Table 1.
Article summary of study results included in this review
| Assessed conditions | ||||||||
|---|---|---|---|---|---|---|---|---|
| Authors | Study type | Sample | Maternal factors | Pregnancy complications | Obstetric complications | Neonatal factors | Main findings | Quality rating |
| Cnattingius et al. 1999 | Case-control study with randomly selected control (Swedish Inpatient Register—1973–1984) | AN (n = 781), HCs (n = 3905) | Maternal age | Hypertensive disease, diabetes, bleeding during pregnancy, multiparity | Inertia uteri, forceps, or vacuum delivery, preterm rupture of the membranes, cesarean section, cephalohematoma, other trauma | Gestational age, birth weight and birth weight for gestational age, Apgar score, jaundice | AN diagnosis was associated with maternal age, very preterm birth (< or = 32 weeks), small for gestational age in very preterm birth, cephalohematoma | 5 |
| Morgan et al. 2000 | Case-control study | BN (n = 935), patients with AN history and BN diagnosis (n = 227) general population data | – | – | – | Month of birth |
BN diagnosis is not associated with a specific month of birth AN history and BN diagnosis are associated with a peak season of birth in March |
4 |
| Shoebridge & Gowers et al. 2000 | Case-control study | AN (n = 40) and HCs (n = 40) | – | Antepartum hemorrhage | Toxemia, previous obstetric complications, forceps delivery, cesarean section | Mean birth weight, gestation length (full-term, 36 weeks, 32 weeks) APGAR score, the baby looked after in a special care baby unit | No association was found | 2 |
| Foley et al. 2001 | Twin cohort study (Virginia Twin Registry, n = 2352) | AN (n = 7), broadly defined AN (bdAN) (n = 71), BN (n = 42), broadly defined BN (bdBN) (n = 100), twin control (n = 1586) | – | High blood pressure, vaginal bleeding, seizures or toxemia, German measles, any other complication, any prenatal complication | Premature contractions, labor lasting more than 24 h, breech delivery, cesarean delivery, forceps delivery, cord wrapped around the neck, blue at birth, required an incubator, any perinatal complication | Birth weight, gestational age, jaundice, failure to breathe at first, convulsions, blood transfusion |
AN: low gestational age, # prenatal maternal complications bdAN: prenatal maternal complications BN: # prenatal maternal complications bdBN: # prenatal maternal complications |
3 |
| Feingold et al. 2002 | Historical prospective cohort (Thomas Jefferson University Hospital 1979–1981, n = 86) | Infants born preterm (n = 84) | Mother older than 40 years, alcohol abuse | Preeclampsia, eclampsia, chorioamnionitis, urinary tract infection, cervicitis, asthma, hypertension, hyperemesis gravidarum, cervical cancer, vaginal bleeding, multiparity | Premature rupture of the membranes, cesarean delivery, breech or transverse delivery, abruptio placentae, placenta previa, incompetent cervix, the total number of complications | Small for gestational age | No association with ED symptomatology for any of the studied variables | 4 |
| Lindberg and Hjern 2003 | Population cohort study (Swedish Register Data 1973–1982 n = 989,871) | AN (n = 1122), HCs (n = 988,749) | Maternal age | Preeclampsia | Premature rupture of the membranes, placental abruption, breech delivery, cephalohematoma, neonatal O2 necessity | Gestational age, size for gestational age, prematurity, Apgar score, distress | AN diagnosis was associated with maternal age (> 25 years), gestational age at birth (23–32 weeks), preeclampsia, premature rupture of the membranes, low Apgar, O2 necessity, and distress, cephalohematoma, breech delivery | 4 |
| Montgomery et al. 2005 | Cohort study (British cohort study 1970 n = 4046) | BN (n = 100), HCs (3946) | Maternal age at delivery, smoke habit, occupation, social class, psychiatric morbidity, BMI | – | – | – | Self-reported diagnosis of BN is associated with maternal smoke during pregnancy | 3 |
| Favaro et al. 2006 | Birth cohort study (Padua birth cohort 1971–1979) |
AN (n = 114), BN (n = 73), HCs (n = 554) |
Maternal age, social class | Bleeding, preeclampsia, maternal diabetes, threatened miscarriages, and anemia | Inertia uteri, breech delivery, premature rupture of the membranes, breech delivery, placental infarction or abruption, placenta previa, meconium staining of the amniotic fluid, forceps or vacuum extraction, the umbilical cord wrapped around the infant’s neck, cephalopelvic need for resuscitation O2, and intubation, disproportion, # of obstetric complications, cephalohematoma | Small for gestation age, birth weight, cyanosis, respiratory and cardiac problems, jaundice, neuromuscular disturbances such as hyporeactivity, hypotonia, and tremors, hypothermia |
