Best Practice & Research Clinical Obstetrics and Gynaecology 29 (2015) 133e144
Contents lists available at ScienceDirect
Best Practice & Research Clinical Obstetrics and Gynaecology journal homepage: www.elsevier.com/locate/bpobgyn
13
Pregnancy and neonatal outcome after bariatric surgery Kent Willis, MD a, *, Nicky Lieberman, MD b, Eyal Sheiner, MD, PhD, Deputy Director General, Professor and Senior Researcher c a
Our Lady of the Lake Regional Medical Center, Baton Rouge, LA, USA Head of Community Medicine Department, Clalit Health Services Ltd., Tel Aviv, Israel c Soroka University Medical Center, Ben-Gurion University of the Negev, PO Box 151, Beer Sheva, Israel b
Keywords: bariatric surgery maternal obesity pregnancy neonatal outcome perinatal outcome pregnancy outcome
The global obesity epidemic is changing the face of maternalefetal medicine. One in five women is obese at time of conception, and increasing numbers of parturients have undergone bariatric surgery. Recent publication of large, population-based studies and comparison studies of preoperative and post-operative pregnancies have highlighted new risks and benefits to the mother and child. Pregnancy after bariatric surgery appears to effectively reduce the risk of complications such as fetal macrosomia, gestational diabetes mellitus, and hypertensive disorders of pregnancy; however, women who become pregnant after bariatric surgery may constitute a unique obstetric population with an increased risk for preterm and small-for-gestational-age infants. In this article, we provide an overview of the current knowledge of the impact of maternal bariatric surgery on neonatal and pregnancy outcomes. © 2014 Elsevier Ltd. All rights reserved.
Introduction Rising rates of obesity have created a worldwide epidemic [1]. As obesity rates in pregnant women rise, the importance of obesity-related complications also increases. Pregravid obesity has been shown
* Corresponding author. Our Lady of the Lake Children's Hospital, 5000 Hennessy Boulevard, Suite 6001, Baton Rouge, LA 70808, USA. Tel.: þ1 2254219570. E-mail addresses: [email protected] (K. Willis), [email protected] (E. Sheiner).
http://dx.doi.org/10.1016/j.bpobgyn.2014.04.015 1521-6934/© 2014 Elsevier Ltd. All rights reserved.
134
K. Willis et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 29 (2015) 133e144
to triple the risk of infertility and increase rates of miscarriage, gestational diabetes mellitus (GDM), gestational hypertension, pre-eclampsia and caesarean delivery (CD) [2e8]. Women of reproductive age are turning to bariatric surgery in increasing numbers. Currently, women aged 18-45 years undergo >50,000 inpatient bariatric surgical procedures a year in the United States [9], accounting for approximately half of all bariatric procedures [10]. This has created a new obstetric population with unique outcomes and risks for mother and child. Recent publication of large, population-based trials [11e13] and studies comparing preoperative and post-operative pregnancies [14,15] have highlighted new risks and benefits of pregnancy after surgery and confirmed prior research in this new and important field of maternalefetal and neonatal medicine. Bariatric procedures Three main classes of bariatric surgeries have been developed to date. These include restrictive, malabsorptive and combined restrictiveemalabsorptive procedures. Restrictive procedures, such as sleeve gastrectomy, vertical banded gastroplasty (VBG) and adjustable gastric banding (ABG), promote weight loss by physically decreasing gastric volume and thereby total food intake. They are generally considered safer and less complicated to perform, but they do have the disadvantage that sugary, high-energy foods may bypass the restriction. Malabsorptive procedures, such as biliopancreatic diversion (BPD), cause weight loss primarily by bypassing a significant portion of the small bowel and inducing malabsorption. Most purely malabsorptive procedures have been linked to significant safety concerns and are now rarely performed [16,17]. In the United States, most procedures are currently of the combined type, with the Roux-en-Y gastric bypass (RYGB) accounting for 93% of all procedures performed in 2000 [17]. Despite its classification, malabsorption is actually not significant in