European Journal of Paediatric Neurology (2004) 8, 21–34

www.elsevier.com/locate/ejpn

ORIGINAL ARTICLE

Effect of positioning on the incidence

of abnormalities of muscle tone in low-risk,
preterm infants
L. Vaivre-Doureta,*, K. Ennourib, I. Jradc, C. Garrecd, E. Papiernike

a
Service de gynécologie obstétrique, Groupe hospitalier Cochin –St Vincent de Paul. Maternité Port-Royal
Baudelocque, Paris, France and INSERM U-483, Université Pierre et Marie Curie, Paris VI and Necker Enfants
Malades, Paris, France
b
INSERM U-483, Université Pierre et Marie Curie, Paris VI and Necker Enfants Malades, Paris, France
c
Service de gynécologie obstétrique, Groupe hospitalier Cochin—St Vincent de Paul. Maternité Port-Royal
Baudelocque, Paris, France
d
INSERM U-483, Paris, Université Pierre et Marie Curie, Paris VI, Paris, France
e
Service de gynécologie obstétrique, Groupe hospitalier Cochin—St Vincent de Paul, Maternité Port-Royal
Baudelocque, Paris, France
Received 21 August 2003; received in revised form 29 September 2003; accepted 2 October 2003

KEYWORDS
Preterm infant; Muscle
abnormalities;
Positioning;
Orthopaedic; Neuro-
development; Evaluation
Summary Background. The preterm infant is subject to the force of gravity: when its
body lies pressed against the mattress on which it is placed.
Aims. The purpose of this study was to investigate short-term effects of varied post-
natal lying positions in order to prevent neuromuscular and postural abnormalities.
Methods. 60 low risk preterm infants of 31– 36 weeks gestational age were enrolled
for this randomised clinical trial. Initially each child underwent neurological and
psychomotor assessments which included tonus and reflex protocols as well as
behavioral, sensory motor and postural examinations. The lying positions of the
treated group were varied (back, prone, and side) using a specially designed moldable
mattress that maintained the functional position of the infant’s body. The control
group was placed on their stomachs, (the standard lying position used in 1994) with a
standard orthopaedic bolster support under their hips. All infants underwent a second
round of examinations upon discharge to assess any changes in neurological and
psychomotor outcomes.
Results. The sensory-motor skills examinations showed significant abnormalities in
the control group: (1) dominance of the extensor muscles due to muscle shortening, (2)
hyper abduction and flexion of the arms, and (3) global neuromuscular rigidity.
Psychomotor and neurological exams of the control and treatment groups showed
delayed developmental muscular acquisitions for infants in the control group.
Conclusion. Regular changes in posture, while retaining correct functional positions,
allowed maintenance of normal neuromuscular and osteo-articular function and
permitted the development of spontaneous and functional motor activity in low-risk
perterm infants.
Q 2003 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights
reserved.

Abbreviations: OR, odds-ratio; CI, confidence interval.

*Corresponding author. Tel.: þ33-44-49-40-14; fax: þ 33-44-49-47-10.
E-mail address: lvaivre@snv.jussieu.fr (L. Vaivre-Douret).

