In this issue of The Journal, Ismaili and associates in Belgium have advanced our understanding of multicystic dysplastic kidney (MCDK), and they have made an important observation that should change practice significantly.
1To appreciate the importance of the Ismaili study, and to place it into perspective, some background must be reviewed.
MCDK, although “cystic” in appearance, is not one of the inherited renal cystic diseases—autosomal dominant (“adult”) or autosomal recessive (“infantile”) polycystic disease. It is really a form of renal dysplasia, in which cystic elements are found along with immature, undifferentiated, primitive tissue. Embryologically, MCDK may result from abnormal renal morphogenesis, likely consequent to abnormalities of developmentally expressed genes. Some degree of ureteropelvic junction obstruction is generally present. MCDK, in addition to cystic elements, may show some hydronephrotic features.
MCDKs are usually unilateral and nonfunctional. In the rare circumstances in which MCDK is bilateral, the condition is generally lethal in the newborn period. The overwhelming majority of isolated MCDKs are sporadic anomalies, for which no genetic explanation is currently available. In other cases, however, MCDK is but one component of a generalized disorder such as the VATER (vertebral defects, imperforate anus, tracheoesophageal fistula, radial and renal dysplasia) association, Zellweger syndrome, or the BOR (branchio-oto-renal) syndrome.
Until the 1980s, MCDK was generally diagnosed in the course of urinary tract imaging that was being obtained for the investigation of problems such as an abdominal mass, urinary tract infection, or hypertension. Thus, virtually all children with MCDK, by definition, had a complication of the condition.
6Pediatric urology texts from this period and before portray MCDK as an unusual condition, generally symptomatic, often associated with other urinary tract anomalies, requiring further diagnostic study, and frequently necessitating nephrectomy. For example, Campbell, in his 1951 text, described an entity “congenital multilocular cysts” (clearly MCDK), which was “rare” and which could cause symptoms from pressure on adjacent organs or bowel obstruction. The condition was to be differentiated from renal tumors, and was treated by surgical removal.
Changes in obstetric practice have altered this situation considerably. Today, virtually all MCDKs are recognized prenatally by ultrasonography. Thus, the typical patient is no longer a child being investigated for a specific symptom.
With the changing epidemiology of MCDK, a shift in management also has occurred. Routine initial nephrectomy for MCDK, once widely practiced, is extremely rare today and is limited to situations in which a real complication of the disorder is occurring.
11The MCDK registry documented a significant drop in the frequency of nephrectomy over the 13 years in which it was collecting data.
12Previously accepted indications for “prophylactic” nephrectomy, such as concern regarding the development of malignancy or hypertension, have disappeared. Again in Syracuse, two nephrectomies for MCDK have been performed in the past 3 years, both because of increasing size of the kidney in older children.
Similarly, the diagnostic approach to the presumed MCDK has been evolving, mainly in the direction of a more conservative approach. At one point, for example, it was generally accepted that some type of functional imaging (ie, renal scan) was needed to differentiate between MCDK (which should be nonfunctioning) and significant ureteropelvic junction obstruction with hydronephrosis, in which some residual function may be present. Today, most pediatric radiologists would agree that a high-quality renal ultrasonogram is usually sufficient to make the diagnosis of MCDK.
The remaining imaging issue for the child with MCDK is the need for a voiding cystourethrogram (VCUG). Mainly because of concern for the presence of vesicoureteral reflux (VUR) in the contralateral kidney, this procedure continues to be recommended
15and performed, although a few studies have recently questioned the need for this.
Interestingly, even studies suggesting the need for this additional imaging on close examination do not always make a compelling case. Flack and Bellinger,
14for example, reported the results of VCUGs in 29 children with MCDK; eight had contralateral VUR despite what were said to be normal ultrasonographic appearance. Most of this VUR, however, was low grade, and all children who had subsequent VCUGs had resolution before 2 years of age. Thus, it is arguable whether any benefit accrued to the children by the recognition of this VUR. One of these children was said to have grade IV VUR and underwent surgical repair. It would be surprising for such a finding to be missed using the protocol described by Ismaili et al.
