Diagnostic accuracy of fetal renal pelvis anteroposterior diameter as a predictor of uropathy: a prospective study

Share Embed


Descrição do Produto

Ultrasound Obstet Gynecol 2004; 24: 745–749 Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/uog.1764

Diagnostic accuracy of fetal renal pelvis anteroposterior diameter as a predictor of uropathy: a prospective study M. C. F. BOUZADA*, E. A. OLIVEIRA†, A. K. PEREIRA‡, H. V. LEITE‡, A. M. RODRIGUES‡, L. A. FAGUNDES†, R. P. GONCALVES† ¸ and R. L. PARREIRAS† *Neonatal and †Pediatric Nephrourology Units, Department of Pediatrics and ‡Fetal Medicine Unit, Department of Gynecology and Obstetrics, Hospital das Cl´ınicas, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil

K E Y W O R D S: fetal; hydronephrosis; prenatal diagnosis; pyelectasis; ROC curve; sensitivity; specificity

ABSTRACT Objective The purpose of this study was to assess the accuracy of prenatal ultrasound measurement of anteroposterior renal pelvis diameter (APD) to discriminate between significant uropathy and idiopathic renal pelvis dilatation. Methods One-hundred-and-three neonates who were found to have fetal renal pelvis dilatation, defined as presence of an APD ≥ 5 mm, underwent systematic investigation for uropathies and were prospectively followed. An ultrasound scan was performed after the first week of postnatal life and all infants underwent a voiding cystourethrogram. Neonates with an APD larger than 10 mm underwent renal scintigraphy. Ultrasound scans, clinical examination and laboratory reviews were scheduled at 6-month intervals. Receiver–operating characteristics (ROC) curves were constructed to determine the best cut-offs for APD to identify renal units with significant uropathy as well as those requiring surgical intervention. Significant uropathy was defined as the presence of wellestablished urinary tract abnormalities or when there was abnormal renal scintigraphy. Results The estimated area under the curve for APD was 0.900 (95% CI, 0.841–0.942) indicating excellent power to discriminate between idiopathic pelvis dilatation and significant uropathy. The sensitivity and specificity for the 7.5 mm cut-off point were 97.9% and 40.6%, respectively. To identify infants who required surgical intervention, the calculated area under the curve was 0.953 (95% CI, 0.908–0.980). Conclusion Our results suggest that measurement of APD is an excellent test to identify fetuses with significant uropathy, as well as those requiring postnatal

intervention. Copyright  2004 ISUOG. Published by John Wiley & Sons, Ltd.

INTRODUCTION Advances in prenatal diagnosis over the last two decades have improved the detection and management of urinary tract abnormalities. The protocol for postnatal evaluation of severe hydroureteronephrosis is well-established and usually includes ultrasound scan, voiding cystourethrogram and diuretic renography1,2 . Nevertheless, a prenatal ultrasound scan frequently reveals isolated renal pelvis dilatation and there have been controversies regarding the threshold of fetal renal pelvis dilatation indicative of significant uropathy3 – 5 . Although anteroposterior fetal renal pelvic diameter (APD) has been the most common parameter used to indicate the need for postnatal investigation, there has been considerable controversy regarding this issue2,6 . Whereas some authors advocate that minimal renal pelvis dilatation requires only postnatal ultrasound scans to identify those cases requiring further investigation, others recommend a complete postnatal investigation in order not to miss vesicoureteral reflux in infants7 – 10 . Those who advocate only an ultrasound investigation are concerned with the possible adverse consequences of invasive procedures in asymptomatic neonates11 – 13 . The purpose of this study was to assess the accuracy of prenatal ultrasound measurement of APD in discriminating between significant uropathy and benign transitory renal pelvis dilatation. The data are based on a series of infants who underwent a systematic protocol of investigation and were prospectively followed.

´ Correspondence to: Dr M. C. F. Bouzada, Avenida Olegario Maciel 2174/1901, 30180-112, Belo Horizonte, Minas Gerais, Brazil (e-mail: [email protected]) Accepted: 20 May 2004

Copyright  2004 ISUOG. Published by John Wiley & Sons, Ltd.

