The Stickler syndromes^1–8 (hereditary arthro-ophthalmopathy; McKusick nos. 108300 and 604841) are one of the more frequently occurring groups of chondrodysplasias and are the commonest inherited cause of rhegmatogenous retinal detachment.⁹ The majority of patients and pedigrees exhibit the type 1 or “membranous” vitreous phenotype^10,11 and harbour mutations in the gene for type II collagen (COL2A1).^12–21 While not all mutations in type II collagen result in the membranous vitreous anomaly,^22,23 when it is exhibited it appears to be congenital and provides a useful basis for mutant locus assignment. This is particularly helpful for sporadic cases where linkage is impossible, and especially in those individuals with mild or minimal systemic involvement where the diagnosis might otherwise be overlooked.^19,24–26 Other pedigrees exhibit a different “beaded” vitreous phenotype and are linked to a different locus.^11,27 We reported the first mutation in the gene encoding the α1 chain of type XI collagen (COL11A1) in one of these pedigrees²⁸ and this locus was confirmed in other pedigrees,^29,30 and is now known as type 2 Stickler syndrome (McKusick no. 604841). The intimate post-translational molecular associations between types II and XI collagen form the foundation of the close clinical overlap between these two sub-groups of Stickler syndrome, but the extent of this clinical overlap and variation remains to be defined. Whilst type 1 Stickler syndrome pedigrees have a particularly high risk of blindness through giant retinal tear and retinal detachment, Annunen et al suggest that patients with COL11A1 mutations are at a low risk of retinal detachment and have a higher incidence of midfacial hypoplasia.²⁹ This would be an important prognostic difference if confirmed. Here we wish to report the first description of the ophthalmic, oro-facial, audiologial, skeletal, and echocardiographic features of a large cohort of patients with type 2 Stickler syndrome in which molecular genetic analysis has confirmed mutations in COL11A1.

METHODS

Five pedigrees with Stickler syndrome, all exhibiting the beaded or type 2 vitreous phenotype, were identified from the Vitreous Research Clinic at Addenbrooke’s Hospital, Cambridge, UK. Ethical approval was granted (LRC92/019) and informed written consent was received in all cases. A sixth pedigree was identified via our collaboration with the University Hospital of Groningen, Netherlands.

The diagnostic criteria used to identify the index cases were modified from those previously published for type 1 Stickler syndrome¹¹ and are as follows: a “major” criterion: “beaded” vitreous anomaly and, in addition, any three of the following “minor” criteria:

1. Myopia with onset before 6 years of age.

2. Rhegmatogenous retinal detachment or paravascular pigmented lattice retinopathy.

3. Joint hypermobility with abnormal Beighton score, with or without radiological evidence of joint degeneration.

4. Audiometric confirmation of sensorineural hearing defect.

5. Midline clefting (bifid uvula, submucous cleft, high arch palate, cleft repair, Pierre Robin sequence).

Ophthalmic, oro-facial, skeletal, and audiologial features were assessed using the methods reported previously¹⁹ in addition to echocardiography. A general ophthalmic history was recorded with particular attention to the age of onset, degree and progression of myopia, cataract, and vitreoretinal disease. A full ophthalmic examination was carried out. In some of the younger patients applanation tonometry and gonioscopy were not possible. Anterior and posterior segment photographs were taken where appropriate.

Orofacial features were assessed according to standard protocols.³¹ Antero-posterior and lateral facial photographs at a standardised scale of 1:8 using a Nikon FM2 camera with Micro Nikon 105 mm medical lens and Kodachrome 64 film at F16. A 1 cm grid was printed and then photographed at a scale of 1:8 to match and clinical measurements of outer canthal distance, inner canthal distance, philtrum length, and middle finger length were also recorded. These measurements were used as controls for the photographic calibration. Control measurements of inner and outer canthal distance, interpupillary distance, and philtrum length from 20 unaffected siblings and 60 age matched controls (recruited from the general ophthalmic clinic) were also recorded.

Joint hypermobility was assessed objectively using the Beighton scoring system.³² A score of 1 or 0 is given for a series of joint manoeuvres and the total sum allocated up to a possible maximum score of 9/9.

An enquiry was made regarding the date and progression of any subjective hearing loss and, in particular, whether this had been a congenital, sudden, or progressive deterioration. A record of the type and duration of noise exposure was made together with any factors likely to contribute a conductive element to any hearing loss. All affected patients underwent bilateral otoscopy and audiometry involving air and bone conduction testing according to standardised procedures.³³

Patients were questioned about their cardiovascular history and underwent a full cardiological examination including auscultation and 2-D echocardiography. Echocardiographic studies were carried out using a Hewlett-Packard Sonos 1000 with 3.5–2.5 MHz phased array transducers. Echocardiographic views consisted of long and short axes, apical four and apical two chambers, incorporating conventional pulsed and colour flow Doppler.

