From MIDWIFERY MATTERS, Issue No.88, Spring 2001
Maren Dietze
EVERY MIDWIFE is expected to understand the mechanism of labour, so this has to be included in the midwifery curriculum. It seems to be generally understood that the mechanism of labour describes the movements of the fetus in response to the bony structures and soft tissues of the birth canal during labour, but as the fetus is largely hidden from view how do we know how the fetus moves during labour? Two of today's major midwifery textbooks (Sweet and Tiran, 1997; Bennett and Brown, 1999) outline the process but provide no references to research underpinning the theory in the relevant chapters.
Until the l7th century beliefs about labour evolved around the idea of an active fetus that 'works its way out' of the womb and separates the pelvic bones in the process (Rhodes, 1995). In 1701 the Dutch obstetrician, Deventer, was the first to depict the pelvic girdle accurately according to today's knowledge. He believed that the pelvic bones could part to some small extent; he described the sacrum and coccyx as swinging backwards; however, he regarded pubic separation as rare and unimportant (Wilson, l995). However, this knowledge, together with the idea that the fetus actively worked his way out of his mother's body, was disputed and increasingly went out of favour amongst obstetricians of the 18th century.
The 18th century saw the rise of obstetrics as a science (Towler and Bramall, l986). The laws of physics were applied to labour in an attempt to gain an improved understanding of the birthing process. However, this mechanical approach could be applied only if the parts involved were reduced to a static pelvis and a passive fetus (Speert, l996). The works of Fielding Ould (1741), William Smellie (1752) and Solares de Renhac (1771) were the main contributors to the formulation of the mechanism of labour (Rhodes, 1995) as we know it today.
The use of anatomical models became fashionable all over Europe in the 18th century (Boschung, 1981) partly because of limited patient material and partly because the mechanical view of labour ignored the role of the woman as a whole. This allowed the teaching of the concept of rotation of the presenting part as it descends through the pelvis, which was especially important when teaching the use of the forceps (Wilson, l995). Models of the pelvis or the woman's lower body are still used today in midwifery and obstetric teaching.
The next logical step was pelvimetry, measuring women's pelves, in order to predict the course of labour for individual women. Pelvimetry, with its emphasis on quantification, precision and standardisation, promoted the recognition of obstetrics as a science (Hiddinga and Blume, 1992). Pelvic measurement was a practical solution to a common problem; rickets with its accompanying pelvic deformity was common in l9th century Europe. According to Shorter (1983) rickets affected one in four women in some areas. This may have influenced the experience of midwives, obstetricians and women considerably.
In the 1920s and 1930s the x-ray offered new possibilities directly to measure internal diameters of the pelvis in living, pregnant women (Hiddinga and Blume, 1992). Radiologists began to offer prognoses for labour. Some obstetricians started to view every primigravid woman as an 'untried pelvis'. This attitude led to the introduction of the 'trial of labour', which included close monitoring of progress; a practice that is still being used today. Recently a systematic review of randomised controlled trials suggested that pelvimetry is not justified because the only significant effect is a rise in the caesarean section rate (Pattinson, l999). No beneficial effect has been shown.
I came to this subject through Morison, a doctor who, in 1971, wrote a simplified explanation of the mechanism of labour for pupil midwives. This was entirely based on Smellie's work, and included Smellie's illustrations. I wondered whether this reworking of ancient obstetric opinion was relevant to today's midwives, and whether the mechanism of labour as depicted in midwifery textbooks was consistent with the latest research. I searched the literature to try and answer these questions.
Although some might think that theoretical research is only of academic interest, I think it is relevant for midwifery practice because midwives use the theory of the mechanism of labour to assess progress during labour, and to explain events during labour. Midwives' thinking and their interpretation of their experiences have a profound impact on their practice (Bryar, l995). However, Bryar found a tendency to disregard the need for theoretical frameworks in some of today's midwifery research, which reduces the usefulness of such studies in terms of generating basic knowledge. This applies to some degree to the HOOP trial (McCandlish et al, 1998) which investigated different delivery techniques while omitting to relate these techniques to the mechanism of labour or any other theoretical explanation.
