- Email: [email protected]
Prenatal Anteroposterior Pelvic Diameter Cutoffs for Postnatal Referral for Isolated Pyelectasis and Hydronephrosis: More is Not Always Better
Prenatal Anteroposterior Pelvic Diameter Cutoffs for Postnatal Referral for Isolated Pyelectasis and Hydronephrosis: More is Not Always Better Giulia Bassanese, Laura Travan, Giuseppina D’Ottavio, Lorenzo Monasta, Alessandro Ventura and Marco Pennesi* From the University of Trieste (GB, AV), and Institute for Maternal and Child HealtheIRCCS “Burlo Garofolo” (LT, GD’O, LM, AV, MP), Trieste, Italy
Abbreviations and Acronyms APPD ¼ anteroposterior pelvic diameter GA ¼ gestational age MAG3 ¼ mercaptoacetyltriglycine US ¼ ultrasound UTI ¼ urinary tract infection VUR ¼ vesicoureteral reflux Accepted for publication May 13, 2013. Study received institutional review board approval. * Correspondence: Department of Pediatrics, Institute for Maternal and Child Healthe IRCCS “Burlo Garofolo,” Via dell’Istria 65/1, 34137 Trieste, Italy (telephone: 39-040-3785263; FAX: 39-040-3785290; e-mail: marco.pennesi@ burlo.trieste.it).
Purpose: Congenital hydronephrosis and isolated pyelectasis are frequently diagnosed by prenatal ultrasound. About 80% of cases resolve spontaneously in early childhood. Currently there is no agreed on protocol for prenatal followup. Most clinicians use a renal pelvis anteroposterior diameter of greater than 4 mm as a threshold for identifying isolated pyelectasis and hydronephrosis at 33 weeks of gestation or anteroposterior diameter greater than 7 mm at 40 weeks of gestation. We sought to determine a fetal renal pelvis diameter cutoff at 20 and 30 weeks of gestation that would be able to predict significant nephron uropathy requiring surgery. Materials and Methods: Our protocol included 2 prenatal ultrasounds at 20 and 30 weeks of gestation and 3 postnatal ultrasounds at ages 1, 6 and 12 months. Between January 2009 and December 2011 we evaluated 149 prenatal cases (130 males, 19 females) of isolated pyelectasis and 41 cases (28 males, 13 females) of hydronephrosis with a renal pelvis anteroposterior diameter of greater than 4 mm at 20 weeks of gestation. Results: For isolated pyelectasis we identified cutoffs of 6 mm at 20 weeks of gestation (100% sensitivity, 84.3% specificity) and 10 mm at 30 weeks of gestation (100% sensitivity, 91.9% specificity). For hydronephrosis we identified cutoffs of 10 mm at 20 weeks of gestation (100% sensitivity, 86.1% specificity) and 12 mm at 30 weeks of gestation (100% sensitivity, 66.7% specificity). Conclusions: Using these thresholds, we could avoid a significant number of followup ultrasounds in the prenatal and postnatal periods, as well as invasive postnatal tests (ie voiding cystourethrography and mercaptoacetyltriglycine scintigraphy) without missing even a single case of obstructive nephropathy requiring surgery. Key Words: hydronephrosis; prenatal diagnosis; postnatal care; pyelectasis; ultrasonography, prenatal
HYDRONEPHROSIS and isolated pyelectasis are the most common prenatal urinary tract anomalies detected by ultrasonography.16 These fetal abnormalities have been observed in 1% to 5% of pregnancies and are frequently caused by a transient or
1858
j
www.jurology.com
chronic obstruction of the urinary tract, which can be functional or anatomical.611 They are often localized at the ureteropelvic junction and less frequently at the ureterovesical junction, or caused by posterior urethral valves in males.12 Prenatal
0022-5347/13/1905-1858/0 THE JOURNAL OF UROLOGY® © 2013 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
http://dx.doi.org/10.1016/j.juro.2013.05.038 Vol. 190, 1858-1863, November 2013 Printed in U.S.A.
POSTNATAL REFERRAL FOR ISOLATED PYELECTASIS AND HYDRONEPHROSIS
hydronephrosis and isolated pyelectasis can also be caused by the presence of a nondynamic segment of ureter, or by a crossing of the lower pole vessels or vesicoureteral reflux. The terms “hydronephrosis” and “isolated pyelectasis” are too often used synonymously in the medical literature because of the lack of consensus on the definition of prenatal hydronephrosis.13 Since the 1990s, several authors have assessed the prenatal renal pelvis threshold to identify those patients whose renal function is at risk, and at the same time to avoid ineffective tests and parental anxiety in the high percentage of patients who are not at risk for renal damage.6,14 Considering that almost 80% of newborns with a prenatal diagnosis of isolated pyelectasis or mild hydronephrosis present as completely normal on ultrasound at birth, some abnormal findings are probably normal variants with minimal or uncertain clinical significance.7 The most common method of diagnosing prenatal uropathy is by measuring the anteroposterior diameter of the renal pelvis via US.1 The currently accepted standard for a clinically significant APPD is based on the original work of Corteville et al, who proposed an APPD cutoff of 4 mm at 33 weeks of GA or 7 mm at 40 weeks of GA.15 Using these cutoffs, they obtained 100% sensitivity for identification of renal pelves that would require postnatal evaluation or surgery. Other studies have described different renal pelvis cutoffs as predictors of obstructive uropathy. Adra et al in 1995 proposed an APPD of 8 mm or greater after 28 weeks of GA,14 while Mandell et al noted persistent postnatal uropathy when APPDs were larger than 6 mm before 20 weeks, larger than 8 mm at 20 to 30 weeks and 8 to 10 mm after 30 weeks of GA.16 Currently there are no international guidelines to identify kidneys that will need postnatal surgery or protocols for prenatal evaluation once isolated pyelectasis or hydronephrosis is identified.17,18 We used the following definitions for isolated pyelectasis and hydronephrosis. Isolated pyelectasis indicates isolated renal pelvis dilatation without associated calyceal dilatation. Hydronephrosis indicates the associated dilatation of the renal pelvis and the calyceal structures. We sought to determine APPD cutoffs at 20 and 30 weeks of GA that are able to predict significant obstructive uropathy requiring surgery in newborns. Our goal was to maximize the detection of significant abnormalities while minimizing unnecessary evaluations that can result in a waste of resources and cause anxiety in the parents.14,19,20
