Evaluation and Follow-up of Fetal Hydronephrosis Deborah M. Feldman, MD, Marvalyn DeCambre, MD, Erin Kong, Adam Borgida, MD, Mujgan Jamil, MBBS, Patrick McKenna, MD, James F. X. Egan, MD Objective. To determine the antenatal course and neonatal follow-up of isolated fetal hydronephrosis. Methods. We reviewed our ultrasonography database from January 1989 to June 1999 for all cases of unilateral or bilateral fetal hydronephrosis that had at least 1 follow-up ultrasonographic examination. Cases were defined as mild, moderate, or severe depending on the renal pelvis anteroposterior diameter and gestational age. Data were analyzed using the χ 2 test with the Fisher exact test where appropriate. Medical records were reviewed, and telephone interviews were performed to determine which infants received follow-up after birth. Results. Of 57,966 ultrasonographic examinations in 20,049 women during the study period, 393 patients met criteria for evaluation. Of these, 347 (88%) had fetuses with mild hydronephrosis. Most of these had complete resolution during the pregnancy. Forty patients had fetuses classified as having moderate hydronephrosis, and 6 patients had fetuses with severe hydronephrosis. Of those classified as moderate hydronephrosis, 15% resolved, 25% improved, 48% remained unchanged, and 12% worsened during the pregnancy. There were no cases of in utero resolution in the severe group; however, 4 of 6 cases improved to moderate or mild, and 2 cases remained unchanged. Of the cases identified prenatally, 25 received consultation by a pediatric urologist in the newborn period, and 7 of these required surgical intervention. Conclusions. Our population-based data suggest that most cases of mild hydronephrosis will resolve before delivery. In contrast, cases of moderate or severe hydronephrosis are less likely to have resolution in utero and are more likely to worsen or remain unchanged. Of those fetuses with persistent hydronephrosis, only a small number required some surgical intervention after birth. This information is useful in counseling the patient whose fetus is noted to have isolated hydronephrosis. Key words: fetal hydronephrosis; prenatal ultrasonography; pregnancy. Abbreviations AP, anteroposterior Received October 30, 2000, from the Division of Maternal-Fetal Medicine, University of Connecticut Health Center, Farmington, Connecticut (D.M.F., A.B., J.F.X.E.); Saint Francis Hospital and Medical Center (E.K., A.B., M.J., J.F.X.E.) and Division of Urology, Connecticut Children s Medical Center (M.D.), Hartford, Connecticut; and Southern Illinois University, Springfield, Illinois (P.M.). Revision requested November 27, 2000. Revised manuscript accepted for publication May 21, 2001. Address correspondence and reprint requests to Deborah M. Feldman, MD, Division of Maternal- Fetal Medicine, Department of Obstetrics and Gynecology, University of Connecticut Health Center, MC-2950, Farmington, CT 06030-2950. Fetal urinary tract abnormalities are among the most commonly diagnosed prenatal malformations, with a reported prevalence as high as 1 per 100 pregnancies. 1 The routine use of prenatal ultrasonography has led to an increase in the number of fetuses identified with urologic anomalies. The benefit is that these infants may be followed in the neonatal period for definitive diagnosis and treatment if necessary. Several studies have been published that attempt to establish a correlation between fetal hydronephrosis and neonatal outcome. 2 7 Limitations of many of these reports include inconsistent definitions of hydronephrosis as well as follow-up information of only those neonates who required surgery after delivery. In 1990, Mandell 8 published a classification system for the prenatal diagnosis of fetal hydronephrosis based on renal pelvis anteroposterior (AP) diameter and gestation- 2001 by the American Institute of Ultrasound in Medicine J Ultrasound Med 20:1065 1069, 2001 0278-4297/01/$3.50
