probability that the valve lesion caused death is high. Finally, what is to be done when the heart is morphologically normal and there is no apparent cause of death? In many such cases the probable cause is a cardiac arrhythmia resulting from physiological rather than anatomical abnormalities. Electrocardiographic abnormalities -a long QT interval and pre-excitation (Wolff-Parkinson-White syndrome) -are associated with sudden death; but if death occurs in someone who has not had electrocardiography the heart will be apparently normal at necropsy and no cause of death will be found. Right ventricular tachycardias are also associated with sudden death; the heart may appear normal or have varying degrees of fibrous scarring.4 When inquiries are made about people who have dropped dead without cause many will be found to have had previous episodes of syncope or palpitations, or both. Some may have been suspected of having petit mal. The distinction between syncope induced by an arrhythmia and minor epilepsy is often difficult in life and may not be resolved at necropsy. After grand mal seizures have been excluded there is still an excess of sudden death in people with epilepsy.'01' Coronary artery spasm is a potential cause of acute myocardial ischaemia and thus sudden death.'2 The combination of a history of chest pain and normal arteries at necropsy may point to the diagnosis. When all these diagnoses have been considered there will still be some people with no history of chest pain, palpitations, or syncope and no cause of death shown by the most careful necropsy. The size of this problem in Britain is not known because there is no agreed nomenclature for categorising and recording the cases. Sudden death without apparent cause during sleep at night has a particularly high incidence in some migrant Asian communities worldwide. 1 14 Some coroners in England and Wales accept a diagnosis of "natural death-cause unascertained"; others do not. A pathologist may be tempted erroneously to ascribe death to ischaemic heart disease. Another temptation is to find minor abnormalities of the conduction system.'5 The proof that abnormalities such as nodoventricular connections cause death is tenuous because they may also be found in the hearts of young people who have died accidentally. A similar difficulty arises with myocardial bridging, in which an
epicardial coronary artery is covered by an external layer of myocardium. In some well documented clinical cases compression of the artery has been shown to cause ischaemia, but to ascribe death to bridging is difficult because of the frequency of the finding in control hearts.'6 It would help bereaved families-and scientific knowlegeif a category of sudden unexpected death syndrome was recognised. Many pathologists and coroners have arrived at this sensible point by mutual agreement. There are certain analogies with the sudden infant death syndrome. Both conditions are probably heterogeneous, and in both necropsy does not explain the cause of death. The existence of the category sudden infant death syndrome has, however, not only helped research but also given families a feeling that they understand, and can begin to accept, the death. M J DAVIES
Professor, St George's Hospital Medical School, London SW17 ORE 1 Thomas AC, Knapman PA; Krikler DM, Davies MJ. Community studv of the causes of "natural" sudden death. B.J 1988;297:1453-6. 2 Davies MJ. Anatomic features in victims of sudden coronary death. Circulation 1992;85:19-24. 3 McKenna WJ, Camm AJC. Suddeni death in hypertrophic cardiomyopathy-assessment of patients at high risk. Circulation 1989;80:1489-92. 4 Fontaine G, Frank R, Fontaliran F, Lascault G, Tonet J. Right ventricular tachycardias. In: Parmley WW, Chatterjee K, eds. Cardiolo'. Vol 1. Philadelphia: J B Lippincott, 1992:1-17. 5 Corrado D, Thiene G, Nava A, Rossi L, Pennelli N. Sudden death in young competitive athletes: clinicopathological correlations in 22 cases. Am3rMed 1990;89:588-96. 6 Lipsett J, Byard RW, Carpenter BF, Jimenez CL, Bourne AJ. Anomalous coronary arteries arising from the aorta associated with sudden death in infancy and childhood. Arch Pathol Lab Med
1991;115:770-3. 7 Mathieu D, Larde D, Vasile N. Primary dissecting aneurysm of the coronary arteries: case report and literature review. Cardiovasc Intervent Radiol 1984;7:71-4. 8 Boudoulas H, Kligfield P, Wooley CF. Mitral valve prolapse: sudden death. In: Boudoulas H, Wooley CF, ed. Mitral valve prolapse and the mitral valve prolapse syndrome. Mount Kisco, NY: Futstra, 1988:591-605. 9 Kligfield P, Devereux RB. Is the mitral valve prolapse patient at high risk of sudden death