RYGB [18]. Instead, neuroendocrine changes are increasingly recognized as an important component of the weight loss effect [16]. A recent 10-year follow-up of a randomized study of laparoscopic Roux-en-Y gastric bypass (LYRGB) and laparoscopic adjustable gastric banding (LAGB) found LYRGB was superior to LAGB in terms of excess weight lost (76.2% vs. 46.2%), but exposed the patient to higher complication rates and more potentially lethal long-term surgical complications [19]. Restrictive procedures also promote a shorter mean time of rapid weight loss than combined procedures (9e12 vs. 12e18 months). The shape of the weight loss curve, in addition to overall amount of weight lost, is different between the two types of procedures [20,21]. Arguably the best long-term study of bariatric surgery is the Swedish obese subjects (SOS) study, a prospective, non-randomized study of 4047 obese persons, of which 2010 underwent a form of bariatric surgery. As reported by €stro € m et al. [22], the average maximum weight loss after 1e2 years was a reduction of 32% for Sjo gastric bypass, 25% for VBG, and 20% for AGB as compared to a 1e2% change in body weight resulting from traditional weight loss techniques. A 10-year follow-up found weight loss had stabilized at 25%, 16%, and 14%, respectively, for the three procedures. Complications related to bariatric surgery have been reported in subsequent pregnancies. Reported complications include internal hernias, bowel obstructions, hyperemesis, cholelithiasis and problems with position and function of the gastric band that may require revision [20,23]. A systematic review of bariatric surgery during pregnancy [24] identified 20 complications requiring surgical intervention, including multiple bowel obstructions, a gastric ulcer, a staple line stricture and several band-related complications. After RYGB, the risk of intestinal obstruction is particularly high [25]. Santulli et al. [26] observed surgery-related complications in about a one-fourth of parturients after RYGB. This is in agreement with an earlier study [27]. In light of the serious potential side effects of bariatric procedures that may arise during pregnancy, it is recommended that a high index of suspicion for gastrointestinal surgical complications be maintained when a post-operative woman presents with abdominal pain during pregnancy [28]. Wax et al. [29] further recommend that any woman with a history of RYGB should ideally have a preconception visit with an obstetrician to discuss the warning signs of small bowel obstruction and the associated risks with pregnancy post-operatively. The early involvement of a bariatric surgeon in cases of suspected bowel obstruction is also strongly encouraged [28].
K. Willis et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 29 (2015) 133e144
135
Type of procedure and pregnancy outcome Several studies have attempted to address whether a particular or subtype of bariatric procedure is superior with regard to pregnancy outcome. At the current time, the body of research is still developing as to what type of procedure or specific procedure is, if any, superior with regard to pregnancy outcome. In general, malabsorptive and restrictive procedures appear to be generally similar in terms of safety and efficacy. Gastric bypass Santulli et al. [26] studied 24 pregnancies following RYGB matched with 120 body mass index (BMI) normal and 120 BMI-matched controls for age and parity. Birthweight was lower in the post-operative cohort than both normal and BMI-matched control groups (2948.2 vs. 3368.2 and 3441.8 g, respectively, P < 0.0001). Post-operative parturients also gained less weight than normal BMI controls (5.8 vs. 13.2 kg, respectively, P < 0.0001) and were more likely to have a CD (25% vs. 9.3%, respectively, P ¼ 0.04). No difference in perinatal complications was noted. Two studies [30,31] examined women after gastric bypass and controls collected from consecutive patients that delivered around the same time matched for obesity and prior operative delivery and stratified for obesity, respectively. Wax et al. [30] found an increased risk of hypertension, but women after surgery were still more likely to be obese than controls. Otherwise, both studies found no significant differences in GDM and hypertensive