Metabolic syndrome in Finnish women 7 years after a gestational diabetes prevention trial

Me abolic synd ome in Finnish women 7 yea s a e a ges a ional diabe es p e en ion ial Ja a Puhkala, 1 Jani Rai anen, 1,2 Päi i Kolu, 1 Pipsa Tuominen, 1 Pauliina Husu, 1 Rii a Luo o 1,2 To ci e: Puhkala J, Rai anen J, Kolu P, e al. Me abolic synd ome in Finnish women 7 yea s a e a ges a ional diabe es p e en ion ial. BMJ Open 2017;7:e014565. doi:10.1136/bmjopen-2016- 014565 ▸P epublica ion his o y o his pape is a ailable online. To iew hese iles please isi he jou nal online (h p://dx.doi.o g/10.1136/ bmjopen-2016-014565). Recei ed 3 Oc obe 2016 Re ised 2 Janua y 2017 Accep ed 13 Janua y 2017 1 UKK Ins i u e o Heal h P omo ion Resea ch, Tampe e, Finland 2 School o Heal h Sciences, Uni e si y o Tampe e, Tampe e, Finland Co espondence o Ja a Puhkala; ja[email p o ec ed] ABSTRACT Backg ound: Risk o de eloping me abolic synd ome (MeS) a e deli e y is high among women wi h ges a ional diabe es melli us (GDM), bu li le is known abou de elopmen o MeS among women wi h isk ac o s o GDM du ing p egnancy. In he p esen s udy, we s udied he p e alence o MeS 7 yea s pos pa um among women wi h GDM isk ac o s du ing p egnancy, women wi h ea ly GDM diagnosis and women wi hou GDM isk ac o s. We also analysed he ea ly p egnancy isk ac o s associa ed wi h MeS. Me hods: A Finnish clus e andomised con olled GDM p e en ion ial was conduc ed in 2007–2009. The p e alence o MeS acco ding o In e na ional Diabe es Fede a ion c i e ia was de e mined in he ollow-up s udy 7 yea s a e o iginal ial. Eligible pa icipan s (n=289) in 4 s udy g oups (in e en ion (n=83) and usual ca e (n=87) wi h GDM isk ac o s; ea ly GDM (n=51), and heal hy con ol wi hou GDM isk ac o s (n=68)) we e e alua ed o MeS. Bina y logis ic eg ession models we e used o analyse isk ac o s associa ed wi h MeS. Resul s: 7 yea s pos pa um, he MeS p e alence was 14% (95% CI 8% o 25%) in he in e en ion g oup; 15% (CI 8% o 25%) in he usual ca e g oup; 50% (CI 35% o 65%) in he ea ly GDM g oup and 7% (CI 2% o 18%) in he heal hy con ol g oup. OR o MeS in women wi h GDM isk ac o s did no di e om he heal hy con ol g oup. Body mass index (BMI)-adjus ed OR o MeS was 9.18 (CI 1.82 o 46.20) in he ea ly GDM g oup compa ed wi h he heal hy con ol g oup. Inc eased p ep egnancy BMI was associa ed wi h MeS (OR, 1.17, CI 1.08 o 1.28, adjus ed o g oup). Conclusions: Inc eased p ep egnancy BMI and ea ly GDM diagnosis we e he s onges isk ac o s o de eloping MeS 7 yea s pos pa um. O e weigh and obese women and especially hose wi h ea ly GDM should be moni o ed and counselled o ca diome abolic isk ac o s a e deli e y. INTRODUCTION Me abolic synd ome (MeS) is a clus e ing o a he oscle o ic isk ac o s, including abdom- inal obesi y, ele a ed plasma iglyce ides, dec eased high densi y lipop o ein (HDL) choles e ol, ele a ed blood p essu e and ele a ed as ing glucose. 12 Insulin esis ance and low-g ade inflamma ion a e cen al d i e s in he pa hogenesis o MeS. 34 In add- i ion o gene ic ea u es, obesogenic li es yle defined by unheal hy die and physical inac i i y inc ease he isk o MeS. 5–9 The p e alence o MeS is inc easing wo ldwide, along wi h he obesi y epidemic and popula- ion ageing. 10 11 Es ima es o MeS p e alence a y be ween 15% and 34% in adul popula- ions, 12–15 and a ound 8–19% in ep oduc - i e-aged women in indus ial coun ies. 16–18 I is ele an o iden i y pa ien s wi h MeS, as hey a e a wice he isk o de eloping ca - dio ascula disease o e he nex 5–10 yea s, and a fi e old li e ime isk o de eloping ype 2 diabe es compa ed wi h indi iduals wi hou he synd ome. 1 Ges a ional diabe es melli us (GDM) is a diso de in glucose and insulin me abolism fi s diagnosed du ing p egnancy. 19 The mos impo an isk ac o s o GDM a e p e- p egnancy o e weigh , high ma e nal age and a amily his o y o ype 2 diabe es. 20 Women wi h a his o y o GDM a e a S eng hs and limi a ions o his s udy ▪Ou s udy was a long- e m ollow-up o a la ge ges a ional diabe es melli us (GDM) p e en ion ial. ▪The s udy was one o he i s ollow-up s udies on he p e alence o me abolic synd ome (MeS) a e deli e y among women wi h isk ac o s o GDM. ▪The esul s include da a o he MeS p e alence among di e en isk g oups. ▪Thi y-se en pe cen o he in i ed women en e ed he measu emen s o MeS, and he ou subg oups we e ela i ely small. ▪The esul s may ha e been mode a ely a ec ed by a heal hy selec ion bias. Puhkala J, e al.BMJ Open 2017;7:e014565. doi:10.1136/bmjopen-2016-014565 1 Open Access Resea ch g oup.bmj.com on May 16, 2017 - Published by h p://bmjopen.bmj.com/Downloaded om inc eased isk o de eloping ca diome abolic diso de s, such as ype 2 diabe es and MeS a e deli e y. 21 22 MeS and GDM sha e mu ual isk ac o s, sugges ing ha women wi h isk ac o s o GDM may be a inc eased isk o u u e MeS, e en in he absence o GDM. 23 24 O e weigh and excessi e ges a ional weigh gain ha e been linked o de elopmen o GDM and la e ca dio- me abolic diso de s. 23 25–27 Long- e m ollow-up s udies on de elopmen o MeS among women a inc eased GDM isk a e lacking. Ou clus e andomised con olled ial (RCT) in Finnish p egnan women a inc eased GDM isk showed ha li es yle counselling was e ec i e in con olling he p opo ion o la ge- o -ges a ional-age new-bo ns and imp o ing he women’s die , and p e en ing excessi e ges a ional weigh gain and dec ease in physical ac i - i y. 28–31 One yea a e deli e y, he p e alence o MeS among women wi h GDM isk ac o s was 16–18% depending on he c i e ia; wi h no di e ences be ween he in e en ion and usual ca e g oup. 24 Ou aim is o s udy MeS and i s componen s 7 yea s pos pa um among Finnish women who in ea ly p eg- nancy we e a inc eased isk o de eloping GDM (in e - en ion o usual ca e), among women wi h ea ly p egnancy GDM diagnosis and among women wi hou GDM isk ac o s du ing p egnancy (heal hy con ol). We also s udied isk ac o s associa ed wi h he de elop- men o MeS 7 yea s pos pa um. METHODS Pa icipan s and s udy design The s udy is a 7-yea ollow-up o a clus e RCT on GDM p e en ion (ISRCTN33885819). De ailed desc ip ions o he design and me hods o he o iginal s udy ha e been published p e iously. 