Abstract Autosomal recessive polycystic kidney dis-ease (ARPKD) is a relatively common form of pediatricpolycystic kidney disease with an incidence of 1:20,000live births. Previous reports, primarily from populationsof European origin, indicate that the clinical presentationand disease course are quite variable. Using a retrospec-tive study design, we sought to determine whether theclinical course and outcome of our multi-ethnic patientcohort differs from the published literature. A 10-year(1990–2000) retrospective study was conducted in whichwe reviewed the clinical, histopathological, and imagingrecords of our 31 ARPKD patients. Patients were diag-nosed between 0 and 14 years of age, with 17 (55%) pre-senting within the 1st month of life. The mean follow-upwas 67 months and age at last follow-up ranged from 0.5to 16 years. Of the 17 patients diagnosed as neonates, 11(65%) had respiratory insufficiency complicated bypneumothoraces. Two died shortly after birth and 2 diedwithin the 1st year of life due to respiratory failure.Among the 13 neonatal survivors, 7 (54%) developedprogressive renal insufficiency, whereas 6 of 14 (43%) of

those children who presented beyond 1 month of age de-veloped renal insufficiency. Hypertension was present in55% of our patients, with nearly all neonatal survivorsrequiring antihypertensive management. Evidence ofportal hypertension was found in 10 (37%) of the 27 pa-tients who survived the 1st year of life. In our multi-eth-nic ARPKD cohort, the 1-year survival rate (87%) andthe clinical variability are comparable to those previous-ly reported. With the recent identification of the PKHD1gene, characterization of disease-causing mutationsshould provide new insights into the molecular basis forthis phenotypic variability.

Keywords Autosomal recessive polycystic kidney disease · Phenotypic variability · Multi-ethnic population

Introduction

Autosomal recessive polycystic kidney disease (ARPKD)is an often-devastating form of polycystic kidney diseasewith an estimated incidence of 1 in 20:000 live births [1].The typical onset occurs in neonates and infants and ischaracterized by cystic dilatation of the renal collectingducts and dysgenesis of the biliary ductal plate.

Blyth and Ockenden [2] were the first to recognizeand characterize the phenotypic variability in ARPKD.They stratified patients into four phenotypic groups ac-cording to the age of presentation, the proportion of di-lated renal collecting ducts, and the degree of biliary fi-brosis. They concluded that within a given sibship thephenotypic manifestations were uniform, and thus theyhypothesized that ARPKD involved mutations in fourdistinct genes. However, re-evaluation of their cases castdoubt on the validity of this rigid classification [3].Moreover, while disease expression among affected sib-lings is often similar, several reports have described sib-ships with discordance in disease onset and clinical phe-notypes [4, 5, 6].

These observations prompted Zerres et al. [7] andothers to propose that ARPKD is caused by mutations in

R. Capisonda · V. Phan · J. W. BalfeDivision of Nephrology, Department of Pediatrics,The Hospital for Sick Children,Toronto, Ontario M5G1X8 Canada

J. Traubuci · A. DanemanDepartment of Diagnostic Imaging,The Hospital for Sick Children,Toronto, Ontario M5G 1X8 Canada

L. M. Guay-WoodfordDivision of Genetic and Translational Medicine,University of Alabama at Birmingham,Birmingham, AL 35294 USA

L. M. Guay-Woodford (✉)Division of Genetic and Translational Medicine, Departments of Medicine and Pediatrics,University of Alabama at Birmingham, Kaul 702,1530 3rd Avenue South 19th Street, Birmingham, AL 35294 USAe-mail: lgw@uab.eduTel.: +1-205-9347308; Fax: +1-205-9755689

Pediatr Nephrol (2003) 18:119–126DOI 10.1007/s00467-002-1021-0

O R I G I N A L A RT I C L E

Rhona Capisonda · Veronique PhanJeffrey Traubuci · Alan DanemanJ. Williamson Balfe · Lisa M. Guay-Woodford

Autosomal recessive polycystic kidney disease: outcomes from a single-center experience

Received: 5 July 2002 / Revised: 11 September 2002 / Accepted: 2 October 2002 / Published online: 21 January 2003© IPNA 2003

a single gene. Genetic studies have confirmed that ARPKD involves a single gene, PKHD1, which maps tothe short arm of chromosome 6 (6p21-p12) [8]. The fullspectrum of phenotypic variants appears to involve mu-tations in this gene [9]. With the recent identification ofthe PKHD1 gene, characterization of disease-causingmutations should provide new insights into the molecu-lar basis for this phenotypic variability [10, 11].

