Extract
Idiopathic scoliosis is a structural, lateral curvature of the spine of
unknown etiology that can occur in otherwise normal children and adolescents.
It is a condition that is affected by growth during the first two decades of
life. Harrenstein, in
19361, coined the
term infantile idiopathic scoliosis. He related it to rickets, stating that
the curve responded well to bracing. In 1951,
James2, in his
preliminary report on infantile idiopathic scoliosis, described the cases of
thirty-three infants who were three years of age or less and had a structural
left thoracic curve with no apparent etiology. Most of the infants were boys.
The scoliosis in these infants occasionally resolved, but when it did progress
it tended to progress to a very serious deformity. Later, in 1954, he
classified idiopathic scoliosis into three groups on the basis of age of
onset: infantile scoliosis with onset up to three years of age, juvenile
scoliosis with onset from four to eight years of age, and adolescent scoliosis
from ten years of age until the end of growth, with no reference to the ages
between eight and ten
years3. This
classification was eventually related to prognosis and outcome, and was soon
integrated with the previously described classification of Ponseti and
Friedman, defining different curve patterns according to the location of the
apical vertebra4.
According to James, the earlier the onset, the worse would be the final
curvature and, hence, the
prognosis3. The
accuracy of his observation is well recognized in clinical practice and in the
literature, as infantile and juvenile progressive curves are among the most
challenging problems in spinal deformity care and have possible implications
for the development of pulmonary parenchyma and, secondarily, a decreased life
span5.
Idiopathic scoliosis is a structural, lateral curvature of the spine of
unknown etiology that can occur in otherwise normal children and adolescents.
It is a condition that is affected by growth during the first two decades of
life. Harrenstein, in
19361, coined the
term infantile idiopathic scoliosis. He related it to rickets, stating that
the curve responded well to bracing. In 1951,
James2, in his
preliminary report on infantile idiopathic scoliosis, described the cases of
thirty-three infants who were three years of age or less and had a structural
left thoracic curve with no apparent etiology. Most of the infants were boys.
The scoliosis in these infants occasionally resolved, but when it did progress
it tended to progress to a very serious deformity. Later, in 1954, he
classified idiopathic scoliosis into three groups on the basis of age of
onset: infantile scoliosis with onset up to three years of age, juvenile
scoliosis with onset from four to eight years of age, and adolescent scoliosis
from ten years of age until the end of growth, with no reference to the ages
between eight and ten
years3. This
classification was eventually related to prognosis and outcome, and was soon
integrated with the previously described classification of Ponseti and
Friedman, defining different curve patterns according to the location of the
apical vertebra4.
According to James, the earlier the onset, the worse would be the final
curvature and, hence, the
prognosis3. The
accuracy of his observation is well recognized in clinical practice and in the
literature, as infantile and juvenile progressive curves are among the most
challenging problems in spinal deformity care and have possible implications
for the development of pulmonary parenchyma and, secondarily, a decreased life
span5.
When infantile idiopathic scoliosis was first described, for a child to
qualify as having the disease, the onset of disease had to occur by the time
the child reached three years of age. This age limit was based on the fact
that different curve patterns occurred after this age. In the initial study by
James, the diagnosis of idiopathic scoliosis was made by the age of ten months
for almost all of the thirty-three children, after ten months and up to two
years of age for a few, and after two years of age for only two
patients2. In 1954,
he added nineteen additional patients to his previous series; in most of those
children, the curve developed by eighteen months of age. The justification to
add these additional patients to the initial series was that they shared
similar characteristics with respect to pattern of deformity (left thoracic
curves) and gender
(male)3.
Making use of the term infantile scoliosis, Scott and Morgan, in
19556, suggested
decreasing the upper limit of onset to two years of age. They were trying to
minimize the accidental inclusion of acquired conditions of known etiology,
such as paralytic scoliosis, neurofibromatosis, or other neurological
conditions, as the curves associated with those conditions tend to occur later
in life. In this series of thirty-five patients, they found that left-sided
curves and thoracic curves (characteristics that James had described) were
much less frequent when the onset was at more than two years of age. They
described two different groups: a progressive infantile idiopathic scoliosis
group (twenty-eight patients), and a resolving infantile idiopathic scoliosis
group (seven patients). While the deformity in the children in their resolving
group tended to decrease with age, the true prevalence of this type in the
general population was unknown, as many cases probably were unreported.
Although Scott and Morgan were reporting on a small group of only seven
patients, they observed that some curves had gradual resolution. They
classified these as type I and the others, which had a stationary period
before proceeding to resolution, as type II. In contrast to the type-II
resolving curves,
Mau7 described a
benign type of progressive curve as "an intermediate group" for
which treatment could modify the course. Taking into consideration the
multiple clinical courses observed, and accepting the same etiology, we may
have a spectrum of curve severity from mild or benign, to moderate, and,
finally, to severe or malignant.
It is difficult to ascertain the true natural history of infantile
idiopathic scoliosis as there are few mature patients who remain untreated. In
some series, the natural course was considered to consist of the time until
the patients received treatment, which is quite often before ten years of
age3,6,8.