identifiable? Cardiovasc Clin 1990;21:143-60. 10 Jay GW, Ieestrama JE. Sudden death in epilepsy: a comprehensive review of the literature and proposed mechanisms. Acta Neurol Scand [Suppl] 1981;82:63-6. 11 Brown SW, Mawer GE, Lawler W, Taylor DC, Shorvon S, Betts TA, et al. Sudden death and epilepsy. Lancet 1990;335:606-7. 12 Myerburg RJ, Kessler KM, Mallon SM, Cox MM, de Marchene E, Interian A, et al. Life threatening arrhythmias in patients with silent myocardial ischemia due to coronary artery spasm. N Eng1'lMed 1992;326:1451-5. 13 Kirscher RH, Eckner FAO, Baron RC. The cardiac pathology of sudden unexplained death in South East Asian refugees. JAMA 1986;256:2700-5. 14 Goh KT, Chao TC, Chew CH. Sudden nocturnal death (SUND) among Thai construction workers in Singapore. Lancet 1990;335:1154-5. 15 Bharati S, Lev M. Congenital abnormalities of the conduction system in sudden death in young adults. J Am Coll Cardiol 1986;8:1096-104. 16 Ferreira AG, Trotter SE, Konig B, Decourt LV, Fox K, Olsen EGI. Myocardial bridges: morphological and functional aspects. BrfHeartJ7 1991;66:364-7.
Reducing aortocaval compression: how much tilt is enough? Do as much as possible in the lateral position During late pregnancy the mother's adoption of the supine position allows the uterus to compress her aorta and vena cava. Such compression may produce potentially adverse physiological disturbances in both the mother and the fetus. Although initially thought to be a problem confined to the recumbent position, aortocaval compression has now also been shown in standing' and semirecumbent2 positions. The effects of such compression are exacerbated during regional anaesthesia and labour. Obstetricians have traditionally sought to reduce these disturbances by tilting the mother away from the supine position, but is the amount of tilt traditionally used enough? Around a tenth of women spontaneously develop the supine hypotensive syndrome or "revealed caval compresBMJ
VOLUME
305
5 SEPTEMBER 1992
sion." In others supine compression of the vena cava is concealed since there is no change in maternal blood pressure3'4 and only a small decrease in cardiac output.4 These women, however, remain more susceptible to hypotension from factors that impair normal circulatory control. The sympathetic blockade of regional anaesthesia prevents reflex venoconstriction and augments the deleterious effects of venous obstruction on cardiac output. Compression of the vena cava also makes maternal resuscitation from cardiac arrest much more difficult, and it may actually increase the fraction of intravenous bupivacaine delivered to the myocardium.5 The fetus too is at risk of reduced oxygen delivery during supine aortocaval compression because of decreased utero539
placental perfusion pressure and intervillous blood flow. The factors implicated include maternal systemic hypotension during the supine hypotensive syndrome or anaesthesia and compression of both the aorta and vena cava.3 Labour imposes an additional stress on the fetus as uterine contractions reduce uteroplacental blood flow. Supine aortic compression is augmented by contractions and potentiates this reduction in blood flow, although, paradoxically, compression of the vena cava is probably reduced during contractions, with less likelihood of the supine hypotensive syndrome. As well as causing maternal systemic hypotension regional anaesthesia might interact in other ways with aortocaval compression to affect the fetus. Firstly, a recent study during caesarean section under spinal anaesthesia showed that some women had a fall in cardiac output which was not reflected in maternal hypotension but was associated with fetal acidosis.6 Other indirect effects of regional anaesthesia during labour include uterine hypertonus and changes in uterine artery tone. Conflicting effects of epidural anaesthesia on uterine perfusion as assessed by Doppler ultrasound have emerged, with reports that arterial resistance may increase or decrease or show no change.7 Secondly, any direct effects of bupivacaine on the fetus and neonate may be augmented if aortocaval compression makes the fetus more acidotic, as more of the local anaesthetic transfers through the placenta under these conditions.8 Finally, epidural analgesia during labour may increase the likelihood of aortocaval compression by interfering with normal maternal behaviour. Women do not spontaneously lie supine during uncomplicated labour,9 but pain relief during contractions may permit them to take up positions that are usually