disorders. Gastric banding A recent study by our group [15] compared each subject's preoperative pregnancy with two subsequent post-operative pregnancies (109 women; 327 paired pregnancies: 109 preoperative pregnancies and 218 post-operative pregnancies). The post-operative parturients had all undergone restrictive bariatric surgery (87% had laparoscopic gastric banding and 13% had silastic ring vertical gastroplasty). Surgery led to a significantly lower prepregnancy and predelivery BMI (36.7 ± 4.4 vs. 31.5 ± 5.5 kg/m2, P < 0.001; 40.6 ± 5.5 vs. 35.3 ± 6.1 kg/m2, P < 0.001, respectively). This effect was preserved in the second post-operative surgery as well (31.5 ± 5.5 vs. 31.3 ± 6.3 kg/m2, P ¼ 0.609, and 35.3 ± 6.1 vs. 35.1 ± 5.9 kg/m2, P ¼ 0.706, respectively). Rates of hypertensive disorders and GDM were significantly lower in both post-operative pregnancies. The rate of macrosomic newborns was also significantly lower in the second pregnancy after surgery. Three studies [32e34] reported similarly lowered rates of GDM, maternal weight gain and preeclampsia following surgery. A larger and more recent study of 133 parturients after LAGB concluded that post-operative pregnancies resulted in babies that were as healthy as the general population [35]. Mixed procedure studies A Danish register-based cohort study identified infants born after bariatric surgery (84.4% gastric bypass) were more likely to have a shorter mean gestation, lower mean birthweight (3312 vs. 3585 g; P < 0.001) and have a higher risk of being born small for gestational age (SGA) (adjusted odds ratio (OR) 2.29, 95% confidence interval (CI) 1.32e3.96). Infants born after surgery had a reduced risk of being large for gestational age (LGA), however (adjusted OR, 0.31; 95% CI 0.15e0.65) [12]. In one of the largest population-based matched cohort studies to date, post-operative women were more likely to have a premature (9.7% vs. 6.1%, OR 1.7, 95% CI 1.4e2.0; P < 0.001) or SGA (5.2% vs. 3.0%, OR 2.0, 95% CI1.5e2.5; P < 0.001) infant than normal or obese controls [11]. A previous study of 298 post-operative pregnancies compared to the local community rates, as it aimed to evaluate if bariatric surgery increased the risk of adverse perinatal outcome, specifically birth defects. While no significant differences were noted in adverse perinatal outcome and birth defects, the study found an increased risk of GDM (9.4% vs. 5.0%, P < 0.001), macrosomia (9.4% vs. 4.6%, P < 0.001) and CD post-operatively (25.2% vs. 12.2%, P < 0.001), but no difference in birthweight, pre-eclampsia or pregnancy complications. Importantly, post-operative women were more likely to be obese
136
K. Willis et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 29 (2015) 133e144
(BMI 30 kg/m2) than the comparison group (10.7% vs. 1.2%, P < 0.001) [6]. This difference probably explains the higher rates of GDM, macrosomia and CD as well. Comparative studies A study by our group [36] compared 394 pregnancies after a restrictive procedure with 55 pregnancies after a malabsorptive procedure. We demonstrated a higher mean birthweight following malabsorptive procedures as compared to restrictive procedures (3332.8 ± 475.5 g after RYGB compared to 3104.3 ± 578.7 g after LAGB). Other outcomes, such as Apgar scores, perinatal mortality and the prevalence of low and high birthweight neonates, were equivalent between both groups. Two smaller studies [20,37] also demonstrated comparable short-term neonatal outcomes between both procedures. However, in one study [37], infants born after RYGB had a significantly lower birthweight than after LAGB (2993 vs. 3253 g, P ¼ 0.02). Surgery-to-conception interval The American Congress of Obstetricians and Gynecologists (ACOG) currently recommends to avoid pregnancy following surgery for 12e24 months after bariatric surgery (2009, reconfirmed 2013) [28]. There is concern for potential nutritional deficiencies and adverse effects on the fetus during the period of rapid weight loss that typically lasts 6e18 months after surgery. During this period, the maternal intake is markedly reduced and the nutritional requirements of the fetus are rapidly increasing [25]. While there is a strong theoretical basis for this recommendation, the current body of research does not provide supporting evidence. Our