Studies that have examined the occurrence of headaches early after childbirth suggest that these are experienced by 20–40% of women on one or more days during the first week (Grove, 1973, Pitt, 1973, Stein, 1981 and Stein et al., 1984). Garcia & Marchant (1993), in an observational study of postnatal health at 8 weeks among 90 women, found that 23% had experienced headaches at some time since the birth. Longer term studies in postpartum populations have also found frequent headaches to be common. In one large observational study of a variety of health problems after childbirth, 419 (4%) of 11,701 women reported frequent headaches which had begun for the first time within 3 months of the delivery and lasted for over 6 weeks, and a further 5% reported similar frequent headaches which they had also had sometime before. The corresponding proportions reporting migraine were 1% and 6%, indicating that migraine was less likely to be a new postpartum problem (MacArthur et al 1991). Glazener et al (1995), in a longitudinal study, examined maternal postnatal morbidity in over 1200 women, who comprised a 20% random sample of deliveries in the Grampian region of Scotland between June 1990 and May 1991. All women were given a questionnaire whilst on the postnatal ward, a postal questionnaire at 8 weeks, and half were sent another questionnaire 12–18 months after the birth. Headaches of any duration, including new and recurrent symptoms, were reported by 14% of women whilst in hospital, by 22% between then and 8 weeks and by 15% after this. A large cluster randomised controlled trial (RCT) of a new model of midwifery-led postnatal care, which focused on the identification and management of common health problems up to 12 weeks after the birth, included the same health problems in a questionnaire completed by women 12 months after the birth. At 12 months headaches were reported by 26% of women overall, with no difference in symptoms between the intervention and control groups (MacArthur et al 2003). As with the study by Glazener et al (1995), women were not asked to distinguish between new and ongoing symptoms.

A longitudinal survey of Italian and French women’s health after childbirth found that headache prevalence had increased at 12 months postpartum compared with at 5 months: 45% of Italian women reported headaches at 12 months compared with 22% at 5 months, and 38% of French women had headaches at 12 months compared with 21% at 5 months (Saurel-Cubizolles et al 2000). Reasons for this were not postulated.

An additional finding of MacArthur et al (1991) was that some headaches, occurring with musculoskeletal symptoms, were more likely to start during the first postnatal week and were associated with epidural analgesia for pain relief. Frequent headaches or migraine in women without musculoskeletal symptoms were associated with younger age, multiparity and lower social class, and not with epidural analgesia, and were probably related more to factors within the social environment than to the delivery. Headaches in the first postpartum week were found to be associated with psychological morbidity by Stein et al (1984). In this study, the occurrence of headache was documented daily among 71 women who were asked to complete a self-rating questionnaire, which also included questions on the presence of tension, depression and feelings of weepiness. Women who developed a headache on at least one day were more depressed and had more tension than those who did not. Details of the self-rating schedule were not described, thus its validity cannot be assessed.

Studies of headache after hospital discharge were all investigating postnatal health more generally, and relied on self-reports of headaches, without obtaining detailed classificatory information of the sort used to distinguish different types of headache as in the ICHD (2004). It is likely, however, that most postpartum headaches are tension headaches, the next most common being migraine, most of which will be reported by women who have a previous history of this. The main role of the midwife is to assess the headache in order to refer those that are due to other causes. The management of tension headaches and migraine is the same as for general population groups (Barrett 1996), but taking care that any analgesia is appropriate for breastfeeding mothers.

A postdural puncture headache (PDPH) may occur following the administration of an epidural or spinal needle for pain relief in labour or for caesarean section. In the case of spinal anaesthesia the dura is punctured deliberately, whilst an accidental dural puncture occurs occasionally during the insertion of the epidural needle. The diagnosis of an accidental dural puncture is usually made by the anaesthetist during insertion of the epidural, when cerebrospinal fluid (CSF) is observed, but some cases are diagnosed retrospectively following the onset of headache in the puerperium.

Headache after dural puncture results from a loss of CSF, with subsequent traction on the meninges, and has several typical presenting characteristics. The most significant is its postural nature; the headache gets dramatically worse when the patient moves from the supine to the upright position, and conversely is markedly diminished or relieved totally when lying down (Katz & Aidinis 1980). The associated symptoms of neck stiffness, visual disturbances, vomiting and auditory effects may also be reported, but are more common with severe PDPH (Lybecker et al., 1995 and Banks et al., 2001).

