Changes in fetal motility as a result of congenital disorders: an overview

Authors

  • J. I. P. de Vries,

    Corresponding author
    1. Department of Obstetrics and Gynaecology, Institute of Fundamental and Clinical Human Sciences, VU University Medical Center, Amsterdam, The Netherlands
    • Department of Obstetrics and Gynaecology, Institute of Fundamental and Clinical Human Sciences, VU University Medical Center, Postbox 7057, 1007 MB Amsterdam, The Netherlands
    Search for more papers by this author
  • B. F. Fong

    1. Department of Obstetrics and Gynaecology, Institute of Fundamental and Clinical Human Sciences, VU University Medical Center, Amsterdam, The Netherlands
    Search for more papers by this author

Abstract

After 35 years of real-time two-dimensional sonography, can we now identify changes in fetal motility indicative of malfunction of the central nervous system in high-risk pregnancies? A literature search on motor assessment procedures (movements specified for body part (SMP), quantity, quality and behavioral state), and motor milestones obtained per fetus (with various congenital disorders) yielded 48 articles describing motility of 104 fetuses. In 67%, SMPs (especially isolated arm or leg movements, breathing and general movements) were assessed. Quantitative aspects were examined in 76%, qualitative aspects in 62% and behavioral states in two fetuses. Abnormal motility can be divided into two main subcategories: hypo- and hyperkinetic, demonstrating decreased variation in qualitative performance (reduced or increased, respectively, amplitude, speed and number of participating body parts, abnormal quantity (reduced or increased, respectively) and reduced differentiation into SMPs. Posture was affected in 40/60 hypokinetic and 4/44 hyperkinetic moving fetuses. The majority of the disorders resulted in an adverse outcome. Fourteen percent survived with a handicap, depending on the underlying disorder. The 16 disorders with hypokinetic motility had mainly an autosomal recessive etiology with no possibility of invasive prenatal diagnosis or conclusive sonographic structural anomalies, in contrast to the 17 disorders with hyperkinetic motility. Within the limitations of the studies, a deeper understanding of affected milestones in motor development can be obtained. Broadening motor assessment procedures from quantitative only to include qualitative aspects, differentiation of SMPs and behavioral states and emphasizing onset and continuity of motility before and after birth will enhance the reliability and predictive value of motility as a parameter in the assessment of fetal condition. Copyright © 2007 ISUOG. Published by John Wiley & Sons, Ltd.

Introduction

Fetal motility can be considered to be a spontaneous expression of the developing nervous system. The first publications on motor activity that differed from normal, as observed on real-time sonography, appeared in 1978. They concerned motor activity in high-risk populations for fetal growth restriction (FGR)1, 2 and fetuses of mothers with diabetes mellitus3. These publications appeared 7 years after the onset of articles on normal motor activity4.

From that point on there have been a number of studies in this area, some very promising, but on the whole they have not provided systematic evidence. For example, reports on quantification of movement activity were promising in predicting preterm birth5, 6, but were later contradicted7. Also, in a recent systematic review, the potential value of absent breathing movements in predicting preterm labor was noted, but the sample sizes and quality of the methods of the studies were deficient, preventing any final conclusions from being drawn8. Changes in the quality of the performance of movements in FGR were published as an important tool for identifying deterioration in the integrity of the fetal central nervous system (CNS)9 and seemed to have well-proven reproducibility and consistency with the findings after birth10. The interobserver agreement of the qualitative assessment as scored from the three longest general movements of a 1-h observation was high (kappa, 0.90 ± 0.07)10. Moreover, a disruption of behavioral state organization in FGR and in fetuses of mothers with diabetes mellitus was considered to be an early sign of influence on the integrity of the CNS11. In other words, the first steps have been taken towards exploring possibilities for surveillance of the fetus at risk of developing transient or persistent neurological problems. Marsal12, for example, advised on ultrasonic motor assessment methods in 1983. Though these methods have been applied scientifically to obtain insight into changes in motor activity in comparison with normal pregnancies, they have found only limited application in daily practice.

The same holds true for fetuses at high risk for congenital anomalies. Publications on this subject present mainly case histories of rare congenital anomalies involving neuromuscular-skeletal anomalies, presenting various motor problems rather than reporting on the systematic surveillance of fetuses. All authors described their findings within the limits of their own background, addressing their own points of interest for assessment of motor activity. Still, a rather high number of these congenital anomalies cannot be diagnosed with sonographically visualized structural anomalies or invasive techniques. In future, will a more systematic approach to motor assessment assist in the prenatal diagnosis of such anomalies?