AN diagnosis was associated with diabetes, anemia, preeclampsia, placental infarction, Ponderal Index < 25, neonatal cardiac problems, neonatal hyporeactivity, # of complications, the umbilical cord wrapped around the neck BN was associated with neonatal hyporeactivity, early feeding difficulties, placental infarction, low birth weight for gestational age, # of complications |
5 |
| Klump et al. 2006 | Twin cohort study (Michigan State Twin Study) | FSS (n = 113) | – | Levels of prenatal testosterone (indirectly studied by 2D:4D ratio) | – | – | ED symptomatology was associated with lower levels of prenatal testosterone (higher 2D:4D ratio) | 2 |
| Culbert et al. 2008 | Twin cohort study (Michigan State Twin Study) | FSS (n = 304), FOS (n = 59), MOS (n = 54), MSS (n = 165), HCs (n = 69) | – | Levels of prenatal testosterone (indirectly studied by OS and SS twin study) | – | – | Highest levels of disordered eating were observed for fSS twins, followed by fOS twins, mOS twins, and mSS twins | 3 |
| Favaro et al. 2008 | Birth cohort study (Padua birth cohort 1971–1979) | AN (n = 66), BN (n = 44), HCs (n = 257) | Maternal age |
Vaginal bleeding, preeclampsia, maternal diabetes, threatened miscarriages, and anemia |
Inertia uteri, breech delivery, premature rupture of the membranes, placental infarction or abruption, placenta previa, meconium staining of the amniotic fluid, forceps or vacuum extraction, the umbilical cord wrapped around the infant’s neck, cephalohematoma, need for resuscitation O2 and intubation | Small for gestational age, birth weight, cyanosis, respiratory and cardiac problems, jaundice, neuromuscular disturbances such as hyporeactivity, hypotonia, and tremors, hypothermia | Preterm birth and neonatal dysmaturity were associated with high harm avoidance in the offspring affected by EDs | 4 |
| Raevuori et al. 2008 | Twin cohort study (FinnTwin 16, 1975–1979 sample n = 2426) | 2426 female twins with known zigosity (OS dizygotic n = 793, SS dizygotic n = 765, monozygotic n = 868), 1,962 male twins (OS dizygotic n = 717, SS dizygotic n = 705, monozygotic n = 540) | – | Levels of prenatal testosterone (indirectly studied by OS and SS twin study) | – | – | Opposite-sex twin pairs were not significantly different from monozygotic or same-sex dizygotic twins (female) in the association with AN or BN | 4 |
| Baker et al. 2009 | Swedish twin study of child and adolescent development (TCHAD) 1985–1986 | n = 439 identical females, n n = 213 fraternal females, n = 461 identical males, n = 344 fraternal males, n = 371 opposite-gender twin pairs | – | Levels of prenatal testosterone (indirectly studied by twin study) | – | – | ED symptomatology was not associated with twin types | 3 |
| Nicholls and Viner 2009 | Prospective birth cohort (British cohort study, n = 16,567) | AN (n = 101), HCs (n = 11,261) | Smoke habit, ethnicity | Gestational diabetes, maternal anemia | Perinatal hypoxia | Birth weight, gestational age, prematurity (< 37 weeks) | No significant differences between AN and HCs | 3 |
| Smith et al. 2010 | Cross-sectional observational study | Male college students (n = 204) | – | Levels of prenatal testosterone (indirectly studied by 2D:4D ratio) | – | – | Higher testosterone exposure (lower 2D:4D ratio) was associated with less ED symptoms, greater drive for muscularity, diminished drive for leanness | 2 |
| Wehkalampi et al. 2010 | Case-control study (Helsinki Study of Very Low Birth Weight Adults 1978–1985 (n = 255) | Very low birth weight adults (n = 255), term (n = 189) | Maternal age, smoke habit, BMI, education | – | – | Gestational age, very preterm birth (< 32 weeks), very low birth weight—VLBW (< 1500 g) | In both sexes, EDI-2 scores were lower in VLBW individuals than in HCs | 3 |
| Coombs et al. 2011 | Observational study on a Midlands High School (UK) students | n = 132 pupils (age 11–14) | – | Levels of prenatal testosterone (indirectly studied by 2D:4D ratio) | – | – | No strong association was found | 2 |
| Favaro et al. 2011 | Birth cohort study (Padua Birth Cohort 1970–1984, n = 27,682) | AN (n = 402), HCs (n = 26,950) | Maternal education, maternal, and social class | Maternal exposure during pregnancy to chickenpox, measles, rubella, or influenza | – | Month of birth |
AN diagnosis was associated with the exposure to rubella and chickenpox at the 6th of pregnancy. Being born in June was associated with AN |