The incidence of accidental dural puncture during epidural is now between about 0.18% and 3.6%, although it had been higher than this when epidural was first used in routine obstetric practice (Gleeson & Reynolds 1998). Where this type of puncture does occur, however, the relatively large diameter of the epidural needle means that loss of CSF is likely and the incidence of PDPH is high. This high incidence is also attributed to bearing down in the second stage of labour which may exacerbate CSF leakage; to decreased intra-abdominal pressure following delivery which causes the epidural veins to collapse; and to rapid loss of fluid from blood loss, diuresis and lactation (Gutsche 1990). In a case-note analysis of 20 years of women with accidental puncture in one maternity unit, the incidence of typical PDPH was 86% (Stride & Cooper 1993). The majority of the women (69%) in this series developed their headache within the first 2 days of delivery, although it could also appear up to 6 days after (Stride & Cooper 1993).

A lower incidence of PDPH was reported in a large UK survey of outcomes following anaesthetic interventions. The National Obstetric Anaesthetic Database (NOAD) was established in the UK in 1998, to support collection of national data on obstetric analgesia and anaesthesia. Data for the first year aimed to determine the incidence, characteristics, contributing factors and management of postpartum headaches associated with anaesthetic interventions (Chan et al 2003). Data were requested from all members of the Obstetric Anaesthetists Association in the UK in 1999, who were asked to provide complete data for at least 1 month continuously for their unit on anaesthetic activity and number of headaches reported. Symptoms were to have lasted for more than 6 hours, unrelieved by mild analgesics. Data were supported with an anonymous individual case record for each incident reported. Data were collected on 65,348 women from 93 obstetric units, 38,271 (59%) of whom had an epidural; 16,844 (26%) had a spinal; 4926 (8%) had a combined spinal-epidural (CSE); 4203 (6%) had a general anaesthetic (GA); 939 (1%) had a GA and regional block; and 165 (0.2%) had a combination of regional techniques. The incidence of headache ranged from 1.1% to 1.9% between all anaesthetic techniques, which increased to 10.9% for multiple regional anaesthetics. Individual case records were returned for 1101 women, 975 of whom had an anaesthetic intervention. Headaches in the 975 women were divided into cases where a PDPH had been diagnosed (404, 41%) and non-PDPH (571, 59%) based on information provided. As data were only collected on women who were inpatients, and given the potential for incomplete data collection, the incidence of PDPH may have been higher. Other outcomes of interest reported in the survey are described later in this chapter.

Most PDPHs are of relatively short duration, lasting for several days (Crawford 1972). In the study described earlier of health problems after childbirth among 11,701 women, however, 74 were recorded as having an accidental dural puncture and 23% of these reported frequent headaches or migraine lasting for longer than 6 weeks. Information on whether these headaches were of a postural nature was not obtained, although some women also reported neckache or visual or auditory disturbances (MacArthur et al 1993).

Much smaller diameter needles are used for spinal than for epidural analgesia, so that although a spinal always punctures the dura, with current types of spinal needles (see below), similar proportions of PDPHs occur after each procedure. Spinal anaesthesia has been used for a variety of surgical and investigative interventions for 100 years or so, but the early incidence of PDPH was high. In a classic study from the USA of a general series of 10,098 spinals, it was noted that obstetric patients had the highest headache rates (Vandam & Dripps 1956). In 1979, Crawford reported a PDPH rate of 16% in an obstetric unit that was a specialist centre for regional blocks (Crawford 1979). Spinal needles of smaller diameter and designs that spread rather than cut the dural fibres have since become popular and have made a significant impact on the incidence of postspinal headache (Turnbull & Shepherd 2003). A general population-based meta-analysis, which found rates of all headaches ranging from 1% to almost 30%, and rates of severe headaches from none to 12%, concluded that smaller and non-cutting needles were associated with the lowest rates (Halpern & Preston 1994). A randomised controlled trial of spinal anaesthesia for caesarean section compared a 25 gauge diamond-tipped needle with a 24 gauge non-cutting Sprotte needle (Cesarini et al 1990). The trial planned to recruit 100 women to each group but was stopped at 55, when a PDPH rate of 14.5% was shown in the former group, compared with none in the Sprotte group. Other obstetric studies comparing different needles have had similar findings (e.g. Shutt et al 1992). In addition to needle size, the direction of insertion of the needle has also been considered. A meta-analysis based on general population data was undertaken to determine if bevel direction during lumbar puncture would influence the incidence of PDPH (Richman et al 2006). The results indicated that use of a cutting needle with insertion in a parallel/longitudinal fashion may significantly reduce the incidence of PDPH, although reasons for the decrease were unclear.

Most maternity units now report incidence rates of severe headaches requiring blood patch of about 1% or less (Hopkinson et al., 1997 and Madej et al., 1993). Mild headaches have been documented in up to 10% (Hopkinson et al 1997), but the extent to which these might be attributed to the spinal is not known since, as described earlier, headaches in the first few days after birth are generally quite common. Data from the NOAD survey described earlier (Chan et al 2003) found that of the 404 women who had a diagnosed PDPH, severe headaches were reported by 202 (50%), a much higher proportion than reported in the non-PDPH group (p < 0.0001), and 305 (75%) of these women reported that the headache limited their daily activity (p < 0.0001).