In the series by James, most of the patients were treated with exercises,
plaster-of-Paris beds, and orthoses. Curves in forty-three patients progressed
to at least 70° by the time the patient was ten years of age, with several
curves progressing to >100°. Curves of <20° in nine patients
resolved "spontaneously" before two years of
age3.
James et al.8, in
1959, reported on 111 patients who had progressive thoracic infantile
idiopathic scoliosis, probably including the cohort previously described by
those authors. Thirty patients were treated with a Milwaukee brace, and
thirty-one patients required su rgery. Of the forty-seven patients who were
between zero and five years of age at the time of follow-up, the curve in
twenty-three was already =70° and the curve in four was =100°.
Of the thirty-seven patients who were between six and ten years of age at the
time of follow-up, the curve in twenty-seven was =70° and the curve in
fourteen was =100°. Of the twenty-three children who were eleven years
of age and older, few patients had remained untreated and the curve in the
majority (twelve patients) was =100°. Two patients had reached
maturity, and the curve in each of those patients measured >150°.
McMaster and
Macnicol9 reported
on twenty-seven patients who underwent a long-term treatment protocol of
bracing and surgery. For twenty-two patients with single thoracic curves, the
mean age at bracing was four years, three months (range, one year, nine months
to eight years, six months) and the mean initial curve was 63° (range,
29° to 105°). In five patients with double major curves, the brace was
applied at a mean age of seven years, nine months (range, four to thirteen
years) for a mean initial thoracic curve of 45° and a mean lumbar curve of
48° (range, 34° to 55° for thoracic curves and 41° to 60°
for lumbar curves). All patients who had surgery were operated on by ten years
of age. Bracing was more effective for single thoracic curves than for double
curves, with the result related to the severity of the curve. In the patients
with single thoracic curves, the mean best correction was 38% in the first
year and 43% after two years of bracing, with a slow deterioration of 1°
per year after that. After five years of bracing, a more rapid deterioration
occurred, and the final mean correction was 11%. Before treatment, these same
patients had a mean progression of 12° annually (range, 6° to
20°).
Scott and Morgan6
reported on twenty-eight patients with infantile idiopathic scoliosis.
Fourteen patients were followed to maturity, and all had severe curves (mean
curve, 120°). The smallest curve was 84° at eighteen years of age and
96° by twenty-seven years of age. The other fourteen patients were still
growing and, at six years of age, the mean curve was 65°, with the
greatest being 112°. All patients had a small thoracic cage, which
resulted in appreciable reduction in vital capacity and reduced cardiovascular
function. Three patients died in the late second and third decade due to
pulmonary and cardiac abnormalities, and another patient died after surgery at
seven years of age. Younger age at the time of onset and younger age at the
time of curve progression was correlated with worse deformity and greatest
functional disability.
We know that severe scoliosis can lead to pulmonary hypertension and cor
pulmonale. This is the end result of compression on the pulmonary parenchyma
and the increased work of breathing caused by a deformed
chest10,11.
There is also a diminution in size of the thorax and limited excursion of the
ribs that result in alveolar hypoventilation along with hypoxemia and
hypercapnia. Pulmonary hypertension results from vascular bed restriction and
hypoxemia, and cor pulmonale and right heart failure
follow10-13.
This is a complication mostly of early onset deformities, as natural history
studies on adolescent idiopathic scoliosis fail to show this
association14. The
retrospective study on mortality and scoliosis related to infantile idiopathic
scoliosis and juvenile idiopathic scoliosis, by Pehrsson et
al.5, found an
increased risk of death from cardiopulmonary disease in these patients with
early onset curves. In the infantile group, the percentage of severe curves
(=70°) was 86%; in the juvenile group, it was 69%; and, in the
adolescent group, it was 46%. Sixty-nine of the 115 curves were of unknown
etiology and were considered idiopathic. Although an initial radiograph was
not always available to clearly rule out any other etiologies, such as
congenital scoliosis, the authors were able to correlate early onset scoliosis
with early death related to pulmonary compromise. The earlier the onset of
scoliosis, the more profound the effect on normal maturation of alveoli, a
process that is generally complete by eight years of
age15.
The tendency for infantile idiopathic scoliosis to resolve spontaneously in
a considerable proportion of
patients2,6,16-18
is rarely seen in the adolescent type and is even less frequent in the
juvenile
group19-22.
James initially reported on twelve progressive curves, eleven stationary
curves during the observational period, and only four resolving
curves2. Within
three years he had identified five more patients and, in 1959, his series of
seventy-seven resolving curves in 212 patients with infantile idiopathic
scoliosis was
published8. In 1965,
Lloyd-Roberts and
Pilcher16 reported
on 100 patients with idiopathic infantile curves and onset before one year of
age; in ninety-two of those patients, the curve resolved spontaneously. This
was a completely different perspective from that which had been reported six
years previously by James et
al.8. In that study,
in which fifty-five patients were from the Great Ormond Street clinic of
Lloyd-Roberts and Pilcher, the rate of spontaneous resolution was only 36%
(seventy-seven of 212 patients). Authors from different institutions have
reported rates of resolution of between 20% and 80% in their
series6,9,17.
In 1975, James18
reported on ninety patients with resolving scoliosis, the majority of whom had
resolution by three years of age. By six years of age, all curves had
resolved.