ORIGINAL PAPER

Bouzada et al.

746

METHODS One-hundred-and-three neonates born consecutively at Hospital das Cl´ınicas-UFMG (Belo Horizonte, Brazil) between May 1999 and May 2003, who were found to have isolated renal pelvis dilatation by prenatal ultrasound scans after 28 weeks’ gestation, underwent systematic investigation for uropathies and were prospectively followed. The criterion adopted for inclusion in the study was the presence of APD, measured on a transverse section of the fetal renal hilum, ≥ 5 mm14,15 . Fetal anatomy was carefully screened by a trained examiner. Fetuses with hydronephrosis associated with ureteral dilatation or bladder distension, or with oligohydramnios were excluded from the present analysis. Fetuses with aneuploidy or multiple malformations were also excluded. Postnatally, neonates were investigated according to the systematic protocol illustrated in Figure 1. The protocol is based on a previously published study16 and was modified according to the suggestions of Jaswon et al15 . An ultrasound scan was performed after the first week of postnatal life and all infants underwent a voiding cystourethrogram (VCU). Antibiotic prophylaxis was started on the first postnatal day in all infants included in the analysis. If both the postnatal scan and VCU were normal, antibiotic prophylaxis was discontinued and the infant was followed until 12 months of age. Ultrasound scans, clinical examination (including growth and blood pressure measurements), and laboratory reviews were

scheduled at 6-month intervals. The postnatal ultrasound scans were performed by the same trained examiner. If the postnatal scan showed mild pelvis dilatation (APD < 10 mm) and the VCU was normal, a similar followup schedule was applied and antibiotic prophylaxis continued until the involution of renal pelvis dilatation. When the VCU was normal but postnatal ultrasound scans demonstrated renal pelvis dilatation ≥ 10 mm, renal scintigraphy (99 m-Tc dimercaptosuccinic acid (DMSA), 99 m-Tc diethylenetriaminepentaacetic acid (DTPA)) was performed after the first month. DMSA uptake was measured 2 h after intravenous injection of the radioisotope. Relative uptake was calculated according to the equation proposed by Konda et al.17 . The dilated renal units were classified into one of three functional groups depending on DMSA uptake determined by renal scan. A poorly functioning unit was designated as having < 20% overall function, a moderately functioning kidney as 20–39%, and preserved renal function as > 40%. The renal units were classified into one of three groups depending on interpretation of the DTPA washout curves. Group I was defined as a non-obstructed unit with a descending curve, Group II was classified as an intermediate pattern with a flat curve, and Group III was defined as a rising curve in the renogram (obstructed units). A non-operative approach was attempted in renal units with good function (> 40%) as ascertained by DMSA scan and a non-obstructed pattern on DTPA. Otherwise,

Fetal renal pelvis APD ≥ 5mm

US and VCU

Both normal stop prophylaxis

VUR

US 6 and 12 months

DMSA prophylaxis

Both normal discharged

Hydronephrosis without VUR

Postnatal APD < 10 mm

Postnatal APD ≥ 10 mm

US 6 and 12 months

DMSA and DTPA

Renal uptake < 40% obstructive pattern

Renal uptake > 40% intermediate pattern

Pyeloplasty

Serial US

Figure 1 Protocol for the investigation of infants with antenatal renal pelvis dilatation. APD, anteroposterior renal pelvis diameter; DMSA, dimercaptosuccinic acid; DTPA, diethylenetriaminepentaacetic acid; US, ultrasound scan; VCU, voiding cystourethrogram; VUR, vesicoureteral reflux.

Copyright  2004 ISUOG. Published by John Wiley & Sons, Ltd.

Ultrasound Obstet Gynecol 2004; 24: 745–749.

