DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY REVIEW

Early sucking and swallowing problems as predictors of

neurodevelopmental outcome in children with neonatal brain injury:
a systematic review
JUSTINE SLATTERY 1 | ANGELA MORGAN 2 | JACINTA DOUGLAS 1

1 School of Human Communication Sciences, La Trobe University, Melbourne, Victoria; 2 Murdoch Childrens Research Institute, Parkville, Victoria, Australia.
Correspondence to Justine Slattery, School of Human Communication Sciences, La Trobe University, Melbourne Campus, Bundoora, 3086 Victoria, Australia. E-mail: j.slattery@latrobe.edu.au

PUBLICATION DATA AIM Early sucking and swallowing problems may be potential markers of neonatal brain injury
Accepted for publication 13th March 2012. and assist in identifying those infants at increased risk of adverse outcomes, but the relation
Published online 19th May 2012. between early sucking and swallowing problems and neonatal brain injury has not been estab-
lished. The aim of the review was, therefore, to investigate the relation between early measures of
ABBREVIATIONS sucking and swallowing and neurodevelopmental outcomes in infants diagnosed with neonatal
HIE Hypoxic–ischaemic encephalopathy brain injury and in infants born very preterm (<32wks) with very low birthweight (<1500g), at risk
IVH Intraventricular haemorrhage of neonatal brain injury.
NHMRC National Health and Medical METHOD We conducted a systematic review of English-language articles using CINAHL, EMBASE,
Research Council and MEDLINE OVID (from 1980 to May 2011). Additional studies were identified through manual
NOMAS Neonatal Oral–Motor Assessment
searches of key journals and the works of expert authors. Extraction of data informed an assessment
Scale
of the level of evidence and risk of bias for each study using a predefined set of quality indicators.
RESULTS A total of 394 abstracts were generated by the search but only nine studies met the inclu-
sion criterion. Early sucking and swallowing problems were present in a consistent proportion of
infants and were predictive of neurodevelopmental outcome in infancy in five of the six studies
reviewed.
LIMITATIONS The methodological quality of studies was variable in terms of research design,
level of evidence (National Health and Medical Research Council levels II, III, and IV), populations
studied, assessments used and the nature and timing of neurodevelopmental follow-up.
CONCLUSIONS Based upon the results of this review, there is currently insufficient evidence to
clearly determine the relation between early sucking and swallowing problems and neonatal brain
injury. Although early sucking and swallowing problems may be related to later neurodevelop-
mental outcomes, further research is required to delineate their value in predicting later motor out-
comes and to establish reliable measures of early sucking and swallowing function.

Brain injury acquired during the perinatal or prenatal period
can manifest in numerous adverse neurodevelopmental out-
comes including motor, sensory, language, and cognitive diffi-
culties.1,2 More specifically, early brain injury is reported to
have a deleterious impact on oral motor functions, including
sucking and swallowing.3 Assessment of the motor system in
early infancy can provide an indication of central nervous
system function and the need to facilitate early intervention
following neonatal brain injury.4,5 Although responsive mea-
sures of sucking and swallowing function in this population
may assist in identifying those infants at increased risk of
adverse outcomes, there is currently little empirical evidence
regarding the relation between early sucking and swallowing
problems and neonatal brain injury.6
Despite advances in medical science, neonatal brain injury
remains common.7–9 As many as 50% or more of preterm
infants born with very low birthweight will be affected by peri-
ventricular leukomalacia (PVL),9 while neonatal and perinatal
strokes reportedly affect 28.6 per 100 000 infants born after
31 weeks.7,10 Although many infants will survive neonatal
brain injury, early and accurate identification of adverse neuro-
developmental outcomes remains a major challenge.11 Neuro-
logical symptoms may not manifest in the neonatal
period,12,13 and routine cranial ultrasound may not be sensitive
enough to determine the nature of early brain injury (in partic-
ular cerebral white matter lesions).14,15 Magnetic resonance
imaging (MRI) provides detailed, objective diagnostic infor-
mation regarding early brain injury,16 but it is costly and not
always accessible.17 Valid and reliable clinical assessment of an
infant’s motor system, including sucking and swallowing func-
tions may provide an inexpensive and widely available measure
of early neurological function. In addition, sensitive measures
of early sucking and swallowing function may have the poten-
tial to assist in identifying those infants most likely to experi-
ence long-term, adverse developmental outcomes as a
consequence of neonatal brain injury.

