Introduction
When providing care and support to women suffering from moderate or severe NVP, HCPs need to consider the evidence on effectiveness and safety as well as the direct and indirect costs to the NHS and to the women. The effectiveness of any intervention is likely to be dependent on the severity of symptoms. Ideally, the severity of NVP should be assessed in a reliable and valid way (e.g. using the PUQE, RINVR or Nausea Questionnaire score). A mild, moderate or severe score, alongside an assessment of how the individual woman is coping, will help HCPs make appropriate recommendations.
The following findings include evidence of effectiveness from the results of the review and associated treatment costs from the economic evaluation presented in Chapter 18, Economic evaluation. No reliable data on fetal outcomes [fetal or neonatal death, congenital abnormalities, low birthweight (< 2.5 kg), preterm birth (before 37 weeks’ gestation) or small for gestational age (birthweight < 10th centile)] were identified as part of the systematic review. All safety data presented below are derived from large population-based observational studies. This evidence is indirect in that it relates to pregnancy but not specifically to NVP.
The interventions that can be recommended to women fall into three categories:
Self-help, ‘over-the-counter’ interventions which can be recommended by GPs, midwives, practice nurses and pharmacists, in information leaflets and on NHS approved websites.
Primary care interventions which can be prescribed by GPs. These may be in addition to or instead of self-help interventions.
Secondary care interventions which are delivered in a hospital setting. Some women are referred to hospital by primary care/community HCPs after failure of self-help and/or primary care interventions. However, increasingly, women refer themselves to hospital maternity services (via day/maternity assessment units) often without any prior intervention. Depending on symptom severity and duration, sufferers may need to progress from one category to another and possibly try a number of different, or combinations of interventions.
Findings
First-line ‘over-the-counter’ interventions and alternative therapies
Ginger
Fresh root ginger may be effective in improving mild symptoms, especially nausea, but findings are not conclusive. When compared with acupressure, ginger may be more effective, but the quality and quantity of evidence is limited. There are too few data to suggest ginger is any better or worse than vitamin B6 or pharmacological interventions.
Ginger appears to be safe for women during pregnancy. Data from two birth cohorts and several small RCTs suggest there is no evidence of an increased risk of congenital anomalies or adverse perinatal outcomes.136,137 In a birth cohort of 68,522 women, 1020 reported using ginger, with no associated increased risks of congenital malformations, stillbirth, perinatal death, preterm birth or low birthweight babies.137
The estimated weekly cost of using a ginger preparation was £2.20.
Acupressure at pericardium point P6
Acupressure using correctly placed ‘sea-bands’ may be effective in improving symptoms of NVP in women with mild to moderate symptoms. Acupressure at KID21 may also be effective but data are limited. The quality of the evidence is low and it is difficult to ascertain whether acupressure is more or less effective than other interventions.
Acupressure is a non-invasive intervention and no safety data are available. The price of sea-bands is approximately £10.
Acupuncture
The evidence of effectiveness of acupuncture is inconclusive. The cost of acupuncture consultations and treatments range between £35 and £70. These data suggest that acupuncture should not be recommended.
Vitamin B6 (pyridoxine)
Vitamin B6 appears to have some effectiveness at improving NVP in women with mild to moderate symptoms, especially at higher doses (≥ 10 mg three times daily for at least 3–4 days).
There is little information regarding the safety of vitamin B6 in isolation. However, the safety of vitamin B6 in combination with antihistamines (doxylamine) has been studied extensively. The results of two meta-analyses, which involved over 200,000 women, found no evidence of increased risk of congenital malformations (relative risk 0.95, 95% CI 0.88 to 1.04138 and OR 1.02, 95% CI 0.66 to 1.55).139 These findings have since been confirmed by several epidemiological studies.
The weekly cost of vitamin B6 preparations has been estimated at between £0.12 and £2.60.
Second-line interventions prescribed by general practitioners in primary care settings
All quoted medication costs will require the addition of £45 for the GP consultation.
Doxylamine plus pyridoxine
A combination of high-dose pyridoxine and doxylamine may be as effective as metoclopramide at relieving moderate symptoms of NVP. However, a combination of these medications has not been shown to be as effective as ondansetron. Given other data, it can be inferred that this therapy is more effective than no treatment (proxied by placebo in trials).
Evidence suggests that the combined therapy of doxylamine and pyridoxine is safe to use during pregnancy (see Appendix 7).
The cost of this combined therapy would be £0.12–2.60 plus cost of doxylamine [price currently not available as not routinely used in the UK NHS, price of alternative antihistamine (e.g. cyclizine) is £0.74–2.22].
Diclectin (delayed release doxylamine 10 mg plus pyridoxine 10 mg) appears to be more effective than placebo when given to treat moderate symptoms and when given pre-emptively to women with a history of moderate NVP in a previous pregnancy. However, Diclectin is currently not available in the UK.
Antihistamines (hydroxyzine, meclizine, cyclizine)
Evidence from the review suggests that these drugs appear to be more effective in treating mild symptoms of NVP compared with no treatment.
A large meta-analysis which involved over 200,000 women who took antihistamines during pregnancy found no evidence of an increased risk of teratogenicity (OR for major malformations was 0.76, 95% CI 0.60 to 0.94), with no other serious maternal or fetal outcomes.140
Estimated weekly cost of treatment with oral cyclizine was £0.74–2.22.
