A Modification to the McHale Procedure Reduces Operative Time and Blood Loss.

BACKGROUND: Treatment of symptomatic spastic hip dislocations in adolescent patients with cerebral palsy includes a variety of described salvage type procedures. In 1990, McHale and colleagues described a technique involving a femoral head resection, valgus-producing proximal femoral osteotomy, and advancement of the lesser trochanter into the acetabulum. We have modified this technique in 3 ways by: performing it in the lateral position with a more posterior approach, not advancing the lesser trochanter into the acetabulum, and closing the capsule over the acetabulum. The purpose of this paper is to describe our technique and to compare the results to Castle type procedures and McHale procedures performed as originally described. METHODS: We retrospectively reviewed all salvage type procedures performed at our institution for spastic hip dislocations in children with cerebral palsy from 2003 to 2013. Preoperative and postoperative pain, estimated blood loss, operative time, length of stay in the hospital, and postoperative pelvis radiographs were reviewed for heterotopic ossification formation and proximal femoral migration. RESULTS: Twenty-six patients with 30 hip procedures were reviewed. The modified McHale technique had shorter operative times when compared with the supine McHale technique and the Castle procedure (134, 171, and 139 min, respectively). There was a trend toward less blood loss in the modified McHale technique, but this was not significant. There was no difference in length of stay in the hospital. The majority of McHale patients (>63%) had pain relief postoperatively, where half of the Castle patients required a revision surgery for pain (4 of 8). There was less heterotopic ossification seen in the modified McHale technique (6.25%) when compared with supine McHale and Castle techniques (both 50%). However, there was more proximal femoral migration in the modified McHale group. CONCLUSIONS: The modified McHale technique is faster with otherwise equivocal results in the immediate operative periods. There is less heterotopic bone formation but more proximal femoral migration with this new technique. LEVEL OF EVIDENCE: Level IV-case series.

Calf circumference discrepancies in patients with unilateral clubfoot: Ponseti versus surgical release.

BACKGROUND: Talipes equinovarus is the most common congenital lower limb abnormality. Decreased calf size has been found to have negative impacts on patients' subjective appraisals of long-term outcomes. This study compares calf circumference ratios in 2 groups of patients with unilateral clubfoot, those treated according to the Ponseti method and those treated with extensive surgery, to determine whether the current standard of care achieves better anatomic outcomes. METHODS: Patients >1 year after treatment for unilateral clubfoot were recruited during normal follow-up appointments and both calves were measured using a standardized protocol. A questionnaire concerning their treatment history was also completed. Data were analyzed by comparing calf circumference ratios between treatment modalities. RESULTS: Thirty-five patients with unilateral clubfoot were recruited after satisfying inclusion criteria. Twenty-four (69%) were included in the Ponseti-managed group, and 11 (31%) were in the extensive surgery group. The affected legs were on average 3% to 10% smaller than the control legs across all groups. The surgery group's average calf ratio was significantly less at 90.8%±3.5% compared with 94.4%±3.3% in the Ponseti group. CONCLUSIONS: The calf circumference of limbs affected by clubfoot is significantly smaller in those treated with extensive surgery as compared with those treated with the Ponseti method alone, with or without percutaneous tenotomy. This supports the Ponseti method as the standard of care for achieving the most favorable anatomic outcome. LEVEL OF EVIDENCE: Level I.

Bisphosphonate use in children with pediatric osteoporosis and other bone conditions.

Bisphosphonates (BPs) are used most commonly in children with osteogenesis imperfecta, resulting in increased trabeculae and cortical thickness, increased bone density as measured by DXA (Dual Energy X-ray Absorptiometry), and improved vertebral morphology. Less well documented in controlled trials are decrease in long bone fractures, improved strength and motor function, and decreased pain. Outside of children with osteogenesis imperfecta, use of bisphosphonates in children is increasing, all of which is off-label. This is seen in children with other chronic conditions resulting in pediatric osteoporosis and insufficiency fractures. Additional indications include steroid dependency with progressive loss of bone density, avascular necrosis of bone, and chronic regional pain syndrome. This review highlights the potential benefits and risks of the use of bisphosphonates in these unique children at risk for fracture or bone collapse.

The use of the Tsuge procedure for pedal macrodactyly: relevance in pediatric orthopedics.