1090-3798/$ - see front matter Q 2003 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.ejpn.2003.10.001
22
Introduction
Preterm infants are usually placed lying prone1,2
during intensive care and this same position was
most commonly used for low-risk, preterm infants
when this study began in 1994. This position is
reported to have beneficial effects, including
increased lung compliance and tidal volume,
regularity of breathing rate and augmented partial
pressure of oxygen.3
New guidelines have been recently issued which
were designed to prevent Sudden Infant Death
Syndrome (SIDS).4 – 7 These guidelines strongly rec-
ommend that infants under the age of one be placed
on their backs for sleeping. As a result, many
maternity units have changed the lying position of
low-risk preterm infants from their stomachs to
their backs. Our interest lies in the sensory and
motor consequences of placing preterm infants
almost exclusively either on their stomachs or
backs, as it relates to possible future abnormalities
in psychomotor development.
Whether positioned on their backs or on their
stomachs, it is inevitable that infants will lie with
all four limbs flattened against the mattress in
external rotation as a result of gravity, and will
have to turn his/her head to avoid facing into the
mattress (if lying prone). This forced positioning has
been shown to lead to early orthopaedic hip
deformities in neonates. Grenier8 investigated
methods of preventing hip deformities in brain
damaged infants. His research demonstrated that
excessive flexion of the legs in a prone position is a
non-functional position (frog-leg position). If this
position is maintained, over time, the flexor and
adductor muscles of the hip shorten within weeks.
This shortening is completely reversible if the brain
is intact, and normal motor activity is subsequently
restored. However, if there is an ante or perinatal
brain injury, early orthopaedic deformities (sub-
luxation, and/or crossed extension of the legs with
spastic diplegia) are not reversible and will exacer-
bate any existing motor handicap(s).
These findings were confirmed by Monterosso
et al.9 who investigated the use of a postural
support to reduce the frog-leg position in preterm
infants less than 31 weeks. They demonstrated
that, short-term, there was a significantly lower
incidence of the frog-leg position in the treated
group. Similarly, Downs et al.10 demonstrated the
benefits of changing the functional position of the
hips to avoid the frog-leg position. In their study, a
population of 45 preterm infants born at less than
33 weeks gestation and observed until full-term to
about 37 –44 weeks (mean 40 weeks), were divided
into two groups. Findings revealed significant
L. Vaivre-Douret et al.
differences between the two groups, with normal
postural examination results for the infants in the
treated group due to legs being placed in a
functional position. Updike et al.2 discussed other
possible consequences of the frog-leg position in
preterm infants which include the tendency to
retract the shoulders and arms, demonstrated
difficulty in axel rotation of the neck, and increased
hyperextension of the neck muscles especially
evident during prolonged periods of assisted venti-
lation. As such, Updike et al.2 suggested that
regularly changing the position of preterm infants
(prone, back, side) could eliminate these postural
deviations.
The studies of Georgieff et al.11 and Georgieff
and Bernbaum12 demonstrate that 46% of a group of
preterm infants weighing less than 1750 g, exam-
ined at the ages of 3, 6, 12, and 18 months
(corrected for gestational age) exhibited scapular
retraction, particularly evident in very low birth-
weight infants who had been intubated. Muscle
tone and development of the shoulders was shown
to delay acquisition of the ability to sit and handle
objects, an important psychomotor milestone.13,14
Our own observations of full-term infants who
had been placed on their stomachs from birth
confirms these findings15,16 and we attribute them
to a peripheral, postural origin. In terms of
postural abnormalities, we observed a retraction
of the scapula, particularly the trapezoid muscles,
a tendency to opisthotonos (hyperextension of
trunk and head), arms in flexion/abduction and
external rotation of legs and orthopaedic deform-
ity of the feet. As a result, the following motor
abnormalities were noticed: Head control exclu-
sively by extensor muscles, inability to reach with
arms and late sitting and early standing
accompanied by behavioral hyperexcitability with
resistance to all postural change. One of our
studies, on a cohort of full-term infants placed
lying on their backs,15,16 demonstrated that tem-
porary muscular abnormalities develop rapidly.
Abnormalities such as hyperextension of the
trunk, make it difficult to interpret the assessment
of motor function and can lead to confusion
between disorders of central origin and postural
abnormalities of peripheral origin. However, we
found that using prophylactic orthopaedic position-
ing in these preterm infants provided psychomotor
benefits. In this report we focus on the posture and
eventual use of orthopaedic positioning of the
preterm infant as part of the overall function of
the infant.
In our 1994 study on premature infants, our
purpose was to test an intervention involving
routine positional changing of these infants
Effect of positioning on the incidence of abnormalities of muscle tone in low-risk, preterm infants
(switching between back, prone, and side sleeping
positions) in an effort to prevent the onset of
secondary acquired muscular anomalies. Our pri-
mary goal was to promote normal motor develop-
ment from the outset (without the defects which
result from acquired muscle shortening) and
secondarily to provide for the most optimum
conditions necessary to conducting an objective
neurological and psychomotor evaluation. This
study investigated the neuromotor differences
between a group of low-risk preterm infants
receiving changes in physiologically functional
supported positions with a similar group of preterm
infants receiving no such intervention. The pro-
cedure consisted of changing the lying position of
the preterm neonate randomly between front,
back, and side positioning, while maintaining
physiologically functional positions. The control
preterm group did not undergo any particular
procedure other than the standard care provided
by the unit, that is, they were placed on their
stomachs with a small bolster (hand rolled sheet)
under their hips to raise the pelvis and prevent hip
deformities.
Materials and methods
This study was a prospective randomized clinical
trial and the design has been approved by the
Ethical Committees identified by number GH/FL/CC
93016.
Sample
We initially recruited 62 infants, however, two of
them were subsequently transferred to another
neonatal centre. As such, the study included 60 low
risk preterm infants, 31 – 36 weeks gestational age.
All infants were born at the Baudelocque maternity
ward and admitted directly to the neonatal unit
during 1994. The definition for a low-risk preterm
infant is what was used for the inclusion criteria,
that is, infants between 31 and 36 weeks in
gestational age without congenital or genetic
neurological anomalies and with normal cranial
ultrasound. Five infants were under observation for
2 days for assisted mechanical ventilation, how-
ever, none actually required mechanical venti-
lation. In regards to the perinatal data between
the control and treated groups, no significant
differences existed (Table 1).
The exclusion criteria excluded preterm infants
who had initially received treatment in an
intensive-care unit; multiple births (triplets or
23
above); breach deliveries; deformities of prenatal
origin;> fractures, sub-luxation(s); genetic abnorm-
alities; hereditary diseases; neurological abnorm-
alities detected within the first week of life;
mechanical ventilation; refusal from the parents
to continue the study; or referral of the infant to
another department during convalescence.
Intervention
Nursing staff were asked to reposition the infants
every 3 – 4 h (corresponding to mealtimes). Change
of position was reported by the nurse on call on a
planning chart for each infant. They were
instructed to vary the lying positions equally.
Infants in the treatment group were placed on a
special moldable mattress while lying on their sides
(Fig. 1) or back only.17 The mattress was able to
support the infant in lateral decubitus by molding
around the rear of the body to support the head in
line with the trunk. When the infant was lying on its
back, the mattress was specially shaped to hold the
shoulders back while keeping the knees bent
forward. While lying prone he/she did not require
the special bolster as the hospital supplied bolster
was used and placed under their hips as with all
admitted preterm infants.
Control situation
The control group was positioned in a prone
position, according to the usual practice and care
in the unit (in the year 1994). A preventative
orthopaedic position of the pelvis and legs was used
for all the children in this group (hand rolled sheets
used to form a small bolster), as suggested by
Grenier18 to prevent the frog-leg position. We used
the same hand rolled sheets to raise the pelvis,
without flexing the hips to more than 908, ensuring
no external rotation or adduction of the legs.
Procedure
This study received Internal Review Board approval.
Consent was obtained from all parents at the time
of enrollment in the study. During the study, there
were two different examiners, one for the neuro-
logical assessment and another for the psychomotor
assessment. The neurological examination was
conducted by the same pediatrician throughout
the course of this study and the psychomotor
examination was conducted by a psychomotor
therapist trained by one of the authors (L. V-D).
The perinatal data (conditions during pregnancy,
delivery, and history of any abnormalities linked to
birth) was included in the infants’ charts as a matter
24 L. Vaivre-Douret et al.

Table 1 Birth variables: differences between treatment group and control group.