796 DOI: 10.1111/j.1469-8749.2012.04318.x ª The Authors. Developmental Medicine & Child Neurology ª 2012 Mac Keith Press
 Sucking and swallowing form critical components of the
infant’s motor repertoire and are essential for successful feed-
ing, growth, and development.18 Sucking and swallowing
abnormalities in early infancy have long been viewed as poten-
tial markers of neonatal brain injury.17,19,20 Moreover, it has
been suggested that prolonged dysphagia or swallowing diffi-
culties in very preterm infants may represent an early marker
of undiagnosed brain injury.20 Successful infant feeding
depends upon the coordination of sucking, swallowing, and
breathing.21 The neurological controls that support this motor
activity are complex and rely upon the integration of cortical,
subcortical, brainstem, and cerebellar inputs.20,22,23 It is gener-
ally agreed that neonatal brain injury can damage these neural
pathways, with resultant sucking and swallowing prob-
lems.19,20,23,24 However, there is currently limited evidence
regarding this concurrent relation, in particular the frequency
of occurrence and the specific patterns of early sucking and
swallowing abnormalities associated with different types of
neonatal brain injury.
In addition, it is argued that a relation exists between early
sucking and swallowing difficulties and later neurodevelop-
ment.19,25 In a recent review, Poore and Barlow19 contend that
early sucking skills may predict later neurodevelopmental out-
comes, particularly in speech and language. They describe the
potential links between the neural networks that control suck-
ing, swallowing, and speech and call for further research to
identify early predictors and treatment techniques to improve
sucking and speech development. However, the predictive
relation between early sucking and swallowing difficulties and
later neurodevelopment remains contentious. da Costa et al.6
argue that there is currently insufficient empirical data to sup-
port the claim that early sucking and swallowing measures can
predict subsequent developmental outcome.
It is important that diagnostic and intervention decisions for
high-risk infants are informed by empirical evidence. There-
fore, the objective of this systematic review was to describe the
relation between early measures of sucking and swallowing
and neurodevelopmental outcomes in neonates diagnosed with
brain injury and infants born very preterm (<32wks) with very
low birthweight (<1500g), at risk of neonatal brain injury. The
review focused on two aims: (1) to describe the concurrent
relation (i.e. the frequency of occurrence and specific features)
of early sucking and swallowing outcomes associated with neo-
natal brain injury; and (2) to evaluate the predictive relation
between early sucking and swallowing measures and later neu-
rodevelopmental outcomes in this population.
METHOD
Search strategy
A systematic search was conducted of relevant computerized
databases: CINAHL (1980–May 2011), EMBASE (1980–May
2011), and MEDLINE OVID (1980–May 2011). The search
strategy included the following Medical Subject Heading
(MeSH) terms and text words for (‘brain injury’ OR ‘hypoxic
ischaemic’ ‘OR ‘stroke’ OR ‘periventricular leukomalacia’ OR
‘encephalomalacia’ OR ‘anoxia’ OR ‘cranial haemorrhage’ OR
‘ischaemia’ OR ‘cerebrovascular disorders’ OR ‘leukoence-
What this paper adds
• This is the first systematic review examining the concurrent relation of early
sucking and swallowing outcomes associated with neonatal brain injury.
• It presents a critical appraisal of research examining the predictive relation
between early sucking and swallowing problems and later neurodevelopmental
outcomes.
• The paper presents a systematically generated list of assessment tools avail-
able for measuring early sucking and swallowing behaviour in children with
neonatal brain injury.
phalopathies’) AND (‘neurological examination’ OR ‘treat-
ment outcome’ OR ‘prognosis’ OR ‘predictive values’ OR ‘risk
factors’ OR ‘psychomotor performance’ OR ‘neurodevelop-
ment’) AND (‘feeding behaviours’ OR ‘sucking behaviours’
OR ‘swallowing behaviours’ OR ‘deglutition disorders’ OR
‘dysphagia’ OR ‘feeding and eating disorders’). The search
was further limited to neonates, including those born preterm,
and infants between the ages of 1 and 23 months. Additional
manual searches of key journals and the works of expert
authors were conducted.
Inclusion criteria
Studies were eligible for inclusion if they included infants
diagnosed with peri-, pre-, or early postnatal brain injury
and ⁄ or infants born very preterm (<32wks) with very low
birthweight (<1500g), at risk of neonatal brain injury; reported
primarily on suck ⁄ swallowing behaviours in the neonatal per-
iod; reported on neurodevelopmental outcomes; provided
empirical data; and were published in English. Studies not
meeting at least one of these criteria were excluded.
Data extraction
The database search yielded 389 abstracts, and a further five
were identified by manual search, resulting in a total yield of
394 papers. The three authors independently read each of the
titles and abstracts and assessed them in relation to the inclu-
sion criteria. Based upon consensus rating, 12 papers were
identified for more detailed review. Of the 12 papers that met
the preliminary inclusion criteria, a further three were
excluded because they did not report on early sucking and
swallowing outcomes or include empirical data. The final nine
studies that met the inclusion criteria were then read and
appraised by the first two authors to assess the level of evi-
dence and methodological quality. Any disagreement between
the reviewers’ ratings of these studies was discussed until con-
sensus was reached.
First, the included studies were assessed to establish the
level of evidence using the National Health and Medical
Research Council (NHMRC) evidence hierarchy.26 Next,
each of the studies was assessed for methodological quality.
The main aims of this review necessitated an assessment of
methodological quality that could be applied to varied aetiol-
ogy and prognostic research. The Critical Appraisal Skills
Program (CASP)27,28 and the Methodological Evaluation of
Observational Research (MORE)29 were used to generate the
quality indicators in the critical appraisal. Each study was
assessed for risk of bias in relation to the following: sampling
method and frame; accuracy of eligibility criteria; source and
Review 797
 Table I: Appraisal of studies reporting early sucking and swallowing problems associated with neonatal brain injury

Critical appraisal criteria

26
NHMRC level of Measures
evidence Sampling Eligibility
f
Study a
(research design) method (frame)b criteriac Source Reliabilityg Blindingd Follow-upe

Mizuno and Ueda30 II (prospective Convenience Yes Objectively measured Interrater Yes Yes
cohort) (health care based) with diagnostic
methods for the
purpose of study
Bier et al.31 II (prospective Convenience Yes Objectively measured Interrater Yes Yes
cohort) (health care based) with diagnostic
methods for the
purpose of study
Barkat-Masih et al.34 III-2 (retrospective Convenience Yes Obtained from medical records NR No Yes
cohort) (medical records)
36
Quattrocchi et al. IV (case series) Convenience Yes Obtained from medical records NR No Yes
(health care based)
Braun and IV (case series) Convenience Yes Objectively measured NR No Yes
Meyer Palmer37 (NR) with diagnostic
methods for the
purpose of study

a
The studies are ordered according to level of evidence and methodological quality. bSampling method adopted by investigators. cWere the
inclusion criteria, i.e. diagnosis of neonatal brain injury, described? dWere the early sucking and swallowing assessments administered without
knowledge of infant’s diagnosis? eWere withdrawals from the study were adequately explained? fSource of early sucking and swallowing
assessment measure. gReliability of the early sucking and swallowing assessment measure. NHMRC, National Health and Medical Research
Council; NR, not reported.