Lately, the need for an epistemology of midwifery knowledge has been voiced (Siddiqui, 1994; Kelly, 1997). Midwives need to consider how knowledge is generated and what kinds of knowledge they are using. This study lays strong emphasis on knowledge derived from scientific research. However, I also attempted to consider knowledge gained from midwives' experience, so far as it was identified by the literature search, because this kind of knowledge can be regarded as true midwifery knowledge, i.e. not 'borrowed' from other professions (Walker and Simpson, 1998).
My literature search and review aimed to identify any literature investigating or testing the theory of the mechanisms of labour. I approached this systematically according to Chalmers and Altman (1995). However, it was difficult to define inclusion criteria for a subject not limited to discovering the effectiveness of a treatment. I searched the MEDLINE database using the key words: 'descent', 'engagement', 'station', 'rotation', 'flexion', 'extension', and 'attitude' in conjunction with 'labour' or fetal head'.
I chose Myles Textbook for Midwives and Mayes Midwifery - A Textbook for Midwives for analysis and comparison with the literature I identified.
Both midwifery textbooks dedicate a separate section to the mechanism of labour. Analysis of these sections suggested the following framework: engagement and descent, rotation, flexion/extension, and movement through the pelvic outlet were the main movements described.
The studies obtained by this search were analysed and evaluated. Many of the articles identified in the midwifery journals contained anecdotal evidence only. They were only included if the evidence related to an argument made by some other identified research, because of the relative weakness of anecdotal evidence.
Descent is recognised in both textbooks as one of the main components of fetal movement through the birth canal. Methods of assessment for descent and engagement vary; abdominal palpation and vaginal examination are used. Little is known about their correlation which makes it difficult to generalise or to compare study findings.
Knight et al investigated the use of abdominal versus vaginal examination when assessing the feasibility of an instrumental delivery; their findings suggest that abdominal palpation is the superior method because moulding of the fetal head can suggest that descent is advanced, although the widest diameter of the fetal skull is still above the pelvic brim. However, these findings may not apply antenatally or in early labour when moulding has not yet occurred.
In the UK and Australia abdominal assessment is the preferred method. Most studies investigating the significance of engagement were US-based, using vaginal assessment.
All studies suggested that engagement in nulliparous women is less common antenatally than described in the textbooks (see table 1 ).
| authors | not engaged | n | time of assessment |
|---|---|---|---|
| Weekes & Flynn (1975) | 50% | 462 | 38-42 weeks |
| Takahashi & Suzuki (1982) | 75.4% | 175 | early labour |
| Kushtagi (1993) | 83% | 71 | admission to labour ward |
| Diegmann et al (1995) | 69% | 101 | early labour |
| Murphy et al (1998) | 78% | 132 | early labour |
| Roshanfekr et al (1999) | 71% | 803 | onset of active labour |
The significance of non-engagement remains uncertain; studies investigating this topic do not have consistent results. Takahashi and Suzuki (1982) and Diegmann et al (1995) found no significant differences in labour outcomes between nulliparous women with engaged or non-engaged fetal head at the onset of labour. Roshanfekr et al (1999) found a significant difference with 14% of those unengaged and 5% of those engaged undergoing caesarean section. Murphy et al (1998) differentiated between a 'floating' fetal head (station -3 or above); a 'dipping' one (-2 or -1 above the ischial spines); and an 'engaged' one. Significantly higher rates of caesarean section were only detected in the group with a 'floating' fetal head: 27 % versus 6.8%. The observed increase in caesarean section may be due to the Rosenthal-effect: many obstetricians believe that non-engagement may lead to problems during labour and they may have been more inclined to perform a caesarean section on these women.
Cephalo-pelvic disproportion, a major cause of non-engagement, now affects only about 1% of pregnant women in developed countries but historically it may have influenced obstetric belief which then spilled over into current practice. This interesting topic is beyond the scope of this article. Nevertheless, its influence on the topic of engagement needs to be acknowledged.
Friedman and Sachtleben (1976) looked at engagement/non-engagement in terms of descent rates, and partograms were developed on the basis of their work. However, they used average values and therefore their results do not present the range of normality.