Evaluation and Follow-up of Fetal Hydronephrosis al age. At 15 to 20 weeks gestation, those fetuses with renal AP diameters of 4 to 7 mm are considered to have mild hydronephrosis, and those with AP diameters greater than 7 mm are classified as having moderate hydronephrosis. At more than 20 to 30 weeks gestation (Fig. 1), renal AP diameter may have mild (5 8 mm), moderate (9 15 mm), or severe (>15 mm) dilatation. Finally, at a gestational age of more than 30 weeks to term, the values increase slightly to 7 to 9 mm for mild, 10 to 15 mm for moderate, and 16 mm or greater for severe. Since the initial presentation of these data, patients evaluated at Saint Francis Hospital and Medical Center who have a diagnosis of fetal hydronephrosis based on prenatal ultrasonography are classified as mild, moderate, or severe according to the criteria of Mandell. 8 Because of the lack of consensus on the treatment during pregnancy of patients with fetal hydronephrosis, our aim in this study was to determine the antenatal course of those patients whose fetuses had a diagnosis of isolated hydronephrosis on the basis of the criteria of Mandell. 8 Specifically, we sought to determine the natural course of mild fetal hydronephrosis with the future goal of developing a protocol for follow-up of these patients during the pregnancy. In addition, we sought to evaluate the number of neonates who received any type of follow-up after birth. Figure 1. Unilateral mild hydronephrosis at 30 weeks gestation. Arrow indicates left renal pelvis. Materials and Methods We performed a retrospective review of our computerized ultrasonography database at Saint Francis Hospital and Medical Center, a tertiarylevel community hospital performing approximately 3300 deliveries annually. The study period was between January 1989 and June 1999. The database was initially searched for all renal anomalies. Patients were then included in the study if they had documented unilateral or bilateral fetal hydronephrosis, if they had at least 1 follow-up ultrasonographic examination after the initial ultrasonography, and if they had planned to have delivery at our institution. Excluded from the study were multiple gestations or fetuses with other known structural or chromosomal anomalies. Also excluded were those patients with other genitourinary abnormalities involving the bladder, ureter, or architecture of the kidney. Cases were classified as mild, moderate, or severe hydronephrosis on the basis of the findings of their initial ultrasonographic examinations using the classification criteria of Mandell. 8 These findings were compared with subsequent ultrasonographic findings to determine whether the hydronephrosis had resolved or whether the classification had remained the same or changed (improved or worsened) during the pregnancy. Statistical analysis was performed using the χ 2 test with the Fisher exact test where appropriate, with P <.05 considered significant. To obtain neonatal follow-up data, we reviewed records from the Department of Pediatric Urology at Connecticut Children s Medical Center, which is the only referral center in the region. Those patients who had persistent hydronephrosis at the time of their last ultrasonography but whose children did not receive consultation or treatment at this center were telephoned by the investigators to determine whether follow-up was performed elsewhere. Results A total of 57,966 ultrasonographic examinations were performed on 20,049 women during the study period. Of these, 393 (2%) met criteria for inclusion in the study. The antenatal courses of all 393 patients are summarized in Table 1. Most fetuses identified with hydronephrosis were classified as having unilateral or bilateral mild hydronephrosis at initial ultrasonography 1066 J Ultrasound Med 20:1065 1069, 2001
Feldman et al Table 1. Summary of the Antenatal Course of 393 Patients With Fetal Hydronephrosis Classification Total, n (%) Resolved, n (%) Improved, n (%) Unchanged, n (%) Worsened, n (%) Mild 347 (88.3) 178 (51)* 0 (0) 134 (39) 35 (10) Moderate 40 (10.2) 6 (15) 10 (25) 19 (48) 5 (12) Severe 6 (1.5) 0 (0) 4 (67) 2 (33) (0) *P <.001 when compared with resolution of moderate or severe hydronephrosis. (347 [88%]). Of these, 51% resolved completely, 39% remained unchanged, and 10% worsened before delivery. Figures 2 and 3 illustrate the antepartum courses of unilateral and bilateral mild fetal hydronephrosis, respectively. There were 40 fetuses identified with moderate unilateral or bilateral hydronephrosis (10%) at initial ultrasonography. Only 6 (15%) of these cases resolved before