The aim of this article was to evaluate the literature on fetuses determined to be at high risk for motor anomalies, particularly congenital ones, by means of real-time two-dimensional sonography. The assessment procedure and the examination of results found are identical to those in our overview of normal fetal motility4. Here we describe milestones in development in these high-risk fetuses and discuss the present applicability of these milestones in clinical practice through an unequivocal assessment procedure.

Methods

A literature search in Pubmed® and Medline® was performed as described previously4. To the original search terms (‘fetal’, ‘movements’, ‘motility’, ‘movement patterns’, ‘ultrasound’ and ‘sonography’), we added the terms ‘congenital malformation’, ‘congenital abnormality’, ‘fetal hypokinesia’, ‘akinesia’, ‘antenatal diagnosis’, ‘activity’ and ‘fetal joint contractures’. Finally, on the basis of the references in the articles found, a manual search was performed. As described in the introduction, the populations examined included human fetuses at high risk for congenital anomalies.

Assessment procedure

The articles were examined for assessment of the following: specific movement patterns (SMPs), non-SMPs (NSMPs), and both SMPs and NSMPs; quantitative aspects of movement; qualitative aspects of movement; behavioral states; duration of observation (≤ 15 min, 16–59 min, ≥ 60 min and unknown); and gestational age at which the fetus was examined (≤ 20 weeks, > 20 weeks). These items are defined in our previous overview4. The presence of postural anomalies was also scored.

Aims of the studies

One of the following four aims4 was distinguished for each article: to study emergence of movements; to study development of movements with age; to study continuity of movements after birth; to study mutual relationships between motor parameters and relationships with other parameters applied for the assessment of fetal condition.

During the evaluation of the literature on congenital anomalies, the general impression arose that a large proportion of abnormal motor activity could be divided into two subcategories with respect to qualitative and quantitative aspects: hypokinetic movements, consisting of qualitative aspects demonstrating reduced amplitude, speed, or number of participating body parts and/or quantitative aspects being normal, reduced or absent; hyperkinetic movements, consisting of qualitative aspects relating to increased speed, amplitude, number of participating body parts up to seizures, and/or quantitative aspects from normal, burst–pause patterns to continuous activity. We categorized the various anomalies as hypo- or hyperkinetic.

Raw data are presented per fetus. This was necessary because articles on congenital abnormalities describe only one fetus or a few fetuses with diverse anomalies, producing various motor anomalies and assessed with different procedures.

Results

With the literature search beginning in 1970, 48 articles were found, examining 104 fetuses. In 60 cases (36 articles13–48) the motor activity was categorized as hypokinetic and in 44 (15 articles17, 42, 44, 49–60) it was hyperkinetic; three articles reported fetuses with both hypokinetic and hyperkinetic movements17, 42, 44. The 5-yearly distribution of the articles has been stable since 1980.

Assessment procedure

The various aspects that were examined in the assessment of motor activity are presented for all fetuses in Table 1. In the SMP research, isolated leg and arm movements were studied most, followed by breathing movements and general movements (Table 2). The literature described, in 60 fetuses, 16 different underlying disorders that induced hypokinetic movements, and in 44 fetuses, 17 disorders that induced hyperkinetic movements. Table 3 presents, for each disorder, whether SMPs were applied, and if so whether differentiation was used, and application of qualitative and quantitative aspects in SMPs and NSMPs. Also described in the table is the presence of abnormal posture; it was noteworthy that the assessment procedures differed considerably (Table 1). Behavioral states, continuity, and relationships with other parameters were not main aims investigated in the literature studied (Table 1).

Table 1. Distribution of motor assessment procedures in congenitally abnormal fetuses with hypokinetic or hyperkinetic motor patterns
 Fetuses (n)
HypokineticHyperkineticTotal
  1. (N)SMP, (non-)specific movement pattern.

