Article from Dr. Fearon - Sagittal Cranio: Surgical Outcomes and Long-Term Growth
In trying to get second, third, and fourth opinions on what to do about Ben's less than stellar post-op outcome, Dr. Fearon in Dallas sent me this article, and I wanted to share:
Sagittal Craniosynostosis: Surgical Outcomes
and Long-Term Growth
Jeffrey A. Fearon, M.D.
Emily B. McLaughlin, M.D.
John C. Kolar, Ph.D.
Dallas, Texas
Craniosynostosis is usually readily identifiable
as the result of inhibited growth occurring
perpendicular to affected sutures.
Sagittal synostosis is reported to be the most
common type of craniosynostosis, occurring with
an estimated incidence of somewhere between
one in 2000 and one in 5000 live births.1,2 Fusion
of the sagittal suture may occur either as an
isolated event or as one of multiple fused sutures.
Children who are born with multiple sutural
synostoses (usually syndromic conditions)
are known to have abnormal skull growth and
will typically require multiple operations to compensate
for this inherent growth disturbance.
However, for children with an isolated fusion of
the sagittal suture, craniofacial surgeons will typically
counsel parents that their child’s skull
should grow normally after correction and that a
single operation will be all that is needed to
correct this anomaly.
Many studies have been published examining
the efficacy of varying techniques used to normalize
skull shape in children who have developed
scaphocephaly from sagittal synostosis.
However, to our knowledge, there have been no
evaluations published that explore long-term
skull growth following surgical correction. This
retrospective study was undertaken not only to
assess the outcome of a large series of surgically
corrected patients but also to answer questions
related to the long-term cranial growth in children
born with isolated sagittal synostosis. After
cranial remodeling, does the skull grow normally
or does growth remain impaired, despite correction?
Does the associated compensatory frontal
bossing need to be directly surgically addressed;
does it improve, or does it worsen over time?
PATIENTS AND METHODS
This retrospective analysis was undertaken after
exemption approval was obtained from the
Institutional Review Board at North Texas Hospital
for Children at Medical City Dallas Hospital.
The sample group was compiled from a database
containing all patients who presented to our center
in Dallas between 1990 and 2003 (software was
designed exclusively for the Craniofacial Center
by M.A. Herbert, Ph.D., Medical City Dallas Hospital).
All patients given the diagnosis of sagittal
synostosis, on the basis of a clinical evaluation,
were included in this review. A total of 132 patients
were identified. Forty-three patients were subsequently
disqualified from this study: eight were
treated by the senior author (J.A.F.) at another
hospital in Dallas and had incomplete records, 12
patients had sagittal synostosis as a component of
a multisutural synostosis, and 23 patients were excluded
for nonoperative treatment (ridging without
significant scaphocephaly, family decided not
to treat, or patient treated by another surgeon).
From this group, the records of 89 patients with
nonsyndromic sagittal synostosis were retrospectively
reviewed.
Our standard treatment protocol includes a
preoperative computed tomography scan and a
full set of anthropometric measurements as components
of a full craniofacial team evaluation. After
postoperative swelling has abated (about 6
weeks postoperatively), a short series of anthropometric
measurements are taken of the cranial
vault to record the postsurgical changes in cranial
morphology. Additional measurements are taken
annually until age 4, then biennially until puberty,
to record postoperative craniofacial growth. All
measurements were taken by one of the authors
(J.C.K.).
Five measurements, which describe the basic
dimensions of the cranial vault, were selected for
analysis from the full battery of measurements.
These were maximum cranial breadth (eu-eu),
maximum cranial length (g-op), minimum cranial
breadth (ft-ft), head circumference, and auricular
head height (v-po).3 The cephalic index (width/
length) was calculated to describe cranial vault
shape. All anthropometric findings were compared
with sex- and age-matched normal standards
and converted to standard (Z) scores for
comparative purposes. Multiple anthropometric
measurements were obtained for 69 of the 89 patients
in this series. Long-term growth was assessed
from an analysis of patients who had a minimum
postoperative follow-up of 3 years (n 22; mean
follow-up, 4.7 years; range, 3 to 11 years). The
early, postswelling measurements (6 to 12 weeks)
were compared with the latest follow-up assessment.
To assess any effect of the evolution of surgical
techniques over the course of this series, the
early postoperative cephalic index in the first 10
patients treated was compared with that of the last
10 patients. Comparisons between groups were
made using paired Student t tests.