Despite these recent genetic advances, the clinicalcourse of ARPKD patients remains a matter of debate.Initial reports stated that most children with ARPKD diein infancy [2, 12, 13]. However more recent reports in-volving cohorts of patients primarily of European origin[14, 15, 16, 17, 18] indicate a better long-term prognosis.

Therefore, we undertook a retrospective analysis ofall ARPKD patients referred to our Canadian tertiarycenter. We have identified 31 ARPKD patients represent-ing different ethnic groups, and in this report we describetheir clinical presentation, disease course, and survivalcharacteristics.

Patients and methods

A retrospective study was conducted on data previously gatheredin a prospective fashion for all patients with ARPKD diagnosed atthe Hospital for Sick Children from 1990 through 2000. Only pa-tients diagnosed with ARPKD by 18 years of age were included.Initially we consulted the Health Department Database to identifyall children diagnosed with congenital renal anomalies (ICD-9-CM).We then examined these medical records to identify patients withenlarged, echogenic kidneys without gross cysts who met at leasttwo additional diagnostic criteria for ARPKD: (1) evidence of bili-ary fibrosis and/or portal hypertension, (2) normal renal ultraso-nography in both parents (available in 5), (3) compatible histopa-thology in renal biopsy, nephrectomy, or postmortem specimens(available in 13 patients), (4) history of ARPKD in a sibling [17].Patients were excluded if they had urinary tract malformations ormajor congenital anomalies of other systems suggesting a diagno-sis other than ARPKD.

All clinical records and histopathological reports were exam-ined. All the imaging studies were reviewed in a blinded fashion(J.T., A.D.). Serial changes in the kidneys, liver, and spleen wereassessed sonographically. Progressive changes in renal length, cor-tical echotexture, pelvicalyceal dilatation, renal calcifications, aswell as the evolution of macroscopic cysts were evaluated. Theage of presentation, growth parameters, age at the initiation of an-tihypertensive medication, as well as the age and cause of death, ifapplicable, were recorded. Hypertension was defined as averagesystolic and/or average diastolic blood pressure (BP) ≥95th per-centile for age and sex adjusted to height and weight measured on

at least three occasions [19]. Serial studies of renal and liver function were obtained. The glomerular filtration rate (GFR) wasassessed as creatinine clearance estimated using the method of Schwartz, adjusted for age [20, 21]. Mild, moderate, and severe renal insufficiency were defined as estimated GFR of50–80 ml/min per 1.73 m2, 25–50 ml/min per 1.73 m2, and lessthan 20 ml/min per 1.73 m2, respectively. The presence of portalhypertension was based on sonographic criteria, including visual-ization of portal vein, splenomegaly, hepato-fugal blood flow, andthe presence of varices.

Kaplan-Meier analysis in SPSS version 8.0 (Chicago, Ill.,USA) was used to determine survival and the survival probabilityof hypertension and end-stage renal disease (ESRD).

Results

Clinical presentation

Thirty-one patients (14 males and 17 females) were diag-nosed with ARPKD between 1990 and 2000. The ethnicdistribution within this cohort was dominated by Cauca-sian children of European descent (n=22, 71%). Howev-er, other ethnic and racial groups were also represented,with 2 children (6%) of East Indian origin, 1 (3%) of Af-rican descent, and 6 (19%) from different Asian groups.

The mean age of diagnosis was 7 years, ranging frombirth to 14 years of age (Tables 1 and 2). Seventeen pa-tients (55%) were diagnosed within the 1st month of life.Of these, 10 children (59%) were diagnosed antenatallybetween the 24th and 35th weeks of gestation. Fetalsonograms revealed enlarged, echogenic kidneys and athigh resolution, multiple tiny renal cysts.

Among the infants diagnosed before 1 month of age,the most frequent clinical presentation (n=11) was respi-ratory insufficiency complicated by pneumothoraces.Two neonates died shortly after birth. The remaining 9required mechanical ventilation, and 2 of these infantsdied during the 1st year of life from respiratory compli-cations. Among older children, i.e., those presenting after 1 month of age, the most frequent presenting signswere hypertension and hepatomegaly (Tables 1 and 2).

The diagnosis was made in 13 children (42%) by his-topathological analysis of renal and/or liver tissues ob-tained by biopsy, nephrectomy, or autopsy. Three fami-lies have 2 affected children and two other families had 3affected siblings. In the latter group, histopathologicaldiagnosis was based on a sibling autopsy.

