There is a dose dependent association between smoking and outcomes such as abruption, stillbirth, recurrent pregnancy loss, decline in ovarian reserve and fertility, intrauterine growth restriction and placental insufficiency. The evidence is particularly strong with smoking > 20 cigarettes per day. The recommendation is that women stop smoking in pregnancy and evidence suggests that those that do are likely to permanently quit smoking. A review comparing different options of nicotine replacement to help women to stop suggests that after the exclusion of studies with bias, there is no difference in the fetal outcomes between the treatment and the placebo group [8]. The SNAP trial of nicotine replacement reported that there was survival without developmental disorder at the 2 year follow up for babies, making this a safe option for women, but no more likely to be successful than placebo [9].

There is heavy public advertising and marketing of e-cigarettes as a safe alternative to smoking. Though e-cigarettes reduce the harm from carbon monoxide, carcinogens and toxins that contribute to lung cancer, e-cigarettes are still addictive secondary to nicotine release. A published article from 2016 confirms the positive effect of reduction of harm secondary to the carcinogens and toxins in the smoke [10]. There is no published evidence of benefit of use in pregnant women. Until more evidence is available, the advice would be not to recommend e-cigarettes due to variable amounts of caffeine and absorption of inhalational agents of unproven safety.

Alcohol affects fertility and pregnancy outcomes in a dose dependent manner with the well recognised fetal alcohol syndrome relating to excessive alcohol consumption in pregnancy. Consumption of greater than 7 units of alcohol/week is associated with growth restriction and can lead to increased risk of behavioural problems and learning difficulties in children [11]. The current national guidelines also comments that there is no health benefit related to alcohol [12]. Standard advice should be to stop alcohol consumption ideally prior to a planned pregnancy and that there is no safe limit in pregnancy.

There is no published evidence to suggest an adverse pregnancy outcome secondary to excessive exercise. Due to the detrimental effects of obesity in pregnancy the advice is to continue with moderate exercise during pregnancy. The review of the effect of aerobic exercise on pregnancy is that it improves maternal fitness [13].

There is a dose related effect on pregnancy with an increased risk of miscarriage at higher levels. Women should be advised to limit the amount of caffeine to 150 mg/day; equivalent to two cups of normal coffee or three cups of black tea. One study has shown an adverse profile if levels are greater than 300 mg/day [17]. However, the Cochrane review concluded that there was insufficient evidence to correlate the fetal outcomes with maternal caffeine consumption due to the low quality of the studies [18].

Acupuncture is a well-established mode of complimentary therapy that aims at helping women cope with the stress of subfertility, miscarriage and pregnancy. There are well-established studies that have shown improvements in coping with stress after acupuncture but not in the prevention of miscarriage [19].

Reproductive outcomes are worse in both underweight (BMI < 18) and obese women (BMI > 30). Any further excessive weight gain in obese women accentuates the adverse perinatal and neonatal outcomes. There is an increased risk of intra-uterine growth restriction, stillbirth, operative deliveries and increased morbidity secondary to infection and thromboembolism. Unfortunately diet and exercise interventions in pregnancy have not shown an impact on neonatal outcome (Cochrane meta-analysis 2015) [20]. However, lifestyle interventions do reduce excessive maternal weight gain and development of maternal hypertension and possibly reduce the risk of caesarean section. Therefore pre-pregnancy advice on diet and exercise to women with high BMI should be routine.

There is no established safety profile for appetite suppressants in pregnancy. The recently published EMPOWAR study [21] did not show any benefit in terms neonatal outcomes for metformin supplementation in women with obesity. The study published in NEJM randomised women with BMI > 35 to metformin supplementation or placebo showed a reduction in maternal weight gain but no difference in the neonatal weight in the treatment arm [22].

