- Email: [email protected]
eclampsia in Enugu, Nigeria: a retrospective observational study
International Journal of Obstetric Anesthesia (2005) 14, 108–113 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijoa.2004.10.011
ORIGINAL ARTICLE
Maternal and fetal outcome of anaesthesia for caesarean delivery in preeclampsia/eclampsia in Enugu, Nigeria: a retrospective observational study U. V. Okafor, O. Okezie Departments of Anaesthesia and Obstetrics and Gynaecology, University of Nigeria Teaching Hospital, Enugu, Nigeria Background: Maternal mortality, for which preeclampsia is a major cause, is a problem in Nigeria. Accurate data are available for caesarean sections in the University of Nigeria Teaching Hospital, Enugu. We therefore studied the outcome of caesarean section among these high-risk patients. Method: We conducted a retrospective survey of hospital records of patients with preeclampsia/eclampsia who had caesarean delivery in this unit over a four-year span from July 1998 to June 2002. Results: There were 3926 deliveries and 4036 births (3611 live births), with 898 women (23%) delivered by caesarean section. Of these, 125 (14%) had preeclampsia/eclampsia, 103 (82.4%) presenting for emergency caesarean delivery and 22 (17.6%) elective. General anaesthesia was used in 116 patients (92.8%) and spinal in nine. The major indications for surgery were severe preeclampsia/eclampsia in patients with unfavourable cervix (68%), fetal distress/intrauterine growth restriction (7.2%) and previous caesarean section (6.4%). There were six maternal deaths, all with general anaesthesia, giving a case fatality rate of 5.2% of general anaesthetics or 4.8% of caesarean deliveries. The cause of death was anaesthetic in three patients, cerebrovascular accident and pulmonary oedema in two and intraoperative haemorrhage in one. There were 13 stillbirths and 10 neonatal deaths. Conclusion: Maternal and fetal mortality were high. Poverty, late presentation, lack of equipment and inexperienced management were major contributory factors. Use of spinal anaesthesia should be encouraged in view of recent favourable reviews and cheaper cost. 2004 Elsevier Ltd. All rights reserved. Keywords: Preeclampsia; Eclampsia; Caesarean section; Mortality; Nigeria
of both mother and fetus and may account for up to 80% of maternal deaths in some parts of the developing world.3 The leading causes of maternal mortality in Nigeria depend on cultural and socio-demographic patterns, and as there is no national database on maternal mortality, estimates are not precise. In parts of southern Nigeria the main causes of maternal mortality are obstetric haemorrhage, sepsis, obstructed labour and eclampsia4–6 and in the North they are eclampsia, obstetric haemorrhage and anaemia.7,8 Nigeria, with 37 000 maternal deaths per year, has the second highest number in the world after India (160 000/year).9 Maternal mortality rates vary from 497 per 100 000 live births in Anambra state in Southern Nigeria5 to 2420 per 100 000 live births in Kano state in Northern Nigeria.7 A review of maternal deaths in University of Nigeria Teaching Hospital (UNTH), Enugu, (1991–2000) gave a maternal mortality of 1046 per
INTRODUCTION Preeclampsia is defined as proteinuric hypertension developing after the 20th week of pregnancy and regressing after delivery.1 Preeclampsia is a disorder unique to human pregnancy and may involve the maternal cardiovascular, renal, coagulation and hepatic systems and is associated with increased fetal mortality and morbidity. It occurs in 5–10% of all pregnancies.2 Both severe preeclampsia and eclampsia can seriously endanger the life Accepted October 2004 U.V. Okafor, Department of Anaesthesia; O. Okezie, Department of Obstetrics and Gynaecology, University of Nigeria Teaching Hospital, Enugu, Nigeria. Correspondence to: DR U.V. Okafor, P.O. Box 1521, Enugu, Enugu State, Nigeria. Tel.: +23442253532, +2348036765372. E-mail: [email protected]. 108
Anaesthesia in preeclampsia in Enugu, Nigeria 109 100 000 with sepsis the predominant cause of mortality.4 Sepsis is often due to septic abortions which take their toll on our nation’s women.10 The delivery suite of our centre is attended by two consultant anaesthetists in addition to one senior medical officer, seven senior registrars and ten registrars, ten consultant obstetricians with eleven senior registrars, sixteen registrars and one senior house officer, and thirteen consultant paediatricians, fourteen senior registrars and fifteen registrars. These are rostered in units to cover the delivery suite. A senior registrar should have had at least 24 months of postgraduate training including a pass in the part one fellowship examination of a postgraduate college (Nigerian or West African postgraduate medical colleges or equivalent). They are also supposed to be well-versed in the management of preeclampsia/eclampsia. There are seventeen trained midwives (one chief nursing officer, one senior matron, six senior nursing officers and three nursing officers) covering the delivery suite in rotation. Antenatal care in UNTH, Enugu for normal parturients involves four-weekly visits up to the 28th week, then fortnightly visits till the 36th week and weekly visits from the 37th week till delivery. Preeclamptic patients are seen one- to two-weekly, depending on the disease severity, by the consultant until delivery. Severe preeclampsia is considered an obstetric emergency and if the patient is not in established labour, she is admitted to the antenatal ward. During each antenatal visit, haemoglobin measurement, urinalysis and a physical examination including blood pressure measurement are carried out. In severe preeclampsia, the following investigations are also performed: bedside clotting time, hourly urine output, full blood count (including platelets), blood urea, electrolytes and creatinine, uric acid, transaminases, plasma protein (albumin), calcium, magnesium, 24-h urine for protein, creatinine, calcium and an electrocardiogram. Urgent delivery is indicated regardless of the gestational age if any of the following are present: severe hypertension persisting after 24 h of treatment, liver dysfunction, signs of impeding eclampsia or evidence that the fetus is in jeopardy. The most expedient mode of delivery is often required in these patients, because the delivery of the feto-placental unit is believed to stem the progress of the disease.