An interesting study by Juntunen and Kirkinen (1994) investigated descent in the subgroup of grand multiparas. They found the station of the presenting part to be significantly higher than in the control group of multiparas during first stage, with 7% of fetal heads non-engaged at 7 cm of dilation, compared to none in the control group. All subjects had normal vaginal deliveries. In the light of this research, it appears important that midwives appreciate the range of normal descent that they may encounter in order to be able to assess the progress of labour.
The literature search revealed one aspect of the mechanism of labour that has been omitted by the textbooks: the movements of the sacroiliac joint in relation to the movements of the fetus. Borell and Fernstroms' (1957a) x-ray studies showed that the sacroiliac joint moved during labour in relation to the descent of the fetus, and that these movements were not brought about by a change of maternal position. They found that as the fetal head passes the pelvic inlet, i.e. at engagement, a movement of rotation occurs within the sacroiliac joint that increases the sagittal diameter of the pelvic inlet. At the time the fetus passes the pelvic outlet, this movement of rotation is reversed, increasing the sagittal diameter of the pelvic outlet (fig 1).
Fig 1: Movement of the sacrum to increase the pelvic inlet (a) and (b) movement of the sacrum to increase the pelvic outlet. From Palastanga N, Field D and Soames (1994). Anatomy and Human Movement: Structure and function (2nd edn) butterworth Heinemann p407
Ohlsen (1973) re-examined the radiographic plates taken by Borell and Fernstrom; he found them to be of good quality. He suggests that pressure of the fetal head contributes to what has been termed 'pelvic moulding', i.e. the bony parts of the birth canal gradually adapt their dimensions as the fetal head passes through them. His argument is that less pelvic moulding can be observed when the fetal head is small.
Until the invention of radiography much had been speculated about pelvic joint mobility and fetal movements through the pelvis. Borell and Fernstrom's numerous radiographic studies (1957b; 1957c; 1958; 1959 and 1967) are widely referenced and comprise the main evidence for pelvic joint mobility. Most of them have not been replicated by other researchers, and they would be impossible to repeat today for ethical reasons. All of Borell and Fernstrom's studies were undertaken using similar methods; utilising a specially constructed delivery table that permitted simultaneous x-rays to be taken laterally and antero-posteriorly.
One limitation lies in the studies' small sample sizes: from 21 to 40 women. Neither selection or exclusion criteria for the samples are specified, nor is it clear whether all women who entered the studies were included in the analysis. The fact that all women were in an unusual delivery position (dorsal recumbent with knees supported) further limits the generalisation of the studies. It is also possible that the circumstances in which these women had to give birth, i.e. complete restriction of movement, repeated examinations and lack of privacy may have influenced the birthing process.
In a theoretical exploration of the role of the uterine ligaments in the mechanism of labour, Tourn‚ (1985) suggests that the utero-sacral ligaments, which connect the uterus with the lower part of the sacrum, bring about the first rotational movement at engagement, aided by the pelvic floor muscles that insert into the coccyx.
The literature search revealed anecdotal evidence (Sutton, 1996a; 1996b), describing the same process as an 'opening of the back', which has been observed by birth attendants who attend mothers in upright or forward leaning birthing positions. In these positions the sacrum can be seen to move backwards during the final phase of labour. Sutton warns against the supine or semi-recumbent position. She argues that these would hinder sacral movements because of maternal weight resting on the sacrum.
This point was picked up by Gastaldo (1992) who pointed out that Borell and Fernstrom (1957c) have sometimes been misinterpreted: the maternal position in which they observed sacral movements has been reported as dorsal recumbent or even lithotomy. In fact, during labour they only filmed women in a 'modified' dorsal recumbent position, in which women were hanging by their knees with their buttocks slightly lifted. This position would have allowed the sacrum to move, in contrast to any other supine position where the woman's weight rests on the sacrum (Rydberg, 1954).
Borell and Fernstrom did not investigate sacral movements during labour in any other than the modified dorsal recumbent position. Hard evidence is lacking concerning the effect of other maternal positions on the mechanism of labour.