All surgical procedures were performed at Medical
City Dallas Hospital from 1990 to 2003 by one of
two craniofacial surgeons; 86 percent were performed
by the senior author (JAF), and 14 percent
were performed by Dr. Ian Munro (retired in
1996), together with one of five pediatric neurosurgeons
(93 percent of procedures were performed
by three of these neurosurgeons). Single-stage posterior
cranial vault remodeling was the procedure
of choice; this technique differed somewhat between
craniofacial surgeons and evolved slightly
over the course of this series, but it has remained
essentially unchanged over the last 5 years. The
preferred age for primary surgical correction was 4
months. An anterior approach was used only for a
minority of patients who had undergone previous
surgery elsewhere and had a significant anterior
deformity, or for an unusual presentation (anterior
fusions with accentuated frontal bossing and minimal
posterior involvement). All procedures were
performed under general anesthesia administered
by pediatric anesthesiologists, with two indwelling
intravenous catheters and an arterial line (no central
lines were used). A first-generation cephalosporin
was administered preoperatively. Access to the
skull was achieved utilizing a scalloped pattern coronal
incision, which was opened with amicroneedle.4
Recently, the incision length has been shortened to
avoid a scar in the temporal scalp. The osteotomy
design for the posterior cranial vault remodeling
used in this series has previously been described
(Fig. 1).5 A rectangular biparietal craniotomy was
performed, leaving as much of the periosteum intact
as possible to minimize blood loss during the
initial dissection. Vertical strips of bone were removed
anterior to the lambdoid sutures. The width
of these strips was designed to correlate with the
desired amount of overall anteroposterior cranial
reduction. An occipital craniotomy was performed,
and radial incisions were executed to flatten the
conical shape. Osteotomies were then made inferiorly,
from the vertical perilambdoid excisions, and
continued anteriorly to create bilateral anteriorly
based parietal-temporal bone flaps. The radially cut
occiput was rotated 180 degrees and re-attached to
the skull base in a more superior and more anterior
location. The vertex biparietal bone flap was then
cut into four strips lengthwise, and the two central
strips were re-attached to the frontal bones and
then shortened posteriorly, to complete the anteroposterior
reduction. The remaining two lateral
bone strips were then shortened appropriately and
replaced. This anteroposterior calvarial reduction
results in a compensatory posterior widening, which
occurs by virtue of the anteriorly hinged parietaltemporal
bone flaps. On average, the posterior biparietal
distance was increased about 2 cm with this
maneuver. Osteosynthesis was achieved with absorbable
sutures, except in patients treated very early in
this series. The surgical technique did not vary with
the age of the patient, and in all cases, the correction
resulted in a completely intact calvaria, without
any skull defects left unfilled with autogenous bone.
After reconstruction, the scalp was irrigated with an
antibiotic solution and closed in two layers with
absorbable sutures. No postoperative dressings or
drains were used. The patients were observed overnight
in the pediatric intensive care unit, and typically
1 additional day on the floor, before being
discharged home.
RESULTS
The average age at the time of surgery in this
series was approximately 8 months, with a range
from just under 2 months to 5 years. Seven of the
89 patients were older than 18 months of age; four
were between 3 and 5 years old. The average
length of surgery was 2 hours and 16 minutes and
ranged from 1.5 to 7 hours. The average weight of
the child at the time of surgery was 8.1 kg, and the
average estimated blood loss was 165 cc, with a
range of 50 to 375 cc. The average volume of blood
transfused (allogenic) was 119 cc, and 70 percent
received banked blood. After the institution (in
2001) of preoperative erythropoietin administration,
and use of a cell saver for blood recycling, the
overall transfusion rate decreased from 92 percent
to 21 percent of patients. The average volume of
recycled autogenous blood transfused was 40 cc
and ranged from 0 to 110cc. The average preoperative
hemoglobin level was 12.1 g/dl (hematocrit
of 39.1 percent), with 51 percent of patients in
this series receiving preoperative erythropoietin.
The average immediate postoperative hemoglobin
level was 10 g/dl (hematocrit of 29.8 percent).
The average hospital length of stay was 2.8 days (all
patients were kept in the hospital for a minimum
of 2 days).
There were no deaths, major complications, or
infections in this series of patients. Only one patient
underwent a secondary related operation,
many years postoperatively, for removal of palpable
wires (placed before the routine use of absorbable
suture osteosynthesis). Two patients experienced
small, isolated (1 cm) wound
dehiscences; one healed by secondary intention
and the other was repaired in the office. Other
complications included one patient each with syndrome
of inappropriate secretion of antidiuretic
hormone, reactive airway disease, urinary retention,
and a seroma (aspirated in the office). Representative
examples of outcomes are seen in Figures
2 through 5.