Diagnostic accuracy of fetal renal pelvis diameter

RESULTS A total of 103 infants were included in the study. Eighty were males (77.7%) and 23 were females (22.3%). Fiftyone infants presented bilateral renal pelvis dilatation, corresponding to 154 renal units for study. Of the 154 dilated renal units, 48 (31%) presented with a significant uropathy and 106 (68.8%) with idiopathic transitory dilatation (Table 1). There was a predominance of ureteropelvic junction obstruction (22%) among renal units with a significant uropathy. Seventeen renal units (11%) required surgical intervention for relief of the obstruction. The area under the ROC curve was 0.900 (95% CI, 0.841–0.942) (Figure 2). A cut-off of 7.5 mm for APD detected most neonates with a significant uropathy. The sensitivity was 97.9% (88.9–99.7%), the specificity was 40.6% (31.1–50.5%), and the positive and negative likelihood ratios were 1.65 and 0.05, respectively. The relative risk for significant uropathy was 18.2 (95% CI, 2.6–98.1) in renal units with APD ≥ 7.5 mm. The sensitivity and specificity for the traditional 10 mm cut-off were 87.5% and 72.6%, respectively.

Copyright  2004 ISUOG. Published by John Wiley & Sons, Ltd.

Table 1 Postnatal diagnosis of urinary tract anomalies and surgical interventions

Patients n (%)

Postnatal diagnosis Idiopathic transitory dilatation Ureteropelvic junction obstruction Hypoplastic kidney Ureterocele Vesicoureteral reflux Megaureter Ureteral duplication Horseshoe kidney Total

Renal units requiring surgical Renal units management n (%) n (%)

59 (57.3) 106 (68.8)

0

31(30)

34 (22)

14 (82)

4 (3.9) 3 (2.9) 2 (1.9) 2 (1.9) 1 (1) 1 (1) 103 (100)

4 (2.6) 3 (1.9) 2 (1.3) 2 (1.3) 2 (1.3) 1 (0.64) 154 (100)

0 3 (18) 0 0 0 0 17 (100)

APD (mm) 100 7.5 10

80

9.5

12 Sensitivity

the patients were managed surgically by Anderson–Hynes dismembered pyeloplasty18 . Receiver–operating characteristic (ROC) curves were analyzed to determine the optimal APD at which to diagnose renal units with significant uropathy and also to identify those units at risk of significant postnatal obstruction, requiring pyeloplasty or other urological surgical intervention. The goals were: (1) to choose the threshold with the highest sensitivity for predicting uropathy and to minimize the chance of missing a clinically significant urinary tract abnormality, and (2) to choose the threshold with the best accuracy for predicting the need for postnatal surgery, thus providing information for counseling the patient’s family. Combined data from postnatal ultrasound, VCU, and renal scintigraphy were regarded as the reference standard. Significant uropathy was defined as the presence of well-established urinary tract abnormality or the presence of an abnormal DMSA scan and/or a diuretic renogram. In contrast, renal pelvis dilatation was defined as idiopathic transitory dilatation and regarded as a non-significant uropathy in the analysis. Therefore, after a minimal followup of 6 months, renal units were classified into three groups for accuracy analysis: (1) idiopathic transitory dilatation; (2) significant uropathy amenable to clinical management; and (3) significant uropathy requiring surgical intervention. The prevalence of disease (pretest probability) was regarded as 28% according to the findings of Ismaili et al.5 . The index test (renal pelvic APD) and the reference standard tests (VCU and renal scintigraphy) were performed by examiners who were blind to the results of the other tests. Informed consent was obtained from the parents and the study was approved by the ethics committee of the university.

747

60 16 40

22

20

0 0

20

40 60 100 − Specificity

80

100

Figure 2 Receiver–operating characteristics (ROC) curve based on the fetal renal pelvic anteroposterior diameter (APD) index as an indicator of significant uropathy.