120

Table 1 Clinical characteristicsat presentation Events 0–1 month >1 month to 1 year >1 year to 5 years >5 years

(n=17) (n=6) (n=3) (n=5)

Antenatal diagnosis 10Family history 5Pneumothorax 11Flank mass 7 2Hepatomegaly 1 1 3Hypertension 2 3 1Renal insufficiency 8 0 0 3Urinary tract infection 1Portal hypertension 1 1

Sonographic features

Twenty-seven patients had two or more sonograms avail-able for review. The initial sonogram was performed inpatients ranging from 1 day to 14 years of age. The timeinterval between the initial and the last investigationranged from 4 months to 14 years, with an average of42 months.

Increased renal size was observed in 12 of 13 patientswho maintained normal renal function. In this group, re-nal calcifications were observed in 6 patients initiallyand 3 additional patients at follow-up (Fig. 1). Six pa-tients had a relatively sonolucent cortical rim (Fig. 1);pelvicalyceal dilatation was present in 5 children, andmacroscopic cysts were observed in 11 subjects at thelast observation. Of the 5 patients who developed ESRD,3 had renal calcifications on initial evaluation, with anadditional patient manifesting renal calcifications on fol-low-up examination. A relatively sonolucent cortical rimwas observed in 2 of these patients and 1 child had pelvi-calyceal dilatation. All 5 children had macroscopic cystsevident at the last observation.

In total, we observed dilatation of the pelvicalycealsystem in 8 (26%) patients at initial evaluation. An addi-

tional 5 patients developed dilatation, giving a total of 13 (42%) by the end of the study. Voiding cystourethro-grams were performed in 6 of these patients. Only 1 hadevidence for vesicoureteral reflux. The other 5 had nor-mal studies.

Among the 27 subjects, a coarse liver echotexture wasobserved in 18 patients on their first sonogram and in 22patients at follow-up evaluation. Intrahepatic biliary ductdilatation, consistent with Caroli disease, was seen in 8 (26%) patients initially. By the end of the study period,this was observed in 12 (44%) of the 27 surviving pa-tients.

Co-morbidities

Systemic hypertension

Seventeen (55%) patients developed systemic hyperten-sion requiring antihypertensive therapy. Of the 13 sur-viving neonates, 11 (85%) developed hypertension, withan age of diagnosis ranging from 4 days to 36 months.Six older patients had hypertension at initial presenta-tion. Of the 17 hypertensive children, 8 (47%) required

121

Table 2 Clinical outcome (GFR glomerular filtration rate, PD peritoneal dialysis, NA not available)

Patient Age at Follow-up Diagnosis of GFR GFR Outcome Age and cause(gender) diagnosis months) hypertension (at diagnosis) (at last of death

(years) (follow-up)

1 (F) 13 30 14 13 Renal transplant2 (F) 0.1 91 55 155 Alive3 (M) 6 6 6 years 44 10 PD4 (M) Antenatal 82 8 months 34 124 Alive5 (M) Antenatal 82 17 days <10 68 Alive6 (M) 0.9 82 79 122 Lost to follow-up7 (M) 0.2 170 2 months 50 29 Alive8 (M) At birth 200 19 days 41 12 Liver/renal transplant9 (M) Antenatal 5 11 25 Died 5 months; respiratory

failure10 (M) 0.4 140 8 months 105 63 Alive11 (M) 5 45 6 years 148 133 Died 9 years; sepsis12 (F) At birth NA 13 NA Died 15 h; respiratory failure13 (F) 0.8 148 90 78 Liver transplant14 (M) 9 76 14 17 Renal transplant15 (F) Antenatal 109 10 days 26 103 Alive16 (F) 5 6 127 NA Alive17 (M) 7 12 108 93 Alive18 (F) 0.8 46 10 months 79 103 Alive19 (M) At birth 170 10 days 21 <10 Renal transplant20 (M) Antenatal 6 1 month 23 85 Alive21 (M) 3 85 148 143 Alive22 (F) At birth 183 4 days 67 <10 Alive23 (F) At birth NA 22 NA Died 16 h; respiratory failure24 (M) Antenatal 7 26 days 64 49 Died 7 months; respiratory

failure25 (F) 8 10 134 130 Alive26 (M) At birth 22 9 days 72 41 Alive27 (F) Antenatal 77 3 years 90 135 Alive28 (F) Antenatal 59 7 months 62 78 Alive29 (M) Antenatal 17 8 months 69 84 Alive30 (F) Antenatal 52 86 154 Alive31 (F) At birth 78 85 122 Alive

GFR rate expressed as ml/min per 1.73 m2 and calculated using the method of Schwartz [17, 18]

multi-agent therapy for blood pressure (BP) manage-ment, including angiotensin converting enzyme inhibi-tors (ACEI), calcium channel blockers, vasodilators, betablockers, and diuretics. BP control, e.g., BP less than90th percentile for sex, age, and height [19], wasachieved in 6 patients with monotherapy, ACEI in 5 chil-dren, and calcium channel blockers in 1 child. Hyperten-sion resolved in 3 patients who presented at less than2 months of age. Two were 7 years old and 1 was14 years of age at last contact. Life-table survival analy-sis showed the probability of requiring antihypertensivemedication was 50% at 5 years and 58% at 15 years.