MATERIALS AND METHODS This prospective study was performed at our institute in collaboration with the pediatric nephrology and prenatal
1859
diagnosis services, and received institutional review board approval. During routine prenatal ultrasonography we detected 149 isolated pyelectasis and 41 hydronephrosis cases with an APPD of 4 mm or greater at 20 weeks of GA, using the criteria of Corteville et al,15 between January 2009 and December 2011. All mothers with a fetus with an APPD of 4 mm or larger at 20 weeks of GA were included in the study independently of age, ethnic origin or social status. Fetuses with other associated anomalies were excluded. All fetuses enrolled were otherwise healthy. To avoid bias, the same trained gynecologist, supported by a pediatric nephrologist, performed ultrasounds and diameter acquisitions. Transabdominal ultrasound was performed with 4 to 8 MHz 3-dimensional array using a VolusonÔ E8 Expert ultrasound system. Differentiation between unilateral and bilateral uropathy is not reported because in the statistical analysis we considered single renal units whose development was independent of the contralateral kidney. Our monitoring protocol included 2 prenatal US evaluations at 20 and 30 weeks of GA and 3 postnatal followup examinations consisting of US and nephrology evaluation performed at ages 1, 6 and 12 months (fig. 1). Patients with isolated pyelectasis or hydronephrosis suspected to be caused by obstructive uropathy (pelvis dilatation greater than 15 mm or poor parenchyma quality) underwent MAG3 scintigraphy. Two parameters were evaluated by MAG3 scintigraphy, ie split renal function (less than 40% differential) and half-time greater than 20 minutes.21 Both parameters were considered equally when assessing the degree of obstruction and need for surgery. In patients with unilateral isolated pyelectasis or hydronephrosis renal scan data were compared to the normal contralateral kidney. In those presenting with bilateral dilatation we considered only half-time as an indication for surgery. In these patients VUR was excluded by cystourethrography. VUR evaluation was also performed in those children who presented with a postnatal UTI, irrespective of the degree of dilatation. Renal isolated pyelectasis and hydronephrosis detected by prenatal ultrasound were classified as mild (4 to 7 mm), moderate (7.1 to 9 mm) or severe (greater than 9 mm). Postnatal outcomes were defined as resolved, improving or worsened. All 149 isolated pyelectasis and 41 hydronephrosis cases diagnosed prenatally were included in the analysis. No patient was lost to followup. Sensitivity and specificity values, ROC curves and cutoff values were calculated and identified through bivariate logistic regressions. All analyses were carried out with Stata/ICÔ, version 11.2 for WindowsÒ.
RESULTS We collected 149 isolated pyelectasis and 41 hydronephrosis cases by prenatal US with an APPD of 4 mm or greater at 20 weeks of GA. The male-tofemale ratio was 7:1 (87% males, 13% females) for isolated pyelectasis and 3:1 (68% males, 32% females) for hydronephrosis. Only 3 children presented with a single episode of UTI, of whom 2 had
1860
POSTNATAL REFERRAL FOR ISOLATED PYELECTASIS AND HYDRONEPHROSIS
Figure 1. Algorithm for prenatal and postnatal followup in patients with isolated pyelectasis and hydronephrosis. T1/2, half-time.
isolated pyelectasis and 1 had hydronephrosis. None of these patients had VUR. Isolated Pyelectasis At 20 weeks of GA mild dilatation was present in 137 cases, moderate in 7 and severe in 5. At 30 weeks of GA dilatation had worsened in 95 patients (69%) with initially mild isolated pyelectasis, remained stable in 19 (14%) and resolved or improved in 23 (17%). Of patients with initially moderate isolated pyelectasis dilatation had worsened in 5 and improved or remained stable in 2. Finally, dilatation had worsened in 4 patients with initially severe isolated pyelectasis and remained stable in 1. Postnatally only 3 of 149 kidneys (2%) required surgery for obstructive uropathy, while the remainder improved or were stable. All patients with isolated pyelectasis treated surgically were in the severe group (2 males, 1 female) (see table). For isolated pyelectasis we identified an APPD of 6 mm
or greater as the new cutoff at 20 weeks of GA (100% sensitivity, 84.25% specificity, AUC 0.8950, fig. 2, A) and an APPD of 10 mm or greater as the cutoff at 30 weeks of GA (100% sensitivity, 91.78% specificity, AUC 0.9623, fig. 2, B). Hydronephrosis At 20 weeks of GA mild dilatation was present in 21 cases, moderate in 8 and severe in 12. At 30 weeks of GA 17 patients (81%) with initially mild hydronephrosis had worsening dilatation and 4 (19%) remained stable. Meanwhile, 7 patients (87.5%) with initially moderate hydronephrosis showed worsening dilatation and 1 (12.5%) remained stable. Finally, all 12 patients with initially severe hydronephrosis demonstrated worsening dilatation. Postnatally 5 of 41 patients with hydronephrosis required surgery for obstructive uropathy (see table). All patients with surgically treated hydronephrosis were in the severe group and were male.
