Background. The branchio‐oto‐renal (BOR) syndrome is an autosomal dominant disease characterized by hearing loss of early onset, preauricular pits, branchial clefts, and early progressive chronic renal failure in up to 40% of family affected members. So far, it has not received due attention in the adult European nephrology literature and because of the combination of deafness with chronic renal failure it may be confused with the Alport syndrome. The BOR syndrome is caused by mutations in the EYA1 gene that maps on chromosome 8q13.3.
Methods. A three‐generation, 20‐member large BOR Greek‐Cypriot family has been studied and followed up clinically over a 27‐year period. The findings in four individuals who developed early onset renal failure are described in detail. Genetic DNA linkage studies have also been carried out.
Results. Of the 15 family members at risk, 14 were tested with DNA linkage analysis. Ten members were genetically affected and four were normal. All 10 affected members developed early‐onset deafness. Some had different ear lobe abnormalities. Nine affected members had preauricular pits. In some of the patients these pits were deep and prominent while in others they were minor and superficial. Eight affected members had early‐onset branchial clefts that needed early corrective surgery without the correct familial diagnosis ever being made. End‐stage renal disease (ESRD) developed in four members at ages 36, 14, 17, and 17 with minimal proteinuria, if any. This was due to unilateral renal agenesis with contralateral hypodysplasia or bilateral, severe renal hypodysplasia.
Conclusion. The BOR syndrome is an infrequent but well‐described entity that combines early‐onset renal failure and deafness together with branchial clefts and preauricular pits. Renal agenesis and dysplasia are the causes of ESRD in these individuals. Other renal abnormalities include bifid kidneys with double ureters, vesico‐ureteric reflux and pelvi‐ureteric stenoses. The BOR syndrome should be included in the differential diagnosis of deafness and chronic renal failure in childhood and adolescence.
Patients combining different types of hearing loss, ear anomalies and branchial fistulae, recently defined as the branchio‐otic (BO) syndrome (OMIM 120502) have been written up and described in the medical literature since the mid‐19th century . The description and addition of the renal abnormalities to complete the whole spectrum of the branchio‐oto‐renal (BOR) syndrome (OMIM 113650), is credited to the work of Melnick et al. [2,3] and Fraser et al. , beginning in the mid 1970s. Although a number of publications on the BOR and BO syndromes have appeared since then, these are largely restricted to genetics journals, and only a few papers have appeared in the adult nephrology literature [5–8]. Most of the published work comes from North America, the only major European contributor being the Department of Genetics at the Institut Pasteur [9–11]. The BOR syndrome seems to be relatively unknown in adult European nephrology, and such families have been confused with the Alport syndrome .
The BOR syndrome is characterized by: (i) various types of deafness, (ii) preauricular pits, (iii) cervical branchial fistulae, and (iv) various renal abnormalities. The most important renal abnormalities that lead to end‐stage renal disease (ESRD) include unilateral renal agenesis with contralateral hypodysplasia or bilateral hypodysplasia characterized by decreased renal volume and size, loss of ultrasound normal cortico‐medullary differentiation, and hyperechogenicity of the renal cortex. Histologically there is glomerular hyalinization, mesangial proliferation, and basement membrane splitting . Bilateral renal agenesis is the extreme, leading to miscarriage or immediate neonatal death . Other renal abnormalities include hydronephrosis associated with ureteropelvic junction obstruction or vesicoureteral reflux. There can also be bifid kidneys with double ureters and calyceal anomalies .
Genetically, most families with the BOR syndrome have genetic mutations on the EYA1 gene on chromosome 8q13.3 [14,15]. At the same time, some BO families also show linkage to the same EYA1 gene on chromosome 8q13.3, suggesting that the BOR and BO syndromes may be allelic phenotypes of mutations in this single gene. Some BOR families and some BO families, however, do not link to this gene, indicating genetic heterogeneity. A second gene locus is currently being looked for. Very recently, a family with the BO syndrome has been identified and published, showing linkage to a new locus on chromosome 1q31, indicating that these two phenotypes, BO and BOR may also correspond to two different genetic entities .
The EYA1 gene is expressed very early, between the 4th and 6th weeks of human development. Deafness relates to abnormalities in the three ossicles of the middle ear derived from the first and second branchial arches, while the branchial fistulae relate to abnormalities of the second, third and fourth arches. In the embryonic human kidney the EYA1 gene is expressed strongly, and in the BOR syndrome there is an inductive fault between the ureteric bud and the metanephric mesenchymal mass as the ureteric bud branches into the renal parenchyma.