reliability of assessment measures; blinding of assessors admin-
istering measures; and adequacy of participant follow-up.
The appraisal was further organized in line with the main
aims of this review. Studies that addressed the concurrent rela-
tion between early sucking and swallowing problems and dif-
ferent types of neonatal brain injury were evaluated using the
specific criteria described in Table I. Studies that explored the
predictive relation between early sucking and swallowing
problems and later neurodevelopmental outcomes for infants
were evaluated using the specific criteria described in Table II.
Finally, data extraction focused on describing the partici-
pants, early sucking and swallowing assessments, neonatal
brain injury diagnoses, neonatal comorbidities, and neuro-
developmental outcomes in each study.
RESULTS
Nine studies that examined the relation between early sucking
and swallowing problems and various aspects of neonatal brain
injury were included in the review. All nine studies adopted
observational research methodologies. Five were prospective
cohort studies,25,30–33 two were retrospective cohort stud-
ies,34,35 and the remaining two studies were case series.36,37
These research designs were consistent with level II, III-3, and
IV evidence respectively (NHMRC26). All of the studies had
small sample sizes ranging from 1137 to 84 participants.34
Five studies reported on the concurrent relation of early
sucking and swallowing problems with different types of neo-
natal brain injury. Six studies explored the predictive relation
between early sucking and swallowing problems and later
neurodevelopmental outcomes. The included studies are
summarized in Tables III and IV.
Early sucking and swallowing problems associated with
neonatal brain injury (concurrent marker)
There were five studies germane to the first aim exploring the
concurrent relation between early sucking and swallowing
problems, (i.e. frequency and specific features) and neonatal
brain injury. Two of these studies used a prospective cohort
design,30,31 (level II; Table I), one used a retrospective cohort
design34 (level III; Table I), and the remaining two studies
were case series36,37 (level IV; Table I).
Early sucking and swallowing problems occurred in a
relatively consistent proportion of infants with neonatal
brain injury. Four of the five studies reported data on the
proportion of infants in their cohorts that had sucking or
swallowing problems during the neonatal period.30,31,34,36
Barkat-Masih et al.34 observed that 48.8% of the 84 neo-
nates with arterial ischaemic stroke included in their study
had sucking or swallowing problems. Almost two-thirds of
the sample presented with sucking and swallowing difficul-
ties during the neonatal period, with the remaining one-
third experiencing problems within the first 3 months of
life. Quattrocchi et al.36 reported that 42% of their sample
of 43 infants with neonatal hypoxic–ischaemic encephalopa-
thy (HIE) had sucking–swallowing alterations and aspiration
episodes, consistent with a diagnosis of oral motor dysfunc-
tion confirmed by imaging. Using instrumental measures,
Mizuno and Ueda30 documented moderate to severe suck-
ing difficulties in 35% of the neonates they studied with
mixed diagnoses of brain injury. Braun and Meyer Palmer37
reported that 45% of their smaller sample of 11 infants
presented with oral motor disorganization or dysfunction.
Their assessment was based upon observation of sucking

798 Developmental Medicine & Child Neurology 2012, 54: 796–806

 Table II: Appraisal of studies investigating the relation of early sucking and swallowing problems to later neurodevelopmental outcomes

Critical appraisal criteria

Measures
NHMRC26 level of Sampling Early suck ⁄ swallowing Neurodevelopment
evidence methodb Eligibility
a f g h
Study (research design) (frame) criteriac Source Reliability Source Reliabilityi Blindingd Follow-upe

Mizuno and Ueda30 II (prospective Convenience Yes Objectively measured with Interrater Objectively measured with Acceptable according No Yes
cohort) (health care diagnostic methods for the diagnostic methods for the to previous
based) purpose of study purpose of study published analysis
Tsai et al.25 II (prospective Convenience Yes Objectively measured with NR Objectively measured with Acceptable according No Yes
cohort) (health care diagnostic methods for the diagnostic methods for the to previous
based) purpose of study purpose of study published analysis
Medoff-Cooper II (prospective Convenience Yes Objectively measured with NR Objectively measured with Acceptable according No Yes
and Gennaro32 cohort) (subset of diagnostic methods for the diagnostic methods for the to previous
another purpose of study purpose of study published analysis
study)
Murray et al.33 II (prospective Convenience Yes Obtained from medical NR Objectively measured with Acceptable according No Yes
cohort) (health care based) records diagnostic methods for the to previous
purpose of study published analysis
Meyer Palmer III-3 (retrospective Convenience Yes Obtained from medical NR Obtained from medical NR No Yes
and Heyman35 cohort) (medical records records
records)
Barkat-Masih et al.34 III-3 (retrospective Convenience Yes Obtained from medical NR Obtained from medical NR No Yes
cohort) (medical records records
records)

a
The studies are ordered according to level of evidence and methodological quality. bSampling method adopted by investigators. cWere the inclusion criteria, i.e. diagnosis of neonatal brain injury,
and ⁄ or very low birthweight infants described? dWere the neurodevelopmental outcomes assessed without knowledge of infant’s early sucking and swallowing problems? eWere withdrawals from
the study explained? fSource of early sucking and swallowing assessment measure. gReliability of the early sucking and swallowing assessment measure. hSource of neurodevelopment assessment
measure. iReliability of neurodevelopmental assessment measures. NHMRC, National Health and Medical Research Council; NR, not reported.

Review 799
 Table III: Summary of studies reporting early sucking and swallowing problems associated with neonatal brain injury

Participants Early suck ⁄ swallowing outcomes

26
NHMRC level of Gestational age
evidence (wks) mean Diagnosis neonatal Age (wks) (mean
Study (research design) n (range) brain injury and ⁄ or range) Assessment Main findings

Mizuno and Ueda30 II (prospective 70 37.8 Severe IVH, intraparenchymal 1 and 2 post Instrumental 35% infants had moderate to severe
cohort) haemorrhages, moderate to commencement sucking and swallowing problems
severe hydrocephalus (45%) of oral feeds during neonatal period