Population Treatment group Control group p

Total ðN ¼ 60Þ Total ðN ¼ 30Þ 31– 33 weeks 34 –36 weeks Total ðN ¼ 30Þ 31–33 weeks 34– 36 weeks
(A) ðn ¼ 10Þ (B) ðn ¼ 20Þ (A) ðn ¼ 11Þ (B) ðn ¼ 19Þ

Gender
Boy 30 (50%) 14 (47%) 6 (60%) 8 (40%) 16 (53%) 7 (64%) 9 (47%) 0.61
Girl 30 (50%) 16 (53%) 4 (40%) 12 (60%) 14 (47%) 4 (36%) 10 (53%)
Average 34.09 þ 1.30 34.00 þ 1.40 32.38 þ 0.83 34.82 þ 0.77 34.17 þ 1.12 32.86 þ 0.65 34.94 þ 0.67 0.61
gestional
age (week)

Apgar
1 min Average 8.50 þ 1.24 8.27 þ 1.48 8.60 þ 1.51 8.10 þ 1.48 8.73 þ 0.91 8.64 þ 0.92 8.79 þ 0.92 0.15
5 min Average 9.65 þ 0.91 9.53 þ 1.11 9.90 þ 0.32 9.35 þ 1.31 9.77 þ 0.68 9.82 þ 0.60 9.74 þ 0.73 0.33
Average birth 2174 þ 464 2114 þ 465 1823 þ 435 2259 þ 416 2235 þ 462 2002 þ 324 2369 þ 484 0.32
weight (g)

Pregnancy
Single 36 (60%) 20 (67%) 7 (70%) 13 (65%) 16 (53%) 9 (82%) 7 (37%) 0.3
Multiple 24 (40%) 10 (33%) 3 (30%) 7 (35%) 14 (47%) 2 (18%) 12 (63%)
Birth type
Spontaneous vaginal 29 (48%) 16 (53%) 6 (60%) 10 (50%) 13 (43%) 8 (73%) 5 (26%)
Induced vaginal 7 (12%) 5 (17%) – 5 (25%) 2 (7%) – 2 (11%) 0.22
Caesarian 24 (40%) 9 (30%) 4 (40%) 5 (25%) 15 (50%) 3 (27%) 12 (63%)

Prematurity
abnormalities
Eutrophia 45 (75%) 23 (77%) 7 (70%) 16 (80%) 22 (73%) 7 (64%) 15 (79%) 0.77
(normal stature)
Hypotrophia 15 (25%) 7 (23%) 3 (30%) 4 (20%) 8 (27%) 4 (36%) 4 (21%)
(small weight)
Coiling of 3 (5%) 1 (3%) 1 (10%) – 2 (7%) – 2 (11%) 0.56
ombilical cord
Other complications 5 (8%) 3 (10%) 1 (10%) 2 (10%) 2 (7%) – 2 (11%) 0.61
(icterus, infection)

