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

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

Scoliosis and pregnancy

RR Betz, WP Bunnell, E Lambrecht-Mulier and GD MacEwen
J Bone Joint Surg Am. 1987;69:90-96.

This information is current as of April 18, 2011

Reprints and Permissions
Publisher Information
Click here to order reprints or request permission to use material from this
article, or locate the article citation on jbjs.org and click on the [Reprints and
Permissions] link.
The Journal of Bone and Joint Surgery
20 Pickering Street, Needham, MA 02492-3157
www.jbjs.org
 Copyright 1987 by The Journal
Scoliosis and Pregnancy*f
BY RANDAL R. BETZ, M.D4, WILLIAM P. BUNNELL,
AND G. DEAN MAcEWEN,
From the Alfred 1. duPont Institute, Wilmington
ABSTRACT: The effects of pregnancy on patients
who have idiopathic scoliosis were investigated in terms
of increased risk of progression of the curve. The charts,
radiographs, and other pertinent data on 355 affected
women who had reached skeletal maturity (Risser Grade
4) before 1975 were reviewed and analyzed. One
hundred and seventy-five patients had had at least one
pregnancy each (Group A) and 180 patients had never
been pregnant (Group B). The groups were comparable
with regard to the treatments that they had received.
After skeletal maturity was reached, the curve pro
gressed more than S degrees in 25 per cent and more
than 10 degrees in 10 per cent of the patients in each
group. The age of the patient at the time of the first
pregnancy did not influence the risk of progression, and
the stability of the curve before pregnancy did not de
crease the risk of its progression during pregnancy. In
patients who had had a spinal fusion, progression in the
unfused portion ofthe spine was negligible in both Group
A and Group B. The presence of a pseudarthrosis did
not result in progression of the curve during pregnancy.
The effects of scoliosis on pregnancy and delivery
were evaluated in the 175 women in Group A. No specific
problems that were directly related to the scoliosis were
noted except for four patients, in whom delivery posed
difficulties. The incidence of cesarean section was one
halfofthe national average, and no sections were directly
related to the mother's scoliosis. The incidence of deliv
ery by cesarean section in patients who had undergone
a posterior spinal fusion was approximately 3 per cent.
Patients who had undergone posterior fusion for idio
pathic scoliosis appeared to be at no increased risk for
pain in the back during pregnancy.
Since the incidence of progressive idiopathic scoliosis
is approximately eight times higher in female patients than patients who have severe kyphoscoliosis. In seven such
in male patients, the possibility that there is an adverse
@ relationship between pregnancy and scoliosis has been complications, an incidence of one in
* No benefits in any form have been received or will be received from labor and an increased frequency of instrumental delivery,
a commercial party related directly or indirectly to the subject ofthis article.
No funds were received in support of this study.
1-ReadattheAnnualMeeting oftheScoliosis Research Society, San
Diego, California, September 19, 1985.
@ Shriners Hospital forCrippled Children. 8400 Roosevelt Boulevard,
Philadelphia, Pennsylvania 19152.
§Alfred I. duPont Institute, P.O. Box 269, Wilmington, Delaware
19899. Please address requests for reprints to Dr. Bunnell.
C Waversbaan 77, 3030 Leuven, Belgium.
90
of Bone and Join: Surgery. irnorporated
M.D.@, ELIZABETH LAMBRECHT-MULIER, M.DAI,
M.D.@, WILMINGTON, DELAWARE
considered27, although it has not been extensively investi
gated. Whether pregnancy increases the risk that a curve
will progress after skeletal maturity is ofcritical importance.
Few reports dealing with this question have appeared in the
literature, and all of them contained small numbers of pa
tients. Keim reported on patients whose curves increased
by 6 to 8 degrees with each pregnancy, the maximum pro
gression being 30 degrees. Berman et al. followed eight
patients through one pregnancy each, and found that only
three had progression of the curve. They concluded that
patients who have a curve of more than 25 degrees have a
greater risk of progression of the curve during pregnancy.
Nachemson et al.23 reviewed the cases of brace-treated pa
tients in whom the curve progressed when they were in their
early twenties, but the curves did not progress after the
patients were twenty-four years old or more. Blount and
Mellencamp reviewed the cases of ten patients who had a
total of nineteen pregnancies. In patients in whom the curve
had been stable before pregnancy no progression was noted;
however, in patients in whom the curve had been progress
ing when they became pregnant, there was some acceler
ation of that progression.
Other authors have found no changes in curvatures after
pregnancy. Carr et al. reported on twenty-four patients who
had been treated with a Milwaukee brace in whom no sig
nificant progression of a curve was noted after pregnancy.
In a study of the natural history of idiopathic scoliosis,
Bunnell reported that there were no detrimental effects of
pregnancy in twenty-seven patients.
Reports of long-term follow-up of patients who under
went spinal fusion for idiopathic scoliosis either did not deal
with the problems that were encountered during pregnancy
or delivery or else mentioned them only briefly81421. In the
obstetrical literature, discussion of pregnancy and scoliosis
usually has been limited to the obstetrical complications of
reports, a total of sixty-one of 250,000 deliveries involved
The problems included one transverse fetal arrest during
perineal tearing, and subsequent genital prolapse.
The primary goal of this study was to determine
whether or not pregnancy increases the risk of progression
of a curve in a scoliotic patient after she has reached skeletal
maturity. In addition, we sought to determine whether preg
nancy has any adverse effects on the fused or unfused por
THE JOURNAL OF BONE AND JOINT SURGERY
 SCOLIOSIS AND PREGNANCY 91
TABLE I
EVALUATIONSLastBaselineSkeletalAdolescentFirstCurrentEvaluationMaturityVisitPregnancyFollow-upGroupA14.516.418.322.126.7(n
AvERAGE AGES OF THE PATIENTS (IN YEARS)AT SELECTED RADIOGRAPHIC