Imaging findings in kidneys treated surgically Ultrasonography
MAG3 Scintigraphy
Diagnosis
Pt No.dSex
Dilatation (mm) at 20 Wks GA
Dilatation (mm) at 30 Wks GA
Postnatal Dilatation (mm)
Uptake (%)
Half-Time (mins)
Pyelectasis Pyelectasis Pyelectasis Hydronephrosis Hydronephrosis Hydronephrosis Hydronephrosis Hydronephrosis
1dmale 2dmale 3dfemale 1dmale 2dmale 3dmale 4dmale 5dmale
6.5 8 7 18 13.9 15 11 10.2
10 15 17 15 33 21 18 12
30 25 27 26 50 30 26 22
48 46 49 38 45 49 49 46
40 38 48 40 60 46 41 44
POSTNATAL REFERRAL FOR ISOLATED PYELECTASIS AND HYDRONEPHROSIS
B
0.75 0.00
0.00
0.25
0.50
Sensitivity
0.50 0.25
Sensitivity
0.75
1.00
1.00
A
1861
0.00
0.25
0.50 1 - Specificity
0.75
1.00
0.00
Area under ROC curve = 0.8950
0.25
0.50 1 - Specificity
0.75
1.00
Area under ROC curve = 0.9623
Figure 2. ROC curve for determining isolated pyelectasis cutoff at 20 weeks (A) and 30 weeks (B) of gestation
For hydronephrosis we identified an APPD of 10 mm or greater as the new cutoff at 20 weeks of GA (100% sensitivity, 86.11% specificity, AUC 0.9472, fig. 3, A) and an APPD of 12 mm or greater as the cutoff at 30 weeks of GA (100% sensitivity, 66.67% specificity, AUC 0.9306, fig. 3, B).
DISCUSSION This study confirms the data reported in the international literature that prenatal urinary tract anomalies are more common in males,22e24 with a male-to-female ratio of 7:1 for isolated pyelectasis and 3:1 for hydronephrosis. Moreover, we found that males have worse outcomes than females. All surgically treated hydronephrotic kidneys and two thirds of those surgically treated for isolated pyelectasis were observed in males.
VUR as the cause of isolated pyelectasis or hydronephrosis was excluded in the most severe cases requiring surgery and in children presenting with UTI. In fact, there is growing evidence regarding the low sensitivity and poor predictive value of fetal renal pelvis size in predicting the presence and severity of VUR.25,26 In the study by Hothi et al the rate of VUR was low, at approximately 4% to 6% in the mild group and up to 18% in the moderate third trimester group.7 Moreover, the recent American Academy of Pediatrics and National Institute for Health and Care Excellence guidelines as well as the Italian recommendations excluded the need for any intervention—antimicrobial prophylaxis or surgery—for VUR less than grade IV, concluding that in mild and moderate isolated pyelectasis and hydronephrosis the research on VUR becomes clinically irrelevant.27e29
B
0.50
Sensitivity
0.25
0.50
0.00
0.25 0.00
Sensitivity
0.75
0.75
1.00
1.00
A
0.00
0.25
Area under ROC curve = 0.9472
0.50 1 - Specificity
0.75
1.00
0.00
0.25
0.50 1 - Specificity
0.75
Area under ROC curve = 0.9306
Figure 3. ROC curve for determining hydronephrosis cutoff at 20 weeks A and 30 weeks B of gestation
1.00
1862
POSTNATAL REFERRAL FOR ISOLATED PYELECTASIS AND HYDRONEPHROSIS
Excluding hydroureteronephrosis, it is likely that most of the undiagnosed VUR cases in our study were lower or medium grades, and children presenting with them are not expected to have a higher ratio of UTI. To define the APPD cutoffs at 20 and 30 weeks of GA able to predict significant obstructive uropathy, we focused particular attention on the sensitivity of the methods rather than the specificity because we believe it is mandatory not to miss even a single case that would require surgery postnatally. According to our data, it is possible to use an APPD cutoff of 6 mm at 20 weeks of GA and 10 mm at 30 weeks without losing any case that would need corrective surgery. This new 20-week GA threshold is appreciably higher than the cutoffs currently recommended in the international literature, which range from 4 to 5 mm.15,16 Our new cutoff appeared to perform quite well, with 100% sensitivity and 84.25% specificity (AUC 0.8950, fig. 2, A). Similar results were obtained with a cutoff of 10 mm at 30 weeks of GA, which yielded 100% sensitivity and 91.78% specificity (AUC 0.9623, fig. 2, B), compared to current recommendations of 6 to 7 mm.15,16 Practically speaking, by using these new cutoffs, we could save about 89% of the total control ultrasounds performed in the prenatal and postnatal periods, resulting in a gain not only in terms of resources, but also in terms of sparing parental anxiety caused by the discovery of a potential fetal abnormality. The same consideration is true for hydronephrosis. According to our data, an APPD cutoff of 10 mm at 20 weeks of GA would have 100%
sensitivity, 86.11% specificity and an AUC of 0.9472, while a cutoff of 12 mm at 30 weeks of GA would have 100% sensitivity, 66.67% specificity and an AUC of 0.9306 (fig. 3). Although this group of patients was numerically smaller, we could spare 58% of the total control ultrasounds performed in the prenatal and postnatal periods using this new cutoff, even if the 30-week GA cutoff specificity is not higher like the others we proposed. Moreover, in our study all patients who underwent surgery had exhibited a clear worsening of pelvis dilatation during prenatal and postnatal followup (see table), revealing that ultrasound data seem to be useful in predicting surgical outcomes.
