- Email: [email protected]
Kyphosis and fractures in children and young adults with cystic fibrosis
Kyphosis and fractures in children and young adults with cystic fibrosis Richard C. Henderson, MD, PhD, a n d Barbara B. S p e c t e r , MD From the Departments of Orthopaedics, Pediatrics, and Radiology, University of North Carolina, Chapel Hill The purpose of this study was to examine children and adolescents with cystic fibrosis for an increased frequency of fracture and excessive thoracic kyphosis, which may result from inadequate skeletal mineralization. In a survey of 143 patients (ages 4.7 to 21.9 years; mean, 11.3 years), the fracture rate for male patients from birth to 5 years of a g e was higher than for female patients and both rates were c o m p a r a b l e with those for normal children. In contrast, female patients 6 to 16 years of a g e with cystic fibrosis had a higher-than-normal fracture rate and a higher rate than their male counterparts. Review of the chest radiographs showed that thoracic kyphosis correlated with a g e and with disease severity as judged by Brasfield scoring. In the >1S-year-old a g e group, kyphosis exceeding 40 degrees, the upper limit of normal, was found in 77% of the female patients and 36% of the male patients. The cause of these findings is uncertain and perhaps multifactorial, but osteopenia is likely a contributing factor. As the life expectancy of patients with cystic fibrosis continues to increase, the skeletal consequences, particularly in female patients, may b e c o m e increasingly significant. (J PEDIATR1994;125:208-12)
The major potential clinical consequence of osteopenia is fracture. In the spine this can result in wedge-shaped deformity of the thoracic vertebral bodies and produces the excessive kyph0sis commonly found in the elderly. In patients with cystic fibrosis a high prevalence of diminished bone mineralization has been documented with single-photon absorptiometry of the forearm and quantitative computed tomography of the lumbar spine. 13 If there is osteopenia in a large percentage of children and adolescents with cystic fibrosis, this may be clinically manifest as overt fractures and as increased thoracic kyphosis. The purpose of this study was tO examine a large population of children and adolescents with cystic fibrosis to detect these clinical problems.
Submitted for publication Nov. 17, 1993; accepted Feb. 22, 1994. Reprint requests: Richard C. Henderson, MD, 237 Burnett-Womack Building, CB No. 7055, Chapel Hill, NC 27599-7055. Copyright ® 1994 by Mosby-Year Book, Inc. 0022-3476/94/$3.00 + 0 9/20/55475
208
METHODS The cystic fibrosis database maintained by the department of pediatrics at the University of North Carolina provided a listing of 150 patients with birthdays between Jan. 1, 1970, and Dec. 31, 1986. All the families were mailed a short questionnaire asking whether the patient had ever sustained a fracture. A follow-up questionnaire and direct telephone or personal contact in our clinic were used to obSee related article, p. 201. tain information from families who did not initially respond. Ultimately, 143 (95.3%) of the 150 eligible subjects were contacted. The 71 male and 72 female respondents ranged in age from 4.7 to 21.9 years (mean + SD, 11.3 + 4.5 years). A total of 36 patients responded that they bad sustained a fracture and all were contacted directly to discuss their fracture(s). Information obtained included the location of the fracture, date and mechanism of injury, treatment rendered, and ultimate functional outcome. One fracture
The Journal of Pediatrics Volume 125, Number 2
Henderson and Specter
209
Table II. Fracture rates
Table I. Location of fracture Site
No. of fractures
Hand/carpals Radius/ulna Humerus Clavicle Foot Tibia Femur Axial skeleton TOTAL
7 20 8 2 1 3 1 2 44
occurred in a patient with severe spastic quadriplegic cerebral palsy who had never ambulated. His skeletal health was likely more dependent on the cerebral palsy and he was excluded from further analyses. Another respondent was uncertain that he had sustained a wrist fracture; the radiologic report at the treating hospital stated "no fracture seen," and this patient was moved to the negative respondent group. There remained 34 patients who had one or more fractures and 108 patients in the group without a fracture. To assess the thoracic kyphosis of this population, the most recent lateral radiograph of the chest was measured according to the standard method of Cobb by one of us (R.C.H.).4 Of the initial 150 potential subjects, the one with severe cerebral palsy was excluded and three subjects did not have chest radiographs available. Thoracic kyphosis of more than 40 degrees is considered abnormal for a population of children and young adults. 