Introduction
In developing these conclusions, we worked with a range of individuals and organisations to ensure their relevancy to patients and practitioners. Women who had suffered from NVP and/or HG and representatives of key patient advocacy groups in the field were members of our Project Steering Group, and were consulted in shaping both the review implications and our recommendations for future research. Our Steering Group and Project Management Group also included a number of clinicians working with women sufferers of NVP and/or HG in a range of health-care settings, representing midwifery, general practice, obstetrics, teratology, paediatrics and clinical pharmacology.
Implications for women and for practitioners
Nausea and vomiting in pregnancy occurs frequently. Even without treatment, the symptoms for some women will resolve. For women who do experience more persistent problems there are some simple approaches that they could adopt themselves that might improve symptoms and/or QoL. These could include increasing oral fluid intake, eating small frequent meals, eating bland foods/protein-predominant meals, avoiding spicy, odorous and fatty foods and stopping iron-containing multivitamins. However, for others there are a range of treatments available to women where there is some evidence that they might help.
Where symptoms are mild, there are a number of first-line over-the-counter or self-purchased therapies where there is some evidence that they work. These therapies may also be considered as initial treatments if a woman sees a doctor, nurse, midwife or other HCP. These treatments are ginger supplements and vitamin B6. These are generally low cost and there is no evidence that they are a risk to the health of the mother or baby. There is no proof that a more expensive version of either ginger or vitamin B6 is any better than a lower cost option. Higher doses of vitamin B6 were found to be more effective than lower doses in reducing symptoms (0.64 mg twice daily vs. 1.28 mg per day). There is also some evidence that acupressure may be effective, but the evidence is very limited. The available evidence suggests that any benefit derived from these over-the-counter therapies will be evident within 3–4 days. Thus, women need to know that, in such cases, it is worth trying something else or consulting with a GP as there are other treatments available via prescription. There is no evidence that these treatments are unsafe to use. Further details on dosage and effects are in
Chapters 4–7 and
Tables 8–
11.
Where symptoms are mild to moderate and/or if the above over-the-counter therapies have not proved helpful to women, a number of second-line interventions are available via prescription. Of the drugs that a GP might prescribe there is evidence that an antihistamine can help reduce symptoms (either alone or combined with vitamin B6) compared with no treatment. Limited data suggests that metoclopramide and promethazine may also help when symptoms are moderate. Droperidol may also be effective, especially at higher doses. Available evidence indicates these treatments are likely to be safe, but more research is needed to clarify this (further details on dosage and effects are in
Chapter 9 and
Table 13).
A GP may also prescribe ondansetron when other treatments have failed. This treatment may be effective for some women in reducing symptoms. The drug appears to be safe in pregnancy but experience is limited and more research is needed. Where women are given large doses of ondansetron and have a risk of some cardiac conditions, they may need an ECG and to have their blood chemistry checked (further details on dosage and effects are in
Chapter 8 and
Table 12).
Where symptoms are more severe or persistent, based on assessment using a validated measurement scale, care may be provided in hospital, and for these women a further range of treatments are available. Evidence suggests that where available these treatments can be provided as an outpatient or day case patient rather than requiring an admission, and usually involve rehydration with i.v. fluids. Dextrose saline appears to be more effective at reducing nausea when compared with normal saline, but must be administered with care as excessive dextrose can exacerbate the potential problem of Wernicke’s encephalopathy (see
Chapter 12 and
Table 16 for more details).
Treatment with corticosteroids does work but would generally only be used following failure of other treatments, when the woman is suffering from severe symptoms and during an inpatient admission. This is because doctors and other HCPs are cautious about the safety of these treatments and the side effect they may have for both the mother and the fetus. Therefore, doctors would like to use other options first if possible. Ideally, more evidence is needed comparing corticosteroids to other antisickness medications (further details on dosage and effects are in
Chapter 15 and
Table 19).
Where symptoms are extremely severe, and when the prolonged effects of severe nausea and vomiting have made women extremely ill, assisted feeding either by tube directly into the stomach or, in very rare cases, by i.v. catheter may be effective. As there are risks associated with this type of treatment, it is restricted to women suffering from very severe symptoms (see
Chapter 16 and
Table 20 for more details).
Recommendations for research
Trajectory of research
Any recommendations for further research need to take into account the trajectory of new research. Research on treatments and management of NVP/HG is ongoing: since the date of the last systematic update of published research for this study (September 2014) new evidence has been published. Although not assembled systematically, scoping searches conducted in early December 2014 identified two potentially relevant studies. The first was a review of the use of gabapentin in pregnancy150 and the second was a RCT comparing day care with inpatient management.151 Neither study was formally assessed, but the authors’ conclusions were that further clinical trials were needed on the use of gabapentin for HG, and that day case care was acceptable and reduced length of stay. These conclusions are consistent with the findings reported in this review.