The literature search revealed little research that has tested the theory of why and how internal rotation of the fetal head and shoulders occurs. Rotation is generally interpreted as a fetal movement of adaptation to the available space in the pelvis. However, the evidence indicates that the soft parts play a greater role in effecting rotation than does the bony pelvis (Stoddart, Nicholson and Popham, 1994). Borell and Fernstrom concluded that the levator ani causes rotation, because of the level at which rotation usually occurs and because the small fetal head that passes through the pelvis with plenty of space around it, also rotates. Juntunen and Kirkinen (1994 ) found that in grand multiparae spontaneous vaginal delivery may occur without prior internal rotation (four out of 42 cases).
Obviously the soft parts would be more effected by repeated pregnancies than the bony pelvis. Stoddart et al, in a well-controlled randomised prospective study, showed that epidural anaesthesia effects rotation because it relaxes the pelvic floor.
Borell and Fernstrom's radiographic study of the rotation of the fetal shoulders during labour reveals that the shoulders rotate at a lower level than the fetal head: while the head rotates anteriorly at a level between the ischial spines and the ischial tuberosities in 75% of cases, the shoulders rotate to lie in the anterior-posterior diameter at a level 3-4 cm below the ischial tuberosities or at extrusion in 50 % or 60 % of cases. In the remaining cases the shoulders are born lying obliquely (30%) or transversely (5%). Borell and Fernstrom infer from these findings that the rotation of the shoulders is effected by a different mechanism than the rotation of the fetal head, i.e. not by the levator ani.
Borell and Fernstrom's findings agree with experiential evidence from an obstetric text written earlier in the century (Edgar, 1911). These suggested that although rotation of the shoulders usually occurs, it is not always the case. For midwives it is important to know that variability in the amount of rotation that the shoulders undergo exists, especially if their hands are guiding the head during the delivery as illustrated in the textbooks. Unnecessary strain on the maternal tissues or the baby's body could be caused by a forced rotation that would not have happened if the baby had been left to deliver unassisted.
Borell and Fernstrom's radiological investigations revealed that the fetal thoracic and cervical spine extend during the last stage of labour while the fetal head remains flexed until after expulsion. Unfortunately Myrfield et al (1997) did not consider this evidence in their theoretical clarification of the debate concerning how best to reduce perineal trauma. Myrfield et al analyse the mechanism of labour, referring to textbook descriptions. They propose that the fetus has to negotiate a 90-degree curve formed by the birth canal. They define the axis of the birth canal as a curve drawn at right angles to the planes of the pelvic inlet, its cavity and its outlet. If the bony pelvis is used in this way to define the course and shape of the birth canal, the 90-degree curve lies at the level of the ischial spines, and from here the fetus pivots around the pubic bone in order to emerge. This pivotal movement could only be achieved by extension. In their view the fetus extends his head and neck (no joints are specified), and this movement starts before crowning. Borell and Fernstrom (1959) found that the curve of the birth canal lies outside the bony pelvis and is entirely formed by the soft parts (figure 2). This contradicts the theory used by Myrfield et al that is widely found in textbooks, but not supported by evidence.
Fig 2: Selheim (1913, above) places the pivot point (X) well above that shown by Borrell and Fernstrom (1957b, below) (X - added by present editor)
Now that midwifery training has moved into higher education research based teaching is increasingly encouraged. Theory may need to be simplified to teach its main components to beginners, to help them form a knowledge base from which to explore research; however, if a theory is to be useful for midwives, its validity needs to be tested, traditional beliefs may need to be challenged.
Anatomical knowledge is often regarded as factual and static, with little room for change, and the mechanism of labour has been placed in this realm. Midwives need to be aware that in the l9th century obstetricians believed in the existence of a biological basis of social and moral order, and that this belief influenced the interpretation and presentation of their anatomical findings.
