Background. Difficulties with ambulation in patients with myelomeningocele often lead to physical inactivity, osteoporosis, and subsequent development of pathologic fractures.
Objective. The purpose of this study was to examine bone mineral density and biochemical markers of bone metabolism in patients with myelomeningocele.
Design and Methods. A total of 35 patients between 6 and 19 years of age with myelomeningocele (ambulatory and nonambulatory) were randomly chosen at the Texas Scottish Rite Hospital for Children. We measured bone mineral density of the distal radius in these patients using single photon absorptiometry and measured the biochemical markers of bone metabolism including parathyroid hormone, 1,25 vitamin D, osteocalcin, urinary pyridinolines/deoxypyridinolines, and urinary calcium excretion.
Results. Bone mineral density of the distal radius in the patients with myelomeningocele was ∼1 to 2 standard deviation units below the mean of the normal population. There were no significant differences between ambulators and nonambulators. However, bone mineral density of the 8 patients who suffered multiple fractures (19) was significantly lower than that for those remaining patients without fractures. Elevated urinary pyridinoline levels, which indicate elevated bone reabsorption, were found more frequently in both non- and limited ambulators than in full-time ambulators. Urinary calcium excretion also was greater than twofold higher in nonambulatory patients versus ambulatory patients. There were no other differences in the biochemical markers of bone metabolism (osteocalcin, parathyroid hormone, 1,25 vitamin D, and urinary deoxypyridinolines) between ambulators and nonambulators. Bone mineral density rises in normal growing children 6 to 19 years of age. When the boys and girls were considered separately, bone mineral density rises with age in boys, but not in girls.
Conclusion. Patients with myelomeningocele have decreased bone mineral density and are at risk of suffering pathologic bone fractures. The measurement of bone mineral density may help to identify those patients at greatest risk of suffering multiple fractures. The urinary calcium excretion of nonambulators was higher than that of ambulators and likely contributes to their decreased bone mineral density. Bone mineral density increases with age in boys, but not in girls.
Lower extremity sensory and motor deficits are found frequently in patients with myelomeningocele. These deficits can impair normal ambulation and lead to variable degrees of physical inactivity, osteoporosis, and development of pathologic fractures.1–6 Such fractures lead to a vicious cycle in which further inactivity continually heightens the risk of repeated fractures. Frequent spine surgeries followed by spica cast immobilization also exacerbate the development of such pathologic fractures.3,,4
Despite the high risk of fractures in children with myelomeningocele, little is known about their bone mineral density and its relationship to fractures. Rosenstein et al found that the neurologic level of spinal cord involvement and ambulatory status affected bone mineral density of both the upper and the lower extremities as measured by125I photon absorptiometry of the distal radius, tibia, and first metatarsal, respectively.7 Specifically, ambulatory patients who had sacral involvement had greater bone mineral density than those who were nonambulatory and had higher level thoracic involvement. However, this study only examined and compared bone mineral density measurements between different subgroups of patients with myelomeningocele. No comparisons of bone mineral density were made with those of normal children and no biochemical markers of bone metabolism were examined.
In this study, we report the results of bone mineral density measurements in patients with myelomeningocele and compare them with results for North American children. Measures of metabolic bone activity also are examined.
Sixty patients with myelomeningocele were randomly chosen from a roster of >700 patients with myelomeningocele followed at the Texas Scottish Rite Hospital for Children in Dallas, TX. Patients with metabolic acidosis, renal insufficiency, or other metabolic bone disease (ie, hyperparathyroidism) were excluded from this study. Of these 60 patients, 43 consented to participate in the study. Because known bone mineral density standards in children are available for the 6- to 19-year-old age group, we limited our study to the 35 patients (18 males and 17 females) in this age range. The remainder were younger than 6 years of age. The average age was 11.1 ± 3.6 years.
The ambulatory status of individuals within this group varied greatly and was categorized into three broad groups, defined as follows8: 1) full-time wheelchair (FTWC) patient, nonambulator who uses a wheelchair full time; 2) limited ambulator (LA), individual who walks with physical limitations and usually only for short distances (home, classroom, or during physical therapy); uses a wheelchair for longer distances; and 3) full-time ambulator (FTA), individual who walks everywhere and never needs a wheelchair.