=175)Group
B13.515.717.0—23.8(n
=180)

tion of the spine after a posterior spinal fusion. We also and comparisons were derived from this program, and the
studied the effects of the scoliosis and posterior fusion on data were plotted with a Houston Instrument digital plotter
pregnancy and delivery. (model DMP-7).

Materials and Methods Results

The charts of 1600 patients who had a diagnosis of Of the 355 patients who participated in this study, 175

idiopathic scoliosis and were seen at the Alfred I. duPont (Group A) had been pregnant at least once and 180 (Group
Institute before 1975 were reviewed. A questionnaire deal B) had never been pregnant. Similar regimens of treatment
ing with pregnancy and treatment for scoliosis was sent to prevailed for the groups. Seventy-five Group-A and eighty
the women, and it was returned by 700 of them. Of the three Group-B patients were merely followed, without treat
total of 700, 355 were eligible for inclusion in this study ment. Thirty-seven Group-A and fifty-five Group-B patients
for these reasons: they had had at least one radiograph made had worn an orthosis, and sixty-three Group-A and forty
at a minimum of one year before reaching skeletal maturity, two Group-B patients had undergone posterior spinal fusion.
one radiograph made at or near the time of skeletal maturity The average age at the five points in time (Table I)
(Risser Grade 4), a current radiograph, and a clinical eval that were of importance was comparable in the two groups
uation by one of us that was done specifically for the pur with the exception of the current age, which was 26.7 years
poses of this study. for Group A and 23 .8 years for Group B.
At least four radiographs from each patient's file were The effect of pregnancy on scoliosis (Table II) was
analyzed: the initial radiograph that had been made at the first evaluated in the 250 patients ( 112 in Group A and 138
time of diagnosis, the radiograph that had been made at the in Group B) who had not had a spinal fusion. The average
time of skeletal maturity, the last radiograph that had been progression ofthe curve between the last visit and the current
made during late adolescence or before pregnancy, and the one was 2.2 degrees in both groups (range for Group A,
current radiograph . For many patients , additional radio —¿ 12 to 25 degrees and for Group B, —¿9 to 39 degrees).

graphs that had been made between the time of skeletal There was progression of the curve of more than 5 degrees
maturity and the first pregnancy were available for analysis. in thirty (27 per cent) of the patients in Group A and thirty
All measurements of curves were made using the Cobb six (26 per cent) of those in Group B and of more than 10
method and, in addition, the pattern, direction, Risser sign, degrees in eleven (10 per cent) of the patients in Group A
angle of vertebral rotation, degree of thoracic kyphosis and and ten (7 per cent) of the patients in Group B. Five (4 per
lumbar lordosis, and balance of the trunk as determined by cent) of the patients in Group A and two ( I per cent) of the
the alignment of the torso over the mid-point of the sacrum patients in Group B had progression of more than I 5 de
were recorded. grees . There was an improvement of the curve that exceeded
A detailed history also was obtained, including these 5 degrees in eight (7 per cent) of the patients in Group A
items of specific interest: the number of pregnancies, age and in seven (5 per cent) of those in Group B . Seventy-four
of the patient at the time of each pregnancy, duration of (66 per cent) of the 1 12 patients in Group A and ninety

each pregnancy (in trimesters), problems (including pain in five (68 per cent) of the I 38 patients in Group B had no
the back) during pregnancy, problems during delivery, and change in the curve.
health of the child or children. The 250 patients who did not have a posterior spinal
The severity and frequency of pain in the back were fusion were compared according to whether they had worn
rated by the patient, each on a scale of zero to 3 points, an orthosis. For those who had not, the risk of progression
with zero indicating the absence of pain. Each patient's of the curve of as much as 10 degrees was the same for
symptoms were assigned to one of four categories of pain: Groups A and B. The patients in Group B who had been
none, mild, moderate, or severe. The maximum numerical treated with an orthosis, however, had a statistically sig
total for each patient therefore was 6 points. nificant (p < 0.05) decrease in the risk of progression (Table
Computer analysis was performed with a specially de III), especially when progression of more than 10 degrees

signed interactive data-base-management system on a Dig was calculated (a risk of 2 per cent compared with 11 per
ital Equipment Corporation 1 1/23 computer. Flow sheets, cent). Whether the patterns of the curves in each subgroup
bar graphs, scattergrams, statistical analyses, summaries, were different did not seem to matter (Table V).

