Nephrology Dialysis Transplantation

Transcription

1 Nephrol Dial Transplant (2000) 15: Original Article Nephrology Dialysis Transplantation Sonographic pattern of recessive polycystic kidney disease in young adults. Differences from the dominant form Carlos Nicolau1, Roser Torra2, Celia Badenas3, Laureano Pérez2, Jesús A Oliver4, Alejandro Darnell2 and Concepció Brú1 1Imaging Diagnosis Center (Ultrasound Unit), Departments of 2Nephrology and 3Genetics, Hospital Clínic, IDIBAPS (Institut d Investigacions Biomèdiques August Pi i Sunyer), University of Barcelona and 4Department of Nephrology, Hospital Juan XXIII, Tarragona, Spain Abstract Background. To study the sonographic pattern of Introduction autosomal recessive polycystic kidney disease Autosomal recessive polycystic kidney disease (ARPKD) in early adulthood in order to identify ( ARPKD) is a hereditary disorder that affects the imaging criteria to diagnose this disease and to distinkidney and the liver. Its occurrence ranges from 1 in guish between recessive and autosomal dominant polycystic kidney disease (ADPKD) in that age group to 1 in births [1,2]. As a recessive disorder Methods. An abdominal ultrasound was performed on both parents of an affected child carry one copy of the four ARPKD subjects (with a mean age of 20.2) and defective gene but are not clinically affected. The on 33 ADPKD subjects in early adulthood (29 without disease is caused by mutation in a gene located in renal failure with a mean age of 20.5, and four with chromosome 6 which has yet to be identified. The renal failure with a mean age of 26.5). Linkage studies main characteristic in the kidneys is the presence of with ADPKD and ARPKD markers were compatible multiple cysts resulting from the dilatation of collecting with the clinical diagnosis in all cases. ducts [2 4]. In the liver, the disease is diffuse and Results. The renal sonographic features in ARPKD presents as portal fibrosis, dilatation of the bile ducts subjects included multiple small cysts in a normal-sized (Carolı s disease), or a combination of both [2,3]. This kidney, increased cortical echogenicity and loss of disease shows a wide clinical variability ranging from corticomedullary differentiation. In ADPKD subjects perinatal death to diagnosis in adolescence or in without renal failure, sonographic features included adulthood, in progressively milder forms [4 6]. In the few or multiple cysts of different sizes, in normal-sized latter cases, ARPKD can be clinically indistinguishable kidneys in 22 out of 29 patients (75.8%), normal from autosomal dominant polycystic kidney disease cortical echogenicity and conserved corticomedullary (ADPKD) because there is an overlap in the clinical differentiation, except in patients with nephromegaly. presentation. Therefore, it may be difficult to establish All ADPKD subjects with renal failure had nephromegaly adifferential diagnosis between them [4]. It is particuhepatic and loss of corticomedullary differentiation. The larly difficult when there is no definitive family history sonographic features in ARPKD patients of these diseases, the possibility of illegitimate paternity included portal fibrosis and in some cases Caroli s or spontaneous mutation of ADPKD [2]. Moreover, disease, while in ADPKD patients a normal hepatic in these cases the genetic linkage analysis is useless as echostructure was detected in all but one case, in well as being expensive, and unavailable in most addition to simple hepatic cysts in a few cases. hospitals. Thus, in these rare cases, the diagnosis is Conclusions. The evaluation of the sonographic fea- based on imaging of the involved organs (kidney and tures of the kidneys and those of the liver may help in liver) or eventually kidney and/or liver biopsy. the differential diagnosis between ARPKD and Sonography is the most useful non-invasive imaging ADPKD in early adulthood. technique in the evaluation of patients with polycystic kidney disease. In this article we present Keywords: ADPKD; ARPKD; Caroli s disease; hepatic ultrasonographic renal and hepatic findings of fibrosis; sonography; ultrasound ARPKD in early adulthood and compare them with ultrasonographic findings of ADPKD in the same age group and/or same clinical presentation in order to Correspondence and offprint requests to: Dr Carlos Nicolau, Department of Radiology, Hospital Clinic, Villarroel 170, characterize these conditions and to assist in differen- Barcelona, Spain. cnicolau@clinic.ub.es tiating them European Renal Association European Dialysis and Transplant Association