Of the 35 study subjects, 12 were FTWC, 10 were LA, and 13 were FTA patients.
The level of spinal cord involvement in our study patients included 10 with thoracolumbar involvement, 20 with midlumbar involvement, and 5 with lumbosacral involvement. Within each ambulatory category, the level of spinal cord involvement was as follows: for FTWC patients, 10 had thoracic involvement and 2 had lumbar involvement; for LA patients, 10 had lumbar involvement; and for FTA patients, 8 had lumbar involvement and 5 had sacral involvement.
In all study patients, blood and urine samples were collected at the Texas Scottish Rite Hospital for Children and sent for analysis at either the hospital's clinical laboratory or the Nichols Institute (San Juan Capistrano, CA). In all 35 patients, laboratory evaluation of the blood sample consisted of measurement of serum electrolytes, including BUN and creatinine, venous blood gas, parathyroid hormone (PTH), di-hydroxy 1,25 vitamin D, and serum osteocalcin (OSTEO). PTH and OSTEO levels were measured at the Nichols Institute via a immunochemiluminometric assay and radioimmunoassay, respectively. Osteocalcin is a small peptide released into circulation during bone mineralization and therefore represents bone formation. The di-hydroxy 1,25 vitamin D level was measured at the Nichols Institute via a radioreceptor assay after extraction from serum and chromatography separation. Urine was collected over a 24-hour period and sent for measurement of pyridinoline and deoxypyridinoline levels. Urinary pyridinolines (PYR) and deoxypyridinolines (dPYR) are excreted as a result of the breakdown of collagen in bone and therefore is a marker of bone resorption. PYR and dPYR values are measured at the Nichols Institute via high performance liquid chromatography.
Bone Densitometry Measurements
Bone density measurements were obtained with a Norland Digital Model 278 125I absorptiometer (Norland, Fort Atkinson, WI). A highly collimated low energy 125I photon beam (27.4 keV) and its corresponding scintillation detector are scanned across the forearm on opposite sides of the distal radius (one-third the distance from the ulnar styloid process to the olecranon). Changes in the beam intensity noted after traversing the radius are inversely proportional to bone mineral content and bone mineral density.9,,10 The absolute bone mineral content also varies according to the size of the forearm. By measuring bone width and calculating bone mineral content/bone width, a corrected or “normalized” bone mineral density value is obtained.9,,10 Absorptiometric determinations of bone mineral content are very accurate (1% to 3% error) and reproducible (<2% error).9–11 Bone width and bone mineral content/bone width are calculated by bone densitometer.
z Scores were calculated for bone mineral density measurements via the following formula12: zscore = (patient's measurement − population mean)/standard deviation (SD).
Different parameters such as bone mineral content, bone mineral density (bone mineral content/bone width), and biochemical markers of bone metabolism were compared using analysis of variance.12 The frequency of such abnormal biochemical markers within each ambulatory group was compared with Fisher's exact test.12 The urinary calcium excretion between ambulators and nonambulators was compared using an unpaired t test. Correlation between bone mineral density versus age for the different sexes were obtained via linear regression.12 Linear regression lines were compared via comparison of the two β coefficients.
The study was approved by the institutional review board at the University of Texas Southwestern Medical Center. Informed consent was obtained from all parents or guardians.
Bone Mineral Density
z Scores of bone mineral content and bone mineral density for all patients in each ambulatory category are shown in Fig 1. Bone mineral content and bone mineral density for all patients with myelomeningocele in each of the three ambulatory categories are 1 to 2 SD units below the mean for the normal population. The z scores of bone mineral content and bone mineral density were nearly identical across all three ambulatory categories.
Eight of our 35 patients (5 FTWC and 3 LA patients) suffered 19 lower extremity fractures. Bones fractured include the femur (11), metatarsal bones (2), patella (4), and lower tibia/fibula (2). The number of fractures per patient ranged from one in 2 patients to four in 2 patients. Bone mineral density z score of the 8 patients with bone fractures (−3.13 ± 1.3 SD) were significantly lower than that for the remaining patients (−1.03 ± 0.38 SD;P < .05) (Fig 2). Bone mineral density z scores among the 8 patients with multiple fractures ranged from −0.46 to −10.22.