Diedrich et
al.23, in a
twenty-five-year follow-up study on thirty-four patients with resolving
infantile idiopathic scoliosis, were able to demonstrate that curves, once
resolved, did not recur during the adolescent growth spurt. This study
included fourteen patients who received only physiotherapy and twenty patients
who were managed with orthoses. No relevant disability was found in either
group of patients at the time of final follow-up.
Recently,
McMaster24 reported
a declining prevalence of new patients with infantile idiopathic scoliosis in
Edinburgh, which is a major referral center for scoliosis in Scotland. Between
1968 and 1972, this unit received a mean of sixteen to seventeen new patients
with infantile idiopathic scoliosis per year, with 34% (twenty-two of
sixty-five) having progressive curves. After 1972, there was a steady decline
in the number of referrals, to a mean of two patients a year from 1980 to
1982, and approximately half of these patients had progressive curves. On the
contrary, referrals for adolescent idiopathic scoliosis at this center
increased continuously from 1968 to 1982, and the referrals for juvenile
idiopathic scoliosis remained unchanged.
Following the published study of Scott and
Morgan6, several
European
studies17,24,25
were published that reinforced the
hypothesis6 of Scott
and Morgan that infantile idiopathic scoliosis was a much more frequent
condition in Europe (12.8%, 28 infantile and 218 idiopathic cases) than in
North America (0.25%, 1 infantile and 404 idiopathic cases). This higher
European prevalence raised questions regarding the etiology of infantile
idiopathic scoliosis.
Wynne-Davies25
analyzed 180 medical records from the Edinburgh Scoliosis Clinic from which
she selected 114 patients with idiopathic scoliosis and studied the prevalence
of scoliosis in their first, second, and third-degree relatives. She noted
only two peak periods of onset for scoliosis: before the first year of life,
and during adolescence. The period between two and seven years of age was
associated with very few new cases; thus, only two groups were available for
comparison: an early onset group (onset before eight years of age), and a
late-onset group (onset at eight years of age or older). In the early onset
group, 88% had left thoracic curves and there was a predominance of boys with
the disease (1.21 or approximately five boys to four girls). In the late-onset
group, 90% had right thoracic curves and there was a predominance of girls
with the disease (0.15, or one boy to seven girls). Wynne-Davies was able to
find an increased prevalence of scoliosis in the relatives of early onset
patients (2.6% in first-degree relatives) compared with that found in the
general population (0.39%), but the association was not as strong as the one
verified for the adolescent type (6.94%). A very important finding was that
plagiocephaly was present in all of the patients with infantile idiopathic
scoliosis, both in the resolving and the progressive types, compared with its
presence in only 11% in a control group. There was no relation between
parental age and infantile idiopathic scoliosis. Mental retardation and
epilepsy were the commonest of the associated conditions (fifteen children),
and families of these children had a greater prevalence of uncomplicated
idiopathic scoliosis. Developmental dislocation of the hip was present in four
patients, but only one subject had a relative who had both scoliosis and
developmental dislocation of the hip.
Hooper26 also
described this association in 1980, when he found a 6.4% prevalence of
developmental dislocation of the hip among 156 patients with infantile
scoliosis, which represents ten times the prevalence of developmental
dislocation of the hip found in the general population. For no apparent
reason, plagiocephaly, scoliosis, and hip dislocation were always on the same
side. Ceballos et
al.17 reported an
even greater prevalence of hip dysplasia (25% of 113 patients). This
association was most often in patients with resolving curves and in girls,
with no correlation between the side of the hip dysplasia and the convexity of
the spinal deformity. While there might be a relation between infantile
idiopathic scoliosis and adolescent idiopathic scoliosis, as both groups had
relatives with the other type of scoliosis, no definitive pattern of inherence
was found.
In 1965, Lloyd-Roberts and
Pilcher16 described
the molded baby syndrome. They suggested that intrauterine molding might be
the cause of all abnormalities. They recognized the consistent association
between scoliosis, plagiocephaly, rib-cage molding, and pelvic obliquity. This
theory, although attractive, did not explain the gender ratio, the difference
between geographic regions, and the fact that the curve usually is not present
at birth.
An environmental etiology was also proposed because of the increased
prevalence of infantile idiopathic scoliosis in Europe, suggesting the
presence of some factors acting in Europe and not in North
America7,24.
This would explain the great difference in the rates of infantile idiopathic
scoliosis that can be found between two regions in which the population
genetics most likely share similarities. This theory is related to how
children were nursed during the first year of
life7,24.
In North America, infants were placed in the crib in a prone position, with
its concomitant decompression of the spine, in contrast to most European
countries in which babies were usually put to sleep in the supine position. In
this position, children tend to turn to a slight oblique position, with some
preference for the oblique right
position27.
Mau7 suggested that
plagiocephaly, plagiothorax, and pelvic obliquity were caused by constant
pressure on the soft bones of infants. He added four more elements to the
molding theory: unilateral contracture of the neck muscles, associated with
oblique posture of the head; adduction contractures of the hips with slight
dysplasia; calcaneus foot deformity; and the subsequent development of a fixed
dorsolumbar kyphosis. Although these features can sometimes be transient,
their presence can serve as a guide to the early diagnosis of infantile
idiopathic scoliosis. The hypothesis proposed by Mau was that a combination of
increased bone plasticity and a slight retardation in maturation of the
central nervous system could explain the osseous deformities and the
asymmetrical contractures seen in the syndrome. In fact, mental retardation
and epilepsy (13% in 134 patients) were the most frequent infantile idiopathic
scoliosis-associated findings reported by
Wynne-Davies25.