Epidural analgesia remains a common intervention in labour, with most units now offering lower-dose infusions which enable women to mobilise. In a recent Cochrane systematic review of CSE with epidural analgesia in labour which included data from 14 trials (2047 women), a total of 25 outcomes were analysed (Hughes et al 2003). No difference was found between CSE and epidural techniques with regard to the incidence of PDPH or blood patch (see below), hypotension, urinary retention, mode of delivery or admission of the baby to the neonatal unit.

Another Cochrane review compared outcomes following spinal or epidural for caesarean section (Ng et al 2004). Ten trials were included in the review, which provided data on a total of 751 women. The reviewers could not draw any conclusion about intraoperative side-effects and postoperative complications, such as PDPH, due to small numbers reported.

Since many women are now discharged from hospital before a PDPH is likely to develop, it is important that the midwife is able to identify this type of headache (Cooper 1999), and if the headache is severe, referral must be made to the appropriate healthcare professional, according to local policy. In the case of a known accidental dural puncture the anaesthetist may have already instituted prophylactic measures or begun treatment if symptoms were severe.

A clinical review of the treatment of PDPH in the general population notes that conservative treatment is recommended in the first instance, which includes simple analgesics (e.g. paracetamol), bed rest and hydration. Bed rest seems to alleviate symptoms, although this has not been found to be effective in preventing a headache occurring (McSwiney & Phillips 1995). An evaluation of the literature (including clinical studies, letters, abstracts and case reports) on the pharmacological management of PDPH in general populations concluded that intravenous and oral caffeine are effective treatments, and at least non-invasive, but that more clinical studies are required to properly evaluate other pharmacotherapies. In the meantime, the authors suggest that therapy be guided by clinical judgement (Choi et al 1996).

Epidural blood patch is now considered by many obstetric anaesthetists to be an effective treatment for severe PDPH after accidental puncture or spinal anaesthesia, but this is unsupported by evidence of effectiveness from RCTs. A blood patch involves injecting 10–20 ml of the woman’s blood into the epidural space around the site of the dural puncture; the blood coagulates and seals the leak of CSF. A second blood patch is sometimes given in the event of failure. A success rate of over 90% has been reported in some observational studies, although others report permanent symptom relief in about two-thirds (Stride and Cooper, 1993 and Taivainen et al., 1993).

There is controversy about the use of blood patching as a prophylactic measure where accidental dural puncture is known to have occurred (Cooper 1999). Berger et al (1998), in a survey of the management of accidental dural puncture during labour in 36 centres in North America, found that almost half used prophylactic blood patching. In the NOAD survey (Chan et al 2003), 242 (60%) of the 404 women with a diagnosed PDPH received a blood patch, with 101 (42%) performed within 2 days of the onset of symptoms and almost all (91%) performed within 8 days. Data on outcome following blood patch were not provided. A Cochrane library systematic review of epidural blood patching for prevention and treatment of PDPH (Sudlow & Warlow 2001) included three trials with data from 77 general population patients. Methodological details were generally incomplete. Although the results of the analyses suggested that both prophylactic and therapeutic epidural blood patching may be of benefit, the very small number of patients and outcome events, as well as uncertainties about trial methodology, precluded reliable assessments of the potential benefits and harms of blood patching. Further, adequately powered randomised trials (including at least a few hundred patients) are required before reliable conclusions can be drawn about the effectiveness of epidural blood patching.

Pregnancy-induced hypertension (PIH) is commonly reported and was defined in a clinical review of the hypertensive disorders of pregnancy as the recording of a blood pressure of 140/90 mmHg or more on two occasions 4 or more hours apart after the 20th week of pregnancy, in a previously normotensive woman (Broughton Pipkin 1995). The hypertensive disorders of pregnancy comprise a spectrum of conditions that are usually classified into four categories: (i) gestational hypertension, a rise in blood pressure during the second half of pregnancy; (ii) pre-eclampsia, usually hypertension with proteinuria (protein in urine) during the second half of pregnancy; (iii) chronic hypertension, a rise in blood pressure prior to pregnancy or before 20 weeks’ gestation and (iv) pre-eclampsia superimposed on chronic hypertension (Gifford et al 2000).

Pre-eclampsia is considered to occur when PIH is associated with significant proteinurea (300 mg/l in 24h) (Davey & MacGillivray 1988). The precise relationship between PIH and pre-eclampsia, however, is still unclear. The most common definition of eclampsia is convulsions, plus the usual signs and symptoms of pre-eclampsia, where other causes of convulsion have been excluded (Douglas & Redman 1994). Fourteen maternal deaths due to eclampsia or pre-eclampsia were reported in the UK between 2000 and 2002 (Lewis & Drife 2004), seven of which occurred after the birth.