Total6044104
SMP333770
NSMP281341
Both167
Quantity512879
Quality273764
Postural abnormalities40444
Observation duration 
 ≤ 15 min044
 16–59 min142236
 ≥ 60 min161329
 Unknown30535
Gestational age 
 ≤ 20 weeks262248
 > 20 weeks342256
Aim 
 Emergence17118
 Development434386
Table 2. Ranking of specific movement patterns (SMPs) studied in 48 articles on congenital abnormalities, separated into hypokinetic and hyperkinetic movement patterns
Hypokinetic (no. of fetuses)Hyperkinetic (no. of fetuses)
Isolated leg movement (22)Isolated leg movement (17)
Isolated arm movement (18)Breathing (16)
Breathing (10)Isolated arm movement (14)
General movement (7)General movement (11)
Eye movement (5) 
Sucking and swallowing, finger, trunk/chest (4)Trunk/chest, retroflexion head, rotation head, startle (9)
Head, eye, jaw opening (3)Sucking and swallowing, stretch, jaw opening, anteflexion head, hiccup, yawn (8)
Wrist (2)Eye movement (7)
Ankle, knee, spine, tongue, retroflexion head, anteflexion head, rotation head, hiccup, startle, stretch (1)Vomiting (2)
Table 3. Hypo- and hyperkinetic movement disorders and gestational age of first occurrence of abnormal movements
DisorderEtiologyFetus (n)GA (weeks)GA > 24 weeks (n)Diff. in SMP (n)Qual. (n)Quant. (n)Postural abn. (n)Prenatal death/ neonatal death/alive (n/n/n)*
  • *

    Termination of pregnancy or spontaneous fetal death/died within days after birth/alive at the time of writing.

  • Associated with transient ischemic/hypoxic insults.

  • At 3 years no spontaneous respiration, mental retardation.

  • §

    Seizure-like activity.

  • Vomiting-like activity.

  • #

    Extrapolation from the article reporting the outcome being biliary atresia and trisomy 18. Aut, autosomal; Diff. in SMP, differentiation in specific movement patterns; dom, dominant; GA, gestational age; Postural abn., postural abnormalities of the extremities and/or face; Qual., qualitative assessment of movement; Quant., quantitative assessment of movement; rec, recessive.

Hypokinetic
Fetal hypo-/akinesia deformation sequenceAut rec - various3611–38121313332727/8/1
Distal arthrogryposisAut dom317–23033330/0/3
Congenital contractural arachnodactylyAut dom125111110/0/1
Non-lethal arthrogryposis multiplex congenitaAut rec - various118011010/0/1
Cerebro-oculofacioskeletal syndromeAut rec121010111/0/0
Generalized amyoplasiaSporadic119000111/0/0
Moebius syndromeSporadic137110100/0/1
HoloprosencephalyIsolated - aut dom419–37220413/1/0
Schwartz–Jampel syndromeAut rec117011110/0/1
Seckel syndromeAut rec135110100/0/1
Restrictive dermopathyAut rec318–28232210/3/0
Nemaline myopathySporadic136100110/1/0
Prader–Willi syndromeChromosome 15 deletion at q11-q13 (75%)135111100/1/0
Spinal muscular atrophyAut rec136111110/1/0
Smith-Lemli–Opitz syndromeAut rec136111110/1/0
Spina bifida apertaVarious337–39133000/0/3
Hyperkinetic
AnencephalyVarious1410–38714914013/1/0
Trisomy 21 617–20006006/0/0
Trisomy 1q 220022202/0/0
Monosomy 5p 319033303/0/0
Trisomy 5p 119111101/0/0
Trisomy 4p 119011101/0/0
Trisomy 18 318–34233221/2/0
Spina bifida apertaVarious118011000/0/1
Fanconi anemiaAut rec122111101/0/0
Olivopontocerebellar hypoplasiaSporadic/ aut rec/dom/X-linked128111§000/1/0
Myoclonic encephalopathyVarious127111§110/1/0
HydranencephalyVarious128111§101/0/0
MicrocephalyVarious135101100/0/1
Apert syndromeAut rec130111110/0/1
HydrocephalyVarious424–3634400Not known
Tracheal atresiaVarious139111000/1/0
Esophageal atresiaVarious234222000/2/0#