The Cochrane review suggests that luteal phase progesterone during assisted reproduction improves pregnancy and live birth rates [23]. However, once pregnancy is achieved it is less clear when to discontinue the treatment. Offering luteal phase support for an extended period of time do not appear to result in more clinical benefits, or to cause more harm, than a short period of luteal phase support. While the evidence on this is limited, NICE suggests that it is biologically plausibile for luteal phase support to be effective for up to 8 weeks after embryo transfer, after which time the pregnancy is self supporting.

A review of 14 randomized controlled trials (2158 women) found no evidence that routine use of progestogens can prevent miscarriages [24]. No difference in the incidence of adverse effects on either the mother or baby was apparent. There was evidence that women who have suffered three or more miscarriages may benefit from progestogen during pregnancy. Four trials showed a decrease in miscarriage compared with placebo or no treatment in these women; however, the trials were of poorer methodological quality so these findings should be interpreted with caution. The recently published robust large multicentre double blinded randomised controlled trial (PROMISE trial) [25] did not show any reduction in the miscarriage rates or improvement in live birth rates in women who suffered from recurrent miscarriages and randomised to progesterone support or placebo in early pregnancy

In the case of threatened miscarriage, a systematic review of trials located four randomised studies involving 421 women that compared the use of progestogens in the treatment of threatened miscarriage with either placebo or no treatment [26]. The limited evidence suggests that the use of a progestogen does reduce the rate of spontaneous miscarriage. Two trials reported that treatment with progestogens did not increase the occurrence of congenital abnormalities in the newborns and the women did not have any significant difference in incidence of pregnancy-induced hypertension and antepartum haemorrhage. Further larger studies are warranted for firmer conclusions. The on-going PRISM trial is powered to definitively answer the question as to whether there is a role for progesterone in early pregnancy bleeding [27].

The Cochrane database suggests that in antiphospholipid syndrome and recurrent miscarriage, unfractionated heparin is effective at preventing miscarriage but this was not confirmed in the one trial using low molecular weight heparin (LMWH) [28]. However, the evidence from unfractionated heparin was of such a large magnitude that it is now routine practice to give LMWH to women with antiphospholipid syndrome and recurrent miscarriage. There is a dearth of evidence as to the management of women with inherited thrombophilia with conflicting results from the published studies. The on-going multicentre randomised controlled trial (ALIFE2) [29] will shed light on the treatment of women with inherited thrombophilia especially with recurrent pregnancy loss. Peri-implantation LMWH may improve implantation in IVF but the trials are of insufficient quality to draw firm conclusions [30]. LMWH has been demonstrated to have no effect at preventing miscarriage during pregnancy in idiopathic recurrent miscarriage in several trials [31].

Low dose aspirin is indicated for women who are considered high risk for developing pre-eclampsia. In the United Kingdom low dose aspirin is recommended to be commenced from 12 weeks of pregnancy until labour in women at high risk of hypertensive disease in pregnancy [32]. A systematic review showed no benefit from low dose aspirin in preventing miscarriage in unexplained recurrent miscarriage. In one large randomised controlled trial there was a lower live birth rate in women with unexplained recurrent miscarriage in women taking aspirin than placebo and so should not be used for this indication [33].

One small study suggested an improvement in pregnancy outcomes in women treated with prednisolone with raised uterine natural killer cells and recurrent miscarriage [34]. Prednisolone in early pregnancy has been associated with gestational diabetes, preterm birth. Until large randomised controlled trails have established the efficacy of prednisolone, its use as a treatment option remains in research settings only.

The immune mechanisms of recurrent implantation failure and recurrent pregnancy loss postulate the rejection of the embryo by the mother. Injection of paternal leucocytes in early pregnancy was done initially to overcome the postulated rejection phenomenon. Intravenous immunoglobulin has been subject to a series of randomised controlled trials and has potential side effects including anaphylaxis. Paternal cell immunization, third-party donor leukocytes, trophoblast membranes, and intravenous immunoglobulin provide no significant beneficial effect over placebo in improving the live birth rate. TNF-Alpha use is associated with severe reactions such as immunosuppression and granulomatous disease.