CONCLUSIONS The critical point in the management of prenatally detected isolated pyelectasis and hydronephrosis is to differentiate clinically significant abnormalities from those that may resolve without treatment. Our proposed new cutoffs might represent a good predictive model, allowing valid prenatal discernment between kidneys that will require postnatal surgery and those that will not need further followup. Using these new cutoffs, we can significantly decrease parental anxiety, minimize the burden of unnecessary prenatal and postnatal investigations in healthy infants, and avoid unnecessary costs. The adoption of our proposed cutoffs in a larger multicenter study could confirm our findings and might prove that more is not always better.
REFERENCES 1. Yamac‚ake KG and Nguyen HT: Current management of antenatal hydronephrosis. Pediatr Nephrol 2013; 28: 237.
7. Hothi DK, Wade AS, Gilbert R et al: Mild fetal renal pelvis dilatation: much ado about nothing? Clin J Am Soc Nephrol 2009; 4: 168.
12. Babu R and Sai V: Postnatal outcome of fetal hydronephrosis: implications for prenatal counselling. Indian J Urol 2010; 26: 60.
2. Estrada CR Jr: Prenatal hydronephrosis: early evaluation. Curr Opin Urol 2008; 18: 401.
8. Kitchens DM and Herndon CD: Prenatal intervention for lower urinary tract obstruction. ScientificWorldJournal 2009; 9: 390.
13. Scott JE and Renwick M: Screening for fetal urological abnormalities: how effective? BJU Int 1999; 84: 693.
9. Kent A, Cox D, Downey P et al: A study of mild fetal pyelectasia—outcome and proposed strategy of management. Prenat Diagn 2000; 20: 206.
14. Adra AM, Mejides AA, Dennaoui MS et al: Fetal pyelectasis: is it always “physiologic”? Am J Obstet Gynecol 1995; 173: 1263.
3. Mouriquand PD, Whitten M and Pracros JP: Pathophysiology, diagnosis and management of prenatal upper tract dilatation. Prenat Diagn 2001; 21: 942. 4. Fefer S and Ellsworth P: Prenatal hydronephrosis. Pediatr Clin North Am 2006; 53: 429. 5. Mandell J, Blyth BR, Peters CA et al: Structural genitourinary defects detected in utero. Radiology 1991; 178: 193.
10. Langer B, Simeoni U, Montoya Y et al: Antenatal diagnosis of upper urinary tract dilation by ultrasonography. Fetal Diagn Ther 1996; 11: 191.
15. Corteville JE, Gray DL and Crane JP: Congenital hydronephrosis: correlation of fetal ultrasonographic findings with infant outcome. Am J Obstet Gynecol 1991; 165: 384.
6. Lee RS, Cendron M, Kinnamon DD et al: Antenatal hydronephrosis as a predictor of postnatal outcome: a meta-analysis. Pediatrics 2006; 118: 586.
11. Passerotti CC, Kalish LA, Chow J et al: The predictive value of the first postnatal ultrasound in children with antenatal hydronephrosis. J Pediatr Urol 2011; 7: 128.
16. Mandell J, Peters CA and Retik AB: Perinatal urology. In: Campbell’s Urology, 7th ed. Edited by PC Walsh, AB Retik, ED Vaughan et al. Philadelphia: WB Saunders Co 1998; pp 1601e1618.
POSTNATAL REFERRAL FOR ISOLATED PYELECTASIS AND HYDRONEPHROSIS
17. Sidhu G, Beyene J and Rosenblum ND: Outcome of isolated antenatal hydronephrosis: a systematic review and meta-analysis. Pediatr Nephrol 2006; 21: 218. 18. Damen-Elias HA, Luijnenburg SE, Visser GH et al: Mild pyelectasis diagnosed by prenatal ultrasound is not a predictor of urinary tract morbidity in childhood. Prenat Diagn 2005; 25: 1239. 19. Ismaili K, Avni FE, Piepsz A et al: Current management of infants with fetal renal pelvis dilation: a survey by French-speaking pediatric nephrologists and urologists. Pediatr Nephrol 2004; 19: 966. 20. Harding LJ, Malone PS and Wellesley DG: Antenatal minimal hydronephrosis: is its followup an unnecessary cause of concern? Prenat Diagn 1999; 19: 701. 21. Conway JJ and Maizels M: The “well tempered” diuretic renogram: a standard method to
examine the asymptomatic neonate with hydronephrosis or hydroureteronephrosis. A report from combined meetings of the Society for Fetal Urology and members of the Pediatric Nuclear Medicine Councilethe Society of Nuclear Medicine. J Nucl Med 1992; 33: 2047. 22. Ardissino G, Dacco V, Testa S et al: Epidemiology of chronic renal failure in children: data from the ItalKid project. Pediatrics 2003; 111: e382. 23. Scott JE and Renwick M: Urological anomalies in the Northern Region Fetal Abnormality Survey. Arch Dis Child 1993; 68: 22. 24. Leung VY, Chu WC and Metreweli C: Prenatal ultrasound study of gender difference in urinary tract morphology in normal fetuses. HK J Paediatr 2008; 13: 165. 25. Anderson NG, Wright S, Abbott GD et al: Fetal renal pelvic dilatation—poor predictor of familial vesicoureteral reflux. Pediatr Nephrol 2003; 18: 902.
1863
26. Phan V, Traubici J, Hershenfield B et al: Vesicoureteral reflux in infants with isolated antenatal hydronephrosis. Pediatr Nephrol 2003; 18: 1224. 27. Roberts KB: Urinary tract infection: clinical practice guideline for the diagnosis and management of the initial UTI in febrile infants and children 2 to 24 months. Pediatrics 2011; 128: 595. 28. National Institute for Health and Care Excellence (NICE) Guideline. Urinary tract infection: diagnosis, treatment and long-term management of urinary tract infection in children. Available at http://www.nice.org.uk/cg54. Accessed March 7, 2013. 29. Ammenti A, Cataldi L, Chimenz R et al: Febrile urinary tract infections in young children: recommendations for the diagnosis, treatment and follow-up. Acta Paediatr 2012; 101: 451.