intolerable. To avoid the problems of supine aortocaval compression the lateral position has been advocated for maternal management during labour, for fetal blood sampling, for establishing regional anaesthesia,"' and even for caesarean section under general anaesthesia. " For obstetricians, midwives, and anaesthetists, however, management is sometimes easier with the mother supine, and methods of reducing aortocaval compression while maintaining this position were introduced into clinical practice in the 1970s for epidural analgesia during labour and at caesarean section. Lateral table tilt and pelvic tilt, which are usually used in the United Kingdom, had better effects than the unmodified supine position.'2 Left tilt is normally preferable to right. The maximum amount of tilt is usually 15°-that is, the angle of the upper plane of the "Crawford wedge"-although a steeper angle has been used for the Cardiff resuscitation wedge,' constructed for use during maternal cardiopulmonary resuscitation. This is a rigid wedge with an upper plane angled at 270, the maximum amount of tilt found to be consistent with effective external cardiac massage.'3 There is evidence, however, that 150 of tilt is insufficient to relieve aortocaval compression completely. Both left and right pelvic tilt failed to reverse the decreased blood flow in the leg associated with the supine position,3 and at caesarean section cardiac output increased towards the value found in the left lateral position when left sided table tilt was supplemented with uterine displacement. '4 Aortic compression has recently been shown during labour with up to 340 of tilt,2 and even in the semilateral position leg blood flow is less than in the lateral
position.
540
What are the implications of these findings? Aortic compression is always concealed on routine maternal monitoring. Until we can identify the degree or effect of aortocaval compression in each woman-for example, by using specific monitoring of aortic compression2 or cardiac output0-a residual degree of compression should be suspected in the supine position even when pelvic or lateral tilt is used. When the mother is recumbent the lateral position should be used for her management during labour, especially if epidural analgesia is being used. The lateral position may, however, be inappropriate for caesarean delivery. Nevertheless, during preparation before final positioning for operation attention must be given to maintaining uteroplacental blood flow, most importantly for operations performed during labour2 or when fetal hypoxia is present. On the journey to the operating room the mother should be in the lateral position rather than simply tilted, and regional anaesthesia can easily, and should, be induced while the patient is lateral."' The woman should remain on her side until the last moment before the start of surgery. The infant should be delivered as quickly as possible, and during delivery the maximum amount of tilt consistent with the surgeon's ability to operate should be used. It may be necessary to reconsider the use of a side support to the operating table, as in the original description by Ansari et al,'2 to achieve this degree of tilt. Although more important for high risk and emergency cases, these measures should also be practised for elective procedures. Attempts to prevent maternal hypotension during regional anaesthesia should be vigorously promoted, with vasopressors rather than intravenous fluid'5 and eventually with drugs or combinations that produce anaesthesia without sympathetic blockade. S M KINSELLA Senior registrar in anaesthetics J G WHITWAM Professor and director, department of anaesthetics
Hammersmith Hospital, London W12 ONN J A D SPENCER Senior lecturer in obstetrics and gynaecology University College and Middlesex School of Medicine, London WC1E 6HX I Shneider KTM, Bollinger A, Huch A, Huch R. The oscillating "sena ca a syndrome" during quiet standing-an unexpected observation in late pregnancy. BrJ Obstes Gsnaecol 1984;91:766-80. 2 Kinsella SM, Whitwam JG, Spencer JAD. The use of the Finapres digital blood pressure instrument for monitoring aortic compression by the uterus. Brj Obstet Gynaecol 1990;97:700-5. 3 Kinsella SM, Lee A, Spencer JAD. MIaternal and fetal effects of the supine and pelv ic tilt positions in late pregnancy. Eurj Obstet Gynecol Reprod Biol 1990;36: 11-7. 4 Clark SL, Cotton DB, Pivarnik JM, Lee W, Hankins GDV, Benedetti TJ, et al. Position change and central hemodynamic profile during normal third-trimester pregnancy and post partum. Am
J Obstet Gvnecol 1991;164:883-7.