Tradition has overemphasised the rigidity of the bony pelvis. My literature review revealed research that pointed towards a significant role played by the soft parts in the mechanism of labour including the pelvic floor as well as other structures, such as the lower uterine segment (Borrell and Fernstrom, 1957a; Stoddart et al, 1994), the cervix (Olah and Neilson, 1994), the psoas muscles (Parsons, 1998) and the uterine ligaments (Tourné, 1985). In contrast, both Mayes and Miles overstress the importance of the bony pelvis. Midwives commonly use rigid pelves when teaching the mechanism of labour. These could be replaced by models that allow for pelvic joint mobility to be replicated, or include additional soft parts. Virtual reality computer programmes could be developed to model the mechanism of labour; these could be a new way of teaching as well as researching this complex process. The theory of the mechanism of labour is not just an academic question. Midwives' understanding of the mechanism of labour will affect how they practise. It will help them to decide how labour should be monitored, how progress should be assessed; it will influence the significance they ascribe to engagement, descent, rotation and attitude. The theory will also affect delivery technique. Understanding the mechanism of labour has direct implications for midwifery practice.
Myles Textbook for Midwives recommends investigating the causes of non-engagement in nulliparas after 38 weeks of gestation but research suggests that non-engagement is common so this advice is perhaps over cautious and a waste of resources. The latest research may be able to explain a discrepancy between abdominal and vaginal assessment of engagement; traditional doctrine cannot. Further research is needed. The expectation that non-engagement at the onset of labour is associated with problems may cause practitioners to intervene unnecessarily, e.g. by performing a caesarean section more readily. In the mean time differentiating between a 'floating', a 'dipping' and an 'engaged' fetal head as suggested by Murphy et al (1998) may be useful, with only the finding of a 'floating' head warranting concern.
The apparent scarcity of good quality research on the subject of what constitutes normal descent, begs the question of how and why this is monitored. Obviously the fetal head needs to descend to be born, therefore continuing descent must be an indicator of progress. Arrest of descent may indicate obstructed labour. Supposedly, midwives possess experiential knowledge of this process, and acquire the ability to judge what is normal. For the novice practitioner (midwife or doctor), more research on the subject could provide helpful knowledge to guide their practice. Knowledge of rare variations of the normal is also important, as indicated by Juntunen and Kirkinen's finding of a different descent patterns in grand multiparas (1994), in order to prevent unnecessary intervention.
Appreciation of sacroiliac joint movement in relation to descent may lead midwives to attribute a different importance to maternal postures adopted throughout labour. However, further research into the subject is needed. Meanwhile, postures where all of the mother's weight rests on her sacrum should be avoided, without restricting the mother's freedom to move and assume comfortable positions. In order to help raise midwives' awareness, the occurrence of sacroiliac joint movement in response to pressures of the descending fetal head should be included in textbook's depiction of the mechanism of labour.
Concerning rotation. Midwives' awareness should shift from overemphasising the role of the pelvis to increasing consideration of the soft parts, for example, when looking after women receiving epidural anaesthesia. More research is needed on the subject: for example, does previous serious perineal trauma affect rotation in subsequent labours? Anecdotal evidence suggests that osteopathy which manipulates the tone of the psoas or pelvic floor muscles during labour can facilitate vaginal delivery (personal communication, 1999). The theory of increasing flexion during fetal descent, and extension during delivery of the head forms the theoretical basis for the delivery technique described in both textbooks, widely used in the UK today. Borell and Fernstrom's research could be regarded as evidence that the fetal head remains in extreme flexion until it emerges, giving weight to the argument that maintaining flexion might reduce perineal trauma. However, the research is restricted to the supine maternal position, and may not apply to other positions.
Concerning the delivery of the shoulders. The textbooks are possibly confusing manipulation and the naturally occurring mechanism. The HOOP trial staff training video (NPEU, 1994) shows a case where the shoulders emerged simultaneously in a transverse position when they were allowed to deliver without manipulation. It is possible that midwives cause or exacerbate perineal damage when assuming that rotation has to take place and that the anterior shoulder emerges first, because they aim to facilitate these movements.