The results of the comparison between early
postoperative and long-term postoperative anthropometric
findings are presented in Table 1.
Overall, the results of this long-term analysis demonstrated
a statistically significant deficiency in
growth. Growth in both cranial breadth and cranial
length was less than predicted (p 0.001 and
p0.05, respectively). However, growth was more
deficient in cranial breadth than in cranial length,
which resulted in an overall regression of the cephalic
index over time (p0.001). The minimum
frontal breadth measurement (one indicator of
frontal bossing) improved immediately postoperatively
(although this area wasnot directly surgically
addressed), but it did not significantly narrow over
the studied time period. The head circumference
also showed a statistically significant diminished capacity
for growth. The head circumference was significantly
above normal preoperatively, and over the
study period fell to less than 1 SD over the mean.
Finally, the cranial vault height grew less than predicted,
although this was not significant.
A preoperative to postoperative comparison of
the first 10 patients in this series with the last 10
patients (Table 2) showed that both groups demonstrated
a statistically significant improvement in
their scaphocephaly with normalization of the cephalic
index. Although the last 10 patients treated
at the end of this series had a greater improvement
in their cephalic index (more brachycephalic),
this difference was not statistically significant.
DISCUSSION
Sagittal craniosynostosis may be associated
with more different surgical repairs than any
other type of craniosynostosis. The earliest treatment
for sagittal craniosynostosis relied on suturectomies,
such as those described by
Lannelongue6 and Lane.7 These simple sutural
excision techniques gradually evolved toward a
more extensive strip craniectomy and then on to
an extended strip procedure modified by active
anteroposterior shortening (“pi” procedure).8,9
Subsequent modifications of the pi procedure
have been described, with more extensive osteotomies
designed to remodel the entire
skull.10 –12 Some surgeons believe that it is necessary
to treat sagittal synostosis with staged procedures
(operating on the posterior half of the
skull first and then the anterior half of the skull
at a later date), to also correct the frontal
bossing.13 To our knowledge, this two-stage procedure
has been reported only for the treatment
of older children presenting with scaphocephaly;
however, we are aware of some surgeons who
use this two-stage approach in infants. Recently,
there has been a renewed interest in a modified
strip procedure. Jimenez and Barone14 reported
using an endoscope to perform a strip craniectomy;
then, instead of actively remodeling the
skull shape during surgery, changes in skull
shape are performed by an orthotist during a
prolonged period of active head-banding.
The patients in our series were treated with a
posterior remodeling procedure; the majority was
treated with a technique that we have previously
described.5 This technique is based on the creation
of bilateral anteriorly hinged parietal-temporal
bone flaps. These flaps permit a bilateral
compensatory widening that occurs commensurate
with anteroposterior shortening (with the re-
modeled occipital bone replaced in a higher and
more anterior position). The rationale for this
single-stage posterior cranial vault remodeling
procedure is based in the desire to address the
most affected region of the skull, as well as to select
the smallest and safest procedure that will reliably
normalize the abnormal skull shape. A more extensive
approach was not used on any patients in
this series. Isolated strip procedures were also not
performed on any patients because they are more
likely to result in skull defects that might later
require surgical intervention; moreover, published
results for strip procedures have shown that
the cephalic index is infrequently normalized, and
that a more extensive remodeling procedure
more reliably improves both the cephalic index
and appearance.15–23 At the other end of the spectrum,
total cranial vault procedures were not performed
because of the risks associated with total
craniectomies, placing the patient in the “seal position,”
and the increased blood loss. There have
also been no studies showing that a total calvarial
vault procedure produces a better result than a
more limited posterior procedure. A two-stage approach
was not planned for any patients in this
series because it seems intuitive that a two-stage
operation would double the risk for the child.
Moreover, it can be reasonably argued that a second
staged procedure should be held in abeyance
until the child is older, at which time both the
child and the parents may determine whether
frontal remodeling is necessary (and none did in
this series). Although a few centers have begun to
evaluate the endoscopically assisted craniectomy
technique, this procedure is not currently being
used at our center in Dallas because (1) the clinical
results with the endoscopic technique appear
to be less consistent than the reproducibly predictable
results achievable with remodeling, (2)
patients must wear a headband for up to 1 year
postoperatively, versus immediate correction with
remodeling, and (3) we are concerned that performing
craniectomies in infants with such a limited
exposure places the patient at a greater risk
for serious complications than does a more open
procedure. The endoscopic procedure’s primary
advantage may be the avoidance of a low temporal
scar in boys. Motivated by this cosmetic benefit,
the incision used in Dallas has been shortened to
keep the scar off the low temporal scalp.
