J Med Genet 1991; 28: 781-785

Case reports

Transmission of Proteus syndrome from father toson?

J Goodship, A Redfearn, D Milligan, D Gardner-Medwin, J Burn

AbstractWe present a male infant with cranial hemi-hypertrophy, a lymphangioma, a lipoma, andepidermal naevi. A diagnosis of Proteus syn-drome was made. His father had had a largelymphangioma resected from the right side ofthe face as a child. We propose that Proteussyndrome has been transmitted from fatherto son.

The features of Proteus syndromeiare partial gigant-ism of the fingers and toes, hemihypertrophy,macrocephaly, plantar hyperplasia, haemangiomas,lipomas, lymphangiomas, and epidermal naevi. Thesyndrome was named by Wiedemann et aPt in 1983after the Greek god, Proteus, who was able to changehis shape at will to avoid capture. The phenotypeevolves over time. Joseph Merrick, the ElephantMan, now accepted to be the first documented caseof Proteus syndrome, had no obvious abnormalitiesat birth. The evolution of the facial dysmorphismcan be seen in the figures of case 2 described byCohen.2 In Cohen's three carefully documentedcases the pathognomonic features of digital gigan-tism and plantar hyperplasia were not present before1 year of age. The naevi and plantar hyperplasiausually become apparent after the neonatal period.The sex ratio is equal. The parents of all reportedcases have been normal and there have been noaffected sibs. We have found only one published

Divisions of Human Genetics and Child Health,Newcastle upon Tyne.J Goodship, A Redfearn, D Milligan, D Gardner-Medwin,J BurnCorrespondence to Dr Goodship, Division of HumanGenetics, University of Newcastle upon Tyne, 19/20Claremont Place, Newcastle upon Tyne NE2 4AA.

Received for publication 28 February 1991.Accepted for publication 22 April 1991.

report of an affected subject having offspring.' Thechild in this case was normal.We describe a child with severe Proteus syndrome

whose father had a lymphangioma of the right sideof the face.

Case reportThe proband was the first child ofunrelated parents.The mother had four children by a previous consortand the father had one daughter by a previousconsort. The father was 30 and the mother 38 whenthe proband was born. Routine ultrasound scans at11 and 18 weeks' gestation had not detected anyabnormalities. A further ultrasound scan was per-formed at 38 weeks' gestation because of suspectedpolyhydramnios. It confirmed polyhydramnios andalso indicated a large, loculated, cystic swellingarising from the thoracic region on the right poster-iorly and a dilated right kidney. A placental biopsyshowed a normal karyotype. The mother went intospontaneous labour at term resulting in breech de-livery of a male weighing 3400 g. The Apgar scorewas 4 at one minute and 7 at five minutes.At birth the baby had obvious enlargement of the

left side of the face and cranium, a large fluctuantmass over the right lumbar area, and wide big toeclefts (figs 1 and 2). The OFC was 400 cm (abovethe 90th centile). All structures on the left side oftheface were larger, including the palpebral fissure, ear,and nostril. A difference of facial skin texture wasevident with midline demarcation. Similar asym-metry was apparent in the alveolar ridge and tongue(fig 3). The mandibular central incisor on the lefterupted at 10 weeks and the maxillary central incisoron the left at 26 weeks, both much earlier than theircounterparts on the right.The asymmetry of the face has increased with

time (fig 4). At 7 months of age the head circumfer-ence was 51-7 cm (still above the 90th centile). Theright ear was 4-2 cm long and the left 5-5 cm. Thehair on the right side of the head was dark and curly

781

copyright. on January 31, 2022 by guest. P

rotected byhttp://jm

g.bmj.com

/J M

ed Genet: first published as 10.1136/jm

g.28.11.781 on 1 Novem

ber 1991. Dow

nloaded from

Goodship, Redfearn, Milligan, Gardner-Medwin, Burn

Figure I The proband in theneonatal period. The left cheek wasenlarged and there was a largefluctuant swelling over the rightposterior thorax.

but the hair on the left was pale and fine. The righteyebrow had a clearly defined shape and was dark,but the left eyebrow was poorly delineated with fairhair extending up the forehead. Ophthalmologicalexamination showed a choroidal coloboma on theleft.His leg length of 30 cm and total length of 70 cm at

7 months were on the 50th centile. His weight of9.4 kg was on the 75th centile.

I

Figure 2 A deep transverse plantar crease originating froma wide big toe cleft.

'NFigure 3 Hypertrophy of the alveolar ridge and tonguewith midline demarcation and premature tooth eruption onthe hypertrophied side.

Figure 4 The proband at 7 months showing enlargement ofthe soft tissues and sparse eyebrow and scalp hair on theleft.