The typical curve described by James was a low thoracic curve with
compensatory curves cephalad and
caudad2. The major
curve is almost always in the thoracic region, with variable degrees of
rotation. Secondary curves are not initially present in this type of
scoliosis, and it has been shown that when they appear, progression will
follow8,28.
The rate of convexity to the left has varied between 56% and
93%5,7,17,23.
Scott and Morgan, however, reported that 85% of curves were to the right when
the onset was after two years of
age6. Boys are more
often affected than girls, with the ratio being two or three boys to every
girl; only one series reported more girls than
boys17. Age of
onset is usually during the first year of life; curves are detected less often
in the second and even less often during the third year. Some curves will
progress and others will resolve spontaneously for no apparent reason.
Distinction between these two patterns usually can be established by the time
the child has reached the age of two years, when almost all of the resolving
curves are improving or double primary curves are developing, which, as stated
above, can be considered as a sign of progression.
In a progressive curve (Table
I), deterioration may take place rapidly and continuously from the
age of onset (Figs. 1-A through
1-D), or the curve may remain stable for several years before
deterioration begins. Any minor fluctuations seen during this period may be
related to flexibility of the spine and measurement error. When the curve
starts to progress, it will do so in a fairly constant way, and the end result
will depend primarily on the age at which the progression began. According to
Scott and Morgan6,
the rate of progression in their twenty-eight patients was 5° a year, with
great variation occurring between the time of curve recognition and the full
extent of progression.
Most progressive curves are thoracic, although double curves tend to be
more frequent in this group. There is no appreciable difference in gender
distribution, and right-sided curves occur more frequently when compared with
the curves seen in children with infantile
scoliosis17,23,29.
The age of onset appears to be older in these patients than in children with
resolving curves, with a mean age of onset around thirteen months and with an
important number being diagnosed at older than eighteen months of
age6,8,16,18,21,25.
When left untreated, these curves tend to progress to very serious deformity
with implications for pulmonary parenchymal development and decreased general
growth and life span.
The main characteristic of the resolving curve
(Table II) is its spontaneous
regression without treatment, usually in the first few years of life. Most of
the curves are thoracic. Secondary curves are rare in this group, although
thoracolumbar and lumbar curves have been reported in appreciable
numbers6,8,9,16,17,28-30.
The age of onset in patients with resolving curves is earlier than with the
progressive type; most curves are diagnosed before the first year of life,
with a mean age of approximately six
months6,23,26,29.
There is no appreciable difference in gender distribution and laterality of
the single thoracic curves, and there is a stronger association with
plagiocephaly when compared with the progressive type
(Table
III)8,9,16,17,25.
After regression, curves seem not to recur during the adolescent growth
spurt, even when resolution is incomplete. Diedrich et
al.23 reported that
32% (eleven of thirty-four) of their patients entered adolescence with curves
still measuring between 5° and 10°, and three additional patients had
curves >10°. None of the curves progressed. This is an intriguing
observation, as we would expect that incompletely resolved curves would tend
to worsen during the adolescent growth spurt. This raises the possibility of a
different etiology for resolving infantile scoliosis.
As most patients with infantile idiopathic scoliosis present during the
first year of life and most are diagnosed by six months of age, one of the
first goals is to rule out a congenital etiology. Early treatment of
progressive curves is important in order to avoid severe progression and
secondary complications. Treatment may also help stable curves to resolve.
Most of our treatment decisions are made on the basis of progression of the
Cobb angle from one visit to the next, as prognosis is difficult to determine
from a single visit.
As previously mentioned, the development of compensatory curves is
correlated with a worse prognosis. Progressive curves can be thoracic, double
major, or lumbar, whereas resolving curves are usually thoracic. Curves with
increased vertebral rotation and shorter length tend to start to progress at
an early age6.
In 1972,
Mehta31, in
analyzing the relationship between the apical rib and the apical vertebra
(Fig. 2), was able to correlate
this relationship with the risk of curve progression. If the apical vertebra
is overlapped by the corresponding rib head, the curve is said to be in phase
II and the chance of progression is enhanced. If the apical rib does not
overlap the vertebral body, the curve is said to be in phase I and the risk of
curve progression is lower. As the majority of the curves are in phase I at
detection, the way to anticipate progression is to evaluate the rib-vertebra
angle difference (RVAD), as measured at the apical vertebra. This corresponds
to the difference between the angle formed between the rib and a line
perpendicular to the apical vertebra on the concave side, where this angle is
usually greater, and that formed on the convex side. An RVAD of greater than
20° correlates with curve
progression31. This
observation has been confirmed by other
authors17,23,28
who identified the limited ability of the Cobb angle to predict the behavior
of the
curve28,31.