800 Developmental Medicine & Child Neurology 2012, 54: 796–806

Mild hydrocephalus,
PVH without HIE (55%)
Bier et al.31 II (prospective 61: group 1 Group 1=26; IVH grade III ⁄ IV: group 1=13%; 38–40 NOMAS NOMAS scores lower at term for croup
cohort) (LBW intubated >1wk)=15; group 2=28; group 2=1%; group 3=0% 1 (p<0.001)
group 2 (LBW intubated group 3=40 Proportion of abnormal or suspect
<1wk)=36; group 3 neurological examination at term
(control)=10 higher in group 1 (p<0.05)
Length of oxygen dependence greatest
predictor of feeding outcome
Barkat-Masih et al.34 III-2 (retrospective 84 36 or greater Neonatal arterial NR Chart review 48.8% feeding disorders (i.e. sucking
cohort) ischaemic stroke, and ⁄ or swallowing difficulties)
unilateral (78.6%) and bilateral 15 had aspiration ⁄ increased risk
(21.4%) of aspiration
Perinatal respiratory problems
significantly correlated with feeding
problems (p=0.004)
Braun and Meyer IV (case series) 11 32 (27–40) Positive neurological signs, 39.5 (35–46) NOMAS Five with oral motor disorganizationa
Palmer37 10; IVH, 6 Instrumental or dysfunction;b 100% (n=5) with oral
motor problem also had brain injury
Two infants with brain injury had
normal NOMAS examination
Quattrocchi et al.36 IV (case series) 43 (29 term ⁄ 37 Pre-peri or post natal HIE NR Imaging 42% oral motor dysfunction.
14 preterm) diagnosed on MRI mean age 77.8% infants with oral motor
5 mths (54% supratentorial; dysfunction had infratentorial
46% infratentorial) lesions on MRI (p<0.001)

a
Disorganization of sucking refers to lack of rhythm of total sucking movement. bDysfunction refers to interruption of the feeding process by abnormal tongue and jaw movements.
IVH, intraventricular haemorrhage; PVH, periventricular haemorrhage; HIE, hypoxic–ischaemic encephalopathy; VLBW, very low birthweight (<1500g); NOMAS, Neonatal Oral–Motor Assessment
Scale;
NR, not reported; MRI, magnetic resonance imaging.
 Table IV: Summary of studies investigating the relation of early sucking and swallowing problems to later neurodevelopmental outcomes

Early suck ⁄ swallow Neurodevelopment

Participants outcomes follow–up

NHMRC26 level of Gestational age

evidence (research (wks), mean Neonatal
Study design) n (range) comorbidities Age (wks) Assessment Age (mo) Assessment Main findings

Mizuno and II (prospective 70 37.8 45% high risk 1 and 2wks post Instrumental 18 BSID All infants with severe feeding
Ueda30 cohort) (severe IVH, commencement of problems in neonatal period had
intraparenchymal oral feeds severe disability at 18mo on BSID
haemorrhages, Correlation between second
moderate to severe feeding assessment and
hydrocephalus) neurodevelopment outcome at
55% low risk (mild 18mo (p<0.001)
hydrocephalus, Feeding pattern and changes in
PVH without HIE) feeding pattern showed higher
sensitivity (91.7% and 95.8%) and
specificity than ultrasound (75%)
as screening test for adverse
neurodevelopmental outcome
at 18mo
Tsai et al.25 II (prospective 27 Group 1 32.5; VLBW and LBW 34 onward: group NOMAS 6 and 12 BSID At 6mo: risk of developmental
cohort) group 2 31.7 infants 1 – normal sucking delay lower in group 1 than in
59% RDS or CLD 37wks; group group 2 (p=0.05). Group 1 had
2 – disorganized higher scores on PDI (p=0.04).
sucking pattern at At 12mo rate of developmental
37wks delay between the two groups
was not significant
Medoff- II (prospective 19 29.1 VLBW infants 34 Instrumental 6 BSID Infants abnormal sucking
Cooper cohort) SD 1.9 measures at 34wks and BSID
and scores <1SD at 6mo. Sucking
Gennaro32 measures showed 78% specificity
and 80% sensitivity in
determining developmental
outcomes at 6mo of age
Murray II (prospective 57 40.1 HIE 100% NR Days to 24 Griffiths Mental Time to establish oral feeding
et al.33 cohort) establish development (days from birth) correlated
oral feeding Scales – Revised significantly with neurological
outcome (r=0.40, p=0.008)
In 14% oral feeding was not
established
Meyer III-3 (retrospective 18 28.6 SD4.2 61% diagnosed NR NOMAS 6, 12, 18, Range of measures NOMAS prediction of
Palmer cohort) perinatal brain 24, 36 including speech dysfunctional feeding associated
and injury including and feeding with developmental delay.
Heyman35 IVH, porencephaly, standardized Normal NOMAS during neonatal
ventriculomegaly, measures referred period significantly associated
asphyxia and to but no results with normal development
hypotonia; 66% reported (p<0.02). Disorganized score on
RDS or BPD NOMAS predicted later
developmental delay in only 56%
of cases

Review 801
 using the Neonatal Oral–Motor Assessment Scale

predictor of speech delay (p=0.97)

Neonatal feeding problems not a

Neonatal feeding problems not a

(NOMAS).

IVH, intraventricular haemorrhage; PVH, periventricular haemorrhage; HIE, hypoxic–ischaemic encephalopathy; NOMAS, Neonatal Oral–Motor Assessment Scale; BSID, Bayley Scales of Infant
These five studies were heterogeneous in terms of sample

predictor of cerebral palsy

size, mean gestational age, and the nature of neonatal brain
injury. A total of 269 infants were reported across the five
studies. All but one of the studies36 adequately described
clinically convenient samples. The mean gestational age of

Main findings
the infants studied ranged from 26 to 37.8 weeks. The aeti-

(p=0.70)
ology of neonatal brain injuries also varied across the stud-
ies. Two of the five studies examined early sucking and
swallowing problems in cohorts of infants with specific neu-
rological diagnoses, including neonatal arterial stroke34 and

Development; RDS, respiratory distress syndrome; BPD, bronchopulomary dysplasia; CLD, chronic lung disease; VLBW, very low birthweight; NR, not reported.
Feeding, speech, and

outcomes reported. neonatal HIE.36 The remaining three studies,30,31,37

described early sucking and swallowing problems in mixed
cerebral palsy
Neurodevelopment

No standard
Assessment

samples of term-born and preterm infants, with varied aeti-

measures
follow–up

ologies of brain injury, including intraventricular haemor-

rhage (IVH), intraparenchymal haemorrhages, mild
hydrocephalus, and periventricular haemorrhage without
HIE or hypoxic brain injury.
Age (mo)

The assessment measures used to identify early sucking and

swallowing difficulties varied across the five studies. The two
12

prospective and relatively higher-quality studies30,31 (level II;