of custom and access to this information was
authorized by the pediatric Chief of Staff for the
purpose of this study. Randomly numbered exam-
ination tables were used to set up two randomized
groups (treatment and control). Randomization was
stratified by gestational age at delivery into one of
two-groups: 31 – 33 weeks gestation (referred to as
Group A) or 34 –36 weeks gestation (referred to as
Group B). Infants were differentiated into two
groups due to obvious maturation differences at
birth. The data were analyzed using Chi square with
Yates’s correction (without regression) or Student’s
t-test. Odds-ratio (OR) was also used.
Neurological and psychomotor assessment
Neurological assessment
The neurological assessment protocol was, in part,
inspired from the neurological examination using
the methods of André-Thomas and Saint-Anne
Dargassies,18 Amiel-Tison and Grenier,19 Prechtl
and Beintema20 and Brazelton.21 This is a collection
of muscle tonus and primitives reflex examinations
used to specifically assess neurological maturation.
Rationals for items included and the reliability and
validity of each assessment were previously high-
lighted and discussed22,23 and due to their length,
have been omitted here. The selected items were
few as the groups did not include infants with
neurological abnormalities and as such, neurologi-
cal items selected were those traditionally assessed
only on normal infants (see Appendix A). Due to
differences in neurological maturity, as related to
gestational age, we calculated a neurological
maturity score based22,23 on level of maturity
using the following items: Ocular motor items þ
passive muscle tone of the limbs þ active muscle
tone þ primitives reflexes. Scores increase with
gestational age (Appendix A).
Psychomotor assessment
Each infant underwent a psychomotor assessment
protocol for newborn infants.22,23 This examination
was comprehensive and included assessment of
behavior, visual/auditory function, sensory-motor
skills, postural control, and orthopaedic lower
Effect of positioning on the incidence of abnormalities of muscle tone in low-risk, preterm infants
Figure 1 Premature newborn lying on side on the BONDOSR moldable mattress.
limbs position. The psychomotor assessment proto-
col included normed reference scores for each
category which were obtained for each infant
by adding the various sub-scores from each cat-
egory22,23 (see Appendix B). Using their individual
score, the infant could then be assessed by
comparing their score to the reference score
(Table 2).
The first examination, performed within the first
week after birth, including both the neurological
protocol and only orthopaedic lower limbs examin-
ation of the psychomotor assessment due to
maturational considerations. The second examin-
ation, performed at the time of discharge from the
unit, consisted of the neurological and psychomotor
protocol because the coordination required to
complete test items was now present due to overall
maturation. For all examinations, the infant was
brought and placed on a special examination table
by a nurse to insure that the examiner was blind to
the infants sleep position in the incubator or bed.
Results
Initial neurological and orthopaedic lower limbs
testing of infants did not show any significant
difference between the control and treatment
group.
All children were assessed when they were
discharged after convalescence with a mean period
of 38 weeks of gestational age (SD ¼ 3 weeks, 3
days) for the treated group and of 38.4 weeks of
25
gestational age (SD ¼ 1 week, 4 days) for the
control group. After inspecting the protocols of
the positioning schedule, we found that the children
in the treated group spent 60% of their time lying on
their sides, 30% lying on their back and 10% on their
stomach. Recall that infants where repositioned
every 3 – 4 h, corresponding with mealtimes.
Neurological assessment
All infants had a maturity score (see Section 2) in
the normal range for their gestational age when
they were discharged after 38 weeks, with a score
of 67 – 74 points.22,23 This further confirmed that our
sample did not include any children with abnormal
neurological signs. In the case of the clinical upper-
trunk and axial passive tone, we found significantly
more dysfunction in the control group (Table 3).
With regard to the items which assess the balance
between the flexors and the extensors of the neck,
there was a greater tendency in the control group to
exhibit an imbalance of tone favoring the extensor
muscles of the neck (tð60Þ ¼ 3:5; p , 0:01). This was
additionally revealed by the poor response to
traction of flexors muscles. There was a significant
difference between the passive tone in the treated
and control groups ðx2ð1; N ¼ 60Þ ¼ 11:4; p , 0:03Þ:
Active tone was also significantly different in the
treated and control groups ðx2ð1; N ¼ 60Þ ¼
37:1; p , 0:01Þ:
These findings demonstrate a strong tendency in
the control group towards abnormalities
with excessive tone in the posterior muscles of
26
Table 2 Validated scores (points) of psychomotor assess-
ment.22,23
Evaluation Normal Moderate Major
deviation deviation
Behavioral during 33 32– 22 #21
the examination
Visual/auditory 9 8– 6 #5
Sensory-motor skill 46–44 43– 28 #27
Postural control 20 19– 10 #10
Orthopaedic 12 11– 8 #7
of lower limb
the trunk and neck, revealing an imbalance
between the extensor and flexor muscles.
Gestational age intra-group analysis
of the data
If we consider the age groups, we observe that some
items were significantly different in the postured
(treated) and non-postured (control) infants.
(a) In the youngest infants (31 – 33 weeks gesta-
tional age designated as Group A), exploration of
the passive tone for a given gestational age
revealed exaggerated extension versus flexion of
the trunk in the control group (63.6%) compared to
treated group (no exaggeration) ðx2ð1; N ¼ 21Þ ¼
9:5; p , 0:03Þ: In addition, the incidence of induced
opisthotonos was significantly lower in the treated
group (27.3%) than in the control group (not
induced) ðx2ð1; N ¼ 21Þ ¼ 11:7; p , 0:03Þ: The con-
trol group had a high headscarf score (i.e. limited
extensibility of shoulder muscles for gestational
age) which differed significantly from the score in
the treated group (Mann – Whitney’s U-test ð19; N ¼
21Þ ¼ 23:5; p , 0:04Þ: The scores for the heel/ear
angle and the popliteal angle were lower (corre-
sponding to narrower angles) in the control group,
and significantly different from those in the treated
group (Control: Mann – Whitney’s U-test ð19; N ¼
21Þ ¼ 22:5; p , 0:01Þ; Treated: Mann – Whitney’s U-
test ð19; N ¼ 21Þ ¼ 31:5; p ¼ 0:05Þ: The data for
passive tone at 31 –33 weeks gestational age showed
that the expected scores are in favor of the treated
group. Active tone (see Section 2) was better in the
treated group (normal in 100% of the infants) than
in the control group (normal in only 36.3% of the
infants). This difference was significant ðx2ð19; N ¼
21Þ ¼ 9:5; p , 0:04Þ:
(b) In the case of active tone, the treated group
differed from the control group by having a
significantly higher score for the extensors of the
neck (Mann – Whitney’s U-test ð19; N ¼ 21Þ ¼
28; p ¼ 0:04Þ:
L. Vaivre-Douret et al.
(c) In the group of the oldest infants (34 –36
weeks gestational age designated as Group B), the
same conclusions were drawn regarding exagger-
ated extension versus flexion of the trunk in the
control group for induced opisthotonos ðx2ð1; N ¼
39Þ ¼ 9:2; p , 0:01Þ resulting in a lower score for
the extensors of the neck (Mann – Whitney’s U-test
ð37; N ¼ 39Þ ¼ 109; p , 0:02Þ: The findings for active
tone were significantly better in the treated group
(normal in 95% of the children) than in the control
group (in which it was normal in only 10.5% of the
children). The maturity score was also significantly
better in the treated group, who had a mean score
of 70 points; than those in the control group, who
scored 68 points (Mann – Whitney’s U-test ð37; N ¼
39Þ ¼ 110; p ¼ 0:03Þ: However, all scores remained
within the normal range for the gestational age
(Appendix A).
Concluding on the neurological assessment, we
report that in this population as a whole, there were
moderate muscular abnormalities such as exagger-
ated straightening and predominance of the exten-
sors of the neck in the control group. The treated
group displayed a balance between the flexors and
extensor muscles of the axis in the assessment
carried out when the infant was discharged from
the unit.
Psychomotor assessment
We found that scores on the protocol showed
varying degrees of impairment, and that some
abnormalities were more marked than others in
the control group (see Table 4). In regards to
behavior, we found mild overexcitability in the
control group: 57 versus 17% in the treated group,
with infants in the control group more difficult to
calm (between 5 and 60 s corresponding to a
moderate to high consolibility level).
The visual and auditory function data for the
control group exhibited a mild decrease in response
to visual and auditory stimulation (40% of the
infants as opposed to 7% of the treated group
Student’s t-test ð60Þ ¼ 3:26; p , 0:01Þ: There was
no significant difference between the abilities of
the infants in the two groups when following a
moving and contrasted object with their eyes,
although this was more difficult in the control
group as evidenced by a lower score compared to
the treated group.
Sensory-motor skills were severely affected in
the control group. These infants exhibited excessive
muscular extension of the posterior axis of the body
(see Table 3) which restricted centripetal forward
coordination movements, such as rolling over and
sitting up using lateral support from the arm (Fig. 2).
Effect of positioning on the incidence of abnormalities of muscle tone in low-risk, preterm infants 27

Table 3 Comparison between control and treated groups for the passive tone data for the main axis of the body.