Subjects and methods
The Greek‐Cypriot family described in this study spans three generations and includes 20 members (Figure 1). Grandfather I‐2 died in 1974 from uraemia at age 37. Three additional members also reached ESRD and have since been transplanted. Their histories and findings are described in detail. All 19 alive members of the family underwent a complete physical examination with attention to the presence of (i) hearing loss, (ii) preauricular pits and ear‐lobe anomalies, (iii) cervical branchial fistulae, and (iv) renal abnormalities. A renal ultrasound examination was carried out in all 15 at‐risk individuals. The distribution of the four clinical features is shown on the family tree (Figure 1).
After informed consent, 18 living family members were tested by DNA linkage analysis using microsatellite markers D8S1807 and D8S530 that are located around the region of the EYA1 gene . The diseased gene frequency was assumed to be 1 in 10000, and because of the variable penetrance and expression of the disease phenotype, the risk for healthy descendants of affected individuals was used as 25%. The responsible mutation has not yet been looked for.
The presence or absence of each of the four major features of the BOR syndrome (hearing loss, preauricular pits, cervical–branchial fistulae, and renal abnormalities) is shown for every family member on the family tree (Figure 1). Genetic linkage analysis was carried out with the use of two microsatellite markers, D8S1807 and D8S530, around the EYA1 gene. The LOD score with D8S1807 reached a maximum of 3.02 at recombination fraction 0.0 and with D8S530 it reached 2.3, also at recombination fraction 0.0, thus documenting linkage to this locus (Figure 1). Ten of the 14 tested at‐risk family members had inherited the affected haplotype and shared some or all signs and symptoms of the BOR syndrome. Four at‐risk members had inherited the normal haplotype and had no suggestive signs or symptoms. One very young child with no relevant signs or symptoms has not yet been tested genetically. The systematic US studies in all 15 at‐risk individuals were normal in 14. One member had a bifid kidney. All 10 affected individuals had hearing loss. This was of variable severity and not always symmetrical or bilateral. Preauricular pits were present in nine of the 10 affected members and were always bilateral. Some pits were obvious and deep (Figure 2a), while some were minor and superficial. Some patients also had associated ear skin tags (Figure 2b). Ear‐lobe abnormalities were also present (Figure 3). These ear‐lobe abnormalities included asymmetrically positioned unequal ears, often cup‐shaped, resembling bat ears. Eight of the 10 affected at‐risk members had cervical branchial clefts, cysts, or fistulae, mostly bilateral, that had required corrective surgery earlier in life without the correct diagnosis ever being made. Only four members exhibited all four features of hearing loss, preauricular pits, branchial fistulae, and progressive renal failure. These four patients developed ESRD at ages 36, 14, 17, and 17.
Patient I‐2 was born on 24/12/1937. He died in August 1974 after 15 months of dialysis for ESRD, just as haemodialysis was being introduced in Cyprus. He was 37 at the time and from the history and old records, he had deafness, preauricular pits, and branchial fistulae.
Patient II‐3 was born on 24/2/1960. He is currently aged 41, alive and well, with a second kidney graft. He required haemodialysis at age 14 for ESRD and at age 16, on 25/10/1976, he received his first living related transplant from his unaffected mother. This functioned well for 25 years. By early 2001 he had gradually reached ESRD again. A second kidney transplant was carried out from a brother aged 41 without removing the failing first graft. The donor brother II‐2 is one of the 10 genetically affected members of the family with only hearing loss and preauricular pits. He has no branchial fistulae and no renal abnormalities. Careful assessment of his renal tract and renal function showed no abnormalities and because he was 41, it was considered very unlikely that he would develop renal failure at this age. He was therefore accepted as a kidney donor.
Patient II‐8 was born on 30/10/1963. In December 1980, at age 17, she began haemodialysis for ESRD. In 1981 she received her first kidney graft from a maternal aunt. This did not function and she returned to dialysis. In November 1983 she received a second, cadaveric graft. This did not function either. The patient underwent a transplant nephrectomy and returned to dialysis until November 1989 when she received a third kidney graft, from a cadaver. This was successful. She has since remained in excellent health and her serum creatinine remains at 1.1 mg%. She is now aged 37 but has not married.