2 1374 Subjects and methods Subjects The sonographic findings of 37 young adults with polycystic kidney disease were retrospectively evaluated. We classified patients into three groups, using linkage analysis, age, and presence of renal failure. The first group consisted of four patients with ARPKD (one man, three women; age range years, mean age 20.2). The second group consisted of all the ADPKD patients who visited our department during the last 3 years who were in the same age group as the patients with ARPKD (29 patients, 15 men, 14 women; age range years, mean age 20.5). None of these 29 subjects had renal failure. The third group consisted of all the ADPKD young adults (younger than 30 years) who visited our department during the last 3 years, with renal failure (four patients, three men, one woman; age range years, mean age 26.5). Methods A complete medical history, physical examination, measure- ment of arterial blood pressure and serum creatinine and a complete abdominal ultrasound study were performed on these patients. Using ultrasound, the kidneys (kidney size, number of cysts per kidney (<6, 6 15, >15), cysts size, cortical echogenicity, corticomedullary differentiation), liver (echostructure, signs of hepatic fibrosis, presence of cysts) and biliary tree (dilatation) were specifically studied. All studies were performed and evaluated by a single consultant radiologist, with a Toshiba SSA 140 ( Toshiba, Japan) equipped with a 3.7 MHz transducer. The diagnosis of ARPKD and ADPKD was confirmed by linkage genetic studies in each case. Linkage analysis was performed by using microsatellite markers flanking the PKHD1 (D6S269, 272, 436, 427, 465, 423, 466, 295, 294, 257), PKD1 (AC2.5- D16S291, KG8-PKD1 and CW2-D16S663) and PKD2 (D4S423, D4S1534 and D4S1542) genes as previously described [7,8]. We obtained informed consent from all subjects who participated in the study. The study was approved by the Ethical Committee of the Hospital. C. Nicolau et al. inal ultrasound examinations in both parents and two sisters were normal. Case 2. A 19-year-old woman underwent a blood test due to asthenia and amenorrhoea during the previous 3 months. The blood analysis showed clear signs of ESRD and the patient underwent haemodialysis. There were no clinical signs of portal hypertension. The patient s younger brother had died at 2 months of age from ARPKD. Results of abdominal ultrasound examinations of both parents and another brother were normal. Case 3. This 20-year-old woman had been diagnosed of ARPKD at birth. She was known to have severe portal hypertension and renal failure in her teens. At 18 years of age she had gone to the emergency room because of upper digestive bleeding due the rupture of oesophageal varices. The intensity of haemorrhage precluded treatment with sclerosis, so she underwent a portocaval shunt that prevented any subsequent variceal bleeding. The patient s renal function is severely impaired (serum creatinine of 6 mg/dl ) but she still does not need renal replacement therapy. Results of abdominal ultrasound examinations of both parents, two sisters, and one brother were strictly normal. Case 4. A 22-year-old woman was diagnosed of ARPKD at birth. At 5 years of age she had upper digestive bleeding caused by rupture of oesophageal varices. After various bleeding episodes she underwent a splenorenal shunt. On admission, her renal function was impaired and serum creati- nine levels were of 2.8 mg/dl. Her brother and parents showed normal abdominal sonography scans. Group 2. ADPKD young adults None of the 29 ADPKD patients in this group had renal failure. Five of them had hypertension. None of them had a history of haematuria or renal calculi. They were initially examined in our department because of family history of ADPKD. Group 3. ADPKD patients <30 years old with renal failure All four of these patients had hypertension. Two of them entered ESRD at 28 and 26 years of age. The other two, who were 24 and 25 years old, had serum creatinine levels of 6 and 4mg/dl respectively. Results Comparison of ultrasonographic findings of ARPKD and ADPKD subjects are shown in Table 1. Clinical data Group 1. ARPKD patients All subjects were products of nonconsanguineous marriages. Group 1. ARPKD patients Linkage analysis of the four ARPKD families was Case 1. A 20-year-old man was admitted to our hospital consistent with ARPKD linked to the PKHD1 locus. because of biochemical features of advanced renal failure. His medical history included palpable masses in both flanks Renal sonography revealed multiple (more than 15) and hepatomegaly at birth. At that time, an exploratory small cysts (smaller than 1.5 cm) in a normal-sized laparotomy had showed enlarged kidneys with multiple cysts. kidney, and loss of corticomedullary differentiation in His growth and development were apparently normal. There all patients with ARPKD (Figure 1). In case 2 there was no family history of renal