Chronic renal insufficiency

The 7 (54%) neonatal survivors who presented withspontaneous pneumothorax at birth rapidly developed re-nal insufficiency. Four (31%) neonates developed mildrenal insufficiency, with a GFR between 50 and80 ml/min per 1.73 m2 and 1 has moderate renal insuffi-ciency, with a GFR of 41.3 ml/min per 1.73 m2 at lastfollow-up. Two patients required dialysis and subsequentrenal transplantation at 8 and 9 years of age, respective-ly. The latter had a combined liver and renal transplant.The remaining 6 (46%) neonatal survivors maintainednormal renal function throughout the observation period.Their ages ranged from 6 months to 10 years.

Of 6 patients who presented between 1 and 12 monthsof age, 50% continued with normal renal function, with

ages ranging from 5 to 9 years at last observation. Threepatients presented between 1 and 5 years of age. All arenormotensive. One of these patients died of sepsis sec-ondary to an invasive, gram-negative enterocolitis at9 years of age. This event was not thought to be relatedto his renal or biliary disease. His GFR at last follow-upwas 133 ml/min per 1.73 m2. Of the other 2 patients, 1 has normal renal function and the data are not avail-able for the other.

Of the 5 patients who presented after 5 years of age, 2 have normal renal function, 1 is currently on peritonealdialysis, and 2 have undergone a renal transplant.

Actuarial renal survival of the 26 survivors showedthe probability of survival without ESRD was 100% at5 years of age and 18% at 15 years of age.

Portal hypertension

Hepatomegaly was documented in 16 (52%) patients andsplenomegaly in 11 (36%). Sonographic evidence forportal hypertension, characterized by visualization ofportal vein, splenomegaly, hepato-fugal blood flow, andthe presence of varices, was noted in 10 (37%) of the 27 patients who survived the 1st year of life. Bleedingesophageal varices occurred in 3 patients, and all re-quired sclerotherapy to control hematemesis. Two pa-tients underwent liver transplantation at 3 and 11 yearsof age respectively, due to recurrent variceal bleedingthat was refractory to medical management and porto-systemic shunting. Both had sonographic evidence ofportal hypertension by 2 years of age.

Other complications

Hyponatremia was observed in 3 patients during the im-mediate postnatal period and all were hypertensive. Thiselectrolyte abnormality resolved by 1 month of age butthe hypertension persisted.

Urinary tract infection due to enteric organisms, e.g.,Escherichia coli, Enterobacter, and Klebsiella, was doc-umented in 6 patients, 4 girls and 2 boys. One of theboys had unilateral grade II vesicoureteral reflux.

Patient outcome

Of the 31 patients, 24 (77%) are alive and followed at our center. Their ages range from 6 months to16 years. One patient transferred to an adult center for continuing care and one was lost to follow-up. Four (13%) patients died during the 1st year of life secondary to respiratory complications, and 1 patientdied at age 9 years due to gram-negative sepsis. Themean duration of follow-up was 67 months (range6–200 months). Kaplan-Meier analysis indicated thatthe probability of survival was 87% at 1 year and 80%at 9 years of age (Table 3).

122

Fig. 1 Radiological findings associated with autosomal recessivepolycystic kidney disease. High-resolution renal sonography in asymptomatic 5-year-old girl reveals relatively sonolucent cortex(white arrowheads), dilated tubules (black arrow), and calcifica-tions (white arrow)

Discussion

This retrospective study represents one of the largest sin-gle-center series reported to date. We acknowledge thatour study design biased patient ascertainment towardsneonates who survived at least a few hours of life, andthus we cannot estimate the incidence of ARPKD amongall live-born infants in Canada. However, our datasetdoes allow us to describe the clinical characteristics anddisease course of 31 neonates and infants who were di-agnosed and managed within the last decade at The Hos-pital for Sick Children.

The 1-year survival rate of 87% in our series wascomparable to most previous studies [13, 14, 15, 16, 17](Table 3). Among the specific causes of death, respirato-ry failure and sepsis were the most common in our se-ries, as in previous reports.