5-7 The Brasfield scoring system 8 was used to categorize the 146 lateral and corresponding posteroanterior chest radiographs. Decreasing Brasfield scores correlate with increasing severity of lung disease) The Brasfield scoring was performed by the other author (B.B.S.) without knowledge of the kyphosis measurement. The severity of lung disease is judged by various radiographic findings. From a maximum score of 25, points are deducted for the degree of hyperinflation, peribronchial thickening, nodular cystic structures (bronchiectasis), areas ofatelectasis or pneumonia, and an assessment of overall severity. Statistical analyses. The continuous variables of age, kyphosis, and Brasfield scores were correlated by means of linear regression analyses. The relation between kyphosis and Brasfield scores was further analyzed independent of age by use of a partial correlation analysis. Kyphosis categorized as normal or greater than 40 degrees was also compared between sexes and by age groups with the MantelHaenszel chi-square test. Comparison fracture rates reported for the normal population are based on the National Center for Health Statistics, National Health Interview Survey 1970-1977, which included more than 1 million households. 9
Cystic Fibrosis Study Group Fracture Total rate Fractures person(No./lO0 Normal (No.) years* person-years) subjectst Age birth to 5 yr Male 9 Female 3 Age 6-16 yr Male 13 Female 19
416 426
2.2 0.7
1.7 0.8
345 386
3.8 4.9
4.3 3.0
*Totalnumberof person-yearsprovidedby the studypopulationof 70 male and 72 femalesubjects. tNormal annualfracturerates per 100 personsbasedon data fromthe National Health InterviewSurvey? RESULTS There were 44 fractures among the 34 patients who reported that they had sustained a fracture. Fractures involving more than one bone in the same area, such as the radius and ulna, or multiple fractures in the same bone were counted as a single fracture. Two patients had fractures in different extremities as a consequence of a single injury. As in the report based on the National Health Survey, these were counted as separate fracture events.9 An additional five patients had two fractures each and one patient had three fractures in unrelated injuries. Fractures were grouped according to site of fracture (Table I) and age at fracture and sex (Table II). Upper extremity fractures accounted for the majority (37 fractures, 84%). The oldest age at fracture was 16.5 years, but only 15% of the study group (21 subjects) were older than 16 years of age. The majority of the fractures were treated with a simple cast or splint (38 fractures, 86%). At the time of the survey, all patients except one, who had sustained his fracture just 6 weeks earlier, reported complete recovery from their injury without sequelae. The mechanism of injury was most commonly falling from standing height, usually while engaged in play activities (18 fractures, 41%). Another frequent mechanism of injury was a fall from a height, such as playground equipment (16 fractures, 36%). In all cases except one, the resulting fracture was reasonably consistent with the reported mechanism of injury. The exception was a 16-yearold female adolescent who fractured her femur by falling while playing tennis; such an injury does not typically produce the force necessary to fracture the femur of a teenager. Fracture rates typically vary with sex and age. Table II shows the fracture rates for male and female patients with cystic fibrosis in the age groups birth to 5 years and 6 to 16 years. For both sexes the fracture rate was lower in the
2 10
Henderson and Specter
The Journal of Pediatrics August 1994
70 U
60
0
50
o
(13
°!
40 0 "113. >._ v
.......
: .....
o
-o. . . . . . om
•
==|
30
o
oo"
20
......
o o
•
•
•
o o
o ~
o
o
: .......
~-
o •
o
o • o
o =• • o=
=wn •
- ..............
o o
,~,
• o
. o
o
mm
o •
o=
.....
o • o • • o o=% o • • •ooo o o o• • =•o • o o • o o
o o
o
•
0-2---:
•
o1= o
• o o
• o
o
o
o
10
0
I
I
I
L
I
I
I
I
I
I
4
6
8
10
12
14
16
18
20
22
.
ACE
Figure. Age versus thoracic kyphosis measured by lateral chest radiographs of 72 male (O) and 74 female (IN) patients with cystic fibrosis. Dashed line at 40 degrees shows upper limit of normal kyphosis.
Table III. Thoracic kyphosis Male subjects
Female subjects
Age
No. of subjects
No. >40 degrees* (%)
Kyphosis't (degrees)
No. of subjects
No. >40 degrees* (%)
Kyphosist (degrees)
<10 yr 10-15 yr >15 yr
37 24 11
2 (5%) 2 (8%) 4 (36%)
27.7 -- 1.3 29.8 + 1.6 35.2 + 4.1
36 25 13
3 (8%) 7 (28%) 10 (77%)
28.4 +_ 1.3 35.4 +_ 2.1 43.7 + 3.0
Kyphosis
values
are
*Mantel-Haenszel tTwo-way
analysis
expressed chi-square of variance
as mean test
+_ S E .
for kyphosis
by sex and
age
>40 group:
degrees
by sex and
sex, p = 0.004;
age
age
group;
group,
younger age group. In the younger age group the fracture rate was higher in male than in female subjects, but the relation was reversed in the 6- to 16-year-old age group. There were no fractures after 16 years of age; however, there were only 43 person-years of fracture risk. The thoracic kyphosis is shown as a function of age in the Figure and is categorized by age and sex in Table III. Of the 146 patients, 28 (19.2%) had kyphosis exceeding 40 degrees, the upper limit of normal, s7 Linear regression analyses showed a positive relation between kyphosis and age (p = 0.0001). In the age groups 10 to 15 years and >15 years, the prevalence of excessive kyphosis and the average kyphosis were higher in female patients than in male patients. In patients less than 10 years of age the thoracic kyphosis was similar in male and female patients. Thoracic kyphosis was greatest in the female patients older than 15
sex, p = 0.01;
age
group,
p = 0.01.