Biochemical Markers of Bone Metabolism
The serum electrolytes and venous blood gas measurements in all patients were within normal limits. The average values for all biochemical markers of bone metabolism within each ambulatory category are shown in Table 1. Table 2 shows normal values for biochemical markers of bone metabolism. PYR and dPYR are collagen cross-link byproducts of bone breakdown and are excreted during bone reabsorption. In contrast, OSTEO (bone Gla protein) is released into circulation during bone formation.
Within each specific ambulatory group, a few patients were noted to be outside the normal range for age (Table 1). In the FTWC group (12 patients), 3 subjects had elevated pyridinoline and deoxypyridinoline levels, 1 had an elevated PTH level, 4 had elevated urinary calcium excretions, and 2 had elevated vitamin D levels, whereas 1 had decreased osteocalcin and 1 had decreased vitamin D levels. In the LA group (10 patients), 3 subjects had elevated PYR levels, 2 had elevated dPYR levels, 1 had an elevated PTH level, and 1 had elevated urinary calcium excretion. In the FTA group (13 patients), only 1 patient each had an elevated deoxypyridinoline level, a PTH level, and a urinary calcium excretion level, whereas 3 patients had elevated vitamin D levels, and 1 had a decreased pyridinoline level. Using Fisher's exact test to compare the biochemical parameters of bone metabolism, abnormally elevated PYR levels were found more frequently in both the FTWC and the LA subjects than in the FTA subjects (P = .05). Urinary calcium excretion also was significantly higher in the FTWC subjects when compared with the ambulators (combination of the LA plus the FTA subjects) (P < .05).
Relationship of Bone Mineral Density With Increasing Age
Bone mineral density in normal children rises with increasing age until its peak at 19 to 20 years of age. Bone mineral density at any given age usually is higher in boys than in girls, except between the ages 12 and 14 years, when the girls have already undergone pubertal maturation.10 We examined the relationship of bone mineral density with age in the boys and girls with myelomeningocele separately. As shown in Fig 3, bone mineral density in boys with myelomeningocele is below normal, but nevertheless, rises with advancing age until its peak at 19 years of age (r = 0.49; P ≤ .05). In contrast, there is no significant rise in bone mineral density in girls with advancing age (Fig 3) (r = 0.24;P > .35). A comparison of the two regression lines reveals that they are statistically different with two different β coefficients (P = .005).
Since the 1960s, the falling mortality rate of patients with myelomeningocele has allowed increasing numbers to survive into adolescence and early adulthood.13,,14 Among the many new challenges faced by these patients are osteoporosis and pathologic bone fractures.1–6 The frequency and severity of such bone fractures in patients with myelomeningocele has been well documented by a number of orthopedists caring for these children. Quilis reported 55 fractures of the lower extremities occurring in 15 of 130 children, and Drennan and associates reported 58 fractures occurring in 25 of 84 children.1,,2 Similarly, James reported 44 lower extremity fractures occurring in 22 of 122 patients.4 All authors felt that disuse of the limbs and overall physical inactivity contributed greatly to the increased risk of bone fractures.1,,2,4 Likewise, we found 19 lower extremity fractures occurring in 8 of 35 patients. Despite these reports of frequent bone fractures in patients with myelomeningocele, few quantitative studies of bone mineral content or bone mineral density have been conducted in this patient population. Yet, bone mineral density has been shown to correlate with bone strength and the incidence of fractures.15,,16 This study is the first to examine and compare bone mineral density (via single photon absorptiometry) of patients with myelomeningocele to that of normal children.
Single photon absorptiometry is a well substantiated technique to measure bone mineral density in children and is more accurate and reproducible than radiographic morphometry.9,,10 Mazess et al studied 763 North American children and established normative bone density data of the upper extremities as measured by photon absorptiometry for the 6- to 19-year-old age group.9,,10 He found that bone mineral density (bone mineral content/bone width) increased with age and peaked at 18 to 19 years of age.
