Abstract
Dual energy x-ray absorptiometry (DEXA) is the current standard for measuring bone mineral density (BMD) in children. The International Society for Clinical Densitometry recommends scanning the total body and spine in children. However, in orthopedics, the total-body and spine DEXA scans are often rendered useless by the presence of metallic hardware and/or contractures.
The lateral distal femoral DEXA scan was developed as a scan mode for children such as those with cerebral palsy who have contractures or metallic implants, which make it impossible to do total-body or spine scans. Unlike other DEXA scans, a single scan of the lateral distal femoral illustrates the density of the metaphyseal cancellous (newer) bone, the transitional, and the cortical (older) bone in 1 image. Because of this, we hypothesized that an individual lateral distal femoral scan could provide a map of bone health over time.
The lateral distal femoral scans of 40 children whose bone growth was tainted by distant chemotherapy (chemotherapy group) were compared to the lateral distal femoral scans of 40 children whose bone environment had remained relatively stable over time (control group).
The hypothesis was not confirmed by the data. The “Z-score difference,” the difference between the Z-scores of the cancellous and cortical bone, for the chemotherapy group (0.16) and the control group (0.32) were not statistically different.
While these results did not confirm the hypothesis, the lateral distal femoral scan remains a reproducible and useful DEXA scan in pediatric orthopedic clinical practice.
While dual-energy x-ray absorptiometry (DEXA) scanning is the gold standard for assessment of bone density in children, clinical use of this tool is in its infancy. Interpretation of pediatric DEXA scans is complicated by differential growth and maturation rates of children. Peak bone density is not reached until the third decade,1 the norm against which to measure is continuously changing, and disagreement exists about how scanning should be done.1-6
The International Society for Clinical Densitometry recommends that regions of interest for DEXA scanning in children should be the total body and spine.2 In a clinical setting, especially orthopedics, this is impractical: many physically challenged children (those most in need of DEXA scanning) have contractures that make positioning for lumbar scan impossible, or they have metallic orthopedic implants or other implants such as baclofen pumps that render the total body scan inaccurate.4
The lateral distal femoral scan was developed for children for whom standard whole-body or spine scans cannot be done.3 The child lies on his or her side, a position that is comfortable and nonthreatening (Figure 1). Finding it highly reproducible in sequential scans, even in children with spastic quadriplegia, we have expanded the use of the lateral distal femoral to virtually all children.
Other regions used in DEXA scanning are composite zones: the hip and the spine scan each superimpose cancellous and cortical, physeal and metaphyseal bone in 1 image to display a single bone density measurement and give a single corresponding Z-score. In contrast, a single lateral distal femoral scan shows the density of the metaphyseal cancellous, the transitional, and the cortical bone in 1 view. Each region produces an individual bone density measurement and Z-score.
We asked if a single lateral distal femoral DEXA scan could reveal the history of a child’s bone growth. The cortical bone (R3) is the oldest bone, the transitional bone (R2) is of intermediate age, and the metaphyseal cancellous bone (R1) is most recently formed by the physis and subject to the most recent health history.
To test our hypothesis, we sought a population who had experienced an osteotoxic environment that persisted for a clearly defined period of time and was now gone. Could we see bad bone in the cortical region and better bone health in the metaphyseal region reflected by improved Z-scores?
The Lateral Distal Femoral Scan in Children With Distant Pediatric Cancers
Long-term studies have demonstrated decreased bone mineral density (BMD) in many children who have undergone chemotherapy for pediatric cancers. At our center, children who have undergone chemotherapy are routinely evaluated for low BMD using DEXA scans, including spine, hip, and lateral distal femoral regions.
The Z-score compares a child’s BMD to an age- and gender-matched mean5,6: in theory, there would be no reason for the Z-score to vary from cancellous to cortical region in a child with a stable health milieu. Children who have had osteotoxic chemotherapy, often including steroids, which are also detrimental to bone density, have not had a stable health milieu: indeed, the lateral distal femoral scans of some of these patients demonstrate an upside down pattern of Z-scores on the lateral distal femoral scan. In the upside down pattern, the cortical bone, which corresponded to bone formed during the time the child received chemotherapy, had the lowest Z-score, and the transitional and cancellous regions demonstrated progressively higher Z-scores. This suggests that BMD improvements after the termination of chemotherapy occurred in a chronological order corresponding to growth. If this were consistently seen, it would affirm that evolving bone health is reflected in the lateral distal femoral scan.
Background
At our institution, pediatric DEXA scanning is performed for clinical and research purposes. All DEXA scans are performed by a certified DEXA technologist using a Hologic Delphi W scanner (Hologic, Inc, Bedford, Massachusetts). Normal scan modes include the spine, hip, and lateral distal femoral. Scans are interpreted by a board-certified, fellowship-trained pediatric orthopedic surgeon (E.A.S.) who is also certified by the International Society for Clinical Densitometry as a clinical densitometrist. Pediatric reference data for the spine and hip are based on the reports of Faulkner et al7 and Southard et al,8 while lateral distal femoral Z-scores are calculated using Henderson et al’s4 database.