p = 0.0001.
years of age, in whom mean kyphosis was 43.7 degrees and 77% had kyphosis greater than 40 degrees. Only 36% of the male patients in this same age group had kyphosis greater than 40 degrees. Two-way analysis of variance showed the differences in kyphosis between the sexes (p = 0.004) and age groups (p = 0.0001) to be statistically significant. Similarly, the prevalence of excessive kyphosis greater than 40 degrees differed between sexes (p = 0.01) and age groups (p = 0.01; Mantel-Haenszel chi-square test). The Brasfield scores, like kyphosis, were found to correlate with age. Linear regression analyses showed a negative relation; scores decreased as age increased. Separate analyses for male and female patients did not reveal a significant difference between sexes (p = 0.8). Because both kyphosis and Brasfield scores correlate with age, partial correlation analysis was used to examine the relation between
The Journal of Pediatrics Volume 125, Number 2
these two variables independent of age. Increasing kyphosis was found to correlate with increasing severity of disease as judged by Brasfield scores (p = 0.002). DISCUSSION The fractures sustained by this group of patients with cystic fibrosis were typical for fractures in a normal pediatric population. Upper extremity fractures were the most common and trauma during routine play activities was the most common mechanism of injury. The fractures readily healed, usually with just simple treatment measures, and there were no permanent sequelae. Fracture rates in normal pediatric populations significantly vary with age and sex. The American Academy of Orthopaedic Surgeons, based on data from the National Health Interview Survey, has reported annual fracture rates for male and female patients in the age groups birth to 5 years and 6 to 16 years.? In children less than 6 years of age and male patients 6 to 16 years of age, the cystic fibrosis and normal populations had similar fracture rates. However, in female patients 6 to 16 years of age the annual fracture rate in normal persons is 3.0 fractures per 100 persons compared with 4.9 fractures per 100 person-years in the population with cystic fibrosis. Although the fracture rate in female patients 6 to 16 years of age with cystic fibrosis appeared to be higher than normal, the lack of a parallel control group within this study somewhat limits the reliability of this observation. The potential impact of differences in study design can be minimized, however, by determining relative fracture rates within each study. In normal pediatric populations, female patients have a lower fracture rate than male patients. In the report from the American Academy of Orthopaedic Surgeons, the relative fracture rate for female to male patients aged 6 to 16 years is 0.70. 9 Similarly, in a population-based study in Rochester, Minn., the relative fracture rate for normal female to male patients aged 5 to 14 years was 0.66. l° In sharp contrast, a higher fracture rate was found in female patients 6 to 16 years of age with cystic fibrosis than in male patients. Thus female adolescents with cystic fibrosis have an elevated fracture rate compared with ageand sex-matched normal subjects and compared with agematched male patients with cystic fibrosis. Whether a fracture results from a traumatic episode depends on several factors, including the force of the injury and the mechanical strength of the bone. Differences in fracture rates do not necessarily indicate differences in skeletal fragility. For example, the higher fracture rate in healthy boys compared with girls is not the result of a weaker skeleton. However, it seems improbable that the apparently increased fracture rate in female adolescents with cystic fibrosis is the result of increased participation in
Henderson and Specter
211
trauma-producing activities. The case of our 16-year-old adolescent girl who sustained a femur fracture with a fall from standing height further suggests that the increased fracture rate is related to diminished skeletal strength. Excessive thoracic kyphosis of more than 40 degrees was found in 19% of patients with cystic fibrosis. This is similar to the findings of Denton et al., 7 who found kyphosis greater than 40 degrees in 15.l% of 91 children and adolescents with cystic fibrosis. In our study excessive kyphosis was uncommon in children less than 10 years of age but increased in prevalence with age; a similar relation has been observed by others. 11 The cause of excessive kyphosis in cystic fibrosis is uncertain and is possibly multifactorial. Denton et al. 7 suggested that the excessive work of breathing leads to fatigue, causing the shoulders and spine to slump forward. Rose et a1.12 found that many patients with cystic fibrosis have limited trunk flexibility and, in some cases, weakness of trunk musculature. They suggested that this may promote a flexed posture of the spine. Another possible explanation for the excessive kyphosis is that it occurs with hyperinflation of the lungs. 3 These factors may all contribute to apparent kyphosis, but they alone cannot account for the wedgeshaped deformity of the thoracic vertebral bodies seen in this study and others. TM 12 Other theories on the cause of the excessive kyphosis attempt to explain the deformity of the vertebrae. 