Van Buchem disease (VBD) or “hyperostosis corticalis generalisata familiaris” is a rare, autosomal recessive, bone sclerosing dysplasia, first described in 1955.1 Another 13 cases were subsequently identified, all among inhabitants of a small village in the Netherlands.2, 3 The phenotype of patients with VBD is very similar to that of patients with sclerosteosis, a bone-sclerosing dysplasia found mainly among the Afrikaners in South Africa.4 Both diseases are the result of defects of the SOST gene, which is located on chromosome 17q12–q21 and encodes sclerostin, an osteocyte-produced negative regulator of bone formation.5 Whereas sclerosteosis is caused by homozygous mutations in the SOST gene, VBD is due to a homozygous deletion of a 52-kb regulatory element 35 kb downstream of the SOST gene, which leads to impaired production of sclerostin.6–9
The clinical and biochemical features of patients with sclerosteosis and healthy disease carriers of this genetic abnormality have been well characterized,4, 10, 11 but less is known about VBD and carriers of the disease. Available data suggest clinical differences between the two sclerosing disorders,12 although information is lacking about potential biochemical differences between the two diseases and possible relationships between biochemical parameters and clinical and radiographic disease characteristics. The purpose of the present study was to characterize the demographic, clinical, biochemical, and densitometric features of patients with VBD and of related carriers of the gene mutation, and to study the relationship between circulating sclerostin and bone turnover markers levels and bone mineral density (BMD). A further aim of the study was to obtain insight in the natural course of VBD by retrospectively studying the clinical course of known patients with this disorder.
Patients and Methods
All known adult patients with VBD living in the Netherlands and their first degree relatives were invited to participate in this study. The majority of subjects were seen at their place of residence whereas those residing in other parts of the country were seen at the Leiden University Medical Center or at the University Medical Center Groningen.
A complete medical history was obtained from all patients, with special emphasis on complaints potentially associated with VBD. Physical examination including detailed neurologic examination and fundoscopy was performed in all patients but only when indicated in their relatives. Z-scores for height and head circumference were calculated and compared to the Dutch 1997 height-for-age and head circumference–for-age surveys using Growth Analyzer 3.5 (Dutch Growth Foundation, Rotterdam, the Netherlands).
Nonfasting blood samples were obtained from all subjects for DNA analysis and for the measurement of biochemical parameters of calcium and bone metabolism and of sclerostin. Subjects were seen at different times of the day, but patients and their relatives were seen at the same time on a particular day. BMD of the spine and of the hip were measured by DXA (Lunar Prodigy; GE Healthcare, Hoevelaken, The Netherlands). Audiometry results and data from previous radiologic examinations were obtained from hospital records.
The study was approved by the Ethics Committee of the Leiden University Medical Center and written informed consent was obtained from all adult participants and from parents of children included in the study.
DNA was tested for the presence of a homozygous or heterozygous 52-kb deletion 35 kb downstream of the SOST gene on chromosome 17q12–q21, known to be the underlying genetic defect for VBD, in patients and carriers.
Blood was collected from all patients and heterozygous carriers and measured for serum calcium adjusted for albumin binding, phosphate, and creatinine, using semiautomated techniques. Alkaline phosphatase (ALP) was measured by a fully-automated P800 modulator system (Roche BV, Woerden, The Netherlands). Parathyroid hormone (PTH) was measured using the Immulite 2500 assay (Siemens Diagnostics, Breda, The Netherlands). procollagen type 1 amino-terminal propeptide (P1NP) and β cross-linked C-telopeptide (β-CTX) were determined by the E-170 system (Roche BV). 25-Hydroxyvitamin D (25-OHD) was measured by the Liaison 25-OH Vitamin D TOTAL assay (DiaSorin S.A./N.V., Brussels, Belgium).
