Moira S. Cheung

Osteogenesis Imperfecta: Update on presentation and management

Osteogenesis Imperfecta (OI) is a rare heritable condition characterized by bone fragility and reduced bone mass. Traditionally OI was classified into OI types I to IV and thought to be only due to a defect in the collagen gene, however through the discovery of the new types of OI–V to VII, breakthroughs have been made in understanding the pathophysiology of autosomal recessive OI and new genetic mutations, such as in CRTAP and P3H1 genes. OI can present at any age and be difficult to diagnose because of the wide phenotypic variation. Awareness of the new forms of OI, the differential diagnosis and the limitations of diagnostic tools, all help to correctly diagnose and manage a patient with OI. Cyclical intravenous pamidronate is now the standard of care for moderately to severely affected children with OI, given in combination with good orthopedic, physiotherapy and rehabilitation programs. The benefits and short term safety of cyclic bisphosphonates have been amply reported in the literature; however their long term effects are still under investigation. Newer more potent forms of bisphosphonates such as zoledronic acid have undergone and are still being subject to international multicentric drug trials and are beginning to replace pamidronate in some centers.

Risedronate in the Treatment of Mild Pediatric Osteogenesis Imperfecta: A Randomized Placebo‐Controlled Study

Intravenous pamidronate is the most widely used treatment for moderate to severe osteogenesis imperfecta (OI). Currently, there is no medical treatment for patients with mild OI. We conducted a single‐center randomized double‐blind placebo‐controlled trial to examine the efficacy and safety of oral risedronate in the treatment of pediatric patients with mild OI. A total of 26 children and adolescents (age, 6.1–17.7 yr; 11 girls) with OI type I were randomized to either placebo (N = 13) or risedronate (N = 13) for 2 yr. Risedronate doses were 15 mg once per week in patients weighing <40 kg and 30 mg once per week in patients weighing >40 kg. After 2 yr of treatment, risedronate decreased serum levels of the bone resorption marker collagen type I N‐telopeptide by 35% compared with a 6% reduction with placebo (p = 0.003). Risedronate increased lumbar spine areal BMD Z‐scores by 0.65, whereas patients receiving placebo experienced a decrease of 0.15 (p = 0.002). In contrast, no significant treatment differences in bone mass and density were found at the radial metaphysis and diaphysis, the hip, and the total body. Histomorphometric analysis of transiliac bone biopsies at the end of the study period did not show a significant treatment difference in cortical width, trabecular bone volume, or parameters of bone turnover. Similarly, there was no detectable treatment effect on vertebral morphometry, second metacarpal cortical width, grip force, bone pain, or number of new fractures. Regarding safety, risedronate was generally well tolerated, and the incidence of clinical or laboratory adverse experiences was similar among treatment groups. These results suggest that the skeletal effects of oral risedronate are weaker than those that are commonly observed with intravenous pamidronate treatment but still lead to an increase in lumbar spine areal BMD. Future studies should investigate whether oral risedronate is effective in reducing fracture rates in children and adolescents with mild OI type I.

Osteogenesis imperfecta type V: marked phenotypic variability despite the presence of the IFITM5 c.−14C>T mutation in all patients

Background Osteogenesis imperfecta (OI) type V is an autosomal dominant bone fragility disorder that we had described a decade ago. Recent research has shown that OI type V is caused by a recurrent c.-14C>T mutation in IFITM5. In the present study, we assessed all patients diagnosed with OI type V at our institutions for the presence of the IFITM5 mutation. Methods IFITM5 exon 1 was analysed by Sanger sequencing in genomic DNA from 42 patients with OI type V (age: 2–67 years; 18 female). Results The c.−14C>T mutation of IFITM5 was detected in all individuals. Indicators of disease severity varied widely: Height z-scores (n=38) ranged from −8.7 to −0.1, median −3.5. Median final height was 147 cm in men (N=15) and 145 cm in women (N=10). Lumbar spine areal bone mineral density z-scores in the absence of bisphosphonate treatment (n=29) were between −7.7 and −0.7, median −5.3. Scoliosis was present in 57%, vertebral compression fractures in 90% of patients. Conclusions Even though the disease-causing mutation is identical among patients with OI type V, the interindividual phenotypic variability is considerable.

