There are various possible etiological processes for nausea and vomiting of pregnancy (NVP), none of which or all of which may play a part. Perhaps, NVP is the final common pathway. The pathogenesis of NVP is poorly understood, and proposed etiologies abound (e.g., endocrine, psychological, immunological, genetic incompatibility, carbohydrate metabolism, vitamin B6, gastric dysrhythmia, hyperolfaction). This may be due, at least in part, to the fact that radiological and pharmacological investigations have been hampered y concerns about teratogenicity, especially during the first trimester of pregnancy.
Fifty to 90% of women suffer from NVP. It is a problem that has been recognized for a very long time; the first report is in a papyrus from about 2000 B.C. It should also be considered that there must be an evolutionary drive for such a common condition, that is, a benefit for the species. Obviously, the chemoreceptor trigger zone and the vomiting centre in the medulla must be involved because that is how one feels nauseated and vomits. However, the processes leading to that point are unclear.
NVP has been associated with many clinical risk factors such as younger maternal age, nulliparity, low socioeconomic status, unplanned pregnancy, non-smoking status in some studies, passive smoking in others, previous pregnancy complicated by NVP, dependent personality trait, increased body mass index, eating disorders, ethnicity (e.g., Polynesian women in New Zealand have more nausea and vomiting than non-Polynesians), and female foetal gender, possibly indicating an immune mechanism.1-5 However, none of these factors allow for prediction of either the occurrence or severity of NVP in an individual patient, and none contribute to our understanding of the underlying pathogenesis. Idiopathic NVP must be distinguished from NVP of known etiology (e.g., hydatidiform mole, multiple gestation). Non-pregnancy related causes (e.g., gastrointestinal, genitourinary, central nervous system, toxic/metabolic) must also be considered and ruled out.
During pregnancy women complain that they either cannot stand the smell of coffee or the taste of this or that. They could normally sit in the back of the car and read, but now they need their eyes fixed directly on the road ahead. That is, there are specific trigger odours for certain women and hypersensitivity to motion in others. It is postulated that these repetitively described symptoms are nature's protective measures, making the woman more aware of potentially noxious agents in her environment so that she will protect herself; perhaps she will stay closer to home if she feels sick. Teleologically, is the priming of these modalities a protective mechanism?6
NVP has also been proposed to be due to either blunted autonomic nervous system function or dysregulation of gastric rhythms, as measured by cutaneous electrogastrograms (EGGs).7-10 If one is going to vomit, it is obvious that some degree of gastric dysrhythmia must be present. But is gastric dysrhythmia cause or effect? EGGs are subject to confounders such as high intra- or inter-observer variability and poor specificity.
There is no proven role for gestational hormones in the etiology of NVP. No consistent correlation has been found between the severity of NVP and the serum levels of human chorionic gonadotropin (hCG), despite the temporal relationship (i.e., that hCG levels peak during the first trimester), and despite the fact that conditions associated with high levels of hCG (e.g., molar pregnancy, multiple gestation) are associated with NVP. Neither oestrogen nor 17-hydroxyprogesterone levels have been correlated with NVP; although a correlation has been found with oestradiol levels (e.g., women who suffer nausea from the combined birth control pill are said to have an increased risk of having nausea and vomiting when pregnant, and situations where increased oestrogens are associated with NVP such as nulliparity, increased body mass index and non-smoking). However, the nature of the NVP-oestradiol relationship has been inconsistent amongst authors and studies. Relaxin also seems to have the right time frame to be a possible etiological agent.
Raised thyroid hormone levels have been observed, but not consistently.11-14 When observed, it can be unclear as to whether the patient has thyrotoxicosis that is causing nausea and vomiting, or whether hCG, especially asialo-hCG, is stimulating the thyroid directly.15-17 Anti-thyroid medication should not be administered if the patient is clinically euthyroid.
NVP has been associated with a deficiency of vitamin B6 and vitamin B complex (pyridoxine is a component of Diclectin). These observations have led directly to changes in therapy, possibly through optimization of maternal nutrition. However, it is not known whether deficiency of vitamin B6 has a causative effect for NVP, or whether raised levels of the vitamin in pregnant women without NVP are pharmacological in alleviating symptoms.18
Psychological factors such as depression, anxiety, anticipatory fear, and eating disorders have also been observed among patients with NVP. However, it is not clear whether or not these preceded or resulted from severe symptomatology.
The question of whether NVP is an immunological response in pregnancy has also been mentioned in the literature. Fairweather, in the 1960s, proposed that NVP might be an allergic phenomenon, i.e., an immunological phenomenon. There has been little investigation of this possible process and no direct associations have been made. However, pregnancy is associated with various immune changes, such as depressed cell-mediated immunity and primed non-specific immunity.
In summary, the etiology of NVP has not been established, and it is likely to be multifactorial. I fear that our view to date has been, too often, down a single track and too simplistic. I believe that there is no single etiological factor for NVP other than pregnancy itself. Viewpoints are sometimes held with a vigour that is inversely proportional to the objective evidence. This is an area in need of active research. It is possible that patients with different underlying abnormalities may respond differently to various therapies which can then be targeted more appropriately.