PATIENTS AND METHODS The hospital records (case notes, labour ward and theatre records) of patients with preeclampsia/eclampsia who had caesarean delivery under anaesthesia at the UNTH, Enugu, Nigeria, from July 1998 to June 2002, a 4-year span, were retrospectively reviewed. Data collected in-
cluded the patients’ demographics, obstetric records, including maternal and fetal mortality, and the anaesthetic techniques used. In our centre, a parturient has mild preeclampsia when she presents with the following; a blood pressure of 140/90 mmHg on two occasions 6 h or more apart, or a rise of 30 mmHg systolic or 15 mmHg diastolic from mid-trimester values; proteinuria above (+) on two consecutive urine specimens and significant nondependent oedema. The senior registrar should be informed. For severe preeclampsia, the blood pressure is persistently above 160/110 mmHg and proteinuria above 5 g/24 h (+++) and symptoms of headache, blurring of vision, epigastric pain and oliguria. Fig. 1 shows the hospital protocol for management of these patients. Methyldopa is the drug of choice for control of hypertension in these patients. Nifedipine and b-blockers are added on the advice of physicians. Hydralazine is used to manage hypertensive crisis. It is given as 10-mg i.v. bolus injections 6-hourly if the diastolic blood pressure is P110 mmHg. Sometimes, hydralazine 20 mg is added to 500 mL of 5% dextrose and titrated according to the blood pressure response. The infusion is stopped if the diastolic blood pressure is below 110 mmHg. There is a protocol for the prophylactic use of magnesium sulphate but because of costs and unavailability, diazepam is commonly used instead. For patients who can afford it, magnesium is given by 6-g bolus i.v. over 15 min or by 3-g/h continuous infusion. This is accompanied by monitoring of patellar reflexes, respiratory rate, pulse, and arterial oxygen saturation. Intravenous diazepam is given as 40 mg in one litre of 5% dextrose at 30–60 drops per minute depending on the sedative effect on the patient. Eclamptic fits are aborted with intravenous diazepam with emphasis on maintenance of a Preeclampsia
yes
SEVERE DISEASE
MILD DISEASE Fetus mature?
No
Monitor until fetus mature or disease severe
Yes
DELIVER
Fig. 1 Management plan for patients with preeclampsia in UNTH, Enugu.
110 International Journal of Obstetric Anesthesia patent airway in addition to preventing injury to the patient, oxygenation with a facemask or in severe cases, intubation. Local guidelines for anaesthetic management of preeclampsia/eclampsia involve an initial assessment of the patient by the senior house officer or registrar who informs the senior registrar. The senior registrar may then decide to call the consultant in very severe cases when he/she considers consultant intervention necessary. General anaesthesia has been the most popular technique, because of the unavailability of epidural sets, and the previously held view that spinal anaesthesia may result in severe hypotension. Rapid sequence induction with Sellick’s manoeuvre and a relaxant technique are used. Sodium thiopentone 4–6 mg/kg and suxamethonium 1–2 mg/kg are used for induction and laryngoscopy. The pressor response to laryngoscopy is attenuated with i.v. lidocaine 1.5 mg/kg or hydralazine 5 mg 5 minutes before laryngoscopy. Anaesthesia is maintained with atracurium, halothane, nitrous oxide/oxygen, oxygen/air mixtures or 100% oxygen in patients with arterial oxygen saturations <95%. Analgesia is provided with pethidine, pentazocine, tramadol or more recently, fentanyl, depending on availability. Facilities for invasive monitoring are inadequate and rarely used for obstetric cases. Nevertheless, spinal anaesthesia is sometimes used in patients with platelet counts >100 · 109/L and normal clotting time (3–11 min), provided they are judged stable, that is the blood pressure is 6150/100 mmHg and proteinuria <1 g in 24 h. Either hyperbaric 5% lidocaine or 0.5% bupivacaine are used, with the aim of achieving a sensory block to T8, in order to avoid blocking the cardioaccelerator nerves (T1-T4). A moderate preload of 5% dextrose is given and ephedrine is used to reverse hypotension. Postoperatively, stable patients are nursed in postnatal wards equipped with oxygen cylinders, but critically ill patients (four in this series) in need of organ support are nursed in the intensive care unit. Postoperative pain relief is provided with parenteral opioids like pethidine, tramadol or pentazocine.
Table 1. Numbers of deliveries and patients with preeclampsiaeclampsia, and numbers delivered by caesarean section Total deliveries (n) Total number of births Total number of live births Number of mothers with preeclampsia/ eclampsia Total number of caesarean sections among all deliveries Number with preeclampsia/eclampsia Mild preeclampsia Severe preeclampsia Eclampsia
3926 4036 3611 176 (4.5%) 898 (23%) 125 (13.9% of CS) 34 (27%) (1 death) 76 (61%) (5 deaths) 15 (12%)
Table 2. Age, parity and gestational age of preeclamptic/eclamptic patients presenting for caesarean delivery Number (%) (total number: 125) Age range (years)
619 20–25 26–30 31–35 36–40 P41
4 (3.2%) 19 (15.2%) 47 (37.6%) 31 (24.8%) 23 (18.4%) 1 (0.8%)
Parity
Nulliparous 1 2 3 4 P5
71 (56.8%) 21 (16.8%) 12 (9.6%) 10 (8%) 6 (4.8%) 5 (4%)
Gestational age at delivery (weeks)
>26 weeks 26–31weeks 32–36 weeks 37–42 weeks >42 weeks
Nil 22 (17.6%) 47 (37.6%) 54 (43.2%) 2 (1.6%)
The indications for caesarean section are given in Table 3. One hundred and three (82.4%) presented for emergency caesarean section, while 22 (17.6%) were counted as elective. The latter patients were not in labour, but presented with previous caesarean section (8), placenta praevia (3), elderly primip/bad obstetric history (3), intrauterine growth restriction only (5), post-dates (1), severe preeclampsia with Table 3. Indications for caesarean delivery
RESULTS The total number of deliveries in the four years of the study, the numbers of caesarean sections, the numbers with preeclampsia/eclampsia, the numbers of patients with preeclampsia/eclampsia delivered by caesarean section and the severity of disease, are shown in Table 1. Ninety patients (72%) had been booked for antenatal care, while 35 (28%) were unbooked. The average age of these patients was 30.2 years. Table 2 shows their age distribution, parity and gestational age at delivery.
Indication Severe preeclampsia/eclampsia with unfavourable cervix Fetal distress/intrauterine growth restriction Previous caesarean section Prolonged labour/poor progress Failed induction of labour/failed vacuum extraction Antepartum haemorrhage Elderly primip/bad obstetric history Premature rupture of membrane Post dates Big baby
Number (%) 85 (68%) 9 8 6 6
(7.2%) (6.4%) (4.8%) (4.8%)
4 3 2 1 1
(3.2%) (2.4%) (1.6%) (0.8%) (0.8%)
Anaesthesia in preeclampsia in Enugu, Nigeria 111 maternal distress (1) and big baby (1). There were seven twin pregnancies (132 babies but only 119 live births). General anaesthesia was used in 116 patients (92.8%) and spinal anaesthesia in the remaining 9 (7.2%). In non-preeclamptic patients, spinal anaesthesia was used in an even smaller percentage of caesarean deliveries (1.2%). There were no deaths among patients who received spinal anaesthesia. There were six deaths during general anaesthesia giving a case fatality rate of 5.2% of general anaesthetics or 4.8% of caesarean sections. The six deaths represented 55.5% of all maternal deaths following caesarean delivery during the period under review. The indication for surgery in five of the patients who died was severe preeclampsia, while one patient had mild preeclampsia complicated by placenta praevia and asthma. The deaths were due to complications of anaesthesia in three patients, preeclampsia/eclampsia (cerebrovascular accident and pulmonary oedema ) in two patients and intraoperative haemorrhage in the patient with placenta praevia and asthma. All patients with eclampsia survived. There were 13 stillbirths and 10 neonatal deaths, giving a perinatal mortality rate in this population of 174/ 1000 births.