Table I) provided details regarding the interrater reliability of
Chart review
Assessment

the assessments used. Mizuno and Ueda30 reported 100%

Early suck ⁄ swallow

interrater agreement on feeding pattern classification, based

outcomes

upon a sample of randomly selected traces generated by

instrumental assessment. Bier et al.31 also reported high levels
of interrater reliability (0.98) between two examiners using the
Age (wks)

NOMAS during the neonatal period. All but one of the five
studies34 provided sufficient detail on the sucking and swal-
NR

lowing measures used, thus permitting replication of the study

methods. Of the four studies that reported detailed assess-
respiratory distress
and bilateral; 46%

ments, the NOMAS,31,37 instrumental sucking measures,30,37

arterial ischaemic
stroke, unilateral

and imaging (i.e. fibreoptic laryngoscopy and upper gastroin-

100% neonatal
comorbidities

testinal contrast radiography)36 were employed. With one

Neonatal

exception, sucking and swallowing was assessed once during

the neonatal period. Mizuno and Ueda30 completed two
assessments of early sucking and swallowing carried out at 1
and 2 weeks after commencement of oral feeding.30
Gestational age

Three studies31,34,37 of varied methodologies and levels of

36 or greater
(wks), mean

evidence (levels II, III, and IV; Table I) explored the associa-
Participants

(range)

tion between perinatal comorbidities and early sucking and

swallowing problems. The two cohort studies31,34 reported a
significant association between perinatal factors, in particular
respiratory comorbidities, and neonatal sucking and swallow-
84
n

ing problems. However, Braun and Meyer Palmer37 did not

find an association between the total number of perinatal risk
evidence (research

III-3 (retrospective
NHMRC26 level of

factors, or any particular risk factor, and oral motor disorgani-

zation or dysfunction during sucking in their case series. The
case series by Quattrocchi et al.36 (level IV; Table I) was the
cohort)
design)

only study that examined the statistical association between

Table IV: Continued

early sucking and swallowing problems and neonatal brain

injury. Quattrocchi et al.36 reported a significant association
Barkat-Masih

between oral motor dysfunction, assessed using fibreoptic lar-

yngoscopy and upper gastrointestinal contrast radiography,
et al.34
Study

and infratentorial or brainstem lesions diagnosed on MRI, in a

sample of infants with neonatal HIE.