disease. The patient had been were also two cysts bigger than 1.5 cm in each kidney. apparently well until 3 months earlier, when he began to Increase of the cortical echogenicity was studied and experience fatigue, anorexia, weight loss, and muscle cramps. demonstrated in cases 2 and 4. The other two cases On admission he presented with end-stage renal disease had both kidneys massively full of cysts, preventing ( ESRD) and non palpable kidneys, and he underwent haemo- the examination of cortical echogenicity. In all cases dialysis. Due to the sonographic findings and the presence there were signs of hepatic fibrosis (confirmed by liver of pancytopenia, an endoscopy was indicated. Endoscopy excluded oesophageal varices, and the patient refused a biopsy in case 1). In case 1 hepatosplenomegaly and haemodynamic study to evaluate portal hypertension. The diffuse heterogeneous liver echostructure were present. patient underwent a liver biopsy that showed periportal No duct ectasia were seen. In cases 2 and 4 there were fibrosis. After the procedure the patient presented with severe multiple diffuse echogenic areas scattered throughout hepatic bleeding that required surgery. Results from abdom- the liver and compatible with hepatic periportal thick-

3 Sonography of polycystic kidney disease in young adults 1375 Table 1. Ultrasonographic findings in young adults with ARPKD or ADPKD ARPKD ADPKD ADPKD (n=4) without renal failure with renal failure (n=29) (n=4) Nephromegaly 0 7 (24.1%) 4 (100%) Renal cysts (n): <6 0 2 (6.8%) (58.6%) 0 >15 4 (100%) 10 (34.4%) 4 (100%) Size of the biggest cyst (cm): <1.5 3 (75%) 4 (13.8%) 0 >1.5 1 (25%) 25 (86.2%) 4 (100%) Increased cortical echogenicitya 2/2 (100%) 0/22 Loss of corticomedullary differentiation 4 (100%) 5 (17.2%) 4 (100%) Presence of hepatic cysts 0 6 (20.6%) 4 (100%) Signs of hepatic fibrosis 4 (100%) 0 1 (25%) Caroli s disease 2 (50%) 0 1 (25%) ait was not possible to evaluate cortical echogenicity in two ARPKD subjects, in seven ADPKD subjects without renal failure, and in all ADPKD subjects with renal failure because both kidneys were massively full of cysts. Fig. 1. A longitudinal ultrasound scan of the right kidney in a 20- year-old man with ARPKD shows a normal-sized kidney with multiple small cysts, and loss of corticomedullary differentiation. Fig. 2. An ultrasound scan of the right lobe of the liver in a 19- year-old woman with ARPKD shows multiple echogenic areas scattered throughout the liver. These areas represent the thickened fibrosed portal tracts that are characteristics of hepatic fibrosis. ening ( Figure 2). In case 3, hepatosplenomegaly was present with increase of the hepatic echogenicity and increase of cortical echogenicity was detected in any multiple intrahepatic duct ectasia showing a characteristic patient in whom it was possible to evaluate this (not echogenic central dot corresponding to fibrovas- evaluated in seven patients because the kidneys were cular bundles containing a portal vein radicle and an almost completely full of cysts). Simple hepatic cysts accompanying branch of the hepatic artery (confirmed were found in six subjects (20.6%) ( Figure 5). No by Doppler ultrasound) compatible with Caroli s disease signs of hepatic fibrosis, portal hypertension or bile (Figure 3). In case 4, there were also intrahepatic ducts ectasia were seen in any patient. duct ectasia. No simple hepatic cysts were observed in any patient. Group 3. ADPKD patients with renal failure Group 2. ADPKD patients without renal failure Linkage analysis was consistent with ADPKD linked to the PKD1 gene in each patient. Moderate or severe Linkage analysis of all the families was consistent with nephromegaly (at least >15 cm) was seen in all sub- ADPKD linked to the PKD1 gene. Nephromegaly was jects. The parenchyma was almost completely replaced only seen in seven patients (24.1%) while kidney size by innumerable cysts of different sizes, distributed was normal in 22 (75.9%) ( Figure 4). Loss of cortico- throughout all the renal parenchyma with loss of medullary differentiation was detected in only five corticomedullary differentiation, not allowing the patients (all of them having nephromegaly). No examination of the cortical echogenicity. There were