Ceballos et al. noted that 92% of the seventy-three resolving curves had an
RVAD of
<20°17,
which was a proportion very similar to that which Ferreira and James had
presented
previously28. The
six curves that made up the other 8% had an RVAD of =20° and were
improving at the time of the three-month follow-up. By the time of the
six-month follow-up, all of the seventy-three curves had a decrease in the
RVAD. On the contrary, of the seven single thoracic curves that progressed,
four initially had an RVAD of >20°, and these measurements remained
stable at the time of the three-month follow-up. By the time of the six-month
follow-up, all seven curves had an increase in the RVAD.
Mehta recognized special radiographic features that were helpful in
predicting progression in the early stage of disease in forty patients with
double thoracic and lumbar
curves31. Some of
the curves showed that the ribs at the apex remain almost horizontal for a
long time. These curves can easily be misinterpreted as benign curves because
they have an RVAD near 0° or even with a negative value. However, by
looking at the rib vertebral angles at the twelfth thoracic vertebra, a
significant asymmetry can be seen, with the rib on the concave side becoming
more vertical than the rib on the convex side, yielding a negative-value
RVAD.
In the series by Ceballos et al., nineteen of the twenty-six progressive
curves were double
curves17.
Initially, the curves had an apical RVAD of nearly 0°. Assessing the RVAD
at the twelfth thoracic vertebra, Ceballos et al. also noted a negative value
in these patients. This happens because the twelfth vertebra, being initially
part of the upper curve, becomes the apex of a curve that is developing caudal
to the first curve, and consequently the rib that is on the concavity of the
upper curve starts to drop as the convexity and vertebral rotation develop at
this level.
Kristmundsdottir et
al.32 reviewed 169
patients with infantile scoliosis, both progressive and nonprogressive and
with or without treatment. These authors were able to correlate a convex
rib-vertebra angle of <68° with progression. Although they recommended
further study with more patients, their results were significant enough for
them to recommend use of this measurement alone as opposed to calculating the
difference between the concave and convex rib-vertebra angles.
In comparison with infantile idiopathic scoliosis, when evaluating juvenile
idiopathic scoliosis, the value of the initial RVAD in predicting curve
progression seems to be more
debatable19-22.
Mannherz et al. found that increments above 10° in RVAD, over the course
of several measurements, and kyphosis of <20° were associated with
curve progression in forty-three patients in their
study21. In
twenty-two double thoracic curves, Figueiredo and
James20 also found
a low or negative value for the RVAD at the apical vertebra of the thoracic
curve. More recently, in a study involving 109 patients, Robinson and
McMaster22 found
that the curves that progressed had a mean initial RVAD of 31° (range,
12° to 45°), while the curves that resolved had a mean RVAD of 9°
(range, 5° to 20°).
The prevalence of neural axis abnormalities in infantile idiopathic
scoliosis is still unknown. Infantile idiopathic scoliosis is always a
diagnosis of exclusion, relying on a complete history and physical
examination. For some authors, these steps are adequate to rule out an
underlying neurological etiology. This is generally true with adolescent
idiopathic scoliosis, as the patient with this disease usually has a typical
history, a normal physical examination, and a typical curve pattern, thus
precluding the need for further imaging studies. With infantile idiopathic
scoliosis the case is different. The prevalence of neurological abnormalities
in patients with scoliosis who are twelve years of age or younger has been
reported to be as high as
20%31-34,
and most of such patients are asymptomatic or have very subtle physical
findings.
Lewonowski et
al.33, in a
magnetic resonance imaging study of twenty-six consecutive patients who had
idiopathic scoliosis and were ten years of age or younger, were able to
demonstrate a terminal lipoma in a four-month-old boy with a right lumbar
curve, among patients in the infantile age group. A three-year,
seven-month-old girl with right thoracic and left lumbar curves had a syrinx.
Two other patients, in the juvenile age range and with double curves, also had
neural axis pathology. A ten-year, six-month-old girl had an Arnold-Chiari
type-I malformation and hydromyelia. Of the five patients (19%) with
neuropathology, only two patients had atypical curves.
With use of magnetic resonance imaging, Evans et
al.34, in their
prospective trial of thirty-one patients who had scoliosis and were between
four and twelve years of age, found neural axis abnormalities in eight
patients (26%; mean age, 8.3 years). There were seven thoracic curves and one
thoracolumbar curve. Four of the thoracic curves had a left convexity. One
patient had nystagmus and ataxia and another had abnormal abdominal
reflexes.
Gupta et al.35
conducted prospective and retrospective studies to evaluate the prevalence of
neural axis abnormalities in patients who were ten years of age or younger and
who had idiopathic scoliosis and a normal neurological examination. In the
prospective study group, which totaled thirty-four patients with a mean age of
nine years (range, three to eleven years), 18% (six patients with a mean age
of seven years, three months [range, four to ten years]) had abnormalities. In
the retrospective review group, which comprised sixty-four patients with a
mean age of seven years, eight months (range, one to ten years), 20% had
neural abnormalities. Six of these patients were four years of age or younger.
Two patients had an Arnold-Chiari type-I malformation that was associated with
a cervicothoracic syrinx, and one had diffuse dural ectasia. In no patient
were the results of the neurological examination abnormal. Two patients
required surgery to address the neurological lesion. One of these patients had
a right thoracic curve and a left thoracolumbar curve, and the other patient
had a left thoracolumbar curve.