Treatment group ðn ¼ 30Þ n (%) Control group ðn ¼ 30Þ n (%) x2 p

Trunk extension/flexion
Normal (extension ¼ flexion) 29 (97%) 14 (47%) 18.5 0.01
Exaggerated (extension ¼ flexion) 1 (3%) 16 (53%)
Opisthotonos (extension of head
and trunk)
Absent 29 (97%) 13 (43%) 20.3 0.01
Induced 1 (3%) 17 (57%)
Extension of the forearm
Complete 30 (100%) 24 (80%) 6.67 0.01
Incomplete 0 6 (20%)

In addition, in the ventral suspension test, the whole
body straightened excessively and movement was
not continuous but dissociated during active – pas-
sive turning over (60 versus 7% in the infants in the
treated group) (Fig. 3).
The findings of the postural examination were
correlated with those of the neuro-psychomotor
examination (0.80) and showed that when placed on
the side position, the control group did not maintain
balance of their axis with problems affecting
posterior muscles of the body axis (Fig. 4). All test
items were significantly different in the treatment
and control groups ðp , 0:01Þ: The abnormal pos-
ition in which arms are continuously in flexion and
abduction was also seen in the control group with a
50% incidence of a tendency towards this posture
and a 40% incidence of the attitude of this posture
(Fig. 5) compared to the treated group. There was
also a tendency to keep the head turned preferen-
tially to one side with a tendance towards plagio-
cephaly (asymmetrical flattening of the head) in
30% of the control group children vs. 10% of the
treated group ðx2ð1; N ¼ 60Þ ¼ 3:75; p , 0:06Þ:
The orthopaedic lower limbs examination
showed that, in general, the infants with a greater
gestational age (group B), had a greater tendency
towards foot deviations than group A (42.9 vs.
38.4%). As such, the assessment upon discharge

Table 4 Abnormalities (%) with ORs from the psychomotor assessment protocol of the whole population.

Population ðN ¼ 60Þ

Treatment ðn ¼ 30Þ n (%) Control ðn ¼ 30Þ n (%) OR (95% CL) x2 p

Assessment of behavior
Normal 21 (70%) 11 (37%) 0.25 (0.085–0.73) 6.7 0.01
Moderate deviation 9 (30%) 19 (63%)

Visual and auditory function

Normal 28 (93%) 18 (60%) 0.11 (0.02–0.55) 9.3 0.01
Moderate deviation 2 (7%) 12 (40%)
Sensory-motor skills
Normal 26 (87%) 3 (10%) 0.017 (0.003–0.084) 35.7 0.01
Moderate deviation 4 (13%) 22 (73%)
Major deviation 0 5 (17%)

Postural control
Normal 27 (90%) 1 (3%) 0.0038 (0.00037–0.039) 45.5 0.01
Moderate deviation 3 (10%) 23 (77%)
Major deviation 0 6 (20%)
Orthopaedic lower limbs
Normal 25 (83% 11 (37%) 0.12 (0.04–0.4) 13.6 0.01
Moderate deviation 5 (17%) 19 (63%)

CL indicates confidence interval.

 28
L. Vaivre-Douret et al.
Figure 2 Score based on sensory-motor skills items between treatment group and control group (Psychomotor assessment).
 Effect of positioning on the incidence of abnormalities of muscle tone in low-risk, preterm infants
Figure 3 Score based on postural control items between treatment group and control group (Psychomotor assessment).