4 1376 C. Nicolau et al. simple hepatic cysts in all cases. In one case we also detected multiple intrahepatic duct ectasia compatible with Caroli s syndrome as well as signs of hepatic fibrosis, with cavernous transformation of the portal vein. ARPKD is characterized by cystic dilatation of the renal collecting ducts and hepatic biliary dysgenesis with periportal fibrosis [2,3,9]. For those patients who survive the perinatal period, the chance of being alive at 15 years of age is considered to be from 50 to 80% [10]. In one of the most representative prognostic analyses reported, 55 ARPKD patients were retrospectively reviewed; 23 patients (42%) presented within the first month of life, and 12 of them (about 50%) survived beyond 2 years of age; of this selected study group survival rates revealed that 46% survived to at least 15 years of age [11]. In a retrospective study that included 14 patients in whom the diagnosis of ARPKD was made before 2 weeks of life, nine children (64%) were still alive at the moment of the report: five of them remained with normal serum creatinine (aged between 8 months and 7 years), while four presented with renal insufficiency (aged between 7 and 14 years) [12]. Additionally, there are some case reports of lateonset symptoms in ARPKD patients who survived into adulthood [7,13]. The differentiation of ARPKD from ADPKD is important for genetic counseling and prognosis. A patient with ARPKD is not going to pass the disease to the following generation while the children of an ADPKD patient have a 50% chance of suffering from the disease. A positive family history is the most important information in establishing the diagnosis of ADPKD, and in most cases, ultrasonography of both Fig. 3. An ultrasound scan of the right lobe of the liver in a 20- year-old woman with ARPKD shows a typical case of Caroli s disease with saccular dilatations of the bile ducts containing intralum- inal bulbar protrusions and cross-bridges. Arterial flow inside a bulbar protrusion of a ductal ectasia was detected using Doppler ultrasound. Discussion Fig. 4. Longitudinal ultrasound scan of the left hepatic lobe and of the right kidney in a 19-year-old man with ADPKD shows a branch of the left portal vein in a normal liver with homogeneous echostructure and a normal-sized kidney with multiple cortical cysts of parents is a reliable way to distinguish between different sizes. The renal corticomedullary differentiation and cortical ADPKD and ARPKD. However, when paternity is in echogenicity remain conserved. doubt, or in cases of spontaneous mutations (accounting for up to 10% of ADPKD cases) parental sonograms are irrelevant. The linkage technique is also useful in diagnosing both entities. However, it is an expensive technique not available in most hospitals, not useful in sporadic cases, and frequently not informative in small families. When mutational detection in the PKD genes becomes available the differential diagnosis of these diseases will become easier. Until then, in these rare cases, a classification problem arises, and the diagnosis must be performed using radiological methods or biopsy of the involved organs [13,14]. Computed tomography (CT) and magnetic resonance imaging ( MRI) have been used with success in the evaluation of ARPKD patients [14 16], especially in older children who are able to co-operate during the performance of the test. These techniques can delineate fine details of the renal architecture and may add useful information to the ultrasound data. However, Fig. 5. Simple hepatic cysts in a 20-year-old woman with ADPKD. Ultrasound scan of the liver showed two simple cysts in a liver with normal echostructure. the intravenous contrast used in CT scans can be dangerous for patients with renal failure. Thus, for

5 Sonography of polycystic kidney disease in young adults 1377 these patients, ultrasound is the most useful non- between ADPKD and ARPKD. In patients with invasive diagnostic tool. ARPKD whose disease becomes manifest in later The sonographic findings in children with ARPKD childhood, hepatic disease is more severe than in early are well known. Sonography reveals diffusely increased onset [5,18,19]. ARPKD is always associated with cortical echogenicity when compared with the hepatic hepatic fibrosis and may occasionally be associated parenchyma as well as loss of the corticomedullary with Caroli s disease. Liver histopathology reveals bilidifferentiation [17]. Kidney size in ARPKD patients ary dysgenesis or ductal plate malformation, characpeaks at 1 2 years of age, gradually becoming smaller terized by the fibrous enlargement of the portal tracts and stabilizing by 4 5 years of age, while echogenicity or by portal portal bands of fibrous tissue containing tends to normalize over time [17]. Furthermore, sec- abnormal bile ducts. In addition, various segments of ondary cortical cyst formation can be found in older the intrahepatic bile ducts may be dilated, producing children and young adults with ARPKD, producing a bile stasis and cholangitis. The sonographic findings pattern that has been described as being similar to that observed in the four patients with ARPKD in our seen in sonograms of ADPKD patients [9,17]. study represent the spectrum of hepatic involvement