The lateral distal femoral scan is performed on both femurs unless an orthopedic implant or recent fracture in the bone precludes scanning of that region. The child lies on his or her side and the leg is positioned to image the femur from midthigh to knee (Figure 1). The technician maps the region of interest manually (Figure 2). The physician enters the BMD as demonstrated by the densitometer into Henderson et al’s4 database to arrive at a Z-score for the cancellous, transitional, and cortical regions of each distal femur (Table 1).
Children are referred to our center for DEXA scanning for many clinical indications, including cerebral palsy, muscular dystrophy, cystic fibrosis, osteogenesis imperfecta, and multiple fractures. Healthy children are scanned as part of various research protocols.
All children having completed treatment for pediatric cancers are referred for routine DEXA scanning, which includes the lateral distal femoral scan mode.
Children who undergo DEXA scanning are asked if their scan may be used in a research database. If they and their parents or guardians consent, they become part of the Carrie Tingley Pediatric Bone Health Project database.
Materials and Methods
Approval was obtained from the Human Research Review Committee of the University of New Mexico Health Sciences Center to perform a retrospective study of DEXA scans of the lateral distal femoral region. We compared the lateral distal femoral scans of children whose bone growth was tainted by distant chemotherapy (chemotherapy group) to the lateral distal femoral scans of children with stable health situations such that their bone environment had remained relatively stable over time (control group).
The control group included healthy children who had undergone DEXA scanning for any reason, such as inclusion in research studies, and other children whose bone health milieu we judged to be stable over time. Children with conditions such as scoliosis were included as long as there had been no changes in medications, diet, activities, or general health in the previous 5 years. Excluded from the control group were children on medications affecting bone density such as steroids, seizure medications, or hormones. Also excluded from the study were children who had fractured either femur or who had orthopedic implants in the distal femur.
One scan for each child was selected. If several existed, the most recent lateral distal femoral scan available was used. For the chemotherapy group, scans were performed between 6 months and 15 years following termination of chemotherapy.
For both groups, the right and left distal femoral Z-scores were averaged for the 2 bone regions: cortical and cancellous. If only 1 femur was scanned, those Z-scores were used.
Statistical Analysis
On each scan we subtracted the Z-score of the cancellous bone from that of the cortical bone to give 1 score, the Z-score difference, for each subject. These were grouped into the chemotherapy group and the control group and compared to each other using a Student t test. Statistical calculations were made with Statgraphics Plus for Windows Version 4.1 (Manugistics, Inc, Rockville, Maryland). Data management was carried out using Microsoft Excel 2002 (Microsoft Corporation, Redmond, Washington). Two-tailed tests and a Type I error rate of 0.05 were used throughout.
Results
Eighty-six subjects were evaluated in this study, ranging in age from 3 to 17 years (Table 2).
We expected that a disparity would be evident in the chemotherapy group, where older bone formed during chemotherapy (cortical bone, farthest from the physis) would be weak, while bone formed recently (juxtaphyseal cancellous bone) would be relatively stronger as reflected by a better Z-score. In children in the control group with stable bone situations, we expected to see no difference in Z-scores.
Our hypothesis was not confirmed by the data. The Z-score difference for the chemotherapy group (0.16) and the control group (0.32) differed but were not statistically different (P=.42; 95% confidence interval –0.23 to 0.54). In fact, the Z-score difference was greater in the control group than in the chemotherapy group, the opposite of our expectation.
Table 3 illustrates the mean Z-scores for each region, cortical and cancellous, in both the chemotherapy and control groups. Of note is that for both groups of patients, the Z-scores cluster at the mean for age and gender.
Our study had an 80% probability of detecting a difference in Z-scores of 0.55, assuming a Type I error rate of 0.05, a group size of 43, and a standard deviation of 0.9.
Discussion
Pediatric DEXA scanning presents challenges as well as promise as a tool in the treatment of children’s bone conditions. For the first 2 decades of life, children’s bones are changing and developing, and the rate at which this occurs depends entirely on the individual.1 There is no densitometric definition for osteoporosis in children because it has not been determined at what Z-score a child is at increased risk for fracture given the variations in bone morphology, age, maturity, and demands of living.1,9 Densitometry offers a valuable clinical tool, and the greater the understanding of it, the more it offers to the clinical practitioner.
Whole-body, spine, and hip DEXA scans radiographically project an amalgam of cortical and cancellous elements within a specific bone area, such that one cannot identify structural characteristics unique to each region. Diseases and medications affect cortical and cancellous bone selectively and differently, so the ability of a scan to differentiate between the different regions is valuable.9 It was our intent to explore the use of the lateral distal femoral in the chronology of bone health, thinking that an individual scan would provide a map of bone health over time.