Sclerostin was measured in serum with an electrochemiluminescence assay (MSD 96-well MULTI-ARRAY Human Sclerostin Assay; Meso-Scale Discoveries, Gaithersburg, MD, USA) as described.11 Using this assay, with a detection limit of ± 1 pg/mL, we previously showed that sclerostin was undetectable in the serum of all of 19 patients with sclerosteosis tested.11 In every run, a control sample from a patient with sclerosteosis and 3 control samples with different values were included. Results of these measurements were highly reproducible (mean values in pg/mL [CV] in 18 runs with different batches were as follows: 20.9 [7.4%], 37.7 [6.2%], and 106.3 [7.4%]). For comparison of sclerostin and P1NP levels we used as control a group of 77 healthy subjects, as reported in a previous study.13 Mean age of the control group was 50.3 (range, 20–77) years, and mean body mass index (BMI) was 25.2 (range, 19.0–36.5) kg/m2.
Statistical analysis was performed using the SPSS 17.0 software (SPSS, Inc., Chicago, IL, USA). Differences between groups were analyzed using ANOVA or Student's t test as appropriate. A Games-Howell post hoc test was used because of inequality in sample sizes. Correlations between sclerostin, biochemical markers, age, and BMD values were assessed by Pearson's correlation testing. P1NP and CTX data were log-transformed because of skewness. A p-value below 0.05 was considered to be statistically significant.
Patients with VBD
Fifteen of the 18 known Dutch patients with VBD consented to participate in the study (12 adults and 3 children). The diagnosis was confirmed in all 15 patients by the presence of the VBD deletion on chromosome 17q12–q21. Of the 32 first-degree relatives who participated in the study, 28 adults were found to be carriers of the disease and were included in the analysis. Demographic characteristics of patients and heterozygous carriers are shown in Table 1.
Table 1. Van Buchem Disease: Characteristics of Patients and Carriers of the Disease
BMI = body mass index; BMD = bone mineral density; FN = femoral neck; PTH = parathyroid hormone.
Patients versus carriers.
39.07 ± 20.9
36.0 ± 16.9
182.9 ± 9.5
175.2 ± 8.2
0.31 ± 0.81
–0.16 ± 0.88
27.1 ± 4.2
25.0 ± 5.8
BMD FN (g/cm2)
2.16 ± 0.41
1.17 ± 0.16
BMD FN Z-score
8.7 ± 2.1
0.9 ± 1.0
BMD spine (g/cm2)
2.13 ± 0.54
1.31 ± 0.17
BMD spine Z-score
9.5 ± 1.9
1.3 ± 1.5
8.8 ± 5.9
5.7 ± 4.0
2.42 ± 0.22
2.45 ± 0.15
1.12 ± 0.33
1.10 ± 0.22
25 OH D (nmol/L)
53.9 ± 20.3
58.4 ± 14.7
77.5 ± 40.0
70.9 ± 12.1
All 15 patients had experienced in the past one or more episodes of facial palsy. The median age at first occurrence was 2.5 years. Facial palsy was already present at birth in 2 patients, and was observed within the first year of life in another 2. Unilateral or bilateral surgical decompression of the facial nerve was performed in 6 of the 15 patients. Five of the 15 patients reported recurrent episodes of ipsilateral facial palsy. Fourteen patients reported some degree of hearing impairment. In 6 of these, this preceded or occurred concurrently with a middle ear infection. The age at which hearing loss became noticeable differed markedly among patients, developing in early childhood in some, whereas it was only noticed in late adulthood in others. Five of the 15 patients used hearing aids and 6 had undergone previous surgery (mastoidectomy because of chronic otitis media, removal of exostosis from the auditory canal, placement of bone anchored hearings aids [BAHAs]). No patient reported visual impairment. Two patients reported a decreased sense of smell. Three patients had previously experienced symptoms of raised intracranial pressure. Following lumbar puncture, symptoms completely resolved in 1 patient, whereas in another they improved over several months with concomitant treatment with acetazolamide. In the third patient a ventricular peritoneal drain had to be placed and treatment with prednisone was given leading to complete resolution of symptoms.14 One patient had sustained a fracture of the wrist on two separate occasions, one at the age of 14 years after falling from a tree, and the other at the age of 17 years after a motorcycle accident. No fractures were reported by any other patient, despite involvement of 5 in major accidents (eg, fall from a height of 3 m, collision with a tree while riding a motorcycle at very high speed, fall of a 200-kg weight on the back). All patients reported to have trouble keeping afloat in water, only able to keep from sinking by active and continuous swimming.