Tooth Extraction Socket Healing in Pediatric Patients Treated with Intravenous Pamidronate

Osteonecrosis of the jaw (ONJ) has been described as a complication of bisphosphonate therapy in adults. In the present study, we did not find a case of ONJ among 278 pediatric patients who had received intravenous pamidronate during childhood or adolescence.

Natural history of hyperplastic callus formation in osteogenesis imperfecta type V.

Hyperplastic callus formation was assessed in 23 patients with osteogenesis imperfecta type V. Hyperplastic callus mostly affected long bones in the lower extremities and occurred predominantly during phases of rapid growth.

Cortical and Trabecular Bone Density in X-Linked Hypophosphatemic Rickets

CONTEXT X-linked hypophosphatemic rickets is caused by mutations in PHEX. Even though the disease is characterized by disordered skeletal mineralization, detailed bone densitometric studies are lacking. OBJECTIVE The aim of the study was to assess volumetric bone mineral density (vBMD) in X-linked hypophosphatemic rickets using forearm peripheral quantitative computed tomography. SETTING The study was conducted in the metabolic bone clinic of a pediatric orthopedic hospital. PATIENTS Thirty-four patients (age, 6 to 60 years; 24 female) with PHEX mutations were studied, of whom 7 children (age, 6 to 11 years) were actively being treated with calcitriol and phosphate supplementation. Twenty-one patients (age, 16 to 40 years) had received the same therapy before but had discontinued the treatment; 6 patients (age, 12 to 60 years) had never received this treatment. MAIN OUTCOME MEASURES Trabecular and cortical vBMD of the radius. RESULTS Trabecular vBMD was elevated (mean age-specific and sex-specific z-score: +1.0) when all patients were analyzed together, due to very high results in currently treated patients (mean z-score: +2.4) and slightly above-average mean values in the other patients. Cortical vBMD was low when the entire cohort was analyzed together (mean z-score: -3.3), but was higher in currently treated patients (mean z-score: -1.3) than in patients who had discontinued therapy (mean z-score: -3.8) or who had never been treated (mean z-score: -4.1). CONCLUSIONS Patients with PHEX mutations have elevated trabecular vBMD at the distal radius while receiving calcitriol and phosphate supplementation, but low cortical vBMD at the radius diaphysis. Low cortical vBMD presumably reflects the underlying mineralization defect that is not entirely corrected by current treatment approaches.

Cranial base abnormalities in osteogenesis imperfecta: phenotypic and genotypic determinants.

Cranial base abnormalities are an important complication of osteogenesis imperfecta (OI), a hereditary bone fragility disorder that in most patients is caused by mutations affecting collagen type I. To elucidate which clinical characteristics are associated with the occurrence of cranial base abnormalities in OI, we compared cephalometric results of 187 OI patients (median age 12.0 years, range 3.4 to 47 years; 96 female) with those of 191 healthy subjects and related findings to clinical descriptors of the disease. Overall, 41 patients (22%) had at least one unambiguously abnormal skull base measure. Multivariate logistic regression analysis in patients with OI types I, III, and IV (n = 169) revealed that height Z‐score [odds ratio (OR) = 0.53, 95% confidence interval (CI) 0.43–0.66, p < .001]—but not age, gender, scleral hue, lumbar spine areal bone mineral density, or a history of bisphosphonate treatment—was a significant independent determinant of skull base abnormalities. Among patients with a height Z‐score below –3, 48% had a skull base abnormality regardless of whether they had received bisphosphonate treatment in the first year of life or not. Genotype‐phenotype correlations were evaluated in patients with detectable mutations in COL1A1 or COL1A2, the genes coding for collagen type I (n = 140). Skull base abnormalities were present in 6% of patients with haploinsufficiency (frameshift or nonsense) mutations, in 43% of patients with helical glycine substitutions caused by COL1A1 mutations, in 32% of patients with helical glycine substitutions owing to COL1A2 mutations, and in 17% of patients with splice‐site mutations affecting either COL1A1 or COL1A2. However, multivariate logistic regression analysis showed that height Z‐score but not the type of collagen type I mutation was independently associated with the prevalence of skull base abnormalities. In conclusion, this study shows that clinical severity of OI, as expressed by the height Z‐score, was the strongest predictor of skull base abnormalities. We did not find evidence for the hypothesis that bisphosphonate treatment protects against skull base abnormalities.