Hypokinetic group13–48

Emergence. The onset of abnormal motility was examined in 17 fetuses (15 fetal hypo-/akinesia deformation sequences (FAHS), one restrictive dermopathy and one Schwartz–Jampel syndrome). In 15 FAHS in which the onset of abnormal motility was examined, 13 NSMPs and two SMPs were quantified; all were found to exhibit reduction and one was qualified. In the latter fetus, SMPs were examined between 15 and 35 weeks. The quality of the general movements and differentiation into SMPs became abnormal from 27 weeks onwards and the quantity of general movements reduced after 28 weeks32. Four studies investigated the emergence of abnormal motor development in a series of pregnancies. 1. One of these studies31 investigated a case with restrictive dermopathy, in which the first pregnancy showed at 29 weeks a fetus with abrupt bursts of swallowing and a continuously open mouth. Except for FGR there were no structural anomalies. The second fetus was examined between 18 and 30 weeks; the quality of the general movements was suspect or abnormal from 18 weeks onwards (reduced participation of body parts, with less waxing and waning of general movements as the gestational age increased, but short-lasting abrupt movements initiated in legs, hips and lower back). No reduction in SMP differentiation occurred, whereas the quantity of the general movements reduced after 29 weeks. A spontaneous delivery occurred at 30 weeks. In the third pregnancy, sonographic examination was performed at admittance for preterm labor at 28 weeks, and hands, elbows and feet showed contractures in the presence of fetal body and mouth movements. All three neonates died within hours after birth. 2. In another study37, in a first pregnancy FAHS was demonstrated at 18 weeks, in the second it was demonstrated at 11 weeks, and in the third at 11 weeks. Whereas in the first two pregnancies, movements and posture of upper and lower extremities were involved, in the third pregnancy, leg movements (extension only) were still present at this point, and there were normal arm movements and posture; reduced movements were observed at 14 weeks and there was akinesia at 16 weeks. The fourth pregnancy was uneventful. 3. In the third study14, in a first pregnancy FAHS was diagnosed at 21 weeks. In the second, normal movements were observed at 10 weeks, but FAHS was diagnosed at 15 weeks. In the third, normal motility was observed at 9, 11, 13 and 16 weeks, but FAHS was diagnosed at 18 weeks. The fourth pregnancy was unaffected. The fetuses in this case resembled phenotypically Neu–Laxova syndrome (which is autosomal recessive). 4. Finally25, 10 fetuses from women with a prior pregnancy affected by FAHS as confirmed by postmortem examination were examined between 8 and 16 weeks. In seven, FAHS was diagnosed within this period and in one it was diagnosed at 29 weeks. One fetus with Schwartz–Jampel syndrome23 was examined for both SMPs (head, arms, legs) and NSMPs, exhibiting reduced differentiation of SMPs, qualitative abnormalities and reduced quantities of movements at 17 weeks. Thus, motility can worsen over time14, 31, 32, 37, can have an early onset13, 14, 25, 37 and demonstrates abnormalities in various motor aspects. There have also been studies describing a total absence of motility, with no possibility of defining the specificity of movement patterns14, 25, 37, 39, 48.

Development. The quantity of movement was reduced in all but three of the 43 fetuses studied for developmental aspects. In one of the 21 cases with FAHS that were studied for developmental aspects, which had underlying Neu–Laxova syndrome (according to postmortem examination)25, a burst–pause pattern in activity was found despite a reduced quantity. Qualification of the movement was performed in 23/43 (53%) cases that were examined for developmental aspects and all were abnormal. In four of the FAHS24, 31, 32 cases, the researchers found not only a reduction in the quality of movements (amplitude, speed, participation of body parts) but also that abrupt movements occurred. In three cases of FAHS seizure-like movements occurred45.

The time of diagnosis of abnormal movements in all 36 fetuses with FAHS, for example, varied from 11 to 38 weeks' gestation, though the majority (n = 24) were diagnosed before 24 weeks (Table 3).

Qualitative examination in FAHS revealed hypokinetic movements in all cases (13/13), but also abrupt movements in three and even seizures in another three fetuses.

In 27/36 cases with FAHS, postural anomalies were described. The posture of the extremities, often combined with facial anomalies, was affected in 40/60 fetuses. Of the 36 cases with FAHS, one case involved only upper extremities46, one involved abnormal arm movements at 35 weeks, 8 weeks later than the appearance of abnormal movements of the lower extremities32, and one case with an abnormal arm posture was found at 14 weeks, 3 weeks later than the appearance of an abnormal leg posture37.