Pedal macrodactyly is a rare clinical entity that poses a challenge to practicing pediatric orthopedic surgeons. Many treatment options have been proposed. In 1967, Kenya Tsuge proposed a method to decrease the length, width, and circumference of a macrodactylous digit, while maintaining the cosmetic benefit of keeping the nail. We retrospectively reviewed our experience with using this technique in four children (six toes) over a 4-year period. The surgery is described and our results reviewed. We believe that the Tsuge procedure is a technically feasible, effective, single-stage reconstructive technique for pedal macrodactyly that pediatric orthopedic surgeons should have in their armamentarium.

Positive communication paradigm decreases early recurrence in clubfoot treatment.

BACKGROUND: The Ponseti method has become the treatment standard for idiopathic clubfoot. Deformity recurrence is most commonly attributed to premature abandonment of the requisite abduction orthosis. A study in 2009 from our center revealed a high rate of deformity recurrence in our patient population. It was surmised that the importance of bracing to maintain correction had not been adequately communicated to some families, especially Native Americans. As a result, the principal investigator developed a different communication protocol for parents of infants. METHODS: All children treated for clubfoot at the University of New Mexico Carrie Tingley Hospital, Albuquerque, NM, from 2008 to 2010 were reviewed. They were compared with a historical control group from this institution, the subjects of the 2009 study, and were analyzed for the rate of recurrence and Pirani score improvement. RESULTS: Our study cohort comprised 69 infants (104 clubfeet), all of whom were treated with the new communication style. The recurrence rate for the new communication paradigm was 2.88% compared with 18.2% in the control group (P<0.001). The Pirani score improvement was 4.0 in the treatment group compared with 3.5 in the control group (P=0.001). Native American recurrence was zero in the treatment group and 41% in the control group (P=0.011). CONCLUSIONS: A positive, rather than a negative communication style, emphasis on the brace as the most important aspect of treatment, and a more culturally sensitive family education paradigm, resulted in a lower rate of deformity recurrence when treating children with clubfeet using the Ponseti method. LEVEL OF EVIDENCE: Level III.

Dual 8-plate technique is not as effective as ablation for epiphysiodesis about the knee.

BACKGROUND: Lower extremity length inequality can be problematic in children and is often addressed surgically. Several techniques have traditionally been utilized for epiphysiodesis, the goal being physeal ablation. Recently, 8-plates, initially developed for hemiepiphysiodesis, have been extended to epiphysiodesis by placing the plates on both medial and lateral sides of the physis. No prior studies have compared 8-plates with physeal ablation techniques. METHODS: Between January 2003 and August 2009, 27 patients underwent epiphysiodesis surgery using either physeal ablation or 8-plate technique. Sixteen patients had physeal ablation and 11 had dual 8-plates. A retrospective chart review sought demographic data, outcomes, and complications. Radiographs were reviewed to measure pretreatment and posttreatment limb lengths. RESULTS: The median improvement in limb length discrepancy was 15.5 mm in the physeal ablation group and 4 mm in the 8-plate group (P<0.001). This difference was maintained following linear regression factoring out the effect of time (10.78 mm for ablation vs. 5.62 mm for 8-plates; P=0.016). There was no statistically significant difference in complication rate between the groups (P=0.112). CONCLUSIONS: Our study demonstrated physeal ablation to be a significantly superior treatment compared with dual 8-plates for epiphysiodesis. Despite theoretical advantages of 8-plates to perform epiphysiodesis about the knee, this study does not recommend the use of medial and lateral 8-plates to effect epiphysiodesis. LEVEL OF EVIDENCE: Therapeutic III.

Prevention of postoperative osteopenia using IV pamidronate: a pilot study.