delivery, whereas 10 (25%) improved to the mild classification, 19 (48%) remained unchanged, and 5 (12%) worsened to severe. Six cases of severe hydronephrosis were identified; none had in utero resolution. Four of the 6 cases improved to moderate or mild, and 2 cases remained unchanged throughout the pregnancy. When compared with cases of moderate or severe hydronephrosis, mild cases were more likely to have resolution in utero (P <.001). Of the 347 fetuses identified with mild hydronephrosis, 221 were found to have unilateral hydronephrosis, and 126 had bilateral hydronephrosis. In evaluating the antenatal courses of those with unilateral hydronephrosis, 113 cases (51.2%) had complete resolution, 84 (37.7%) remained the same, and 24 (10.6%) worsened to the moderate or severe classification at some point during the pregnancy. Of the 126 fetuses identified with bilateral hydronephrosis, 64 (50.8%) had complete resolution, whereas 50 (39.8%) remained the same and 12 (9.3%) worsened. The differences in the rates of resolution among those with unilateral versus bilateral mild hydronephrosis were not statistically significant. Of the 40 cases of moderate hydronephrosis, 15 (37.5%) were unilateral and 25 (67.5%) were bilateral. Follow-up data on the cases of unilateral moderate hydronephrosis revealed resolution in 1 case (6.7%), improvement to mild in 5 (33.3%), no change in 8 (53.3%), and worsening to severe in 1 (6.7%). Of the 25 cases of bilateral moderate hydronephrosis, 5 cases (20%) resolved, 5 (20%) improved to mild, 11 did not change (44%), and 4 worsened (16%). Cases of bilateral moderate hydronephrosis were significantly more likely to resolve than cases of unilateral moderate hydronephrosis (P <.05). There were no cases of in utero intervention among any patients in the study group. Neonatal follow-up data were available after 1996, at which point the referral network to the Pediatric Urology Division was well established. There were 110 cases of persistent hydronephrosis identified from 1996 to 1999. Of these, we were able to identify 25 (23%) patients who received consultation or treatment by a pediatric urologist in the newborn period. Follow-up data were available on these patients for up to 6 years. Table 2 summarizes the diagnoses and outcomes of these patients. Seven of these patients required surgery, including ureteropelvic junction obstruction repair in Figure 2. Outcomes of 221 cases of unilateral mild hydronephrosis. Figure 3. Outcomes of 126 cases of bilateral mild hydronephrosis. J Ultrasound Med 20:1065 1069, 2001 1067
Evaluation and Follow-up of Fetal Hydronephrosis Table 2. Summary of 25 Cases of Pediatric Urologic Follow-up No. of Diagnosis* Cases Surgery Outcome Follow-up, y Hydronephrosis 10 0/10 5 resolved, 2 6 3 improved, 2 stable Reflux 5 2/5 2 resolved, 2 4 3 improved Ureteropelvic junction 6 5/6 6 resolved 2 3 obstruction Duplication 2 0/2 1 resolved, 3 6 1 improved Megaureter 2 0/2 2 resolved 2 3 *Primary diagnosis for neonate. 5 patients and nephrectomy in 2. There was no significant association between the need for surgery and the severity of hydronephrosis at the last ultrasonographic examination (P =.32) or whether the hydronephrosis worsened during the pregnancy (P =.65). Of the 7 cases requiring surgery, 5 were identified as mild hydronephrosis, and 2 were identified as severe at initial ultrasonography. Four of the mild cases worsened to moderate hydronephrosis, and 1 remained mild. The 2 severe cases remained unchanged throughout the pregnancies. Of the remaining 85 patients whose newborns did not receive consultation or treatment with a pediatric urologist at our referral center, 36 (42%) were reached by telephone, and followup information was completed. By patient report, neonatal renal ultrasonography was performed on an outpatient basis in 19 cases, with normal results reported in 16 cases and mild hydronephrosis, which was followed by a pediatric nephrologist, in 3 cases. Seventeen patients received no neonatal follow-up according to our telephone survey. In all of these cases, parents reported that the children were healthy with no known medical problems. Discussion Prenatal ultrasonographic detection of a fluid collection in the fetal renal pelvis can often represent a diagnostic