32 The s udy was conduc ed in 14 p ima y heal hca e ma e ni y clinics in Wes e n Finland in 2007–2009, a 1-yea pos pa um ollow-up in 2009–2011 and he 7-yea ollow-up in 2014–2016. The p ima y aim o he ial was o p e en GDM and excessi e ges a ional weigh gain among p egnan women wi h an inc eased isk o GDM. Ma e ni y clinics we e andomised in o se en in e en ion and se en con ol clinics. P egnan women we e ec ui ed by nu ses a hei fi s isi (8–12 weeks’ges a ion) in ma e ni y clinics. Women we e eligible i hey had a leas one o he ollowing GDM isk ac o s: age≥40 yea s, p ep egnancy body mass index (BMI) ≥25 kg/m², GDM o any sign o glucose in ole ance in any p e ious p egnancy, a mac osomic baby (≥4500 g) in any p e i- ous p egnancy o diabe es in fi s -deg ee o second- deg ee ela i es. The main exclusion c i e ia we e age<18 yea s, a GDM diagnosis a 8–12 weeks’ges a ion, win p egnancy, physical es ic ions ha p ecluded exe cise o a clinical his o y o ch onic disease. A diag- nosis o GDM was based on a 2-hou 75 g o al glucose ole ance es (OGTT) acco ding o a leas one o he ollowing c i e ia: as ing plasma glucose o ≥5.3 mmol/L, >10.0 mmol/L a 1 hou o >8.6 mmol/ La 2hou s. 33 The o iginal in e en ion included indi idual counsel- ling on weigh gain, die and physical ac i i y by public heal h nu ses du ing fi e ou ine isi s o ma e ni y clinics. The women in he con ol clinics ecei ed usual ma e nal ca e, which also included some li es yle ad ice. O 888 p egnan women who pa icipa ed in he base- line assessmen s a 8–12 weeks’ges a ion, 442 (50%) we e eligible o he o iginal s udy (in e en ion and usual ca e g oup) (figu e 1). Fu he , 374 we e excluded, o whom 174 due o a GDM diagnosis a 8–12 weeks’ges a ion (ea ly GDM g oup). The heal hy con ol g oup consis ed o 176 women who did no mee he inclusion c i e ia. Se en hund ed and eigh y-eigh women we e in i ed o pa icipa e in he ollow-up s udy 7 yea s (mean 7.2, ange 5.6–8.3) a e deli e y, and 289 women pa icipa ed. The s udy was conduc ed in acco dance wi h he Decla a ion o Helsinki (2000), and i was app o ed by Pi kanmaa Uni e si y Hospi al Dis ic E hics (Re . numbe R14039). The pa icipan s comple ed a w i en in o med consen be o e pa icipa ion. Labo a o y measu emen s In o ma ion on ma e nal measu emen s by he nu ses was ob ained om he s anda d ma e ni y ca ds. Heigh was measu ed a he fi s ma e ni y ca e isi (8– 12 weeks’ges a ion), and weigh was measu ed a each ma e ni y ca e isi , and a 1 and 7 yea s pos pa um. Wais ci cum e ence was measu ed a 1 and 7 yea s pos - pa um, and he a e age o h ee measu emen s was used in he analysis. Blood p essu e was measu ed in duplica e a each ma e ni y ca e isi and a 1 and 7 yea s pos pa um. Blood samples o glucose, choles e ol, HDL choles- e ol and iglyce ide analysis a e a 12-hou as , and a 2-hou OGTT we e collec ed a baseline and a 1 and 7 yea s pos pa um. All blood analyses we e pe o med a he UKK Ins i u e. Du ing he OGTT, he pa icipan s d ank 75 g glucose in 330 mL wa e (Glucodyn, Ul imed, Finland), and he samples we e aken a e 60 and 120 min. Plasma glucose concen a ions we e mea- su ed esh wi hin 24 hou s a e he OGTT. Plasma samples o lipid analysis we e s o ed ozen a −80°C un il analysed. Ci ic acid/fluo ide and EDTA ubes we e used o glucose and lipid analysis o enous blood. Glucose, o al choles e ol, HDL choles e ol and iglyce - ide concen a ions we e de e mined in enzyma ic assays using a Roche Cobas Mi a Plus analyse . All es ing was pe o med in duplica e. MeS was diagnosed acco ding o In e na ional Diabe es Fede a ion (IDF) c i e ia. 1 The diagnosis was se i any h ee o he fi e c i e ia we e p esen : inc eased wais ci cum e ence (>80 cm); ele a ed igly- ce ides (≥1.7 mmol/L), o specific medica ion; educed HDL choles e ol (<1.3 mmol/L), o specific medica ion, ele a ed as ing glucose (≥5.6 mmol/L), o specific 2Puhkala J, e al.BMJ Open 2017;7:e014565. doi:10.1136/bmjopen-2016-014565 Open Access g oup.bmj.com on May 16, 2017 - Published by h p://bmjopen.bmj.com/Downloaded om medica ion, and ele a ed blood p essu e (≥130 o 85 mm Hg), o specific medica ion. S a is ical analyses The backg ound cha ac e is ics and desc ip i e in o ma- ion on componen s and he p e alence o MeS a e epo ed as means and SDs o equencies and p opo - ions. The Wilson sco e me hod wi hou con inui y co ec ion was used o calcula e 95% CIs o p e alence. Aχ 2 es was used o in es iga e whe he dis ibu ions o ca ego ical a iables di e ed om one ano he . Con inuous a iables we e no mally dis ibu ed. We used one-way analysis o a iance o compa e means ac oss g oups. Di e ences in con inuous a iables be ween he heal hy con ol o he ea ly GDM g oup compa ed wi h he o he h ee g oups me ged oge he we e es ed by independen samples - es . Bina y logis ic eg ession models we e used o ob ain ORs and hei 95% CIs o s udy associa ions be ween MeS and i s explana o y a i- ables. Explana o y a iables we e g oup ( he in e en- ion, he usual ca e, he ea ly GDM and he heal hy con ol as a e e ence g oup), age (con inuous), BMI (con inuous) and he fi e GDM isk ac o s which we e used