Scant literature exists on the benefits of the lateral distal femoral scan, although in small studies it has been shown to be a reliable measure of bone density. In 2002, Henderson et al4 published pediatric reference data for the lateral distal femoral scan using 256 healthy children from 3 to 18 years. The distal femur correlated strongly with bone density in the proximal femur (r>0.90) and the lumbar spine (r>0.83). Khoury and Szalay5 noted that the lateral distal femoral DEXA scan could be performed on patients who could not be scanned at hip and spine due to spinal instrumentation, scoliosis, hip dislocations, or previous surgery. Harcke et al,3 examining the lateral distal femoral scans of 34 children with cerebral palsy, found that it was well tolerated and an excellent alternative to whole-body, hip, and spine scans.
In a 2006 review of pediatric DEXA scanning, 25 of 119 children (21%) referred for DEXA scanning had either metallic implants or contractures such that lumbar spine or whole-body studies were not possible.6 All of these children were successfully scanned using the lateral distal femoral regional scan.
Our initial hypothesis—that an individual lateral distal femoral scan might provide a chronological history of bone health—was not confirmed by the data. This suggests that older cortical bone may be metabolically active in children to a degree comparable to that of newer cancellous bone, so that over time the changing metabolic environment affects both areas. It also may suggest that our control group did not have the stable bone health situations that we originally thought. Some of the children in the control group had diseases that affected bone metabolism, and this may have contributed to the greater Z-score difference in that group. It may also be that this study was retrospective and nonrandomized, had a relatively small sample size, and was a single-event comparison rather than a longitudinal study. Nevertheless, despite a null hypothesis, the fact that the 2 groups were statistically similar attests to the reliability and reproducibility of the lateral distal femoral scan.
Conclusion
Pediatric health problems are increasingly acknowledged as the genesis of adult osteoporosis. As more children survive childhood cancers and other chronic illnesses, the demand for accurate monitoring of long-term effects on bone health increases. While the single lateral distal femoral scan may not appear to document a chronology of bone health through years of growth as we had hoped, it is again shown to be a simple and reproducible tool with which to measure bone density in children. Increasing experience with this technique in children may yet confirm our hypothesis that the lateral distal femoral scan can demonstrate a chronology of bone health over time. This phenomenon may require a specific physiologic scenario that our clinical vignette did not adequately expose. Regardless, more widespread use of and familiarity with the lateral distal femoral scan will lead to expansion of existing databases, further refining the accuracy and increasing the acceptance of this technique.
References
- Sawyer AJ, Bachrach LK. Rationale for bone densitometry in childhood and adolescence. In: Sawyer AJ, Bachrach LK, Fung EB, eds. Bone Densitometry in Growing Patients: Guidelines for Clinical Practice. Totowa, NJ: Humana Press; 2007:1-13.
- Khan AA, Bachrach L, Brown JP, et al. Standards and guidelines for performing central dual-energy x-ray absorptiometry in premenopausal women, men, and children. J Clin Densitom. 2004; 7(1):51-64.
- Harcke HT, Taylor A, Bachrach S, Miller F, Henderson RC. Lateral femoral scan: an alternative method for assessing bone mineral density in children with cerebral palsy. Pediatr Radiol. 1998; 28(4):241-246.
- Henderson RC, Lark RK, Newman JE, et al. Pediatric reference data for dual energy x-ray absorptiometric measures of normal bone density in the distal femur. AJR Am J Roentgenol. 2002; 178(2):439-443.
- Khoury DJ, Szalay EA. Bone mineral density correlation with fractures in nonambulatory pediatric patients. J Pediatr Orthop. 2007; 27(5):562-566.
- Szalay EA, Harriman D. Adapting pediatric DEXA scanning to clinical orthopaedics. J Pediatr Orthop. 2006; 26(5):686-690.
- Faulkner RA, Bailey DA, Drinkwater DT, McKay HA, Arnold C, Wilkinson AA. Bone densitometry in Canadian children 8-17 years of age. Calcif Tissue Int. 1996; 59(5):344-351.
- Southard RN, Morris JD, Mahan JD, et al. Bone mass in healthy children: measurement with quantitative DEXA. Radiology. 1991; 179(3):735-738.
- Crabtree NJ, Leonard MB, Zemel BS. Dual-energy x-ray absorptiometry. In: Sawyer AJ, Bachrach LK, Fung EB, eds. Bone Densitometry in Growing Patients: Guidelines for Clinical Practice. Totowa, NJ: Humana Press; 2007:41-58.
Authors
Ms Tryon and Dr Szalay are from the University of New Mexico Carrie Tingley Hospital, Albuquerque, New Mexico.
Ms Tryon received a research scholarship from the Carrie Tingley Hospital Foundation; and Dr Szalay has no relevant financial relationships to disclose.
Correspondence should be addressed to: Elizabeth Szalay, MD, Department of Orthopedics & Rehabilitation, UNM Carrie Tingley Hospital, 1127 University Blvd NE, Albuquerque, NM 87102.