Behavioral states were examined in two cases. No behavioral states were found in the case of Smith-Lemli–Opitz syndrome38 and abnormal behavioral states were found in the case of Prader–Willi syndrome19.

Hyperkinetic group17, 42, 44, 49–60

Emergence. One article described the quality of SMPs (startles, isolated arm movement, general movement) being abnormal as early as the first trimester (10 weeks) in an anencephalic fetus59. The examination was performed within a research set-up relating 30 min of motor observations at 10 weeks to postmortem findings in spontaneous abortions.

Development. A reduction in differentiation was described in six anencephalic fetuses during the second trimester (16–19 weeks)60.

Quantity was examined in 26/43 fetuses for developmental aspects. In those with anencephaly six showed increased, four showed normal and four showed reduced movements. Increased and reduced movements were described in trisomy 18, hydranencephaly and Apert's syndrome42, 51, 53, 56. Burst–pause patterns were described in seven fetuses of which five were anencephalic, one had Fanconi anemia and one had myoclonic encephalopathy. This was found to be correlated histologically with absence of the hindbrain in the five anencephalics60. One trisomy 18 fetus, two trisomy 1q fetuses and one trisomy 5p fetus demonstrated continuous activity51, 56.

Qualitatively, all 37 examined were found to be abnormal. Four fetuses had seizure-like activity: myoclonic encephalopathy55, olivopontocerebellar hypoplasia58, hydranencephaly53, and microcephaly57. It is noteworthy that, with respect to those with chromosomal anomalies, six of 11 cases of trisomy 21 exhibited an abnormal quality of NSMP (at 17 and 20 weeks)51. Qualitatively abnormal movements in two structural anomalies were limited to one SMP. The first was in a case with tracheal atresia showing large-amplitude, high-speed (jerky, vigorous) breathing49, while the second was with two fetuses with esophageal atresia which had the general appearance of vomiting (mouth opening, extension of the neck, tongue protrusion)52.

Posture was described as affected in five fetuses (two with trisomy 1851, 56, one with myoclonic encephalopathy55, one with hydranencephaly53 and one with Apert's syndrome42).

Discussion

This review of the literature adds significantly to our knowledge of how comprehensive is the influence of congenital anomalies on fetal motility. Hyper- and hypokinetic changes in motor aspects in relation to normal motor development were found to be associated with congenital anomalies. The anomalies inducing hypokinetic motility consisted mainly of autosomal recessive disorders with neuromuscular–skeletal–skin disorders and those inducing hyperkinetic disorders consisted mainly of sporadic or low-recurrence-rate structural or chromosomal disorders with CNS involvement. The majority had an adverse outcome (14% alive with a handicap varying according to the underlying anomaly). With respect to the assessment procedure, the following changes in motor development were determined. First, the differentiation into SMPs was reduced in both cases of congenital abnormalities with hypokinetic movements and in those with hyperkinetic movements (FAHS and anencephaly). Second, qualitative changes were investigated in 62% of cases. Third, a quantitative change of movement, either increased or decreased, was investigated in 76% of cases. The normal distribution of increase and decrease in ongoing activity during one burst of general movements can disappear in fetuses with congenital abnormalities. Abnormal temporal patterning was found in fetuses with hyperkinetic movements: a burst–pause pattern in activity was observed in those with anencephaly, Fanconi anemia and myoclonic encephalopathy; continuous activity was observed in fetuses with chromosomal anomalies (trisomy 18, trisomy 1q and trisomy 5p). Posture was affected in 42% of cases and this was not limited to fetuses with hypokinetic movements.

Finally, behavioral states were studied twice only and were found to be disrupted in fetuses with hypokinetic movements.