DISCUSSION The caesarean section rate among all mothers during the four years of this study was 23%. This is higher than the rates of 18 and 20% reported recently for northern Nigeria,11,12 and reflects an overall increase in the caesarean section rate in Nigeria in the past decade.11,12 Iyaiya et al. from Illorin, Nigeria,12 reported that 9.3% of the parturients presenting for caesarean delivery had preeclampsia or eclampsia. In this study, preeclampsia and eclampsia patients represented 14% of parturients who presented for caesarean delivery and 4.5% of all deliveries. The caesarean section rate among our patients was 71%, which is greater than other reported caesarean section rates among such patients.12–14 Ozumba and Okaehialam14 reported in 1993 that 42% of preeclamptic/eclamptic patients managed in our hospital were delivered by caesarean section. Clearly this figure has since become even higher. Severe maternal and fetal complications often make urgent delivery necessary, and caesarean delivery is the expedient mode in most of these patients. It has been reported from Africa that operative and instrumental delivery may reduce maternal and perinatal mortality.15 In this study, however, the mortality rate, at about 5%, was higher among these high-risk patients than in the total population.
General anaesthesia was used in the great majority of the patients, while only 7.2% delivered under spinal anaesthesia. Epidural anaesthesia, the preferred choice of many anaesthetists, was not used due to the absence of usable epidural sets during the study period. Epidural sets have since become available. Epidural anaesthesia has for long been considered the technique of choice in preeclampsia, in part because it is known to increase intervillous blood flow and its slow onset helps in controlling the blood pressure.16 Recent studies have shown that both general17,18 and spinal anaesthesia19–21 are generally safe if properly conducted. Spinal anaesthesia is known to produce more consistent anaesthesia than epidural, and preeclamptics are actually less prone to hypotension than normal individuals.22,23 There were no maternal deaths among those receiving spinal anaesthesia in this study, but it must be acknowledged that numbers were very small. Fenton et al. in their study of 8070 caesarean sections in Malawi, showed that spinal anaesthesia may be safer than general,24 but a randomised study would be needed to ascertain this. Although not all the deaths in these high-risk patients were attributed solely to anaesthesia, three deaths were directly related to anaesthetic complications, two deaths due to cerebrovascular accident and pulmonary oedema, and one death due to intraoperative haemorrhage. Some of these deaths might have been prevented by the use of invasive monitoring, including central venous pressure, blood pressure and pulmonary artery pressure monitoring. Drugs like alfentanil, sufentanil and esmolol may obtund the pressor response to intubation better than either lidocaine or hydralazine, which causes a reflex maternal tachycardia. Spinal anaesthesia in stable patients avoids some of these complications, with the added benefit of a relatively simpler technique and cheaper cost. It also requires less expensive monitoring than does safe delivery of general anaesthesia. Cost is important in an environment where the cost of a disposable epidural set is about 25% of the minimum monthly wage for a government employee. Because of inadequate monitoring facilities, irregular supply of general anaesthetic drugs, patient poverty, and the complications of general anaesthesia, some authors consider spinal anaesthesia a preferable alternative to epidural or general anaesthesia,25,26 particularly in West Africa, because of its cheaper cost. Spinal anaesthesia was used successfully in six patients with severe preeclampsia and one with eclampsia, and may also benefit elective patients. An even smaller proportion of non-preeclamptic parturients (1.2%) delivered under regional anaesthesia during this study period. General anaesthesia has hitherto been the popular technique for caesarean delivery in this centre, where most patients present as emergencies and drugs like ketamine are cheap and easy to use and
112 International Journal of Obstetric Anesthesia relatively safe for most non-hypertensive parturients. However, we now emphasise regional techniques, unless there are contraindications to their use or time constraints. The case fatality rate in this study (4.8%) is higher than the case fatality rate reported from the United States for preeclampsia/eclampsia (6.4 per 10 000),27 but lower than that reported from six sub-Saharan African countries (18.4%).28 The latter studies were carried out in the capital cities of Niger, Mauritania, Burkina Faso, Ivory Coast and Mali and in two towns and a major city in Senegal.28 A report from Ethiopia put the case fatality rate at 21%.15 Contributory factors in the high rate in this study may be: Poor remuneration of staff, leading to a flight of manpower overseas.29–31 This department lost six consultant anaesthetists to the Gulf States, Europe and better paying jobs in Nigeria since the late 1980s. Poverty among patients and late presentation to the hospital.32,33 The fact that 69 deliveries (55.2%) were pre-term underlines the severity of the disease. Extensive use of general anaesthesia despite grossly inadequate supply of drugs and monitoring facilities. Haemodynamic monitoring equipment, capnography and pulse oximetry, and drugs such as alfentanil, esmolol and fentanyl are rarely available. Hospital administrators need to be better informed on the importance of anaesthesia in modern medicine, to persuade them to allocate more funds to meet these needs. Five of the 104 patients managed by trainee anaesthetists died in this study. There was one fatality in the 21 cases managed by the consultants. Although this gives a similar fatality rate, it must be remembered that consultants manage the sickest patients, cases considered beyond the scope of senior registrars. All the deaths were intraoperative. The 13 stillbirths and 10 neonatal deaths in this study gave a perinatal mortality rate of 174/1000 births. Preeclampsia and eclampsia are a leading cause of fetal loss especially in the third world, due to placental abruption, placental insufficiency and birth asphyxia.
CONCLUSION While maternal and fetal mortality following anaesthesia for caesarean delivery in patients with preeclampsia/eclampsia has fallen over the years in the Western world, the figures in this study show that it remains a major cause of loss of life in West Africa. Poverty, late presentation of patients, lack of equipment and drugs and inexperienced staff were probably major contributory factors. A case can be made for the use of spinal anaesthesia in stable patients in view of recent favourable reviews and the cheaper cost. Further studies in this
and other Nigerian hospitals will help present a clearer picture. Finally, a multidisciplinary approach is vital, with close co-operation between the anaesthetist, obstetrician and midwife. The preeclampsia protocol which was developed by Robson et al.34, has been adopted by many obstetric units and can be modified to suit individual units. ACKNOWLEDGEMENT We thank the staff of the records department, labour ward and obstetric theatre of the University of Nigeria teaching hospital, Enugu for their co-operation during this study. We also extend our gratitude to Miss Mary Nwodo who graciously typed the manuscript.