in en ance c i e ia o he o iginal s udy: BMI ≥25 kg/m², age≥40 yea s, GDM o any sign o glucose in ole ance in any p e ious p egnancy, a mac osomic baby (≥4500 g) in any p e ious p egnancy, and diabe es in fi s -deg ee o second-deg ee ela i es. Based on he ule o 10 e en s pe a iable, mul i a iable logis ic eg ession models we e adjus ed only o BMI and GDM o any sign o glucose in ole ance in any p e ious p eg- nancy. The esul s we e conside ed o be s a is ically sig- nifican i p<0.05. All analyses we e pe o med wi h SPSS so wa e (V.20.0). RESULTS Backg ound cha ac e is ics Table 1 shows he cha ac e is ics o he 289 women who pa icipa ed in he ollow-up s udy 7 yea s pos pa um. The mean age o he pa icipan s was 37.8 yea s ( ange 25–52 yea s) and he mean numbe o deli e ies was 2.5 ( ange 1–9). Twen y-nine (10%) women smoked e- quen ly o occasionally. The mos common inclusion c i- e ia o he s udy we e p ep egnancy o e weigh o obesi y, and diabe es in fi s -deg ee o second-deg ee ela i es. Women wi h ea ly GDM ulfil he inclusion c i- e ia o being o e weigh (p<0.001) and ha ing a his o y o GDM (p=0.002) mo e o en han o he s. Heal hy con- ols did no mee inclusion c i e ia (GDM isk ac o s) a baseline. The p e alence o o e weigh and obesi y 7 yea s pos pa um (53%, 69%, 80%) had s ayed a ound he same le el, o sligh ly inc eased, compa ed wi h ea ly p egnancy (48%, 68%, 78%) among women in he in e en ion, usual ca e and ea ly GDM g oup, espec - i ely. Fou women (2 in he ea ly GDM and 2 in he heal hy con ol g oup) had medica ion o ype 2 dia- be es. Six een (5%) women had medica ion o dyslipi- daemias and eigh (3%) o hype ension, wi h no Figu e 1 Flow diag am o he s udy. The wo in e en ion g oups (in e en ion and usual ca e), ea ly ges a ional diabe es melli us and heal hy con ol g oups we e in i ed o ollow-up measu emen s 7 yea s a e in e en ion. Puhkala J, e al.BMJ Open 2017;7:e014565. doi:10.1136/bmjopen-2016-014565 3 Open Access g oup.bmj.com on May 16, 2017 - Published by h p://bmjopen.bmj.com/Downloaded om di e ence be ween g oups. Eigh y-one (28%) women epo ed ho monal con acep ion. Women wi h ea ly GDM had highe mean weigh (p<0.001) and BMI (p<0.001) compa ed wi h women in he o he h ee g oups. Significan associa ion be ween g oup and o e weigh (p<0.001) and obesi y (p<0.001) was ound, so ha women wi h ea ly GDM we e mo e o en o e weigh and obese han women in he o he g oups. Heal hy con ols, on he o he hand, had lowe mean weigh (p<0.001) and lowe BMI (p<0.001). They we e less o en o e weigh o obese compa ed wi h women in he o he h ee g oups, bo h o he associa- ions being significan (p<0.001). The e we e no o he be ween-g oup di e ences in he backg ound cha ac e - is ics. Compa ed wi h baseline (a 8–12 weeks’p eg- nancy), he women had gained weigh on a e age 3.1 kg (SD 7.0, ange −33.2 o 25.1 kg). The g oups did no di e s a is ically significan ly om each o he . The heal hy con ol g oup seemed o ha e had gained less weigh han he o he h ee g oups (1.8 s 3.5, p=0.13). Twel e (9%) women o he in e en ion g oup and se en (5%) women o he usual ca e g oup and none o he heal hy con ol g oup had been diagnosed o GDM a 26–28 weeks’p egnancy. MeS and i s componen s 7 yea s pos pa um The p e alence o MeS among all women was 19% (95% CI 15% o 25%) ( able 2). Among women wi h GDM isk ac o s, he p e alence was 14% (95% CI 9% o 21%), wi h no di e ence be ween he in e en ion (14%, 95% CI 8% o 25%) and he usual ca e g oup (15%, 95% CI 8% o 25%). The p e alence was highes among he ea ly GDM g oup (50%, 95% CI 35% o 65%) and lowes among he heal hy con ol g oup (7%, 95% CI 2% o 18%). Two (11%) o he 19 women wi h he GDM diagnosis a 26–28 weeks’ges a ion had MeS a 7 yea s pos pa um. Se en y- wo pe cen o he women exceeded he wais ci cum e ence limi o 80 cm (inc eased wais ci cum e - ence) and almos hal exceeded he limi o 88 cm (abdominal obesi y). HDL was educed (≤1.3 mmol/L) among hal o he women. All women wi h he MeS diagnosis exceeded he wais ci cum e ence o 80 cm. Women in he ea ly GDM g oup had a la ge mean wais ci cum e ence (p<0.001), highe as ing glucose (p<0.001), highe sys olic blood p essu e (p=0.047) and highe TG (p<0.0001). They mo e o en exceeded he limi s o inc eased wais ci cum e ence and abdominal obesi y (p<0.001 and p=0.002), impai ed as ing glucose (p<0.001) and inc eased TG (p<0.001) compa ed wi h he h ee o he g oups. Heal hy con ols had a smalle mean wais ci cum e ence (p<0.001), lowe sys olic (p=0.01) and dias olic (p=0.03) blood p essu e and highe HDL (p=0.02), and hey less o en exceed he limi s o inc eased wais ci cum e ence (p<0.001) and dec eased HDL (p=0.03) compa ed wi h he o he Table 1 Backg ound cha ac e is ics o women a 7 yea s pos pa um In e en ion (n=83) Usual ca e (n=87) Ea ly GDM (n=51) Heal hy con ol (n=68) All (n=289) Age (yea s) 37.7±4.5 38.0±4.9 38.6±4.3 37.0±5.2 37.8±4.8 Weigh (kg) 75.6±17.3 76.9±13.1 84.5±16.3 62.9±7.6 74.6±15.7 BMI (kg/m 2 ) 27.2±5.5 28.1±5.1 30.1±5.6 22.7±2.5 26.9±5.4 BMI 25–29.9 kg/m 2 17 (20) 39 (45) 41 (35) 12 (18) 86 (29) BMI ≥30 kg/m 2 27 (33) 21 (24) 23 (45) 0 (0) 71 (25) Smoking No 73 (91) 77 (91) 45 (94) 57 (84) 252 (90) Occasionally o daily 7 (9) 8 (9) 3 (6) 11 (16) 29 (10) Educa ion Basic o seconda y educa ion 25 (31) 25 (28) 19 (37) 20 (31) 89 (31) Poly echnic educa ion 28 (34) 40 (46) 20 (39) 29 (45) 117 (41) Uni e si y deg ee 29 (35) 23 (26) 13 (25) 16 (25) 81 (28) Pa i y 2.4±1.0 2.6±1.2 2.6±1.6 2.5±1.0 2.5±1.2 1 8 (10) 12 (14) 6 (13) 6 (9) 32 (12) 2–3 66 (84) 60 (71) 35 (74) 50 (75) 211 (76) ≥4 5 (6) 12 (14) 6 (13) 11 (16) 34 (12) GDM isk c i e ia (a 8–12 weeks’ p egnancy) BMI ≥25 kg/m 2 40 (48) 59 (68) 39 (78) NA 138 (48) Mac osomic child in any p e ious p egnancy 3 (4) 3 (3) 5 (10) NA 11 (4) GDM in any p e ious p egnancy 13 (16) 7 (8) 13 (39) NA 33 (12) Diabe es in i s -deg ee o second-deg ee ela i es 53 (64) 43 (49) 25 (50) NA 121 (42) Age ≥40 yea s 3 (4) 5 (6) 0 (0) NA 8 (3) Means and SDs o equencies (and p opo ions) o pa icipan s (n=289). BMI, body mass index; GDM, ges a ional diabe es melli us. 4Puhkala J, e al.BMJ Open 2017;7:e014565. doi:10.1136/bmjopen-2016-014565 Open Access g oup.bmj.com on May 16, 2017 - Published by h p://bmjopen.bmj.com/Downloaded om g oups. The e we e no di e ences in he componen s MeS be ween he in e en ion and usual ca e g oups. Fi y-fi e (27%) women did no mee he c i e ia o any MeS componen : 14 (23%) women in he in e en- ion g oup, 14 (22%) in he usual ca e g oup, 7 (18%) in he ea ly GDM g oup and 20 (46%) in he heal hy con ol g oup. Mos women (n=115, 56%) me one o wo MeS componen s. Risk ac o s o MeS Uni a iable bina y logis ic eg ession models In uni a iable logis ic eg ession models, he OR o de eloping MeS 7 yea s pos pa um among women wi h inc eased GDM isk du ing p egnancy (in e en ion and usual ca e) did no di e s a is ically significan ly om he heal hy con ol g oup ( able 3). OR o MeS among women wi h ea ly GDM was 21.0 (95% CI 4.47 o 98.7) compa ed wi h he heal hy con ols, and 6.0 (95% CI 2.7 o 13.2) compa ed wi h he in e en ion and he usual ca e g oups (no p esen ed). P ep egnancy BMI (OR 1.20, 95% CI 1.11 o 1.29, p<0.001), p ep egnancy o e weigh o obesi y (OR 8.06, 95% CI 3.21 o 20.2, p<0.001) and GDM o any sign o glucose in ole ance in any p e ious p egnancy (OR 3.21, 95% CI 1.34 o 7.69, p<0.01) we e associa ed wi h inc eased occu ence o MeS. The o he GDM isk ac o s (mac osomic baby in any p e ious p egnancy, diabe es in ela i es o age) we e no associa ed wi h he MeS occu ence 7 yea s pos pa um. Mul i a iable bina y logis ic eg ession models In a mul i a iable eg ession model 1, adjus ed o p e- p egnancy BMI, he OR o de eloping MeS 7 yea s pos pa um among women wi h inc eased GDM isk du ing p egnancy (in e en ion and usual ca e) did no di e significan ly om he heal hy con ol g oup ( able 3). Women wi h ea ly p egnancy GDM had inc eased odds o MeS compa ed wi h he heal hy con ol g oup (OR 9.18, 95% CI 1.82 o 46.20, p=0.007). Also inc eased p ep egnancy BMI was associa ed wi h inc eased odds o MeS (OR 1.20, 95% CI 1.11 o 1.29, p<0.001). In a mul i a iable eg ession model 2, adjus ed o BMI and GDM o any sign o glucose in ole ance in any p e ious p egnancy, ea ly GDM when compa ed wi h he heal hy con ol g oup (OR 97.75, 95% CI 1.50 o 40.0, p=0.014), and inc eased p ep eg- nancy BMI (OR 1.19, 95% CI 1.09 o 1.31, p<0.001) we e associa ed wi h he MeS occu ence. GDM o any sign o glucose in ole ance in any p e ious p egnancy was no associa ed wi h he MeS occu ence 7 yea s pos pa um. D opou analyses Da a o 7-yea ollow-up o MeS we e a ailable o 37% o he women who we e in i ed o pa icipa e a e 7 yea s (n=788). O he women in i ed o ollow-up s udy, 17% we e no willing o pa icipa e, 26% we e ou o each and 20% only comple ed a ollow-up ques ion- nai e. D opou analyses showed ha he women pa ici- pa ing he 7-yea ollow-up s udy we e olde a he baseline (30.2 s 29.2, p<0.01), we e less o en equen o occasional smoke s be o e p egnancy (17% s 30%, p<0.001) and we e highe educa ed (uni e si y deg ee 28% s 19%, p<0.009) compa ed wi h he non- pa icipan s. The e we e no di e ences be ween he g oups in o he backg ound cha ac e is ic (BMI, pa i y Table 2 Componen s o me abolic synd ome (MeS) and he p e alence o MeS acco ding o In e na ional Diabe es Fede a ion c i e ia In e en ion (n=64–74) Usual ca e (n=66–78) Ea ly GDM (n=40–45) Heal hy con ol (n=44–51) All (n=217–248) Missing alues (N)* Wais ci cum e ence (cm)* 89.9±13.0 90.9±12.8 97.9±14.6 79.5±6.5 89.5±13.5 0, 3, 0, 0 Wais ≥80 cm 52 (70) 59 (79) 42 (93) 24 (47) 177 (72) Wais ≥88 cm 41 (55) 42 (56) 32 (71) 6 (12) 121 (49) Fas ing glucose (mmol/L) 5.2±0.5 5.2±0.4 5.6±0.5 5.2±0.4 5.3±0.5 5, 12, 3, 6 Fas ing glucose ≥5.6 mmol/L o medica ion 10 (15) 7 (11) 21 (49) 4 (9) 42 (19) Sys olic p essu e (mm Hg) 119±13 118±12 122±11 114±14 118±12 2, 3, 1, 4 Dias olic p essu e (mm Hg) 77±10 78±10 79±7 75±10 77±10 2, 3, 1, 4 Blood p essu e ≥130 o ≥85 mm Hg o medica ion 10 (14) 13 (17) 8 (18) 6 (13) 37 (16) HDL choles e ol (mmol/L) 1.30±0.27 1.28±0.38 1.28±0.36 1.41±0.30 1.32±0.34 10, 0, 2, 1 HDL choles e ol ≤1.3 mmol/L o medica ion 33 (52) 41 (53) 25 (58) 18 (36) 117 (50) T iglyce ides (mmol/L) 0.91±0.34 1.05±0.67 1.36±0.88 0.95±0.38 1.05±0.61 10, 0, 2, 1 T iglyce ides ≥1.7 mmol/L 1 (2) 9 (12) 12 (28) 5 (10) 27 (12) Me abolic synd ome 9 (14) 10 (15) 20 (50) 3 (7) 42 (19) 10, 7, 5, 7 Means and SDs o equencies (and p opo ions) o pa icipan s. GDM, ges a ional diabe es melli us. *Numbe o missing alues in he in e en ion, usual ca e, ea ly GDM, heal hy con ol g oups, espec i ely. Puhkala J, e al.BMJ Open 2017;7:e014565. doi:10.1136/bmjopen-2016-014565 5 Open Access g oup.bmj.com on May 16, 2017 - Published by h p://bmjopen.bmj.com/Downloaded om o GDM isk c i e ia), labo a o y measu emen s o g oup dis ibu ion. DISCUSSION Se en yea s pos pa um, MeS was diagnosed among 14% (95% CI 9% o 21%) o he women wi h isk ac o s o GDM du ing p egnancy. The MeS p e alence was highes among women wi h a diagnosis o ea ly GDM (50%, 95% CI 35% o 65%), and lowes among he heal hy con ol g oup (7%, 95% CI 2% o 18%). Diagnosis o ea ly GDM and inc eased p ep egnancy BMI we e he s onges isk ac o s o de eloping MeS. P ep egnancy and p esen o e weigh o obesi y we e common among women wi h GDM isk ac o s, and especially among hose wi h ea ly GDM. In a p ospec i e Finnish popula ion s udy among women aged 36 and 39, he MeS p e alence was 18% and 23% acco ding o he IDF defini ion. 18 Compa ed wi h hese le els, he MeS p e alence in ou s udy was somewha lowe among women wi h an inc eased isk o GDM. The e o e, inc eased GDM isk seemed no o be associa ed wi h a highe MeS p e alence 7 yea s pos - pa um. Compa ed wi h he women aged 35–45 and 25– 60 yea s in a Finnish popula ion s udy in 2012, howe e , he women wi h inc eased GDM isk we e mo e o en obese (28% s 16%), 34 and abdominally obese (wais ci - cum e ence >88 cm) (56% s 30%). 