11, 12 Compressive forces are known to inhibit bone growth and can lead to angular deformities.13 Postural factors, hyperinflation, or other unrecognized factors in cystic fibrosis could produce excessive compression on the anterior portion of the vertebral bodies, causing a growth abnormality that ultimately results in wedge-shaped deformity and kyphosis. Osteopenia could contribute to both the elevated fracture rate and the excessive kyphosis found in cystic fibrosis. Several studies have documented a high prevalence of decreased skeletal mineralization in patients with cystic fibrosis)-3, 14, 15 Consistent with the findings of this study, Mischler et a l ) measured bone mineralization in patients with cystic fibrosis and concluded that adolescent girls were the subgroup at greatest risk for decreased mineralization. Others have also found female patients to be more affected than male patients] 4 but not all such studies have found a sex difference), 15 The reasons for sex differences in fracture rate, kyphosis, and bone mineralization are unknown. Female patients with cystic fibrosis also tend to have a shorter life span than male patients, and this is unexplained. 16 Osteoporosis in the elderly is characterized by loss of calcium from the skeleton. In children and adolescents with cystic fibrosis the condition develops because accretion of mineral does not occur normally. Diminished longitudinal
2 12
Henderson and Specter
growth of the skeleton, manifested as short stature, is additional evidence that normal skeletal development can be adversely affected in cystic fibrosis, ~7, 18 Many of the potential causes of osteopenia in cystic fibrosis are interrelated and include delayed puberty and endocrine development, malabsorption, low serum vitamin D levels, poor nutrition, decreased physical activity, and chronic respiratory acidosis.l-3, 17,19, 2o The cause of the increased fracture rate and kyphosis is uncertain and possibly multifactoral. However, osteopenia, is one of the most plausible explanations. As the life expectancy of patients with cystic fibrosis continues to improve, consequences of the condition not apparent at a young age may become clinically quite significant at an older age. The implication of this study that clinical manifestations of osteopenia may begin to appear as early as the second decade of life strongly indicates the need for more detailed study of skeletal growth and development in cystic fibrosis. We thank Drs. Gerald Fernald, Robert Wood, and the other members of the University of North Carolina Division of Pediatric Pulmonary Medicine, and Michelle Wagner and Ralph DeMasi of the Department of Biostatistics, for their help with this study. REFERENCES
1. Mischler EH, Chesney P J, Chesney RW, Mazess RB. Demineralization in cystic fibrosis. Am J Dis Child 1979; 133:632-5. 2. Hahn T J, Squires AE, Halstead LR, Strominger DB. Reduced serum 25-hydroxyvitamin D concentration and disordered mineral metabolism in patients with cystic fibrosis. J PEDIATR 1979;94:38-42. 3. Gibbens DT, Gilsanz V, Boechat MI, Dufer D, Carlson ME, Wang C. Osteoporosis in cystic fibrosis. J PEDIATR 1988; 113:295-300. 4. Cobb R J. Outline for the study of scoliosis. Am Acad Orthop Surg 1948;5:261-75.
The Journal of Pediatrics August 1994
5. Propst-Proctor SL, Bleck EE. Radiographic determination of lordosis and kyphosis in normal, and scoliotic children. J Pediatr Orthop 1983;3:344-6. 6. Fon GT, Pitt M J, Thies AC. Thoracic kyphosis: range in normal subjects. Am J Roentgenol 1980;134:979-83. 7. Denton JR, Tietjen R, Gaerlan PF. Thoracic kyphosis in cystic fibrosis, Clin Orthop 1981;155:71-4. 8. Brasfield D, Hicks G, Soong S J, Tiller RE. The chest roentgenogram in cystic fibrosis: a new scoring system. Pediatrics 1979;63:24-9. 9. Grazier KL, Holbrook TL, Kelsey JL, Stauffer RN. The frequency of occurrence, impact, and cost of musculoskeletal conditions in the United States. Chicago: American Academy of Orthopaedic Surgeons, 1984:72-80. 10. Garraway WM, Stauffer RN, Kurland LT, O'Fallon WM. Limb fractures in a defined population. I. Frequency and distribution. Mayo Clin Proc 1979;54:701-7. 11. Erkkila JC, Warwick WJ, Bradford DS. Spine deformities and cystic fibrosis. Clin Orthop 1978;131:146-50. 12. Rose J, Gamble J, Schultz A, Lewiston N. Back pain and spinal deformity in cystic fibrosis. Am J Dis Child 1987; 141:1313-6. 13. Arkin AM, Katz JF. The effects of pressure on epiphyseal growth. J Bone Joint Surg Am 1956;38:1056-66. 14. Stead RJ, Houlder S, Agnew J, et al. Vitamin D and parathyroid hormone and bone mineralization in adults with cystic fibrosis. Thorax 1988;43:190-4. 15. Hanley JG, McKenna MJ, Quigley C, Freaney R, Muldowney FP, Fitzgerald MX. Hypovitaminosis D and response to supplementation in older patients with cystic fibrosis. Q J Med 1985;219:377-85. 16. Wood RE. Prognosis. In: Taussig LM, ed. Cystic fibrosis. New York: Thieme-Stratton, 1984:434-60. 17. Landon C, Rosenfeld RG, Short stature and pubertal delay in male adolescents with cystic fibrosis. Am J Dis Child 1984; 138:388-91. 18. Sproul A, Huang N. Growth patterns in children with cystic fibrosis. J PEDIATR 1964;65:664-76. 19. Reiter EO, Brugman SM, Pike JW, et al. Vitamin D metabolites in adolescents and young adults with cystic fibrosis: effects of sun and season. J PEDIATR 1985;106:21-6. 20. Reiter EO, Stern RC, Root AW. The reproductive endocrine system in cystic fibrosis. Am J Dis Child 1981;135:422-6.