APD cut-offs were also calculated to identify renal units requiring surgical intervention. The area under the curve estimated by ROC was 0.953 (95% CI, 0.908–0.980) (Figure 3). A cut-off of 15 mm had the best diagnostic performance to identify the renal units requiring surgical intervention. The sensitivity was 82.4% (56.0–96.0%), the specificity was 87.9% (81.4–92.8%), and the positive and negative likelihood ratios were 6.83 and 0.20, respectively. The relative risk for surgical intervention was 17.8 (95% CI, 5.4–58) in renal units with APD ≥ 15 mm.

DISCUSSION The widespread use of ultrasound evaluation during pregnancy has increased the detection of genitourinary

Ultrasound Obstet Gynecol 2004; 24: 745–749.

Bouzada et al.

748 APD (mm) 100

13 15

80 18

Sensitivity

22 60 27 40

20

0 0

20

40

60

80

100

100 − Specificity

Figure 3 Receiver–operating characteristics (ROC) curve based on the fetal renal pelvic anteroposterior diameter (APD) index as an indicator of the need for surgical intervention.

anomalies in utero. The spectrum of urinary tract abnormalities identified prenatally is wide, ranging from minimal pyelectasis to urethral obstruction1 . Fetal pyelectasis is the most common urinary tract abnormality detected by prenatal sonography19 . Studies of unselected populations have demonstrated that the incidence of a minor degree of antenatal pyelectasis ranges from 2% to 4.5% in fetuses submitted to routine prenatal ultrasonography5,15 . Although the prenatal sonographic findings of these entities may be similar, the management and prognosis of each are different. Some factors such as bladder enlargement, megaureter, and oligohydramnios are well-established as markers of complex uropathies and of a worse prognosis20,21 . However, the natural history of isolated renal pelvis dilatation has not been fully determined and there has been controversy regarding issues related to the measurement of renal dilatation22 . Physicians who take care of the fetus and infant are challenged by the following question: when does dilation of the renal collecting system indicate a serious pathology with potential for renal deterioration?20,21 . APD has been the most common parameter used to indicate a potential urinary tract abnormality. Our findings support the notion that the APD is a simple and efficient method for the screening of possible significant uropathies in otherwise healthy fetuses. The estimated area under the curve was 0.900 (95% CI, 0.841–0.942), the sensitivity and specificity for the 7.5 mm cut-off were 97.9% and 40.6%, respectively. Previously, some authors have advocated that an anteroposterior diameter of < 10 mm for the renal pelvis is physiological and would not require postnatal investigation23 . However, recent prospective studies with a more complete postnatal investigation have demonstrated that minor renal pelvis

Copyright  2004 ISUOG. Published by John Wiley & Sons, Ltd.

dilatation can be indicative of significant uropathy. Corteville et al.24 demonstrated a sensitivity of 91% and a specificity of 66% for the threshold of 7 mm beyond 24 weeks’ gestation. Adra et al.25 reported a sensitivity of 85% and specificity of 41% for the cut-off of 8 mm beyond 28 weeks’ gestation. Considering these data as a whole, it can be seen that lower limits for the APD cut-off presented greater sensitivity but, as expected, they generated an increase in the number of false-positive cases. This is an important question that should be thoroughly examined. Harding et al.7 showed that detection of any fetal abnormality, minimal pyelectasis included, generates considerable parental anxiety and distress. Although they agreed that minimal fetal renal pelvis dilatation should not be ignored, they advocated a less intensive prenatal and postnatal management for these patients. Our argument in favor of maximal sensitivity is the following: the major benefit of prenatal diagnosis is to alert the pediatrician to perform an early evaluation of the urinary tract, thus preventing urinary infections and deterioration of an obstructed renal unit. We believe that thresholds for tests for early diagnosis of diseases in asymptomatic individuals should maximize the true positive rate since a false-negative labeling can be potentially harmful in this context. Thus, the possible benefits of early identification of significant uropathies must be set against the execution of expensive and invasive tests in healthy infants. The second part of our study confirmed the power of fetal APD for discriminating between fetuses that will require postnatal surgical intervention and those that will benefit from conservative management. This is an important question for appropriate antenatal counseling of parents2 . Seventeen renal units (11%) required surgical intervention, most of them being submitted to pyeloplasty for relief of ureteropelvic junction obstruction. Similar findings were also reported by Ouzounian et al.26 . The area under the curve was 0.953 (95% CI, 0.908–0.980), indicating an excellent discrimination. The sensitivity and specificity for the 15 mm cut-off were 82.4% and 87.9%, respectively. A 20 mm measurement of fetal or postnatal APD threshold has been proposed as a useful parameter for distinguishing those infants who would benefit from surgical intervention27,28 . Our results indicate that, if this threshold is used, some cases requiring surgical intervention would be missed. It is also important to note that some cases may need surgical relief of the obstruction in the presence of a smaller degree of pelvis dilatation. As an example, there were two infants in our series with APD less than 20 mm (one with ureterocele and the other with primary obstructive megaureter) who had to undergo surgery. In conclusion, our data support the use of APD to identify fetuses with significant uropathy, as well as those requiring postnatal intervention. ROC curve analysis demonstrated that fetal APD has sufficient discriminative power to be useful in the clinical setting. Our results suggest that the larger the renal pelvis during the third trimester, the greater the risk that postnatal intervention will be required. However, we believe that