Over the years, a stable number of publications has appeared relating to congenitally abnormal fetuses with sonographic identification of motor and or postural abnormalities. From their assessment procedures it was evident that most of the investigators put considerable effort into using adequate observation times in the set-up of their studies; most examinations lasted > 15 min. Adequate duration of observation is gestational-age-related; 15 min can be considered sufficient in observations of fetuses aged < 20 weeks' gestation, 30 min is required between 20 and 30 weeks and ≥ 60 min thereafter4. There was a difference in the assessment procedures performed between the group of fetuses with hypokinetic and the group with hyperkinetic motility. The hypokinetic group was quantified more frequently than it was qualified (85% and 45%, respectively), while the reverse was true for the hyperkinetic group (64% and 84%, respectively). It is important to realize that a larger proportion of the quantified hypokinetic group, compared with the hyperkinetic group, had unknown observation periods, while important conclusions such as reduced activity or akinesia are presented in these articles. On the other hand, these hypokinetic fetuses were examined more often than were hyperkinetic ones for the emergence of abnormal movements. This interest in the emergence of abnormal movements fits with the effort to offer prenatal diagnostic possibilities in the absence of invasive diagnostics or conclusive diagnosis by sonographically visualized structural abnormalities in this diverse group of neuromuscular–skeletal–skin disorders with varying onset of motor and postural anomalies.

A number of critiscisms of the various published studies must be made at this point, because the reproducibility of the assessment procedure was influenced negatively by several aspects of the research, which hampered comparison between investigations. First, although SMPs were applied in the majority of motor examinations, their differentiation was only studied for a limited number of congenital abnormalities, despite the fact that their early and strongly age-related differentiation allows investigation as early as the first trimester. The attention to general movement, the most frequently occurring SMP, was also limited, despite its higher reproducibility compared with the NSMPs. Moreover, in addition to the higher reproducibility, judgement of the quality of general movements has proved to be of prognostic value for the integrity of the nervous system and later neurological outcome in congenitally normal but growth-restricted fetuses9, 10.

Second, qualitative changes were described in various inexact ways, using such adjectives as ‘feeble’, ‘vigorous’, ‘jerky’ and ‘abrupt’. In cases of congenital abnormalities, the movement repertoire is often damaged to such an extent that systematic comparison of the different qualitative aspects with cases of normal motility can help to elucidate how differently the various SMPs, including general movement, are performed in abnormal cases. Observing abnormal motility demonstrates impressively how much of the normal variation in amplitude, speed, participating body parts, fluency, and increasing and decreasing activity is lacking. Reporting these aspects more precisely creates the baseline characteristics of the quality of motility and therefore enhances reproducibility.

Third, though quantitative changes have been examined extensively, definitions of single movements or their duration vary greatly. It has been determined that the best interval between bursts of normal motor activity is 1–3 s in order to allow discrimination between short-lasting and long-lasting activity bursts (in general movements or body movements)61. Despite the fact that it is widely accepted that a characteristic of normal motility is the wide range in activity, most effort has still been placed on simple assessment of quantity. There is a growing awareness that quantitative assessment seems to be of more value for demonstrating trends in the wide variance of activity than for locating individual fetuses at high risk. In 34% of the studies reviewed, the duration of movements was not even presented, making quantification a non-reproducible result. Only extremes in abnormal quantity will be helpful to determine if an activity is not within the normal range, for example, continuous activity or absence of or severe reduction in motility over a well-defined observation period. Temporal patterning has only been studied in a limited way, and the organization of behavioral states, as one of the most important milestones in temporal patterning, was studied only twice.

Fourth, fetal posture, as an important result of neuromuscular development, was reported in only a minority of the studies. Yet, its importance is reflected in the finding that postural anomalies were reported not only in disorders inducing hypokinetic but also in those inducing hyperkinetic movements.

Finally, though population characteristics were, for the most part, defined thoroughly in the studies reviewed, they are not presented in this overview because most publications were case histories, with a few exceptions reporting the same assessment procedures in a number of fetuses with the same congenital abnormalities, thus enhancing the reproducibility of the assessment44, 50, 51, 60. There is also a realistic possibility of bias in the presented articles on congenital abnormalities, if only cases in which the abnormal findings of motor anomalies were found to be associated with congenital abnormalities were submitted for publication.

Neuromuscular development

Despite the limitations of the studies described here, the following milestones in neuromuscular development could be determined.

Differentiation into SMPs was found to be reduced from the onset of the disorder, regardless of in which of the three trimesters this occurred. The start and subsequent differentiation of the movement patterns during the first trimester appeared to be based on a minimal neural and myogenetic structure62. However, the development of the neuromuscular system must be sufficient to facilitate spontaneous motility from about 7 weeks' gestation onwards, and the emergence of all the different SMPs, in the 7 weeks thereafter, which remain recognizable throughout gestation. The heterogeneous etiology of FAHS may start its destructive effects on various systems at different gestational ages, and there must be variable expression since the offspring of one family showed variance of a few weeks in the onset of the disease14, 37. This has also been illustrated in a series of proven central core disease myopathies (histologically and by mutations in the ryanodine receptor gene)39. This disease is considered as a relatively benign, non-progressive hypotonia during early childhood. However, it can also have severe forms, with fetal akinesia and postural anomalies, like the one of 36 fetuses with FAHS in this overview reported to be alive. This fetus needed ventilatory assistance during the 1st month, had swallowing difficulties for 5 months and received intensive orthopedic care thereafter.