The major potential clinical consequence of osteopenia is fracture. In the spine this can result in wedge-shaped deformity of the thoracic vertebral bodies and produces the excessive kyph0sis commonly found in the elderly. In patients with cystic fibrosis a high prevalence of diminished bone mineralization has been documented with single-photon absorptiometry of the forearm and quantitative computed tomography of the lumbar spine. 13 If there is osteopenia in a large percentage of children and adolescents with cystic fibrosis, this may be clinically manifest as overt fractures and as increased thoracic kyphosis. The purpose of this study was tO examine a large population of children and adolescents with cystic fibrosis to detect these clinical problems.
Submitted for publication Nov. 17, 1993; accepted Feb. 22, 1994. Reprint requests: Richard C. Henderson, MD, 237 Burnett-Womack Building, CB No. 7055, Chapel Hill, NC 27599-7055. Copyright ® 1994 by Mosby-Year Book, Inc. 0022-3476/94/$3.00 + 0 9/20/55475
208
METHODS The cystic fibrosis database maintained by the department of pediatrics at the University of North Carolina provided a listing of 150 patients with birthdays between Jan. 1, 1970, and Dec. 31, 1986. All the families were mailed a short questionnaire asking whether the patient had ever sustained a fracture. A follow-up questionnaire and direct telephone or personal contact in our clinic were used to obSee related article, p. 201. tain information from families who did not initially respond. Ultimately, 143 (95.3%) of the 150 eligible subjects were contacted. The 71 male and 72 female respondents ranged in age from 4.7 to 21.9 years (mean + SD, 11.3 + 4.5 years). A total of 36 patients responded that they bad sustained a fracture and all were contacted directly to discuss their fracture(s). Information obtained included the location of the fracture, date and mechanism of injury, treatment rendered, and ultimate functional outcome. One fracture
The Journal of Pediatrics Volume 125, Number 2
Henderson and Specter
209
Table II. Fracture rates
Table I. Location of fracture Site
No. of fractures
Hand/carpals Radius/ulna Humerus Clavicle Foot Tibia Femur Axial skeleton TOTAL
7 20 8 2 1 3 1 2 44
occurred in a patient with severe spastic quadriplegic cerebral palsy who had never ambulated. His skeletal health was likely more dependent on the cerebral palsy and he was excluded from further analyses. Another respondent was uncertain that he had sustained a wrist fracture; the radiologic report at the treating hospital stated "no fracture seen," and this patient was moved to the negative respondent group. There remained 34 patients who had one or more fractures and 108 patients in the group without a fracture. To assess the thoracic kyphosis of this population, the most recent lateral radiograph of the chest was measured according to the standard method of Cobb by one of us (R.C.H.).4 Of the initial 150 potential subjects, the one with severe cerebral palsy was excluded and three subjects did not have chest radiographs available. Thoracic kyphosis of more than 40 degrees is considered abnormal for a population of children and young adults. 5-7 The Brasfield scoring system 8 was used to categorize the 146 lateral and corresponding posteroanterior chest radiographs. Decreasing Brasfield scores correlate with increasing severity of lung disease) The Brasfield scoring was performed by the other author (B.B.S.) without knowledge of the kyphosis measurement. The severity of lung disease is judged by various radiographic findings. From a maximum score of 25, points are deducted for the degree of hyperinflation, peribronchial thickening, nodular cystic structures (bronchiectasis), areas ofatelectasis or pneumonia, and an assessment of overall severity. Statistical analyses. The continuous variables of age, kyphosis, and Brasfield scores were correlated by means of linear regression analyses. The relation between kyphosis and Brasfield scores was further analyzed independent of age by use of a partial correlation analysis. Kyphosis categorized as normal or greater than 40 degrees was also compared between sexes and by age groups with the MantelHaenszel chi-square test. Comparison fracture rates reported for the normal population are based on the National Center for Health Statistics, National Health Interview Survey 1970-1977, which included more than 1 million households. 9