Considering the complexity of the subject and the period of time for which scientists have investigated it, the scarcity of recent good quality research is remarkable. This justifies the inclusion and weight attributed to Borell and Fernstrom's research could hardly be described as being 'recent'. A historical study investigating the origins and formulation of the mechanism of labour in relation to the wider historical context could provide another route towards evaluating the theory, and defining its place in contemporary midwifery practice. Comparing it with other theories about labour, revealing contradictions, would be a useful topic for future investigations.
Bendixen KD (1995). 'Historical representations of the pelvis', British Journal of Obstetrics and Gynaecology , 102 , 6, 442-444 In: MIDIRS Midwifery Digest, 5, 4, 390-391.
Bennett V R and Brown L K (eds) (1999). Myles Textbook for Midwives l3th edn, Churchill Livingstone, London.
Borell U and Fernstrom I (1957a). 'The movements at the sacroiliac joints and their importance to changes in the pelvic dimensions during parturition', Acta Obstetricia et Gynecologica Scandinavica , 36 , 42-57.
Borell U and Fernstrom I (1957b). 'Shape and course of the human birth canal - A radiographic study in the human', Acta Obstetricia et Gynecologica Scandinavica , 36 , 166-179.
Borell U and Fernstrom I (1957c). 'The movements in the mechanism of disengagement with special reference to the attitude of the foetal head', Acta Obstetricia et Gynecologica Scandinavica , 36 , 347.
Borell U and Fernstrom I (1958). 'Radiographic studies of the rotation of the foetal shoulders during labour', Acta Obstetricia et Gynecologica Scandinavica , 37 , Fasc l, 54-61.
Borell U and Fernstrom I (1959). 'Internal anterior rotation of the foetal head - A contribution to its explanation', Acta Obstetricia et Gynecologica Scandinavica , 38 , 103-109.
Borell U and Fernstrom I (1967). 'The mechanism of labour', Radiologic Clinics of North America, 5, 73-85.
Bryar R M (1995). Theory for midwifery practice , MacMillan Press, Basingstoke.
Chalmers I and Altman D G (eds) (1995). Systematic Reviews , BMJ Publishing Group, London.
Diegmann E K, Chez R A, Danclair W G (1995). 'Station in early labor in nulliparous women at term', Journal of Nurse Midwifery , 40 , 4, 382-385.
Edgar J C (1911). The practice of obstetrics , 3rd edn, Rebman, London, cited in: Mortimore V R and McNabb M (1997) 'A six-year retrospective analysis of shoulder dystocia and delivery of the shoulders', Midwifery , 1998, 14 , 162-173.
Friedmann E A and Sachtleben M R (1976). 'Station of the fetal presenting part IV: Arrest of descent in nulliparas', Obstetrics and Gynecology , 47 , 129-136.
Friedmann E A and Sachtleben M R (1965). 'Station of the fetal presenting part. I. Pattern of descent', American Journal of Obstetrics and Gynecology , 93 , 4, 522-529.
Friedmann EA and Sachtleben M R (1970). 'Station of the fetal presenting part. II. Slope of descent', American Journal of Obstetrics and Gynecology, 107 , 7, 1031-1034.
Gastaldo T D ( 1992). 'Squatting during labour - The significance of maternal position on pelvic outlet diameters' [correspondence]. Birth , 19 , 4, 230-232.
Gautreau G, osteopath, personal communication during lecture on 'Non-pharmacological pain relief in labour: manual techniques' on 9.4.1999.
Hillan E M ( 1999). 'Physiology and management of the second stage of labour' in: Bennett V R and Brown L K (eds) Myles Textbook for Midwives . l3th edn. Churchill Livingstone, London. pp 447-64.
Juntunen K and Kirkinen P (1994). 'Partogram of a grand multipara: different descent slope compared with an ordinary parturient', Journal of Perinatal Medicine , 22 , 3, 213-218.
Knight D, Newnham J P, McKenna M, Evans S (1993). 'A comparison of abdominal and vaginal examinations for the diagnosis of engagement of the fetal head', Australian and New Zealand Journal of Obstetrics and Gynaecology , 33 , 2, 154-158.
Kushtagi P (1993). 'Pattern of descent of foetal head in normal labour', Journal of the Indian Medical Association, 93 , 9, 336-339.