VOL. 69-A, NO. I. JANUARY 1987

92 R. R. BETZ, W. P. BUNNELL, ELIZABETH LAMBRECHT-MULIER, AND G. D. MACEWEN

TABLE 11
PROGRESSIONOF THE CURVEINANORTH0sIS*Curve
PATIENTS WHOWERE OBSERVED ORWERE TREATED WITH

ProgressionImprovement of
DegreesGroup Unchanged>5
Degrees> 10 Degrees> 15 >5 DegreesCurve

(66%)Group
A30/1 12 (27%)1 1/112 (l0@)5/1 12 (4%)8/1 12 (7%)74/1 12
B36/138 (2691110/138 (7%)2/138 (1%)7/138 (5%)95/138 (68%)

* The patients whose curves progressed more than 10 or 1 5 degrees are included in the group that had progression of more than 5 degrees, and
those whose progression was more than 10 degrees includes those who had a curve that progressed more than 15 degrees.

However, when we analyzed the data according to the were analyzed, the distribution of findings as to progression
severity of the curve at the last visit during adolescence, or regression of the curve was virtually identical among
there were several notable findings. Each patient was placed Group-A and Group-B patients.
in one of three subcategories, as described by Weinstein One hundred and five patients (sixty-three in Group A
and Ponseti, with regard to the curve: 30 degrees or less, and forty-two in Group B) underwent posterior spinal fu
31 to 49 degrees, and 50 degrees or more. The risk of sion, and forty-nine patients in Group A and thirty-five
progression of the curve was nearly identical for both groups patients in Group B did not have internal fixation. Fourteen
of patients within each category, with no statistically sig patients in Group A and seven in Group B had had prior
nificant differences (Table IV). However, the more severe treatment with a brace.
the curve was, the higher the risk of progression was; this The average ages at the times when radiographs were
was more evident when the increase was more than 5 degrees made (before fusion, after fusion, at the last visit as an
as compared with more than 10 degrees. adolescent, and at the current follow-up examination) were
The 112 patients in Group A who had not had a spinal comparable in both groups, as were the average degrees of
fusion were divided into three subgroups based on age at curvature at those times (Table VI).
the time of the first pregnancy. Twenty-four patients had Four patients in Group A had a documented pseudar
been eighteen years old or less; fifty-seven, nineteen to throsis . Two of them underwent posterior spinal fusion with
twenty-three years old; and thirty-one, twenty-four years internal fixation and repair of the pseudarthrosis before preg
old or older. There was a low risk of progression of the nancy. In the other two patients with a pseudarthrosis, no
curve in all three groups, but it was highest among those progression in the fused segment of the spine was seen
whose first pregnancy occurred when they were twenty-four during pregnancy.
years old or older. In five patients ( 16 per cent) the curve Pseudarthrosis developed in two Group-B patients who
progressed more than 10 degrees, and of all of the patients had been treated with posterior fusion without instrumen
whose curve progressed more than 10 degrees, two-thirds tation. In neither patient was progression of the curve doe
were twenty-four years old or older at the time of the first umented between the time when the last radiographs had
pregnancy. When the first pregnancy occurred at the age of been made and the most recent visit.
eighteen or earlier, there was hardly any risk of progression; Analysis of the alignment of the seventh cervical ver
only one patient (5 per cent) of the twenty-four in this group tebra with the midline of the sacrum revealed no differences
had progression of the curve of more than 10 degrees. The between Group A and Group B . The presence or absence
findings did not change regardless of whether only full-term of abnormal kyphosis or lordosis and the height and weight
pregnancies or all pregnancies were included in the analysis. of the patients were also analyzed for both groups, and no
There was a slight increase in the risk of progression significant differences were found.
of a curve of more than 10 degrees in the patients who had The 175 Group-A patients delivered a total of 238
had three pregnancies compared with those who had had children, and sixty-nine additional pregnancies ended in
only one, but the difference was not statistically significant. either spontaneous or elective abortion. A total of 159 chil
When limited to full-term pregnancies, the same analysis dren were born to the patients who had not had a spinal
revealed the same results. fusion, and there were forty-seven additional pregnancies
When changes in a curve between the time when skel among these patients. None of these patients reported any
etal maturity was reached and the status before pregnancy problems related to scoliosis during pregnancy. Seventy