group recently found no significant difference between 104 women who conceived during the first year after surgery and 385 that conceived after the first year. A shorter time to conception was not associated with an increased risk of pregnancy complications [38]. A similar study performed in Denmark on 286 women after RYGB, of whom 158 conceived during the first year after surgery and 128 later, found no statistically significant difference (P > 0.05) with regard to birthweight, gestational age, risk of preeclampsia, GDM, labour induction, caesarean section, postpartum haemorrhage, preterm birth, SGA, LGA, Apgar score or in the need of neonatal intensive care between women who conceived during or after the first year after surgery [39]. A recent multicenter French cohort study [37] also found neonatal outcomes were unaltered by the interval from surgery to conception. Similarly, in the largest population-based cohort study to date, no effect modification was noted by procedure type or interval from surgery to delivery was detected for preterm or SGA birth [11]. Several older studies have also demonstrated no increased risk of complications associated with fetal malnutrition, such as intrauterine growth restriction (IUGR) and malformation [27,31,33,40]. Neonatal outcomes The three most commonly reported neonatal outcomes in recent studies were prematurity, SGA and LGA/macrosomia. A selection of notable studies of neonatal outcome from the last 5 years may be seen in Table 1 [11,12,14,15,26,36e38,41]. Prematurity Obesity has been linked to an increased risk of preterm delivery. A meta-analysis [42] concluded obese and overweight women have an increased risk of having a preterm (relative risk (RR) 1.24, 95% CI 1.13e1.37) and induced preterm birth (RR 1.30, 95% CI 1.23e1.37). As preterm infants are at risk of a number of complications, lowering a mother's prepregnancy BMI with bariatric surgery has been generally regarded as reasonable. Recent research suggests postsurgical women may be at a higher risk of preterm delivery than was previously thought. In the largest study to date, Roos et al. [11] found post-operative deliveries resulted in a preterm birth in a significantly higher percentage than matched controls: 9.7% versus 6.1 (risk difference 3.6%, 95% CI 2.4e4.9%; P < 0.001). The risks were elevated for both spontaneous and
Study
Inclusion criteria Intervention
Sheiner et al., 2009 [36], Israel
Santulli et al., 2010 [26], France Lapolla et al., 2010 [41], Italy
Sheiner et al., 2011 [38], Israel Aricha-Tamir et al., 2012 [14], Israel Ducarme et al., 2013 [37], France
Amsalem et al., 2013 [15], Israel
Kjaer et al., 2013 [12], Denmark Roos et al., 2013 [11], Sweden
202 deliveries after LAGB 136 deliveries after SRVG 56 deliveries after VBG 24 pregnancies after RYGB 69 women (83 pregnancies) after LAGB
104 pregnancies < 1st postop year 144 postop pregnancies 63 neonates born after LAGB 43 neonates born 1st postop year 144 preop pregnancies in same women 31 neonates born after RYGB 51 neonates born >1 year postop 109 preop pregnancies
1277 matched controlsa 12,379 matched controlsb 12,338 matched controlsb
SGA
LGA/Macrosomia
Cases
Controls
Cases
Controls
Cases
n (%)
n (%)
n (%)
n (%)
n (%)
n (%)
20 (9.9) 11 (8.1) 3 (5.4) NR
2 (3.6)
19 (9.4) 12 (8.8) 3 (5.4) 2 (8.3)
4 (7.3)
9 4 3 0
2 (3.6)
15 (17.6)
1 (1.4) 31 (3.6)
3 (2.9)
14 (3.6)
NR
1 (0.8) 6 (5.0) 20 (2.3)
(4.5) (2.9) (5.4) (0)
7 (8.1)
5 (4.3)
Controls
3 (2.5) 13 (10.8) 51 (6.0) 11 (9.4)
NR
12 (11.5)
32 (8.3)
NR
6 (4.2)
8 (5.6)
5 (7.9) 1 (2.5)
1 (3.2) 5 (9.8)
8 (7.1) 1 (1.1)
12 (11.1)
5 (7.9) 6 (13.9)
4 (12.6) 3 (5.9)
11 (17.5) 10 (22.5)
18 (16.7) 12 (10.8)
11 (10.1)
NR
29 (8.6) 243 (9.7)
85 (6.7) 750 (6.1)
24 (7.1)
37 (2.9)
8 (2.4)
93 (7.3)
131 (5.2)
369 (3.0)
105 (4.2)
895 (7.3)
NR ¼ Not reported. LAGB ¼ Laparoscopic adjustable gastric banding. SRVG ¼ Silastic ring vertical gastroplasty VBG ¼ vertical banded gastroplasty a Matching factors: BMI, parity, maternal age and date of delivery. b Matching factors: maternal age, parity, early pregnancy body mass index, early pregnancy smoking status, education level and year of delivery.
10 (32.3) 9 (17.6)
K. Willis et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 29 (2015) 133e144
Table 1 Selected studies of neonatal outcome in the last 5 years.
