, first year 84 days, second year 19 days; recurrent UTIs; weight0.4th centileYesPUVAlive (1 year)Mild (123) NoNoResection of valves twice, circumcision, right orchiopexyDMSA non functioning right kidney, first 6 weeks 19 days in hospital, first year 20 daysYesPUVDied at 10 months of age renal failureESRF (342)Peritoneal at 6 weeks-VesicostomyNever dischargedNo shuntPUVAlive (1 year)Moderate (119)NoNoVesicostomy, resection of valves and bladder closureFisrt 6 weeks 24 days in hospital;Conservative(n=4)DiagnosisOutcome (final follow-up)Renal fun

10 ction at final follow-up (Nadir Cr mol/l
ction at final follow-up (Nadir Cr mol/l)DialysisTransplantSurgeryMorbidity 1.13 (0.65, 1.94)[0.52, 2.44]0.86 (0.34, 2.15)0.57 (0.05, 7.14)1.14 (0.32, 3.98)1.70 (0.35, 8.17)1.36 (0.19, 9.57)1.50 (0.46, 4.87) Riskratio(95% confidence interval)PLUTO trial 2012McFadyen et al 1984Lipitz et al 1993Hutton et al 1997Freedman et al 1996Anumba et al 2005Combined RR Author and year Favours no interventionFavours interventionRisk ratio (95% CI)RCT and observational data –abnormal perinatal renal function Bayesian analysis Prob(substantiallyincreasedsurvival)0.15Prob(increasedsurvival)0.79Prob(decreaedsurvival)0.21 Probabilitydensity 0.2 1.5 5 RR(survival)(a)Priorbeliefsexperts Prob(substantialincreasedsurvival)=0.25Prob(increasedsurvival)=0.86Prob(decreasedsurvival)=0.12 Probabilitydensity 0.2 1.5 5 RR(survival)(b)Posteriorestimatesusingexpertprior Prob(decreasedsurvival)=0.08Prob(increasedsurvival)=0.92Prob(substantiallyincreasedsurvival)=0.67 Probabilitydensity 0.2 0.5 1.5 5 RR(survival)(c)PosteriorestimatesusingvaguepriorBayesian prior and posterior estimates of relativerisk of survival to age 28 days. The prior distribution :(a)was obtained by eliciting prior distributionsfrom 52 experts, averaging the distributions, and fitting a normal distribution. The posterior distribution :(b) is based on combining the elicited prior with the observed ITT results. The posterior distribution:(c) is based on combining an informative priorwith the observed ITT results. The probabilities quoted are based on results of analysis in winBUGs. 12 Kaplan-Meier survival: PLUTO •RCT data

11 trend towards improved perinatal surviv
trend towards improved perinatal survival with shuntingbut with an increased risk of pregnancy loss. In RCT, the ‘as treated analysisdemonstrated that VAS improved perinatal survival to 60% as compared to no intervention (19%; p=0.03).•At 2 years of age only 2 babies of 7 (28%) survivors post-VAS were alive with no renal impairment. (Morris et al, Lancet. 2013;382(9903):1496-506) LUTO amenable to treatment is a PLUTO trial closed early due to poor recruitment:-Rare condition-Patients opting for TOP (53.4%)-Clinicians not in equipoisePreliminary results:Observational and RCT data trend towards improved perinatal survival with shuntingbut with an increased risk of pregnancy loss-Evidence for improvement of renal function is inconclusive but at 2 years of age only 2 babies of 7 (28%) survivors post-VAS were alive with no renal impairment (1 shunt, 1 cons.) Future:-Long term FU to determine infant mortality and morbidity-Economic analysis with decision analytic modeling with long term data PLUTO Collaborative GroupHTA and Wellbeing of Women•The women and babies who agreed to take part in our PLUTO 10%20%30%40%50%60%70%80%90%100% �=24 weeks 4 weeks 10%20%30%40%50%60%70%80%90%100% �=5th percentile 5th percentileElected conservative more likely to have a normal liquor volume p=0.02Elected conservative more likely to be later gestation 13 Aim:To gain an insight into the experiences of pregnant women approached to take part in PLUTOPurposeful sample of six womenMethods:Semi-structured interviews analysed using a phenomenological interpretive and themati

12 c approachFindings:Data saturation was n
c approachFindings:Data saturation was not achieved hence further interviews will be undertakenFour themes have been identified so far:Antenatal diagnosisParticipation in the studyEmotional impact of diagnosisSources of Support Within study analysis: No QALY as no consensus re methodology in childrenOutcome expressed as ‘Cost per additional disability free life year gained’Patient specific resource use and costsModel based analysis: Will allow projection of costs and benefits beyond trial follow up periodData from the epidemiology study and systematic review of shunt outcome studies will provide additional data to inform the model Effect of VASon mortality, %Frequency Density -40-2002040PaediatricNephrology Fetal Medicine PaediatricUrologyNo effectVAS is harmfulVAS is beneficial Study or Subgroup2.1.1 randomised shuntPLUTOSubtotal (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.56 (P = 0.58)2.1.2 registered shuntPLUTOSubtotal (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.89 (P = 0.37)2.1.3 randomised conservativePLUTOSubtotal (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 0.91 (P = 0.36)2.1.4 registered conservativePLUTOSubtotal (95% CI)Total eventsHeterogeneity: Not applicableTest for overall effect: Z = 1.89 (P = 0.06)Total (95% CI)Total eventsHeterogeneity: Chi² = 0.62, df = 3 (P = 0.89); I² = 0%Test for overall effect: Z = 2.32 (P = 0.02)Test for sub g rou differences: Chi² = 0.61 , df = 3 ( P = 0.90 ), I² = 0%EventsTotalEventsTotalWeight29.5%29.5%15.1%15.1%16.0%