Ultrasound Obstet Gynecol 2004; 24: 745–749.

Diagnostic accuracy of fetal renal pelvis diameter judicious postnatal management must be established after a thorough physical examination and comprehensive ultrasonography of the urinary tract of the neonate. Prospective studies that include data on urinary tract infections, renal growth, renal damage and a prolonged follow-up of infants with prenatally detected renal pelvis dilatation, are necessary to refine the diagnostic criteria for the selection of the best approach to these patients.

ACKNOWLEDGMENTS ˜ de This study was partially supported by Funda¸cao Amparo a` Pesquisa do Estado de Minas Gerais (FAPEMIG), Conselho de Desenvolvimento Cientı´fico ´ ´ e Tecnologico (CNPq) and Pro-Reitoria de PesquisaUFMG.

REFERENCES 1. Elder JS. Antenatal hydronephrosis. Fetal and neonatal management. Pediatr Clin North Am 1997; 44: 1299–1321. 2. Woodward M, Frank D. Postnatal management of antenatal hydronephrosis. BJU Int 2002; 89: 149–156. 3. Kent A, Cox D, Downey P, James SL. A study of mild fetal pyelectasia – outcome and proposed strategy of management. Prenat Diagn 2000; 20: 206–209. 4. Sairam S, Al-Habib A, Sasson S, Thilaganathan B. Natural history of fetal hydronephrosis diagnosed on mid-trimester ultrasound. Ultrasound Obstet Gynecol 2001; 17: 191–196. 5. Ismaili K, Hall M, Donner C, Thomas D, Vermeylen D, Avni FE. Results of systematic screening for minor degrees of fetal renal pelvis dilatation in an unselected population. Am J Obstet Gynecol 2003; 188: 242–246. 6. Sherer DM. Is fetal hydronephrosis overdiagnosed? Ultrasound Obstet Gynecol 2000; 16: 601–606. 7. Harding LJ, Malone PS, Wellesley DG. Antenatal minimal hydronephrosis: is its follow-up an unnecessary cause of concern? Prenat Diagn 1999; 19: 701–705. 8. 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. 9. McIlroy PJ, Abbott GD, Anderson NG, Turner JG, Mogridge N, Wells JE. Outcome of primary vesicoureteric reflux detected following fetal renal pelvic dilatation. J Paediatr Child Health 2000; 36: 569–573. 10. Brophy MM, Austin PF, Yan Y, Coplen DE. Vesicoureteral reflux and clinical outcomes in infants with prenatally detected hydronephrosis. J Urol 2002; 168: 1716–1719. 11. Misra D, Kempley ST, Hird MF. Are patients with antenatally diagnosed hydronephrosis being overinvestigated and overtreated? Eur J Pediatr Surg 1999; 9: 303–306. 12. Vates TS, Shull MJ, Underberg-Davis SJ, Fleisher MH. Complications of voiding cystourethrography in the evaluation of

Copyright  2004 ISUOG. Published by John Wiley & Sons, Ltd.