137
138
K. Willis et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 29 (2015) 133e144
medically indicated preterm delivery (5.2% vs. 3.6%; OR 1.5, 1.2e1.9; P < 0.001 and 4.5% vs. 2.5%; OR 1.8, 1.4e2.3; P < 0.001, respectively). In women with a BMI 35 kg/m2, increasing BMI made preterm delivery more likely (P ¼ 0.01). When the study group (mean BMI 32.6 kg/m2) was compared to cohorts of women with BMI 35 kg/ m2 and BMI 40 kg/m2, there was an attenuation of risk, but the risk of preterm birth remained elevated for both groups. This creates an important distinction, as the data suggest the elevated risk of preterm infants is significant only for women with an early pregnancy BMI of
Contents lists available at ScienceDirect
Best Practice & Research Clinical Obstetrics and Gynaecology journal homepage: www.elsevier.com/locate/bpobgyn
13
Pregnancy and neonatal outcome after bariatric surgery Kent Willis, MD a, *, Nicky Lieberman, MD b, Eyal Sheiner, MD, PhD, Deputy Director General, Professor and Senior Researcher c a
Our Lady of the Lake Regional Medical Center, Baton Rouge, LA, USA Head of Community Medicine Department, Clalit Health Services Ltd., Tel Aviv, Israel c Soroka University Medical Center, Ben-Gurion University of the Negev, PO Box 151, Beer Sheva, Israel b
Keywords: bariatric surgery maternal obesity pregnancy neonatal outcome perinatal outcome pregnancy outcome
The global obesity epidemic is changing the face of maternalefetal medicine. One in five women is obese at time of conception, and increasing numbers of parturients have undergone bariatric surgery. Recent publication of large, population-based studies and comparison studies of preoperative and post-operative pregnancies have highlighted new risks and benefits to the mother and child. Pregnancy after bariatric surgery appears to effectively reduce the risk of complications such as fetal macrosomia, gestational diabetes mellitus, and hypertensive disorders of pregnancy; however, women who become pregnant after bariatric surgery may constitute a unique obstetric population with an increased risk for preterm and small-for-gestational-age infants. In this article, we provide an overview of the current knowledge of the impact of maternal bariatric surgery on neonatal and pregnancy outcomes. © 2014 Elsevier Ltd. All rights reserved.
Introduction Rising rates of obesity have created a worldwide epidemic [1]. As obesity rates in pregnant women rise, the importance of obesity-related complications also increases. Pregravid obesity has been shown
* Corresponding author. Our Lady of the Lake Children's Hospital, 5000 Hennessy Boulevard, Suite 6001, Baton Rouge, LA 70808, USA. Tel.: þ1 2254219570. E-mail addresses: [email protected] (K. Willis), [email protected] (E. Sheiner).
http://dx.doi.org/10.1016/j.bpobgyn.2014.04.015 1521-6934/© 2014 Elsevier Ltd. All rights reserved.
134
K. Willis et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 29 (2015) 133e144
to triple the risk of infertility and increase rates of miscarriage, gestational diabetes mellitus (GDM), gestational hypertension, pre-eclampsia and caesarean delivery (CD) [2e8]. Women of reproductive age are turning to bariatric surgery in increasing numbers. Currently, women aged 18-45 years undergo >50,000 inpatient bariatric surgical procedures a year in the United States [9], accounting for approximately half of all bariatric procedures [10]. This has created a new obstetric population with unique outcomes and risks for mother and child. Recent publication of large, population-based trials [11e13] and studies comparing preoperative and post-operative pregnancies [14,15] have highlighted new risks and benefits of pregnancy after surgery and confirmed prior research in this new and important field of maternalefetal and neonatal medicine. Bariatric procedures Three main classes of bariatric surgeries have been developed to date. These include restrictive, malabsorptive and combined restrictiveemalabsorptive procedures. Restrictive procedures, such as sleeve gastrectomy, vertical banded gastroplasty (VBG) and adjustable gastric banding (ABG), promote weight loss by physically decreasing gastric volume and thereby total food intake. They are generally considered safer and less complicated to perform, but they do have the disadvantage that sugary, high-energy foods may bypass the restriction. Malabsorptive procedures, such as biliopancreatic diversion (BPD), cause weight loss primarily by bypassing a significant portion of the small bowel and inducing malabsorption. Most purely malabsorptive procedures have been linked to significant safety concerns and