Cystic Fibrosis Study Group Fracture Total rate Fractures person(No./lO0 Normal (No.) years* person-years) subjectst Age birth to 5 yr Male 9 Female 3 Age 6-16 yr Male 13 Female 19
416 426
2.2 0.7
1.7 0.8
345 386
3.8 4.9
4.3 3.0
*Totalnumberof person-yearsprovidedby the studypopulationof 70 male and 72 femalesubjects. tNormal annualfracturerates per 100 personsbasedon data fromthe National Health InterviewSurvey? RESULTS There were 44 fractures among the 34 patients who reported that they had sustained a fracture. Fractures involving more than one bone in the same area, such as the radius and ulna, or multiple fractures in the same bone were counted as a single fracture. Two patients had fractures in different extremities as a consequence of a single injury. As in the report based on the National Health Survey, these were counted as separate fracture events.9 An additional five patients had two fractures each and one patient had three fractures in unrelated injuries. Fractures were grouped according to site of fracture (Table I) and age at fracture and sex (Table II). Upper extremity fractures accounted for the majority (37 fractures, 84%). The oldest age at fracture was 16.5 years, but only 15% of the study group (21 subjects) were older than 16 years of age. The majority of the fractures were treated with a simple cast or splint (38 fractures, 86%). At the time of the survey, all patients except one, who had sustained his fracture just 6 weeks earlier, reported complete recovery from their injury without sequelae. The mechanism of injury was most commonly falling from standing height, usually while engaged in play activities (18 fractures, 41%). Another frequent mechanism of injury was a fall from a height, such as playground equipment (16 fractures, 36%). In all cases except one, the resulting fracture was reasonably consistent with the reported mechanism of injury. The exception was a 16-yearold female adolescent who fractured her femur by falling while playing tennis; such an injury does not typically produce the force necessary to fracture the femur of a teenager. Fracture rates typically vary with sex and age. Table II shows the fracture rates for male and female patients with cystic fibrosis in the age groups birth to 5 years and 6 to 16 years. For both sexes the fracture rate was lower in the
2 10
Henderson and Specter
The Journal of Pediatrics August 1994
70 U
60
0
50
o
(13
°!
40 0 "113. >._ v
.......
: .....
o
-o. . . . . . om
•
==|
30
o
oo"
20
......
o o
•
•
•
o o
o ~
o
o
: .......
~-
o •
o
o • o
o =• • o=
=wn •
- ..............
o o
,~,
• o
. o
o
mm
o •
o=
.....
o • o • • o o=% o • • •ooo o o o• • =•o • o o • o o
o o
o
•
0-2---:
•
o1= o
• o o
• o
o
o
o
10
0
I
I
I
L
I
I
I
I
I
I
4
6
8
10
12
14
16
18
20
22
.
ACE
Figure. Age versus thoracic kyphosis measured by lateral chest radiographs of 72 male (O) and 74 female (IN) patients with cystic fibrosis. Dashed line at 40 degrees shows upper limit of normal kyphosis.
Table III. Thoracic kyphosis Male subjects
Female subjects
Age
No. of subjects
No. >40 degrees* (%)
Kyphosis't (degrees)
No. of subjects
No. >40 degrees* (%)
Kyphosist (degrees)
<10 yr 10-15 yr >15 yr
37 24 11
2 (5%) 2 (8%) 4 (36%)
27.7 -- 1.3 29.8 + 1.6 35.2 + 4.1
36 25 13
3 (8%) 7 (28%) 10 (77%)
28.4 +_ 1.3 35.4 +_ 2.1 43.7 + 3.0
Kyphosis
values
are
*Mantel-Haenszel tTwo-way
analysis
expressed chi-square of variance
as mean test
+_ S E .
for kyphosis
by sex and
age
>40 group:
degrees
by sex and
sex, p = 0.004;
age
age
group;
group,
younger age group. In the younger age group the fracture rate was higher in male than in female subjects, but the relation was reversed in the 6- to 16-year-old age group. There were no fractures after 16 years of age; however, there were only 43 person-years of fracture risk. The thoracic kyphosis is shown as a function of age in the Figure and is categorized by age and sex in Table III. Of the 146 patients, 28 (19.2%) had kyphosis exceeding 40 degrees, the upper limit of normal, s7 Linear regression analyses showed a positive relation between kyphosis and age (p = 0.0001). In the age groups 10 to 15 years and >15 years, the prevalence of excessive kyphosis and the average kyphosis were higher in female patients than in male patients. In patients less than 10 years of age the thoracic kyphosis was similar in male and female patients. Thoracic kyphosis was greatest in the female patients older than 15
sex, p = 0.01;
age
group,
p = 0.01.
p = 0.0001.