TABLE III
RESULTS OFTREATMENT ORTH0SISGroup
IN PATIENTS WHO WERE OBSERVED OR WERE TREATEDWITH AN

AGroup BRisk
ofProgressionProgressionProgressionProgressionUnchangedof
ofRisk
ofRisk ofRisk

DegreesObservation50/75 >5 Degreesof > 10 DegreesUnchangedof >5 Degreesof > 10

(11%)Orthosis24/37 (67SF)20/75 (27%)7/75 (9sf)54/83 (61%)28/83 (34%)9/83

(65ch)10/37 (27%)4/37 ( 11%)46/55 (83%)8/55 (14%)1/55 (2%)

THE JOURNAL OF BONE AND JOINT SURGERY

 SCOLIOSIS AND PREGNANCY 93
TABLE IV
SUBGROUPSGroup
SEVERITY OF THE CuRVE COMPARED BY TREATMENT

BRisk AGroup
ofSeverityProgressionProgressionProgressionProgressionof
ofRisk ofRisk ofRisk

DegreesObservation@30
the CurveNo.of >5 Degreesof > 10 DegreesNo.of >5 Degreesof > 10

(6%)31-49
degrees336 (18%)2 (6%)317 (22%)2
(6%)@50
degrees267 (27%)3 (12%)328 (25%)2
(25%)Orthosis@30
degrees167 (44%)2 (12%)2013 (65%)5

(0%)31-49
degrees183 (17%)1 (6%)312 (6%)0
(5%)@50
degrees124 (33%)2 (17%)216 (29%)1
degrees73 (43%)2 (29%)30 (0%)0 (0%)

seven per cent reported having had backache during preg
nancy, although only 12 per cent considered the pain to
have been severe. Two of 159 deliveries were marked by
problems related to the scoliosis, in that spinal anesthesia
could not be successfully administered. Cesarean section
was necessary in twelve (7.4 per cent) of the 159 deliveries;
in none of them was this related to the scoliosis.
Congenital disorders were reported in the offspring of
three patients: one child had congenital dislocation of both
hips, one had an abnormality of the ureterovesical junction,
and one had esotropia. In four children idiopathic scoliosis
was noted to have developed before they were ten years
old.
Of the Group-A patients, two-thirds reported having
had some backache before pregnancy and three-quarters,
during pregnancy and at the time of the current evaluation.
Severe pain in the back was reported by 12 per cent of the
patients during pregnancy and by 9 per cent at the time of
the current evaluation. Halfofthe Group-B patients reported
that they had pain in the back at the time of the current
evaluation, with severe pain being reported by 3 per cent.
Of the 175 Group-A patients, sixty-three underwent
spinal fusion. These sixty-three patients delivered seventy
nine infants, and twenty-two additional pregnancies ended
in either spontaneous or elective abortion. According to the
patients, there were no problems during pregnancy that were
directly due to the scoliosis other than pain in the back,
which was reported by about two-thirds.
Two patients reported having had problems during de
livery that were minor but were directly related to the spinal
fusion. One was a failure of attempted spinal anesthesia,
and the other involved difficulty with proper positioning.
Two (2 .5 per cent) of the seventy-nine infants were delivered
by cesarean section, but in neither instance was this related
to the mother's scoliosis.
Four of the seventy-nine infants that were delivered by
mothers who had had a spinal fusion were reported to have
a congenital anomaly: one each had pectus excavatum, con
genital dislocation of the hip, tracheoesophageal fistula, and
club foot. No children showed signs of scoliosis at the time
of the latest follow-up evaluation.
Forty-one per cent of the patients who had been treated
non-surgically reported mild, moderate, or severe pain in
the back before the first pregnancy; 63 per cent, during
pregnancy; and 70 per cent, at the time ofthe current follow
up evaluation. Current pain in the back appeared to be
present to approximately the same degree in the study group
as it is in the general population (women in the age groups
under consideration). Pain in the back was also analyzed
for severity and frequency, according to each of the follow
ing categories: the patient's occupation, the patient's current
weight, the weight gain during pregnancy, the level of fu
sion, and the presence of internal fixation. No significant
differences were noted between Group-A and Group-B pa
tients.
Discussion
The outstanding finding in this study was that preg
nancy does not have an adverse effect on scoliosis; that is,
it does not increase the risk of progression of the curve. In
many patients in this series the curve did increase; 25 per
cent had progression of the curve of 5 degrees or more

TABLE V
PATTERNSOF FUSIONGroup
THE CURVES IN PATIENTS WHO DIDNOT HAVE A SPINAL

BRisk AGroup
ofProgression
of Risk ofRisk of Risk
Progressionof ProgressionProgression
DegreesThoracic7/31 >5 Degrees of > 10 Degreesof >5 Degrees of > 10