802 Developmental Medicine & Child Neurology 2012, 54: 796–806

Early sucking and swallowing problems associated with
later neurodevelopmental outcomes (predictive marker)
There were six studies pertinent to the second aim, examining
the predictive relation between early sucking and swallowing
problems and neurodevelopmental outcomes. Prospective
cohort designs were used in four of these studies25,30,32,33
(level II; Table II) while the remaining two used retrospective
cohorts34,35 (level26 III-3; Table II). Participants in two of
these studies were exclusively preterm, very low-birthweight
(VLBW) infants.25,32 The remaining four studies examined
prospective neurodevelopmental outcomes in mixed groups
of term and preterm infants with varied diagnoses of neona-
tal brain injury, including HIE,33 neonatal arterial ischaemic
stroke,34 and mixed neurological impairments.30,35 A total of
275 infants were included across the six studies. Conve-
nience sampling techniques were used to recruit participants
to these studies and participant samples were adequately
described.
Five of the six studies reported a predictive relation between
early sucking and ⁄ or swallowing measures and neurodevelop-
mental outcome in later infancy.25,30,32,33,35 Four of these
studies used prospective cohorts25,30,32,33 (level II; Table II)
and one was a retrospective cohort study35 (level III-3;
Table II). Instrumental measures of early sucking behaviours
appeared to be more sensitive predictors of neurodevelop-
ment, in both term and preterm infants, with and without a
neonatal diagnosis of brain injury. The prospective cohort
study by Mizuno and Ueda30 described a positive correlation
between the second neonatal feeding assessment and perfor-
mance on standardized developmental measures when
followed up at 18 months of age. All infants diagnosed with
severe sucking difficulties during the neonatal period were
reported to have severe disability on the Bayley Scales of
Infant Development (BSID) at follow-up. Mizuno and Ueda30
found that changes in sucking patterns between the first and
second instrumental sucking assessment had higher specificity
(91.7%) and sensitivity, (95.8%) than cranial ultrasound as a
predictor of infant neurodevelopmental outcomes at
18 months of age. Medoff-Cooper and Gennaro32 also
reported that early instrumental sucking measures had 78%
(reasonable) specificity and 80% sensitivity in predicting
neurodevelopmental outcomes at 6 months corrected age in a
sample of 19 VLBW infants.
The two studies that used the NOMAS as an observational
measure of early sucking behaviour provided contradictory
findings in relation to later neurodevelopmental out-
comes.25,35 Tsai et al.25 (level II; Table II) reported a positive
association between disorganized sucking patterns, observed
at 37 weeks’ gestation using the NOMAS, and developmental
delay at 6 and 12 months corrected age, in a sample of 27 pre-
term infants without diagnosed brain lesion. Furthermore,
Tsai et al.25 observed that the disorganized rating on the
NOMAS during the neonatal period was significantly associ-
ated with developmental delay at 6 months on the Psychomo-
tor Developmental Index of the BSID and at 12 months on
both the Mental and Psychomotor Developmental Index.
Conversely, in a retrospective cohort study, Meyer Palmer and
Heyman,35 (level III-3; Table II) found that a classification of
disorganized feeding (lack of rhythm of total sucking move-
ment) on the NOMAS was equivocal with respect to later
developmental outcomes, predicting delay in only 56% of
infants. These authors reported that a NOMAS classification
of dysfunctional feeding (atypical patterns of jaw and tongue
movement) was strongly associated with developmental delay,
while a normal feeding pattern was significantly associated
with normal development (p<0.02).
The retrospective cohort study by Barkat-Masih et al.34
(level III-3; Table II) was the only study that did not report a
relation between early sucking and ⁄ or swallowing problems
and later neurodevelopmental outcome. They observed that
neonatal sucking and swallowing difficulties were not a predic-
tor of later speech delay or cerebral palsy (CP) in their sample
of 84 infants with neonatal arterial ischaemic stroke. The mea-
sures used to diagnose early sucking and swallowing difficul-
ties and speech delay in this study were not described in
sufficient detail to enable replication.
DISCUSSION
The main aims of this systematic review were: (1) to describe
the concurrent relation of early sucking and swallowing out-
comes associated with neonatal brain injury; and (2) to evalu-
ate the predictive relation between early sucking and
swallowing measures and later neurodevelopmental outcomes
in this population. Nine studies relevant to these two aims
were identified through systematic searching of the literature
published from 1980 to May 2011.
Our review highlighted that the concurrent relation
between early sucking and swallowing problems and neonatal
brain injury has not been studied extensively. The few relevant
studies examined here were heterogeneous in terms of the
research design, levels of evidence (levels II, III, and IV), the
infant populations described, and the assessment measures
used. Early sucking and swallowing problems were reported
to be present in 35 to 48% of infants with different types of
neonatal brain injury. These data indicated a slightly lower
proportion of early sucking and swallowing difficulties associ-
ated with neonatal brain injury than past findings in CP.3
Specifically, Reilly et al.3 noted that sucking (57% prevalence
rate) and swallowing difficulties (38% prevalence rate) were
common in the first year of life in children with a diagnosis of
CP. The epidemiology of paediatric dysphagia remains inade-
quately documented.21,38 The discrepancy between the results
described in this review and previous research may be attrib-
uted in part to sampling issues, including small sample sizes,
predominance of convenience samples rather than whole-
cohort studies, and variability in the measures used to define
and diagnose early sucking and swallowing problems.21
The specific features of early sucking and swallowing prob-
lems associated with different diagnoses of neonatal brain
injury have not been consistently described. This may be due
partly to the different diagnostic assessments and variable defi-
nitions of sucking and swallowing difficulties present across
studies. Population-specific studies of early sucking and swal-
lowing difficulties associated with neonatal brain injury were
Review 803
limited to neonatal arterial ischaemic stroke34 and neonatal
HIE,33 both commonly reported mechanisms of brain injury
that predominantly affect term-born infants.15,39 The remain-
ing studies described early sucking and swallowing problems
in mixed samples of term and preterm infants with varied
aetiologies of neonatal brain injury, including a small number
of infants with IVH, but no infants with PVL.30,31,37 Given
that IVH and PVL are two of the most common brain lesions
associated with adverse neurodevelopmental outcomes in pre-
term infants,15 the nature of early sucking and swallowing
skills in this population remains largely unexplored.
The extent to which early sucking and swallowing problems
can be seen as a concurrent marker of neonatal brain injury
depends upon evidence of significant associations and consid-
eration of perinatal comorbidities that are likely to confound
this relation. Only one study specifically examined the associa-
tion between early sucking and swallowing difficulties and the
site of neurological lesion, and this was a case series (level IV).
Quattrocchi et al.36 found a significant association between
oral motor dysfunction and dorsal brainstem lesions on MRI
in a sample of neonates with HIE. Although this was a small
retrospective case series, these findings concur with previously
held views that lesions located in the lower brainstem contrib-
ute to neonatal dysphagia.20,22,40 Perinatal comorbidities, par-
ticularly a history of respiratory disease, appear to increase the
likelihood of early sucking and swallowing problems in infants
with neonatal brain injury.31,34 The exact nature of the rela-
tion between these early sucking and swallowing problems,
respiratory disease, and neurological difficulties continues to
be debated. Further research must address the mechanisms
underlying early sucking and swallowing difficulties in these
high-risk neonatal populations.41
Early sucking and swallowing measures predicted neurode-
velopmental outcome in later infancy in five of the six studies
reviewed.25,30,32,33,35 The body of evidence exploring the pre-
dictive relation between early sucking and swallowing mea-
sures and later neurodevelopment remains small. Of the six
studies, four used a prospective cohort design (level II); the
remaining two studies adopted retrospective cohort methodol-
ogies (level III-3). The studies were heterogeneous in terms of
the infant populations described, the sucking and swallowing
measures used, and the neurodevelopmental assessments
administered at follow-up.
Instrumental measures of early sucking and swallowing
function30,32 reported high levels of sensitivity and specificity
in predicting developmental outcomes for infants on the BSID
at 632 and 18 months.30 Conversely, non-instrumental mea-
sures of early sucking and swallowing function did not consis-
tently predict later outcomes.25,33–35 The NOMAS is an
observational checklist that rates infant feeding behaviour as
normal, disorganized, or dysfunctional based on observation
of tongue and jaw movements during both non-nutritive and
nutritive sucking. Studies that used the NOMAS reported
conflicting findings in relation to disorganized sucking ratings
and later developmental outcomes.25,35 The NOMAS is
widely used in clinical and research environments; however, its
psychometric properties continue to be debated.42,43
804 Developmental Medicine & Child Neurology 2012, 54: 796–806
Early sucking and swallowing are important oral motor
functions and it is logical to hypothesize that maturational
delays of these skills may correlate with later delays in other
gross and fine motor activities. Only three of the six studies
exploring the predictive relation between early sucking and
swallowing problems and later neurodevelopmental outcomes
examined the nature of the infant’s broader developmental
profiles later in life. Both Medoff-Cooper and Gennaro32 and
Tsai et al.25 reported that the VLBW infants with early suck-
ing difficulties were more likely to have delays on the Psycho-
motor Developmental Index of the BSID, which measures
motor skills, at 6 and 12 months of age, lending some support
to the motor skills hypothesis. Conversely, Barkat-Masih
et al.34 reported that neonatal feeding difficulties in their
cohort of infants with neonatal ischaemic stroke were not a
predictor of later speech delay or CP. The different popula-
tions examined and assessments used in these studies may con-
tribute to these discrepant findings.
Limitations of the current review
There are a number of caveats when interpreting the results of
this systematic review. The studies included here were all pub-
lished in English, and despite extensive searching there is a risk
of publication bias which increases the likelihood of including
studies reporting positive findings.44 The extent to which
results of this review can be generalized and applied in clinical
practice is influenced by the quality of the included studies.
Several methodological limitations were apparent. These limi-
tations included the heterogeneous populations studied, vari-
ability in the definition and assessments used to diagnose early
sucking and swallowing difficulties, and variability in the pres-
ence, nature, and timing of later developmental follow-up.
Sampling issues were also evident. Sample sizes were small,
ranging from 11 to 85, and convenience sampling was used in
most of the studies limiting the extent to which the findings
can be generalized to the larger population of infants.45
Across the studies included in this review, there was a lack
of consensus regarding what constitutes early sucking and
swallowing problems and the most valid and reliable way to
measure these functions. One of the major challenges in
applying these data to clinical practice is the variability in the
measures of early sucking and swallowing behaviours used
throughout this research. Only two of the nine studies
reported intra- or interrater reliability data and assessor blind-
ing.30,31 While instrumental measures of early sucking and
swallowing functions reported high sensitivity and specificity
in terms of predicting later developmental outcomes,30,32 these
tools are not accessible to all and are not routinely used in
neonatal units. The NOMAS could arguably be used more
readily in clinical practice, yet the user must obtain the neces-
sary certification and training, which is not easily available to
clinicians working outside the USA. Further, the psychometric
properties of the NOMAS continue to be debated, with
respect to the reliability of the normal and disorganized rating
categories,42,43 and further validation of the components of
the NOMAS that identify infants with neurological abnormal-
ities is required.42
Future research
A number of potential areas for future research were identified
in the course of this review. Population-specific studies of the
early sucking and swallowing problems using valid and reliable
measures are required in inception cohorts of very preterm
infants either with or at risk of IVH and PVL. Future studies
also need to compare measures of early sucking and swallow-
ing function with concurrent neuroimaging in order to
elucidate the relation between the nature, site, and severity of
neurological impairment and the clinical neonatal sucking and
swallowing examination.
There is also a need for further well-designed prospective
cohort studies, particularly in high-risk infant populations,
employing repeated, reliable measures of early sucking and
swallowing function to examine the critical diagnostic time
points for optimizing the sensitivity and specificity of predic-
tions regarding later developmental outcome. Finally, the
hypothesis that early sucking and swallowing problems may
predict later adverse motor outcomes including motor speech
difficulties requires prospective investigation using valid and
reliable measures of later motor functions.
CONCLUSION
This systematic review revealed that there is currently insuffi-
cient evidence to clearly determine the relation between early
sucking and swallowing problems and neonatal brain injury.
What can be said is that early sucking and swallowing prob-
lems do affect a consistent proportion of infants (35–48%)
with varied aetiologies of neonatal brain injury and that peri-
natal comorbidities, in particular respiratory disease, increase
the likelihood of early sucking and swallowing difficulties in
these infants. The significance of early infant sucking and
swallowing problems in predicting later neurodevelopment
has received most of the research attention. The data included
in this systematic review support the assertion that early suck-
ing and swallowing measures may predict later neurodevelop-
ment outcome. Although there is a small body of evidence
pointing to a particular association between early sucking and
swallowing problems and delays in measures of psychomotor
neurodevelopment, more high quality research is required to
substantiate this association. Currently clinicians working in
neonatal care are not able to access the instrumental sucking
measures reported to be most sensitive and specific for pre-
dicting later neurodevelopmental outcomes, and further work
is required to establish the sensitivity and reliability of existing
non-instrumental measures of early sucking and swallowing
function.
ACKNOWLEDGEMENTS
The preparation of this paper for publication was supported by a La
Trobe University Faculty of Health Sciences publication support
grant, the Victorian Government’s Operational Infrastructure Sup-
port Program, and the National Health and Medical Research Coun-
cil grant no. 607315 awarded to AM.