4 |
| Quinton et al. 2011 | Case-control study | AN (n = 25), BN (n = 26), HCs (n = 99) | – | Levels of prenatal testosterone (indirectly studied by 2D:4D ratio) | – | – |
AN was associated with low 2D:4D ratio (higher prenatal testosterone) BN: high 2D:4D ratio was associated with lower prenatal testosterone |
2 |
| Nosarti et al. 2012 | Historical population- based cohort study (Swedish Birth Register 1973–1985 and Hospital Discharge Register n = 1,301,522) | ED (n = 997), other psychiatric disorder and HCs | Maternal age, maternal education, maternal psychiatric family history | Parity | – | Gestational age, birth weight for gestational age, newborn sex, Apgar score at 5 min | ED: gestational week (<32 weeks) | 4 |
| Lydecker et al. 2013 | Three twin cohort study (MATR (US), NIPHTP (Norway), STAGE (Sweden)) |
US: OS (n = 481), SS (n = 1022) Norway: OS (n = 345), SS (n = 1430) Sweden: OS (n = 2433), SS (n = 7000) |
– | Levels of prenatal testosterone (indirectly studied by OS and SS twin study) | – | – | No association between co-twin sex and EDs | 4 |
| Allen et al. 2013 | Population cohort (Western Australian Pregnancy Cohort (Raine) n = 2900) | ED (n = 98), HCs (n = 428) | Maternal age, BMI, drinking alcohol, smoking cigarettes, education, family income | Serum 25(OH)D level at 18 weeks, kidney disease or dysfunction, urinary tract infection, thyroid dysfunction | – | Gestational age at birth, season of birth, weight, preterm birth (< 37 weeks), newborn sex |
EDs were associated with low quartile serum 25(OH)D level and season of birth (spring), kidney disease or dysfunction, sex (female) BN was associated with low quartile serum 25(OH)D level and season of birth (spring) |
3 |
| Culbert et al. 2013 | Twin cohort study (Michigan State Twin Study) | FSS, fOS, mOS, mSS (n = 394), HCs (n = 63) | – | Levels of prenatal testosterone (indirectly studied by OS and SS twin study) | – | – |
No differences were observed in the levels of disordered eating attitudes in opposite-sex and same-sex twins in pre-early puberty. During later phases of puberty, females from opposite-sex twin pairs exhibited lower disordered eating attitudes than females from same-sex twin pairs |
3 |
| Taborelli et al. 2013 | Case-control study (Sister Pair Study) | AN (n = 94), BN (n = 63), HCs (n = 157) | Anxiety during pregnancy (questionnaire) | – | Anxiety during pregnancy was associated with AN, but not BN | 4 | ||
| Vellisca et al. 2013 | Case-control study | AN (n = 210), general population data | – | – | – | Month of birth | AN diagnosis was not associated with the month of birth | 4 |
| Winje et al. 2013 | Case-control study | AN (n = 4045), HCs with the same year of birth, sex, and region of birth | – | – | – | Month of birth | No significant differences between AN and HCs | 4 |
| Goodman et al. 2014 | Population cohort study (Swedish Register Data, 1975–1998, n = 2,135,279) | AN (n = 7351), BN (n = 2804), other eating disorders (n = 10,408), HCs (n = 2015,862) | Maternal age, maternal education, number of full siblings, number of half-siblings, eating disorder in mother, multiple smoke habits | Multiparity | Premature rupture of the membranes, cesarean section, instrumental delivery, cephalohematoma, other birth trauma | Gestational age, birth weight for gestational age, birth length for gestational age, APGAR score |
AN was associated with maternal age, multiparity, lower gestational age (dose-response). Weak evidence of cesarean and instrumental delivery and birth trauma other than cephalohematoma AN was negatively associated with higher maternal weight and smoking BN was associated with higher birth weight for gestational age (dose-response) |
5 |
| Romero-Martínez and Moya-Albiol 2014 | Observational study | AN (n = 34), HCs (n = 40) | Maternal age, maternal BMI, right 2D:4D ratio, education, marital status of children | Levels of prenatal testosterone (indirectly studied by 2D:4D ratio) | – | – | Low 2D:4D ratio (higher prenatal testosterone) in AN. Salivary testosterone was negatively related to the 2D:4D ratio | 3 |
| Culbert et al. 2015 | Michigan State University Twin Registry |
Study 1 (2D:4D ratios): monozygotic (n = 229) and dyzigotic (n = 180) Study 2 (OS-F study): 1538 males and females and 131 non-twin females as an additional comparison group |
– | Levels of prenatal testosterone (indirectly studied by 2D:4D ratio and by OS-F twin study) | – | – | In both studies, higher prenatal testosterone exposure (lower 2D:4D, females from opposite-sex twin pairs vs controls) predicted lower disordered eating symptoms in early adolescence and young adulthood | 3 |