782

Armulo,

copyright. on January 31, 2022 by guest. P

rotected byhttp://jm

g.bmj.com

/J M

ed Genet: first published as 10.1136/jm

g.28.11.781 on 1 Novem

ber 1991. Dow

nloaded from

Transmission of Proteus syndrome from father to son?

Figure S Epidermal naevuson the left side of the face andneck.

The large swelling over the right lumbar regionwas partially resected and was shown to be a lym-phangioma. There was a separate, smaller, firmmass, thought clinically to be a lipoma, over thelateral rib cage on the right. On abdominal examina-tion the left lobe of the liver was palpable. Thegenitalia were normal.The feet were noted to be abnormal in the neo-

natal period with widely spaced toes and a deepcrease between the hallux and second toe (fig 2).There was a dimple on the lateral border of both feetwhich was more pronounced when he flexed his toes(fig 6). The fingers and toes were symmetrical withno gigantism at 7 months of age.The skin lesions, not apparent at birth, were first

noted at 4 months. At 7 months there was a raisedvelvety naevus on the left side anterior to the left earextending down the neck and under the chin. Aseparate swirl from the lesion narrowed to a thinpigmented line terminating at the angle of the mouth(fig 5). There was a pigmented macule 1 cm indiameter lateral to the left eye, a 1-5 x 1 cm cafe aulait patch on the lateral border of the right foot (fig6), and a smaller pigmented macule on the medialaspect of the right thigh. The right side of the thoraxand abdomen had irregular areas of slightly hyper-pigmented skin juxtaposed with hypopigmentedareas.

Cranial CT scan showed left hemimegalencephalydisplacing the falx to the right of the midline (fig 7).There was ipsilateral enlargement of the lateralventricle. The left cerebral hemisphere was grosslyabnormal with enlarged hyperdense frontal andoccipital lobes and relative loss of volume, probably

owing to schizencephaly, in the parietal lobe. Theright cerebral hemisphere had numerous small areasof low density and smaller focal areas of high dens-ity.He developed seizures on day 3. Partial control

has been achieved with a combination of carbamaz-epine, vigabatrin, and clobazam. EEGs have repeat-edly shown a focus in the left centrotemporal regionand hemispherectomy is being considered.At 7 months he moves all four limbs but reaches

for objects with his left hand. Tone is slightlyincreased on the right. He reacts to noise and followsobjects through a short distance.The father of the index case was noted to have

facial asymmetry (fig 8). He had had a similar facial

Figure 6 Deep dimple on the lateral border of the rightfoot exaggerated by toe flexion. A cafi au lait patch isvisible on the dorsum of the foot.

783

copyright. on January 31, 2022 by guest. P

rotected byhttp://jm

g.bmj.com

/J M

ed Genet: first published as 10.1136/jm

g.28.11.781 on 1 Novem

ber 1991. Dow

nloaded from

Goodship, Redfearn, Milligan, Gardner-Medwin, Burn

enlargement as an infant and had undergone numer-ous operative procedures through childhood to re-duce this swelling (fig 9). Histology of the resectedtissue was of a lymphangioma. The skin overlyingthe lymphangioma appeared normal. His hands andfeet were normal, he had no pigmented naevi orhamartomata, and did not have hemimegalen-cephaly on CT scan.

DiscussionThe differential diagnosis considered in this childincluded Klippel-Trenaunay syndrome, Bannayan-Zonana syndrome, Maffucci syndrome, epidermalnaevus syndrome, encephalocraniocutaneous lipo-matosis, and Proteus syndrome. The overlappingmanifestations of these disorders, which complicatediagnosis, have been the subject of a recent review.4Hotamisligil4 suggested a diagnostic rating scale forProteus syndrome with different features beingawarded a number of points. The highest possiblescore for a patient with all the features was 19-5. Hefelt that 13 points were required to diagnose Proteussyndrome with 100% certainty. The child presentedin this paper scores 15-5 on Hotamisligil's scale.The asymmetry in our case and the skin lesions

exclude the diagnosis of Bannayan-Zonana syn-drome. Maffucci syndrome is characterised byenchondromatosis with classical radiographicchanges which were not present in our case. Thepresence of a lymphangioma and a lipoma in ourcase excludes Klippel-Trenaunay syndrome and epi-dermal naevus syndrome. Encephalocraniocuta-neous lipomatosis (ECCL) is a syndrome with unila-teral cutaneous and ophthalmological lesions with

Figure 7 CT scan showing enlargement and abnormalmorphology of the left hemisphere and left lateral ventriclewith a partial schizencephaly.

Figure 8 The proband's father.

Figure 9 Photograph of the father aged 8 after the firstsurgical procedure for removal of the lymphangioma fromthe right cheek.

