Walking and Dietary Modification for Recurrent Early Miscarriages

Overview

This study is part of a big one aiming to evaluate how lifestyle interventions during pregnancy affect obstetric results, neonatal metabolism and the intelligence of the offspring (study not yet completed). Data regarding obstetric and neonatal results were entered in NCT01409382, but we decided to split results in two for the sake of clarity. A cohort of women with early pregnancy losses without antiphospholipid antibodies was selected for two reasons. One is that these women follow strictly the recommendadtions. The second is that no medication has been shown to increase the rate of take-home babies in women with early miscarriages who test negative for antiphospholipid antibodies. We decided to focus on the fibrinolytic system because trophoblast migration and placental vasculogenesis and angiogenesis depend on plasmin-dependent extracellular matrix remodeling. Plasminogen activator inhibitor (PAI)-1 inhibits the generation of plasmin. Since both glucose and insulin increase PAI-1 synthesis, hyperglycemia itself, or by stimulating insulin production, reduces plasmin generation, which may impair placentation. Abnormalities in glucose metabolism may be also deleterious to embryos by causing epigenetic changes. Chromosomal abnormalities are considered an important cause of early pregnancy losses. Several lines of evidence lend support to the hypothesis that carbohydrate metabolism abnormalities contribute to the pathogenesis of recurrent early pregnancy losses. One is that of the pregnancies of the women with polycystic ovary syndrome, around 30 and 50% end with first-trimester miscarriages. Hyperinsulinemia is a prevalent feature of the syndrome, and interventions proven effective in reducing insulin levels, such as metformin, have been shown to reduce the rate of early miscarriages. The other is that patients with body mass index of ≥25 kg/m2 have significantly higher odds of early miscarriage, regardless of the method of conception. The investigator's hypothesis was that a balanced diet combined to regular exercise, by improving glucose homeostasis, would increase the take-home baby rate in women with consecutive early miscarriages. Moderate exercises are usually well tolerated not only by the mother, but also by the fetus, as indicated by tests of fetal well-being, including umbilical artery systolic to diastolic ratio.

Women aged 18 to 40 years trying to conceive spontaneously were eligible if they had two or more consecutive pregnancy losses in the first trimester, documented by pathology or ultrasound-confirmed gestational sac. All participants underwent ultrasound examination before inclusion in the study. Exclusion criteria were any of the following: anatomic anomalies that may increase the risk of pregnancy losses, not amenable to surgical correction during pregnancy, such as uterine septum; antiphospholipid antibodies; prior second- or third-trimester losses; current multiple gestation; disabilities such as hemiplegia or paraplegia; renal or liver failure; conditions requiring a priori anticoagulation. Participants were enrolled by staff at the participating center. Randomization to the intervention protocol Walking and Diet (W\&D) or to a control group was performed before pregnancy occurred by a statistician using a computer-generated random-number table. This was not a double blind study, but care was taken to ensure that appointments of the patients assigned to the intervention protocol did not coincide with those of controls. The intervention was standardized by training of research staff. Women assigned to W\&D were instructed to walk at a moderate pace (4 km/h) for at least 40 minutes, seven days a week. Besides, those who remained seated most of the day were advised to walk 25 to 30 minutes twice a day, avoiding hence more than 12 hours of physical inactivity. Walking could be replaced by stationary bicycle rides or swimming when convenient, which often occurred near term and when the mother was obese. Strenuous exercises were discouraged. Patients assigned to protocol W\&D were also informed of the importance of a balanced diet and recommended to avoid high-glycemic index meals (high-carbohydrate, low-fiber). Sucralose could be used as a sweetener. As a strategy to promote satiety and reduce carbohydrate intake, W\&D participants were also advised to eat at least two daily servings of protein-rich food. The intervention began when participants wished to conceive, continuing until delivery. Careful instructions about walking speed and diet were given to participants assigned to W\&D at enrolment and at each consultation. During exercise, neither fetal nor maternal cardiac rate were assessed. Non-adherence to the intervention protocol was suspected when non-obese participants assigned to W\&D gained \> 1 kg in 4 weeks until the 28th week of gestation, \> 1.5 kg from the 28th to the 32nd week, and \> 2 kg in 4 weeks thereafter, in the absence of edema. The threshold was lower for obese participants: \> 700 g in 4 weeks until the 28th week of gestation, \> 1 kg from the 28th to the 32nd week, and \> 1.5 kg in 4 weeks thereafter21. Excessive weight gain aroused the suspicion of protocol violation because high carbohydrate consumption, especially when combined with physical inactivity stimulates the pancreas to overproduce insulin, a hormone that promotes fat storage. To enhance adherence to the protocol, W\&D participants who gained excessive weight were recommended to increase the frequency, duration and intensity of the physical activity and to increase the protein intake. Participants of the W\&D group who had a successful pregnancy volunteered to encourage mothers assigned to the intervention protocol, especially those who gained excessive weight. At enrollment and during first-trimester consultations, W\&D participants were explained that antiemetics such as ondansetron should be taken before nausea became severe, in order to help them tolerate balanced meals. No recommendations regarding diet or physical activity were given to controls. Antiemetics such as ondansetron were given to controls who complained of hyperemesis. All participants were given folic acid tablets 5 mg daily until 14 weeks of gestation, as prevention of neural tube defects. In both groups, participants reporting abdominal pain, cramps, and vaginal bleeding during the first-trimester were medicated with vaginal progesterone. Subcutaneous heparin was given to all participants whose pregnancies were complicated with nephrotic range proteinuria or any evidence of placental insufficiency. Antihypertensive medications included methyldopa, amlodipine and hydralazine. No patient received aspirin or metformin in this study. Appointments were scheduled according to the routine. Maternal weight and blood pressure were assessed at every appointment and all mothers were screened for gestational diabetes according to the American Diabetes Association recommendations. Obstetric and neonatal outcomes were obtained from the hospital records. Neonates were classified as small, appropriate or large for gestational age according to Olsen et al. growth curves. Written informed consent was obtained from each participant after a full explanation of the study.