Anda di halaman 1dari 6

This is an enhanced PDF from The Journal of Bone and Joint Surgery

The PDF of the article you requested follows this cover page.

Fractures in patients who have myelomeningocele

TR Lock and DD Aronson
J Bone Joint Surg Am. 1989;71:1153-1157.

This information is current as of May 27, 2010

Reprints and Permissions Click here to order reprints or request permission to use material from this
article, or locate the article citation on and click on the [Reprints and
Permissions] link.
Publisher Information The Journal of Bone and Joint Surgery
20 Pickering Street, Needham, MA 02492-3157
Copyright 1989 by The Journal ofBone and Joint Surgery, Incorporated

Fractures in Patients Who Have Myelomeningocele*


From the Department of Orthopaedic Surgery. Children’s Hospital of Michigan,

Wayne State University School of Medicine, Detroit

ABSTRACT: Thirty-seven (20 per cent) of 186 chil- which may create a problem in differentiating between Os-
dren who had myelomeningocele whose records were teomyelitis, septic arthritis, cellulitis, neoplasm, and frac-
reviewed had sustained a total of seventy-six fractures. ture”3’8”2’13”5’18#{176}. The radiographs are the key to the correct
The frequency with which the fractures occurred was diagnosis, as the fracture patterns are unique and are dif-
related directly to the level of neurological involvement. ferent from those seen in children who are neurologically
Thirteen (41 per cent) of the patients who had involve- intact. Diaphyseal or metaphyseal fractures are easily rec-
ment at the thoracic level, fifteen (36 per cent) who had ognized. Epiphyseal separations may be confusing, how-
involvement at the upper lumbar level, eight (10 per ever, as they may show only a widened physis, subperiosteal
cent) who had involvement at the lower lumbar level, new-bone formation, or a widened, irregular metaphy-
and one (3 per cent) who had involvement at the sacral sis8”8’20. Awareness of this appearance helps to avoid un-
level sustained fractures. necessary biopsy or antibiotic therapy8’9”820.
Sixty-five (86 per cent) of the fractures occurred The treatment of fractures in patients who have mye-
before the child was nine years old, fifty-eight (76 per lomeningocele is complicated by the occurrence of patho-
cent) were judged to be secondary to the limb being in logical fractures at adjacent sites. Previous reports in the
a cast, and seventy-four (97 per cent) involved the lower literature have identified some of these problems, yet there
extremity. Eleven patients, all of whom had thoracic or have been few recommendations for treatment4’6’7”3”4’20.
upper involvement,
lumbar sustained fractures of mul- The purpose of this study was to review our experience with
tiple extremities. All fractures of the lower extremity fractures in children who had myelomeningocele, for the
were distal to the level of neurological involvement; they purpose of formulating recommendations for treatment and
occurred predominantly in the femur in patients who prevention of the fractures.
had thoracic involvement and in the tibia in patients who
had lumbar involvement. Materials and Methods

All of the metaphyseal and diaphyseal fractures We reviewed the records of 186 patients who, at the
healed satisfactorily, whether they were treated by im- time of writing, were being followed in the Myelomenin-
mobilization in a plaster cast or in a bulky Webril dress- gocele Care Center at Children’s Hospital of Michigan. The
ing, although there were fewer complications in the latter ages of the patients ranged from five months to twenty-three
group. The seven fractures that involved the physeal years and nine months, the average age being eight years
plate were a major problem, as three (43 per cent) had and six months. Eighty-three patients were male and 103
delayed union and two (29 per cent) developed prema- were female.
ture growth arrest. The patients were divided into four groups, according
to the most caudad functioning motor level: thirty-two pa-
Fractures complicate the management of children who tients had involvement ofthoracic levels; forty-two, of upper
have myelomeningocel&’3’46’10”215’20. A fracture may occur lumbar levels; seventy-eight, of lower lumbar levels; and
at night, while the child is not wearing an orthosis, when thirty-four, of sacral levels. The medical records and ra-
the child simply turns in .
&‘ It usually results in painless diographs were analyzed to determine the incidence, etiol-
swelling, warmth, and hyperemia of the extremity4”2”820. ogy, and pattern of the fractures.
It may also cause a low-grade fever, an elevated leukocyte
count, and an elevated erythrocyte-sedimentation rate, Incidence