In our cohort, 55% of the patients presented duringthe neonatal period. Neonatal respiratory distress or pal-pable, enlarged kidneys was the most common present-ing sign, consistent with the results from previous stud-ies.

The reported incidence of severe respiratory problemsin ARPKD neonates varies from 13% to 75% [5, 14, 22].In our experience, 11 (65%) neonates had severe respira-

tory distress and associated spontaneous pneumothora-ces. Four of these patients died in the 1st year of life.Previous studies [2, 5, 12] have suggested that severe re-nal functional impairment in the affected fetus leads todecreased fetal urine output and secondary oligohydram-nios. With the resulting pulmonary hypoplasia, most ofthe infants die from respiratory insufficiency shortly af-ter birth. Lonergan et al. [23] reported an inverse correla-tion between the volume of amniotic fluid and renal sizeand suggested that this correlation may predict both therisk of neonatal respiratory compromise as well as long-term renal outcome. While our retrospective dataset can-not address this issue directly, we found no correlationbetween renal size and respiratory function in our neona-tal cohort.

Cole et al. [14] observed that in ARPKD patients di-agnosed as neonates, renal clearances improved duringthe initial 2 years of life and remained stable for manyyears, until progressively declining during adolescence.We have made a somewhat similar observation in our co-hort. All 13 neonatal survivors had improvement in renalfunction. However, the 7 surviving neonates who pre-sented with spontaneous pneumothorax at birth subse-quently developed renal insufficiency. The other 6 neo-nates without significant respiratory complications have

123

Table 3 Comparison of our results with the published literature (Rx treatment, SCr serum creatinine, ESRD end-stage renal disease)

Length of Our series Kaariainen et al. Cole et al. Gagnadoux et al. Kaplan et al. Zerres et al.study 1990–2000 [13] [14] [15] [16] [17]

(10 years) 1974–1983 1970–1984 1962–1986 1950–1987 1987–1993(9 years) (14 years) (24 years) (37 years) (6 years)

Patients (n) 31 73 17 33 55 115(18 neonatalsurvivors)

Age at 32% prenatal 72% <1 month Diagnosis 1st 33% <1 month 42% <1 month 10% prenataldiagnosis year of life 41% <1 months

(inclusion criteria)

23% <1 month 6% ≤1 year 55% 1–18 months 42% 1–12 months19% 1–12 months 22% >1 year 12% 6–11 years 16% >1 year 23% 1–12 months26% >1 year 26% >1 year

Renal 51% GFR 82% (9/11) 35% GFR 42% GFR 58% SCr 72% GFR function <80 ml/min GFR <40 ml/min per <80 ml/min per >100 µmol/l <3rd percentile

per 1.73 m2 <90 ml/min per 1.73 m2 1.73 m2 for age16% ESRD 1.73 m2 29% ESRD 21% ESRD 10% ESRD

Hypertension 55% on drug Rx 61% (11/18) on 100% >95th 76% on drug Rx 65% on drug Rx 70% on drug Rxdrug Rx percentile or

drug RxEvidence of 37% 11% (2/18) 35% 39% 47% 46%portal hypertension and/or hepatomegalySurvival rate 1 year 87% 1 year 19% (14/73) 1 year 88% 1 year 91% 1 year 79% 1 year 94% (M),

82% (F)9 years 80% 10 years 51%

15 years 46% 3 years 94% (M), 79% (F)

Death rate in 13% 81% (59/73) 12% 9% 24% 9%the 1st year of life

maintained normal renal function. Therefore, we proposethat the coincidence of respiratory insufficiency andpneumothoraces requiring mechanical ventilation is apoor prognostic marker for long-term patient as well asrenal outcome.

It has been proposed that later-onset ARPKD is associ-ated with longer patient survival and, while renal functionmay decline, renal insufficiency often remains mild formany decades [14, 16, 18, 24]. In our cohort, 9 patientspresented between 1 month and 5 years of age. Seven ofthese patients had either stable or improved GFR at lastobservation, with a mean follow-up of 80 months (range45–148 months). While the other 2 patients experienced adecline in renal function, neither required renal replace-ment therapy. Among the 5 patients who presented after5 years of age, 1 required renal replacement therapy within 2 months and another patient presented with chron-ic renal insufficiency and progressed to ESRD within1 year. Another patient required renal replacement therapyat 16 years of age, 8 years after his initial diagnosis. Wetherefore agree with others [5, 15] that, contrary to the as-sertion first made by Blyth and Ockenden [2], there is nocorrelation between the age of ARPKD diagnosis and therate of progression of renal insufficiency.