29
30
Figure 4 Premature newborn (control group) aged 38
weeks, placed lying on side: unable to stay in this position
because the head falls back and arms are in flexion and
abduction posture.
showed that children in control group B had more
severe lateral deviations of the feet; which we
believe was attributable to the maintenance of an
unchanging position and to the pull of gravity in the
prone position during the period in the incubator.
By considering these age groups, we can see that
the same items were affected in both the control
and treated groups.
As can be seen from Table 4, at the time of the
conclusion of the study there were more abnorm-
alities in the control group than in the treated
group. The biggest difference between these two
groups was postural assessment, with a consider-
able incidence (20%) of major deviations. The next
most severely affected assessment was that of
sensory-motor skills (17% of major abnormalities in
the control group). The behavioral and orthopaedic
assessments were moderately affected (63% mod-
erate deviations, particularly involving metatarsus
valgus in the control group versus 17% in the treated
group). The visual and auditory function was
L. Vaivre-Douret et al.
Figure 5 Examination of sitting-position of a premature
newborn aged 38 weeks (control group): Arms are in
flexion and abduction postural attitude.
the least affected, with a 40% incidence of
moderate deviations in the control group versus
8% in the treated group.
We also calculated the OR for the various general
conclusions based on the psychomotor protocol
(Table 4). We found that in all cases the OR was
significant and less than 1. As such, we concluded
that our postural change program was beneficial in
preventing secondary deformities.
Discussion
When the infants left the neonatal unit at a
gestational age close to full-term all had a posture
of a neonate, with the pelvis raised and their knees
naturally bent forwards while lying on their side. Our
conclusions are similar to those of Downs et al.10 who
showed that the results of the neurological and
postural examinations carried out at full term were
better in the treated children simply due to hip
posturing. It can be concluded that the bolster
recommended initially by Grenier.8 to raise the
pelvis in the prone position provides preventative
hip support. However, as demonstrated by various
Effect of positioning on the incidence of abnormalities of muscle tone in low-risk, preterm infants
neurological and psychomotor assessments, this
intervention is not enough to prevent abnormalities
affecting other parts of the body and in fact, may
even significantly accentuate these anomalies due
to the shifting of weight to the top of the body if
children are placed exclusively on their stomachs.
This can encourage an hyperextension of the neck
and lateral rotation of the head in addition to
positioning of the arms in flexion and abduction
when bearing weight. These findings confirm our
earlier studies about the effects on motor function
of lying infants on their stomachs15,16 as well as the
suggestions advanced by Updike et al.2 As the
muscular system of the preterm infant matures,
they become increasingly active in opposing gravity,
for example by supporting themselves by their arms
and attempting to lift their head (similar to ‘push-
ups’). The prone position encourages this strength-
ening as the extensors of the trunk are more
developed by exercise with tightening of the
insertion points (clavicles and acromion). However,
this normal posture can lead to confusion between
hypertonicity of central and peripheral origins,
making it impossible to reach a diagnosis within
the first year of life.15
In regards to sensory-motor skill, the scores of
the treated group were significantly better than
those of the control group. Muscular and osteo-
articular deformities and acquired muscular short-
ening (of the muscles sterno-cleido-mastoid, the
shoulder girdle, the trapezoid muscle along with
abduction and external rotation muscles of the
shoulder and biceps of the arm) blocks active
movements, allowing ungainly and non-functional
movements to develop (e.g. extending the head
backwards while turning over and failure of the arm
to follow the motion).
In this population, we also considered the
orthopaedic consequences to posturing. We found
that infants placed on their stomachs had a higher
incidence of orthopaedic abnormalities of the feet
in the control group. Deformed postural position(s)
may have been present before birth,24 without
being noticed as an actual deformity, particularly in
the older group of preterm infants, or they may
have been acquired (particularly valgus deviation of
the feet), and then maintained by the weight of
gravity of the body when the child was lying in the
prone position. Katz et al.25 and Katz et al.1
observed orthopaedic deformities in preterm
infants and showed that they tended to display
lateral rotation of the legs, external tibial rotation,
and everted feet at birth.
Infants in our study whose position was changed
regularly had balanced assessments which suggest
that what is needed is not restraint of the body at
31
all times, but freedom of movement in order to
achieve normal tonic/postural balance as soon as
possible such as, molding the head (against plagi-
ocephaly), balance of the body axis, and free
movement of the limbs.
Our results also reveal more abnormalities in
infants at a lower gestational age, although this
difference was not significant. This can be explained
by the slowness of the physiological maturation
process along the caudal/cerebral axis in preterm
infants.19 The younger the child, the more it is in a
globally hypotonic (small weight for gestational
age) phase. Similarly, we found no relationship
between anomalies and hypotrophy in these infants.
By calculating the correlation between the
scores from the maturity, behavior, visual and
auditory, sensory-motor skills, orthopaedic, and
postural assessments based on the protocols
used,22,23 we were able to identify significant links
ðp , 0:01Þ between the behavior score and the
sensory-motor score ðr ¼ 0:68Þ; between the sen-
sory-motor score and the visual and auditory score
ðr ¼ 0:70Þ and between the sensory-motor score
and the postural score ðr ¼ 0:82Þ: On the basis of
this study we conclude that changing the sleeping
posture of preterm infants has a protective effect
as revealed by the OR. We think that this tendency
as well as the findings of the various assessments,
suggest that changes in posture may have a
beneficial effect on behavior that is evidenced by
a better response on our psychomotor tests, less
excitability, movement easy to elicit, etc.
Abnormal muscle tone interferes with behavior
control, overexcitability and more difficulties for
consolibility in the control group and then can
interfere with a mothers’ ability to properly
establish holding contact with their infant as the
infant is often trying to achieve dorsal hyperexten-
sion (Fig. 4). This is an overexcitability of the Moro
reflex (with cry) as the nape of the neck is
constantly stimulated. As a result of this stimu-
lation, the head, which contains the vestibular
system (the organ of balance), is constantly sway-
ing. In addition, the child’s arms are often elevated
or in a ‘hand’s up’ posture caused mainly by the
retraction of the trapezoid muscles. As such, the
infant adopts a stiff, rather immobile posture which
makes it higher difficult to hold.
Our results demonstrate that muscular short-
ening develops rapidly if one adopts a single laying
position (ex. prone) and that disrupts functional
motor organization making coordination between
limbs and gaze difficulty.
One possible bio-physiological explanation for
this is provided by the experimental work of Tabary
et al.26 in which they maintain that a muscle adapts