35 These di e ences a e ele an , as especially abdominal obesi y is he mos impo an independen ac o in he de elopmen o MeS. 10 The finding may indica e u u e isk in ou pa i- cipan s. O e weigh was also he mos common inclusion c i e ia (GDM isk c i e ia) in ou s udy, which appa - en ly had an influence on he high p e alence o o e - weigh and obesi y pos pa um. Howe e , he in e en ion (14%, 95% CI 8% o 25%) and usual ca e (15%, 95% CI 8% o 25%) g oup did no di e om each o he in he p e alence o MeS o i s componen s a e 7 yea s, e en hough counselling had posi i e e ec s on women’s weigh gain and li es yle du ing p egnancy. 28–31 Howe e , 7 yea s a e a long ime, and se e al ac o s ha e a ec ed women’s li es yle. Mo he s wi h small child en may ha e had lack o ime and may ha e go i ed o ollow he ecommended li e- s yle habi s which we e discussed du ing he in e en- ion. Fu he , pa icipa ion in a GDM p e en ion s udy may ha e imp o ed heal h consciousness and li es yle habi s in he in e en ion and he usual ca e g oup. Addi ionally, also usual ca e includes some li es yle coun- selling in Finland. D opou a e was a he high, bu since i was non-di e en ial o g oup, he possibili y o bias in he di e ences be ween g oups emains low. Ou subg oups di e ed in he p e alence o MeS. Women in he ea ly GDM and women in he heal hy con ol g oups ep esen ed he ex emes (50% s 7%). Ea lie s udies ha e ound ha his o y o GDM is s ongly associa ed wi h a highe isk o u u e MeS and o he ca diome abolic diso de s. 36 37 In a Finnish Table 3 Logis ic eg ession models wi h ORs and 95% CIs and p alues (n=217) o occu ence o me abolic synd ome (acco ding o In e na ional Diabe es Fede a ion c i e ia) by i s explana o y a iables (g oup, age, body mass index and i e ges a ional diabe es melli us isk ac o s) Uni a iable logis ic eg ession models Mul i a iable logis ic eg ession model 1 Mul i a iable logis ic eg ession model 2 OR (95% CI) p Value OR (95% CI) p Value OR (95% CI) p Value G oup ( e e ence=heal hy con ol) In e en ion 3.44 (0. 71 o 16.8) 0.13 1.64 (0.31 o 8.79) 0.56 1.29 (0.23 o 7.18) 0.77 Usual ca e 3.15 (0.65 o 15.3) 0.16 1.39 (0.26 o 7.41) 0.70 1.17 (0.21 o 6.41) 0.85 Ea ly GDM 21.0 (4.47 o 98.7) <0.001 9.18 (1.82 o 46.20) 0.007 7.75 (1.50 o 40.0) 0.014 Age (ea ly p egnancy, con inuous) 1.07 (1.00 o 1.15) 0.052 BMI (p ep egnancy, con inuous) 1.20 (1.11 o 1.29) <0.001 1.17 (1.08 o 1.28) <0.001 1.19 (1.09 o 1.31) <0.001 GDM isk ac o s du ing ea ly p egnancy BMI ≥25 kg/m 2 (p ep egnancy) 8.06 (3.21 o 20.2) <0.001 A mac osomic baby (≥4500 g) in any p e ious p egnancy 0.74 (0.09 o 6.34) 0.78 GDM o any sign o glucose in ole ance in any p e ious p egnancy 3.21 (1.34 o 7.69) <0.01 2.07 (0.74 o 5.78) 0.17 Type 1 o 2 diabe es in i s -deg ee o second-deg ee ela i es 1.04 (0.52 o 2.09) 0.91 Age ≥40 yea s 0.74 (0.09 o 6.34) 0.78 GDM, ges a ional diabe es; BMI, body mass index. Uni a iable logis ic eg ession models: sepa a e models o each explana o y a iables wi hou adjus men s. Mul i a iable logis ic eg ession model 1: adjus ed o one explana o y a iable (BMI). Mul i a iable logis ic eg ession model 2: adjus ed o wo explana o y a iables (BMI, and GDM o any sign o glucose in ole ance in any p e ious p egnancy). 6Puhkala J, e al.BMJ Open 2017;7:e014565. doi:10.1136/bmjopen-2016-014565 Open Access g oup.bmj.com on May 16, 2017 - Published by h p://bmjopen.bmj.com/Downloaded om coho among 240 women, he isk o de eloping MeS 2–6 yea s a e p egnancy complica ed by GDM was 2.4- old highe compa ed wi h no mal p egnancy. 21 In ou ea lie epo o he same o iginal ial, he MeS p e alence among women wi h ea ly GDM was al eady ele a ed (31%) a 1-yea pos pa um compa ed wi h he in e en ion g oup (11%). 24 Se en yea s pos pa um, he MeS p e alence among women wi h ea ly GDM had inc eased o 50%. Fu he , women wi h ea ly GDM had inc eased odds o MeS compa ed wi h o he h ee g oups (in e en ion, usual ca e and heal hy con ols). The associa ion emained a e adjus men o BMI. Also his o y o GDM was associa ed wi h he MeS occu ence. These findings a es he la e ca diome abolic isk associa ed wi h GDM. Ne e heless, GDM diagnosis a 26–28 weeks’ges a ion among he 19 women in he in e en ion and usual ca e g oup seemed no o be associa ed wi h highe MeS p e alence in ou da a. The e o e, ou s udy sug- ges s ha GDM was associa ed wi h an inc eased MeS p e alence 7 yea s pos pa um i i was diagnosed in ea ly p egnancy (8–12 weeks). Women wi h ea ly GDM a e a special high isk g oup o la e ca diome abolic diso de s. A la ge me a-analysis ound ha women wi h ea ly p egnancy GDM we e a a wo old isk o u u e ype 2 diabe es compa ed wi h women wi h la e GDM. 26 Se e al ca diome abolic isk ac o s we e clus e ed in he women in he ea ly GDM g oup. In ea ly p egnancy, hey had been mo e o en o e weigh o obese, and 7 yea s pos pa um, hey s ill we e mo e o en o e weigh o obese, and had componen s o MeS mo e equen ly han he o he women. Fu he , compa ed wi h Finnish emale popula ion, he women wi h ea ly GDM in ou s udy we e h ee imes mo e o en obese (45% s 16%), 34 and o e wo imes mo e o en abdominally obese (wais ci cum e ence >88cm) (71% s 30%). 