There were seventy-six fractures in thirty-seven (20

* No benefits in any form have been received or will be received from
a commercial party related directly or indirectly to the subject ofthis article.
per cent) of the 186 patients. There was a direct relationship
No funds were received in support of this study. between the frequency with which the fractures occurred
t Department ofOrthopaedic Surgery, Wayne State University School
and the level of neurological involvement. The incidence
of Medicine, Hutzel Hospital, 4707 St. Antoine, Detroit, Michigan 48201.
Please address requests for reprints to Dr. Aronson. of fracture was 41 and 36 per cent in patients who had

VOL. 71-A, NO. 8, SEPTEMBER 1989 1153



Patients Who
Had Fractures No. of
Motor No. of
Level Patients No. Per Cent Fractures

Thoracic 32 13 41 28
Upper lumbar 42 15 36 37
Lower lumbar 78 8 10 10
Sacral 34 1 3 1
Total 186 37 20 76

involvement of thoracic and upper lumbar levels, respec-

tively, while in patients who had lower lumbar and sacral
involvement, it was 10 and 3 per cent (Table I).
There was a strong correlation between the patient’s
age and the incidence of fracture. Sixty-five fractures (86
per cent) occurred in the 100 patients who were less than
nine years old. Only one fracture occurred after puberty
(Table il).


Fifty-eight (76 per cent) of the fractures occurred while

the patient was in a postoperative or corrective plaster cast.
A fracture was considered to be related to the effects on the
limb of being in the cast if it occurred while the extremity
was in the cast or within six weeks after removal of the
cast. Fifteen (26 per cent) of the cast-related fractures oc-
curred while the patient was in a hip-spica cast and thirty
(52 per cent), while the patient was in a club-foot cast (an
above-the-knee cast in approximately 80 to 90 per cent of
patients). In eleven patients in whom the initial fracture was
treated by immobilization in a plaster cast, multiple fractures
subsequently developed. We have used the term clusters to
describe these secondary fractures, which were either in the
ipsilateral extremity, or, if a hip-spica cast had been used,
often in both extremities. These clusters of fractures oc-
FIG. 1-A
curred exclusively in patients who had thoracic or upper Figs. 1-A, 1-B, and 1-C: This four-year-old boy, who had upper lumbar
lumbar involvement. For only seven (9 per cent) of the myelomeningocele, was seen because of a three-month history of swelling
in the proximal part of the left thigh. He had had two previous fractures
seventy-six fractures was there a history of known trauma. of the left femoral shaft, which had healed without difficulty.
Fig. 1-A: The differential diagnosis from the radiograph included in-
Location fection, neoplasm, and a Salter-Harris Type-I fracture with abundant callus.

Seventy-four of the seventy-six fractures involved the

lower extremity and only two were in the upper extremity. one, in the patella; and two, in the proximal part of the
Forty fractures were in the femur; thirty-three, in the tibia; humerus. The location of the fracture correlated closely with

TABLE II the level of motor involvement. Fractures occurred predom-

inantly in the femur in patients who had involvement of
ACCORDING TO AGE OF THE PATIENTS thoracic levels and in the tibia in patients who had involve-
ment of lumbar levels (Table III).
Motor 0-9 Yrs. 9-14 Yrs. > 14 Yrs.
Forty-two fractures were in the metaphysis, twenty-
Level (100 Patients) (45 Patients) (41 Patients)
seven were in the diaphysis, and seven involved the physis.
Thoracic 21 7 0
The location of the fracture within the bone, however, did
Upper lumbar 36 1 0
not correlate with the level of neurological involvement.
Lower lumbar 7 2 1
Sacral 1 0 0 Treatment
Total 65 (86%) 10 (13%) 1 (1%)
All of the fractures were treated non-operatively, ac-



1. :1

FIG. 1-B
An open biopsy of the left femoral neck revealed fracture callus with areas of avascular necrosis, and cultures were negative. After two months of
treatment in a hip-spica cast, there is early healing of the fracture.