Systemic hypertension was prevalent in our cohort,with 11 (85%) of the 13 neonatal survivors and 17 (55%)of the total cohort requiring antihypertensive medication.Numerous previous studies [15, 17, 18, 25] have report-ed that ARPKD-related hypertension is severe at onset,particularly in infants, but can decrease in severity oreven completely remit with disease progression. Wehave noted this same phenomenon. In 3 patients, appro-priate BP control was achieved with reduced doses ofantihypertensive agents, while in 3 others, treatment wasdiscontinued. In the latter group, each had mild-to-mod-erate renal insufficiency and moderately enlarged kid-neys at presentation. Each patient was normotensive by2 years of age, and 2 of the 3 had normal renal function.

The pathogenesis of systemic hypertension in ARPKDremains unclear. Previous investigators have suggestedthat volume overload associated with poor renal functionmay be a causal factor [15, 16]. Kaplan et al. [16] notedthat the onset of severe hypertension within the first fewmonths of life was associated with hyponatremia. Thissame association was evident in 3 of our ARPKD neo-nates. While there is compelling evidence that hyperten-sion in autosomal dominant polycystic kidney disease involves a renin-mediated mechanism [26], the limiteddata in ARPKD suggest that renin levels are low in hy-pertensive neonates, particularly in those with concomi-tant hyponatremia [18]. Therefore, while the currentstandard for hypertension management in ARPKD pa-tients includes ACEI and calcium channel blockers, the mechanism of action of these agents in controllingARPKD-related hypertension remains unclear.

Congenital hepatic fibrosis is an invariant feature ofARPKD, and in many patients leads to portal hyperten-sion and its sequelae. We identified portal hypertensionusing sonographic criteria in 37% of our patient popula-

tion, a prevalence consistent with previous reports (Table 3). We also noted that despite progressive portalhypertension, hepatocellular function was rarely altered,and only 2 (7%) of our patients had mild elevations inserum transaminase levels. One patient underwent acombined liver and renal transplant as his progressiveportal hypertension was refractory to porto-systemicshunting. Tsimaratos et al. [27] recommended combinedliver and renal transplantation in ARPKD patients withporto-systemic shunts who progress to ESRD, due to thesignificant risk of hepatic encephalopathy developing inthese patients.

We identified five sets of siblings in our cohort. Theclinical disease presentation in three families was concor-dant. The sibling pairs in each of these families presentedin the immediate perinatal period with respiratory insuffi-ciency and early onset hypertension. Clinical disease ex-pression was discordant in two other families. In the firstfamily, 1 sibling presented as a neonate with respiratoryinsufficiency, while the other 2 siblings were diagnosed at6 months of age due to the onset of hypertension. At lastobservation, both of these surviving sisters have main-tained normal renal function. In the second family, the in-dex patient was identified at 8 years of age followingevaluation for splenomegaly. His renal function subse-quently deteriorated and he underwent renal transplanta-tion at 15 years of age. His 2 older sisters were diagnosedduring early childhood by screening sonography and bothhave remained asymptomatic throughout the observationperiod. Therefore, our data are consistent with previousreports that suggest the clinical phenotypes, even amongARPKD siblings, can be quite variable [3, 6].

Our experience indicates that high-resolution sono-graphy is a robust imaging tool for the longitudinal eval-uation of ARPKD patients. Progressive increase in renalsize was observed in 12 of our 13 patients who continueto have normal renal function at the last observation, butalso in 3 of the 5 patients who progressed to ESRD. Ourdata support the assertion by Blickman et al. [28] thatthere is no correlation between renal length and renalfunction in ARPKD. However, we caution that neitherstudy measured renal volume, and we note that in ADPKD patients, recent data have suggested a directcorrelation between large renal volumes and more accel-erated renal disease [29].

Previous sonographic studies have reported asymmet-ric renal size [30], renal calcifications [31], and a rela-tively sonolucent cortical rim [32] in ARPKD patients.In our cohort, no asymmetry in renal size was observed,either at presentation or longitudinal follow-up. In addi-tion, we found no significant difference between patientswith normal renal function and those who progressed toESRD, with respect to the prevalence of renal calcifica-tions or a relatively sonolucent cortical rim. Therefore,our data do not support a specific correlation betweenclinical progression of ARPKD and sonographic evi-dence of either renal calcifications or a sonolucent corti-cal rim as proposed by others [31, 33]. In fact, our dataare consistent with the assertion by Currarino et al. [32]

124

that the sonolucent cortical rim may represent a zone ofmore severe tubular dilatation, and this finding does nothave prognostic relevance.