32
its number of sarcomers to the length imposed on
it. As such, immobility can cause a muscle to loose
up to 40% of its sarcomers within 3 weeks. Even
though this phenomenon is reversible, it can be a
cause of concern until the normal development of
the child leads to an equilibrium established in
response to appropriate stimuli from the environ-
ment. Recent studies have shown the importance of
the first regular sleep position as neonates habitu-
ate very quickly during the first 3 months of life and
do not readily accept a change of sleeping position
imposed upon them as evidenced by duration of cry
and time until consolability.27,28
Our findings indicate that a properly supported
position is a posture that ensures functional
support of all the parts of the body as well as
ensuring physical safety. In all positions (back and
prone), the knees must be held forward and lower
than the pelvis and this has been amply demon-
strated by the work of Grenier.8 Apart from the
fact that placing an infant flat on its stomach pits
it against gravity and therefore increases tidal
respiratory volume, we do not think that this
position has any advantage from a developmental
care point of view.15,16 Measures must be taken to
support the head in line with the main axis of the
body and with the body slightly curled as soon as
the baby’s physiological flexion condition permits
this, when lying either on its side or on its back.
The shoulders and knees must also be held forward
when the baby is lying on its back. This position is
reminiscent of the fetal lateral position as this
soothes the infant in a centripetal arrangement of
the body and brings the shoulders rounded and
knees forward to facilitate the organization of the
initial proximal and distal movements (the hands
are drawn towards each other in midline and
towards the mouth). This helps to maintain the
balance of the main axis of the body and the initial
coordination around the axis, encouraging auton-
omy of the baby for its future developmental
sensory-motor acquisitions.15 Furthermore, the
infants’ sight both (in the lateral position) with
regard to oculomotor activity and the ability to
take an interest in the whole body, is constructive
in developing the child’s knowledge of its own
body (somatognosis)17,23,29 and consequent neur-
onal organization.
Based on our findings, it appears that physiologi-
cal postural prevention for preterm infants, includ-
ing low-risk infants, is effective. We demonstrated
that infants in the treated group had more positive
and clear assessments than those in the control
group, i.e. they did not display any further
abnormalities during their convalescence period.
This also demonstrates the usefulness of being able
L. Vaivre-Douret et al.
to carry out an objective evaluation when the child
is discharged from the neonatal unit. We note that
infants with the same initial neurological status,
(such as those in the control group) presented with
only isolated abnormalities, compared to those in
the treated group. Given the uniformity of the
group at the outset, we can confirm our hypothesis
that these abnormalities were acquired during
convalescence; and that they are therefore of
postural, peripheral origin. It is important to note
that some neurological test items, such as the
headscarf sign and opisthotonos, should be con-
sidered for their functional impact rather than just
considering their global score. The neurological
maturity score we established provides no diagnos-
tic information for mild disorders, since all infants
had the score expected for their age when
discharged. In contrast, the psychomotor assess-
ment22,23 was useful and revealed moderate devi-
ations from normal, which all too often remain
unnoticed.
The findings of this study, we believe, can
serve as a guide to sensitize neonatal practices to
the importance of early psychomotor develop-
ment in low-risk premature infants. The globally
functional posture of these infants is a means of
promoting correct psychomotor development for
the future and offers an important insight in
positioning high-risk, preterm populations. High-
risk infants spend even more time in imposed
positions than do their low-risk counterparts, and
this can be particularly prejudicial for their future
functional potential especially if they have
inherent neuro-physiological difficulties. Due to
gravitational force and the preterm infants
inability to resist (due to motor immaturity),
their functional lying position can create, main-
tain and/or facilitate durable motor difficulties.
This can be significantly ameliorated by correct
physiological positioning using support materials
dedicated to this end.
A limitation of this study concerned the bolsters
and mattress that were used to support physiologi-
cally correct lying positions. Due to frequent and
normal movement by these neonates, these sup-
porting elements were often moved in such a way as
to no longer maintain the physiologically correct
position desired. As such, constant surveillance was
required to rearrange both the bolster and the
mattress to ensure adherence to the protocol
demanded by this study. This limitation, however,
did encourage one of the authors (L.V-D) to invent a
new postural support mattress30,31 which main-
tains, at all times, the globally correct body
positions required.
Effect of positioning on the incidence of abnormalities of muscle tone in low-risk, preterm infants
Acknowledgements
This work was supported by a grant from the French
society for paralysed children and children with
multiple deformities (SESEPP). We would like to
thank Dr C. Amiel-Tison and Dr F. Lebrun for their
assistance in carrying out this study, as well as
INSERM U. 149 (G. Bréart) and Carla Dos Santos for
statistical consultation and recommendations.
Appendix A
Items for the maturity score of the Neurological
Assesment Protocol
1. Oculo-motor: follow the black and white cible
with eyes or eyes þ head;
2. Axial passive tone: (a) repeated ventral flexion of
the head with normally no change in the
resistance of antagonist muscles; (b) amplitude
of the movement of rotation of the head to each
side; (c) ventral flexion of the trunk; maneuver
compared with dorsal extension of the trunk
maneuver which should balance each other; (d)
absent or induced opisthotonos should not occur;
3. Passive tone of the arms: (a) return of the arms to
a flexed position; (b) the headscarf maneuver
which explores the extensibility of the shoulder
muscles (near the midline of body’s axis close to
term) ; (c) the angle of extensibility of the forearm
relative to the arm (approximately 175 –1808);
4. Passive tone of the legs: (a) return of the legs to a
flexed position; (b) extension of the leg; (c) the
heel/ear angle; (d) the popliteal angle; (e) the
adductors angle; (f) the angle of dorsiflexion of
the foot;
5. Active tone: (a) evaluation of neck flexors with
raise to sit maneuver and return backward to
evaluate neck extensors. Normally at term, flexor
and extensor tone is balanced resulting in a head
control aligned with the axis of the trunk during
about 3 s;
6. Primitive reflexes: (a) sucking/swallowing reflex;
(b) crossed stretching reflex; (c) grasping; (d)
response to traction by the arms, which is a grip
reaction of the child’s fingers which makes it
possible to raise the child, normally to a position
of global flexion of the body. (e) automatic
walking; (f) Moro reflex. Neurological maturity
score of normal neurological assessment(22, 23)
Gestational 6 7 7(1/2) 8 8(1/2) 9
age (months)
Maturity 8 25 40 56 67 74
score (points)
33
Appendix B
Items for the maturity score of the Psychomotor
Assesment Protocol
(a) The behavioral score (33 points) consisted of
the exploration of Prechtl and Benteima (1964)
of behavioral states (3 points) with lability of
