Participants

In Gothenburg and Stockholm, Sweden, white pregnant women without previous VTE were given written information, at around gestational week 12, about the study. Enrollment proceeded from October 1998 to September 2002 when 500 carriers had been identified. The only additional criterion for inclusion was Swedish language comprehension. Women with low-dose aspirin medication or known coagulation disorders, with the exception of prothrombin gene mutation, were excluded ( Figure ).

Excluded participants at different stages of evaluation

Figures within brackets denote number of excluded participants.

APC , activated protein C; ASA , low-dose aspirin; FVL , factor V Leiden; LMWH , low-molecular-weight heparin; VTE , venous thromboembolism.

Kjellberg. FVL mutation and pregnancy. Am J Obstet Gynecol 2010 .

Women with APC resistance were genotyped for the FVL mutation.

All FVL carriers were given personal information about the thrombophilia and known consequences during pregnancy and other situations. The women were recommended to use compression stockings during pregnancy. A special questionnaire, aimed at identifying blood coagulation disorders and heredity for VTE, was filled out by participants. Intercurrent diseases and medications were recorded. Every fifth woman who tested negative for APC resistance served as a control.

Follow-up duration after delivery was at least 6 weeks. The FVL carriers were compared with controls concerning gestational age at delivery; gestational hypertension; mild and severe preeclampsia; the syndrome of hemolysis, elevated liver enzymes and low platelets (HELLP); eclampsia; IUFD; birthweight deviation; late miscarriage; placental abruption; neonatal asphyxia; and blood loss at delivery.

VTE and superficial thrombophlebitis and the number of clinical consultations on suspicion of VTE were recorded. Anticoagulant therapy or thromboprophylaxis lasting more than 3 days antenatally, multiple pregnancy, and miscarriage before 23 gestational weeks were considered to be confounders at comparison of VTE, blood loss, and obstetric complications. Chromosomal disorders, IUFD, and neonatal death because of prolonged premature rupture of membranes were considered to be confounders at comparison of birthweight deviation. The women with these confounders were excluded at the time of the respective calculations ( Figure ).

Four hundred twenty-three of the FVL carriers (85%) were tested for the prothrombin G20210 A allele.

According to local guidelines, low-molecular-weight heparin (LMWH) was recommended as thromboprophylaxis during pregnancy for conditions entailing increased risk of VTE, such as air travel, varicose veins, cesarean section, and immobilization.

The local Ethics Committee for Medical Research at the University of Gothenburg approved the study (June 18, 1997, L151-97).

Laboratory methods

Blood samples for determination of APC resistance by analysis of modified APC ratio were taken at gestational week 17 at the routine ultrasound visit. Blood samples were taken by careful venipuncture from a cubital vein following a 20 minute rest, using butterfly syringe and Vacutainer (Venoject, Terumo, Leuven, Belgium) techniques. The blood was centrifuged at 2000 × g for 20 minutes at room temperature within 1 hour of sampling. The plasma was carefully removed without stirring the buffy coat and aliquotted into polystyrene tubes. All plasma samples were stored at −20°C until analysis, carried out within a week.

The modified APC ratio was determined in plasma samples prediluted with Factor V-deficient plasma (COATEST APC Resistance V; Chromogenix AB, Mölndal, Sweden). The FVL mutation was confirmed by polymerase chain reaction (PCR) technique if the modified APC ratio was less than 2.0. The prothrombin G20210 A mutation was confirmed by PCR technique. The PCR assays were performed in an accredited laboratory (ISO 15189). The following controls were used in each run: 2 wild type, 2 heterozygous, 2 homozygous, and 2 blanks.

Criteria for diagnosis of complications

Blood pressure was measured in the sitting subject’s left arm at heart level after a 20 minute rest.

Chronic hypertension: a diastolic blood pressure of 90 mm Hg or greater before 20 weeks of gestation or ongoing antihypertensive medication.