In a multicenter study on infantile idiopathic scoliosis, Dobbs et
al.36 found a 22%
prevalence of neural axis abnormalities in forty-six otherwise asymptomatic
patients with curves that were =20°. The mean age of these patients at
presentation was seventeen months (range, two to thirty-seven months). There
was no relation with gender, curve magnitude, or location or direction of the
curve. Five patients with a main thoracic curve had an Arnold-Chiari type-I
malformation, three patients with syringomyelia had unspecified curves, and in
two patients with a lumbar curve, one had a low-lying conus and the other had
a brainstem tumor. Eight of the ten patients with neural axis abnormalities
required surgery. Based on these observations, the authors recommended a total
spine magnetic resonance imaging evaluation for all infants with infantile
idiopathic scoliosis who had a curve of =20°.
Infantile idiopathic scoliosis can be one of the most severe forms of
scoliosis. It is characterized by severe vertebral rotation. At skeletal
maturity, untreated progressive curves are usually >100°. Patients with
infantile idiopathic scoliosis tend to have an increase in morbidity and
mortality secondary to restrictive lung disease and cor
pulmonale5,6.
Age—both at onset and when the curve starts to progress—is the
crucial factor in the development of pulmonary disease. In the early years of
life, pulmonary alveoli are developing and can be affected by the presence of
a severe chest wall
deformity15.
Alveolar development continues until eight years of age, although, by four
years of age, a substantial amount of alveoli are already
formed37,38.
The possible effect on alveolar development is the main reason why
Dickson39 chose
five years of age as the separating point between early onset and late onset
curves. Although possible implications for pulmonary development should always
be considered when delineating a treatment plan for a child with scoliosis, it
is our opinion that defining a classification on the basis of this risk alone
might be misleading in terms of possible etiology, the risk of progression,
the possible association with neurological abnormalities, and the potential
morbidity and mortality.
Infantile idiopathic scoliosis presents some unique characteristics that
are still difficult to interpret: male predominance, left convexity, the
tendency to show spontaneous resolution, the lack of progression during the
adolescent growth spurt after
resolution23, and
the difference in prevalence between Europe and the United
States6,24,25.
Until larger series can be studied, other currently unknown factors, extrinsic
in nature perhaps, may be the main reason for the geographic asymmetry. One
key question is: Is resolving scoliosis, as opposed to progressive scoliosis,
a different entity with a different etiology, or does infantile idiopathic
scoliosis have a full spectrum of presentations, as the current literature
seems to
support6,40,41?
This spectrum seems to range from the most benign forms to the most malignant
ones, which are those that have temporarily stable curves that may resolve or
progress later in time. In this context, the early conservative management of
some patients who have risk factors for progression may help resolve any
stable or less malignant curves by taking advantage of the rapid growth of the
child at this age41
(Figs. 3-A through 3-D).
The literature on the natural history of infantile idiopathic scoliosis is
difficult to interpret because patient demographics are poorly defined. Some
studies are based on the same previously reported population with new cases
added. Most series have only a few untreated patients, and many reports do not
discuss treatment. Although some authors consider that left thoracic curves in
boys are associated with a better prognosis, evaluating the literature is
confusing. Some authors report that resolving and progressive curves are
associated with almost the some proportion of boys and girls or left and right
curve convexity. Age of onset seems to be the key factor in terms of
prognosis. The age of detection for resolving curves is usually before one
year of age, whereas, for progressive curves, it is usually after one year of
age. Lloyd-Roberts and Pilcher demonstrated that 92% of curves resolved in
infantile idiopathic scoliosis when the onset was prior to one year of
age16. As there is
probably some delay between curve onset and detection, most resolving curves
are probably present before the age of six months. This may be the period when
some degree of delay in maturation of the nervous system, associated with poor
muscular tone, could act in conjunction with other unknown factors to cause
the development of the
deformity42.
The rib-phase classification system of Mehta and the rib-vertebra angle
difference can be related to
prognosis29. In all
published studies, the finding of at least one rib in phase II and an RVAD of
>20° for ribs in phase I was consistently present in all progressive
curves. Unfortunately, all of these studies were retrospective in nature. In
her study, Mehta reported on forty-six resolving curves and eighteen
progressive
curves31. However
the mean age of the patients at the time that the first radiograph was made
was significantly different between the two populations (p = 0.00043).
Patients with progressive curves had the first radiograph made at a mean age
of 13.4 months, whereas patients with resolving curves had a first radiograph
made at a mean age of 8.7 months. Hence, we may argue that an appreciable
number of progressive curves could have been classified as nonprogressive if
they had been identified earlier in time.
Until we have prospective studies to document the true predictive value of
these measurements, the presence of at least one rib in phase II and an RVAD
of >20° for ribs in phase I has to be regarded as a prognostic sign for
progression. An RVAD close to zero is also a sign of caution because it may
indicate the development of a double curve, which has a greater tendency to
progress. Tracking the Cobb angle may be an unreliable way to follow these
patients, as the maximum curve that has been reported to resolve spontaneously
measured 46°. A variation in RVAD in the interval between two different
clinical appointments seems to be a key factor when judging the risk of
progression, as most curves had a major change in RVAD at the time of the
three-month follow-up and a definitive change at the time of the six-month
follow-up.