| Lofrano-Prado et al. 2015 | Cross-sectional observational study | College students (n = 408) | Mother’s age > 25 years | Number of obstetric complications, cesarean delivery | Low birth weight (< 2500 g), no breastfeeding, not first in the birth order |
Mother’s age lower than 25 years old is associated with AN symptoms BN symptoms are associated with # of obstetric complications |
4 | |
| Micali et al. 2015 | Cohort study (very preterm cohort n = 476) | VPT, < 33 weeks (n = 143) | – | – | Cesarean section, vaginal delivery | Birthweight, gestational age, ventricular dilatation, neonatal complications | ED symptomatology was not associated with any of the factors. Cesarean delivery was associated with compensatory behaviors. | 2 |
| St-Hilaire et al. 2015 | Cohort study (Project Ice Storm cohort, n = 54) | Teenagers born from cohort mothers (n = 54) | Maternal age, maternal education, social class, maternal stress exposure, trimester of stress exposure (Storm 24, IES-R scales) | – | – | Birth weight, length of gestation | Higher EAT-26 scores are associated with maternal exposure to stress in the third trimester | 3 |
| Tenconi et al. 2015 | Birth cohort study (Padua new Birth Cohort 1969–1997 and previous 1971–1979) |
New cohort: AN (n = 150), BN (n = 35), HCs (n = 73) Whole cohort: AN (n = 264), BN (n = 108), HCs (n = 624) |
Maternal age, social class |
Bleeding, preeclampsia, maternal diabetes, threatened miscarriages, and anemia |
Inertia uteri, breech delivery, premature rupture of the membranes, breech delivery, placental infarction or abruption, placenta previa, meconium staining of the amniotic fluid, forceps or vacuum extraction, umbilical cord wrapped around the infant’s neck, and cephalopelvic disproportion, # of obstetric complications, need for resuscitation, O2 and intubation, cephalohematoma | Small for gestation age, birth weight, cyanosis, respiratory and cardiac problems, jaundice, neuromuscular disturbances such as hyporeactivity, hypotonia, and tremors, hypothermia |
AN maintains its associations as shown in Favaro (2006), and higher maternal age and weight gain during pregnancy. Maternal diabetes and anemia lost their association BN is associated with neonatal hyporeactivity, early feeding difficulties, short and small for gestational age |
5 |
| Matinolli et al. 2016 | Population cohort study (ESTER 1985–1989 e AYLS 1985–1986) | Early preterm (n = 185), late preterm (n = 348), term-born control (n = 637) | Maternal weight, maternal smoke habit, socioeconomic status, education | Gestational diabetes, hypertension, preeclampsia, eclampsia | – | Early preterm birth (< 32 weeks) and late preterm birth (32–37 weeks) | EDI-2 scores were significantly lower in early-born preterms than in HCs , in particular in “Body dissatisfaction” and “Drive for thinness” subscales | 4 |
| Sacks et al. 2016 | Population cohort study (Soroka University Medical Center, Israel 1991–2014 (n = 231,271) | EDs (n = 486) | Maternal age, maternal obesity | Gestational diabetes mellitus (type A1 and type A2) | – | Gestational age at birth | ED and other psychiatric disorders in the offspring were associated with gestational diabetes mellitus | 2 |
| Su et al. 2016 | Population-based cohort (Denmark 1973–2000 (n = 1,034,539) and Sweden 1970–1997 (n = 1,246,560) | AN (n = 5878), BN (n = 1722), mixed ED (n = 3159), HCs (n = 2110,755) | Maternal loss of a close relative 1 year prior to or during pregnancy | – | – | – | ED, BN, and mixed ED were associated with maternal exposure to prenatal loss | 4 |
| Razaz et al. 2018 | Retrospective Swedish cohort study (1992–2002, n = 486,688) | AN (n = 2414) | Maternal BMI, years of education, maternal age at delivery, smoke habit | Multiparity | Vaginal delivery, vaginal instrumental delivery, elective cesarean section, emergency cesarean section | Birthweight for gestational age, gestational age at delivery |
AN was associated with higher maternal age, higher maternal education, multiparity, and preterm birth The rate of AN decreased with maternal overweight and obesity in a dose-response manner |
5 |
AN anorexia nervosa, BN bulimia nervosa, HC healthy control, ED eating disorder, EDI-2 Eating Disorder Inventory-2, EAT-26 Eating Attitude Test-26, BITE Bulimic Investigatory Test, FSS female same-sex twins, FOS female opposite-sex twins, MOS male opposite-sex twins, MSS male same-sex twins, SS same sex, OS opposite sex, VPT very preterm, BMI body mass index