Gestational hypertension: a diastolic blood pressure of 90 mm Hg or greater on two occasions, at minimum six-hour interval, after gestational week 20.

Mild preeclampsia: a diastolic blood pressure of 90 mm Hg or greater on 2 occasions and proteinuria (reading on dipstick of ≥1) on 2 occasions, at a minimum 6 hour interval, and negative urine culture.

Severe preeclampsia: mild preeclampsia but with 1 of the following additional findings:

• •
  

  Diastolic blood pressure of 120 mm Hg or greater once or diastolic blood pressure of 110 mm Hg or greater on 2 occasions, at minimum 6 hour interval.

  
  
• •
  

  Proteinuria of 3 g or greater per 24 hours.

  
  
• •
  

  Symptoms/signs of organ disorder, such as renal insufficiency, pulmonary edema, eclampsia, visual disturbances, IUGR, or platelet count less than 100 × 10 ⁹ /L.

HELLP syndrome: classified according to Martin et al (ie, epigastric pain, hemolysis, elevated serum alanine aminotransferase, and thrombocytopenia with or without other signs of preeclampsia). Hemolysis was not always analyzed. Women with the HELLP syndrome are included among those with severe preeclampsia.

Birthweight deviation: sex-specific intrauterine growth curves based on ultrasonically estimated fetal weights were used as a reference. The average birthweight deviation, in percent, from the reference fetal weight at the pertinent gestational age was determined as follows and compared in FVL carriers and controls: 100 × mean value of ([weight – mean weight for the same gestational age and gender]/[mean weight for the same gestational age and sex]).

Large for gestational age (LGA) and small for gestational age (SGA): birthweight above or below 2 SD, respectively. SGA served as a proxy for IUGR.

Placental abruption: abdominal pain, uterine tenderness, frequent small contractions, and elevated uterine tone together with vaginal bleeding or a retroplacental clot (ie, a condition leading to immediate delivery).

Neonatal asphyxia: Apgar score less than 7 at 5 and 10 minutes.

Late miscarriage: the term late in this study refers to miscarriages occurring during gestational weeks 17-22.

VTE: deep venous thrombosis (DVT) was diagnosed by triplex ultrasonography (combination of venous compression, color Doppler, and pulsed Doppler ultrasonography), magnetic resonance venography, or phlebography. Pulmonary embolism (PE) was diagnosed by spiral computed tomography or ventilation-perfusion lung scintigraphy according to Biello.

Superficial thrombophlebitis: diagnosed clinically (ie, palpable, tender vein, accompanied by a surrounding area of localized inflammation). If the thrombophlebitis was located near a perforating vein, DVT was ruled out by triplex ultrasonography.

Blood loss: routinely estimated and recorded by the midwife at delivery.

Statistics

The incidences of pregnancy-induced hypertension (preeclampsia included), premature delivery, and SGA in Sweden have been reported at 5.0%, 6.0%, and 2.5%, respectively. To achieve 80% power in finding a doubled relative risk of the complication with the lowest incidence in a 2-tailed test, 430 cases and 860 controls were required.

The 2-sample Student t test was used for analyses of age, body mass index, thromboprophylaxis frequency, birthweight difference, and blood loss in milliliters. Parity is a discrete variable, and because the number of observations in both groups was large, we assumed that the means in the groups were normally distributed and used the Student t test for comparisons.

Occurrence probability was compared for analyses of differences in smoking and essential hypertension. The relative risks for obstetric complications, blood loss greater than 1000 mL, superficial thrombophlebitis, and consultations for suspected VTE were calculated by cross-tabulation and with a 95% confidence interval. Blood loss at delivery is a positive variable with considerable relative variation and was thus analyzed on a logarithmic scale.

Six major comparisons of the primary endpoint outcomes in carriers and noncarriers were undertaken. Taking multiple testing into consideration, P values less than .008 were considered to be statistically significant.

All P values are 2 tailed except those for comparisons of VTE, thrombophlebitis, and blood loss, which are 1 tailed.