REFERENCES 1. Mushambi M C, Halligan A W, Williamson K. Recent development in the pathophysiology and management of pre-eclampsia. Br J Anaesth 1996; 76: 133–148. 2. Macgillivary J. Some observations in the incidence of preeclampsia. Br J Obstet Gynaecol 1958; 65: 536–539. 3. Conz P A, Catalano C. Pathophysiology of preeclampsia. G Ital Nefrol 2003; 20: 15–22. 4. Okaro J M, Umezulike A C, Onah H E, Chukwuali L I, Ezegwu O F, Nweke P C. Maternal mortality at the University of Nigeria Teaching Hospital, Enugu. Before and After Kenya. Afr J Reprod Health 2001; 5: 90–97. 5. Chukwudebelu W O, Ozumba B C. Maternal mortality in Anambra State of Nigeria. Int J Gynaecol Obstet 1988; 27: 365–370. 6. Ariba A J, Inem A V, Biersack G, Aina A O, Ayankagbe O O, Adetoro O O. Pattern of obstetric mortality in a voluntary agency hospital in Abeokuta, Southwest Nigeria. Nigerian Med Practitioner 2004; 45: 83–90. 7. Adamu Y M, Salihu H M, Sathiakumar N, Alexander G R. Maternal mortality in Northern Nigeria; a population based study. Eur J Obstet Gynecol Reprod Biol 2004; 109: 153–159. 8. Wall L L. Dead mothers and injured wives: the social context of maternal morbidity and mortality among the Hausa of northern Nigeria. Stud Fam Planning 1998; 29: 341–359. 9. Maternal mortality in 2000 – Estimates developed by WHO, UNICEF and UNFDA. 2003. 10. Ikechebelu J I, Okoh C C. Morbidity and mortality following induced abortions in Nnewi, Nigeria. Trop Doct 2003; 33: 170–172. 11. Aisien A O, Lawson J O, Adebayo A. Caesarean section in a Northern Nigeria University Teaching Hospital. Niger Postgrad Med J 2002; 9: 146–150. 12. Iyaiya M A, Aboyeji P A. Caesarean delivery: the trend over a ten-year period at Illorin, Nigeria. Nig J Surg Res 2001; 3: 11–18. 13. Grau T, Conradi R, Wacker J, Martin E. Preeclampsia and anaesthesia. Zentralbl Gynakol 1999; 21: 627–630. 14. Ozumba B C, Okeahlaham M. Incidence and complication of preeclampsia at the University of Nigeria Teaching Hospital (UNTH), Enugu, Nigeria. Orient J Med 1993; 5: 136–138. 15. Yifru B, Ahmed A. Emergency obstetric performance with emphasis on operative delivery: does it reflect the quality of care. Ethiop J Health Dev 2004; 18: 96–106. 16. O’ Sullivan G. Medical aspects of obstetrics. In: Prys-Roberts C, Brown Jr B R, eds. International Practice of Anaesthesia. Oxford: Butterworth-Heinemann; 1996: 2/95/1-20. 17. Moodley T, Jjuuko G, Rout C. Epidural compared with general anaesthesia for caesarean delivery in conscious women with eclampisa. BJOG 2001; 108: 378. p. 378–382. 18. Eldridge E. Obstetric anaesthesia and analgesia. In: Allman K G, Wilson I H, eds. Oxford Handbook of Anaesthesia. Oxford: Oxford University Press; 2001: 695–756.
Anaesthesia in preeclampsia in Enugu, Nigeria 113 19. Gogarten W. Spinal anaesthesia for obstetrics. Best Pract Res Clin Anaesthesiol 2003; 17: 377–392. 20. Hood D D, Curry R. Spinal vs epidural anesthesia for cesarean section in severely preeclamptic patients; a retrospective survey. Anesthesiology 1999; 90: 1276–1282. 21. Ahmed S M, Khan R M, Bano S, Ajmani P, Kumar A. Is spinal anaesthesia safe in preeclamptic toxaemia patients? J Indian Med Assoc 1999; 97: 165–168. 22. Brodie H, Malinow A M. Anesthetic management of preeclampsia/ eclampsia. Int J Obstet Anesth 1999; 8: 110–124. 23. Aya A G, Mangin R, Vialles N, et al. Patients with severe preeclampsia experience less hypotension during spinal anesthesia for elective cesarean delivery than healthy parturients: a prospective cohort comparison. Anesth Analg 2003; 97: 867–872. 24. Fenton P M, Whitty C J M, Reynolds F. Caesarean section in Malawi: prospective study of early maternal and perinatal mortality. BMJ 2003; 327: 587–591. 25. Farragher R, Datta S. Recent advances in obstetric anesthesia. J Anesth 2003; 17: 30–41. 26. Shifman E M, Filippovich G V. Safety of subarachnoid anesthesia during cesarean section in pregnant women with preeclampsia. Anesteziol Reanimatol 2003; 5: 38–41.
27. Mackay A P, Berg C J, Atrash H K. Pregnancy related mortality from preeclampsia and eclampsia. Obstet Gynecol 2001; 97: 533–538. 28. Prual A, Bouvier-Colle M H, de Bernis L, Breart G. Severe maternal morbidity from direct obstetric causes in West Africa: incidence and case fatality rates. Bull World Health Organ 2000; 78: 593–602. 29. Stillwell B, Diallo K, Zurn P, et al. Migration of health workers from developing countries; strategic approaches to its management. Bull World Health Organ 2004; 85: 595–600. 30. Raufu A. Nigeria health authorities worry over exodus of doctors and nurses. BMJ 2002; 325: 65. 31. Loewenson R. Structural adjustment and health policy in Africa. International Journal of Health Services 1993; 23: 717–730. 32. Conde-Agudelo A, Kafury-Goeta A C. Case-control study of risk factors for complicated eclampsia. Obstet Gynecol 1997; 90: 172–175. 33. Haelterman E, Qvist R, Barlow P, Alexander S. Social deprivation and access to care as risk factors for severe preeclampsia. Eur J Obstet Gynecol Reprod Biol 2003; 111: 25–32. 34. Robson S C, Redfern N, Walkinshaw S A. A protocol for the intrapartum management of severe preeclampsia. Int J Obstet Anesth 1992; 1: 222–229.