are now rarely performed [16,17]. In the United States, most procedures are currently of the combined type, with the Roux-en-Y gastric bypass (RYGB) accounting for 93% of all procedures performed in 2000 [17]. Despite its classification, malabsorption is actually not significant in RYGB [18]. Instead, neuroendocrine changes are increasingly recognized as an important component of the weight loss effect [16]. A recent 10-year follow-up of a randomized study of laparoscopic Roux-en-Y gastric bypass (LYRGB) and laparoscopic adjustable gastric banding (LAGB) found LYRGB was superior to LAGB in terms of excess weight lost (76.2% vs. 46.2%), but exposed the patient to higher complication rates and more potentially lethal long-term surgical complications [19]. Restrictive procedures also promote a shorter mean time of rapid weight loss than combined procedures (9e12 vs. 12e18 months). The shape of the weight loss curve, in addition to overall amount of weight lost, is different between the two types of procedures [20,21]. Arguably the best long-term study of bariatric surgery is the Swedish obese subjects (SOS) study, a prospective, non-randomized study of 4047 obese persons, of which 2010 underwent a form of bariatric surgery. As reported by €stro € m et al. [22], the average maximum weight loss after 1e2 years was a reduction of 32% for Sjo gastric bypass, 25% for VBG, and 20% for AGB as compared to a 1e2% change in body weight resulting from traditional weight loss techniques. A 10-year follow-up found weight loss had stabilized at 25%, 16%, and 14%, respectively, for the three procedures. Complications related to bariatric surgery have been reported in subsequent pregnancies. Reported complications include internal hernias, bowel obstructions, hyperemesis, cholelithiasis and problems with position and function of the gastric band that may require revision [20,23]. A systematic review of bariatric surgery during pregnancy [24] identified 20 complications requiring surgical intervention, including multiple bowel obstructions, a gastric ulcer, a staple line stricture and several band-related complications. After RYGB, the risk of intestinal obstruction is particularly high [25]. Santulli et al. [26] observed surgery-related complications in about a one-fourth of parturients after RYGB. This is in agreement with an earlier study [27]. In light of the serious potential side effects of bariatric procedures that may arise during pregnancy, it is recommended that a high index of suspicion for gastrointestinal surgical complications be maintained when a post-operative woman presents with abdominal pain during pregnancy [28]. Wax et al. [29] further recommend that any woman with a history of RYGB should ideally have a preconception visit with an obstetrician to discuss the warning signs of small bowel obstruction and the associated risks with pregnancy post-operatively. The early involvement of a bariatric surgeon in cases of suspected bowel obstruction is also strongly encouraged [28].
K. Willis et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 29 (2015) 133e144
135
Type of procedure and pregnancy outcome Several studies have attempted to address whether a particular or subtype of bariatric procedure is superior with regard to pregnancy outcome. At the current time, the body of research is still developing as to what type of procedure or specific procedure is, if any, superior with regard to pregnancy outcome. In general, malabsorptive and restrictive procedures appear to be generally similar in terms of safety and efficacy. Gastric bypass Santulli et al. [26] studied 24 pregnancies following RYGB matched with 120 body mass index (BMI) normal and 120 BMI-matched controls for age and parity. Birthweight was lower in the post-operative cohort than both normal and BMI-matched control groups (2948.2 vs. 3368.2 and 3441.8 g, respectively, P < 0.0001). Post-operative parturients also gained less weight than normal BMI controls (5.8 vs. 13.2 kg, respectively, P < 0.0001) and were more likely to have a CD (25% vs. 9.3%, respectively, P ¼ 0.04). No difference in perinatal complications was noted. Two studies [30,31] examined women after gastric bypass and controls collected from consecutive patients that delivered around the same time matched for obesity and prior operative delivery and stratified for obesity, respectively. Wax et al. [30] found an increased risk of hypertension, but women after surgery were still more likely to be obese than controls. Otherwise, both studies found no significant differences in GDM and hypertensive disorders. Gastric banding A recent study by our group [15] compared each subject's preoperative pregnancy with two subsequent post-operative pregnancies (109 women; 327 paired pregnancies: 109 preoperative pregnancies and 218 post-operative pregnancies). The post-operative parturients had all undergone restrictive bariatric surgery (87% had laparoscopic gastric banding and 13% had silastic ring vertical gastroplasty). Surgery led to a significantly lower prepregnancy and predelivery BMI (36.7 ± 4.4 vs. 31.5 ± 5.5 kg/m2, P < 0.001; 40.6 ± 5.5 vs. 35.3 ± 6.1 kg/m2, P < 0.001, respectively). This effect was preserved in the second post-operative surgery as well (31.5 ± 5.5 vs. 31.3 ± 6.3 kg/m2, P ¼ 0.609, and 35.3 ± 6.1 vs. 35.1 ± 5.9 kg/m2, P ¼ 0.706, respectively). Rates of hypertensive disorders and GDM were significantly lower in both post-operative pregnancies. The rate of macrosomic newborns was also significantly lower in the second pregnancy after surgery. Three studies [32e34] reported similarly lowered rates of GDM, maternal weight gain and preeclampsia following surgery. A larger and more recent study of 133 parturients after LAGB concluded that post-operative pregnancies resulted in babies that were as healthy as the general population [35]. Mixed procedure studies A Danish register-based cohort study identified infants born after bariatric surgery (84.4% gastric bypass) were more likely to have a shorter mean gestation, lower mean birthweight (3312 vs. 3585 g; P < 0.001) and have a higher risk of being born small for gestational age (SGA) (adjusted odds ratio (OR) 2.29, 95% confidence interval (CI) 1.32e3.96). Infants born after surgery had a reduced risk of being large for gestational age (LGA), however (adjusted OR, 0.31; 95% CI 0.15e0.65) [12]. In one of the largest population-based matched cohort studies to date, post-operative women were more likely to have a premature (9.7% vs. 6.1%, OR 1.7, 95% CI 1.4e2.0; P < 0.001) or SGA (5.2% vs. 3.0%, OR 2.0, 95% CI1.5e2.5; P < 0.001) infant than normal or obese controls [11]. A previous study of 298 post-operative pregnancies compared to the local community rates, as it aimed to evaluate if bariatric surgery increased the risk of adverse perinatal outcome, specifically birth defects. While no significant differences were noted in adverse perinatal outcome and birth defects, the study found an increased risk of GDM (9.4% vs. 5.0%, P < 0.001), macrosomia (9.4% vs. 4.6%, P < 0.001) and CD post-operatively (25.2% vs. 12.2%, P < 0.001), but no difference in birthweight, pre-eclampsia or pregnancy complications. Importantly, post-operative women were more likely to be obese
136
K. Willis et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 29 (2015) 133e144
(BMI 30 kg/m2) than the comparison group (10.7% vs. 1.2%, P < 0.001) [6]. This difference probably explains the higher rates of GDM, macrosomia and CD as well. Comparative studies A study by our group [36] compared 394 pregnancies after a restrictive procedure with 55 pregnancies after a malabsorptive procedure. We demonstrated a higher mean birthweight following malabsorptive procedures as compared to restrictive procedures (3332.8 ± 475.5 g after RYGB compared to 3104.3 ± 578.7 g after LAGB). Other outcomes, such as Apgar scores, perinatal mortality and the prevalence of low and high birthweight neonates, were equivalent between both groups. Two smaller studies [20,37] also demonstrated comparable short-term neonatal outcomes between both procedures. However, in one study [37], infants born after RYGB had a significantly lower birthweight than after LAGB (2993 vs. 3253 g, P ¼ 0.02). Surgery-to-conception interval The American Congress of Obstetricians and Gynecologists (ACOG) currently recommends to avoid pregnancy following surgery for 12e24 months after bariatric surgery (2009, reconfirmed 2013) [28]. There is concern for potential nutritional deficiencies and adverse effects on the fetus during the period of rapid weight loss that typically lasts 6e18 months after surgery. During this period, the maternal intake is markedly reduced and the nutritional requirements of the fetus are rapidly increasing [25]. While there is a strong theoretical basis for this recommendation, the current body of research does not provide supporting evidence. Our group recently found no significant difference between 104 women who conceived during the first year after surgery and 385 that conceived after the first year. A shorter time to conception was not associated with an increased risk of pregnancy complications [38]. A similar study performed in Denmark on 286 women after RYGB, of whom 158 conceived during the first year after