FIG. 1-C
After removal of the cast, there were multiple fractures in clusters. After three weeks of treatment in bilateral long bulky Webril dressings, there is
early healing of all fractures.

VOL. 71-A, NO. 8, SEFFEMBER 1989


cording to the preference of the treating physician. Fifty- in whom there was a problem with the differential diagnosis;
three (70 per cent) of the fractures were immobilized in a the child had delayed union and sustained five subsequent
plaster cast for periods ranging from three to twelve weeks. fractures after removal of a hip-spica cast (Figs. 1-A, 1-B,
The plaster cast was worn until union of the fracture was and 1-C).
visible on the radiographs. Twenty-three fractures (30 per
cent) were managed with a bulky Webril dressing (Kendall,
Boston, Massachusetts) wrapped to a thickness of 1 .5 cen- The incidence of fracture in patients who have mye-
timeters, which in turn was wrapped with one or more Ace lomeningocele has been reported to range from 1 1 .5 to 30
bandages (Becton, Dickinson, Rutherford, New Jersey). per cent4’10”2’14’15’20. In our 186 active patients, both the
incidence and the location of the fractures varied directly
with the level of neurological involvement. We found a 20
per cent over-all incidence of fracture, but sixty-five (86 per
cent) of the seventy-six fractures were in patients who had
involvement of the thoracic or upper lumbar level. Fractures
Location involving the lower extremity were distal to the level of
Level Femur Tibia Other neurological involvement; they occurred predominantly in
Thoracic 20 (71%) 7 (25%) 1 (4%) the femur in patients who had thoracic involvement and in
Upper lumbar 16 (43%) 20 (54%) 1 (3%) the tibia in patients who had lumbar involvement. This
Lower lumbar 3 (30%) 6 (60%) 1 (10%) distinction has not received appropriate emphasis in the
Sacral 1 (100%) 0 (0%) 0 (0%) literature. As the level of neurological involvement de-
Total 40 (53%) 33 (43%) 3 (4%) scends, the ability to walk is improved, weight-bearing is
increased, and protective sensation is present. The increased
level of activity strengthens the bone, according to Wolff’s
The duration of treatment with the soft cast ranged law, and decreases the incidence of pathological fracture.
from one to three weeks. The Webril dressing was worn Drennan and Freehafer reported fifty-eight fractures in
only until callus was visible on the radiographs. After the twenty-five patients who had myelomeningocele and found
plaster cast or Webril dressing was removed, the patient that the upper age limit at the time of the initial fracture
was advised to resume pre-fracture activities gradually, de- was six years. Most (86 per cent) of the fractures in our
pending on the amount of callus that was visible on the series occurred before the patient was nine years old. These
radiographs. The average length of follow-up from the time younger patients need a number of surgical procedures to
of fracture to that of review was three years and eleven attempt to improve the ability to walk, but the immobility
months (range, four months to fifteen years). that is caused by the casts that are applied postoperatively
results in disuse osteoporosis and leads to clusters of path-
Results oiogical fractures. As the patients become older, however,
All of the fractures of the diaphysis and metaphysis they need fewer operative procedures, and the level of ac-
healed rapidly, often forming abundant callus. There was tivity decreases, leading to fewer fractures.
no major difference in fracture-healing between the patients Seventy-six per cent of the fractures in our series were
who were treated with a plaster cast and those who were secondary to the use of plaster casts. Other authors have
treated with a Webril dressing, except that the latter method also reported a high incidence of pathological fractures after
resulted in fewer pressure sores and fewer fractures after the use of plaster casts in children who have a paralytic
treatment. None of the patients needed any change in the disorder4’102”46. Club-foot casts (above-the-knee casts, in
orthosis or had any change in level of activity after treatment approximately 80 to 90 per cent of patients) and hip-spica
as a result of the fracture. casts account for most such fractures and are particularly
Seven fractures involved the epiphyseal plate, includ- dangerous in patients who have thoracic or upper lumbar
ing four Salter-Harris Type-I and three Salter-Harris Type- involvement. Furthermore, stiffness of the joints after im-
II fractures18. Three of these seven fractures had delayed mobilization in a cast increases the risk of fracture by caus-
union. Two of these patients, who had a fracture involving ing concentration of forces on the osteoporotic bone adjacent
the distal tibial physis, had a three-month history of swelling to the joints. Anschuetz et al. reported on three patients
of the ankle, which posed a problem with the differential who had myelomeningoceie in whom life-threatening sys-
diagnosis. Radiographs revealed widening ofthe distal tibial temic hypovolemia developed secondary to bilateral fracture
physis consistent with osteomyelitis. One patient had neg- of the femur after prolonged immobilization. We concluded
ative cultures of material that was aspirated from the tibia. that, if at all possible, plaster casts should be avoided in
Both fractures healed after treatment in a non-weight-bear- children who have myelomeningocele. If some degree of
ing cast, but with a premature growth arrest of the distal immobilization is needed, a bulky Webril dressing provides
tibial physis. The third fracture involving the physis that less rigid immobilization than a plaster cast. This still allows
had delayed union was a Salter-Harris Type-I fracture of satisfactory fracture-healing, but perhaps with less disuse
the proximal part of the femur. This fracture was in a child osteoporosis.