Using single photon-absorptiometry in the distal radius of our patient population, bone mineral density of all patients was 1 to 2 SD units below the mean of the normal population (Fig 1). There was a tendency for bone mineral density z scores for FTWC subjects (−1.94 ± 1.11) to be lower than those for FTA subjects (−1.15 ± 0.41), although the difference was not statistically significant. Likely factors responsible for the low bone mineral density are sedentary lifestyle and physical inactivity caused by the ambulatory difficulties of patients with myelomeningocele. The effects of sedentary lifestyle and physical inactivity on bone fractures and bone mineral density have been examined in several studies. James found that in the asymmetrically affected patients, the nonactive or paralytic limb was more likely to fracture compared with the active limb.4 More recently, Lin and colleagues examined hemiplegic patients with cerebral palsy and found that bone mineral content and bone mineral density of the nonactive limb were 5.6% and 21% lower than measures for the active limb.17 Rosenstein et al also studied patients with myelomeningocele and found that nonambulatory patients had lower bone mineral densities than did ambulatory patients.7 Unlike in our study, however, no comparisons of bone mineral density were made with that in normal children. These studies, nevertheless, underscore the importance of physical inactivity or “disuse” of a limb in contributing to both the decreased bone mineral density and the greater risk of fractures in our patients. The association of decreased bone mineral density and pathologic fractures is highlighted in Fig 2. Patients with multiple fractures had greatly diminished bone mineral density compared with those without fractures. Bone mineral density measurements may be an important clinical tool to help identify those patients at risk for multiple pathologic bone fractures. The affordability, low maintenance cost, and ease of use of a modern single photon absorptiometer should help to make bone mineral density measurements readily available. Newer single photon absorptiometers do not require handling of radioactive material and require only minimal training for operation. No trained technician is required, and single photon absorptiometers are typically used by office or clinic personnel.
Disuse of a limb, especially of the lower extremities, however, is unlikely to be the sole factor involved in decreased bone mineralization in patients with myelomeningocele. In our study, we found decreased bone mineral density in the radius of the upper extremity, an extremity often used extensively, especially in FTWC subjects. Overall physical inactivity may have a systemic effect on total body bone mineralization and influence the mineralization of bones not affected directly and locally by the neurologic lesions. Such a systemic effect, rather than a direct and localized effect, also is suggested by the large decrease in bone mineral density of the distal radius across all three ambulatory groups of our patient population (Fig 1). Rosenstein and coworkers found that bone mineral density of the distal radius in nonambulatory patients with myelomeningocele was reduced by 31% over those who were ambulatory, again arguing for a systemic effect of physical inactivity on bone mineralization.7 Similar reductions in bone mineral density were seen in the weight-bearing lower extremities (tibia and first metatarsal) between nonambulators and ambulators.7 Thus, both direct and localized factors as well as indirect and systemic factors affect total body bone mineralization. It has been suggested that a direct neuropathic effect of the neurologic lesions might contribute to decreased bone mineralization.1,,17 However, the decreased bone mineral density of the distal radius, a normally innervated limb in our patients with myelomeningocele, argues against this possibility.
Only one other study has compared bone mineral density of children with neurologic deficits to that of normal children. Henderson et al examined patients with cerebral palsy and found that bone mineral density of the lumbar spine and the proximal femur of these patients also was 1 SD unit below that for age-matched control subjects.18 These findings are in agreement with results of bone mineral densities obtained in our study.
Disorders of bone metabolism may be associated with abnormal values of OSTEO or PYR.19 We examined these biochemical bone parameters and found that FTWC and LA subjects had elevated PYR levels (and therefore greater bone reabsorption) more frequently compared with FTA subjects. No differences in deoxypyridinoline, PTH, vitamin D, and osteocalcin levels exist among the different ambulatory groups. Henderson et al also were unable to find any correlation between serum vitamin D levels and OSTEO with bone mineral density in patients with cerebral palsy.18 In contrast, low OSTEO levels is a useful marker of excessive bone loss in postmenopausal women.16,,19
FTWC patients have higher urinary calcium excretion than the combination of LA and FTA patients. FTWC patients are less physically active than their ambulatory counterparts and therefore suffer from “relative immobilization” and immobilization hypercalciuria.