BACKGROUND: Postoperative bone mineral density (BMD) loss, especially after cast immobilization and/or non-weight-bearing, is a well-known phenomenon in children that can cause fracture. Children with marginal bone density are at greatest risk. This prospective randomized control trial compared the effect of single-dose intravenous (IV) pamidronate versus placebo to prevent postoperative BMD loss. METHODS: Children between the ages of 4 and 18 were included in the study; inclusion criteria included a predisposition to low bone density and hip or lower extremity surgery that would require cast immobilization or non-weight-bearing for at least 4 weeks. Dual-energy x-ray absorptiometry (DXA) scans of the lumbar spine and bilateral distal femora were performed preoperatively and at least 4 weeks postoperatively. Subjects were randomized to receive either a single, low dose of IV pamidronate (1 mg/kg) or placebo, given during the immediate postoperative period. Changes in the BMD were compared using the Mann-Whitney test for significance in the lumbar spine. A multivariate general linear model was used to compare the effect of surgery, DXA region, and treatment on BMD. RESULTS: A total of 24 subjects were included in the study, and 20 completed the protocol. Pamidronate-treated subjects showed a statistically significant difference with a median gain in BMD of 0.029 gm/cm in the lumbar spine compared with the control group, which showed a median loss of 0.025 gm/cm. Treatment did not have a statistically significant effect on BMD loss in the distal femur but trended toward decreased BMD loss (treatment=0.0331 gm/cm, control=0.0416 gm/cm). There were no complications or adverse reactions. CONCLUSIONS: This small pilot study shows that single-dose postoperative pamidronate mitigated postoperative BMD loss in at-risk children, which may in turn decrease postoperative fracture risk. Further investigation into the use of IV pamidronate in postoperative patients is warranted. LEVEL OF EVIDENCE: Level 1 double-blinded randomized control trial.

Relative osteopenia after femoral implant removal in children and adolescents.

Radiographic osteopenia is regularly observed after implant removal from a fracture or femoral osteotomy but has not been objectively quantified. Hardware removal is generally performed months to years after the index event (fracture or osteotomy) when full activity has been resumed. Objectively demonstrable bone mineral deficiency affects fracture risk. Hardware removal may facilitate the return to normal bone mineral density. Children who had dual-energy X-ray absorptiometry scans following femoral implant removal were retrospectively reviewed to assess the percent of change in bone mineral density and change in Z-scores. The femoral neck and the lateral distal femora were scanned, comparing the operated side with unaffected femur as a control. Sixteen children were included. Patients demonstrated up to 15.4% (average, 4.8%) less bone mineral density in the femoral neck region, up to 43% less (average, 16.5%) in the metabolically active distal metaphyseal region, and up to 18.1% less (average, 6.3%) in the transitional region. No statistical difference was noted in the diaphy-seal region. A statistically significant decrease in Z-score was noted when plate and screw constructs (average change, -0.97 SD) as compared with intramedullary nail constructs (average change, -0.33 SD) were used. Children can exhibit statistically significant decreases in bone mineral density in the femoral neck and distal femur following femoral implant removal, with plate and screw constructs demonstrating a greater effect than intramedullary (load sharing) devices. This has implications for return to activity and suggests that implant removal may be important in restoration of bone strength in children.

Efficacy of alendronate in the treatment of low bone density in the pediatric and young adult population.

BACKGROUND: Pediatric osteoporosis is uncommon but can result in painful and debilitating insufficiency fractures. Treatment options for osteoporosis in children are few. Bisphosphonate therapy for children has not been approved by the Food and Drug Administration (FDA) in the United States, but its use in that population has been increasing. Randomized controlled studies have not been done because of the small subject pool and the difficulty in randomizing a child with an insufficiency fracture to a placebo arm of a study. This retrospective case-control study of a population of children with primarily neuromuscular disease was done to review changes in bone mineral density as reflected by dual x-ray absorptiometry (DXA) scanning. METHODS: Medical records and DXA scans were screened to identify children with low bone density who had been treated with alendronate as well as similar control subjects with low bone density for their age who had not received alendronate. Medication acquisition was confirmed by refill records, and cumulative exposure was calculated. Interval DXA scans were reviewed to correlate bone mineral density change in grams per square centimeter as well as the percent change and percent change over time for both alendronate-treated and control subjects. RESULTS: Twenty-eight alendronate-treated subjects and thirty control subjects met the inclusion criteria. No significant improvement in bone mineral density was seen in the alendronate-treated subjects as compared with the control subjects. Some patients in both groups exhibited marked improvement, with improvement of >31% seen only in the alendronate-treated subjects. CONCLUSIONS: Alendronate does not reliably improve bone density in children and young adults with primarily neuromuscular disease and without osteogenesis imperfecta. Individual patients treated with bisphosphonates must be carefully followed to ensure medication compliance and appropriate response.

Pediatric vitamin D deficiency in a southwestern luminous climate.