(9%)Thoracolumbar4/16 (22%) 4/31 (13%)17/52 (32%) 5/52

(0%)Lumbar4/14 (25%) 1/16 (6%)5/34 (15%) 0/34
(10%)Double (28%) 0/14 (0%)3/19 (16%) 2/19
major15/51 (29%) 6/51 (12%)9/33 (27%) 2/33 (6%)

VOL. 69-A,NO. 1.JANUARY 1987

 94 R. R. BETZ, W. P. BUNNELL, ELIZABETH LAMBRECHT-MULIER, AND G. D. MACEWEN

TABLE VI
SELECTED RADIOGRAPHIC EVALUATIONS IN PATIENTS WHO HAD HAD A POSTERIOR
SPINAL FUSION IN GROUPSA AND B*

Before Adolescent
Follow-upABABA
FusionAfter FusionLast VisitFirst PregnancyCurrent
BAv. BABA

24.3Av.
age (yrs.)14.515.615.516.619.1 18.722.4—26.8
(degrees)Fused
curve
34Fusedcurve It5556303332 3433
31Unfused
curve 2@5153272829 2930
curve2*3633171918 2119 20

* The primary fused curve was called Fused curve 1 . If the patient had a double major curve or a compensatory curve that required fusion, this
was called Fused curve 2. If it was not fused, it was referred to as Unfused curve 2. Progression of the curve in the fused portion of the spine was
not noted in any Group-A patient. Only one patient in Group B had progression (1 1 degrees) in Fused curve 2. In two Group-A patients, the curve in
the unfused portion of the spine progressed (6 and 8 degrees).
1@Sixty-three patients in Group A and forty-two in Group B.
@ Forty-five patients in Group A and thirty-two in Group B.
§Eighteen patients in Group A and ten in Group B.

during the decade after skeletal maturity was reached. How the stability of the curve before pregnancy was the most
ever, none of the increases of the curve were of major important factor in determining whether or not a curve would
magnitude (more than 20 degrees). progress during or after pregnancy.
We could not substantiate the finding by Cochran and After pregnancy, no patient who had been treated sur
Nachemson7 and by Nachemson22 that women who become gically showed progression in the fused portion ofthe spine,
pregnant at or after the age of twenty-four years have a and only two of the eighteen patients who had undergone
decreased risk for progression of the curve. In fact, two a fusion had significant progression in the unfused portion
thirds of the patients in our series in whom the curve pro of the spine. Similarly, Nachemson et al.23 reported pro
gressed more than 10 degrees were twenty-four years old gression in the unfused portion of the spine in only three
or older at the time of the first pregnancy. It should be noted of the twenty-eight patients whom they studied. Thus, it
that only five (16 per cent) of the thirty-one Group-A patients appears that pregnancy does not adversely affect the unfused
who were twenty-four years old or older showed progression portion of the surgically treated spine in the majority of
of more than 10 degrees, and 78 per cent of the same group patients.
did not have any progression of the curve at all. Of the five Our study confirms the necessity of follow-up after
patients in question, three had had radiographs made within skeletal maturity for patients who have a curve that measures
the two years preceding pregnancy that revealed that the more than 30 degrees. We currently evaluate patients every
progression did not occur during the late teen-age years. two years until they reach the age of twenty-five and every
The severity of the curve within each of the three age-groups five years thereafter. The evaluation should be done soon
(as regarded pregnancy) was analyzed, and no significant after each delivery to minimize the risk to a fetus of exposure
differences in progression of the curve were noted. to x-rays.
Our study substantiates Keim's statement that scoliosis Another goal of this study was to assess whether sco
progresses in adults. We did not find, however, that curves liosis of a mild to moderate degree has an adverse effect
generally progressed 6 to 8 degrees with each pregnancy, on pregnancy and delivery . No problems during pregnancy
as he suggested. There was equal progression of curves in that were specifically related to scoliosis were reported by
both Group-A and Group-B patients. Berman et al. , in a our patients. During delivery, the only problems (two) that
study of eight patients, suggested that patients who have a were encountered were concerned with the administration
curve of more than 25 degrees have an increased risk of of spinal anesthesia. Cesarean section was done in twelve
progression of the curve with pregnancy, but our data did of 159 deliveries, or 7.4 per cent, which is well below the
not confirm this. national average of 16 per cent30. All ofthe cesarean sections
Blount and Mellencamp, in 1980, reported on ten pa were performed because of either cephalopelvic dispropor
tients who had a total of nineteen pregnancies. In the three tion or fetal distress. In no instance was the surgical pro
patients who had progression of the curve before pregnancy, cedure directly related to the mother's scoliosis.
there was progression during the pregnancy also. In our Congenital anomalies were few (three in 159 off
patients, approximately one-third of the unstable curves re spring), and the incidence was well below the national av
mained unstable, while the remainder stabilized. One erage of 7 per cent20. One infant had a club foot and one
quarter of the stable curves progressed more than 5 degrees had congenital dislocation of both hips, and in so small a
in both Group A and Group B . The fact that a curve was group (159 children) these anomalies may be worthy of note.
stable before pregnancy did not mean that it would not However, due to the small number of patients involved in
progress with pregnancy. These findings are not in accord the study, no valid calculations of incidence can be made.
with those of Blount and Meliencamp, who reported that The incidence of spinal problems in the offspring was