surgery and 128 later, found no statistically significant difference (P > 0.05) with regard to birthweight, gestational age, risk of preeclampsia, GDM, labour induction, caesarean section, postpartum haemorrhage, preterm birth, SGA, LGA, Apgar score or in the need of neonatal intensive care between women who conceived during or after the first year after surgery [39]. A recent multicenter French cohort study [37] also found neonatal outcomes were unaltered by the interval from surgery to conception. Similarly, in the largest population-based cohort study to date, no effect modification was noted by procedure type or interval from surgery to delivery was detected for preterm or SGA birth [11]. Several older studies have also demonstrated no increased risk of complications associated with fetal malnutrition, such as intrauterine growth restriction (IUGR) and malformation [27,31,33,40]. Neonatal outcomes The three most commonly reported neonatal outcomes in recent studies were prematurity, SGA and LGA/macrosomia. A selection of notable studies of neonatal outcome from the last 5 years may be seen in Table 1 [11,12,14,15,26,36e38,41]. Prematurity Obesity has been linked to an increased risk of preterm delivery. A meta-analysis [42] concluded obese and overweight women have an increased risk of having a preterm (relative risk (RR) 1.24, 95% CI 1.13e1.37) and induced preterm birth (RR 1.30, 95% CI 1.23e1.37). As preterm infants are at risk of a number of complications, lowering a mother's prepregnancy BMI with bariatric surgery has been generally regarded as reasonable. Recent research suggests postsurgical women may be at a higher risk of preterm delivery than was previously thought. In the largest study to date, Roos et al. [11] found post-operative deliveries resulted in a preterm birth in a significantly higher percentage than matched controls: 9.7% versus 6.1 (risk difference 3.6%, 95% CI 2.4e4.9%; P < 0.001). The risks were elevated for both spontaneous and
Study
Inclusion criteria Intervention
Sheiner et al., 2009 [36], Israel
Santulli et al., 2010 [26], France Lapolla et al., 2010 [41], Italy
Sheiner et al., 2011 [38], Israel Aricha-Tamir et al., 2012 [14], Israel Ducarme et al., 2013 [37], France
Amsalem et al., 2013 [15], Israel
Kjaer et al., 2013 [12], Denmark Roos et al., 2013 [11], Sweden
202 deliveries after LAGB 136 deliveries after SRVG 56 deliveries after VBG 24 pregnancies after RYGB 69 women (83 pregnancies) after LAGB
104 pregnancies < 1st postop year 144 postop pregnancies 63 neonates born after LAGB 43 neonates born 1st postop year 144 preop pregnancies in same women 31 neonates born after RYGB 51 neonates born >1 year postop 109 preop pregnancies
1277 matched controlsa 12,379 matched controlsb 12,338 matched controlsb
SGA
LGA/Macrosomia
Cases
Controls
Cases
Controls
Cases
n (%)
n (%)
n (%)
n (%)
n (%)
n (%)
20 (9.9) 11 (8.1) 3 (5.4) NR
2 (3.6)
19 (9.4) 12 (8.8) 3 (5.4) 2 (8.3)
4 (7.3)
9 4 3 0
2 (3.6)
15 (17.6)
1 (1.4) 31 (3.6)
3 (2.9)
14 (3.6)
NR
1 (0.8) 6 (5.0) 20 (2.3)
(4.5) (2.9) (5.4) (0)
7 (8.1)
5 (4.3)
Controls
3 (2.5) 13 (10.8) 51 (6.0) 11 (9.4)
NR
12 (11.5)
32 (8.3)
NR
6 (4.2)
8 (5.6)
5 (7.9) 1 (2.5)
1 (3.2) 5 (9.8)
8 (7.1) 1 (1.1)
12 (11.1)
5 (7.9) 6 (13.9)
4 (12.6) 3 (5.9)
11 (17.5) 10 (22.5)
18 (16.7) 12 (10.8)
11 (10.1)
NR
29 (8.6) 243 (9.7)
85 (6.7) 750 (6.1)
24 (7.1)
37 (2.9)
8 (2.4)
93 (7.3)
131 (5.2)
369 (3.0)
105 (4.2)
895 (7.3)
NR ¼ Not reported. LAGB ¼ Laparoscopic adjustable gastric banding. SRVG ¼ Silastic ring vertical gastroplasty VBG ¼ vertical banded gastroplasty a Matching factors: BMI, parity, maternal age and date of delivery. b Matching factors: maternal age, parity, early pregnancy body mass index, early pregnancy smoking status, education level and year of delivery.
10 (32.3) 9 (17.6)
K. Willis et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 29 (2015) 133e144
Table 1 Selected studies of neonatal outcome in the last 5 years.
137
138
K. Willis et al. / Best Practice & Research Clinical Obstetrics and Gynaecology 29 (2015) 133e144
medically indicated preterm delivery (5.2% vs. 3.6%; OR 1.5, 1.2e1.9; P < 0.001 and 4.5% vs. 2.5%; OR 1.8, 1.4e2.3; P < 0.001, respectively). In women with a BMI 35 kg/m2, increasing BMI made preterm delivery more likely (P ¼ 0.01). When the study group (mean BMI 32.6 kg/m2) was compared to cohorts of women with BMI 35 kg/ m2 and BMI 40 kg/m2, there was an attenuation of risk, but the risk of preterm birth remained elevated for both groups. This creates an important distinction, as the data suggest the elevated risk of preterm infants is significant only for women with an early pregnancy BMI of