This series of 89 surgically treated patients
with sagittal synostosis compares favorably to other
published series of similar size with respect to complication
rates.20,24,25 There were no mortalities, no
major complications, and no infections. With a
clinical follow-up of as long as 14 years, none of the
patients treated primarily underwent a second cranial
vault remodeling procedure. After surgical
correction, only one of 89 patients (1.1 percent)
underwent a secondary procedure. This patient
was brought back to the operating room for a brief
anesthetic to remove some symptomatic wires that
had been placed at the initial procedure (since
1992, all cranial osteosyntheses have been performed
by the senior author using absorbable
sutures).5 No patients were noted postoperatively
to have any palpable bone defects. After the institution
of routine preoperative recombinant
erythropoietin and blood recycling, almost 80 percent
of infants did not require any allogenic blood
transfusions.26, 27 The average length of hospitalization
was just over 2 days (a time period that the
surgical team considered the minimum safe
length of stay for a child undergoing a craniotomy).
Furthermore, as assessed by the cephalic
index (measured by direct surface anthropometric
measurements), the performed correction was
routinely and reliably effective in correcting the
abnormal skull shape.
Our protocol entailed obtaining preoperative
anthropometric measurements, followed by a first
set of postoperative measurements that immediately
followed the resolution of swelling (before
any significant growth had occurred). Additional
follow-up occurs yearly until the child is 4 years
old, then biennial measurements are continued
until facial growth is complete. The first postoperative
measurement provides an assessment of
the success of the surgical procedure, and the
subsequent measurements evaluate growth. This
investigation found that following the surgical repair
for sagittal synostosis in infancy, skull growth
is not normal. Moreover, we found that not only
is overall skull growth diminished but there is also
less growth than would normally be expected in
cranial width than in cranial length. The overall
effect of this impaired growth is that the skull
gradually becomes more dolicocephalic (the cranial
index worsens). Growth in head circumference
was also impaired, resulting in a trend toward
normalization (declining from above the 98th percentile
toward the 50th percentile). It is tempting
to speculate that the poor growth observed over
the 3- to 11-year follow-up period was an inherent
component of the underlying process that initially
presented as a fused sagittal suture. However, any
potential effects of surgical intervention cannot be
dismissed without following a cohort of untreated
children (and this would have obvious ethical impediments).
Another interesting finding was the
immediate reduction in frontal bossing (as measured
by minimum frontal breadth), especially
considering that a posterior procedure was performed
that did not directly address the frontal
region. We speculate that this frontal narrowing
was a secondary compression effect, which resulted
from the posterior widening. As a result of
the significant reduction in the anteroposterior
dimension, some diminution of frontal height
would also be expected. Also, the forehead will
appear less prominent when viewed from the front,
because of the wider posterior skull. Our assessment
did discover that the frontal bossing does not
worsen over time, and none of the patients in our
series, who were treated primarily with a posterior
procedure, have required frontal surgery.
Considering that normalization of appearance
is the primary motivation for correcting sagittal
synostosis, and that this treatment is typically performed
very early in growth, long-term outcome
evaluations provide valuable information concerning
the efficacy of a particular repair. In this series
of patients treated for scaphocephaly resulting from
sagittal craniosynostosis, posterior remodeling resulted
in normalization of the cephalic index without
the need for a secondary procedure (objectively
determined by direct anthropological measurements,
and subjectively determined by the patient,
the patient’s family, and the surgeon). The results
of this study show that skull growth is not normal
following scaphocephaly correction. Given these
findings of diminished calvarial growth, surgeons
may wish to consider expanding their goals for
treatment beyond the normalization of the cephalic
index to an overcorrection of the presenting deformity.
Infants who are corrected to a normal cephalic
index will develop dolichocephaly with subsequent
growth. Therefore, to provide the best
long-term aesthetic result, a correction that results
in slight brachycephaly at the end of the procedure
should be sought.
Jeffrey A. Fearon, M.D.
7777 Forest Lane, C-700
Dallas, Texas 75230
cranio700@aol.com.
-------------------------------------------------
I SO wish we would have either done this or endo with a helmet instead of the strip craniectomy. I felt like we didn't have that many options, because the strip was just what our surgeon did. Ben's head has reverted back to the narrow sagittal shape and we aren't sure what's going to happen next.
PM me and I'll email you the document as a PDF with pictures. It is too large to be able to attach to this forum. It includes photos of children before and after surgery and other diagrams showing where they cut the skull during surgery. I couldn't get Adobe to let me save just the images to load here, but I'll try again later.
Reply With Quote
Sagittal Cranio Kid