BACKGROUND: Few studies look at vitamin D levels in children living in sunny climates as it is assumed that they receive adequate vitamin D from sun exposure. In light of changing lifestyles of children and studies documenting vitamin D deficiency among children in extreme climates, a study to examine vitamin D levels in healthy children living in a luminous climate was conducted. METHODS: A retrospective chart review of vitamin D levels in healthy children with vague musculoskeletal pain (such as "growing pains") was done. Healthy children, specifically without musculoskeletal pain, were prospectively recruited as controls. RESULTS: Eighty-eight children, 42 children with "pain" and 46 controls were studied. No statistical difference in vitamin D levels was found between the "pain" group (mean vitamin D level 29.1 ng/mL) and the control group (mean vitamin D level 32.4 ng/mL, P<0.52). Overall, 14% of the entire group had levels <20 ng/mL, 49% had levels <30 ng/mL, and 15% had levels >40 ng/mL. CONCLUSIONS: A consensus has yet to be established as to what an "optimal" vitamin D level is for growing children to develop strong bones for a lifetime. This study demonstrated that 14% of children living in a sunny climate had vitamin D levels below 20 ng/mL, a level universally accepted as insufficient, and 49% were below 30 ng/mL, arguably a "desired" level. A sunny climate does not assure vitamin D sufficiency. Virtually all children should be supplemented, with laboratory follow-up for those at high risk for low bone density/those with insufficiency fractures.

Children with spina bifida are at risk for low bone density.

BACKGROUND: Patients with spina bifida frequently sustain lower extremity fractures which may be difficult to diagnose because they feel little or no pain, although the relative contributions of low bone density to pain insensitivity are unclear. Routine dual-energy xray absorptiometry (DXA) scanning is unreliable because these patients lack bony elements in the spine, and many have joint contractures and/or implanted hardware. QUESTIONS/PURPOSES: We asked (1) if the lateral distal femoral scan is useful in spina bifida; (2) whether nonambulatory children with spina bifida exhibit differences in bone mineral density (BMD) compared with an age-and-sex-matched population; and (3) whether Z-scores were related to extremity fracture incidence. METHODS: We retrospectively reviewed 37 patients with spina bifida who had DXA scans and sufficient data. Z-scores were correlated with functional level, ambulatory status, body mass index, and fracture history. RESULTS: The distal femoral scan could be performed in subjects for whom total body and/or lumbar scans could not be performed accurately. Twenty-four of 37 had Z-scores below -2 SD, defined as "low bone density for age." Ten of 35 patients (29%) with fracture information had experienced one or more fractures. Our sample size was too small to correlate Z-score with fracture. CONCLUSION: We believe BMD should be monitored in patients with spina bifida; nonambulatory patients with spina bifida and those with other risk factors are more likely to have low bone density for age than unaffected individuals. The LDF scan was useful in this population in whom lumbar and total body scans are often invalidated by contracture or artifact. Although lower extremity fractures occur regardless of ambulation or bone density, knowing an individual's bone health status may lead to interventions to improve bone health.

Femoral nerve palsy and hip instability in infants with breech birth presentation: a review of the literature and report of 2 cases.

BACKGROUND: Femoral nerve palsy has been described as a result of hip hyperflexion during orthotic treatment for developmental dysplasia of the hip, but femoral nerve palsy in newborns who have not had brace treatment has not been reported. METHODS: Two cases of femoral nerve palsy after breech lie in utero are reviewed. Neither of these infants had undergone treatment of any sort for the hip. RESULTS: In both children, Ortolani-positive hip dislocation was encountered. Both were treated with abduction orthoses with lesser hip flexion than the Pavlik-type brace. Nerve recovery was complete, and the hip dysplasia resolved satisfactorily. CONCLUSIONS: Femoral nerve palsy may be seen after breech lie in utero and may be associated with hip instability. Treatment considerations for the hip must take into account the nerve compromise. LEVEL OF EVIDENCE: Level IV, case series.

The relationship between fractures and DXA measures of BMD in the distal femur of children and adolescents with cerebral palsy or muscular dystrophy.