ORIGINAL ARTICLE
Maternal and fetal outcome of anaesthesia for caesarean delivery in preeclampsia/eclampsia in Enugu, Nigeria: a retrospective observational study U. V. Okafor, O. Okezie Departments of Anaesthesia and Obstetrics and Gynaecology, University of Nigeria Teaching Hospital, Enugu, Nigeria Background: Maternal mortality, for which preeclampsia is a major cause, is a problem in Nigeria. Accurate data are available for caesarean sections in the University of Nigeria Teaching Hospital, Enugu. We therefore studied the outcome of caesarean section among these high-risk patients. Method: We conducted a retrospective survey of hospital records of patients with preeclampsia/eclampsia who had caesarean delivery in this unit over a four-year span from July 1998 to June 2002. Results: There were 3926 deliveries and 4036 births (3611 live births), with 898 women (23%) delivered by caesarean section. Of these, 125 (14%) had preeclampsia/eclampsia, 103 (82.4%) presenting for emergency caesarean delivery and 22 (17.6%) elective. General anaesthesia was used in 116 patients (92.8%) and spinal in nine. The major indications for surgery were severe preeclampsia/eclampsia in patients with unfavourable cervix (68%), fetal distress/intrauterine growth restriction (7.2%) and previous caesarean section (6.4%). There were six maternal deaths, all with general anaesthesia, giving a case fatality rate of 5.2% of general anaesthetics or 4.8% of caesarean deliveries. The cause of death was anaesthetic in three patients, cerebrovascular accident and pulmonary oedema in two and intraoperative haemorrhage in one. There were 13 stillbirths and 10 neonatal deaths. Conclusion: Maternal and fetal mortality were high. Poverty, late presentation, lack of equipment and inexperienced management were major contributory factors. Use of spinal anaesthesia should be encouraged in view of recent favourable reviews and cheaper cost. 2004 Elsevier Ltd. All rights reserved. Keywords: Preeclampsia; Eclampsia; Caesarean section; Mortality; Nigeria
of both mother and fetus and may account for up to 80% of maternal deaths in some parts of the developing world.3 The leading causes of maternal mortality in Nigeria depend on cultural and socio-demographic patterns, and as there is no national database on maternal mortality, estimates are not precise. In parts of southern Nigeria the main causes of maternal mortality are obstetric haemorrhage, sepsis, obstructed labour and eclampsia4–6 and in the North they are eclampsia, obstetric haemorrhage and anaemia.7,8 Nigeria, with 37 000 maternal deaths per year, has the second highest number in the world after India (160 000/year).9 Maternal mortality rates vary from 497 per 100 000 live births in Anambra state in Southern Nigeria5 to 2420 per 100 000 live births in Kano state in Northern Nigeria.7 A review of maternal deaths in University of Nigeria Teaching Hospital (UNTH), Enugu, (1991–2000) gave a maternal mortality of 1046 per
INTRODUCTION Preeclampsia is defined as proteinuric hypertension developing after the 20th week of pregnancy and regressing after delivery.1 Preeclampsia is a disorder unique to human pregnancy and may involve the maternal cardiovascular, renal, coagulation and hepatic systems and is associated with increased fetal mortality and morbidity. It occurs in 5–10% of all pregnancies.2 Both severe preeclampsia and eclampsia can seriously endanger the life Accepted October 2004 U.V. Okafor, Department of Anaesthesia; O. Okezie, Department of Obstetrics and Gynaecology, University of Nigeria Teaching Hospital, Enugu, Nigeria. Correspondence to: DR U.V. Okafor, P.O. Box 1521, Enugu, Enugu State, Nigeria. Tel.: +23442253532, +2348036765372. E-mail: [email protected]. 108
Anaesthesia in preeclampsia in Enugu, Nigeria 109 100 000 with sepsis the predominant cause of mortality.4 Sepsis is often due to septic abortions which take their toll on our nation’s women.10 The delivery suite of our centre is attended by two consultant anaesthetists in addition to one senior medical officer, seven senior registrars and ten registrars, ten consultant obstetricians with eleven senior registrars, sixteen registrars and one senior house officer, and thirteen consultant paediatricians, fourteen senior registrars and fifteen registrars. These are rostered in units to cover the delivery suite. A senior registrar should have had at least 24 months of postgraduate training including a pass in the part one fellowship examination of a postgraduate college (Nigerian or West African postgraduate medical colleges or equivalent). They are also supposed to be well-versed in the management of preeclampsia/eclampsia. There are seventeen trained midwives (one chief nursing officer, one senior matron, six senior nursing officers and three nursing officers) covering the delivery suite in rotation. Antenatal care in UNTH, Enugu for normal parturients involves four-weekly visits up to the 28th week, then fortnightly visits till the 36th week and weekly visits from the 37th week till delivery. Preeclamptic patients are seen one- to two-weekly, depending on the disease severity, by the consultant until delivery. Severe preeclampsia is considered an obstetric emergency and if the patient is not in established labour, she is admitted to the antenatal ward. During each antenatal visit, haemoglobin measurement, urinalysis and a physical examination including blood pressure measurement are carried out. In severe preeclampsia, the following investigations are also performed: bedside clotting time, hourly urine output, full blood count (including platelets), blood urea, electrolytes and creatinine, uric acid, transaminases, plasma protein (albumin), calcium, magnesium, 24-h urine for protein, creatinine, calcium and an electrocardiogram. Urgent delivery is indicated regardless of the gestational age if any of the following are present: severe hypertension persisting after 24 h of treatment, liver dysfunction, signs of impeding eclampsia or evidence that the fetus is in jeopardy. The most expedient mode of delivery is often required in these patients, because the delivery of the feto-placental unit is believed to stem the progress of the disease.
