Table 1 Characteristics of studies included in the systematic review
| Â | Â | Â | Â | Â | Â | Diet assessment | Â | ||
|---|---|---|---|---|---|---|---|---|---|
Reference | Country | N | Cases, n (%) | Population characteristics | Age, yearsa | Method | Period | Course | Outcome |
Cohort studies | Â | Â | Â | Â | Â | Â | Â | Â | Â |
 Brantsæter et al., 2011 [20] | Norway | 33,399 | 1,755 (5.3) Mild PE: 997 (3.0) Severe PE: 514 (1.5) Unclassified PE: 244 (0.7) | Nulliparous pregnant women recruited through a nationwide postal invitation in connection with their first routine ultrasonography examination between 2002 and 2008 | 28.4 ± 4.4 | 255-item FFQb | 17 to 22 weeks’ gestation | Diet only | PEc |
 Goodarzi Khoigani et al., 2012 [22] | Iran | 584 | 23 (3.9) | Pregnant women referred to 18 health centers and 12 private offices in Isfahan between 2009 and 2010 | Cases: 26.48 ± 4.21; non-cases: 25.60 ± 4.44 | 48-hour dietary recall | 11 to 15, 26, and 34 to 37 weeks’ gestation | Diet with supplements | PE |
 Haugen et al., 2009 [24] | Norway | 23,423 | 1,267 (5.4) | Nulliparous pregnant women recruited through a nationwide postal invitation in connection with their first routine ultrasonography examination between 2002 and 2008 | 58% of women age range: 20 to 29 | 255-item FFQb | 17 to 22 weeks’ gestation | Diet with and without supplements | PEc |
 Klemmensen et al., 2009 [26] | Denmark | 57,346 | All subtypes: 1,487 (2.6) Severe PE: 337 (0.6) | Pregnant women recruited when first visiting their general practitioner because of pregnancy between 1996 and 2002 | NR | 360-item FFQb | 25 weeks’ gestation | Diet with supplements | PEc |
 Qiu et al., 2008 [29] | USA | 1,538 | 64 (4.2) | Pregnant women attending prenatal care clinics affiliated with Swedish Medical Center and Tacoma General Hospital in Seattle and Tacoma, Washington between 1996 and 2002 | Mean ± SE: 32.2 ± 0.1 | 122-item FFQb | First trimester | Diet only | PE |
 Triche et al., 2008 [30] | USA | 1,681 | 63 (3.7) | Pregnant women recruited between 1996 and 2000 from 56 obstetric practices and 15 clinics associated with six hospitals in Connecticut and Massachusetts | Majority of women: range: 25 to 34 | Interview (chocolate foods and drinks) | Mean: 14.9 weeks’ gestation (range 6.1 to 24.3) | Diet only | PE |
 Saftlas et al., 2010 [31] | USA | 2,508 | PE: 60 (2.4) GH: 161 (6.4) | Pregnant women recruited at their first prenatal visit between 1988 and 1991 at 13 prenatal care practices in Connecticut | Majority range: 25 to 34 | Interview (chocolate foods and drinks) | First and third trimester | Diet only | PE and GH |
 Geraldo Lopes Ramos et al., 2006 [35] | Brazil | 1,052 | Mild PE: 52 (4.9); severe PE: 16 (1.5); GH: 36 (3.4) | Women who gave birth at Hospital de ClÃnicas de Porto Alegre | Mean sge: GH cases 26.1; mild PE cases 28.1; severe PE cases 23.3; non-cases 25.2 | Dietary interview | 1 day after delivery (retrospective, during pregnancy) | Diet only | PE and GH |
 Borgen et al., 2012 [37] | Norway | 32,933 | 1,703 (5.2) | Nulliparous pregnant women recruited through a nationwide postal invitation in connection with their first routine ultrasonography examination between 2002 and 2008 | 51% of women; age range: 25 to 29 | 255-item FFQb | 17 to 22 weeks’ gestation | Diet only | PEc |
 Brantsæter et al., 2009 [38] | Norway | 23,423 | 1267 (5.4) | Nulliparous pregnant women recruited through a nationwide postal invitation in connection with their first routine ultrasonography examination between 2002 and 2008 | 58% of women; age range: 20 to 29 | 255-item FFQb | 17 to 22 weeks’ gestation | Diet only | PEc |