On the basis of the information reported in the literature, infantile
idiopathic scoliosis can be followed conservatively until resolution or
progression is documented. An age of onset of one year or older, compensatory
curves and an RVAD of >20°, and ribs in phase II according to the
classification system of Mehta should be regarded as suspicious signs of curve
progression.
The incidence of infantile idiopathic scoliosis increased significantly in
Europe during a brief period in the second half of the twentieth century. In
the later part of that century, referrals for idiopathic infantile scoliosis
decreased appreciably in
Scotland24. One
could question whether there could have been a change in the referral system,
although the steady and increasing numbers of referrals of patients with
juvenile and adolescent scoliosis seem to refute this possibility. The
prevalence might be declining because of a resolution of unknown etiological
factors.
There is an increased awareness regarding the need to detect neurological
abnormalities associated with these infantile, early onset curves, even though
most of the patients have a normal clinical
examination33-36.
The need for a magnetic resonance imaging scan is still a matter of debate.
In this age group, these studies must be done with use of general anesthesia,
which is associated with some risks. Also, we would disagree with a
recommendation that has been made for the use of magnetic resonance imaging
for any infantile curve of
>20°36 as it
has been shown that the maximum-degree curve to resolve has been a 46°
curve. We therefore believe that curve progression should be the major
indication for a magnetic resonance imaging scan in patients with early onset
scoliosis. Although curves that have been diagnosed when the patient was
between four and ten years of age are very similar to the adolescent type,
their earlier onset should alert the physician to the increased risk of an
associated neurological abnormality. ?
Harrenstein RJ. Sur la scolioses des
nourrissons et des jeunes enfants. Revue d'Orthopédie.
1936;23:
289.23289
1936
James JIP. Two curve patterns in
idiopathic structural scoliosis. J Bone Joint Surg Br.
1951;33:
399-406.33399
1951
[PubMed]
James JIP. Idiopathic scoliosis; the
prognosis, diagnosis, and operative indications related to curve patterns and
the age at onset. J Bone Joint Surg Br.
1954;36:
36-49.3636
1954
[PubMed]
Ponseti IV, Friedman B. Prognosis in
idiopathic scoliosis. J Bone Joint Surg Am.
1950;32:
381-95.32381
1950
Pehrsson K, Larsson S, Oden A, Nachemson
A. Long-term follow-up of patients with untreated scoliosis. A study of
mortality, causes of death, and symptoms. Spine.
1992;17:
1091-6.171091
1992
[PubMed][CrossRef]
Scott JC, Morgan TH. The natural history
and prognosis of infantile idiopathic scoliosis. J Bone Joint Surg
Br. 1955;37:
400-13.37400
1955
Mau H. Etiology of idiopathic infantile
scoliosis. Reconstr Surg Traumatol.
1972;13:
184-90.13184
1972
[PubMed]
James JIP, Lloyd-Roberts GC, Pilcher MF.
Infantile structural scoliosis. J Bone Joint Surg Br.
1959;41:
719-35.41719
1959
[PubMed]
McMaster MJ, Macnicol MF. The management
of progressive infantile idiopathic scoliosis. J Bone Joint Surg
Br. 1979;61:
36-42.6136
1979
Hanley T, Platts MM, Clifton M, Morris
TL. Heart failure of the hunchback. Q J Med.
1958;27:
155-71.27155
1958
[PubMed]
Block AJ, Wexler J, McDonnell EJ.
Cardiopulmonary failure of the hunchback. A possible therapeutic approach.
JAMA. 1970;212:
1520-2.2121520
1970
[PubMed][CrossRef]
Swank SM, Winter RB, Moe JH. Scoliosis
and cor pulmonale. Spine.
1982;7:
343-54.7343
1982
[PubMed][CrossRef]
Branthwaite MA. Cardiorespiratory
consequences of unfused idiopathic scoliosis. Br J Dis Chest.
1986;80:
360-9.80360
1986
[PubMed][CrossRef]
Weinstein SL, Dolan LA, Spratt KF,
Peterson KK, Spoonamore MJ, Ponseti IV. Health and function of patients with
untreated idiopathic scoliosis: a 50-year natural history study.
JAMA. 2003;289:
559-67.289559
2003
[PubMed][CrossRef]
Reid L. Lung growth [abstract]. In:
Proceedings of the 3rd Symposium of the Institute of Diseases of the
Chest; 1970, London, England.
1970
Lloyd-Roberts GC, Pilcher MF. Structural
idiopathic scoliosis in infancy: a study of the natural history of 100
patients. J Bone Joint Surg Br.
1965;47:
520-3.47520
1965
[PubMed]
Ceballos T, Ferrer-Torrelles M, Castillo
F, Fernandez-Paredes E. Prognosis in infantile idiopathic scoliosis. J
Bone Joint Surg Am. 1980;62:
863-75.62863
1980
James JIP. The management of infants
with scoliosis. J Bone Joint Surg Br.
1975;57:
422-9.57422
1975
[PubMed]
Tolo VT, Gillespie R. The
characteristics of juvenile idiopathic scoliosis and results of its treatment.
J Bone Joint Surg Br.