REFERENCES
1. Anderson PJ, Doyle L. Cognitive and educational deficits in
children born extremely preterm. Semin Perinatol 2008; 32:
51–6.
2. Roze E, Van Braekel K, Maathuis C, Martijn A, Bos A.
Functional outcome at school age of preterm infants with
periventricular haemorrhagic infarction. Pediatrics 2009;
123: 1493–500.
3. Reilly S, Skuse D, Poblete X. Prevalence of feeding problems
and ⁄ oral motor dysfunction in children with cerebral palsy:
a community survey. J Pediatr 1996; 129: 877–81.
4. Prechtl HFR. State of the art of a new functional assessment
of the young nervous system An early predictor of cerebral
palsy. Early Hum Dev 1997; 50: 1–11.
5. Allen MC, Lipkin PH. Introduction: developmental assess-
ment of the fetus and young infant. Ment Retard Dev Disabil
Res Rev 2005; 11: 1–2.
6. da Costa SP, van den Engel-Hoek L, Bos AF. Sucking and
swallowing in infants and diagnostic tools. J Perinatol 2008;
28: 247–57.
7. Estan J, Hope P. Unilateral neonatal cerebral infarction in
full term infants. Arch Dis Child Fetal Neonatal Ed 1997; 76:
F88–93.
8. Volpe JJ, editor, Neurology of the Newborn. 5th edn. Phila-
delphia, PA: Elsevier, 2008.
9. Volpe J. Brain injury in premature infants: complex amalgam
of destructive and developmental disturbances. Lancet Neurol
2009; 8: 110–24.
10. Perlman J. Neurobehavioural deficits in premature graduates
of intensive care – potential medical and neonatal environ-
mental risk factors. Pediatrics 2001; 108: 1339–48.
11. Martinenez-Biarge M, Diez-Sebastian J, Rutherford MA,
Cowan FM. Outcomes after central grey matter injury in
term perinatal hypoxic–ischaemic encephalopathy. Early
Hum Dev 2010; 86: 675–82.
12. Lynch J, Hirtz D, De Veber G, Nelson K. Report of the
National Institute of Neurological Disorders and Stroke
Workshop on Perinatal and Childhood Stroke. Pediatrics
2002; 109: 116–23.
13. Lynch J, Nelson K. Epidemiology of perinatal stroke. Curr
Opin Pediatr 2001; 13: 499–505.
14. Inder T, Warfield S, Wang H, Huppi P, Volpe J. Abnormal
cerebral structure is present at term in premature infants.
Pediatrics 2005; 115: 286–93.
15. Adams-Chapman I. Insults to the developing brain and
impact on neurodevelopmental outcome. J Commun Disord
2009; 42: 256–62.
16. Mathur A, Inder T. Magnetic resonance imaging – insights
into brain injury and outcomes in premature infants. J Com-
mun Disord 2009; 42: 248–55.
17. Anthony MY. The value of the clinical examination revisited.
Dev Med Child Neurol 2010; 52: 231.
18. Forssberg H. Neural control of human motor development.
Curr Opin Neurobiol 1999; 9: 676–82.
19. Poore MA, Barlow SM. Suck predicts neuromotor integrity
and developmental outcomes. Perspect Speech Sci and Orofac
Disord 2009; 19: 44–51.
20. Bingham P. Deprivation and dysphagia in premature infants.
J Child Neurol 2009; 24: 743–9.
21. Arvedson J, Brodsky L. Paediatric Swallowing and Feeding.
New York, NY: Singular Thomson Learning, 2002.
22. Barlow SM. Central pattern generators involved in oral and
respiratory control for feeding in the term infant. Curr Opin
Otolaryngol Head Neck Surg 2009; 17: 187–3.
23. Barlow SM, Estep M. Central pattern generation and the
motor infrastructure for suck, respiration and speech. J Com-
mun Disord 2006; 39: 366–80.
24. Nurse S, Parkes J. Nursing babies at risk of cerebral
palsy in the neonatal period. J Neonat Nurs 2010; 16:
215–20.
25. Tsai S, Chen C, Lin M. Prediction for developmental delay
on Neonatal Oral Motor Assessment Scale in preterm infants
without brain lesion. Pediatr Int 2010; 52: 65–8.
26. National Health and Medical Research Council. Additional
Levels of Evidence and Grades for Recommendations for
Developers of Guidelines. Canberra: Commonwealth Gov-
ernment, 2009.
27. Public Health Resource Unit. (2006) Critical Appraisal Skills
Programme (CASP). Making sense of the evidence. 12 ques-
tions to help you make sense of a diagnostic test study.
http://www.sph.nhs.uk/sph-files/casp-appraisal-tools/Diag
nostic%20Tests%2012%20Questions.pdf (accessed 21
October 2011).
28. Public Health Resource Unit. (2004) Critical Appraisal Skills
Programme (CASP). Making sense of the evidence. 12 ques-
tions to help you make sense of a cohort study. http://
www.sph.nhs.uk/sph-files/casp-appraisal-tools/cohort%2012%
20questions.pdf (accessed 21 October 2011).
29. Shamliyan T, Kane R, Ansari M, et al. Development of the
Quality Criteria to Evaluate Nontherapeutic Studies of Inci-
dence, Prevalence, or Risk Factors of Chronic Diseases: Pilot