Morison C R (1971). 'The mechanism of labour. 1: Normal labour', The Nursing Mirror Midwives' Journal, Oct 15, 133 ,16, 34-38.
Morrin N A (1997). 'Midwifery care in the second stage of labour', in: Sweet BR and Tiran D (eds) Mayes' Midwifery - A textbook for midwives , l2th ed. Bailliere Tindall, London. pp 385-402.
Morris W I C (1955). 'Report of the Newcastle-upon-Tyne Obstetrical and Gynaecological Society - Shoulder Dystocia', Journal of Obstetrics and Gynaecology of the British Empire , 62 , 302-306 cited in: Mortimore V R and McNabb M (1997) 'A six-year retrospective analysis of shoulder dystocia and delivery of the shoulders', Midwifery , 1998, 14 , 162-173.
Murphy K, Shah L and Cohen W R (1998). 'Labor and delivery in nulliparous women who present with an unengaged fetal head', Journal of Perinatology, 8 , 2, 122-125.
Myrfield K, Brook C, Creedy D (1997). 'Reducing perineal trauma: implications of flexion and extension of the fetal head during birth', Midwifery , 13 , 197-201. National Perinatal Epidemiology Unit (1994).
HOOP - Hands On Or Poised - Care of the perineum at delivery (video). Oxford Medical Illustrations, Oxford.
Ohlsen H (1973). 'Moulding of the pelvis during labour', Acta Radiologica Diagnosis, 14 , 417-34.
Olah K S J and Neilson J P (1994). 'Failure to progress in the management of labour', British Journal of Obstetrics and Gynaecology , 101 , l, l-3.
Parsons C (1998). 'Damage to the pelvic floor: causes, prevention and treatment', Practising Midwife , l , 2, 14-16.
Rosenthal R (1976). Experimenter effects in behavioural research (enlarged edn.), Irvington, New York.
Roshanfekr D, Blakemore K, Lee J et al (1999). 'Station at onset of active labor in nulliparous patients and risk of cesarean delivery', Obstetrics and Gynecology , 93 , 3, 329-331.
Rydberg E (1954). The mechanism of labour, Thomas, London. cited in: Mortimore V R and McNabb M (1997) 'A six-year retrospective analysis of shoulder dystocia and delivery of the shoulders', Midwifery , 1998, 14 , 162-173.
Speert H (1996). Obstetric and Gynecologic Milestones Illustrated 2nd edn, The Parthenon Publishing Group, New York.
Stoddart A P, Nicholson K E and Popham P A (1994). 'Low dose bupivacaine/fentanyl epidural infusions in labour and mode of delivery', Anaesthesia , 49 , 12, 1087-1090.
Sutton J (1996a). 'Birth: medical emergency or engineering miracle? A midwifery approach to keeping birth normal', MIDIRS Midwifery Digest , 6 , 2, 170-173.
Sutton J (1996b). 'A midwife's observations of how the birth process is influenced by the relationship of the maternal pelvis and the foetal head', Journal of the Association of Chartered Physiotherapists in Women's Health , 79, 31-33. Reprinted in: MIDIRS Midwifery Digest , Jun 1997, 7 , 2, 189-191.
Sweet B with Tiran D (eds) (1997). Mayes' Midwifery-A Textbook forMidwives , l2th edn, Bailliere Tindall, London.
Takahashi K and Suzuki K (1982). 'Incidence and significance of the unengaged fetal head in nulliparas in early labor', International Journal of Biological Research in Pregnancy , 3 , 1, 8-9.
Tourné C E (1985). 'Mecanique obstetricale. Phase intermediaire et phase d'enfantement', Journal de Gynecologie, Obstetrique et Biologie de la Reproduction , 14 , 8, 965-972.
Weekes A R and Flynn M J (1975). 'Engagement of the fetal head in primigravidae and its relationship to duration of gestation and time of onset of labour', British Journal of Obstetrics and Gynaecology , 82 , 1, 7-11.
This article was originally published in Midwifery Matters ISSUE 88 Spring 2001, p3-8
LW updated February 4, 2005