Abbreviations and Acronyms APPD ¼ anteroposterior pelvic diameter GA ¼ gestational age MAG3 ¼ mercaptoacetyltriglycine US ¼ ultrasound UTI ¼ urinary tract infection VUR ¼ vesicoureteral reflux Accepted for publication May 13, 2013. Study received institutional review board approval. * Correspondence: Department of Pediatrics, Institute for Maternal and Child Healthe IRCCS “Burlo Garofolo,” Via dell’Istria 65/1, 34137 Trieste, Italy (telephone: 39-040-3785263; FAX: 39-040-3785290; e-mail: marco.pennesi@ burlo.trieste.it).
Purpose: Congenital hydronephrosis and isolated pyelectasis are frequently diagnosed by prenatal ultrasound. About 80% of cases resolve spontaneously in early childhood. Currently there is no agreed on protocol for prenatal followup. Most clinicians use a renal pelvis anteroposterior diameter of greater than 4 mm as a threshold for identifying isolated pyelectasis and hydronephrosis at 33 weeks of gestation or anteroposterior diameter greater than 7 mm at 40 weeks of gestation. We sought to determine a fetal renal pelvis diameter cutoff at 20 and 30 weeks of gestation that would be able to predict significant nephron uropathy requiring surgery. Materials and Methods: Our protocol included 2 prenatal ultrasounds at 20 and 30 weeks of gestation and 3 postnatal ultrasounds at ages 1, 6 and 12 months. Between January 2009 and December 2011 we evaluated 149 prenatal cases (130 males, 19 females) of isolated pyelectasis and 41 cases (28 males, 13 females) of hydronephrosis with a renal pelvis anteroposterior diameter of greater than 4 mm at 20 weeks of gestation. Results: For isolated pyelectasis we identified cutoffs of 6 mm at 20 weeks of gestation (100% sensitivity, 84.3% specificity) and 10 mm at 30 weeks of gestation (100% sensitivity, 91.9% specificity). For hydronephrosis we identified cutoffs of 10 mm at 20 weeks of gestation (100% sensitivity, 86.1% specificity) and 12 mm at 30 weeks of gestation (100% sensitivity, 66.7% specificity). Conclusions: Using these thresholds, we could avoid a significant number of followup ultrasounds in the prenatal and postnatal periods, as well as invasive postnatal tests (ie voiding cystourethrography and mercaptoacetyltriglycine scintigraphy) without missing even a single case of obstructive nephropathy requiring surgery. Key Words: hydronephrosis; prenatal diagnosis; postnatal care; pyelectasis; ultrasonography, prenatal
HYDRONEPHROSIS and isolated pyelectasis are the most common prenatal urinary tract anomalies detected by ultrasonography.16 These fetal abnormalities have been observed in 1% to 5% of pregnancies and are frequently caused by a transient or
1858
j
www.jurology.com
chronic obstruction of the urinary tract, which can be functional or anatomical.611 They are often localized at the ureteropelvic junction and less frequently at the ureterovesical junction, or caused by posterior urethral valves in males.12 Prenatal
0022-5347/13/1905-1858/0 THE JOURNAL OF UROLOGY® © 2013 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
http://dx.doi.org/10.1016/j.juro.2013.05.038 Vol. 190, 1858-1863, November 2013 Printed in U.S.A.
POSTNATAL REFERRAL FOR ISOLATED PYELECTASIS AND HYDRONEPHROSIS
hydronephrosis and isolated pyelectasis can also be caused by the presence of a nondynamic segment of ureter, or by a crossing of the lower pole vessels or vesicoureteral reflux. The terms “hydronephrosis” and “isolated pyelectasis” are too often used synonymously in the medical literature because of the lack of consensus on the definition of prenatal hydronephrosis.13 Since the 1990s, several authors have assessed the prenatal renal pelvis threshold to identify those patients whose renal function is at risk, and at the same time to avoid ineffective tests and parental anxiety in the high percentage of patients who are not at risk for renal damage.6,14 Considering that almost 80% of newborns with a prenatal diagnosis of isolated pyelectasis or mild hydronephrosis present as completely normal on ultrasound at birth, some abnormal findings are probably normal variants with minimal or uncertain clinical significance.7 The most common method of diagnosing prenatal uropathy is by measuring the anteroposterior diameter of the renal pelvis via US.1 The currently accepted standard for a clinically significant APPD is based on the original work of Corteville et al, who proposed an APPD cutoff of 4 mm at 33 weeks of GA or 7 mm at 40 weeks of GA.15 Using these cutoffs, they obtained 100% sensitivity for identification of renal pelves that would require postnatal evaluation or surgery. Other studies have described different renal pelvis cutoffs as predictors of obstructive uropathy. Adra et al in 1995 proposed an APPD of 8 mm or greater after 28 weeks of GA,14 while Mandell et al noted persistent postnatal uropathy when APPDs were larger than 6 mm before 20 weeks, larger than 8 mm at 20 to 30 weeks and 8 to 10 mm after 30 weeks of GA.16 Currently there are no international guidelines to identify kidneys that will need postnatal surgery or protocols for prenatal evaluation once isolated pyelectasis or hydronephrosis is identified.17,18 We used the following definitions for isolated pyelectasis and hydronephrosis. Isolated pyelectasis indicates isolated renal pelvis dilatation without associated calyceal dilatation. Hydronephrosis indicates the associated dilatation of the renal pelvis and the calyceal structures. We sought to determine APPD cutoffs at 20 and 30 weeks of GA that are able to predict significant obstructive uropathy requiring surgery in newborns. Our goal was to maximize the detection of significant abnormalities while minimizing unnecessary evaluations that can result in a waste of resources and cause anxiety in the parents.14,19,20