The clinical course of the disease appeared to stabilize in adulthood, with the majority of patients reporting no progression of symptoms or development of complications with time. Three patients (aged 49, 52, and 82 years) did, however, experience a further decrease in hearing with time, and 1 patient (age 43 years) reported continuing increase in the size of her lower jaw. None of the patients reported symptoms related to other organs such as heart, lungs, urogenital, or gastrointestinal tracts, except for an 82-year-old man, who had diabetes mellitus and heart failure associated with hypertension.
On clinical examination all patients had normal stature (Table 1) and all adults had some degree of facial distortion. Eleven of the 12 adult patients had a large, high forehead, and the mandible was enlarged in 10, with 1 patient having had corrective surgery of her lower jaw. The 3 children had no apparent facial dysmorphic features.15 The average circumference of the skull in adults was 63.6 cm (Z-score: + 3.4) for men, and 61.3 cm (Z-score: + 3.2) for women. In the children, skull circumferences were 50.5 cm (Z-score: 0), 49.8 cm (Z-score: 0), and 52.3 cm (Z-score: + 1.3).
None of the patients had any abnormalities of hands or digits, such as syndactyly or nail hypoplasia. Blood pressure was normal in all but 1 patient and all had normal sinus cardiac rhythm.
On neurological examination none of the patients had anosmia, visual field defects or papilledema, except for a boy aged 5 years. Oculomotor function was normal and pupil responses were adequate in all. Sensibility of the face was impaired in 2 patients. All patients had slight to severe facial palsy (House Brackman score II to V) which was unilateral in 2 and bilateral in the remaining 13 patients. On previous otoscopy, exostosis was observed in the bony part of the external auditory canal in 10 patients; in some, the external auditory canal was narrowed to a lumen of less than 2 mm. In 1 patient there was a complete fixation of all the ossicles of the middle ear. Hearing tests had been performed at some stage in all patients. In 3 patients there was no hearing loss, whereas this was mild in 3, moderate in 3, moderately severe in 2, severe in 1, and profound in 3. The type of hearing loss varied between patients, (sensorineural, conductive, or mixed). Deep tendon reflexes were normal and all patients had normal plantar responses.
BMD measured in 11 adult patients, and in 2 children, was increased in all (Table 1). BMD increased with age, but appeared to reach a plateau in late adulthood (Fig. 1). In this cross-sectional study, the small number of elderly patients (only 2 early menopausal women) does not allow any conclusions about potential BMD changes with time.
CT scans of the mastoid, skull, and/or cervical spine had been performed in 9 patients. All scans showed a generalized increased thickness of all skull bones, with calvarial width measurements of greater than 2 cm in adult patients. A CT scan of the mastoid bones has been performed in 7 adult patients, in all of whom there was an evident narrowing or the internal auditory meatus. Narrowing of the internal auditory meatus could already be observed in a 3-year-old child, but absent in a 6-year-old child. Narrowing of the facial nerve canal was observed in 3 patients. A cervical spinal stenosis due to hyperostosis of the vertebrae and to degenerative changes of the intervertebral discs was observed in 3 patients, in 1 of those patients it resulted in a myelopathy. One patient had lumbar spine stenosis due to hypertrophic facet arthrosis and bulging of the intervertebral discs.