5 Kasten GW, Martin ST. Resuscitation from bupivacaine-induced cardiovascular toxicity during partial inferior vena cava occlusion. Anesth Analg 1986;65:341-4. 6 Robson SC, Boys RJ, Rodeck C, Morgan B. Maternal and fetal haemodynamic'effects of spinal and extradural anaesthesia for elective caesarean section. Brj Anaesth 1992;68:54-9. 7 Long MG, Price P, Spencer JAD. Uteroplacental perfusion after epidural analgcsia for elcctive caesarean section. Brj Obstet Gynaecol 1988;95:1081-2. 8 Gaylord DG, Carson RJ, Reynolds F. The effects of umbilical perfusate pH and controlled maternal hypotension on placental drug transfer in the rabbit. Anesth Analg 1990;71:42-8. 9 Carlson JM, Diehl JA, Sachtleben-Murray M, McRae M, Fenwick L, Friedtnan EA. Maternal position during parturition in normal labor. Obstet Gvnecol 1986;68:443-7. 10 Carrie LES. A plea for lateral thinking by obstetric anaesthetists. Anaesthesia 1989;44:444. 11 Waldron KW, Wood C. Cesarean section in the lateral position. Obstet Gvnecol 1971;37:706-10. 12 Ansari I, Wallace G, Clemetson CAB, Mallikarjuneswara VR, Clemetson CDM1. Tilt caesarean section. Journal of Obstetrics and Gvnaecology ofthe British Commonwealth 1970;77:713-21. 13 Rees GAD, Willis BA. Resuscitation in late pregnancy. Anaesthesia 1988;43:347-9. 14 Secher NJ, Arnsbo P, Heslet Andersen L, Thomsen A. Measurements of cardiac stroke volumc in various body positions in pregnancy and during caesarean section: a comparisotn between thermodilution and impedance cardiography. ScandJ GuCln Lab Invest 1979;39:569-76. 15 Epidural block for caesarean section and circulatory changes [editorial]. Lancet(. 1989;ii: 1()76-8.