35 P esen obesi y pa ly explains he inc eased MeS p e a- lence. Clus e ing o li es yle- ela ed isk ac o s is common. 38–40 P egnancy- ela ed weigh e en ion ends o be highes among hose women who ha e been o e - weigh o obese be o e p egnancy. 41 42 Fu he , many women e ain weigh a e p egnancy, and he mean weigh gain associa ed wi h pa i y is a ound 0.5–3.2 kg, wi h a high pe sonal a ia ion. 39 43 44 In a Finnish popu- la ion s udy, pa i y- ela ed weigh e en ion was asso- cia ed wi h isce al obesi y. 45 High p e alence o obesi y p omo ing insulin esis - ance was p obably he common denomina o o he inc eased p e alence o ea ly p egnancy GDM and MeS a e 7 yea s. 46 47 This finding clea ly unde lines ha i is pa icula ly impo an o in es in li es yle counselling among o e weigh and obese p egnan women en e ing ma e ni y ca e. P e en ion o excessi e weigh gain du ing p egnancy and p omo ing pos pa um weigh educ ion by indi idually ailo ed li es yle modifica ions ha e been e ec i e agains weigh e en ion, and in p e- en ing u he ca diome abolic diso de s, such as MeS in pa ous women. 48–52 A weigh educ ion o 5–10% is associa ed wi h significan dec eases in ca diome abolic isk ac o s. 53 54 The possibili ies o eaching p egnan women in many coun ies a e good due o ma e ni y ca e sys ems, bu he e is o en lack o long- e m ollow-up as he heal hca e p o ide s change. 55 Ac ions o inc ease coope a ion be ween di e en p o ide s a e needed. Heal hy con ols we e heal hie han he o he h ee g oups wi h GDM isk ac o s, as assessed by com- ponen s and he p e alence o MeS a 7 yea s pos - pa um. Also compa ed wi h he Finnish popula ion, hey ep esen ed an unexcep ionally heal hy g oup, in wha comes o he p e alence o obesi y (0% s 16%), abdominal obesi y (12% s 30%) as well as MeS (7% s 17–23%). 34 35 The esul s sugges ha keeping no mal weigh and p e en ion o excessi e p egnancy- ela ed weigh e en ion a e impo an in no only p e en ing GDM, bu also in p e en ing la e MeS. S eng hs and limi a ions Ou s udy has se e al s eng hs. I was a long- e m ollow-up, om ea ly p egnancy un il 7 yea s pos - pa um. I was also one o he fi s ollow-up s udies o ges a ional ials, in which he p e alence o MeS has been de e mined among women wi h isk ac o s o GDM. To he bes o ou knowledge, he only p e ious ollow-up epo was ou s on he MeS p e alence among women wi h GDM isk ac o s a 1-yea pos pa um. 24 In gene al, he e is a limi ed amoun o s udies o he MeS p e alence among ep oduc i e-aged emale popula- ions, e en hough obesi y is apidly inc easing wo ld- wide, especially among young adul s. 11 56 Ou s udy esul s include da a o he MeS p e alence among di e - en isk g oups—women wi h an inc eased GDM isk, women diagnosed wi h ea ly GDM and heal hy p egnan women. Also associa ions be ween isk ac o s assessed in ea ly p egnancy and MeS 7 yea s pos pa um ha e no been iden ified ea lie . Ou o iginal ial 32 was one o he la ges RCTs abou p e en ing he de elopmen o GDM in women wi h GDM isk ac o s. A e 7 yea s, he ec ui men o he women o he o i- ginal s udy was di ficul , and e en ually, 37% o he in i ed women en e ed he measu emen s o analyses o MeS. Thi y-se en pe cen we e ei he no willing o pa icipa e o only comple ed he ollow-up ques ion- nai e. D opou analyses showed ha he esul s may ha e been a ec ed by a heal hy selec ion bias, as he women who pa icipa ed in he ollow-up s udy we e less o en smoke s and we e highe educa ed. On he o he hand, he bias a ec ed all ou subg oups en e ing he ollow-up s udy, as hey did no di e om each o he ega ding o he pa icipa ion a e. The subg oups we e ela i ely small, which weakened he powe o compa e he subg oups. The s udy was implemen ed among Finnish women, and hus, he esul s can only be gene - alised o Caucasian popula ions. Puhkala J, e al.BMJ Open 2017;7:e014565. doi:10.1136/bmjopen-2016-014565 7 Open Access g oup.bmj.com on May 16, 2017 - Published by h p://bmjopen.bmj.com/Downloaded om CONCLUSIONS Se en yea s pos pa um, MeS was diagnosed among 14% o he women wi h isk ac o s o GDM du ing p eg- nancy, which is less han a e age. The MeS p e alence was he highes among women wi h a diagnosis o ea ly GDM (50%), and he lowes among he heal hy con ol g oup (7%). Ea ly GDM diagnosis and inc eased p e- p egnancy BMI we e he s onges isk ac o s o de el- oping MeS. P ep egnancy and p esen o e weigh o obesi y we e common among women wi h GDM isk ac o s, and especially among hose wi h ea ly GDM. O e weigh and obese women and especially hose wi h ea ly GDM should be moni o ed and counselled o ca - diome abolic isk ac o s a e deli e y. P e en ion o p egnancy- ela ed weigh e en ion, including excessi e ges a ional and pos pa um weigh gain, is impo an o he p e en ion o MeS. O e weigh and obesi y among ep oduc i e-aged women is inc easing, which ep esen s e en a g ea e challenge in moni o ing and managing isk ac o s o ch onic diseases. La ge popula ion s udies on he MeS p e alence and e alua ion o p e en- ion s a egies a e needed among ep oduc i e-aged women, especially among hose wi h GDM isk ac o s. Acknowledgemen s The au ho s hank hospi al chemis , Anne Rauhio, om Fimlab Labo a o ies who was esponsible o he labo a o y es ing. Päi i Vii anen, Ta u Helenius, Ulla Hakala, Ulla Honkanen and Si ke Rasinpe ä ec ui ed he s udy pa icipan s, pa icipa ed in he da a collec ion and managed he s udy da a. Con ibu o s RL, JP, PK, PT and PH ini ia ed he s udy design; JR and JP pe o med s a is ical analyses; JP w o e he pape . All au