 Chavarro et al., 2011 [39] | Denmark | 63,226 | 2,206 (3.49) | Pregnant women recruited when first visiting their general practitioner because of pregnancy between 1998 and 2003 | 29 ± 4 | 360-item FFQb | 25 weeks’ gestation | Diet only | PE |
 Clausen et al., 2001 [40] | Norway | 3,133 | 85 (2.7); early-onset PE: 27 (0.9); late-onset PE: 58 (1.8) | Caucasian pregnant women, representing all socioeconomic classes, delivering between 1994 and 1996 at Aker Hospital in Oslo | Cases 29.2 ± 4.9; non-cases 29.9 ± 4.5 | 180-item FFQb | Early second trimester | Diet with supplements | PE |
 Longo-Mbenza et al., 2008 [41] | Congo | 238 | PE: 7 (2.9) GH: 4 (1.7) | Pregnant women admitted to the Evangelical Hospital of Kimpese, located in a rural area, between 2002 and 2003 | Cases: 25.5 ± 7.2; non-cases: 27.4 ± 6.4 | Questionnaire (vegetables and meat) | First trimester | Diet only | PE and/or GH |
 Morris et al. 2001 [42] | USA | 4,314 | PE : 326 (7.6) GH: 747 (17.3) | Nulliparous pregnant women enrolled in a randomized clinical trial seeking prenatal care at university medical centers and affiliated clinics/hospitals in five communities | 84% of women mean: <25 | 24-hour dietary recall | 13 to 21 weeks’ gestation | Diet with supplements | PEc and GH |
 Oken et al., 2007 [43] | USA | 1,718 | PE: 59 (3) GH: 119 (7) | Pregnant women recruited at eight offices of Harvard Vanguard Medical Associates, a large multispecialty urban/suburban group practice in eastern Massachusetts, at first prenatal visit between 1999 and 2002 | ~90% of women age range: 20 to 40 | >140-item FFQb | First trimester | Diet with and without supplements | PE and GH |
 Olafsdottir et al., 2006 [44] | Iceland | 488 | PE: 19 (3.9) GH: 30 (6.1) | Randomly selected pregnant women, attending a routine first visit at the Center of Prenatal Care in Reykjavik from 1999 to 2001 | GH cases 29 ± 6; PE cases; 26 ± 4; non-cases: 28 ± 5 | 150-item FFQb | 11 to 15 weeks’ gestation | Diet with supplements | PE and/or GH |
 Ortega et al., 1999 [45] | Spain | 82 | 6 (7.3) | Pregnant women who were to deliver at the Cuenca INSALUD Hospital in Cuenca city area between 1990-1991 | Non-cases: 27.0 ± 3.9; cases; 26.2 ± 3.4 | 5-day dietary record | Third trimester | Diet with and without supplements | GH |
 Richardson et al., 1995 [46] | USA | 9,291 | 83 (3.8) (black women) and 185 (2.6) (white women) | Pregnant women who are members of a prepaid medical insurance plan, residing in the Oakland area of California and who delivered between 1959 and 1967 | Majority range: 20 to 34 | Interview (glasses of milk per day) | During pregnancy | Diet with and without supplements | PE |
 Rifas-Shiman et al., 2009 [47] | USA | 1,777 | 60 (3.4) | Pregnant women recruited at eight offices of Harvard Vanguard Medical Associates, a large multispecialty urban/suburban group practice in eastern Massachusetts, at first prenatal visit between 1999 and 2002 | 64% of women range: 25 to 35 | 166-item FFQb | First and second trimester | Diet only | PE |
 Rumbold et al., 2005 [48] | Australia | 229 | 38 (12) GH or PE, 20 (7) GH, 17 (5) PE | Women attending the antenatal clinic of the Women’s and Children’s Hospital for routine antenatal care in Adelaide between April and July 2001 | Cases: 28 ± 5 cases; non-cases: 28 ± 5 | 166-item FFQb | Mid to late pregnancy | Diet with supplements | PE or GH |
 Skajaa et al., 1991 [49] | Denmark | 965 | 13 (1.3) | Women recruited when attending one of two antenatal clinics in Aarhus between 1988 and 1989 | Median (range): 28.7 (18 to 45) | Interview | 30 weeks’ gestation (retrospective, previous 3 months) | Diet with supplements | PE |