BMJ
VOLUME
305
5
SEPTEMBER
1992
epicardial coronary artery is covered by an external layer of myocardium. In some well documented clinical cases compression of the artery has been shown to cause ischaemia, but to ascribe death to bridging is difficult because of the frequency of the finding in control hearts.'6 It would help bereaved families-and scientific knowlegeif a category of sudden unexpected death syndrome was recognised. Many pathologists and coroners have arrived at this sensible point by mutual agreement. There are certain analogies with the sudden infant death syndrome. Both conditions are probably heterogeneous, and in both necropsy does not explain the cause of death. The existence of the category sudden infant death syndrome has, however, not only helped research but also given families a feeling that they understand, and can begin to accept, the death. M J DAVIES
Professor, St George's Hospital Medical School, London SW17 ORE 1 Thomas AC, Knapman PA; Krikler DM, Davies MJ. Community studv of the causes of "natural" sudden death. B.J 1988;297:1453-6. 2 Davies MJ. Anatomic features in victims of sudden coronary death. Circulation 1992;85:19-24. 3 McKenna WJ, Camm AJC. Suddeni death in hypertrophic cardiomyopathy-assessment of patients at high risk. Circulation 1989;80:1489-92. 4 Fontaine G, Frank R, Fontaliran F, Lascault G, Tonet J. Right ventricular tachycardias. In: Parmley WW, Chatterjee K, eds. Cardiolo'. Vol 1. Philadelphia: J B Lippincott, 1992:1-17. 5 Corrado D, Thiene G, Nava A, Rossi L, Pennelli N. Sudden death in young competitive athletes: clinicopathological correlations in 22 cases. Am3rMed 1990;89:588-96. 6 Lipsett J, Byard RW, Carpenter BF, Jimenez CL, Bourne AJ. Anomalous coronary arteries arising from the aorta associated with sudden death in infancy and childhood. Arch Pathol Lab Med
1991;115:770-3. 7 Mathieu D, Larde D, Vasile N. Primary dissecting aneurysm of the coronary arteries: case report and literature review. Cardiovasc Intervent Radiol 1984;7:71-4. 8 Boudoulas H, Kligfield P, Wooley CF. Mitral valve prolapse: sudden death. In: Boudoulas H, Wooley CF, ed. Mitral valve prolapse and the mitral valve prolapse syndrome. Mount Kisco, NY: Futstra, 1988:591-605. 9 Kligfield P, Devereux RB. Is the mitral valve prolapse patient at high risk of sudden death
identifiable? Cardiovasc Clin 1990;21:143-60. 10 Jay GW, Ieestrama JE. Sudden death in epilepsy: a comprehensive review of the literature and proposed mechanisms. Acta Neurol Scand [Suppl] 1981;82:63-6. 11 Brown SW, Mawer GE, Lawler W, Taylor DC, Shorvon S, Betts TA, et al. Sudden death and epilepsy. Lancet 1990;335:606-7. 12 Myerburg RJ, Kessler KM, Mallon SM, Cox MM, de Marchene E, Interian A, et al. Life threatening arrhythmias in patients with silent myocardial ischemia due to coronary artery spasm. N Eng1'lMed 1992;326:1451-5. 13 Kirscher RH, Eckner FAO, Baron RC. The cardiac pathology of sudden unexplained death in South East Asian refugees. JAMA 1986;256:2700-5. 14 Goh KT, Chao TC, Chew CH. Sudden nocturnal death (SUND) among Thai construction workers in Singapore. Lancet 1990;335:1154-5. 15 Bharati S, Lev M. Congenital abnormalities of the conduction system in sudden death in young adults. J Am Coll Cardiol 1986;8:1096-104. 16 Ferreira AG, Trotter SE, Konig B, Decourt LV, Fox K, Olsen EGI. Myocardial bridges: morphological and functional aspects. BrfHeartJ7 1991;66:364-7.