Patient III‐1 was born on 30/11/1983. In 1994, at age 11, he was found to have a single right kidney on ultrasound and radioactive renogram. There was no proteinuria, the serum creatinine was 1.1 mg% and this single right kidney was small at only 9.3 cm long. The clinical picture was consistent with unilateral left renal agenesis and right hypodysplasia. At age 17, 6 years later, he reached ESRD. There was still no significant proteinuria. The single right kidney was now smaller at 8 cm long, with increased cortical echogenicity and lack of the normal corticomedullary differentiation. In March 2001 the patient was transplanted with a graft from a normal distant living relative. Five months later he is well with a serum creatinine of 1.6 mg%, on triple immunosuppressive treatment with steroids, cyclosporin, and mycophenolate mofetil.
In nephrology, the association of renal failure and deafness immediately brings to mind the Alport syndrome  with which every nephrologist is well acquainted. However, this is not always correct, as the present family study clearly illustrates. This family, known to this department for over 25 years, had initially been wrongly labelled as Alport syndrome in the 1970s, though in retrospect it did not fulfil the criteria for this syndrome. The useful paper by Misra and Nolph in 1998 , referred to a similar patient, initially thought to have Alport syndrome, but which later proved to be the BOR syndrome. This important paper alerted us to our own similar mistake and to the important spectrum of branchial‐arch abnormalities with deafness, associated with renal disease. The whole differential diagnosis of chronic renal failure and deafness in young people was very recently discussed by Richardson et al. , and their paper refers to a great number of syndromes ranging from Alport to Muckle–Wells, Refsum disease, Cockayne syndrome, familial renal tubular acidosis with sensorineural deafness, ichthyosis and prolinuria, Charcot–Marie–Tooth, Alstrom, mitochondrial disorders, and hypo‐ and hyper‐parathyroidism.
A recent search in Pubmed (1960–2001) on the BOR and BO syndromes yielded 69 publications in English with only six in nephrology journals [5–8]. There were 39 other publications in journals dealing with genetics, 15 in ENT and nine in various other journals, confirming the limited attention this syndrome has so far received in adult nephrology journals. Most publications come from North America.
This BOR Greek‐Cypriot family includes one dead affected and 15 alive at‐risk members, 10 of whom are clinically and genetically affected. Only four patients (36.4%) have developed ESRD, grandfather at age 36, a son and a daughter at ages 14 and 17, and a grandson at age 17. The four members that developed renal failure also had deafness, preauricular pits, and branchial fistulae. The grandson III‐1, who has been studied thoroughly, clearly had the agenesis/hypodysplasia complex. Hypodysplasia also seems to be the explanation for the ESRD in the remaining three cases. What remains unclear and puzzling is how and why affected family members with a presumably identical germinal mutation develop different features of the syndrome, while at the same time retaining the genetic ability to pass missing features like renal failure and branchial fistulae to their offspring. This wide, intrafamilial phenotypic variation within BOR families is of great pathophysiological and genetic interest and deserves further study.
One explanation may be attributed simply to non‐genetic stochastic factors that influence the development of kidneys at very early stages. Another explanation, however, may implicate the contribution of another unlinked gene that may co‐operate with the EYA1 gene product in kidney development. Inheritance of one or other gene alone may not be adequate to affect kidney development, whereas co‐inheritance may lead to the renal manifestations of agenesis/hypodysplasia. In the present family, for instance, patient III‐1, who had severe renal involvement and reached ESRD at very early age, may have been the result of co‐inheriting the rare EYA1 mutation from his father, and a more common genetic variant from his healthy mother. Examples of digenic inheritance have been reported recently for a form of progressive hearing loss with linkage to DFNA2 and DFNA12 and for a form of retinitis pigmentosa with the need for co‐inheriting mutations in the unlinked peripherin/RDS and ROM1 genes . A similar mechanism has been proposed for polycystic kidney disease, where it was hypothesized that cyst formation results from the inheritance of one germinal mutation in one of the two PKD1 or PKD2 genes, and the occurrence of a second‐hit, somatic mutation in the other interacting gene, thereby generating a trans‐heterozygous situation in the affected cystic cell . Similarly, in patients with the BOR syndrome with renal involvement, it cannot be excluded that a second hit has not affected the inherited normal allele of the same EYA1 gene.
Correspondence and offprint requests to: Dr Alkis Pierides FRCP, FACP, PO Box 25638, 1311 Nicosia, Cyprus. Email: email@example.com
The authors thank Dr K. Christodoulou for the statistical evaluation of the DNA linkage analysis data. This work was partly funded by the Cyprus Kidney Association.