Dopamine receptor antagonists
There was no evidence of a difference between metoclopramide and promethazine in improving symptoms in moderate cases of NVP. There was also very low-quality evidence from a non-randomised study that droperidol may be effective, especially at higher doses but higher-quality evidence is needed to confirm or refute this.
Dopamine antagonists such as the phenothiazines promethazine and prochlorperazine are regarded as safe: a meta-analysis of eight studies (n = 2948) identified no difference in the risk of major malformations (pooled relative risk 1.03, 95% CI 0.89 to 1.22).141 Other drugs in this class used to treat NVP/HG include domperidone, droperidol, trimethobenzamide and metoclopramide. Limited evidence suggests that trimethobenzamide is safe while a recent large study of metoclopramide used during the first trimester of pregnancy (n = 3458) found no evidence of an increased risk of major malformations (OR 1.04, 95% CI 0.89 to 1.14) or adverse obstetric outcome.142 These findings agreed with those of a large, Danish birth cohort (exposed cases n = 28,486), which reported no increased risk of congenital malformations (OR 0.95, 95% CI 0.88 to 1.03), miscarriages or stillbirths.143 The estimated weekly costs of these drugs are metoclopramide £0.22–0.66, promethazine £0.74–2.22, prochlorperazine £0.47–1.42, domperidone £0.39–1.17.
Serotonin receptor antagonists (ondansetron)
Ondansetron appears to be effective at all levels of symptom severity. However, there is little evidence to say that it is more effective than metoclopramide or antihistamines.
Serotonin antagonists may be safe in pregnancy but experience is limited. One recent case–control study reported an increased risk of cleft palate (adjusted OR 2.37, 95% CI 1.18 to 4.76).144 However, another Danish study, involving over 600,000 pregnancies, found no association with any adverse fetal outcomes.145 Concerns have been raised regarding the risk of cardiac arrhythmias [time between start of the Q wave and of the T wave in the heart’s electrical cycle (QT) prolongation] with large doses of i.v. ondansetron. A recent systematic review identified that out of all cases of QT prolongation, 67% of patients had a significant medical history or were using a concomitant QT prolonging medication. The study concluded that prior screening of electrocardiogram (ECG) and electrolytes should be limited to high risk patients receiving ondansetron intravenously.146
The estimated weekly cost of prescribing oral ondansetron was £60.83.
Second- and third-line interventions delivered in secondary care settings
Outpatient/day case management
Limited evidence suggests that this could be an effective alternative to inpatient admission for women with moderate to severe symptoms where i.v. rehydration is required.
There is no population-based safety data regarding pregnancy outcomes following outpatient/day case management.
The estimated cost of an outpatient attendance, which would involve i.v. rehydration, i.v. antiemetics, appropriate blood and urine tests and an ultrasound scan, would be approximately £290.
Inpatient management costs
The estimated cost of a 2-night inpatient admission, which would involve i.v. rehydration, i.v./oral antiemetics, appropriate blood and urine test, an ultrasound scan, thromboprophylaxis and thiamine supplementation, was approximately £800. If a combination of, or alternative antiemetics were required this cost could rise to £850.
Corticosteroids
Evidence is limited but corticosteroids appear to reduce symptom severity, and appear to be more effective at reducing episodes of vomiting than metoclopramide and Phenergan.
Concerns remain about the safety of corticosteroids. In one meta-analysis, the pooled risk ratio for cohort and case–control studies combined revealed no increased risk of major malformations associated with first trimester exposure (cumulative OR 1.45, 95% CI 0.81 to 2.60). However, a subanalysis of case–control studies revealed an increase in the risk of the fetus developing an oral cleft palate (OR 3.35, 95% CI 1.97 to 5.69) and the results were homogeneous between studies.147 However, other studies do not show this association with cleft palate formation.148,149
Treatment with corticosteroids would generally be initiated following failure of other antiemetic therapies, when the woman is suffering from severe symptoms and during an inpatient admission. The estimated cost of a 5-day inpatient admission in which corticosteroid therapy was given was estimated at £2000–2100.
Enteral nutrition
Enteral feeding appears to be a potentially effective but extreme method of supporting women suffering from very severe symptoms, but no comparative studies were identified.
Costs were not available for enteral feeding but in a recent economic evaluation of total parental feeding the cost per patient of the intervention itself and excluding other hospital based costs was approximately £340 per patient (Kilonzo M, personal communication).
Summary
Overall, there are some data to guide the choice of therapy, but there is little information to guide the choice between therapies. The evidence reviewed suggests that treatments tend to improve symptoms quickly – over a small number of days. Therefore, if a woman gets insufficient relief from a first treatment then this suggests an alternative treatment could be tried. For some treatments there are no data currently available to support their use. Thus, while there may be few safety concerns with some of these treatments, lack of robust effectiveness data can nevertheless guide practitioners in the advice that is given to women.
One obvious concern is around safety. In part this is prompted by the legacy of thalidomide. No reliable evidence directly applicable to the target group of pregnant women was found although evidence from large population sources exists. These data have generally suggested that there is no evidence of any safety concerns with the medications, but this is not the same as ruling out any important differences in safety outcomes. In many cases the CIs are sufficiently wide to include clinically important differences both in favour and against the treatments. Although women need reassurance that treatments are safe, this should not offer false reassurance, therefore advice on safety should be tempered by these findings.