In our series, pelvicalyceal dilatation was observed in~40% of patients and appeared to be a primary structuralabnormality rather than a secondary consequence of co-incident vesicoureteral reflux. Therefore, we proposethat this feature should be added to the sonographic char-acteristics of ARPKD. However, based on our data, wedo not attribute any prognostic importance to the pres-ence of pelvicalyceal dilatation in ARPKD patients.

In summary, our study demonstrates that ARPKD hasa variable clinical presentation and clinical course in amulti-ethnic population. Furthermore, this study supportsthe favorable long-term prognosis for ARPKD patients.Our data indicate that in neonates, the coincidence ofrespiratory insufficiency and pneumothoraces requiringmechanical ventilation is a poor prognostic marker forboth patient and renal survival. This study did not identi-fy any other new prognostic markers for disease progres-sion or confirm the prognostic validity of sonographicfeatures as proposed by others. With the identification ofthe PKHD1 gene, mutational analyses may identify spe-cific phenotype-genotype correlations that could explainsome of the disease variability. If so, this genetic infor-mation may provide new possibilities for developingclinically relevant prognostic markers.

Acknowledgements The authors would like to thank the manyfamilies and their health care providers who have cooperated withthese studies.

References

1. Zerres K, Muecher G, Becker J, Steinkamm C, Rudnik-Schoneborn S, Heikkila P, Rapola J, Salonen R, Germino G,Onuchic L, Somlo S, Avner E, Harman L, Stockwin J, Guay-Woodford L (1998) Prenatal diagnosis of autosomal recessivepolycystic kidney disease (ARPKD): molecular genetics, clini-cal experience, and fetal morphology. Am J Med Genet76:137–144

2. Blyth H, Ockenden BG (1971) Polycystic disease of kidneyand liver presenting in childhood. J Med Genet 8:257–284

3. Kaplan BS, Kaplan P, Chadarevian J-P de, Jequier S, O’ReganS, Russo P (1988) Variable expression of autosomal recessivepolycystic kidney disease and congenital hepatic fibrosis with-in a family. Am J Med Genet 29:639–647

4. Chilton SJ, Cremin BJ (1981) The spectrum of polycystic dis-ease in children. Pediatr Radiol 11:9–15

5. Gang DL, Herrin JT (1986) Infantile polycystic disease of theliver and kidneys. Clin Nephrol 25:28–36

6. Deget F, Rudnik-Schoneborn S, Zerres K (1995) Course of au-tosomal recessive polycystic kidney disease (ARPKD) in sib-lings: a clinical comparison of 20 sibships. Clin Genet47:248–253

7. Zerres K, Volpel MC, Weiss H (1984) Cystic kidneys: genet-ics, pathologic anatomy, clinical picture, and prenatal diagno-sis. Hum Genet 68:l04-l35

8. Zerres K, Muecher G, Bachner L, Deschennes G, EggermannT, Kaariainen H, Knapp M, Lenner T, Misselwitz J, Muhlendahl K, Neumann N, Pirson Y, Rudnik-Schoneborn S,Steinbicker V, Wirth B, Scharer K (1994) Mapping of the genefor autosomal recessive polycystic kidney disease (ARPKD)to chromosome 6p21-cen. Nat Genet 7:429–432

9. Guay-Woodford LM, Muecher G, Hopkins SD, Avner ED,Germino GG, Guillot AP, Herrin J, Holleman R, Irons DA,Primack W, et al. (1995) The severe perinatal form of autoso-mal recessive polycystic kidney disease maps to chromosome6p21.1-p12: implications for genetic counseling. Am J HumGenet 56:1101–1107

10. Ward C, Hogan M, Rossetti S, Walker D, Sneddon T, Wang X,Kubly V, Cunningham J, Bacallao R, Ishibashi M, Milliner D,Torres V, Harris P (2002) The gene mutated in autosomal re-cessive polycystic kidney disease encodes a large, receptor-like protein. Nat Genet 30:259–269

11. Onuchic L, Furu L, Nagasawa Y, Hou X, Eggermann T, RenZ, Bergmann C, Senderek J, Esquivel E, Zeltner R, Rudnik-Schöneborn S, Mrug M, Sweeney W, Avner E, Zerres K,Guay-Woodford L, Somlo S, Germino G (2002) PKHD1, thepolycystic kidney and hepatic disease 1 gene, encodes a novellarge protein containing multiple IPT domains and PbH1 re-peats. Am J Hum Genet 70:1305–1317