VBD disease carriers
Two of 28 studied heterozygous carriers had complaints potentially related to VBD. These 2 carriers, the mother and sister of a studied patient, had both sustained a transient unilateral facial palsy during labor, which was quickly resolved. At the time of the study there were no residual neurological signs of facial paralysis in either. Six of the 28 carriers had sustained a fracture (upper arm 2, leg 2, wrist 1, metacarpal 1), all but 1 having occurred in childhood, and all associated with appropriate trauma. None of the 4 relatives without the mutation reported a fracture.
Physical examination of the 28 carriers was unremarkable. Height and BMI were normal and not significantly different from those of patients (Table 1). All subjects had normal blood pressure and sinus cardiac rhythm.
Mean BMD at the lumbar spine and the femoral neck are shown in Table 1. BMD values were significantly lower than those of patients but varied widely (Fig. 1). Z-scores ranged between –2.2 and + 4.6 at the spine, and between –1.1 and + 2.9 at the femoral neck. At the spine, 19 subjects (70%) had BMD Z-scores above 0, with 6 subjects (22%) exceeding 2 SD. BMD Z-scores of the femoral neck were greater than 0 in 21 carriers (81%), exceeding 2 SD in 3 subjects (12%). There were only 4 menopausal women in the carrier group, not allowing any conclusions about a possible effect of menopause transition on BMD. When disease carriers were divided according to age (younger or older than 30 years) their BMD Z-scores were very similar (<30 years: spine 1.3 ± 1.0, femoral neck 0.81 ± 0.81; >30 years: spine 1.2 ± 1.8, femoral neck 0.83 ± 1.2).
Blood was collected from 14 patients and 28 carriers and from 77 healthy controls for measurement of biochemical markers of calcium metabolism and bone turnover. Mean calcium, phosphate, 25-OHD, and PTH concentrations did not differ between patients and carriers (Table 1). There were no abnormalities in hematological parameters in the 7 patients in whom these were tested.
Serum P1NP levels declined with age in both patients and carriers (Fig. 2). In adult patients, the value was greater than 65 ng/mL in 67% and was markedly elevated in children. Serum P1NP was also increased in 19% of adult heterozygous carriers. Because of the age-related changes and the lack of children in the carrier group, we compared only values of adults. Serum P1NP levels were significantly higher in adult patients with VBD than in carriers of the disease (96.0; 95% confidence interval [CI], 54.6–137.4 ng/mL versus 47.8; 95% CI, 39.4–56.2 ng/mL, p = 0.003) and healthy controls (37.8; 95% CI, 34.5–41.0, p = 0.028) (Fig. 3). There was no significant difference in serum P1NP values between carriers and healthy controls (p = 0.14).
Serum levels of CTX, obtained in the nonfasting state, declined with age in patients, but not in carriers (Fig. 2). These were higher than the upper limit of the normal reference range in 3 patients (25%) but in none of the carriers. Serum CTX levels were higher in patients (mean 447 pg/mL; 95% CI, 266–628 ng/mL) compared to carriers (mean 216 pg/mL; 95% CI, 167–266 pg/mL) (p = 0.020). Absolute values of serum CTX should be interpreted with caution because nonfasting blood samples were obtained. However, comparison between groups is valid because samples were obtained from patients and their relatives under identical conditions. Serum levels of CTX and P1NP were significantly correlated in both patients (r = 0.95, p < 0.001) and carriers (r = 0.71, p < 0.001).
Sclerostin was detectable in the serum of all but 1 patient, and ranged between 1.7 and 16.9 pg/mL. All carriers had detectable serum sclerostin levels, with values ranging between 14.4 and 55.0 pg/mL. VBD patients had significantly lower sclerostin levels (mean 8.0 pg/mL; 95% CI, 4.9–11.0 pg/mL) compared to carriers (mean 28.7 pg/mL; 95% CI, 24.5–32.9 pg/mL, p < 0.001) and to healthy controls (mean 40.0 pg/mL; 95% CI, 34.5–41.0 pg/mL, p < 0.001). Mean sclerostin levels were also significantly lower in carriers than in controls (p < 0.001) (Fig. 3).