Reducing aortocaval compression: how much tilt is enough? Do as much as possible in the lateral position During late pregnancy the mother's adoption of the supine position allows the uterus to compress her aorta and vena cava. Such compression may produce potentially adverse physiological disturbances in both the mother and the fetus. Although initially thought to be a problem confined to the recumbent position, aortocaval compression has now also been shown in standing' and semirecumbent2 positions. The effects of such compression are exacerbated during regional anaesthesia and labour. Obstetricians have traditionally sought to reduce these disturbances by tilting the mother away from the supine position, but is the amount of tilt traditionally used enough? Around a tenth of women spontaneously develop the supine hypotensive syndrome or "revealed caval compresBMJ
VOLUME
305
5 SEPTEMBER 1992
sion." In others supine compression of the vena cava is concealed since there is no change in maternal blood pressure3'4 and only a small decrease in cardiac output.4 These women, however, remain more susceptible to hypotension from factors that impair normal circulatory control. The sympathetic blockade of regional anaesthesia prevents reflex venoconstriction and augments the deleterious effects of venous obstruction on cardiac output. Compression of the vena cava also makes maternal resuscitation from cardiac arrest much more difficult, and it may actually increase the fraction of intravenous bupivacaine delivered to the myocardium.5 The fetus too is at risk of reduced oxygen delivery during supine aortocaval compression because of decreased utero539
placental perfusion pressure and intervillous blood flow. The factors implicated include maternal systemic hypotension during the supine hypotensive syndrome or anaesthesia and compression of both the aorta and vena cava.3 Labour imposes an additional stress on the fetus as uterine contractions reduce uteroplacental blood flow. Supine aortic compression is augmented by contractions and potentiates this reduction in blood flow, although, paradoxically, compression of the vena cava is probably reduced during contractions, with less likelihood of the supine hypotensive syndrome. As well as causing maternal systemic hypotension regional anaesthesia might interact in other ways with aortocaval compression to affect the fetus. Firstly, a recent study during caesarean section under spinal anaesthesia showed that some women had a fall in cardiac output which was not reflected in maternal hypotension but was associated with fetal acidosis.6 Other indirect effects of regional anaesthesia during labour include uterine hypertonus and changes in uterine artery tone. Conflicting effects of epidural anaesthesia on uterine perfusion as assessed by Doppler ultrasound have emerged, with reports that arterial resistance may increase or decrease or show no change.7 Secondly, any direct effects of bupivacaine on the fetus and neonate may be augmented if aortocaval compression makes the fetus more acidotic, as more of the local anaesthetic transfers through the placenta under these conditions.8 Finally, epidural analgesia during labour may increase the likelihood of aortocaval compression by interfering with normal maternal behaviour. Women do not spontaneously lie supine during uncomplicated labour,9 but pain relief during contractions may permit them to take up positions that are usually intolerable. To avoid the problems of supine aortocaval compression the lateral position has been advocated for maternal management during labour, for fetal blood sampling, for establishing regional anaesthesia,"' and even for caesarean section under general anaesthesia. " For obstetricians, midwives, and anaesthetists, however, management is sometimes easier with the mother supine, and methods of reducing aortocaval compression while maintaining this position were introduced into clinical practice in the 1970s for epidural analgesia during labour and at caesarean section. Lateral table tilt and pelvic tilt, which are usually used in the United Kingdom, had better effects than the unmodified supine position.'2 Left tilt is normally preferable to right. The maximum amount of tilt is usually 15°-that is, the angle of the upper plane of the "Crawford wedge"-although a steeper angle has been used for the Cardiff resuscitation wedge,' constructed for use during maternal cardiopulmonary resuscitation. This is a rigid wedge with an upper plane angled at 270, the maximum amount of tilt found to be consistent with effective external cardiac massage.'3 There is evidence, however, that 150 of tilt is insufficient to relieve aortocaval compression completely. Both left and right pelvic tilt failed to reverse the decreased blood flow in the leg associated with the supine position,3 and at caesarean section cardiac output increased towards the value found in the left lateral position when left sided table tilt was supplemented with uterine displacement. '4 Aortic compression has recently been shown during labour with up to 340 of tilt,2 and even in the semilateral position leg blood flow is less than in the lateral
position.