1978;60:
181-8.60181
1978
[PubMed]
Figueiredo UM, James JIP. Juvenile
idiopathic scoliosis. J Bone Joint Surg Br.
1981;63:
61-6.6361
1981
[PubMed]
Mannherz RE, Betz RR, Clancy M, Steel
HH. Juvenile idiopathic scoliosis followed to skeletal maturity.
Spine. 1988;13:
1087-90.131087
1988
[PubMed][CrossRef]
Robinson CM, McMaster MJ. Juvenile
idiopathic scoliosis. Curve patterns and prognosis in one hundred and nine
patients. J Bone Joint Surg Am.
1996;78:
1140-8.781140
1996
[PubMed]
Diedrich O, von Strempel A, Schloz M,
Schmitt O, Kraft CN. Long-term observation and management of resolving
infantile idiopathic scoliosis: a 25-year follow-up. J Bone Joint Surg
Br. 2002;84:
1030-5.841030
2002
[CrossRef]
McMaster MJ. Infantile idiopathic
scoliosis: can it be prevented? J Bone Joint Surg Br.
1983;65:
612-7.65612
1983
[PubMed]
Wynne-Davies R. Familial (idiopathic)
scoliosis. A family survey. J Bone Joint Surg Br.
1968;50:
24-30.5024
1968
[PubMed]
Hooper G. Congenital dislocation of the
hip in infantile idiopathic scoliosis. J Bone Joint Surg Br.
1980;62:
447-9.62447
1980
[PubMed]
Gladel W. Die schrägseitenlage des
säuglings als gemeinsame Ursache der säuglingsskoliose und der
hûftluxation. [Oblique lateral decubitus of infants as a common cause of
infant scoliosis and hip dislocation]. Monatsschr Kinderheilkd.
1969;117: 196-8.
German.117196
1969
[PubMed]
Ferreira JH, James JIP. Progressive and
resolving infantile idiopathic scoliosis. The differential diagnosis. J
Bone Joint Surg Br. 1972;54:
648-55.54648
1972
Thompson SK, Bentley G. Prognosis in
infantile idiopathic scoliosis. J Bone Joint Surg Br.
1980;62:
151-4.62151
1980
[PubMed]
Roels J, Fabry G. Infantile idiopathic
scoliosis. Follow-up study. Acta Orthop Belg.
1977;43:
576-84.43576
1977
[PubMed]
Mehta MH. The rib-vertebra angle in the
early diagnosis between resolving and progressive infantile scoliosis.
J Bone Joint Surg Br.
1972;54:
230-43.54230
1972
[PubMed]
Kristmundsdottir F, Burwell RG, James
JIP. The rib-vertebra angles on the convexity and concavity of the spinal
curve in infantile idiopathic scoliosis. Clin Orthop Relat Res.
1985;201:
205-9.201205
1985
[PubMed]
Lewonowski K, King JD, Nelson MD.
Routine use of magnetic resonance imaging in idiopathic scoliosis patients
less than eleven years of age. Spine.
1992;17(6 Suppl):
S109-16.17S109
1992
[PubMed][CrossRef]
Evans SC, Edgar MA, Hall-Craggs MA,
Powell MP, Taylor BA, Noordeen HH. MRI of `idiopathic' juvenile scoliosis. A
prospective study. J Bone Joint Surg Br.
1996;78:
314-7.78314
1996
[PubMed]
Gupta P, Lenke LG, Bridwell KH.
Incidence of neural axis abnormalities in infantile and juvenile patients with
spinal deformity. Is a magnetic resonance image screening necessary?
Spine. 1998;23:
206-10.23206
1998
[PubMed][CrossRef]
Dobbs MB, Lenke LG, Szymanski DA,
Morcuende JA, Weinstein SL, Bridwell KH, Sponseller PD. Prevalence of neural
axis abnormalities in patients with infantile idiopathic scoliosis. J
Bone Joint Surg Am. 2002;84:
2230-4.842230
2002
Emery JL, Mithal A. The number of
alveoli in the terminal respiratory unit of man during late intrauterine life
and childhood. Arch Dis Child.
1960;35:
544-7.35544
1960
[PubMed][CrossRef]
Thurbeck WM. Postnatal human lung
growth. Thorax. 1982;37:
564-71.37564
1982
[PubMed][CrossRef]
Dickson RA. Early-onset idiopathic
scoliosis. In: Weinstein SL, editor. 2nd ed. The pediatric spine:
principles and practice. Philadelphia: Lippincott Williams and
Wilkins; 2001. p 321-8.321
2001
Mau H. The changing concept of infantile
scoliosis. Int Orthop.
1981;5:
131-7.5131
1981
[PubMed][CrossRef]
Mehta MH. Growth as a corrective force
in the early treatment of progressive infantile scoliosis. J Bone Joint
Surg Br. 2005;87:
1237-47.871237
2005
[CrossRef]
Wynne-Davies R. Infantile idiopathic
scoliosis. Causative factors, particularly in the first six months of life.
J Bone Joint Surg Br.
1975;57:
138-41.57138
1975
[PubMed]
Scott JC. Resolving scoliosis. J
Bone Joint Surg Br. 1959;41:
105-13.41105
1959