ho s con ibu ed o e inemen o he s udy p o ocol and app o ed he inal manusc ip . JP and RL ha e he p ima y esponsibili y o he inal con en . Funding This wo k was suppo ed by Academy Finland (g an numbe 277079), Compe i i e Resea ch Funding o he Tampe e Uni e si y Hospi al (9R030, 9S034, 9N041, 9M053) and he Juho Vainio Founda ion. Compe ing in e es s None decla ed. E hics app o al Pi kanmaa Uni e si y Hospi al Dis ic E hics. P o enance and pee e iew No commissioned; ex e nally pee e iewed. Da a sha ing s a emen All da a p esen ed in his s udy a e a ailable o u he collabo a ion a e eques o RL ([email p o ec ed]). Open Access This is an Open Access a icle dis ibu ed in acco dance wi h he C ea i e Commons A ibu ion Non Comme cial (CC BY-NC 4.0) license, which pe mi s o he s o dis ibu e, emix, adap , build upon his wo k non- comme cially, and license hei de i a i e wo ks on di e en e ms, p o ided he o iginal wo k is p ope ly ci ed and he use is non-comme cial. See: h p:// c ea i ecommons.o g/licenses/by-nc/4.0/ REFERENCES 1. Albe i KG, Eckel RH, G undy SM, e al. Ha monizing he me abolic synd ome: a join in e im s a emen o he In e na ional Diabe es Fede a ion Task Fo ce on Epidemiology and P e en ion; Na ional Hea , Lung, and Blood Ins i u e; Ame ican Hea Associa ion; Wo ld Hea Fede a ion; In e na ional A he oscle osis Socie y; and In e na ional Associa ion o he S udy o Obesi y. Ci cula ion 2009;120:1640–5. 2. Sil a V, S an on KR, G ande AJ. Ha monizing he diagnosis o me abolic synd ome— ocusing on abdominal obesi y. Me ab Synd Rela Diso d 2013;11:102–8. 3. Esse N, Paquo N, Scheen AJ. In lamma o y ma ke s and ca diome abolic diseases. Ac a Clin Belg 2015;70:193–9. 4. Gallaghe EJ, Le oi h D, Ka nieli E. Insulin esis ance in obesi y as he unde lying cause o he me abolic synd ome. M Sinai J Med 2010;77:511–23. 5. Edwa dson CL, Go ely T, Da ies MJ, e al. Associa ion o seden a y beha iou wi h me abolic synd ome: a me a-analysis. PLoS ONE 2012;7:e34916. 6. Mo ei a GC, Cipullo JP, Cio lia LA, e al. P e alence o me abolic synd ome: associa ion wi h isk ac o s and ca dio ascula complica ions in an u ban popula ion. PLoS ONE 2014;9:e105056. 7. Bassi N, Ka agodin I, Wang S, e al. Li es yle modi ica ion o me abolic synd ome: a sys ema ic e iew. Am J Med 2014;127:1242. e1–10. 8. Takaha a M, Shimomu a I. Me abolic synd ome and li es yle modi ica ion. Re Endoc Me ab Diso d 2014;15:317–27. 9. B own AE, Walke M. Gene ics o insulin esis ance and he me abolic synd ome. Cu Ca diol Rep 2016;18:75. 10. Han TS, Lean ME. A clinical pe spec i e o obesi y, me abolic synd ome and ca dio ascula disease. JRSM Ca dio asc Dis 2016;5:2048004016633371. 11. O’Neill S, O’D iscoll L. Me abolic synd ome: a close look a he g owing epidemic and i s associa ed pa hologies. Obes Re 2015;16:1–12. 12. an Vlie -Os ap chouk JV, Nuo io ML, Slag e SN, e al. The p e alence o me abolic synd ome and me abolically heal hy obesi y in Eu ope: a collabo a i e analysis o en la ge coho s udies. BMC Endoc Diso d 2014;14:9. 13. Aguila M, Bhuke T, To es S, e al.P e alenceo heme abolic synd ome in he uni ed s a es, 2003–2012. JAMA 2015;313: 1973–4. 14. Ha is MF. The me abolic synd ome. Aus Fam Physician 2013;42:524–7. 15. Came on AJ, Shaw JE, Zimme PZ. The me abolic synd ome: p e alence in wo ldwide popula ions. Endoc inol Me ab Clin No h Am 2004;33:351–75. able o con en s. 16. B ode s ad AR, Melhus M. P e alence o me abolic synd ome and diabe es melli us in Sami and No wegian popula ions. The SAMINOR—a c oss-sec ional s udy. BMJ Open 2016;6:e009474. 17. Alke wi A, Donneau AF, Sau ageo N, e al. P e alence o he me abolic synd ome in Luxembou g acco ding o he Join In e im S a emen de ini ion es ima ed om he ORISCAV-LUX s udy. BMC Public Heal h 2011;11:4. 18. Raiko JR, Viika i JS, Ilmanen A, e al. Follow-ups o he Ca dio ascula Risk in Young Finns S udy in 2001 and 2007: le els and 6-yea changes in isk ac o s. J In e n Med 2010;267:370–84. 19. Ame ican Diabe es Associa ion. (2) Classi ica ion and diagnosis o diabe es. Diabe es Ca e 2015;(Suppl 38):S8–16. 20. Kampmann U, Madsen LR, Skajaa GO, e al. Ges a ional diabe es: a clinical upda e. Wo ld J Diabe es 2015;6:1065–72. 21. Vilmi-Ke älä T, Palomäki O, Vainio M, e al. The isk o me abolic synd ome a e ges a ional diabe es melli us—a hospi al-based coho s udy. Diabe ol Me ab Synd 2015;7:1–10. 22. Xu Y, Shen S, Sun L, e al. Me abolic synd ome isk a e ges a ional diabe es: a sys ema ic e iew and me a-analysis. PLoS ONE 2014;9:e87863. 23. Du nwald C. Ges a ional diabe es: linking epidemiology, excessi e ges a ional weigh gain, ad e se p egnancy ou comes, and u u e me abolic synd ome. Semin Pe ina ol 2015;39:254–8. 24. Puhkala J, Kinnunen TI, Vasanka i T, e al. P e alence o me abolic synd ome one yea a e deli e y in Finnish women a inc eased isk o ges a ional diabe es melli us du ing p egnancy. J P egnancy 2013;2013:139049. 25. Gilmo e LA, Klempel-Donchenko M, Redman LM. P egnancy as a window o u u e heal h: excessi e ges a ional weigh gain and obesi y. Semin Pe ina ol 2015;39:296–303. 26. Rayanagouda G, Hashi AA, Zamo a J, e al. Quan i ica ion o he ype 2 diabe es isk in women wi h ges a ional diabe es: a sys ema ic e iew and me a-analysis o 95,750 women. Diabe ologia 2016;59:1403–11. 27. Puhkala J, Luo o R, Aho upa M, e al. Pos pa um weigh e en ion is associa ed wi h ele a ed a io o oxidized LDL lipids o HDL-choles e ol. Lipids 2013;48:1227–35. 28. Kinnunen TI, Puhkala J, Rai anen J, e al. E ec s o die a y counselling on ood habi s and die a y in ake o Finnish p egnan women a inc eased isk o ges a ional diabe es—a seconda y analysis o a clus e - andomized con olled ial. Ma e n Child Nu 2014;10:184–97. 