784

copyright. on January 31, 2022 by guest. P

rotected byhttp://jm

g.bmj.com

/J M

ed Genet: first published as 10.1136/jm

g.28.11.781 on 1 Novem

ber 1991. Dow

nloaded from

Transmission of Proteus syndrome from father to son?

ipsilateral cerebral atrophy, seizures, and mentaldeficiency. Superficial lipomas and connective tissuenaevi have been observed in this syndrome andWiedemann and Burgio5 have suggested that ECCLmay be part of the Proteus spectrum. Hemimegalen-cephaly has not been reported in ECCL.The association of hemihypertrophy, a lymphan-

gioma, a lipoma, and pigmented skin lesions in our

patient makes Proteus syndrome the most likelydiagnosis. Widely splayed, abnormal toes and opticcoloboma have also been reported in Proteus syn-

drome.26 Our case does not yet have the characteris-tic enlargement of digits or plantar hyperplasia.Proteus syndrome is an evolving phenotype and notall features are apparent in the neonatal period. Itmay also be argued that it is a contradiction in termsto consider involvement of a particular part of thebody as prerequisite for the diagnosis of Proteussyndrome.A child has been reported who was mosaic for a

duplication of lql 1-q25 and had clinical featuresconsistent with a diagnosis of Proteus syndrome.7She had right hemihypertrophy, macrocephaly,multiple cavernous haemangiomas, scoliosis, and a

coloboma of the optic nerve. No cytogenetic abnor-mality was seen in lymphocytes or fibroblasts in ourpatient.The parents of all previous cases of Proteus syn-

drome have been normal and there have been no

affected sibs. The lack of any affected sibs, now thatmany cases have been described, is against auto-somal recessive inheritance. The reports are com-

patible with all cases representing somatic or germ-

line new mutations.The most intriguing feature of our case is the

presence of facial asymmetry in the father. He doesnot fulfil the diagnostic criteria of Proteus syndromebut the occurrence of an extensive facial lymphan-gioma in the parent of a child with Proteus syn-

drome is unlikely to be a coincidental finding.

A plausible explanation for this observation is thatProteus syndrome is an autosomal dominant dis-order with variable expression. This would be ana-logous to the variable expression seen in neurofibro-matosis which can cause hemihypertrophy evenwhen the mutant gene is inherited in the germline.An alternative explanation is that a mutant genepresent in the father has undergone a further muta-tion leading to more severe disease in the son. Athird explanation is that the father is a mosaic withthe mutation causing overgrowth in a proportion ofhis cells. Transmission of this mutation in the germ-line could account for the greater severity in theproband. This interpretation is at odds with thehypothesis put forward by Happle8 that Proteussyndrome is a manifestation of mosaicism for a muta-tion that would be lethal in the non-mosiac state.However, precedents for this have been describedand characterised at a molecular level. Subjects withmild osteogenesis imperfecta have had severelyaffected offspring and have been shown to be mosaicfor mutations causing collagen abnormalities.9

1 Wiedemann HR, Burgio GR, Aldenhoff P, Kunze J, KaufmannHJ, Schirg E. The Proteus syndrome. Eur J Pediatr1983;140:5-12.

2 Cohen MM. Understanding Proteus syndrome, unmasking theelephant man, and stemming elephant fever. Neurofibromato-sis 1988;1:260-80.

3 Horton WA. Klippel-Trenaunay-Weber syndrome. Birth De-fects 1971;7:315-8.

4 Hotamisligil GS. Proteus syndrome and hamartoses with over-growth. Dysmorphol Clin Genet 1990;4:87-102.

5 Wiedemann HR, Burgio GR. Encephalocraniocutaneous lipo-matosis and proteus syndrome. Am J Med Genet 1986;25:403-4.

6 Clark RD, Donnai D, Rogers J, Cooper J, Baraitser M. Proteussyndrome: an expanded phenotype. Am J Med Genet1987;27:99-1 17.

7 Say B, Carpenter NJ. Report of a case resembling the Proteussyndrome with a chromosome abnormality. Am J Med Genet1988;31:987-9.

8 Happle R. Lethal genes surviving by mosaicism: a possibleexplanation for sporadic birth defects involving the skin.J Am Acad Dermatol 1987;16:899-906.

9 Wallis GA, Starman BJ, Zinn AB, Byers PH. Variable expres-sion of osteogenesis imperfecta in a nuclear family isexplained by somatic mosaicism for a lethal point mutation inthe alphaI(I) gene (COLIAI) of type I collagen in a parent.Am J Hum Genet 1990;46:1034-40.

785

copyright. on January 31, 2022 by guest. P

rotected byhttp://jm

g.bmj.com

/J M

ed Genet: first published as 10.1136/jm

g.28.11.781 on 1 Novem

ber 1991. Dow

nloaded from