Wormian bones in osteogenesis imperfecta: Correlation to clinical findings and genotype†

The presence of a larger than usual number of Wormian bones (accessory skull bones completely surrounded by a suture line) is a well‐known radiographic sign of osteogenesis imperfecta (OI), but the phenotypic and genotypic correlates are not well characterized. In the present study we retrospectively analyzed skull radiographs of 195 OI patients (median age 11.8 years, range 0.4–48 years; 100 female). A significant number of Wormian bones (SNWB, defined as the presence of 10 or more Wormian bones) were found in at least one patient in all of the OI types studied (I, III to VII). SNWB were observed in 35% of patients with OI type I, in 96% of patients with OI type III and 78% of patients with OI type IV. SNWB were present in 28% of patients with haploinsufficiency (nonsense and frameshift) mutations in COL1A1, in 96% of patients with helical glycine substitutions in the alpha 1 chain of collagen type I and in 72% of patients with helical glycine substitutions in the alpha 2 chain of collagen type I. Stepwise multivariate logistic regression analysis showed that height z‐score, an indicator of disease severity, was inversely related with the prevalence of SNWB. SNWB were visible in 19 of the 26 patients who had skull radiographs in the first year of life, including a 2‐week‐old newborn. Thus, it appears that SNWB occur more frequently in more severely affected OI patients and seem to develop mostly in utero.

Muscle Anatomy and Dynamic Muscle Function in Osteogenesis Imperfecta Type I

CONTEXT Results of previous studies suggested that children and adolescents with osteogenesis imperfecta (OI) type I have a muscle force deficit. However, muscle function has only been assessed by static isometric force tests and not in more natural conditions such as dynamic force and power tests. OBJECTIVE The purpose of this study was to assess lower extremity dynamic muscle function and muscle anatomy in OI type I. SETTING The study was performed in the outpatient department of a pediatric orthopedic hospital. PATIENTS AND OTHER PARTICIPANTS A total of 54 individuals with OI type I (6-21 years; 20 male) and 54 age- and sex-matched controls took part in this study. MAIN OUTCOME MEASURES Calf muscle cross-sectional area and density were measured by peripheral quantitative computed tomography. Lower extremity muscle function (peak force per body weight and peak power per body mass) was measured by jumping mechanography through 5 tests: multiple two-legged hopping, multiple one-legged hopping, single two-legged jump, chair-rise test, and heel-rise test. RESULTS Compared with age- and sex-matched controls, patients with OI type I had smaller muscle size (P = .04) but normal muscle density (P = .21). They also had lower average peak force and lower specific force (peak force/muscle cross-sectional area; all P < .008). Average peak power was lower in patients with OI type I but not significantly so (all P > .054). CONCLUSIONS Children and adolescents with OI type I have, on average, a significant force deficit in the lower limb as measured by dynamic force tests. Nonetheless, these data also show that OI type I is compatible with normal muscle performance in some individuals.

Large Osteoclasts in Pediatric Osteogenesis Imperfecta Patients Receiving Intravenous Pamidronate

Intravenous pamidronate is widely used to treat children with moderate to severe osteogenesis imperfecta (OI). Changes in the appearance of osteoclasts have previously been noted in children receiving pamidronate and have been interpreted as signs of toxicity. In this study, we analyzed osteoclast parameters in paired iliac bone specimens before and after 2–4 yr of cyclical intravenous pamidronate therapy in 44 pediatric OI patients (age range: 1.4–17.5 yr; 21 girls). During pamidronate treatment, average osteoclast diameter and the mean number of nuclei present per osteoclast increased by 18% (p = 0.02) and 43% (p < 0.001), respectively. The number of samples containing large osteoclasts (LOcs, diameter > 50 μm) increased from 6 (14%) before treatment to 23 (52%) after pamidronate therapy (p < 0.001 by χ2 test). Post‐treatment samples containing LOcs had a greater core width (p = 0.04) and a higher cancellous bone volume per tissue volume (p < 0.001), because cancellous bone volume had increased more during pamidronate treatment (p < 0.001). Osteoclast number and surface were higher in samples with LOcs, but there was no difference in cancellous bone formation parameters. The presence of LOcs was independent of OI type, type of collagen type I mutation, lumbar spine BMD, and other clinical or biochemical measures. In conclusion, this study did not show any indication that LOcs during pamidronate treatment are indicative of toxicity. It seems more likely that the observed abnormalities in osteoclast morphology are part of the mechanism of action of this drug.