PATIENTS AND METHODS The hospital records (case notes, labour ward and theatre records) of patients with preeclampsia/eclampsia who had caesarean delivery under anaesthesia at the UNTH, Enugu, Nigeria, from July 1998 to June 2002, a 4-year span, were retrospectively reviewed. Data collected in-
cluded the patients’ demographics, obstetric records, including maternal and fetal mortality, and the anaesthetic techniques used. In our centre, a parturient has mild preeclampsia when she presents with the following; a blood pressure of 140/90 mmHg on two occasions 6 h or more apart, or a rise of 30 mmHg systolic or 15 mmHg diastolic from mid-trimester values; proteinuria above (+) on two consecutive urine specimens and significant nondependent oedema. The senior registrar should be informed. For severe preeclampsia, the blood pressure is persistently above 160/110 mmHg and proteinuria above 5 g/24 h (+++) and symptoms of headache, blurring of vision, epigastric pain and oliguria. Fig. 1 shows the hospital protocol for management of these patients. Methyldopa is the drug of choice for control of hypertension in these patients. Nifedipine and b-blockers are added on the advice of physicians. Hydralazine is used to manage hypertensive crisis. It is given as 10-mg i.v. bolus injections 6-hourly if the diastolic blood pressure is P110 mmHg. Sometimes, hydralazine 20 mg is added to 500 mL of 5% dextrose and titrated according to the blood pressure response. The infusion is stopped if the diastolic blood pressure is below 110 mmHg. There is a protocol for the prophylactic use of magnesium sulphate but because of costs and unavailability, diazepam is commonly used instead. For patients who can afford it, magnesium is given by 6-g bolus i.v. over 15 min or by 3-g/h continuous infusion. This is accompanied by monitoring of patellar reflexes, respiratory rate, pulse, and arterial oxygen saturation. Intravenous diazepam is given as 40 mg in one litre of 5% dextrose at 30–60 drops per minute depending on the sedative effect on the patient. Eclamptic fits are aborted with intravenous diazepam with emphasis on maintenance of a Preeclampsia
yes
SEVERE DISEASE
MILD DISEASE Fetus mature?
No
Monitor until fetus mature or disease severe
Yes
DELIVER
Fig. 1 Management plan for patients with preeclampsia in UNTH, Enugu.
110 International Journal of Obstetric Anesthesia patent airway in addition to preventing injury to the patient, oxygenation with a facemask or in severe cases, intubation. Local guidelines for anaesthetic management of preeclampsia/eclampsia involve an initial assessment of the patient by the senior house officer or registrar who informs the senior registrar. The senior registrar may then decide to call the consultant in very severe cases when he/she considers consultant intervention necessary. General anaesthesia has been the most popular technique, because of the unavailability of epidural sets, and the previously held view that spinal anaesthesia may result in severe hypotension. Rapid sequence induction with Sellick’s manoeuvre and a relaxant technique are used. Sodium thiopentone 4–6 mg/kg and suxamethonium 1–2 mg/kg are used for induction and laryngoscopy. The pressor response to laryngoscopy is attenuated with i.v. lidocaine 1.5 mg/kg or hydralazine 5 mg 5 minutes before laryngoscopy. Anaesthesia is maintained with atracurium, halothane, nitrous oxide/oxygen, oxygen/air mixtures or 100% oxygen in patients with arterial oxygen saturations <95%. Analgesia is provided with pethidine, pentazocine, tramadol or more recently, fentanyl, depending on availability. Facilities for invasive monitoring are inadequate and rarely used for obstetric cases. Nevertheless, spinal anaesthesia is sometimes used in patients with platelet counts >100 · 109/L and normal clotting time (3–11 min), provided they are judged stable, that is the blood pressure is 6150/100 mmHg and proteinuria <1 g in 24 h. Either hyperbaric 5% lidocaine or 0.5% bupivacaine are used, with the aim of achieving a sensory block to T8, in order to avoid blocking the cardioaccelerator nerves (T1-T4). A moderate preload of 5% dextrose is given and ephedrine is used to reverse hypotension. Postoperatively, stable patients are nursed in postnatal wards equipped with oxygen cylinders, but critically ill patients (four in this series) in need of organ support are nursed in the intensive care unit. Postoperative pain relief is provided with parenteral opioids like pethidine, tramadol or pentazocine.
Table 1. Numbers of deliveries and patients with preeclampsiaeclampsia, and numbers delivered by caesarean section Total deliveries (n) Total number of births Total number of live births Number of mothers with preeclampsia/ eclampsia Total number of caesarean sections among all deliveries Number with preeclampsia/eclampsia Mild preeclampsia Severe preeclampsia Eclampsia
3926 4036 3611 176 (4.5%) 898 (23%) 125 (13.9% of CS) 34 (27%) (1 death) 76 (61%) (5 deaths) 15 (12%)
Table 2. Age, parity and gestational age of preeclamptic/eclamptic patients presenting for caesarean delivery Number (%) (total number: 125) Age range (years)
619 20–25 26–30 31–35 36–40 P41
4 (3.2%) 19 (15.2%) 47 (37.6%) 31 (24.8%) 23 (18.4%) 1 (0.8%)
Parity
Nulliparous 1 2 3 4 P5
71 (56.8%) 21 (16.8%) 12 (9.6%) 10 (8%) 6 (4.8%) 5 (4%)
Gestational age at delivery (weeks)
>26 weeks 26–31weeks 32–36 weeks 37–42 weeks >42 weeks
Nil 22 (17.6%) 47 (37.6%) 54 (43.2%) 2 (1.6%)
The indications for caesarean section are given in Table 3. One hundred and three (82.4%) presented for emergency caesarean section, while 22 (17.6%) were counted as elective. The latter patients were not in labour, but presented with previous caesarean section (8), placenta praevia (3), elderly primip/bad obstetric history (3), intrauterine growth restriction only (5), post-dates (1), severe preeclampsia with Table 3. Indications for caesarean delivery
RESULTS The total number of deliveries in the four years of the study, the numbers of caesarean sections, the numbers with preeclampsia/eclampsia, the numbers of patients with preeclampsia/eclampsia delivered by caesarean section and the severity of disease, are shown in Table 1. Ninety patients (72%) had been booked for antenatal care, while 35 (28%) were unbooked. The average age of these patients was 30.2 years. Table 2 shows their age distribution, parity and gestational age at delivery.
Indication Severe preeclampsia/eclampsia with unfavourable cervix Fetal distress/intrauterine growth restriction Previous caesarean section Prolonged labour/poor progress Failed induction of labour/failed vacuum extraction Antepartum haemorrhage Elderly primip/bad obstetric history Premature rupture of membrane Post dates Big baby
Number (%) 85 (68%) 9 8 6 6
(7.2%) (6.4%) (4.8%) (4.8%)
4 3 2 1 1
(3.2%) (2.4%) (1.6%) (0.8%) (0.8%)
Anaesthesia in preeclampsia in Enugu, Nigeria 111 maternal distress (1) and big baby (1). There were seven twin pregnancies (132 babies but only 119 live births). General anaesthesia was used in 116 patients (92.8%) and spinal anaesthesia in the remaining 9 (7.2%). In non-preeclamptic patients, spinal anaesthesia was used in an even smaller percentage of caesarean deliveries (1.2%). There were no deaths among patients who received spinal anaesthesia. There were six deaths during general anaesthesia giving a case fatality rate of 5.2% of general anaesthetics or 4.8% of caesarean sections. The six deaths represented 55.5% of all maternal deaths following caesarean delivery during the period under review. The indication for surgery in five of the patients who died was severe preeclampsia, while one patient had mild preeclampsia complicated by placenta praevia and asthma. The deaths were due to complications of anaesthesia in three patients, preeclampsia/eclampsia (cerebrovascular accident and pulmonary oedema ) in two patients and intraoperative haemorrhage in the patient with placenta praevia and asthma. All patients with eclampsia survived. There were 13 stillbirths and 10 neonatal deaths, giving a perinatal mortality rate in this population of 174/ 1000 births.