It is into this context that the Ismaili et al study fits. These investigators had the advantage of working with a cohort of children, whose MCDK was ascertained by prenatal ultrasonography. A close collaboration with their center's obstetric unit allowed early recognition of such children. Each child was subjected to a systematic series of imaging procedures, again uniform for the group. These studies included additional ultrasonograms, renal scans, and VCUGs. Follow-up was complete and comprehensive.
The important finding of this study was the sensitivity and specificity of two renal ultrasonograms, about 1 month apart, in detecting contralateral urinary tract problems. As in other published series, the authors detected a variety of contralateral anomalies in this cohort. On the other hand, virtually all of these anomalies were detected by one of the two renal ultrasonograms. There were, for example, only four children with contralateral VUR who had totally normal ultrasonograms. In each, the reflux was low grade, resolved spontaneously, and was likely inconsequential. The ultrasonography examinations also detected most of the lower urinary tract anomalies. There were only three lower tract anomalies not suspected by ultrasonography; again, none of these was likely to be important.
It must be stressed that all of these ultrasonography examinations were undertaken by highly experienced personnel, using optimal equipment. The studies included measurements such as the pelvic diameter, which may not be a part of routine renal ultrasonography in every center.
On the basis of this study, as well as some of the other studies cited, it is reasonable to draw conclusions about the investigation of infants with prenatal ultrasonograms that suggest MCDK. All of these children should have high-quality urinary tract ultrasonography studies in the first few days of life. The study should be repeated in about a month. These should be performed in an experienced center. No further imaging appears to be necessary if both of these studies show a kidney that has the ultrasonographic features of MCDK, as well as a normal bladder and contralateral kidney. An important caveat to this recommendation is the reminder that the imaging must be performed in a center with interest and expertise in the imaging of the infant's urinary tract.
We typically obtain follow-up studies on such children every 6 to 12 months for the first few years, looking for involution of the MCDK as well as normal growth of the contralateral kidney. Admittedly, the evidence basis for these follow-up recommendations is not as strong as it is for the initial studies. Perhaps ongoing studies of this cohort by Ismaili and associates may answer this question as well.
- Routine voiding cystourethrography is of no value in neonates with unilateral multicystic dysplastic kidney.J Pediatr. 2005; 146: 759-763
- Pathogenesis of dysplastic kidney associated with urinary tract obstruction in utero.Nephrol Dial Transplant. 2002; 17: 37-38
- Normal nephrogenesis occurs in the early stage of bilateral multicystic dysplastic kidneys.Arch Gynecol Obstet. 2002; 266: 133-135
- The molecular control of renal branching morphogenesis: current knowledge and emerging insights.Differentiation. 2002; 70: 227-246
- Syndromic ear anomalies and renal ultrasounds.Pediatrics. 2001; 108: E32
- Neonatal hydronephrosis in the era of sonography.AJR. 1987; 148: 959-963
- Clinical Pediatric Urology.WB Saunders Company, Philadelphia1951
- Unilateral multicystic dysplastic kidney disease: defining the natural history.Acta Paediatr. 2000; 89: 811-813
- Clinical course and outcome for children with multicystic dysplastic kidneys.Pediatr nephrol. 2000; 14: 1098-1101
- Multicystic dysplastic kidney in children: US follow-up.Radiology. 1993; 186: 785-788
- Involution rate of multicystic renal dysplasia.Pediatrics. 1998; 102: E73
- Report of the multicystic kidney registry: preliminary findings.J Urol. 1993; 150: 1870-1872
- Lower urinary tract abnormalities in children with multicystic dysplastic kidney.Radiology. 1997; 203: 223-226
- The multicystic dysplastic kidney and contralateral vesicoureteral reflux: protection of the solitary kidney.J Urol. 1993; 150: 1873-1874
- Contralateral renal abnormalities in multicystic-dysplastic kidney disease.J Pediatr. 1992; 121: 65-67
© 2005 Elsevier Inc. Published by Elsevier Inc. All rights reserved.