For the large families MS1, MS40, MS42, and JH1, linkage analysis was carried out with flanking and/or intragenic markers for the candidate genes COL2A1, COL11A1, and COL11A2 as previously described.³⁰ For MS67 and MS71, analysis of COL11A1 was performed on the basis of vitreous phenotype. Amplification and sequencing of COL11A1 cDNA was achieved using RNA from cultured dermal fibroblasts.^28,30

RESULTS

A total of 31 affected members from six pedigrees were identified. The pedigrees are shown in fig 1.

All patients exhibited the “beaded” vitreous phenotype (fig 2) and had confirmed mutations in the α1 chain of type XI collagen (COL11A1) shown in table 1. The Dutch family JH1 was found to have the same mutation as the British family MS42.

Clinical features

Twenty patients (65%) were female and 11 (35%) male. The ages ranged from 10 to 84 years with an average age of 38. The clinical features are summarised in table 2.

Ophthalmic

A total of 87% of patients were short sighted (myopic) and 14 (52%) of the patients reported their myopia to be stable. Some unaffected individuals exhibiting a normal vitreous phenotype were also myopic, including one of non-identical twins, MS40 IV:5, emphasising the importance of vitreous phenotype as a marker for the disorder.

Some 64% of patients either had cataract or were aphakic or pseudophakic. Of the phakic patients with cataract, 33% (5/15) exhibited the wedge-shaped lens opacity peculiar to Stickler syndrome^34,35 and some sporadic cases of rhegmatogenous retinal detachment (fig 3).

A total of 38% had pigmented paravascular lattice. Thirteen patients (42%) suffered retinal detachment. Six had bilateral retinal detachment, one with bilateral giant retinal tears. The average age at which retinal detachment had occurred was 34 years with a range of 9–55 years. Two patients had retinal detachment under the age of 16.

Orofacial

One third of patients were found to have variable manifestations of mid-line clefting including bifid uvula, high arched palate, and cleft palate. Facial features were in general more subtle than those seen in type 1 Stickler syndrome with mild mid-facial and nasal hypoplasia. In some affected individuals the facial phenotype did not vary significantly from age/sex matched controls³¹ (fig 4).

Seven patients had lateral skull x rays. Four were normal (normal calvaria and frontal sinuses). Of the three abnormal cases, in one the frontal sinus was absent; in another it was small and, in contrast, large in the third.

No patient had any abnormalities of skin, hair, or sweating as reported in Marshall syndrome.³⁶

Skeletal

One third of patients exhibited or reported previous joint laxity (fig 5) with almost half experiencing symptoms of arthritis (most frequently knees, ankles, back, and wrists).

Audiological

A total of 45% of patients reported symptomatic hearing loss but, of those tested, 20/25 (80%) had some degree of high frequency sensorineural hearing loss ranging from mild (30 dB) to moderate (30–60 dB) loss. No patient reported profound deafness in either ear. Three patients had mild or moderate conductive hearing loss in addition to sensorineural loss.

Mitral valve prolapse

None of the 12 patients who underwent echocardiography had mitral valve prolapse.

DISCUSSION

This is the first report describing the clinical features of type 2 Stickler syndrome. All patients have a proven mutation in COL11A1 and exhibited the “beaded” type 2 vitreous phenotype.

The range of clinical features is similar to those in type 1 Stickler syndrome, with variability between and within families, but with a particularly high prevalence of sensorineural hearing loss which is often mild enough to go unnoticed by the patient. In contrast to the study by Annunen et al, this study confirms that individuals are indeed at high risk of retinal detachment, 42% of this group having suffered a retinal detachment in at least one eye at the time of study.

There has been some confusion in the literature regarding the vitreous phenotypes of type 1 and type 2 Stickler syndrome. It is important to recognise that the type 1 anomaly is a congenital and not a degenerative manifestation. Following a case report by Parentin et al,^37,38 McLeod et al³⁹ have suggested that observation of the vitreous phenotype is not reliable in distinguishing between type 1 and type 2 Stickler syndromes. The pedigree reported by Parentin et al^37,38 was classified as type 1 Stickler syndrome but linkage analysis was thought to favour COL11A1 rather than COL2A1, although no mutation analysis was performed. The description in the manuscript of the vitreous phenotype suggested type 1 Stickler syndrome, but the photographs included in the paper did not demonstrate the type 1 vitreous anomaly. All three patients, in whom the vitreous had been commented upon, had bilateral retinal detachment. An alternative explanation is that the type 1 vitreous anomaly was confused with the detached posterior hyaloid membrane. In addition, the implication of COL11A1 was only weakly supported by the linkage analysis, whilst the markers used for the linkage analysis did not adequately exclude a mutation in COL2A1 as they did not flank both sides of the gene.