Metaphyseal and diaphyseal fractures in these patients treated by leaving the fractured extremity in the patient’s
result from minimum trauma, are often incomplete or im- long brace. We are concerned that soft-tissue swelling sec-
pacted, and have intact periosteum. These fractures heal ondary to the fracture might lead to pressure sores in the
rapidly, regardless of the method of treatment, and often brace, but we have no experience with this method of treat-
form abundant callus. The callus may be so abundant that ment.
it may be confused with infection or osteogenic sarcoma, Fractures that involve the epiphyseal plate are a dif-
and this confusion has led to biopsy in some patients. Many ferent entity and carry a poorer prognosis13. These fractures
of the children in this series had unexplained swelling, and are less common (9 per cent, in our series) and are char-
only subsequent radiographs confirmed the presence of frac- acterized by painless swelling without systemic signs. The
ture .
The actual date when the fracture had occurred was radiographs may reveal widening of the physis, and the
often difficult to determine, as was the precise date of frac- appearance may be confused with that of rickets or osteo-
ture-healing. The patients who wore a Webril dressing were myelitis. When subperiosteal new bone is present, the ra-
treated for a shorter duration than those who wore a plaster diographic appearance may be confused with that of
cast, and the radiographs were made at different intervals osteogenic sarcoma. These fractures are secondary to re-
of time. It was our impression, however, that there was no petitive trauma to an insensate growth plate5’9’2#{176}
and may
difference between the two methods in the length of time lead to delayed union or growth arrest if there is a delay in
that it took for the fracture to heal. We did not make scan- diagnosis and treatment.
ograms for these patients; therefore, while it is possible that As in any retrospective study, there is a possibility that
some fractures healed with undetected mild shortening, no some fractures may have gone unreported. Why fractures
patient, whether treated with a Webril dressing or a plaster never developed in some patients while others sustained
cast, had clinically detectable shortening or malunion at multiple fractures is unknown. We believe that the findings
follow-up. We recommend a Webril dressing as the treat- in this study are important for the physician who cares for
ment of choice for fractures in patients who have myelo- children who have rnyelomeningocele.