Children with limited or no ability to ambulate frequently sustain fragility fractures. Joint contractures, scoliosis, hip dysplasia, and metallic implants often prevent reliable measures of bone mineral density (BMD) in the proximal femur and lumbar spine, where BMD is commonly measured. Further, the relevance of lumbar spine BMD to fracture risk in this population is questionable. In an effort to obtain bone density measures that are both technically feasible and clinically relevant, a technique was developed involving dual-energy X-ray absorptiometry (DXA) measures of the distal femur projected in the lateral plane. The purpose of this study is to test the hypothesis that these new measures of BMD correlate with fractures in children with limited or no ability to ambulate. The relationship between distal femur BMD Z-scores and fracture history was assessed in a cross-sectional study of 619 children aged 6 to 18 years with muscular dystrophy or moderate to severe cerebral palsy compiled from eight centers. There was a strong correlation between fracture history and BMD Z-scores in the distal femur; 35% to 42% of those with BMD Z-scores less than -5 had fractured compared with 13% to 15% of those with BMD Z-scores greater than -1. Risk ratios were 1.06 to 1.15 (95% confidence interval 1.04-1.22), meaning a 6% to 15% increased risk of fracture with each 1.0 decrease in BMD Z-score. In clinical practice, DXA measure of BMD in the distal femur is the technique of choice for the assessment of children with impaired mobility.

Physeal stapling versus 8-plate hemiepiphysiodesis for guided correction of angular deformity about the knee.

BACKGROUND: Angular deformity in the lower extremities results in cosmetic deformity, gait disturbance, pain, and early joint degeneration. Corrective osteotomy is the gold standard for angular deformity, but is a major surgical intervention with significant incidence of complication. For these reasons, hemiepiphysiodesis is an attractive alternative in the growing child to allow "guided growth" to correct the angular deformity. Physeal stapling has proven success, but hardware prominence or failure has been problematic. Recently, the tension band plate construct ("8-plate") has been promoted for hemiepiphysiodesis, citing ease of surgical technique and more rapid rate of correction. We sought to test the claim that the 8-plate effected a more rapid correction of angular deformity with a lower complication rate. METHODS: Hemiepiphysiodesis for angular deformity in 63 lower extremities from 2000 to 2007 were retrospectively reviewed. Thirty-nine limbs received staple hemiepiphysiodesis and 24 received 8-plate hemiepiphysiodesis. Angular measurements were compared preoperatively, during the first year postoperatively, and at the time of hardware removal or skeletal maturity. Complications requiring additional surgery for the correction of angular deformity were noted in each group. RESULTS: There was no difference between the 2 groups in the rate of correction (approximately 10 degrees/y, P=0.48). Complication rates were similar (12.8% vs. 12.5%, P=1.0). Patients with abnormal physes (eg, Blount disease, skeletal dysplasias) had a higher complication rate (27.8% vs. 6.7% for patients with normal physes, P=0.04) with no difference between the 8-plate and staple groups (P=1.0). The patients in the 8-plate group were significantly younger than those in the staple group (P=0.04). CONCLUSIONS: The 8-plate is as effective as staple hemiepiphysiodesis for guided correction of angular deformity with respect to rate of correction and complications, even in somewhat younger patients. Higher complication rates are observed in patients with pathologic physes.

Effect of cultural factors on outcome of Ponseti treatment of clubfeet in rural America.

BACKGROUND: Nonoperative management of clubfoot with the Ponseti method has proven to be effective, and it is the accepted initial form of treatment. Although several studies have shown that problems with compliance with the brace protocol are principally responsible for recurrence, no distinction has been made with regard to whether the distance from the site of care affects the early recurrence rate. We compared early recurrence after Ponseti treatment between rural and urban ethnically diverse North American populations to analyze whether distance from the site of care affects compliance and whether certain patient demographic characteristics predict recurrence. METHODS: One hundred consecutive infants with a total of 138 clubfeet treated with the Ponseti method were followed prospectively for at least two years from the beginning of treatment. Early recurrence, defined as the need for subsequent cast treatment or surgical treatment, and compliance, defined as strict adherence to the brace protocol described by Ponseti, were analyzed with respect to the distance from the site of care, age at presentation, number of casts needed for the initial correction, need for tenotomy, and family demographic variables. RESULTS: Of eighteen infants from a rural area who had early recurrence, fourteen were Native American. The families of these children, like those of all of the children with early recurrence, discontinued orthotic use earlier than was recommended by the physician. Discontinuation of orthotic use was related to recurrence, with an odds ratio of 120 (p < 0.0001), in patients living in a rural area. Native American ethnicity, unmarried parents, public or no insurance, parental education at the high-school level or less, and a family income of less than $20,000 were also significant risk factors for recurrence in patients living in a rural area. Intrinsic factors of the clubfoot deformity were not correlated with recurrence or discontinuation of bracing. CONCLUSIONS: Compliance with the orthotic regimen after cast treatment is imperative for the Ponseti method to succeed. The striking difference in outcome in rural Native American patients as compared with the outcomes in urban Native American patients and children of other ethnicities suggests particular problems in communicating to families in this subpopulation the importance of bracing to maintain correction. An examination of communication styles suggested that these communication failures may be culturally related.