DISCUSSION The caesarean section rate among all mothers during the four years of this study was 23%. This is higher than the rates of 18 and 20% reported recently for northern Nigeria,11,12 and reflects an overall increase in the caesarean section rate in Nigeria in the past decade.11,12 Iyaiya et al. from Illorin, Nigeria,12 reported that 9.3% of the parturients presenting for caesarean delivery had preeclampsia or eclampsia. In this study, preeclampsia and eclampsia patients represented 14% of parturients who presented for caesarean delivery and 4.5% of all deliveries. The caesarean section rate among our patients was 71%, which is greater than other reported caesarean section rates among such patients.12–14 Ozumba and Okaehialam14 reported in 1993 that 42% of preeclamptic/eclamptic patients managed in our hospital were delivered by caesarean section. Clearly this figure has since become even higher. Severe maternal and fetal complications often make urgent delivery necessary, and caesarean delivery is the expedient mode in most of these patients. It has been reported from Africa that operative and instrumental delivery may reduce maternal and perinatal mortality.15 In this study, however, the mortality rate, at about 5%, was higher among these high-risk patients than in the total population.
General anaesthesia was used in the great majority of the patients, while only 7.2% delivered under spinal anaesthesia. Epidural anaesthesia, the preferred choice of many anaesthetists, was not used due to the absence of usable epidural sets during the study period. Epidural sets have since become available. Epidural anaesthesia has for long been considered the technique of choice in preeclampsia, in part because it is known to increase intervillous blood flow and its slow onset helps in controlling the blood pressure.16 Recent studies have shown that both general17,18 and spinal anaesthesia19–21 are generally safe if properly conducted. Spinal anaesthesia is known to produce more consistent anaesthesia than epidural, and preeclamptics are actually less prone to hypotension than normal individuals.22,23 There were no maternal deaths among those receiving spinal anaesthesia in this study, but it must be acknowledged that numbers were very small. Fenton et al. in their study of 8070 caesarean sections in Malawi, showed that spinal anaesthesia may be safer than general,24 but a randomised study would be needed to ascertain this. Although not all the deaths in these high-risk patients were attributed solely to anaesthesia, three deaths were directly related to anaesthetic complications, two deaths due to cerebrovascular accident and pulmonary oedema, and one death due to intraoperative haemorrhage. Some of these deaths might have been prevented by the use of invasive monitoring, including central venous pressure, blood pressure and pulmonary artery pressure monitoring. Drugs like alfentanil, sufentanil and esmolol may obtund the pressor response to intubation better than either lidocaine or hydralazine, which causes a reflex maternal tachycardia. Spinal anaesthesia in stable patients avoids some of these complications, with the added benefit of a relatively simpler technique and cheaper cost. It also requires less expensive monitoring than does safe delivery of general anaesthesia. Cost is important in an environment where the cost of a disposable epidural set is about 25% of the minimum monthly wage for a government employee. Because of inadequate monitoring facilities, irregular supply of general anaesthetic drugs, patient poverty, and the complications of general anaesthesia, some authors consider spinal anaesthesia a preferable alternative to epidural or general anaesthesia,25,26 particularly in West Africa, because of its cheaper cost. Spinal anaesthesia was used successfully in six patients with severe preeclampsia and one with eclampsia, and may also benefit elective patients. An even smaller proportion of non-preeclamptic parturients (1.2%) delivered under regional anaesthesia during this study period. General anaesthesia has hitherto been the popular technique for caesarean delivery in this centre, where most patients present as emergencies and drugs like ketamine are cheap and easy to use and
112 International Journal of Obstetric Anesthesia relatively safe for most non-hypertensive parturients. However, we now emphasise regional techniques, unless there are contraindications to their use or time constraints. The case fatality rate in this study (4.8%) is higher than the case fatality rate reported from the United States for preeclampsia/eclampsia (6.4 per 10 000),27 but lower than that reported from six sub-Saharan African countries (18.4%).28 The latter studies were carried out in the capital cities of Niger, Mauritania, Burkina Faso, Ivory Coast and Mali and in two towns and a major city in Senegal.28 A report from Ethiopia put the case fatality rate at 21%.15 Contributory factors in the high rate in this study may be: Poor remuneration of staff, leading to a flight of manpower overseas.29–31 This department lost six consultant anaesthetists to the Gulf States, Europe and better paying jobs in Nigeria since the late 1980s. Poverty among patients and late presentation to the hospital.32,33 The fact that 69 deliveries (55.2%) were pre-term underlines the severity of the disease. Extensive use of general anaesthesia despite grossly inadequate supply of drugs and monitoring facilities. Haemodynamic monitoring equipment, capnography and pulse oximetry, and drugs such as alfentanil, esmolol and fentanyl are rarely available. Hospital administrators need to be better informed on the importance of anaesthesia in modern medicine, to persuade them to allocate more funds to meet these needs. Five of the 104 patients managed by trainee anaesthetists died in this study. There was one fatality in the 21 cases managed by the consultants. Although this gives a similar fatality rate, it must be remembered that consultants manage the sickest patients, cases considered beyond the scope of senior registrars. All the deaths were intraoperative. The 13 stillbirths and 10 neonatal deaths in this study gave a perinatal mortality rate of 174/1000 births. Preeclampsia and eclampsia are a leading cause of fetal loss especially in the third world, due to placental abruption, placental insufficiency and birth asphyxia.
CONCLUSION While maternal and fetal mortality following anaesthesia for caesarean delivery in patients with preeclampsia/eclampsia has fallen over the years in the Western world, the figures in this study show that it remains a major cause of loss of life in West Africa. Poverty, late presentation of patients, lack of equipment and drugs and inexperienced staff were probably major contributory factors. A case can be made for the use of spinal anaesthesia in stable patients in view of recent favourable reviews and the cheaper cost. Further studies in this
and other Nigerian hospitals will help present a clearer picture. Finally, a multidisciplinary approach is vital, with close co-operation between the anaesthetist, obstetrician and midwife. The preeclampsia protocol which was developed by Robson et al.34, has been adopted by many obstetric units and can be modified to suit individual units. ACKNOWLEDGEMENT We thank the staff of the records department, labour ward and obstetric theatre of the University of Nigeria teaching hospital, Enugu for their co-operation during this study. We also extend our gratitude to Miss Mary Nwodo who graciously typed the manuscript.