MATERIALS AND METHODS This prospective study was performed at our institute in collaboration with the pediatric nephrology and prenatal
1859
diagnosis services, and received institutional review board approval. During routine prenatal ultrasonography we detected 149 isolated pyelectasis and 41 hydronephrosis cases with an APPD of 4 mm or greater at 20 weeks of GA, using the criteria of Corteville et al,15 between January 2009 and December 2011. All mothers with a fetus with an APPD of 4 mm or larger at 20 weeks of GA were included in the study independently of age, ethnic origin or social status. Fetuses with other associated anomalies were excluded. All fetuses enrolled were otherwise healthy. To avoid bias, the same trained gynecologist, supported by a pediatric nephrologist, performed ultrasounds and diameter acquisitions. Transabdominal ultrasound was performed with 4 to 8 MHz 3-dimensional array using a VolusonÔ E8 Expert ultrasound system. Differentiation between unilateral and bilateral uropathy is not reported because in the statistical analysis we considered single renal units whose development was independent of the contralateral kidney. Our monitoring protocol included 2 prenatal US evaluations at 20 and 30 weeks of GA and 3 postnatal followup examinations consisting of US and nephrology evaluation performed at ages 1, 6 and 12 months (fig. 1). Patients with isolated pyelectasis or hydronephrosis suspected to be caused by obstructive uropathy (pelvis dilatation greater than 15 mm or poor parenchyma quality) underwent MAG3 scintigraphy. Two parameters were evaluated by MAG3 scintigraphy, ie split renal function (less than 40% differential) and half-time greater than 20 minutes.21 Both parameters were considered equally when assessing the degree of obstruction and need for surgery. In patients with unilateral isolated pyelectasis or hydronephrosis renal scan data were compared to the normal contralateral kidney. In those presenting with bilateral dilatation we considered only half-time as an indication for surgery. In these patients VUR was excluded by cystourethrography. VUR evaluation was also performed in those children who presented with a postnatal UTI, irrespective of the degree of dilatation. Renal isolated pyelectasis and hydronephrosis detected by prenatal ultrasound were classified as mild (4 to 7 mm), moderate (7.1 to 9 mm) or severe (greater than 9 mm). Postnatal outcomes were defined as resolved, improving or worsened. All 149 isolated pyelectasis and 41 hydronephrosis cases diagnosed prenatally were included in the analysis. No patient was lost to followup. Sensitivity and specificity values, ROC curves and cutoff values were calculated and identified through bivariate logistic regressions. All analyses were carried out with Stata/ICÔ, version 11.2 for WindowsÒ.
RESULTS We collected 149 isolated pyelectasis and 41 hydronephrosis cases by prenatal US with an APPD of 4 mm or greater at 20 weeks of GA. The male-tofemale ratio was 7:1 (87% males, 13% females) for isolated pyelectasis and 3:1 (68% males, 32% females) for hydronephrosis. Only 3 children presented with a single episode of UTI, of whom 2 had
1860
POSTNATAL REFERRAL FOR ISOLATED PYELECTASIS AND HYDRONEPHROSIS
Figure 1. Algorithm for prenatal and postnatal followup in patients with isolated pyelectasis and hydronephrosis. T1/2, half-time.
isolated pyelectasis and 1 had hydronephrosis. None of these patients had VUR. Isolated Pyelectasis At 20 weeks of GA mild dilatation was present in 137 cases, moderate in 7 and severe in 5. At 30 weeks of GA dilatation had worsened in 95 patients (69%) with initially mild isolated pyelectasis, remained stable in 19 (14%) and resolved or improved in 23 (17%). Of patients with initially moderate isolated pyelectasis dilatation had worsened in 5 and improved or remained stable in 2. Finally, dilatation had worsened in 4 patients with initially severe isolated pyelectasis and remained stable in 1. Postnatally only 3 of 149 kidneys (2%) required surgery for obstructive uropathy, while the remainder improved or were stable. All patients with isolated pyelectasis treated surgically were in the severe group (2 males, 1 female) (see table). For isolated pyelectasis we identified an APPD of 6 mm
or greater as the new cutoff at 20 weeks of GA (100% sensitivity, 84.25% specificity, AUC 0.8950, fig. 2, A) and an APPD of 10 mm or greater as the cutoff at 30 weeks of GA (100% sensitivity, 91.78% specificity, AUC 0.9623, fig. 2, B). Hydronephrosis At 20 weeks of GA mild dilatation was present in 21 cases, moderate in 8 and severe in 12. At 30 weeks of GA 17 patients (81%) with initially mild hydronephrosis had worsening dilatation and 4 (19%) remained stable. Meanwhile, 7 patients (87.5%) with initially moderate hydronephrosis showed worsening dilatation and 1 (12.5%) remained stable. Finally, all 12 patients with initially severe hydronephrosis demonstrated worsening dilatation. Postnatally 5 of 41 patients with hydronephrosis required surgery for obstructive uropathy (see table). All patients with surgically treated hydronephrosis were in the severe group and were male.