McLeod et al³⁹ also describe two affected members of a type 2 Stickler syndrome pedigree (with a COL11A1 mutation confirmed) and report that the vitreous phenotype appeared to change from a type 2 anomaly to a type 1 anomaly with the development of a posterior vitreous detachment. In contrast to the congenital type 1 vitreous anomaly, the posterior hyaloid membrane is visible only after posterior vitreous detachment and is clearly a separate entity.⁴⁰ Both membranes can be demonstrated in the same eye when a patient with type 1 Stickler syndrome develops a posterior vitreous detachment. The posterior hyaloid membrane differs in its position, its movement, and the degree of surface crinkling. We believe the suggestion that the type 2 phenotype can convert to the type 1 phenotype is misleading.

The “beaded” type 2 vitreous anomaly is less easy to distinguish but, as our study demonstrates, is sufficiently characteristic to be a useful clinical hallmark differentiating type 2 from type 1 Stickler syndrome.

In contrast to the findings of Liberfarb and Goldblatt⁴¹ who found that over 45% of their patients had mitral valve prolapse, Ahmad et al⁴² looked at 78 patients who included both type 1 and the type 2 Stickler syndrome patients included in this study and found that none had mitral valve prolapse.

Other disorders have been reported to share some of the features of Stickler syndrome. Wagner⁴³ reported a large Swiss family with an autosomal dominant vitreoretinal disorder resembling Stickler syndrome but without retinal detachment. Analysis of the original Wagner pedigree has shown linkage to 5q13–q14,⁴⁴ confirming that it is genetically distinct from Stickler syndrome. The original Weissenbacher-Zweymuller syndrome patient⁴⁵ was found to be heterozygous for a mutation in COL11A2.⁴⁶ Although, in cartilage, the α2(XI) collagen forms heterotrimers with α1(XI) collagen, it is not expressed in the eye and thus there are no associated eye changes. The term non-ocular Stickler syndrome (McKusick no. 184840) encompassing COL11A2 disorders has been suggested.

There is continuing debate over the clinical overlap and differential diagnosis of Stickler and Marshall syndromes.^29,47–49 Marshall³⁶ described seven members in a three generation pedigree who were affected with a hereditary “ectodermal dysplasia” with ocular abnormalities and hearing defect. The pedigree showed autosomal dominant inheritance, normal stature but diminished sweating, and abnormal teeth. Hair and nails were normal. All patients were myopic (moderate to high) with fluid vitreous, although the vitreous phenotype was not described in detail. Several affected individuals had congenital or juvenile cataract which underwent sudden maturation, some with lens subluxation and secondary glaucoma. At the age of 43, one patient suffered a retinal detachment, 9 months following traumatic lens dislocation. Otherwise, there were no localised retinal lesions. In contrast to the series of patients described here, all the patients reported by Marshall had a short, depressed nose and an underdeveloped maxilla. x Rays showed a thickening of the outer table of the skull and absent frontal sinus in two siblings. Midline clefting and arthropathy were not reported although one patient had mild postural scoliosis. Shanske et al⁵⁰ argue that photographs published in Marshall’s original paper also show that several of the patients have striking ocular hypertelorism, and confirm that Marshall syndrome is a rare condition, with only eight additional reports since 1958. In two of these reports cited by Shanske et al, the authors did not consider their patients to have Marshall syndrome.^51,52 The distinction between the Stickler and Marshall syndromes is complicated by further reports describing Marshall syndrome but with features that were not described in the original Marshall kindred, and yet are known features in cases of Stickler syndrome confirmed by molecular genetic analysis. Annunen et al²⁹ described a series of patients with mutations in COL11A1 and COL2A1 and found similar clinical findings in patients with mutations in either gene. The notable differences were that those with COL11A1 mutations more commonly had severe hearing impairment and seldom had vitreoretinal degeneration or retinal detachment. Those with COL11A1 mutations were classified as having Marshall syndrome or an overlapping Marshall–Stickler syndrome, whilst those with COL2A1 mutations were considered to have Stickler syndrome. The controversy will continue until the molecular genetic basis of the original Marshall pedigree is resolved.

In the continuing search for a clinical distinction between Stickler syndrome and Marshall syndrome, concentration on subtle facial differences may detract from recognising the serious risk of retinal detachment. This study demonstrates the importance of the vitreous phenotype in the diagnosis of Stickler syndrome, even in those individuals who appear clinically normal in other aspects of the disorder. Recognising the risk to the individual and to members of the family allows appropriate steps to be taken to educate, offer genetic counselling, consider prophylaxis, and offer prompt remedial treatment.