749

13.

14.

15.

16.

17.

18.

19.

20.

21. 22.

23.

24.

25.

26.

27.

28.

infants with prenatally detected hydronephrosis. J Urol 1999; 162: 1221–1223. Yerkes EB, Adams MC, Pope JC 4th, Brock JW 3rd. Does every patient with prenatal hydronephrosis need voiding cystourethrography? J Urol 1999; 162: 1218–1220. Dudley JA, Haworth JM, McGraw ME, Frank JD, Tizard EJ. Clinical relevance and implications of antenatal hydronephrosis. Arch Dis Child Fetal Neonatal Ed 1997; 76: 31–34. Jaswon MS, Dibble L, Puri S, Davis J, Young J, Dave R, Morgan H. Prospective study of outcome in antenatally diagnosed renal pelvis dilatation. Arch Dis Child Fetal Neonatal Ed 1999; 80: 135–138. Oliveira EA, Diniz JS, Cabral AC, Leite HV, Colosimo EA, Oliveira RB, Vilasboas AS. Prognostic factors in fetal hydronephrosis: a multivariate analysis. Pediatr Nephrol 1999; 13: 859–864. Konda R, Sakai K, Ota S, Abe Y, Hatakeyama T, Orikasa S. Ultrasound grade of hydronephrosis and severity of renal cortical damage on 99m technetium dimercaptosuccinic acid renal scan in infants with unilateral hydronephrosis during followup and after pyeloplasty. J Urol 2002; 167: 2159–2163. Anderson JC, Hynes W. Retrocaval ureter: a case diagnosed preoperatively and treated successfully by a plastic operation. Br J Urol 1949; 21: 209–211. Mandell J, Blyth BR, Peters CA, Retik AB, Estroff JA, Benacerraf BR. Structural genitourinary defects detected in utero. Radiology 1991; 178: 193–196. Oliveira EA, Diniz JS, Cabral AC, Pereira AK, Leite HV, Colosimo EA, Vilasboas AS. Predictive factors of fetal urethral obstruction: a multivariate analysis. Fetal Diagn Ther 2000; 15: 180–186. Roth JA, Diamond DA. Prenatal hydronephrosis. Curr Opin Pediatr 2001; 13: 138–141. Persutte WH, Koyle M, Lenke RR, Klas J, Ryan C, Hobbins JC. Mild pyelectasis ascertained with prenatal ultrasonography is pediatrically significant. Ultrasound Obstet Gynecol 1997; 10: 12–18. Grignon A, Filion R, Filiatrault D, Robitaille P, Homsy Y, Boutin H, Leblond R. Urinary tract dilatation in utero: classification and clinical applications. Radiology 1986; 160: 645–647. Corteville JE, Gray DL, Crane JP. Congenital hydronephrosis: correlation of fetal ultrasonographic findings with infant outcome. Am J Obstet Gynecol 1991; 165: 384–388. Adra AM, Mejides AA, Dennaoui MS, Beydoun SN. Fetal pyelectasis: is it always ‘‘physiologic’’? Am J Obstet Gynecol 1995; 173: 1263–1266. Ouzounian JG, Castro MA, Fresquez M, al-Sulyman OM, Kovacs BW. Prognostic significance of antenatally detected fetal pyelectasis. Ultrasound Obstet Gynecol 1996; 7: 424–428. Ransley PG, Dhillon HK, Gordon I, Duffy PG, Dillon MJ, Barratt TM. The postnatal management of hydronephrosis diagnosed by prenatal ultrasound. J Urol 1990; 144: 584–587. Gotoh H, Masuzaki H, Fukuda H, Yoshimura S, Ishimaru T. Detection and assessment of pyelectasis in the fetus: relationship to postnatal renal function. Obstet Gynecol 1998; 92: 226–231.

Ultrasound Obstet Gynecol 2004; 24: 745–749.

Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.