The relationship between serum sclerostin, bone turnover markers, and BMD was studied by pooling data from all adult patients and carriers, in order to obtain a broader range of sclerostin values. In this pooled cohort, serum sclerostin levels were inversely correlated with serum P1NP levels (r = –0.39, p = 0.018) (Fig. 4), but not with serum CTX (r = –0.28, p = 0.11). Serum sclerostin levels were also inversely correlated with BMD (g/cm2) at both the femoral neck (r = –0.69, p < 0.001) and the lumbar spine (r = –0.78, p < 0.001) and with plasma PTH (r = –0.35, p = 0.042). There was no relationship between serum sclerostin values and BMD in the two groups of studied subjects analyzed separately or in carriers divided according to median BMD values. There was a trend for lower serum sclerostin levels in patients with femoral neck BMD values above the median compared to those below the median (5.4 ± 5.1 pg/mL versus 8.5 ± 4.8 pg/mL), but the numbers were small.
Since the first description of VBD, approximately 30 cases have been reported, the majority of whom are Dutch and living in the Netherlands. Before the discovery of the genetic background of the disease, a number of patients from the United States,16, 17 United Kingdom,18–20 and Italy21 were diagnosed as having VBD, based on their clinical presentation. In some of these cases, however, there was an autosomal dominant inheritance,17, 19, 21 making it more likely that they suffered from a “high bone mass disorder,” or Worth's disease,22 caused by mutations in the LRP5 receptor, as was later confirmed by genetic analysis in an Italian family.23 However, two siblings with genetically-confirmed VBD were recently reported from Germany, although no information about their genetic background was provided.24
The phenotype of our patients with VBD is in keeping with previously described clinical features of the disease.1–3, 15, 22, 25 Our data also confirm that the clinical course of VBD is less severe than that of sclerosteosis, in which the majority of patients develop conductive hearing loss in early childhood and raised intracranial pressure around adolescence or early adulthood,4, 11 the latter being the main cause of death in patients with sclerosteosis. Raised intracranial pressure was much less frequently observed in patients with VBD (20%), and none of the 3 patients with documented increased intracranial pressure required a craniotomy, in contrast to the patients with sclerosteosis.26 There was, however, a large variation in the severity of clinical findings among patients, but none of the measured laboratory parameters could explain these differences.
Facial palsy was a universal finding in studied patients. It developed early, at an age similar to that of patients with sclerosteosis, and was already present at birth in 2 patients. The reported occurrence of facial palsy at birth in a number of patients with VBD suggests that the fallopian canal may already be narrowed in utero. However, the regulatory element of SOST, which is lacking in patients with VBD, was not found to control embryonic SOST transcription,27 which might explain the absence of syndactyly and other digit abnormalities in VBD. Hearing loss was also common, but was less severe than in patients with sclerosteosis. These findings suggest that the extent of bone overgrowth during the first years of life might be similar in VBD and in sclerosteosis, and progression later in life appears to occur more slowly in VBD than in sclerosteosis. Complications associated with bone overgrowth appeared to stabilize after the third decade of life in both diseases.4, 11
Consistent with the clinical findings and similar to the case in sclerosteosis,11 serum P1NP levels declined with age, reaching normal levels in several patients. This was consistent with no further increases in BMD. The stabilization of BMD we observed, contrasts with the findings of a previous study, in which the diaphyseal/metaphyseal ratio and the cortical thickness of metacarpals were reported to increase with age in patients with VBD, and in which an adult patient with progressive growth of the mandible was described.25 The oldest patient in our study, aged 82 years, had an enlarged mandible that demonstrated no progressive growth for decades, and progressive enlargement of the mandible has not been reported in patients with sclerosteosis.