 Tande et al., 2013 [50] | USA | 65 | 13 (20) | Nulliparous pregnant women recruited at the time of their first prenatal visit at a local obstetrics clinic | Mean ± SE: non-cases: 24.2 ± 0.62; cases: 25.3 ± 0.72 | 78-item FFQ | <14 weeks’ gestation | Diet with supplements | PE or GH |
 Timmermans et al., 2011 [51] | Netherlands | 3,187 | PE: 58 (1.8) GH: 165 (5.2) | Pregnant women living in Rotterdam delivering between 2002 and 2006 | 31.6 ± 4.0 | 293-item FFQb | Early pregnancy (median 13.5 wks, IQR 3.4) | Diet only | PE and GH |
Case-control studies | Â | Â | Â | Â | Â | Â | Â | Â | Â |
 Frederick et al., 2005 [21] | USA | 511 | 172 cases, 339 controls | Women delivering at Swedish Medical Center and Tacoma General Hospital in Washington between 1998 and 2001. Controls were normotensive women, delivering on the same day as a case, matched for parity and maternal age | Mean ± SEM: cases: 29.9 ± 0.5; controls: 30.6 ± 0.3 | 121-item FFQb | During postpartum hospital stay (retrospectively, 12 months prior to delivery) | Diet only | PE |
 Gulsen et al., 2012 [23] | Turkey | 247 | 92 cases, 155 controls | Pregnant women from Konya and neighboring cities hospitalized with pre-eclampsia at a clinic between 2004 and 2005. Controls were healthy pregnant women visiting the same institute for routine control | Range: 20 to 34: 65% of cases; 88% of controls | Questionnaire (7 food groups) | During pregnancy | Diet only | PE |
 Kesmodel et al., 1997 [25] | Denmark | 764 | PE 43 cases, 256 controls GH: 179 cases, 256 controls | Population-based nested case-control study of women who delivered at Aarhus University hospital between 1989 and 1991. Controls were evenly distributed over and covering all months of the period corresponding to the recall time in the case groups | NR | FFQ | Postpartum (retrospective, during pregnancy) | Diet with supplements | PE and GH |
 Marcoux et al., 1991 [27] | Canada | 928 | PE: 172 (505 controls) GH: 251 (505 controls) | Pregnant women who delivered in Quebec City or Montreal between 1984 and 1986. Controls were women who delivered immediately after the case in the same hospital and had not more than one elevated blood pressure reading after 20 weeks of pregnancy | PE cases: 26.0 ± 4.8; GH cases: 26.2 ± 4.3; controls: 26.1 ± 4.2 | 20-item FFQ (dairy foods only) | After delivery (retrospective, first 20 weeks’ gestation) | Diet only | PEc and GH |
 Paknahad et al., 2008 [28] | Iran | 92 | 46 cases, 46 controls | Pregnant women attending Al-Zahra and Shaheed Beheshti hospitals in Isfahan. Controls were normotensive pregnant women matched for age and parity | Mean ± SE: cases: 26.5 ± 0.89; controls: 24.6 ± 0.72 | FFQ and 24-hour dietary recall | Mean ± SD 33.7 ± 2.7 weeks’ gestation | Diet only | PE or GH |
 Wei et al., 2009 [36] | Canada | 337 | 92 cases (69 severe PE; 23 mild PE), 245 controls | Nulliparous pregnant women recruited within 48 hours after delivery in four hospitals in Quebec. Controls were normotensive pregnant women delivering during the same period as the case | Cases: 29.0 ± 5.2; controls: 29.1 ± 5.3 | Questionnaire | Within 48 hours after delivery | Diet only | PE |
 Al et al., 1994 [52] | Netherlands | 116 | 29 cases, 87 controls | Nested case-control study of pregnant women recruited at hospitals in the Maastricht region. Cases were matched for parity and hospital with controls with an uncomplicated pregnancy who delivered around the same time | Mean ± SEM: controls: 28.5 ± 0.35; cases: 27.9 ± 0.54 | FFQ and diet history | 22 weeks’ gestation | Diet only | PE |