REFERENCES 1. Mushambi M C, Halligan A W, Williamson K. Recent development in the pathophysiology and management of pre-eclampsia. Br J Anaesth 1996; 76: 133–148. 2. Macgillivary J. Some observations in the incidence of preeclampsia. Br J Obstet Gynaecol 1958; 65: 536–539. 3. Conz P A, Catalano C. Pathophysiology of preeclampsia. G Ital Nefrol 2003; 20: 15–22. 4. Okaro J M, Umezulike A C, Onah H E, Chukwuali L I, Ezegwu O F, Nweke P C. Maternal mortality at the University of Nigeria Teaching Hospital, Enugu. Before and After Kenya. Afr J Reprod Health 2001; 5: 90–97. 5. Chukwudebelu W O, Ozumba B C. Maternal mortality in Anambra State of Nigeria. Int J Gynaecol Obstet 1988; 27: 365–370. 6. Ariba A J, Inem A V, Biersack G, Aina A O, Ayankagbe O O, Adetoro O O. Pattern of obstetric mortality in a voluntary agency hospital in Abeokuta, Southwest Nigeria. Nigerian Med Practitioner 2004; 45: 83–90. 7. Adamu Y M, Salihu H M, Sathiakumar N, Alexander G R. Maternal mortality in Northern Nigeria; a population based study. Eur J Obstet Gynecol Reprod Biol 2004; 109: 153–159. 8. Wall L L. Dead mothers and injured wives: the social context of maternal morbidity and mortality among the Hausa of northern Nigeria. Stud Fam Planning 1998; 29: 341–359. 9. Maternal mortality in 2000 – Estimates developed by WHO, UNICEF and UNFDA. 2003. 10. Ikechebelu J I, Okoh C C. Morbidity and mortality following induced abortions in Nnewi, Nigeria. Trop Doct 2003; 33: 170–172. 11. Aisien A O, Lawson J O, Adebayo A. Caesarean section in a Northern Nigeria University Teaching Hospital. Niger Postgrad Med J 2002; 9: 146–150. 12. Iyaiya M A, Aboyeji P A. Caesarean delivery: the trend over a ten-year period at Illorin, Nigeria. Nig J Surg Res 2001; 3: 11–18. 13. Grau T, Conradi R, Wacker J, Martin E. Preeclampsia and anaesthesia. Zentralbl Gynakol 1999; 21: 627–630. 14. Ozumba B C, Okeahlaham M. Incidence and complication of preeclampsia at the University of Nigeria Teaching Hospital (UNTH), Enugu, Nigeria. Orient J Med 1993; 5: 136–138. 15. Yifru B, Ahmed A. Emergency obstetric performance with emphasis on operative delivery: does it reflect the quality of care. Ethiop J Health Dev 2004; 18: 96–106. 16. O’ Sullivan G. Medical aspects of obstetrics. In: Prys-Roberts C, Brown Jr B R, eds. International Practice of Anaesthesia. Oxford: Butterworth-Heinemann; 1996: 2/95/1-20. 17. Moodley T, Jjuuko G, Rout C. Epidural compared with general anaesthesia for caesarean delivery in conscious women with eclampisa. BJOG 2001; 108: 378. p. 378–382. 18. Eldridge E. Obstetric anaesthesia and analgesia. In: Allman K G, Wilson I H, eds. Oxford Handbook of Anaesthesia. Oxford: Oxford University Press; 2001: 695–756.
Anaesthesia in preeclampsia in Enugu, Nigeria 113 19. Gogarten W. Spinal anaesthesia for obstetrics. Best Pract Res Clin Anaesthesiol 2003; 17: 377–392. 20. Hood D D, Curry R. Spinal vs epidural anesthesia for cesarean section in severely preeclamptic patients; a retrospective survey. Anesthesiology 1999; 90: 1276–1282. 21. Ahmed S M, Khan R M, Bano S, Ajmani P, Kumar A. Is spinal anaesthesia safe in preeclamptic toxaemia patients? J Indian Med Assoc 1999; 97: 165–168. 22. Brodie H, Malinow A M. Anesthetic management of preeclampsia/ eclampsia. Int J Obstet Anesth 1999; 8: 110–124. 23. Aya A G, Mangin R, Vialles N, et al. Patients with severe preeclampsia experience less hypotension during spinal anesthesia for elective cesarean delivery than healthy parturients: a prospective cohort comparison. Anesth Analg 2003; 97: 867–872. 24. Fenton P M, Whitty C J M, Reynolds F. Caesarean section in Malawi: prospective study of early maternal and perinatal mortality. BMJ 2003; 327: 587–591. 25. Farragher R, Datta S. Recent advances in obstetric anesthesia. J Anesth 2003; 17: 30–41. 26. Shifman E M, Filippovich G V. Safety of subarachnoid anesthesia during cesarean section in pregnant women with preeclampsia. Anesteziol Reanimatol 2003; 5: 38–41.
27. Mackay A P, Berg C J, Atrash H K. Pregnancy related mortality from preeclampsia and eclampsia. Obstet Gynecol 2001; 97: 533–538. 28. Prual A, Bouvier-Colle M H, de Bernis L, Breart G. Severe maternal morbidity from direct obstetric causes in West Africa: incidence and case fatality rates. Bull World Health Organ 2000; 78: 593–602. 29. Stillwell B, Diallo K, Zurn P, et al. Migration of health workers from developing countries; strategic approaches to its management. Bull World Health Organ 2004; 85: 595–600. 30. Raufu A. Nigeria health authorities worry over exodus of doctors and nurses. BMJ 2002; 325: 65. 31. Loewenson R. Structural adjustment and health policy in Africa. International Journal of Health Services 1993; 23: 717–730. 32. Conde-Agudelo A, Kafury-Goeta A C. Case-control study of risk factors for complicated eclampsia. Obstet Gynecol 1997; 90: 172–175. 33. Haelterman E, Qvist R, Barlow P, Alexander S. Social deprivation and access to care as risk factors for severe preeclampsia. Eur J Obstet Gynecol Reprod Biol 2003; 111: 25–32. 34. Robson S C, Redfern N, Walkinshaw S A. A protocol for the intrapartum management of severe preeclampsia. Int J Obstet Anesth 1992; 1: 222–229.