THE JOURNAL OF BONE AND JOINT SURGERY

 SCOLIOSIS AND PREGNANCY 95

very low, as only four patients were involved. Since the
children were still very young, the incidence of adolescent
idiopathic scoliosis could not be determined. Currently we
inform patients that the risk of inheritance of scoliosis is
between 15 and 20 per cent'°.
Seventy-seven per cent of the patients who had been
treated non-surgically for scoliosis reported having had pain
in the back during pregnancy. This is more than that reported
by Mantle et al. , who found an incidence of pain in the
back of 48 per cent in a group of 180 women under ob
stetrical care. In our study, 54 per cent of the Group-B
patients had pain in the back at the time of the current
evaluation. Torgerson and Dotter, in 1976, reported an in
cidence of pain in the back in the general population of 60
per cent. The differences in severity of pain in the back and
the varying scoring systems that were used may account for
these different percentages and prohibit accurate comparison
with other studies that have been reported in the literature.
The incidence of pain in the back in scoliotic patients
who were treated non-surgically was reported to be 3 1 per
cent by Collis and Ponseti, 40 per cent by Nachemson et
al.23, @9per cent by Drummond et al. , and 90 per cent by
Nilsonne and Lundgren; however, there were large varia
tions in the ages of the patients among these series. Based
on our data, women whose scoliosis is not treated or is
treated with an orthosis who become pregnant may be at
slightly increased risk for pain in the back, but only a small
percentage have severe pain. With this possible exception,
we think that scoliosis of a mild or moderate degree has no
adverse effect on pregnancy and delivery.
In our series, for patients who underwent posterior
spinal fusion no problems during pregnancy that were see
ondary to scoliosis were reported, and only two problems
arose during delivery. A cesarean section was necessary in
only two of seventy-nine deliveries, or 2.5 per cent. Con
genital problems occurred in four (5 per cent) of seventy
nine infants; this is less than the national average.
Conclusions
Based on our data, we concluded that pregnancy does
not increase the risk of progression of a scoliotic curve in
a patient whose curve is not severe. The risk of progression
of a curve during pregnancy is unaffected by the age of the
patient at the time of the first pregnancy, the number of
pregnancies, or the stability of the curve. Pregnancy has no
significant adverse effect on the spine in a patient who has
had a posterior fusion for idiopathic scoliosis of mild or
moderate degree. Mild to moderate scoliosis does not have
a deleterious effect on pregnancy or delivery. The incidence
of delivery by cesarean section and of health problems in
the children of women who have scoliosis is no greater than
that for the non-scoliotic population. The incidence and
severity of pain in the back during pregnancy appear to be
somewhat higher in scoliotic patients who have not had a
spinal fusion, but those with idiopathic scoliosis who
undergo posterior fusion appear to have no increased risk
of pain in the back or other adverse effects during pregnancy
and delivery.
N0IL: The authors thank Carolyn Hill. Gail Uuss. Susan Keenan. and Diane Ferry for their
assistance in his project.