years of age, in whom mean kyphosis was 43.7 degrees and 77% had kyphosis greater than 40 degrees. Only 36% of the male patients in this same age group had kyphosis greater than 40 degrees. Two-way analysis of variance showed the differences in kyphosis between the sexes (p = 0.004) and age groups (p = 0.0001) to be statistically significant. Similarly, the prevalence of excessive kyphosis greater than 40 degrees differed between sexes (p = 0.01) and age groups (p = 0.01; Mantel-Haenszel chi-square test). The Brasfield scores, like kyphosis, were found to correlate with age. Linear regression analyses showed a negative relation; scores decreased as age increased. Separate analyses for male and female patients did not reveal a significant difference between sexes (p = 0.8). Because both kyphosis and Brasfield scores correlate with age, partial correlation analysis was used to examine the relation between
The Journal of Pediatrics Volume 125, Number 2
these two variables independent of age. Increasing kyphosis was found to correlate with increasing severity of disease as judged by Brasfield scores (p = 0.002). DISCUSSION The fractures sustained by this group of patients with cystic fibrosis were typical for fractures in a normal pediatric population. Upper extremity fractures were the most common and trauma during routine play activities was the most common mechanism of injury. The fractures readily healed, usually with just simple treatment measures, and there were no permanent sequelae. Fracture rates in normal pediatric populations significantly vary with age and sex. The American Academy of Orthopaedic Surgeons, based on data from the National Health Interview Survey, has reported annual fracture rates for male and female patients in the age groups birth to 5 years and 6 to 16 years.? In children less than 6 years of age and male patients 6 to 16 years of age, the cystic fibrosis and normal populations had similar fracture rates. However, in female patients 6 to 16 years of age the annual fracture rate in normal persons is 3.0 fractures per 100 persons compared with 4.9 fractures per 100 person-years in the population with cystic fibrosis. Although the fracture rate in female patients 6 to 16 years of age with cystic fibrosis appeared to be higher than normal, the lack of a parallel control group within this study somewhat limits the reliability of this observation. The potential impact of differences in study design can be minimized, however, by determining relative fracture rates within each study. In normal pediatric populations, female patients have a lower fracture rate than male patients. In the report from the American Academy of Orthopaedic Surgeons, the relative fracture rate for female to male patients aged 6 to 16 years is 0.70. 9 Similarly, in a population-based study in Rochester, Minn., the relative fracture rate for normal female to male patients aged 5 to 14 years was 0.66. l° In sharp contrast, a higher fracture rate was found in female patients 6 to 16 years of age with cystic fibrosis than in male patients. Thus female adolescents with cystic fibrosis have an elevated fracture rate compared with ageand sex-matched normal subjects and compared with agematched male patients with cystic fibrosis. Whether a fracture results from a traumatic episode depends on several factors, including the force of the injury and the mechanical strength of the bone. Differences in fracture rates do not necessarily indicate differences in skeletal fragility. For example, the higher fracture rate in healthy boys compared with girls is not the result of a weaker skeleton. However, it seems improbable that the apparently increased fracture rate in female adolescents with cystic fibrosis is the result of increased participation in
Henderson and Specter
211
trauma-producing activities. The case of our 16-year-old adolescent girl who sustained a femur fracture with a fall from standing height further suggests that the increased fracture rate is related to diminished skeletal strength. Excessive thoracic kyphosis of more than 40 degrees was found in 19% of patients with cystic fibrosis. This is similar to the findings of Denton et al., 7 who found kyphosis greater than 40 degrees in 15.l% of 91 children and adolescents with cystic fibrosis. In our study excessive kyphosis was uncommon in children less than 10 years of age but increased in prevalence with age; a similar relation has been observed by others. 11 The cause of excessive kyphosis in cystic fibrosis is uncertain and is possibly multifactorial. Denton et al. 7 suggested that the excessive work of breathing leads to fatigue, causing the shoulders and spine to slump forward. Rose et a1.12 found that many patients with cystic fibrosis have limited trunk flexibility and, in some cases, weakness of trunk musculature. They suggested that this may promote a flexed posture of the spine. Another possible explanation for the excessive kyphosis is that it occurs with hyperinflation of the lungs. 3 These factors may all contribute to apparent kyphosis, but they alone cannot account for the wedgeshaped deformity of the thoracic vertebral bodies seen in this study and others. TM 12 Other theories on the cause of the excessive kyphosis attempt to explain the deformity of the vertebrae. 