Review 805
 Study of New Checklists. Rockville, MD: Agency for
Healthcare Research and Quality, 2011.
30. Mizuno K, Ueda A. Neonatal feeding performance as a pre-
dictor of neurodevelopmental outcome at 18 months. Dev
Med Child Neurol 2005; 47: 299–304.
31. Bier JA, Ferguson A, Cho C, Oh W, Vohr BR. The oral
motor development of low-birth-weight infants who under-
went orotracheal intubation during the neonatal period. Am
J Dis Child 1993; 147: 858–62.
32. Medoff-Cooper B, Gennaro S. The correlation of suck-
ing behaviours and Bayley Scales of Infant Development
at six months of age in VLBW infants. Nurs Res 1996;
45: 291–6.
33. Murray DM, Bala P, O’Connor CM, Ryan CA, Connolly S,
Boylan GB. The predictive value of early neurological exami-
nation in neonatal hypoxic-ischaemic encephalopathy and
neurodevelopmental outcome at 24 months. Dev Med Child
Neurol 2010; 52: e55–9.
34. Barkat-Masih M, Saha C, Hamby DK, Ofner S, Go-
lomb MR. Feeding problems in children with neonatal
806 Developmental Medicine & Child Neurology 2012, 54: 796–806
arterial ischaemic stroke. J Child Neurol 2010; 25: 867–
72.
35. Meyer Palmer M, Heyman MB. Developmental outcome for
neonates with dysfunction and disorganized sucking patterns:
preliminary findings. Infant-Toddler Intervention. Transdis-
ciplinary J 1999; 9: 299–308.
36. Quattrocchi CC, Longo D, Delfino LN, et al. Dorsal brain
stem syndrome: MR imaging location of brain stem tegmen-
tal lesions in neonates with oral motor dysfunction. AJNR
Am J Neuroradiol 2010; 31: 1438–42.
37. Braun MA, Meyer Palmer M. A pilot study of oral-motor
dysfunction in ‘at-risk’ infants. Phys Occup Ther Pediatr 1985;
5: 13–25.
38. Miller CK. Updates on paediatric feeding and swallow-
ing. Curr Opin Otolaryngol Head Neck Surg 2009; 17:
194–9.
39. Krägeloh-Mann I. Imaging and early brain injury and cortical
plasticity. Exp Neurol 2006; 190: S84–S90.
40. Sarnat HB. Watershed infarcts in the fetal and neonatal
brainstem. An aetiology of central hypoventilation, dyspha-
gia, Moebius syndrome and micrognathia. Eur J Paediatr
Neurol 2004; 8: 71–87.
41. Gewolb IH, Vice FL. Abnormalities in the co-ordination of
respiration and swallowing in preterm infants with broncho-
pulmonary dysplasia. Dev Med Child Neurol 2006; 48: 595–9.
42. Howe T, Sheu C, Hsieh Y, Hsieh C. Pyschometric charac-
teristics of Neonatal Oral-Motor Assessment Scale in healthy
preterm infants. Dev Med Child Neurol 2007; 49: 915–9.
43. da Costa SP, van der Schans C. The reliability of the Neona-
tal Oral–Motor Assessment Scale. Acta Paediatr 2008; 97:
21–6.
44. Cochrane Collaboration Open Learning Material (2002)
http://www.cochrane-net.org/openlearning/html/mod15–
2.htm (accessed 12 October 2011).
45. Portney LG, Watkins MP, editors. Foundations of Clinical
Research: Applications to Practice, 3rd edn. NJ: Prentice
Hall, Upper Saddle River NJ, 2009.