 Atkinson et al., 1998 [53] | Zimbabwe | 374 | 180 cases, 194 controls | Pregnant women delivering at one of nine clinics from the Harare Maternity Hospital located in suburbs of Harare city, between 1995 and 1996. Most patients were from poor urban areas or migrated between rural and urban areas. The first healthy women admitted after each case was used as a control | Cases: 25.6 ± 6.4; controls: 24.8 ± 7.9 | Questionnaire (meat, poultry, fruit, fish, vegetables, and dairy) | During postpartum hospital stay (retrospectively, month prior to birth) | Diet only | PEc |
 Duvekot et al., 2002 [54] | Netherlands | NR | 163 cases | Pregnant women selected from a computer database and patient charts in two hospitals between 1991 and 1996. Controls were matched for age and delivery date | Median ± SD: cases: 28 ± 1; controls: 28 ± 0.3 | Questionnaire (milk consumption and calcium supplement use) | During pregnancy | Diet with supplements | PEc |
 Kazemian et al., 2013 [55] | Iran | 263 | 113 cases, 150 controls | Pregnant women referred to Shahid Akbarabadi Hospital between January and May 2011. Controls were pregnant women with normal blood pressure referred to this hospital for prenatal care, matched for gestational age | Cases: 28.73 ± 6.04; controls: 25.36 ± 4.84 | 148-item FFQb | At diagnosis (retrospectively, past 3 months) | Diet with supplements | GH |
 Reyes et al., 2012 [56] | Colombia | 402 | 201 cases, 201 controls | Pregnant women recruited from six Colombian cities between 2006 and 2009. Healthy pregnant controls were matched for age and selected from the same city of residence and the same hospital of delivery as the case | Cases: 26.45 ± 7.22; controls: 26.71 ± 7.21 | 90-item FFQb | Before delivery (retrospective, last 12 months) | Diet only | PE |
 Richards et al., 2014 [57] | South Africa | 192 | 96 cases, 96 controls | Women who delivered at the Maternity Centre at Groote Schuur Hospital and Mowbray Maternity Hospital in Cape Town between January and November 2010. Healthy pregnant women who delivered a live infant were matched with cases by ethnicity, gravidity, age, and gestational age at delivery | Cases: 24 ± 4.3; controls: 24 ± 4.4 | Questionnaire | After delivery | Diet only | PE |
 Schiff et al., 1996 [58] | USA | 138 | 48 cases, 90 controls | Pregnant women admitted to the EH Crump Women’s Hospital in Memphis, Tennessee between January 1994 and April 1995. Normal outpatients with no evidence of hypertension or proteinuria either at recruitment or delivery in the third trimester served as controls | Cases: 21.5 ± 5.4; controls: 20.1 ± 4.4 | >100-item questionnaire | During pregnancy | Diet with and without supplements | PEc |
 Sharbaf et al., 2013 [59] | Iran | 140 | 40 cases, 100 controls | Nulliparous pregnant women recruited within 48 hours after delivery in two hospitals in Tehran. Controls were normotensive pregnant women delivering during the same time as the case | Cases: 28 ± 4.1; controls: 27 ± 5 | Questionnaire | Within 48 hours after delivery | Diet only | PE |
 Zhang et al., 2002 [60] | USA | 368 | 109 cases, 259 controls | Women delivering at Swedish Medical Center and Tacoma General Hospital in Washington between 1998 and 2000. Controls were normotensive women, delivering on the same day of a case | Mean ± SEM: : cases: 31.1 ± 0.6; controls: 29.9 ± 4.5 | 121-item FFQb | During postpartum hospital stay (retrospectively, 12 months prior to delivery) | Diet only | PE |