VBD is caused by a deficiency of sclerostin synthesis. As shown here, all but 1 of the patients had detectable levels of sclerostin in serum, which were, however, significantly lower than those of heterozygous carriers of the disease. These data contrast with findings in patients with sclerosteosis11 in whom the SOST gene is affected, rendering the synthesis of sclerostin impossible. The regulatory element that is missing in VBD has been shown to be of crucial importance for the transcription of SOST in bone.27 Therefore, it is likely that in VBD, sclerostin production is due either to leaky transcription, the accidental transcription of a gene without prior activation of its regulatory element, or to the production of sclerostin by cells other than osteocytes in which this regulatory element is not essential. In favor of the first hypothesis is the finding of a weak sclerostin signal by immunological staining in a bone biopsy from a patient with VBD.28 Alternatively, circulating sclerostin might be secreted by cementocytes or hypertrophic chondrocytes.28, 29 We have previously shown that sclerostin is not expressed in cementocytes of patients with VBD,28 but there are no data about its potential production by chondrocytes in these patients. In contrast to patients with sclerosteosis who are tall, patients with VBD have normal height, which is not different from that of carriers of the disease. We have previously speculated that the tall stature of patients with sclerosteosis might be due to lack of sclerostin at the growth plate, which might not be the case in patients with VBD and at least part of the measured circulating sclerostin may originate from chondrocytes. Clearly more studies are needed to clarify this issue. Independently of these considerations, the milder clinical phenotype of patients with VBD compared to that of sclerosteosis is in line with the differences in sclerostin levels between the two diseases.
As with sclerosteosis, a gene-dose effect was also present in VBD with patients having lower levels of sclerostin and higher levels of P1NP than their heterozygous carriers. Further, these sclerostin levels were significantly and negatively associated with BMD, underscoring the importance of sclerostin in the process of bone formation. In line with the findings in patients, carriers of VBD had higher levels of serum sclerostin than carriers of sclerosteosis. Consequently, the difference in mean serum P1NP levels between VBD carriers and healthy controls was also smaller, and did not reach significance. There was also a large variation in BMD values in carriers, with some individuals having very high values whereas others had low values, in contrast to the BMD of carriers of sclerosteosis, which was found to be either high normal or increased, with none of the patients having low values.30
Since its discovery a decade ago there has been much interest in sclerostin because of its key role in the regulation of bone formation and its use as target for new bone-building therapies for osteoporosis.31 Our data provide further support to the rationale that sclerostin functions as a negative regulator of bone formation in humans. Not only did patients with VBD have higher P1NP levels in comparison to carriers and controls, but sclerostin levels were also negatively associated with P1NP levels in a pooled cohort of patients and carriers.
The stabilization of the clinical course of VBD with time, and the decline in P1NP levels in patients with VBD and sclerosteosis with time raise the question of whether sclerostin controls bone formation to the same extent throughout life. During childhood, longitudinal growth and strengthening of the skeleton by bone modeling requires high bone formation rates. We believe that sclerostin is extremely important during bone modeling to keep the vast numbers of active osteoblasts in check. Later in life, when the skeleton has matured, bone formation takes place as part of the bone remodeling process, and osteoblast number and activity is far less than that required for the process of bone modeling. Although sclerostin plays a clear role in the regulation of osteoblast activity during bone remodeling, sclerostin deficiency appears to be less important during this process than during bone modeling. Thus, it may well be that the absence of sclerostin during bone remodeling may be compensated to a certain extent by the action of other Wnt-antagonists, such as DKK1, or by other mechanisms maintaining the equilibrium between bone formation and resorption. However, this remains speculative, and further studies are required to examine this hypothesis.
All authors state that they have no conflicts of interest.
This work was supported by the European Commission-FP7 (TALOS: Health-F2-2008-201099). We thank Mr. Abdulrhaman Al-Afandi for his help in the genotyping of patients and carriers.
Authors' roles: AHvL, NATH, and SEP contributed to the conception, design, analysis, and interpretation of the data. AHvL, MEvE, and FGD contributed to the acquisition, analysis, and interpretation of data. EB contributed to the analysis and interpretation of data. All authors participated in drafting and/or revising the manuscript, and all authors approved the final version of the manuscript. AHvL and SEP take responsibility for the integrity of the analysis of the data.