11, 12 Compressive forces are known to inhibit bone growth and can lead to angular deformities.13 Postural factors, hyperinflation, or other unrecognized factors in cystic fibrosis could produce excessive compression on the anterior portion of the vertebral bodies, causing a growth abnormality that ultimately results in wedge-shaped deformity and kyphosis. Osteopenia could contribute to both the elevated fracture rate and the excessive kyphosis found in cystic fibrosis. Several studies have documented a high prevalence of decreased skeletal mineralization in patients with cystic fibrosis)-3, 14, 15 Consistent with the findings of this study, Mischler et a l ) measured bone mineralization in patients with cystic fibrosis and concluded that adolescent girls were the subgroup at greatest risk for decreased mineralization. Others have also found female patients to be more affected than male patients] 4 but not all such studies have found a sex difference), 15 The reasons for sex differences in fracture rate, kyphosis, and bone mineralization are unknown. Female patients with cystic fibrosis also tend to have a shorter life span than male patients, and this is unexplained. 16 Osteoporosis in the elderly is characterized by loss of calcium from the skeleton. In children and adolescents with cystic fibrosis the condition develops because accretion of mineral does not occur normally. Diminished longitudinal
2 12
Henderson and Specter
growth of the skeleton, manifested as short stature, is additional evidence that normal skeletal development can be adversely affected in cystic fibrosis, ~7, 18 Many of the potential causes of osteopenia in cystic fibrosis are interrelated and include delayed puberty and endocrine development, malabsorption, low serum vitamin D levels, poor nutrition, decreased physical activity, and chronic respiratory acidosis.l-3, 17,19, 2o The cause of the increased fracture rate and kyphosis is uncertain and possibly multifactoral. However, osteopenia, is one of the most plausible explanations. As the life expectancy of patients with cystic fibrosis continues to improve, consequences of the condition not apparent at a young age may become clinically quite significant at an older age. The implication of this study that clinical manifestations of osteopenia may begin to appear as early as the second decade of life strongly indicates the need for more detailed study of skeletal growth and development in cystic fibrosis. We thank Drs. Gerald Fernald, Robert Wood, and the other members of the University of North Carolina Division of Pediatric Pulmonary Medicine, and Michelle Wagner and Ralph DeMasi of the Department of Biostatistics, for their help with this study. REFERENCES
1. Mischler EH, Chesney P J, Chesney RW, Mazess RB. Demineralization in cystic fibrosis. Am J Dis Child 1979; 133:632-5. 2. Hahn T J, Squires AE, Halstead LR, Strominger DB. Reduced serum 25-hydroxyvitamin D concentration and disordered mineral metabolism in patients with cystic fibrosis. J PEDIATR 1979;94:38-42. 3. Gibbens DT, Gilsanz V, Boechat MI, Dufer D, Carlson ME, Wang C. Osteoporosis in cystic fibrosis. J PEDIATR 1988; 113:295-300. 4. Cobb R J. Outline for the study of scoliosis. Am Acad Orthop Surg 1948;5:261-75.
The Journal of Pediatrics August 1994
5. Propst-Proctor SL, Bleck EE. Radiographic determination of lordosis and kyphosis in normal, and scoliotic children. J Pediatr Orthop 1983;3:344-6. 6. Fon GT, Pitt M J, Thies AC. Thoracic kyphosis: range in normal subjects. Am J Roentgenol 1980;134:979-83. 7. Denton JR, Tietjen R, Gaerlan PF. Thoracic kyphosis in cystic fibrosis, Clin Orthop 1981;155:71-4. 8. Brasfield D, Hicks G, Soong S J, Tiller RE. The chest roentgenogram in cystic fibrosis: a new scoring system. Pediatrics 1979;63:24-9. 9. Grazier KL, Holbrook TL, Kelsey JL, Stauffer RN. The frequency of occurrence, impact, and cost of musculoskeletal conditions in the United States. Chicago: American Academy of Orthopaedic Surgeons, 1984:72-80. 10. Garraway WM, Stauffer RN, Kurland LT, O'Fallon WM. Limb fractures in a defined population. I. Frequency and distribution. Mayo Clin Proc 1979;54:701-7. 11. Erkkila JC, Warwick WJ, Bradford DS. Spine deformities and cystic fibrosis. Clin Orthop 1978;131:146-50. 12. Rose J, Gamble J, Schultz A, Lewiston N. Back pain and spinal deformity in cystic fibrosis. Am J Dis Child 1987; 141:1313-6. 13. Arkin AM, Katz JF. The effects of pressure on epiphyseal growth. J Bone Joint Surg Am 1956;38:1056-66. 14. Stead RJ, Houlder S, Agnew J, et al. Vitamin D and parathyroid hormone and bone mineralization in adults with cystic fibrosis. Thorax 1988;43:190-4. 15. Hanley JG, McKenna MJ, Quigley C, Freaney R, Muldowney FP, Fitzgerald MX. Hypovitaminosis D and response to supplementation in older patients with cystic fibrosis. Q J Med 1985;219:377-85. 16. Wood RE. Prognosis. In: Taussig LM, ed. Cystic fibrosis. New York: Thieme-Stratton, 1984:434-60. 17. Landon C, Rosenfeld RG, Short stature and pubertal delay in male adolescents with cystic fibrosis. Am J Dis Child 1984; 138:388-91. 18. Sproul A, Huang N. Growth patterns in children with cystic fibrosis. J PEDIATR 1964;65:664-76. 19. Reiter EO, Brugman SM, Pike JW, et al. Vitamin D metabolites in adolescents and young adults with cystic fibrosis: effects of sun and season. J PEDIATR 1985;106:21-6. 20. Reiter EO, Stern RC, Root AW. The reproductive endocrine system in cystic fibrosis. Am J Dis Child 1981;135:422-6.