13 16.0%39.4%39.4%100.0%M-H, Fixed, 95% CI1
16.0%39.4%39.4%100.0%M-H, Fixed, 95% CI1.62 [0.30, 8.65]1.62 [0.30, 8.65]3.00 [0.26, 33.97]3.00 [0.26, 33.97]3.07 [0.27, 34.37]3.07 [0.27, 34.37]3.79 [0.95, 15.07]3.79 [0.95, 15.07]2.91 [1.18, 7.19]24 weeks&#x-8.2;=24 weeksRisk Ratio (Non-event)Risk Ratio (Non-event)M-H, Fixed, 95% CI 0.01 0.1 Favours 24 weeksFavour&#x-6.7;s =24 weeks 14 Study or Subgroup3.1.1 randomised shuntPLUTOSubtotal (95% CI)Heterogeneity: Not applicableTest for overall effect: Z = 0.51 (P = 0.61)3.1.2 registered shuntPLUTOSubtotal (95% CI)Heterogeneity: Not applicableTest for overall effect: Z = 1.03 (P = 0.30)3.1.3 randomised conservativePLUTOSubtotal (95% CI)Heterogeneity: Not applicableTest for overall effect: Z = 0.72 (P = 0.47)3.1.4 registered conservativePLUTOSubtotal (95% CI)Heterogeneity: Not applicableTest for overall effect: Z = 3.10 (P = 0.002)Total (95% CI)Heterogeneity: Chi² = 6.77, df = 3 (P = 0.08); I² = 56%Test for overall effect: Z = 3.32 (P = 0.0009)Test for sub g rou differences: Chi² = 5.88 , df = 3 ( P = 0.12 ), I² = 49.0%EventsTotalEventsTotalWeightM-H, Fixed, 95% CI1.36 [0.41, 4.53]1.36 [0.41, 4.53]2.40 [0.46, 12.61]2.40 [0.46, 12.61]1.33 [0.61, 2.91]1.33 [0.61, 2.91]8.63 [2.20, 33.74]8.63 [2.20, 33.74]2.55 [1.47, 4.43]5th percentile&#x-100;=5th percentileRisk Ratio (Non-event)Risk Ratio (Non-event)M-H, Fixed, 95% CI 0.01 100Favours 5th percentileFavour&#x-12.;　s =5th percentile Table: Antenatal prognostic features and outcomePrognostic featureAllocated to shunt n=16OutcomeAllocated to conservative n=15OutcomeGestational age at diagnosis 24 weeks13/16 (81.3%)14/15

14 (93.3%)Perinatal death8/13 (61.5%)11/14
(93.3%)Perinatal death8/13 (61.5%)11/14 (78.6%)Survival to 28 days with CRF4/5 (80%)3/3 (100%)Survival to 28 days and well1/5 (20%)0Gestational age at diagnosis 24 weeks3/16 (18.8%)1/15 (6.7%)Alive at 28 days 2/3 (66.7%)1/1 (100%)Survival to 28 days with normal renal function1/2 (50%)0/1Poor prognosis for pulmonary hypoplasia at randomisationHigh73 NND due to PH, IUD at 16 weeks, 3 alive and well at 28 days6 NND due to PH, 2 alive and wellModerate3NND, TOP, alive and well at 28 days1NND due to PHMild64 alive at 28 days, 2 TOP62 alive at 28 days, 1 NND due to PH, I IUD, 2 TOPAntenatal progression of prognosis for pulmonary hypoplasiaWorsening of prognosis to high 42 TOP, 1 IUD, 1 alive at 28 days51 TOP, 2 NND due to PH, 2 alive at 28 daysImprovement in prognosis high to moderate1Alive at 28 days2Alive at 28 daysPoor prognosis for renal function at randomisation survivorsHigh42 CRF at 6-8 weeks, 2 alive and well at 1 year1CRF at 28 daysModerate1CRF at 28 days, peritoneal dialysis, died at 10 months2CRF at 1 year awaiting transplant at 2 years; CRF at 28 days died at 7 weeksMild31 alive and well at 2 years, 1 CRF at 2 years, 1 CRF at 28 days1CRF at 28 days A ntenatal progression of poor prognosis for renal function in survivorsWorsening moderate to high1CRF with peritoneal dialysis at 28 days, died at 10 months1CRF at 1 yearWorsening mild to high01CRF at 28 days TreatmentUrinalysis (Johnson etal)OutcomeShuntAbnormalAlive with CRFShuntNormalTOP 19+6ShuntAbnormalDied 2 hrsShuntAbnormalAlive with CRFShuntAbnormalAlive and well1 yrShuntAbnormalDied at 24 hrsShuntNormalTOP