29. Ai asalo M, Rai anen J, Kinnunen TI, e al. Is in ensi e counseling in ma e ni y ca e easible and e ec i e in p omo ing physical ac i i y among women a isk o ges a ional diabe es? Seconda y analysis o a clus e andomized NELLI s udy in Finland. In J Beha Nu Phys Ac 2012;9:104. 8Puhkala J, e al.BMJ Open 2017;7:e014565. doi:10.1136/bmjopen-2016-014565 Open Access g oup.bmj.com on May 16, 2017 - Published by h p://bmjopen.bmj.com/Downloaded om 30. Kinnunen TI, Rai anen J, Ai asalo M, e al. P e en ing excessi e ges a ional weigh gain–a seconda y analysis o a clus e - andomised con olled ial. Eu J Clin Nu 2012;66:1344–50. 31. Luo o R, Kinnunen TI, Ai asalo M, e al. P ima y p e en ion o ges a ional diabe es melli us and la ge- o -ges a ional-age newbo ns by li es yle counseling: a clus e - andomized con olled ial. PLoS Med 2011;8:e1001036. 32. Luo o RM, Kinnunen TI, Ai asalo M, e al. P e en ion o ges a ional diabe es: design o a clus e - andomized con olled ial and one-yea ollow-up. BMC P egnancy Childbi h 2010;10:39. 33. Ame ican Diabe es Associa ion. Diagnosis and classi ica ion o diabe es melli us. Diabe es Ca e 2006;(29 Suppl 1):S43–8. 34. Bo odulin K, Le älah i E, Saa ikoski L, e al. Kansallinen FINRISKI 2012— e eys u kimus—Osa 2: Tu kimuksen aulukkolii e. Te eyden ja hy in oinnin lai os, Helsinki. (In Finnish). 2013; 22. h p://u n. i/URN:ISBN:978-952-302-054-2 (accessed 30 Jun 2016). 35. Männis ö S, Laa ikainen T, Va iainen E. Suomalais en liha uus ennen ja ny . Tu kimukses a ii iis i 4, ma askuu 2012.Te eyden ja hy in oinnin lai os, Helsinki. (In Finnish). 2012. h p://u n. i/URN: ISBN:978-952-245-792-9 (accessed 30 Jun 2016). 36. Sulli an SD, Umans JG, Ra ne R. Ges a ional diabe es: implica ions o ca dio ascula heal h. Cu Diab Rep 2012;12:43–52. 37. Akinci B, Cel ik A, Yene S, e al. P edic ion o de eloping me abolic synd ome a e ges a ional diabe es melli us. Fe il S e il 2010;93:1248–54. 38. Al huizen E, an Poppel MN, de V ies JH, e al. Pos pa um beha iou as p edic o o weigh change om be o e p egnancy o one yea pos pa um. BMC Public Heal h 2011;11:165. 39. Wol e WS, Sobal J, Olson CM, e al. Pa i y-associa ed body weigh : modi ica ion by sociodemog aphic and beha io al ac o s. Obes Res 1997;5:131–41. 40. Ohlin A, Rössne S. Fac o s ela ed o body weigh changes du ing and a e p egnancy: he S ockholm P egnancy and Weigh De elopmen S udy. Obes Res 1996;4:271–6. 41. Siega-Riz AM, Viswana han M, Moos MK, e al. A sys ema ic e iew o ou comes o ma e nal weigh gain acco ding o he Ins i u e o Medicine ecommenda ions: bi hweigh , e al g ow h, and pos pa um weigh e en ion. Am J Obs e Gynecol 2009;201:339. e1–14. 42. Ma ge ison Zilko CE, Rehkop D, Ab ams B. Associa ion o ma e nal ges a ional weigh gain wi h sho - and long- e m ma e nal and child heal h ou comes. Am J Obs e Gynecol 2010;202:574.e1–8. 43. B own WJ, Hockey R, Dobson AJ. E ec s o ha ing a baby on weigh gain. Am J P e Med 2010;38:163–70. 44. Go e SA, B own DM, Wes DS. The ole o pos pa um weigh e en ion in obesi y among women: a e iew o he e idence. Ann Beha Med 2003;26:149–59. 45. Luo o R, Männis ö S, Rai anen J. Ten-yea change in he associa ion be ween obesi y and pa i y: esul s om The Na ional FINRISK Popula ion S udy. Gend Med 2011;8:399–406. 46. Hakka ainen H, Huopio H, Cede be g H, e al. The isk o me abolic synd ome in women wi h p e ious GDM in a long- e m ollow-up. Gynecol Endoc inol 2016;32:920–5. 47. Mission JF, Ma shall NE, Caughey AB. P egnancy isks associa ed wi h obesi y. Obs e Gynecol Clin No h Am 2015;42:335–53. 48. Hoedjes M, Be ks D, Vogel I, e al. P e e ences o pos pa um li es yle counseling among women sha ing an inc eased ca dio ascula and me abolic isk: a ocus g oup s udy. Hype ens P egnancy 2011;30:83–92. 49. Phelan S. Windows o oppo uni y o li es yle in e en ions o p e en ges a ional diabe es melli us. Am J Pe ina ol 2016;33:1291–9. 50. Guo J, Chen JL, Whi emo e R, e al. Pos pa um li es yle in e en ions o p e en ype 2 diabe es among women wi h his o y o ges a ional diabe es: a sys ema ic e iew o andomized clinical ials. J Womens Heal h (La chm ) 2016;25:38–49. 51. an de Plig P, Willcox J, Heske h KD, e al. Sys ema ic e iew o li es yle in e en ions o limi pos pa um weigh e en ion: implica ions o u u e oppo uni ies o p e en ma e nal o e weigh and obesi y ollowing childbi h. Obes Re 2013;14:792–805. 52. Phelan S, Phipps MG, Ab ams B, e al. Does beha io al in e en ion in p egnancy educe pos pa um weigh e en ion? Twel e-mon h ou comes o he Fi o Deli e y andomized ial. Am J Clin Nu 2014;99:302–11. 53. Ha ing on M, Gibson S, Co ell RC. A e iew and me a-analysis o he e ec o weigh loss on all-cause mo ali y isk. Nu Res Re 2009;22:93–108. 54. Klein S, Bu ke LE, B ay GA, e al. Clinical implica ions o obesi y wi h speci ic ocus on ca dio ascula disease: a s a emen o p o essionals om he Ame ican Hea Associa ion Council on Nu i ion, Physical Ac i i y, and Me abolism: endo sed by he Ame ican College o Ca diology Founda ion. Ci cula ion 2004;110:2952–67. 55. Pie ce M, Modde J, Mo agy I, e al. Missed oppo uni ies o diabe es p e en ion: pos -p egnancy ollow-up o women wi h ges a ional diabe es melli us in England. B J Gen P ac 2011;61: e611–19. 56. NCD Risk Fac o Collabo a ion (NCD-RisC). T ends in adul body-mass index in 200 coun ies om 1975 o 2014: a pooled analysis o 1698 popula ion-based measu emen s udies wi h 19.2 million pa icipan s. Lance 2016;387:1377–96. Puhkala J, e al.BMJ Open 2017;7:e014565. doi:10.1136/bmjopen-2016-014565 9 Open Access g oup.bmj.com on May 16, 2017 - Published by h p://bmjopen.bmj.com/Downloaded om