Imaging findings in kidneys treated surgically Ultrasonography
MAG3 Scintigraphy
Diagnosis
Pt No.dSex
Dilatation (mm) at 20 Wks GA
Dilatation (mm) at 30 Wks GA
Postnatal Dilatation (mm)
Uptake (%)
Half-Time (mins)
Pyelectasis Pyelectasis Pyelectasis Hydronephrosis Hydronephrosis Hydronephrosis Hydronephrosis Hydronephrosis
1dmale 2dmale 3dfemale 1dmale 2dmale 3dmale 4dmale 5dmale
6.5 8 7 18 13.9 15 11 10.2
10 15 17 15 33 21 18 12
30 25 27 26 50 30 26 22
48 46 49 38 45 49 49 46
40 38 48 40 60 46 41 44
POSTNATAL REFERRAL FOR ISOLATED PYELECTASIS AND HYDRONEPHROSIS
B
0.75 0.00
0.00
0.25
0.50
Sensitivity
0.50 0.25
Sensitivity
0.75
1.00
1.00
A
1861
0.00
0.25
0.50 1 - Specificity
0.75
1.00
0.00
Area under ROC curve = 0.8950
0.25
0.50 1 - Specificity
0.75
1.00
Area under ROC curve = 0.9623
Figure 2. ROC curve for determining isolated pyelectasis cutoff at 20 weeks (A) and 30 weeks (B) of gestation
For hydronephrosis we identified an APPD of 10 mm or greater as the new cutoff at 20 weeks of GA (100% sensitivity, 86.11% specificity, AUC 0.9472, fig. 3, A) and an APPD of 12 mm or greater as the cutoff at 30 weeks of GA (100% sensitivity, 66.67% specificity, AUC 0.9306, fig. 3, B).
DISCUSSION This study confirms the data reported in the international literature that prenatal urinary tract anomalies are more common in males,22e24 with a male-to-female ratio of 7:1 for isolated pyelectasis and 3:1 for hydronephrosis. Moreover, we found that males have worse outcomes than females. All surgically treated hydronephrotic kidneys and two thirds of those surgically treated for isolated pyelectasis were observed in males.
VUR as the cause of isolated pyelectasis or hydronephrosis was excluded in the most severe cases requiring surgery and in children presenting with UTI. In fact, there is growing evidence regarding the low sensitivity and poor predictive value of fetal renal pelvis size in predicting the presence and severity of VUR.25,26 In the study by Hothi et al the rate of VUR was low, at approximately 4% to 6% in the mild group and up to 18% in the moderate third trimester group.7 Moreover, the recent American Academy of Pediatrics and National Institute for Health and Care Excellence guidelines as well as the Italian recommendations excluded the need for any intervention—antimicrobial prophylaxis or surgery—for VUR less than grade IV, concluding that in mild and moderate isolated pyelectasis and hydronephrosis the research on VUR becomes clinically irrelevant.27e29
B
0.50
Sensitivity
0.25
0.50
0.00
0.25 0.00
Sensitivity
0.75
0.75
1.00
1.00
A
0.00
0.25
Area under ROC curve = 0.9472
0.50 1 - Specificity
0.75
1.00
0.00
0.25
0.50 1 - Specificity
0.75
Area under ROC curve = 0.9306
Figure 3. ROC curve for determining hydronephrosis cutoff at 20 weeks A and 30 weeks B of gestation
1.00
1862
POSTNATAL REFERRAL FOR ISOLATED PYELECTASIS AND HYDRONEPHROSIS
Excluding hydroureteronephrosis, it is likely that most of the undiagnosed VUR cases in our study were lower or medium grades, and children presenting with them are not expected to have a higher ratio of UTI. To define the APPD cutoffs at 20 and 30 weeks of GA able to predict significant obstructive uropathy, we focused particular attention on the sensitivity of the methods rather than the specificity because we believe it is mandatory not to miss even a single case that would require surgery postnatally. According to our data, it is possible to use an APPD cutoff of 6 mm at 20 weeks of GA and 10 mm at 30 weeks without losing any case that would need corrective surgery. This new 20-week GA threshold is appreciably higher than the cutoffs currently recommended in the international literature, which range from 4 to 5 mm.15,16 Our new cutoff appeared to perform quite well, with 100% sensitivity and 84.25% specificity (AUC 0.8950, fig. 2, A). Similar results were obtained with a cutoff of 10 mm at 30 weeks of GA, which yielded 100% sensitivity and 91.78% specificity (AUC 0.9623, fig. 2, B), compared to current recommendations of 6 to 7 mm.15,16 Practically speaking, by using these new cutoffs, we could save about 89% of the total control ultrasounds performed in the prenatal and postnatal periods, resulting in a gain not only in terms of resources, but also in terms of sparing parental anxiety caused by the discovery of a potential fetal abnormality. The same consideration is true for hydronephrosis. According to our data, an APPD cutoff of 10 mm at 20 weeks of GA would have 100%
sensitivity, 86.11% specificity and an AUC of 0.9472, while a cutoff of 12 mm at 30 weeks of GA would have 100% sensitivity, 66.67% specificity and an AUC of 0.9306 (fig. 3). Although this group of patients was numerically smaller, we could spare 58% of the total control ultrasounds performed in the prenatal and postnatal periods using this new cutoff, even if the 30-week GA cutoff specificity is not higher like the others we proposed. Moreover, in our study all patients who underwent surgery had exhibited a clear worsening of pelvis dilatation during prenatal and postnatal followup (see table), revealing that ultrasound data seem to be useful in predicting surgical outcomes.
CONCLUSIONS The critical point in the management of prenatally detected isolated pyelectasis and hydronephrosis is to differentiate clinically significant abnormalities from those that may resolve without treatment. Our proposed new cutoffs might represent a good predictive model, allowing valid prenatal discernment between kidneys that will require postnatal surgery and those that will not need further followup. Using these new cutoffs, we can significantly decrease parental anxiety, minimize the burden of unnecessary prenatal and postnatal investigations in healthy infants, and avoid unnecessary costs. The adoption of our proposed cutoffs in a larger multicenter study could confirm our findings and might prove that more is not always better.