There are also a number of studies recorded as ongoing on searches of ClinicalTrials.gov (www.clinicaltrials.gov) and the International Clinical Trials Registry Platform (http://apps.who.int/trialsearch/default.aspx) (searches of both were conducted in December 2014). Ongoing studies are investigating:
gabapentin versus ondansetron for HG in a double-blind RCT of 80 women with HG (NCT02163434, due to finish in May 2018)
Diclectin as a pre-emptive treatment for NVP versus treatment with Diclectin when symptoms occur in a RCT (NCT00293644, due to end in April 2015)
rapid hydration with hospital admission for HG (ISRCTN24659467)
enteral feeding with oral rehydration for the treatment of HG.
Research recommendations
Given the trajectory of research and the evidence gaps identified by this study, in order of priority, the following recommendations for research are made:
A well-designed multicentre RCT to determine which second-line, hospital-prescribed therapy should be adopted as mainstream provision for the treatment of NVP in the UK NHS. This could compare day case or inpatient delivery of (a) a combination of antiemetic therapy (antihistamines or dopamine receptor antagonists or serotonin receptor antagonists) with i.v. rehydration (rapid i.v. hydration or standard inpatient care) as required; against (b) alternative antiemetic therapy (antihistamines or dopamine receptor antagonists or serotonin receptor antagonists) with i.v. rehydration (rapid i.v. hydration or standard inpatient care).
A well-designed multicentre RCT to test the use of subsequent treatments, such as steroids, as a third-line therapy. This could examine, for example, the effectiveness of corticosteroids versus serotonin receptor antagonists (ondansetron). In this trial, the indication for steroid use could also be investigated. For example, failure of successful treatment following three admissions for i.v. rehydration and medication. The dose, duration of treatment and effectiveness could also be examined.
A well-designed multicentre RCT to determine which second-line, GP-prescribed therapy should be adopted as mainstream provision for the treatment of NVP in UK primary care. This could compare the effectiveness and cost-effectiveness of vitamin B6–antihistamine combination against, for example, a dopamine receptor antagonist.
Six key factors should be incorporated within any future research trial of second-line, third-line or GP-initiated NVP/HG therapies:
Stakeholder co-design: all stakeholders and, in particular, the women themselves, should be involved in the design of future studies to ensure that future research produces information of relevance to them as well as to health services.
Embedded process evaluation in order to examine the views of women participants on the intervention/therapy being trialled, study how the intervention is implemented in practice, investigate contextual factors that affect the delivery of an intervention, and study the way effects vary in subgroups.
Mental health and well-being outcomes: we know from other sources that severe NVP has an impact on mental health both ante- and post-natally. Therefore consideration needs to be given as to how to capture the impact on women regarding these factors. A variety of approaches are possible, but it may for example include qualitative work and/or the use of existing validated tools such as (a) QoL indices such as Short Form questionnaire-36 items (total and subscales); (b) measures of mental health and well-being such as Edinburgh Postnatal Depression Scale; and (c) satisfaction with care as measured by Client Satisfaction Questionnaire-8.
Validated symptom severity scale: given the problems identified in this review with the lack of consistent outcome measures in the existing evidence base, entry to future trials should be based on PUQE criteria as an objective means of establishing severity, with subsequent outcome measurements also based on the same score.
Standard reporting criteria for adverse events, maternal and fetal outcomes: all studies of drugs administered during the first trimester of pregnancy should have obligatory reporting of rates of adverse events and maternal/fetal outcomes to the UKTIS central database to support the availability of reliable prospective controlled data. In addition, all trials of NVP/HG interventions in the UK NHS should include sufficiently long-term participant follow-up periods to enable the capture of relevant maternal and fetal outcomes given the safety profile of the therapies compared. These outcome data should be analysed to determine whether or not there are any associations between particular therapies and/or characteristics of women, in order to guide further research into stratified care.
Economic evaluation of NVP/HG therapies: as there may be trade-offs between cost and effects, future studies should also include an economic evaluation.
A large simple RCT of self-medication. The choice of interventions should be informed by the preferences of women about which treatments are of most relevance to them. Points 1–5 listed above also apply, but the inclusion of an economic component would also need to be justified given the anticipated low cost of the intervention and that a more effective treatment would save subsequent management costs.