12. Lieberman E, Salinas-Madrigal L, Gwinn J, Brennan L, FineR, Landing B (1971) Infantile polycystic disease of the kid-neys and liver: clinical, pathological and radiological correla-tions and comparison with congenital hepatic fibrosis. Medi-cine (Baltimore) 50:277–318

13. Kaariainen H, Jaaskelainen J, Kivisaari L, Koskimies O, NorioR (1988) Dominant and recessive polycystic kidney disease inchildren: classification by intravenous pyelography, ultra-sound, and computed tomography. Pediatr Radiol 18:45–50

14. Cole BR, Conley SB, Stapleton FB (1987) Polycystic kidneydisease in the first year of life. J Pediatr 111:693–699

15. Gagnadoux M-F, Habib R, Levy M, Brunelle F, Broyer M(1989) Cystic renal diseases in children. Adv Nephrol18:33–58

16. Kaplan B, Fay J, Shah V (1989) Autosomal recessive polycys-tic kidney disease. Pediatr Nephrol 3:43–49

17. Zerres K, Rudnik-Schoneborn S, Deget F, Holtkamp U,Brodehl J, Geisert J, Scharer K, Nephrologie AfP (1996) Au-tosomal recessive polycystic kidney disease in 115 children:clinical presentation, course and influence of gender. ActaPaediatr 85:437–445

18. Roy S, Dillon M, Trompeter R, Barratt T (1997) Autosomalrecessive polycystic kidney disease: long-term outcome ofneonatal survivors. Pediatr Nephrol 11:302–306

19. National High Blood Pressure Education Program WGR(1996) Update on the 1987 Task Force Report on High BloodPressure in Children and Adolescents. Pediatrics 98:649–658

20. Schwartz G, Haycock G, Edelmann C, Spitzer A (1976) Asimple estimate of glomerular filtration rate in children de-rived from body length and plasma creatinine. Pediatrics58:259–263

21. Schwartz G, Feld L, Langford D (1984) A simple estimate ofglomerular filtration rate in full-term infants during the firstyear of life. J Pediatr 104:849–854

22. Kaariainen H (1987) Polycystic kidney disease in children: a genetic and epidemiological study of 82 Finnish patients. J Med Genet 24:474–481

23. Lonergan G, Rice R, Suarez E (2000) Autosomal recessivepolycystic kidney disease: radiologic-pathologic correlation.Radiographics 20:837–855

24. Jamil B, McMahon L, Savige J, Wang Y, Walker R (1999) A study of long-term morbidity associated with autosomal recessive polycystic kidney disease. Nephrol Dial Transplant14:205–209

25. Mattoo TK, Khatani Y, Ashraf B (1994) Autosomal recessivepolycystic kidney disease in 15 Arab children. Pediatr Nephrol8:85–87

26. Chapman A, Gabow P (1997) Hypertension in autosomaldominant polycystic kidney disease. Kidney Int [Suppl]61:S71-S73

27. Tsimaratos M, Cloarec S, Roquelaure B, Retornaz K, Picon G,Chabrol B, Guys J-M, Sarles J, Nivet H (2000) Chronic renalfailure and portal hypertension-is portosystemic shunt indicat-ed? Pediatr Nephrol 14:856–858

125

28. Blickman J, Bramson R, Herrin J (1995) Autosomal recessivepolycystic kidney disease: long-term sonographic findings inpatients surviving the neonatal period. Am J Radiol 164:1247–1250

29. Fick-Brosnahan G, Belz M, McFann K, Johnson A, Schrier R(2002) Relationship between renal volume growth and renalfunction in autosomal dominant polycystic kidney disease: alongitudinal study. Am J Kidney Dis 39:1127–1134

30. Kogutt M, Robichaux W, Boineau F, Drake G, Simonton S(1993) Asymmetric renal size in autosomal recessive polycys-tic kidney disease: a unique presentation. Am J Radiol160:835–836

31. Lucaya J, Enriquez G, Nieto J, Callis L, Pena P, Dominguez C(1993) Renal calcifications in patients with autosomal reces-sive polycystic kidney disease: prevalance and cause. Am JRadiol 160:359–362

32. Currarino G, Stannard M, Rutledge J (1989) The sonolucentcortical rim in infantile polycystic kidneys. J Ultrasound Med8:571–574

33. Melson G, Shackelford G, Cole B, McClennan B (1985) Thespectrum of sonographic findings in infantile polycystic kid-ney disease with urographic and clinical correlations. J ClinUltrasound 13:113–119

126