Nevertheless, our study suggests that these sonographic of the disease. The sonographic appearance of hepatic findings may be different enough in comparison with fibrosis can show increase of the echogenicity and those found in ADPKD patients to allow a correct diffuse heterogeneous structure, but the visualization diagnosis. of multiple diffuse echogenic areas that represent the The loss of the corticomedullary differentiation, thickened fibrosed portal tracts encompassing portal increase of cortical echogenicity, and presence of mul- tributaries is characteristic. This finding is non-specific tiple cyst smaller than 1.5 cm throughout the normal- because it has been described in other diseases such as sized kidney are common findings in young adults with schistosomiasis, hepatitis, etc. [21]. In ARPKD ARPKD. The basic pathological lesion in ARPKD is patients thickened fibrosed portal tracts reveal abnorat the level of the collecting ducts. The increased mal development of the biliary tree. The sonographic echogenicity is due to the interfaces produced by the appearance of Caroli s disease is characteristic: saccular walls of ectatic tubules that are too small to be resolved dilatations of bile ducts, which may contain intralumiby sonography [18]. Jain et al. [19] explained that the nal bulbar protrusions and cross-bridges, as the result peripheral layer of cortex remains free of cysts because of an encasement of portal radicles by areas of ductal the peripheral cortex does not have any collecting or ectasia [ 22]. Colour Doppler discloses hepatic arterial connecting ducts. Nevertheless our study does not and portal venous flow within the fibrovascular projeccorroborate this finding in any ARPKD patient tions in the bile ducts [23]. Demonstration of communbecause the peripheral cortex also had cysts, probably ication between sacculi and bile ducts is important in due to secondary cortical cyst formation in older distinguishing Caroli s disease from polycystic liver children and young adults. disease [24]. Our results indicate that the presence of In ADPKD patients multiple or discrete cysts of a polycystic kidney in a young adult with hepatic varying sizes may be seen scattered throughout the cortex, including the subcapsular location, because fibrosis or Caroli s disease is highly suggestive of cysts in ADPKD can arise from anywhere in the ARPKD. In rare cases without any clear clinical signs nephron or collecting system. The natural history of of hepatic fibrosis, or without typical findings in the ADPKD kidneys is well established, with a correlation ultrasound study, a liver biopsy can be useful [13] in between the sonographic findings (number and size of demonstrating congenital fibrosis. Only one of the cysts and size of the kidney) and the age and/or severity patients presented in this study underwent a liver of the disease [20]. As the diseases progresses, the size biopsy. We did not perform this procedure in the other and number of cysts and the renal volume increases. patients, because liver biopsy is an invasive procedure Nevertheless, we have found a high incidence of with a risk of complications (two patients had multiple normal-sized kidneys (75.9%) in young adults with and large intrahepatic duct ectasia), and linkage ana- ADPKD (without the marked increase in kidney size lysis was consistent with ARPKD in all cases. described in other reports) [17,18] and less than 15 However, ADPKD cannot be completely excluded in cysts per kidney ( 65.4%), probably because they were patients with polycystic kidneys and hepatic fibrosis asymptomatic subjects. In this age group, the renal because this association has been reported in ADPKD cortex is almost always conserved, unlike with patients [25,26]. Of the 400 ADPKD subjects studied ARPKD, except in patients with more advanced forms in our department in the last 3 years, we have found of ADPKD in which there is moderate or severe this association in only one case (a subject of the group nephromegaly and the cortex is barely visible. of ADPKD with renal failure), thus confirming that it Detecting nephromegaly can help in differentiating is a rare association. On the other hand, the presence ADPKD from ARPKD, as in the four ADPKD subadult of a normal liver or simple hepatic cysts in a young jects with renal failure in our study. Moreover, in with polycystic kidneys is highly suggestive of general, ADPKD patients with renal failure present ADPKD [3]. Simple hepatic cysts are a common with larger cysts than those in ARPKD patients. feature in ADPKD adults and in 20% of ADPKD Our results also indicate that the sonographic evaluation of the liver is very useful in differentiating patients younger than 30, but are extremely rare in ARPKD patients.