References
1. BERGE, J. E.: Pregnancy Associated with Severe Kyphoscoliosis of the Thoracic Spine. J. Obstet. and Gynaec. British Commonwealth. 69: 81-
98, 1962.
2. BERMAN, A. T.: COHEN, D. L.; and SCHWENTKER,E. P.: The Effects of Pregnancy on Idiopathic Scoliosis. A Preliminary Report on Eight Cases
and a Review of the Literature. Spine, 7: 76-77. 1982.
3. BLOUNT, W. P.. and MELLENCAMP,D. D.: The Effect of Pregnancy on Idiopathic Scoliosis. J. Bone and Joint Surg., 62-A: 1083-1087, Oct.
1980.
4. BUNNELL, W. P. : A Study of the Natural History of Idiopathic Scoliosis. Read at the Annual Meeting of The American Academy of Orthopaedic
Surgeons, Anaheim, California, March 12, 1983.
5. CARR, W. A.; MOE, J. H.: WINTER, R. B.: and LONSTEIN. J. E.: Treatment of Idiopathic Scoliosis in the Milwaukee Brace. J. Bone and Joint
Surg., 62-A: 599-612, June 1980.
6. COBB. J. R. : Outline for the Study of Scoliosis. In Instructional Course Lectures. The American Academy of Orthopaedic Surgeons. Vol. 5, pp.
261-275. Ann Arbor, J. W. Edwards, 1948.
7. COCHRAN, T. , and NACHEMSON,A.: Long-Term Anatomic and Functional Changes in Patients with Adolescent Idiopathic Scoliosis Treated with
the Milwaukee Brace. Spine. 10: 127-133. 1985.
8. COCHRAN, THOMAS;IRSTAM.LARS; and NACHEMSON,ALF: Long-Term Anatomic and Functional Changes in Patients with Adolescent Idiopathic
Scoliosis Treated by Harrington Rod Fusion. Spine, 8: 576-584, 1983.
9. COLLIS, D. K. , and PONSETI, I. V. : Long-Term Follow-up of Patients with Idiopathic Scoliosis Not Treated Surgically. J. Bone and Joint Surg.,
51-A: 425-445, April 1969.
10. COWELL, H. R.; HALL, J. N.; and MACEWEN, G. D.: Genetic Aspects ofldiopathic Scoliosis. Clin. Orthop., 86: 121-131, 1972.
11. DRUMMOND.D. S. : FOWLES, J. V. : ECUYER, SERGE; ROY, LOUIS: and KERNER, M. : Untreated Scoliosis in the Adult. In Proceedings of the
Scoliosis Research Society. J. Bone and Joint Surg. , 58-A: 156, Jan. 1976.
12. DUGAN, R. J. , and BLACK, M. E.: Kyphoscoliosis and Pregnancy. Am. J. Obstet. and Gynec. , 73: 89-93, 1957.
13. FEARL, C. S.: Kyphoscoliosis and Pregnancy. Its Cardiorespiratory Implications. Read at the Annual Meeting of the Pacific Coast Obstetrical and
Gynecological Society. Timberline, Oregon, Nov. 1950.
14. HARRINGTON,P. R. , and DICKSON, J. H.: An Eleven-Year Clinical Investigation of Harrington Instrumentation. A Preliminary Report on 578
Cases. Clin. Orthop.. 93: 113-130, 1973.
15. JONES. D. H.: Kyphoscoliosis Complicating Pregnancy. Lancet, 1: 517-519, 1964.
16. KEIM, H. A.: The Adolescent Spine, pp. 104-109. New York, Springer, 1982.
17. KOPENHAGER,T.: A Review of 50 Pregnant Patients with Kyphoscoliosis. British J. Obstet. and Gynec., 84: 585-587, 1977.
18. MANNING, C. W.; PRIME, F. J.; and ZORAB. P. A.: Pregnancy and Scoliosis. Lancet, 2: 792-795, 1967.
19. MANTLE, M. J.: GREENWOOD.R. M.: and CURREY. H. L. F.: Backache in Pregnancy. Rheumat. and Rehab., 16: 95-101, 1977.
20. MARCH OF DIMES: Facts 1979, p. V-I . National Foundation, March of Dimes, White Plains, New York, 1978.
21 . M0SK0wITz. A.; MOE. J. H.; WINTER, R. B.: and BINNER, H.: Long-Term Follow-up of Scoliosis Fusion. J. Bone and Joint Surg., 62-A: 364-
376,April1980.
22. NACHEMSON,ALF: A Long Term Follow-up Study of Non-Treated Scoliosis. Acta Orthop. Scandinavica, 39: 466-476, 1968.
23. NACHEMSON,ALF; COCHRAN, T. P. : IRSTAM,L. : and FALLSTROM,K. : Pregnancy after Scoliosis Treatment. Read at the Annual Meeting of the
Scoliosis Research Society. Montreal, Quebec, Canada, Sept. 16, 1981.
24. NILSONNE. ULF. and LUNDGREN. K.-D.: Long-Term Prognosis in Idiopathic Scoliosis. Acta Orthop. Scandinavica, 39: 456-465, 1968.
25. PHELAN, J. P.: DAINER. M. J.: and COWHERD. D. W.: Pregnancy Complicated by Thoracolumbar Scoliosis. Southern Med. J.. 71: 76-78, 1978.
26. RISSER, J. C. : Important Practical Facts in the Treatment of Scoliosis. In Instructional Course Lectures. The American Academy of Orthopaedic
96 R. R. BETZ, W. P. BUNNELL, ELIZABETH LAMBRECHT-MULIER, AND G. D. MACEWEN

Surgeons. Vol. 5, pp. 248-260. Ann Arbor, J. W. Edwards, 1948.