Bleck and Kleinman
in patients who have
that fractures
of the
and Carol
authors thank Stephen W.
for her help in preparing
Burke, M.D.
the manuscript.
for his assistance in organizing the data

1 . ANSCHUETZ, R. H. ; FREEHAFER, A. A.; SHAFFER, J. W. ; and DIXON, M. S. , JR. : Severe Fracture Complications in Myelodysplasia. J. Pediat.
Orthop. , 4: 22-24, 1984.
2. BLECK, E. E. , and KLEINMAN, R. G.: Special Injuries ofthe Musculoskeletal System. in Fractures in Children, p. 21 1 . Edited by C. A. Rockwood,
Jr. , K. E. Wilkins, and
R. E. King. Philadelphia, J. B. Lippincott, 1984.
3. BURNEY, D. W. , JR., and HAMSA, W. R.: Spina Bifida with Myelomeningocele. Clin. Orthop., 30: 167-174, 1963.
4. DRENNAN, J. C. , and FREEHAFER, A. A.: Fractures of the Lower Extremities in Paraplegic Children. Clin. Orthop., 77: 211-217, 1971.
5. EDVARDSEN, PER: Physeo-Epiphyseal Injuries of Lower Extremities in Myelomeningocele. Acta Orthop. Scandinavica, 43: 550-557, 1972.
6. EICHENHOLTZ, S. N.: Management of Long-Bones Fractures in Paraplegic Patients. J. Bone and Joint Surg. , 45-A: 299-310, March 1963.
7. FREEHAFER, A. A. , and MAST, W. A.: Lower Extremity Fractures in Patients with Spinal-Cord Injury. J. Bone and Joint Surg. , 47-A: 683-694,
June 1965.
8. GOLDING, CAMPBELL: Museum Pages. III. Spina Bifida and Epiphysial Displacement. J. Bone and Joint Surg., 42-B(2): 387-389, 1960.
9. GYEPES, M. T.; NEWBERN, D. H.; and NEUHAUSER, E. B. D.: Metaphyseal and Physeal Injuries in Children with Spina Bifida and Meningo-
myeloceles. Am. J. Roentgenol. , 95: 168-177, 1965.
10. JAMES, C. C. M.: Fractures of the Lower Limbs in Spina Bifida Cystica: A Survey of 44 Fractures in 122 Children. in Studies in Hydrocephalus
and Spina Bifida. Devel. Med. and Child Neurol. , Supplementum 22: 88-93, 1970.
11 . KATZ, J. F.: Spontaneous Fractures in Paraplegic Children. J. Bone and Joint Surg. , 35-A: 220-226, Jan. 1953.
12. KORHONEN, B. J.: Fractures in Myelodysplasia. Clin. Orthop. , 79: 145-155, 1971.
13. KUMAR, S. J. ; COWELL, H. R.; and TOWNSEND, PETER: Physeal, Metaphyseal, and Diaphyseal Injuries of the Lower Extremities in Children with
Myelomeningocele. J. Pediat. Orthop. , 4: 25-27, 1984.
14. MENELAUS, M. B.: The Orthopaedic Management of Spina Bifida Cystica. Ed. 2, pp. 61-66. New York, Churchill Livingstone, 1980.
15. QUILlS, A. N. : Fractures in Children with Myelomeningocele. Report of 15 Cases and a Review of the Literature. Acta Orthop. Scandinavica,
45: 883-897, 1974.
16. ROBIN, G. C.: Fractures in Poliomyelitis in Children. J. Bone and Joint Surg., 48-A: 1048-1054, Sept. 1966.
17. SALTER, R. B. , and HARRIS, W. R.: Injuries Involving the Epiphyseal Plate. J. Bone and Joint Surg. , 45-A: 587-622, April 1963.
18. SOUTTER, F. E.: Spina Bifida and Epiphysial Displacement. Report of Two Cases. J. Bone and Joint Surg. , 44-B(1): 106-109, 1962.
19. TOWNSEND, P. F. ; COWELL, H. R. ; and STEG, N. L. : Lower Extremity Fractures Simulating Infection in Myelomeningocele. Clin. Orthop. , 144:

255-259, 1979.
20. WENGER, D. R. ; JEFFCOAT, B. T. ; and HERRING, J. A. : The Guarded Prognosis of Physeal Injury in Paraplegic Children. J. Bone and Joint Surg.,
62-A: 241-246, March 1980.

VOL. 71-A, NO. 8, SEPTEMBER 1989