A new look at the incidence of slipped capital femoral epiphysis in new Mexico.

PURPOSE: Past epidemiological studies demonstrated a nearly fivefold lower incidence of slipped capital femoral epiphysis (SCFE) in New Mexico compared with Connecticut. A recent study demonstrated some regional variability but did not address this earlier finding. We sought to reexamine the incidence of SCFE in New Mexico to improve the understanding of the epidemiology and ultimately the disorder itself. METHODS: The discharge databases for the 11 major medical centers in the state were reviewed for the ICD-9 code for SCFE (732.2) for 1995 to 2006. The data were analyzed by comparison with the 2000 New Mexico census data. The incidence data are reported as cases per 100,000 boys aged 10 to 17 years and girls aged 8 to 15 years, as per Kelsey's original article. RESULTS: The incidence of SCFE in New Mexico for the study period was 5.99. This is a doubling of the reported incidence in the 1960s (2.13) and represents a statistically significant change (P < 0.001). More detailed analysis of our data demonstrated a statistically significant increase during 3-year intervals: 1995-1997, 2.27; 1998-2000, 2.75; 2001-2003, 4.73; and 2004-2006, 7.38. The mean age of onset was 12.2 years. There was a male to female ratio of incidence of 1.94:1. Relative frequencies by race were as follows: 4.63x for African Americans, 2.20x for Hispanics, and 2.20x for Native Americans. A preponderance of cases was treated at the state's only tertiary pediatric orthopaedic center: 168 to 15 in the remaining 10 centers. CONCLUSIONS: The incidence of SCFE has increased dramatically in New Mexico since Kelsey's epidemiological study in 1970. Obesity is a patient factor that has changed over this same period. According to the National Health and Nutrition Examination Survey Data for 2003/2004, the rates of obesity have tripled since 1971. In New Mexico, 25% of high-school children are estimated to be overweight. However, according to a recent study examining a national database (compiled from 27 states), the national incidence of SCFE remained fairly constant at 10.8 per 100,000.Interestingly, as more patients are seen at a tertiary center for children's orthopaedics, the rate of diagnosis in New Mexico has risen to resemble national trends. In the 1960, that center was located in a remote site and did not provide acute care for children's musculoskeletal issues. Increased obesity in children and improved access to pediatric orthopaedic evaluation may have contributed to a significant increase in reported incidence of SCFE in New Mexico.

Adolescents with idiopathic scoliosis are not osteoporotic.

STUDY DESIGN: Case controlled study. OBJECTIVE: To explore the relative effects of body mass index (BMI) and the presence or absence of adolescent idiopathic scoliosis (AIS) on bone mineral density (BMD) as evidenced by Z-scores in adolescents. SUMMARY OF BACKGROUND DATA: Prior studies have identified adolescents with idiopathic scoliosis as having "osteoporosis" or "osteopenia," when only a small percentage of subjects in these studies actually had bone density that was clinically abnormal. The terms osteoporosis and osteopenia as used in adults cannot be applied to adolescents and children, as fracture risk has not been well correlated to Z-scores. As we had noted that our scoliosis patients of normal and heavy weight had normal Z-scores, this study was undertaken to explore the relationship of bone mineral density to body mass index in adolescents with and without scoliosis. METHODS: Dual energy x-ray absorptiometry (DXA) scans of 49 adolescents with adolescent idiopathic scoliosis were compared to 40 normal control adolescents. Z-scores were compared to reduce variability when comparing subjects of varying age and genders. Student t test or simple linear regression was used to explore relationships between Z-scores and clinical and demographic variables. RESULTS: In both groups of subjects, Z-score was most strongly correlated with BMI (P < 0.001). The presence of scoliosis had the effect of lowering the Z-score as if the individual had "lost" 3.4 BMI units. CONCLUSION: Z-scores in subjects with and without scoliosis were most strongly correlated to BMI: thin patients had lower bone density, heavy patients had higher. The presence of scoliosis had an effect similar to subtracting 3.4 "BMI units," lowering the Z-score from what might otherwise be predicted. The "scoliosis effect" may be noticeable in thin individuals, pushing them to the "low for age" level, whereas in heavier individuals, the effect is negligible. No subjects in either group met the ISCD definition for osteoporosis.