540
What are the implications of these findings? Aortic compression is always concealed on routine maternal monitoring. Until we can identify the degree or effect of aortocaval compression in each woman-for example, by using specific monitoring of aortic compression2 or cardiac output0-a residual degree of compression should be suspected in the supine position even when pelvic or lateral tilt is used. When the mother is recumbent the lateral position should be used for her management during labour, especially if epidural analgesia is being used. The lateral position may, however, be inappropriate for caesarean delivery. Nevertheless, during preparation before final positioning for operation attention must be given to maintaining uteroplacental blood flow, most importantly for operations performed during labour2 or when fetal hypoxia is present. On the journey to the operating room the mother should be in the lateral position rather than simply tilted, and regional anaesthesia can easily, and should, be induced while the patient is lateral."' The woman should remain on her side until the last moment before the start of surgery. The infant should be delivered as quickly as possible, and during delivery the maximum amount of tilt consistent with the surgeon's ability to operate should be used. It may be necessary to reconsider the use of a side support to the operating table, as in the original description by Ansari et al,'2 to achieve this degree of tilt. Although more important for high risk and emergency cases, these measures should also be practised for elective procedures. Attempts to prevent maternal hypotension during regional anaesthesia should be vigorously promoted, with vasopressors rather than intravenous fluid'5 and eventually with drugs or combinations that produce anaesthesia without sympathetic blockade. S M KINSELLA Senior registrar in anaesthetics J G WHITWAM Professor and director, department of anaesthetics
Hammersmith Hospital, London W12 ONN J A D SPENCER Senior lecturer in obstetrics and gynaecology University College and Middlesex School of Medicine, London WC1E 6HX I Shneider KTM, Bollinger A, Huch A, Huch R. The oscillating "sena ca a syndrome" during quiet standing-an unexpected observation in late pregnancy. BrJ Obstes Gsnaecol 1984;91:766-80. 2 Kinsella SM, Whitwam JG, Spencer JAD. The use of the Finapres digital blood pressure instrument for monitoring aortic compression by the uterus. Brj Obstet Gynaecol 1990;97:700-5. 3 Kinsella SM, Lee A, Spencer JAD. MIaternal and fetal effects of the supine and pelv ic tilt positions in late pregnancy. Eurj Obstet Gynecol Reprod Biol 1990;36: 11-7. 4 Clark SL, Cotton DB, Pivarnik JM, Lee W, Hankins GDV, Benedetti TJ, et al. Position change and central hemodynamic profile during normal third-trimester pregnancy and post partum. Am
J Obstet Gvnecol 1991;164:883-7.
5 Kasten GW, Martin ST. Resuscitation from bupivacaine-induced cardiovascular toxicity during partial inferior vena cava occlusion. Anesth Analg 1986;65:341-4. 6 Robson SC, Boys RJ, Rodeck C, Morgan B. Maternal and fetal haemodynamic'effects of spinal and extradural anaesthesia for elective caesarean section. Brj Anaesth 1992;68:54-9. 7 Long MG, Price P, Spencer JAD. Uteroplacental perfusion after epidural analgcsia for elcctive caesarean section. Brj Obstet Gynaecol 1988;95:1081-2. 8 Gaylord DG, Carson RJ, Reynolds F. The effects of umbilical perfusate pH and controlled maternal hypotension on placental drug transfer in the rabbit. Anesth Analg 1990;71:42-8. 9 Carlson JM, Diehl JA, Sachtleben-Murray M, McRae M, Fenwick L, Friedtnan EA. Maternal position during parturition in normal labor. Obstet Gvnecol 1986;68:443-7. 10 Carrie LES. A plea for lateral thinking by obstetric anaesthetists. Anaesthesia 1989;44:444. 11 Waldron KW, Wood C. Cesarean section in the lateral position. Obstet Gvnecol 1971;37:706-10. 12 Ansari I, Wallace G, Clemetson CAB, Mallikarjuneswara VR, Clemetson CDM1. Tilt caesarean section. Journal of Obstetrics and Gvnaecology ofthe British Commonwealth 1970;77:713-21. 13 Rees GAD, Willis BA. Resuscitation in late pregnancy. Anaesthesia 1988;43:347-9. 14 Secher NJ, Arnsbo P, Heslet Andersen L, Thomsen A. Measurements of cardiac stroke volumc in various body positions in pregnancy and during caesarean section: a comparisotn between thermodilution and impedance cardiography. ScandJ GuCln Lab Invest 1979;39:569-76. 15 Epidural block for caesarean section and circulatory changes [editorial]. Lancet(. 1989;ii: 1()76-8.
BMJ
VOLUME
305
5
SEPTEMBER
1992