In contrast to FAHS, with its onset of abnormal motor activity at various ages, the quality of general movement in anencephalics was found to be influenced from its very emergence, as found in a case at 10 weeks59, explained by the reduced supraspinal influences to induce variance and thus complexity of the movement. Likewise, the reduced differentiation in SMPs occurs in anencephalics from the very beginning of motor activity, and has been demonstrated throughout the period between 10 and 35 gestational weeks60. It is impressive that even with histologically proven absence of cervical cord, isolated arm movements can be observed, suggesting that in the absence of supraspinal connections, parts of the body can move using merely ectopic motoneurons either inside or outside the CNS60. The mechanisms of disintegration of the nervous system can result in various motor abnormalities within a certain disorder. This is clearly demonstrated within the reported anencephalic, FAHS and restrictive dermopathy fetuses, all of which have extensive but variable motor anomalies. In the latter two disorders, a worsening of quality with age was observed. As mentioned earlier, it is important to realize that the underlying cause of FAHS varies and often is not even known. Despite the seriousness of the disorder, it does not always occur in both upper and lower extremities, necessitating very careful follow-up in cases in which only one extremity deformity is found before 24 weeks. This is in contrast to older fetuses or neonates, in which the finding of asymmetrical postural anomalies is considered reassuring. The important overview of Smith's Recognizable Patterns of Human Malformation63 describes underlying disorders of asymmetrical posture deformities as being likely caused by extrinsic factors for intrauterine constraint (e.g. uterine malformation), whereas symmetrical deformities are caused by intrinsic factors (e.g. CNS, peripheral nervous system, muscle, connective tissue, skeleton). Another expression of the varying involvement of motor activity is reflected in the fact that the quality of movements may not undergo the expected reduction in amplitude and speed: short-lasting high-speed bursts of activity (abruptness) have also been encountered in the case of FAHS. While it seems logical that, in cases of restrictive dermopathy, the skin restriction limits the amplitude and speed of movements, abruptly performed general movements were also seen in one fetus31. These abrupt movements must be explained by a problem that is initiated more centrally. This is also the case for leg movements in spina bifida aperta, which can vary from normal to abnormal hypokinetic to hyperkinetic, depending on the supraspinal connections44.

In this review, 16 different congenital disorders inducing mostly hypokinetic motility and 17 inducing mostly hyperkinetic motility have been reported. In comparison to our extensive knowledge for neonates (for example, Smith's Recognizable Patterns in Human Malformation63 reported 52 disorders with frequent hypotonicity and 35 with frequent hypertonicity or seizures), this is just the beginning. When comparing fetal and neonatal data it is important to realize that while findings before birth are confirmed, they may also undergo further development. This is the case, for example, for two anomalies with hypokinetic movements before birth (Smith-Lemli–Opitz and Prader–Willi syndrome). These neonates remain hypokinetic, but can have seizures thereafter as well. The six cases with abnormal quality of motor activity in the chromosomal anomaly, trisomy 21, are in agreement with the findings of Mazzone et al.64 on general movements in children with Down syndrome, who also demonstrated heterogeneity among the children. As such, we do not advocate any application for clinical practice, but point out that even in trisomy 21, qualitative motor changes can be detected.

Quantitative changes in motor activity in the fetuses described was often found to be influenced in such a way that the normal waxing and waning of an ongoing general movement instead became reduced, continuous or burst–pause activity. These phenomena are not transient. Reduced motility is found especially in anomalies with limitations caused by central and peripheral nervous system abnormalities as well as muscular involvement, whereas increased, continuous and burst–pause activity is found particularly when there are abnormalities in the CNS hindering normal patterning.