6 1378 C. Nicolau et al. In conclusion, the evaluation of the sonographic of autosomal-recessive polycystic kidney disease in two sisters. features of the kidneys and those of the liver may help Am J Nephrol 1988; 8: Kern S, Zimmerhackl LB, Hildebrandt F, Uhl M. Rare-MRin the differential diagnosis between ARPKD and urography a new diagnostic method in autosomal recessive ADPKD in early adulthood. polycystic kidney disease. Acta Radiol 1999; 40: Howie JL, Nicholson RL. CT evaluation of infantile polycystic disease. J Can Assoc Radiol 1980; 31: References 16. Berger PE, Munschauer RW, Kuhn JP. Computed tomography and ultrasound of renal and perirenal diseases in infants and 1. Bosniak MA, Ambos MA. Polycystic kidney disease. Semin children. Relationship to excretory urography in renal cystic Roentgenol 1975; 10: disease, trauma and neoplasm. Pediatr Radiol 1980; 9: Zerres K, Rudnik-Schoneborn S, Steinkamm C, Mucher G. 17. Blickman JG, Bramson RT, Herrin JT. Autosomal recessive Autosomal recessive polycystic kidney disease. Nephrol Dial polycystic kidney disease: Long-term sonographic findings in Transplant 1996; 11 [Suppl 6]: patients surviving the neonatal period. Am J Roentgenol 1995; 3. Gang DL, Herrin JT. Infantile polycystic kidney disease of the 164: liver and kidneys. Clin Nephrol 1986; 25: Grossman H, Rosenberg ER, Bowie JD, Ram P, Merten DF. 4. Welling LW, Grantham JJ. Cystic and developmental diseases Sonographic diagnosis of renal cystic diseases. Am J Roentgenol of the kidney. In. Brenner BM (ed.). The Kidney, 5th edn. 1983; 140: Saunders, Philadelphia, 1996; Jain M, Le Quesne GW, Bourne AJ, Henning P. High-resolution 5. Blyth H, Ockenden BG. Polycystic disease of kidneys and liver ultrasonography in the differential diagnosis of cystic diseases presenting in childhood. J Med Genet 1971; 8: of the kidney in infancy and childhood: preliminary experience. 6. Potter EL. Normal and Abnormal Development of the Kidney. J Ultrasound Med 1997; 16: Year Book Medical Publishers, Chicago Gabow PA. Autosomal dominant polycystic kidney disease 7. Perez L, Torra R, Badenas C et al. Autosomal recessive poly- More than a renal disease. Am J Kidney Dis 1990; 5: cystic kidney disease presenting in adulthood. Molecular diaperiportal 21. Medhat A, Nafeh M, Swifee Y et al. Ultrasound-detected hepatic gnosis of the family. Nephrol Dial Transplant 1998; 13: thickening in patients with prolonged pyrexia. AmJ 8. Torra R, Badenas C, Darnell A et al. Linkage, clinical features Trop Med Hyg 1998; 59: and prognosis of autosomal dominant polycystic kidney disease 22. Marchal GJ, Desmet VJ, Proesmans WC et al. Caroli disease: types 1 and 2. J Am Soc Nephrol 1996; 7: high-frequency US and pathologic findings. Radiology 1986; 9. Guay-Woodford L, Galliani CA, Musulman-Mroczek E, Spear 158: GS, Guillot AP, Bernstein J. Diffuse renal cystic disease in 23. Gorka W, Lewall DB. Sonography in the assessment of patients children: morphologic and genetic correlations. Pediatr Nephrol with Caroli s disease. J Clin Ultrasound 1998; 26: ; 12: Miller WJ, Sechtin AG, Campbell WL, Pieters PC. Imaging 10. Gagnadoux MF, Habib R, Levy M et al. Cystic renal disease findings in Caroli s disease. Am J Roentgenol 1995; 165: in children. Adv Nephrol 1989; 18: Torra R, Badenas C, Darnell A, Bru C, Escorsell A, Estivill X. 11. Kaplan BS, Fay J, Shah V. Autosomal recessive polycystic Autosomal dominant polycystic kidney disease with anticipation kidney disease. Pediatr Nephrol 1989; 3: and Caroli s disease associated with a PKD1 mutation. Kidney 12. Deget F, Rudnik Schöneborn S, Zerres K. Course of autosomal Int 1997; 52: recessive polycystic kidney disease in siblings: a clinical comparhepatic 26. Cobben J, Breuninng M, Kate CSL, Zerres K. Congenital ison of 20 sibships. Clin Genet 1995; 47: fibrosis in autosomal dominant polycystic kidney disease. 13. Neumann HPH, Zerres K, Fischer CL et al. Late manifestation Kidney Int 1990; 38: Received for publication: Accepted in revised form: