In Canada, the prevalence of obese adults has increased from 9.7% in 1970–1972 to 14.9% in 1998.1 This trend applies to women of reproductive age and has led to obesity rates increasing among women presenting for prenatal care.2
Obese women appear to be at increased risk of gestational diabetes, preeclampsia, labor induction, cesarean delivery, postpartum hemorrhage, and infectious morbidity.3–7 Groups in the United States England, Sweden, and Denmark have performed retrospective database studies evaluating the effect of prepregnancy obesity on adverse pregnancy outcomes.4–6,8,9 The results show an association between maternal prepregnancy obesity and gestational diabetes, pregnancy-induced hypertension (PIH), and macrosomic infants, but do not consistently show an association between other maternal outcomes such as postpartum hemorrhage and postpartum infectious morbidity. The purpose of this population-based cohort study was to carry out a comprehensive investigation of the relationship between prepregnancy obesity and maternal pregnancy outcomes.
MATERIALS AND METHODS
The Nova Scotia Atlee Perinatal Database is a provincial, population-based, clinically oriented computerized database that collates information on pregnancy outcomes in Nova Scotia, Canada. Maternal and newborn data (including demographic, antenatal, intrapartum, and postpartum variables, and maternal and newborn diagnoses) are consistently available for every pregnancy of 20 weeks or more gestational age, with a birth weight of 500 g or more, occurring in Nova Scotia hospitals and to Nova Scotia residents since 1988. This database was used to investigate the impact of prepregnancy maternal obesity on pregnancy outcomes.
Maternal data included in this study consisted of information from singleton pregnancies delivered in Nova Scotia between January 1, 1988, and December 31, 2002. Women were excluded if there was no maternal weight recorded.
Women were categorized by prepregnancy weight in kilograms (kg). The prepregnancy weight recorded in the database is taken from the prepregnancy weight stated on the standard provincial prenatal form used for all pregnant patients in Nova Scotia. A prepregnancy weight of 90 kg or greater was considered obese. The reference (nonobese) group was defined as those with a prepregnancy weight of 55–75 kg. Women with a weight of less than 55 kg or with a weight greater than 75 kg and less than 90 kg were excluded. Obese women were further categorized into a moderate obesity group, which consisted of women with a prepregnancy weight of 90–120 kg, and a severe obesity group, which consisted of women with a prepregnancy weight greater than 120 kg. Maternal outcomes in the 2 obese groups were compared with those in the nonobese population of women.
Other maternal characteristics studied included maternal age, parity, marital status, smoking, socioeconomic status, weight gain in pregnancy, and pre-existing diabetes. Birth weight and gestational age at delivery were also assessed. Antepartum maternal morbidity outcomes examined included hypertensive disorders. The diagnosis of hypertensive disorders in pregnancy was made if it occurred in the antepartum or postpartum period. Mild PIH included physician-diagnosed mild PIH if in the chart, transient hypertension, or a diastolic blood pressure greater than 90 mm Hg on 2 or more occasions in a 24-hour period. Severe PIH included physician-diagnosed severe PIH, diastolic blood pressure greater than 110 mm Hg on at least 2 occasions within a 6-hour period, if magnesium sulfate was administered for seizure prophylaxis, if there was more than 2+ proteinuria or HELLP syndrome (hemolysis, elevated liver enzymes, low platelets). HELLP syndrome was diagnosed as low platelets (< 100,000/μL), and/or elevated liver enzymes (alanine aminotransferase > 35 U/L, aspartate aminotransferase > 30 U/L) and/or elevated lactate dehydrogenase (> 670 U/L) or physician-diagnosed HELLP syndrome. Eclampsia included physician-diagnosed eclampsia or one or more convulsions not attributable to other cerebral conditions in a patient with hypertension. Chronic hypertension included a history of hypertensive disease when not pregnant, before current pregnancy, or before 20 weeks of gestation in the current pregnancy. Antepartum venous thromboembolism and gestational diabetes were also assessed. Intrapartum maternal outcome variables included preterm birth (< 37 weeks of gestation), very preterm birth (< 32 weeks of gestation), postterm delivery (> 41 weeks of gestation), labor induction, cesarean delivery, operative (forceps- or vacuum-assisted) vaginal delivery, use of regional anesthesia (spinal and/or epidural), need for general anesthesia, anesthesia complications (including postdural puncture headache, paresthesias, hypotension, and back pain), pyrexia in labor (temperature ≥ 38°C), prolonged second stage (> 3 hours), shoulder dystocia, and severe perineal laceration (third- or fourth-degree lacerations). Postpartum maternal outcomes assessed were antibiotic use, postpartum hemorrhage (estimated blood loss > 500 mL after vaginal delivery or > 1,000 mL after cesarean delivery), need for blood transfusion, pyrexia (temperature ≥ 38°C), puerperal morbidity (temperature ≥ 38°C on 2 or more occasions, excluding the first 24 hours postpartum), endometritis, urinary tract infection, wound infection (infected abdominal or episiotomy wound), wound dehiscence, and prolonged hospital stay (> 3 days).
Statistical analysis was performed with SAS 8.0 for Windows (SAS Institute Inc, Cary, NC). The χ2 test and Fisher exact test were used in contrasts involving categorical variables. The Student t test was used to compare the means of continuous variables. Univariate and multivariable logistic regression analyses were performed. Logistic regression adjusted for potential confounders, such as maternal age, marital status, smoking, parity, and socioeconomic status. Chronic hypertension and pre-existing diabetes were considered intermediate steps in the causal pathway between maternal prepregnancy obesity and maternal morbidity and were not initially included as independent variables in the logistic regression analyses.10 The effect of obesity after controlling for pre-existing diabetes and chronic hypertension was then determined for outcomes that were statistically significant in the initial model and which are impacted by these pre-existing conditions. The models with and those without control for pre-existing diabetes and chronic hypertension were expected to show whether the effect of obesity was mediated partly or entirely through its relationship with diabetes and hypertension. Because women could have had more than one delivery during the study period, variance estimates were corrected for the potential nonindependence of such observations with SAS ProcGenmod (SAS 8.2; SAS Institute Inc).11 Odds ratios (ORs), adjusted ORs, and 95% confidence intervals (CIs) were calculated expressing the relationship between obesity groups and specific maternal outcomes. P < .05 was considered statistically significant. Ethics approval for the study was obtained from the Research Ethics Board at the IWK Health Centre in Halifax, Nova Scotia.
There were 159,344 singleton deliveries to Nova Scotia residents between 1988 and 2002, but 16,940 (10.6%) of these women were excluded from our study because no maternal weight was recorded. This left 142,404 singleton deliveries with prepregnancy maternal weight recorded. Of these, 34,447 (24.2%) deliveries were to women weighing less than 55 kg, 79,005 (55.5%) were to women weighing 55–75 kg (nonobese group), 18,818 (13.2%) were to women weighing more than 75 kg and less than 90 kg, 9,355 (6.6%) were to women weighing 90–120 kg (moderate obesity group), and 779 (0.6%) were to women weighing more than 120 kg (severe obesity group). Over the 15-year study period, the number of women in the obese categories (ie, ≥ 90 kg) increased from 3.2% in 1988 to 10.2% in 2002.
Maternal characteristics are summarized in Table 1. The obese women were slightly older. There were fewer primiparous women in the 2 obese categories (nonobese 44.7%, moderate obesity 42.0%, severe obesity 39.8%, P < .01). There was no difference in marital status between groups. Obese women had a significantly greater incidence of pre-existing diabetes (nonobese 0.3%, moderate obesity 0.7%, severe obesity 2.2%, P < .001) and chronic hypertension (nonobese 0.6%, moderate obesity 3.9%, severe obesity 9.1%, P < .001). The mean weight gain in pregnancy was substantially less in the women in the obese groups. Mean gestational age at delivery and birth weight were significantly different between the obese groups and the nonobese group.
Univariate analysis showed obese women to be at increased risk of PIH, antepartum venous thromboembolism, gestational diabetes, induction of labor, cesarean delivery, regional anesthesia use, need of general anesthesia, antibiotic use in labor, shoulder dystocia, antibiotic use postpartum, and wound infection (Table 2). There were no cases of eclampsia in any of the groups.
The results of the initial logistic regression analyses corrected for repeat deliveries to the same woman are summarized in Table 3. Pregnancy-induced hypertension rates were higher among the moderately obese women (adjusted OR 2.38, 95% CI 2.24–2.52) and severely obese women (adjusted OR 3.00, 95% CI 2.49–3.62) compared with the nonobese women. The rates of severe PIH were also higher among obese women (moderate obesity: adjusted OR 1.56, 95% CI 1.35–1.80; severe obesity: adjusted OR 2.34, 95% CI 1.59–3.46). This implies that, relative to nonobese women, there was 1 excess case of PIH per 10 moderately obese women and 1 per 7 severely obese women. For severe PIH, there was 1 excess case per 122 moderately obese women and 1 per 52 severely obese women. The incidence of antepartum venous thromboembolism was increased in obese women and increased with increased maternal weight (moderate obesity: adjusted OR 2.17, 95% CI 1.30–3.63; severe obesity: adjusted OR 4.13, 95% CI 1.26–13.54). Thus, relative to nonobese women, there was 1 excess case of antepartum venous thromboembolism per 857 moderately obese women and 1 per 321 severely obese women. Gestational diabetes was more common in the obese groups (moderate obesity: adjusted OR 2.80, 95% CI 2.54–3.08; severe obesity: adjusted OR 5.43, 95% CI 4.31–6.85); ie, there was 1 excess case of gestational diabetes per 29 moderately obese women and 1 per 13 severely obese women.
There was no difference in preterm birth (< 37 weeks) or very preterm birth rates (< 32 weeks) between the nonobese group and the obese groups. The obese groups were more likely to be induced, with the adjusted OR for induction of labor being 1.94 (95% CI 1.86–2.04) for the moderate obesity group and 2.77 (95% CI 2.39–3.21) for the severe obesity group; ie, there was 1 excess induction of labor per 8 moderately obese women and 1 per 5 severely obese women. Obese women had a higher rate of cesarean delivery, with the adjusted OR increasing with increased maternal weight (moderate obesity: adjusted OR 1.60, 95% CI 1.66–1.83; severe obesity: adjusted OR 2.46, 95% CI 2.11–2.85). Thus, relative to nonobese women, there was 1 excess cesarean delivery per 12 moderately obese women and 1 per 6 severely obese women. Obese women were less likely to have operative vaginal deliveries.
Obesity increased the likelihood of the women receiving regional anesthesia and general anesthesia. Anesthetic complications were increased in the severe obesity group (ie, 1 excess case of anesthetic complications per 68 severely obese women) but not in the moderate obesity group. The risk of pyrexia in labor and prolonged second stage was not increased in the obese groups, but the women in these groups were more likely to receive antibiotics intrapartum. Shoulder dystocia was significantly more common in obese women (ie, there was 1 excess case of shoulder dystocia per 112 moderately obese women and 1 per 94 severely obese women).
Postpartum, the obese groups were more likely to receive antibiotics and experience wound infections, with the risk of both increasing with increased maternal weight. Relative to nonobese women, there was 1 excess case of wound infection per 189 moderately obese women and 1 per 34 severely obese women. Moderately obese women had a slightly higher risk of postpartum hemorrhage although there was no difference in need for blood transfusion between the obese groups and the nonobese group. The obese groups were not at increased risk for postpartum pyrexia, puerperal morbidity, endometritis, postpartum urinary tract infection, or prolonged hospital stay relative to the nonobese group.
The effect of obesity was altered by controlling for pre-existing diabetes and chronic hypertension. These results are summarized in Table 4. Adjusting for pre-existing diabetes and chronic hypertension decreased the adjusted OR slightly for the outcomes of PIH, antepartum venous thromboembolism, labor induction, cesarean delivery, regional anesthesia, and general anesthesia for both obesity groups, but the obese groups remained at significantly increased risk. The adjusted OR for severe PIH became nonsignificant in both groups. Anesthesia complications became more likely in the severely obese women. The risk of wound infection decreased slightly in moderately obese women after adjusting for pre-existing diabetes and chronic hypertension, but the risk increased for severely obese women.
Supplementary analyses showed that women with missing information on prepregnancy weight had the following characteristics: mean maternal age 27.6 years, proportion primiparous 37.7%, married 66.2%, smokers 30.0%, low socioeconomic status 47.4%, pre-existing diabetes 0.4%, chronic hypertension 0.9%, mean gestational age at delivery 38.7 weeks, and mean birth weight 3,336 grams.
In Canada and other industrialized countries, obesity rates are reaching epidemic proportions,1,2 and obesity rates among women presenting for prenatal care have increased dramatically. In our population over the 15-year study period from 1988 to 2002, the proportion of women with a prepregnancy weight 90 kg or greater increased from 3.2% to 10.2%. Women in the obese groups were slightly older and less likely to be primiparous. Increasing age and multiparity are recognized as risk factors for obesity.12 The mean weight gain in pregnancy among obese women was substantially less than among nonobese women. This has been observed in other obese pregnant populations and may be because obese women have sufficient energy stores for pregnancy without the need for similar weight gain (Weiss JL, Malone FD. Caring for obese obstetric patients. Contemp Ob/Gyn 2001;6:12–26). Obese women had a greater incidence of pre-existing diabetes and chronic hypertension, which has also been previously documented.13–15 Similarly, we found a higher risk of PIH, gestational diabetes, labor induction, and cesarean delivery in obese women. These findings are consistent with the results of several other large population-based studies.3–7,16
The rates of operative vaginal delivery were lower in the obese groups. This is likely due to the higher cesarean delivery rates in the obese groups and the reluctance of practitioners to perform operative vaginal deliveries in this population because of the increased risk of shoulder dystocia. We did find an increased risk of shoulder dystocia in the obese groups. As previously reported,12 shoulder dystocia may occur because of the increased incidence of macrosomia in the infants of obese women, which is recognized as a risk factor for shoulder dystocia.
There is conflicting data in the literature regarding maternal obesity and preterm birth, with some studies showing an increased risk,3,6 while others have demonstrated a decreased risk or no change8,16 in the risk of preterm birth. We found no difference between the obese groups and the nonobese group for preterm birth (< 37 weeks) or very preterm birth (< 32 weeks). The reason for this difference between previous studies and our study is unclear but may reflect differences in the study populations.
Obesity is a well-recognized risk factor for venous thromboembolism in pregnancy,17,18 although few studies have looked at the rate of antepartum venous thromboembolism in obese pregnant women.4 We found that obese women had a significantly increased risk of antepartum venous thromboembolism, with the risk increasing with increased maternal weight. Obese pregnant women may warrant prophylactic measures against venous thromboembolism, such as compression stockings or prophylactic low-molecular-weight heparin, especially if exposed to other risk factors (eg, bedrest).
In our study, obese women had higher rates of use of regional and general anesthesia. Severely obese women had higher rates of anesthetic complications. This remained true after controlling for pre-existing diabetes and chronic hypertension. A number of anatomical factors in obese patients likely explain the increased incidence of complications seen in this population. Increased subcutaneous fat increases the difficulty in placing regional anesthesia, increasing the rate of placement failure and thus the need for general anesthesia.17
As expected, our study found obese women to be at a greater risk of wound infection,19,20 but we did not find them to be at greater risk of other postpartum morbidities. As documented in previous studies,4,7 we found that moderate obesity conferred a small increase in the risk of postpartum hemorrhage, but there was no difference in the rates of blood transfusion in the obese groups compared with the nonobese group. The obese groups were not at increased risk over the nonobese group for postpartum pyrexia, puerperal morbidity, endometritis, postpartum urinary tract infection, or prolonged hospital stay. These findings contradict previous studies and may be due to the more frequent use of intrapartum and postpartum antibiotics in the obese groups in our setting.
No single definition of obesity has been universally adopted in the literature. Definitions that have been used include a weight of greater than or equal to 80 kg or 90 kg, a weight for height greater than 20% of a reference standard, a weight greater than 50% over ideal body weight, or a body mass index (BMI) of 30 kg/m2 or greater.12 We were unable to use BMI to define obesity because information on height is not recorded in the Nova Scotia Atlee Perinatal Database. Our categorizations of prepregnancy weight into obese and nonobese groups is consistent with the finding of Wolfe et al,12 who showed that the development of adverse pregnancy outcomes associated with maternal obesity was comparably predicted by either maternal weight or a BMI greater than the 90th percentile. The World Health Organization defines obesity based on BMI, with a BMI of 30 greater being considered obese.21 Our study defined obesity as a prepregnancy maternal weight 90 kg or more. A woman of average height (164 cm)22 with a weight of 90 kg or more would have a BMI of 33.5 or greater. Thus, in defining obesity as a prepregnancy maternal weight of 90 kg or more, it is unlikely that we would misclassify any women as obese if they were not. This also ensures that women in the obesity groups were obese by both BMI and weight classifications. Severe obesity has been defined as a BMI greater than 40 kg/m2.21 We choose 120 kg to define severe obesity because woman of average height (164 cm) with a weight of more than 120 kg would have a BMI of 45 or greater. Women with a prepregnancy weight greater than 120 kg would have to be 174 cm or taller to have a BMI less than 40. Thus, we are confident that women with a weight greater than 120 kg would be severely obese by BMI calculations.
The initial multivariable logistic regression analyses did not control for pre-existing diabetes and chronic hypertension because these conditions were considered intermediate steps in the causal pathway between maternal prepregnancy obesity and maternal morbidity. The subsequent logistic regression analysis did adjust for pre-existing diabetes and chronic hypertension, with little impact on the adjusted OR (except for severe PIH). This suggests that pre-existing diabetes and chronic hypertension are not the only pathways between obesity and adverse maternal outcome. Obesity alone, in the absence of these pre-existing conditions, is a risk factor for adverse maternal outcome.
Our study has some limitations. The categorization of the obese population into moderate and severe obesity groups led to small numbers in the severe obesity group. However, this was done to determine whether the pregnancy-associated risks of maternal obesity increase with increasing degrees of obesity. We were limited by definitions for hypertensive disorders in pregnancy that were used by the database. These definitions are not exactly the same as commonly used definitions, but they do approximate definitions proposed by the Canadian Hypertension Society and other organizations.23 Although retrospective studies in general are limited by the reliability of data, information in the Nova Scotia Atlee Perinatal Database is of high quality. Routine data check and edits are made at the time of data collection by qualified health records personnel, and validation24 and reabstraction studies attest to the quality of the data in the database. Prepregnancy weight was the only variable missing from our data to a large extent. All subjects with missing values were excluded from the regression analyses. The missing information on prepregnancy weight constitutes a limitation of the study. Supplementary analyses showed that the women excluded from the study because of missing information were slightly younger, of higher parity, and of lower socioeconomic status, were delivered slightly earlier, and had slightly smaller infants. Nevertheless, there is no reason to believe that the obesity-maternal outcome relationship observed in our study was biased because of these exclusions. The large number of comparisons carried out increases the potential for a type I error and requires that the P values and confidence intervals be interpreted with caution.
Obese women considering pregnancy should be informed of the risks that maternal obesity confers on a pregnancy. Health care professionals need to encourage and assist obese women to make lifestyle changes to lose weight preconceptually in an attempt to optimize pregnancy outcomes and potentially decrease the risk of complications in pregnancy.
1. Katzmarzyk PT. The Canadian obesity epidemic: an historical perspective. Obes Res 2002;10:666–74.
2. Ehrenberg HM, Dierker L, Milluzzi C, Mercer BM. Prevalence of maternal obesity in an urban center. Am J Obstet Gynecol 2002;187:1189–93.
3. Baeten JM, Bukusi EA, Lambe M. Pregnancy complications and outcomes among overweight and obese nulliparous women. Am J Public Health 2001;91:436–40.
4. Sebire NJ, Jolly M, Harris JP, Wadsworth J, Joffe M, Beard RW, et al. Maternal obesity and pregnancy outcome: a study of 287,213 pregnancies in London. Int J Obes Relat Metab Disord 2001;25:1175–82.
5. Lu GC, Rouse DJ, DuBard M, Cliver S, Kimberlin D, Hauth JC. The effect of the increasing prevalence of maternal obesity on perinatal morbidity. Am J Obstet Gynecol 2001;185:845–9.
6. Jensen DM, Damm P, Sorensen B, Molsted-Pedersen, Westergaard JG, Ovesen P, et al. Pregnancy outcome and prepregnancy body mass index in 2459 glucose-tolerant Danish women. Am J Obstet Gynecol 2003;189:239–44.
7. Usha Kiran TS, Hemmadi S, Bethel J, Evans J. Outcome of pregnancy in a woman with increased body mass index. BJOG 2005;112:768–72.
8. Cnattingius S, Bergstrom R, Lipworth L, Kramer M. Prepregnancy weight and the risk of adverse pregnancy outcomes. N Engl J Med 1998;338:147–52.
9. Cedergren M. Maternal morbid obesity and the risk of adverse pregnancy outcome. Obstet Gynecol 2004;103:219–24.
10. Kleinbaum DG, Kupper LL, Morgenstern H. Epidemiologic research: principles and quantitative methods. Belmont (CA): Lifetime Learning Publications; 1982. p. 257.
11. Zeger SL, Liang K-Y. Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986;42:121–30.
12. Wolfe HM, Zador IE, Gross TL, Martier SS, Sokol RJ. The clinical utility of maternal body mass index in pregnancy. Am J Obset Gynecol 1991;164:1306–10.
13. Must A, Spadano J, Coakley EH, Field AE, Colditz G, Dietz WH. The disease burden associated with overweight and obesity. JAMA 1999;282:1523–9.
14. Formiguera X, Canton A. Obesity: epidemiology and clinical aspects. Best Pract Res Clin Gastroenterol 2004;18:1125–46.
15. Sharma AM, Chetty VT. Obesity, hypertension and insulin resistance. Acta Diabetol 2005;42 suppl:S3–8.
16. Sheiner E, Levy A, Menes TS, Silverberg D, Katz M, Mazor M. Maternal obesity as an independent risk factor for caesarean delivery. Paediatr Perinat Epidemiol 2004;18:196–201.
17. Andreasen KR, Anderson ML, Schantz AL. Obesity and pregnancy. Act Obstet Gynecol Scand 2004;83:1022–9.
18. Walker ID. Venous and arterial thrombosis during pregnancy: epidemiology. Semin Vasc Med 2003;3:25–32.
19. Martens MG, Kolrud BL, Faro S, Maccato M, Hammill H. Development of wound infection or separation after cesarean delivery. Prospective evaluation of 2,431 cases. J Reprod Med 1995;40:171–5.
20. Beattie PG, Rings TR, Hunter MF, Lake Y. Risk factors for wound infection following caesarean section. Aust N Z J Obstet Gynaecol 1994;34:398–402.
21. World Health Organization. Obesity: preventing and managing the global epidemic. Report of a WHO Consultation on Obesity, Geneva, June 3–5, 1997. Geneva: World Health Organization; 1998.
22. Nutrition Canada. Anthropometry Report: Height, Weight and Body Dimensions. Ottawa, Canada: Bureau of Nutritional Sciences, Health Protection Branch, Health and Welfare Canada; 1980.
23. Helewa ME, Burrows RF, Smith J, Williams K, Brain P, Rabkin SW. Report of the Canadian Hypertensive Society Consensus Conference: 1. Definitions, evaluation and classifications of hypertensive disorders in pregnancy. CMAJ 1997;157:715–25.
24. Fair M, Cyr M, Allen AC, Wen SW, Guyon G, MacDonald RC. Validation study for a record linkage of births and infant deaths in Canada. Catalogue No. 84F0013XIE. Ottawa, Canada: Statistics Canada; 1999.