dilemma for the obstetrician and may well lead to unnecessary parental anxiety. In many cases, however, there is a major benefit to alerting the pediatrician and pediatric urologist of a possible fetal urologic malformation so that early postnatal evaluation and treatment can be initiated. Although several ultrasonographic diagnostic criteria have been proposed for the detection of fetal hydronephrosis, 7,9 there is no established universal definition. In addition, although many investigators use the renal pelvic AP diameter as the primary technique for diagnosing fetal hydronehprosis, 7,9,10 others have suggested that the ratio of AP pelvic diameter to AP kidney diameter may be helpful. 10,11 Again, no universally accepted cutoff values of these measurements have been established. The classification system established by Mandell 8 considers the normal growth that occurs in the fetal kidney as gestational age advances, thereby establishing different threshold values at different gestational ages. Although it is widely accepted that a fetal renal pelvic AP diameter of greater than 10 mm is abnormal at any gestational age, setting a lower threshold earlier in gestation may allow for earlier detection of a potential genitourinary anomaly and may allow for appropriate antenatal monitoring. The data we present, using the sonographic classification system of Mandell, 8 indicate that 51% of fetuses with unilateral or bilateral mild hydronephrosis are likely to have complete resolution before delivery, possibly eliminating the need for a postnatal evaluation of the infant. However, because of the small (10%) chance of worsening hydronephrosis, we recommend follow-up ultrasonography before delivery with appropriate postnatal follow-up if needed. Fetuses with moderate or severe hydronephrosis on the basis of this classification system are significantly less likely to undergo resolution. They should be followed antenatally for worsening hydronephrosis, and if the hydronephrosis is associated with oligohydramnios, possible antenatal intervention could be considered. Postnatally, these patients should receive a urologic evaluation. The retrospective nature of this study did not allow us to report on the neonatal outcomes of all the study patients; however, we were able to report on the patients who did receive care at our pediatric urology referral center as well as many patients who received only outpatient follow-up neonatal renal ultrasonography. To obtain more complete outcome data on all patients with the diagnosis of prenatal hydronephrosis, a prospective cohort study is needed. 1068 J Ultrasound Med 20:1065 1069, 2001
Feldman et al References 1. Elder JS. Antenatal hydronephrosis: fetal and neonatal management. Pediatr Clin North Am 1997; 44:1299 1321. 2. Stocks A, Richards D, Frentzen B, Richard G. Correlation of prenatal renal pelvic anteroposterior diameter with outcome in infancy. J Urol 1996; 155: 1050 1052. 3. Barker P, Cave MM, Thomas DFM, et al. Fetal pelviureteric junction obstruction: predictors of outcome. Br J Urol 1995;76: 640-52. 4. Najmaldin AS, Burge DM, Atwell, JD. Outcome of antenatally diagnosed pelviureteric junction hydronephrosis. Br J Urol 1991; 67:96 99. 5. Clautice-Engle T, Anderson NG, Allan RB, Abbott GD. Diagnosis of obstructive hydronephrosis in infants: comparison sonograms performed at 6 days and 6 weeks after birth. AJR Am J Roentgenol 1995; 163:963 967. 6. Arnold AJ, Rickwood AMK. Natural history of pelviureteric obstruction detected by prenatal sonography. Br J Urol 1990; 65:91 96. 7. Corteville JE, Gray DL, Crane, JP. Congenital hydronephrosis: correlation of fetal ultrasonographic findings with infant outcome. Am J Obstet Gynecol 1991; 165:384 388. 8. Mandell J. Prenatal diagnosis and treatment of obstructive uropathies. Probl Urol 1990; 4:547 554. 9. Grignon A, Filion R, Filiatrault D, et al. Urinary tract dilation in utero: classification and clinical applications. Radiology 1986; 160:645 647. 10. Arger PH, Coleman BG, Mintz MC, et al. Routine fetal genitourinary tract screening. Radiology 1985; 156:485 489. 11. Kleiner B, Callen PW, Filly RA. Sonographic analysis of the fetus with ureteropelvic junction obstruction. AJR Am J Roentgenol 1987; 148:359 363. J Ultrasound Med 20:1065 1069, 2001 1069