skin color (Brazelton, 1984). That is, quality of
cry (3 points), consolability in state V (2
points); spontaneous motor activity of the
trunk and the limbs (12 points), isolated
movements of the extremities (12 points);
and abnormal movements (1 point).
(b) Visual/auditory function (9 points) included
the child’s receptive expression to visual and
auditory stimuli produced by the examiner (2
points); visual orientation by use of a black/
white target (5 points) and the child’s reaction
to auditory stimulation using an auditory sound
or hearing test (2 points).
(c) Sensory-motor skills (46 points), which evalu-
ate the child’s dynamic ability to roll over
(head, trunk, overpassing arm…) from lying on
its back to lying on its stomach, induced and
guided by the legs with examiner help (12
points) included the general performance of
this sequential continuous movement, dis-
sociate or in block (4 points); changing from
lying on its stomach to lying on its back with the
examiner guiding the leg or pushing the pelvis
by oblique movement (4 points); active lateral
straightening of the head and body using lateral
push from the arm and examiner support on the
other arm (10 points); global straightening on
one foot with support (8 points); ventral
suspension with observation of head, trunk
and limb position (8 points).
(d) A postural control (20 points) to evaluate the
limbs position and the ability to maintain the
following positions: Supine (7 points); Prone (4
points); Right side (1 point); Left side (1 point);
possible abnormalities of the body axis in these
different positions i.e. throwing the head or
the head and the trunk backward like opistho-
tonos (5 points); relaxation of the arms hanging
next to the trunk when in a maintained
semi-seated position (2 points) rather than an
attitude of arms flexion and abduction or
tendency of this posture.
(e) Orthopaedic lower limbs position (12 points)
consisting of checking abnormal position of the
hips with abduction and external rotation of
the hips (1 point); observation of foot deviation
(10 points) and when appropriate, any other
abnormality (1 point).
34
References
1. Katz K, Krikler R, Wielunsky E, Merlob P. Effects of neonatal
posture on later lower limb rotation and gait in preterm
infants. J Pediatr Orthoped 1991;11(4):520—2.
2. Updike C, Rosemary E, Schmidt RE, et al. Positional support
for preterm infants. Am J Occup Ther 1986;40(10):712—5.
3. Kurlak LO, Ruggins NR, Stephenson TJ. Effect of nursing
position on incidence, type, and duration of clinically
significant apnoea in preterm infants. Arch Dis Childhood-
Fetal Neonatal Ed 1994;71(1):16—19.
4. American Academy of Pediatrics, Changing concepts of
sudden infant death syndrome: implications for infant
sleeping environment and sleep position. Pediatrics 2000;
105(3):650—6.
5. Carroll JL, Siska ES. SIDS: Counseling parents to reduce the
risk. Am Fam Physician 1998;57(7):1566—71.
6. Dehan M, Briand E. En trois ans, le nombre de morts subites
du nourrisson a diminue de pres de 70% [In three years, the
number of SIDS cases have decreased appoximately by 70%].
Rev du Praticien 1998;12(408):13—15.
7. Oyen N, Markestad T, Skjaerven R. Combined effects of
sleeping position and prenatal risk factors in sudden infant
death syndrome: the nordic epidemiological SIDS study.
Pediatrics 1997;100(4):613—21.
8. Grenier A. Prevention des deformations precoces de hanche
chez les nouveau-nes a cerveau lese. [Prevention Of Early
Hip Deformations In Newborns With Cerebral Lesions]. Ann
de Pediatr 1988;35(6):423—7.
9. Monterosso L, Coenen A, Percival P, Evans S. Effect of
postural support nappy on flattened posture of the lower
extremeties in very preterm infants. J Paediatr Child Health
1995;31(4):350—4.
10. Downs JA, Edwards AD, McCormick DC, Roth SC, Stewart AL.
Effect of intervention on development of hip posture in very
preterm babies. Arch Dis Child 1991;66(7):797—801.
11. Georgieff MK, Bernbaum JC, Hoffman-Williamson M, Daft A.
Abnormal truncal muscle tone as a useful early marker for
developmental delay in low birth weight infants. Pediatrics
1986;77(5):659—63.
12. Georgieff MK, Bernbaum JC. Abnormal shoulder girdle
muscle tone in premature infants during their first 18 months
of life. Pediatrics 1986;77(5):664—9.
13. Hunter JG. The neonatal intensive care unit. In: Allen A,
Pratt P, editors. Occupational therapy for children. St Louis,
MO: Mosby; 1996. p. 583—631.
14. Montford K, Case-Smith J. The effects of a neonatal
positioner on scapular retraction. Am J Occup Ther 1997;
51:378—84.
15. Douret L. Deleterious effects of the prone position in the
full-term infant throughout the first year of life. Child Care
Health Dev 1993;19:167—84.
16. Vaivre-Douret L. Motor behaviour change in the prone
position of full-term infants during the first year of life:
effects of early intervention. Infant-Toddler Intervent 1993;
3(4):255—63.
L. Vaivre-Douret et al.
17. Vaivre-Douret L. Confort et prevention postural du nou-
veau-ne: Mise au point d’un matelas Bondosr [Postural
comfort and prevention for the newborn: analysis of the
Bondos mattress]. Rev Eur de Technol Biomed 1996;18(3):
84—8.
18. André-Thomas Saint-Anne Dargassies S. Etudes neurologi-
ques sur le nouveau-né et le jeune nourrisson [Neurological
studies on the newborn and infant]. Paris: Masson; 1952.
19. Amiel-Tison C, Grenier A. Neurological assessment during
the first year of life. New York: Oxford University; 1986.
20. Prechtl H, Beintema D. The neurological examination of the
full term newborn infant. Clin Dev Med 1964;12:74.
21. Brazelton TB. Neonatal behavioral assessment scale, 2nd Ed.
Clinics in developmental medicine, vol. 88. London: Spastics
International, Medical Publications; 1984.
22. Vaivre-Douret L. Intervention precoce en psychomotricite
aupres d’enfants hospitalises en unites de soins intensifs et
en neonatalite. [Early intervention and psychomotor
behavior with infants in neonatal and intensive care
units]. Neuropsychiatr Enfance et Adolesc 1997;45(4—5):
190—208.
23. Vaivre-Douret L. Precis Theorique et Pratique du Devel-
oppement Moteur du Jeune Enfant. [Motor development in
the young infant—theory and practice]. Paris: Elsevier; 1997.
24. Amiel-Tison C, Atlani E. Les deformations foetales liees aux
contraintes in utero.[Fetal deformations due to in utero
constraints]. La Rev du Praticien 1988;13:805—12.
25. Katz K, Naor N, Merlob P, Wielunsky E. Rotational deform-
ities of the tibia and foot in preterm infants. J Pediatr
Orthoped 1990;10(4):483—5.
26. Tabary JC, Tabary C, Tardieu C, Tardieu G, Goldspink G.
Physiological and structural changes in the cat’s soleus
muscle due to immobilization at different lengths by plaster
casts. J Phys 1972;224:231—44.
27. Vaivre-Douret L, Dos Santos C, Richard A, et al. Comporte-
ments des meres face a la position de couchage de leur bebe:
Effets de la derniere campagne de prevention concernant la
mort subite du nourrisson. [Behavior of mothers’ in
reference to laying position: The effects of the latest SIDS
campaign]. Arch de Pediatrie 2000;7:1293—9.
28. Vaivre-Douret L, Dos Santos C, Charlemaine C, Cabrol D.
Effects of sleeping and waking position on infants’ motor
development. Eur Rev Appl Psychol 2003;53(4). in press.
29. Vaivre-Douret L, Burnod Y. Motor development of young
children (0—4 years). Contribution of a functionnal posturo-
motor, locomotor and eye-hand grip rating scale. Child
Care, Health Dev 2001;27(6):515—34.
30. Vaivre-Douret L. Dispositifs de maintien de posture pour
bébés. [Device of postural bearing for baby]. Invention
Patent no 9801028, Official Bulletin of Industrial Property no
277427; 1998.
31. Vaivre-Douret L. Du cocon au coconou ou de l’artisanat a
l’ere industrielle. [From the cocoon to coconou, art in
industry]. Actes des Cinquiemes Journees Franciliennes de
Pediatrie 1999;24—9.