Bone mineral density in normal children of both sexes increases with age until it peaks in early adulthood.9,,10 In our cohort of boys, bone mineral density was 1 to 2 SD units below normal, but, nevertheless, increased with age until the observed peak at 19 years. It is interesting, however, that bone mineral density in girls demonstrates no significant change between the ages of 6 and 19 years. The divergent rise in bone mineral density versus age seen in the boys may be related to a central role of testosterone in postpubertal skeletal maturation. The absence of a rise in bone mineral density versus age in girls may be related to the lack of high levels of serum testosterone. Indeed, osteoporosis in adult men is usually related to testosterone deficiency.20 Alternatively, cultural biases in the different levels of physical activity encouraged between the sexes also may play a role in this divergence in bone mineral density.
It is unknown at this time whether pharmacologic therapy with inhibitors of urinary calcium excretion (ie, thiazide diuretics) or inhibitors of bone reabsorption (ie, bis-phosphonates) would be of help in increasing bone mineral density or decreasing the incidence of multiple bone fractures. Use of thiazide diuretics might be particularly helpful in lowering urinary calcium excretion in FTWC patients. The efficacy of bis-phosphonates to increase bone mineral density seems promising as was demonstrated recently in three nonambulatory patients with cerebral palsy.21 In general, this population of patients would seem appropriate for future therapeutic trials to improve bone disease and quality of life.
In summary, patients with myelomeningocele have decreased bone mineral density (1 to 2 SD units below norm) from 6 to 19 years of age. This reduced bone mineral density may, in large part, explain the higher incidence of bone fractures in patients with myelomeningocele. Measurement of bone mineral density may be an important clinical tool to help identify those patients particularly at high risk for pathologic fractures. This low bone mineral density was associated with a greater frequency of abnormally elevated PYR levels in the FTWC and LA subjects versus the FTA subjects. There also were higher levels of urinary calcium excretion in the nonambulators versus the ambulators. Bone mineral density increased with age in boys, but not in girls.
This study was supported by a grant from the Texas Scottish Rite Hospital for Children of Dallas.
We thank Janell McQuinn and Anne Goodrich for their excellent secretarial assistance; Theresa Weaver, LVN, for her assistance with the bone densitometer; and Richard Browne, PhD, for his assistance with the statistical analysis.
- Received January 26, 1998.
- Accepted April 17, 1998.
Reprint requests to (A.Q.) Department of Pediatrics, University of Texas Southwestern Medicine Center, 5323 Harry Hines Blvd, Dallas, TX 75235-9063.
- FTWC =
- full-time wheelchair patient •
- LA =
- limited ambulator •
- FTA =
- full-time ambulator •
- PTH =
- parathyroid hormone •
- OSTEO =
- serum osteocalcin •
- PYR =
- urinary pyridinolines •
- dPYR =
- urinary deoxypyridinolines
- ↵Anshuetz RH, Freehafer AA, Shaffer JW, Dixon MS Jr. Severe fracture complications in myelodysplasia. J Pediat Orthop. 1984;4:22–24
- James CCM
- ↵Menelaus MB. The Orthopaedic Management of Spina Bifida Cystica. New York, NY: Churchill Livingstone; 1980:61
- Hoffer M,
- Feiwell E,
- Perry R,
- Perry J,
- Bonnett C
- ↵Mazess RB, Cameron JR. Bone mineral content in normal US whites. Presented at the International Conference on Bone Mineral Measurement. National Institutes of Health, US Dept of Health, Education and Welfare; 1973; 75–683
- ↵Chesney RW, Mazess RB, Rose PG, Jax DK. Bone mineral status measured by direct photon absorptiometry in childhood renal disease. Pediatrics. 1977;60:864–872
- ↵Dixon WJ, Massey FJ. Introduction to Statistical Analysis. New York, NY: McGraw Hill;
- Delmas PD
- ↵Jackson JA. Osteoporosis in men. In: Favus MJ, ed. Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism. 2nd ed. New York, NY: Raven Press; 1993:241
- Shaw NJ,
- White CP,
- Fraser WD,
- Rosenbloom L
- Copyright © 1998 American Academy of Pediatrics