Behavioral states develop late in pregnancy (36 weeks) and are considered to be an expression of the integrity of the nervous system. They have been found to be one of the first parameters of fetal well-being that are influenced in cases of FGR caused by uteroplacental insufficiency65, 66. In the severe congenital disorders presented here, their late emergence means that they are of limited diagnostic value. In the case of Prader–Willi syndrome, prolonged inactive periods were described19, and this has been confirmed by others. These prolonged inactive periods with short active periods correspond with the diminished quantity of the fetal movements.

The normal development of joints starts at about 5.5 weeks' gestation. By 7 weeks, many joint spaces exist and movement is possible from 8 weeks onwards. Movement is thought to be essential for the normal development of the joints62. There is a growing awareness of the influence of limitation of movement on posture even under physiological circumstances. Fetuses in cephalic presentation are more likely to have flexed wrists than are fetuses in breech presentation. The wrist of the cephalic fetus is situated in the lower segment, whereas the breech fetus has more room in the fundus of the uterus67 (since both have a preferred elbow flexion, the environmental influence further away from the lower segment and fundus has been shown to be identical).

Contractures are secondary to intrinsic (neuromuscular, skeletal, skin) or extrinsic (fetal crowding and constraint) factors. Intrinsic factors often lead to symmetrical contractures, and extrinsic ones generally to asymmetrical contractures63. Neonates born with arthrogryposis can be divided into three main categories: those with limb involvement only; those with involvement of the trunk, craniofacies or viscera, and limbs; those with severe CNS dysfunction68. An extensive differential diagnosis is provided in this article. In general, the outcome of the first group is good, as is that of the second, depending on the specific diagnosis, while the third group has a poor outcome, with about 50% dying within the 1st year.

Sonographic motor assessment is not implemented routinely in the care of high-risk pregnancies. As obstetricians we wish to be able to distinguish the fetus at risk for developing a motor handicap after birth, for example in the case of FGR, or the fetus at risk of miscarriage because of a congenital abnormality involving the CNS. Better insight into the continuity of movements before and after birth would enhance the possibility of the clinical application of sonographic motor assessment and would make necessary collaboration with pediatricians active in this field. Longitudinal studies on small groups of fetuses at high risk for congenital abnormalities are worthwhile and need to be extended to larger groups for testing the reliability of its application.

Because not all anomalies can be detected by ultrasound, magnetic resonance imaging or chromosomal, DNA or metabolic testing, previous investigators have shown their efforts in the clinical application of motor assessment by publishing case histories. However, these researchers have not systematically studied motor assessment. Most studies were case reports with no longitudinal approach to investigate changes over time or after birth; both are necessary for testing the reliability of fetal motor assessment in these rarely occurring disorders with respect to later neuromuscular outcome. As demonstrated in FAHS, the first sign can be limited to one extremity contracture only. Thus, a repeat examination after 2 weeks would indicate if there has been progressive worsening of a centrally located rather than merely a peripheral disorder.

In 44% of cases the motor anomalies of the various congenital abnormalities were found after 24 weeks' gestation, an important age in daily practice taking into consideration the possibilities of viability and termination of pregnancy. Ranking the prevalence of these SMPs per observation period might be helpful because ranking is found to be a rather consistent age-related phenomenon in a normal population. Ranking might shed more light on the change in distribution of the SMPs than would quantification alone4.

This overview, as was our overview of normal motor development4, is limited by our choice to evaluate only a few items of motor assessment and focus on one main developmental aspect per fetus. However, both reviews reveal that fetal sonographic assessment should include various motor aspects. It is not only the quantitative data, though easy to reproduce, that give insight into motor changes, but also, or even more so, data on quality, emergence and differentiation of SMPs and behavioral states that are important. Which motor aspects change is dependent upon the underlying problem. Both overviews are meant to stimulate future research on various aspects of motor activity with reproducible methods. The next step could be the recommendation to make use of prenatal motor assessment procedures in future routine care. Since we are not yet able to discern fetuses at high risk for developing a motor handicap after birth, we must strive to increase the possibilities of the fetal neurological examination with respect to its functional expression of the CNS. A multidisciplinary approach facilitates linkage with neurological outcomes or possible postmortem findings.

Acknowledgements

For assistance in the literature search we are indebted to the efforts of A. Banga, student of the Faculty of Movement Sciences, Institute of Fundamental and Clinical Movement Sciences, VU University, Amsterdam, The Netherlands.

Ancillary