27. ROGALA. E. J. : DRUMMOND.D. S.; and GURR, JEAN: Scoliosis: Incidence and Natural History. A Prospective Epidemiological Study. J. Bone
and Joint Surg., 60-A: 173-176, March 1978.
28. TORGERSON,W. R. , and DOTTER. W. E.: Comparative Roentgenographic Study of the Asymptomatic and Symptomatic Lumbar Spine. J. Bone
and Joint Surg. , 58-A: 850-853, Sept. 1976.
29. WEINSTEIN. S. L. , and PONSETI. I. V.: Curve Progression in Idiopathic Scoliosis. J. Bone and Joint Surg. . 65-A: 447-455, April 1983.
30. WII.I.IAMs OBSTETRICS:Ed. 16, p. 1082. Edited by J. A. Pritchard and P. C. MacDonald. New York, Appleton-Century-Crofts, 1980.

Copyrtght I957 by The Journal of Born and Joust Surgers. Ineorporwed

Lyme Arthritis in Children

AN ORTHOPAEDIC PERSPECTIVE*

BY RANDALL W. CULP, M.D.t, ANDREW H. EICHENFIELD, M.D.t, RICHARD S. DAVIDSON, M.D.t,

DENIS S. DRUMMOND, M.D.t, MARK R. CHRISTOFERSEN, M.D.t, AND DONALD P. GOLDSMITH, M.D.t,
PHILADELPHIA, PENNSYLVANIA

l@rom the Departments of Orthopedic Surgery aiid Rheumatology. Children's Hospital of Philadelphia, Philadelphia

ABSTRACT: The cases of forty-three
clinical and serological evidence of Lyme arthritis that
was diagnosed between August 1983 and July 1985 were
evaluated. The mean length of follow-up was twenty
months, with a range of five to thirty months. All of the
children lived in or had visited an area where the disease
was known to be endemic. Arthritis was the presenting
feature in more than half of the children, and half of
the children had initially consulted an orthopaedic sur
geon, none of whom made the correct diagnosis. Only
twenty patients had a history of erythema chronicum
migrans, the characteristic rash that precedes the ar
thritis, and for only nineteen children was there any
recollection of having been bitten by a tick. Three pa
tients had Bell palsy and one had a popliteal cyst in
conjunction with the arthritis. All of the patients had
oligoarticular involvement. The knee was involved in all
but two patients. Recurrent attacks of synovitis were
common. Effusion was the only radiographic abnor
mality that was observed, and it was found in thirty-two
patients. The sedimentation rate was elevated in thirty
of thirty-six patients. Immunofluorescent serology for
Lyme disease, which is sensitive and specific, was uni
formly positive. Of thirty-three patients who were
treated with oral administration of penicillin or tetra
cycline alone, thirty-one responded, while two patients
who had recurrent attacks of the disease responded to
parenteral administration of antibiotics. The remaining
ten patients responded to combinations of orally and
parenterally administered antibiotics. Longer follow-up
is needed to further document the apparently low rate
* No benefits in any form have been received or
a commercial party related directly or indirectly to the subject ofthis article.
No funds were received in support of this study'.
t Children's Hospital of Philadelphia. 34th and
levard. Philadelphia. Pennsylvania 19104. Please address requests for re
prints to Dr. Davidson. Departnient of Orthopedic Surgery.
children with of relapse after antibiotic therapy in this young popu
lation.
Lyme disease was first recognized in 197522. It has
been established that the spirochete Borrelia burgdorferi'4
is the etiological agent that is transmitted by the tick Ixodes
dammini (deer tick) or a related ixodid tick'3'4. The disease
classically appears first in the summer, and there is a char
acteristic lesion of the skin, erythema chronicum mi
grans'523. Neurological59, cardiac'9, or articular manifes
tations7'6'5 may supervene later. The diagnosis of Lyme
disease is often made on clinical grounds and should be
confirmed by serological tests for antispirochetal anti
bodies@' l.l2@
The infection has been described in twenty-four of the
United States and in nineteen countries. Endemic foci have
been identified in southern New England, the middle-At
lantic states, Wisconsin, Minnesota, and California'22°.
Usually the articular involvement develops within a few
weeks after the onset of the systemic symptoms, but the
interval may be as long as two years. The typical pattern is
one of brief, intermittent attacks of swelling and pain in one
or more joints, primarily the large ones7'@'t2°.The involve
ment of large joints may become chronic in about 10 per
cent of untreated patients7'6'7. Recently, Lyme disease has
been diagnosed in patients who had oligoarticular arthritis,
but no history of a rash or a tick-bite, who had positive
serology2 ‘¿.
Lyme arthritis has been described so recently that most
orthopaedic surgeons may not be familiar with it, and there
have been few reports of it in the orthopaedic literature.
The proved cases of Lyme arthritis that we are reporting in
patients who were treated at two children's hospitals in
will he received from
Philadelphia will serve to alert readers to the disease.
Civic Center Bou Materials and Methods
The forty-three children were examined and treated at

THE JOURNAL OF BONE AND JOINT SURGERY