REFERENCES 1. Yamac‚ake KG and Nguyen HT: Current management of antenatal hydronephrosis. Pediatr Nephrol 2013; 28: 237.
7. Hothi DK, Wade AS, Gilbert R et al: Mild fetal renal pelvis dilatation: much ado about nothing? Clin J Am Soc Nephrol 2009; 4: 168.
12. Babu R and Sai V: Postnatal outcome of fetal hydronephrosis: implications for prenatal counselling. Indian J Urol 2010; 26: 60.
2. Estrada CR Jr: Prenatal hydronephrosis: early evaluation. Curr Opin Urol 2008; 18: 401.
8. Kitchens DM and Herndon CD: Prenatal intervention for lower urinary tract obstruction. ScientificWorldJournal 2009; 9: 390.
13. Scott JE and Renwick M: Screening for fetal urological abnormalities: how effective? BJU Int 1999; 84: 693.
9. Kent A, Cox D, Downey P et al: A study of mild fetal pyelectasia—outcome and proposed strategy of management. Prenat Diagn 2000; 20: 206.
14. Adra AM, Mejides AA, Dennaoui MS et al: Fetal pyelectasis: is it always “physiologic”? Am J Obstet Gynecol 1995; 173: 1263.
3. Mouriquand PD, Whitten M and Pracros JP: Pathophysiology, diagnosis and management of prenatal upper tract dilatation. Prenat Diagn 2001; 21: 942. 4. Fefer S and Ellsworth P: Prenatal hydronephrosis. Pediatr Clin North Am 2006; 53: 429. 5. Mandell J, Blyth BR, Peters CA et al: Structural genitourinary defects detected in utero. Radiology 1991; 178: 193.
10. Langer B, Simeoni U, Montoya Y et al: Antenatal diagnosis of upper urinary tract dilation by ultrasonography. Fetal Diagn Ther 1996; 11: 191.
15. Corteville JE, Gray DL and Crane JP: Congenital hydronephrosis: correlation of fetal ultrasonographic findings with infant outcome. Am J Obstet Gynecol 1991; 165: 384.
6. Lee RS, Cendron M, Kinnamon DD et al: Antenatal hydronephrosis as a predictor of postnatal outcome: a meta-analysis. Pediatrics 2006; 118: 586.
11. Passerotti CC, Kalish LA, Chow J et al: The predictive value of the first postnatal ultrasound in children with antenatal hydronephrosis. J Pediatr Urol 2011; 7: 128.
16. Mandell J, Peters CA and Retik AB: Perinatal urology. In: Campbell’s Urology, 7th ed. Edited by PC Walsh, AB Retik, ED Vaughan et al. Philadelphia: WB Saunders Co 1998; pp 1601e1618.
POSTNATAL REFERRAL FOR ISOLATED PYELECTASIS AND HYDRONEPHROSIS
17. Sidhu G, Beyene J and Rosenblum ND: Outcome of isolated antenatal hydronephrosis: a systematic review and meta-analysis. Pediatr Nephrol 2006; 21: 218. 18. Damen-Elias HA, Luijnenburg SE, Visser GH et al: Mild pyelectasis diagnosed by prenatal ultrasound is not a predictor of urinary tract morbidity in childhood. Prenat Diagn 2005; 25: 1239. 19. Ismaili K, Avni FE, Piepsz A et al: Current management of infants with fetal renal pelvis dilation: a survey by French-speaking pediatric nephrologists and urologists. Pediatr Nephrol 2004; 19: 966. 20. Harding LJ, Malone PS and Wellesley DG: Antenatal minimal hydronephrosis: is its followup an unnecessary cause of concern? Prenat Diagn 1999; 19: 701. 21. Conway JJ and Maizels M: The “well tempered” diuretic renogram: a standard method to
examine the asymptomatic neonate with hydronephrosis or hydroureteronephrosis. A report from combined meetings of the Society for Fetal Urology and members of the Pediatric Nuclear Medicine Councilethe Society of Nuclear Medicine. J Nucl Med 1992; 33: 2047. 22. Ardissino G, Dacco V, Testa S et al: Epidemiology of chronic renal failure in children: data from the ItalKid project. Pediatrics 2003; 111: e382. 23. Scott JE and Renwick M: Urological anomalies in the Northern Region Fetal Abnormality Survey. Arch Dis Child 1993; 68: 22. 24. Leung VY, Chu WC and Metreweli C: Prenatal ultrasound study of gender difference in urinary tract morphology in normal fetuses. HK J Paediatr 2008; 13: 165. 25. Anderson NG, Wright S, Abbott GD et al: Fetal renal pelvic dilatation—poor predictor of familial vesicoureteral reflux. Pediatr Nephrol 2003; 18: 902.
1863
26. Phan V, Traubici J, Hershenfield B et al: Vesicoureteral reflux in infants with isolated antenatal hydronephrosis. Pediatr Nephrol 2003; 18: 1224. 27. Roberts KB: Urinary tract infection: clinical practice guideline for the diagnosis and management of the initial UTI in febrile infants and children 2 to 24 months. Pediatrics 2011; 128: 595. 28. National Institute for Health and Care Excellence (NICE) Guideline. Urinary tract infection: diagnosis, treatment and long-term management of urinary tract infection in children. Available at http://www.nice.org.uk/cg54. Accessed March 7, 2013. 29. Ammenti A, Cataldi L, Chimenz R et al: Febrile urinary tract infections in young children: recommendations for the diagnosis, treatment and follow-up. Acta Paediatr 2012; 101: 451.