Trabalhos Originais Archives - Page 2 of 51 - Revista Brasileira de Ginecologia e Obstetrícia
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(10):773-780
DOI 10.1590/S0100-72032004001000004
PURPOSE: to compare the information obtained with pelvic and transvaginal ultrasonography (USG), hystero-salpingography (HSG), diagnostic hysteroscopy (HSC), pelvic nuclear magnetic resonance imaging (PNMR), three-dimensional hysterosonography (3D HSSNG), to optimize and simplify the investigation about cervical and corporeal uterine factors in conjugal infertility. METHODS: in the period between January and July 2003, fifty women reporting infertility for at least two years were submitted to USG, HSG, HSC, PNMR, and 3D HSSNG as tracking examinations for uterine factor diagnosis. The endocervical canal, as well as the endometrium, myometrium, and the presence of uterine malformations were investigated. The results of each examination were analyzed and compared. RESULTS: of the 50 women studied, 12 (24%) presented alteration in at least one of the examinations. When 3D HSSNG was compared to USG, 3D HSSNG provided additional information in 7 cases (58.3%); when compared to HSG, it provided additional information in 7 cases (58.3%); when compared to HSC, it provided additional information in 4 cases (32.1%), and when compared to PNMR, it provided additional information in 6 cases (50%). There were only two cases in which HSG detected alterations of the endocervical canal that were not visualized using 3D HSSNG. In the other cases 3D HSSNG imparted the same diagnosis; furthermore, it provided additional information in comparison to the other examinations. Statistical analysis using the kappa test demonstrated that the diagnoses obtained by 3D HSSNG were in agreement with those obtained with USG, HSG and PNMR (p<0,05). When the HSG and 3D HSSNG results were combined, all conditions associated with infertility could be precisely diagnosed, using only these examinations. CONCLUSION: the association of the HSG with 3D HSSNG may be sufficient for the diagnosis of cervical and corporeal uterine factors in infertility, reducing the number of examinations for each patient, the total cost, as well as the anxiety and the delay in treatment.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(10):773-780
DOI 10.1590/S0100-72032004001000004
PURPOSE: to compare the information obtained with pelvic and transvaginal ultrasonography (USG), hystero-salpingography (HSG), diagnostic hysteroscopy (HSC), pelvic nuclear magnetic resonance imaging (PNMR), three-dimensional hysterosonography (3D HSSNG), to optimize and simplify the investigation about cervical and corporeal uterine factors in conjugal infertility. METHODS: in the period between January and July 2003, fifty women reporting infertility for at least two years were submitted to USG, HSG, HSC, PNMR, and 3D HSSNG as tracking examinations for uterine factor diagnosis. The endocervical canal, as well as the endometrium, myometrium, and the presence of uterine malformations were investigated. The results of each examination were analyzed and compared. RESULTS: of the 50 women studied, 12 (24%) presented alteration in at least one of the examinations. When 3D HSSNG was compared to USG, 3D HSSNG provided additional information in 7 cases (58.3%); when compared to HSG, it provided additional information in 7 cases (58.3%); when compared to HSC, it provided additional information in 4 cases (32.1%), and when compared to PNMR, it provided additional information in 6 cases (50%). There were only two cases in which HSG detected alterations of the endocervical canal that were not visualized using 3D HSSNG. In the other cases 3D HSSNG imparted the same diagnosis; furthermore, it provided additional information in comparison to the other examinations. Statistical analysis using the kappa test demonstrated that the diagnoses obtained by 3D HSSNG were in agreement with those obtained with USG, HSG and PNMR (p<0,05). When the HSG and 3D HSSNG results were combined, all conditions associated with infertility could be precisely diagnosed, using only these examinations. CONCLUSION: the association of the HSG with 3D HSSNG may be sufficient for the diagnosis of cervical and corporeal uterine factors in infertility, reducing the number of examinations for each patient, the total cost, as well as the anxiety and the delay in treatment.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(10):781-786
DOI 10.1590/S0100-72032004001000005
PURPOSE: to assess the obstetric and perinatal outcomes in cases of term newborns in breech presentation, in patients with previous vaginal deliveries, comparing them to term newborns in vertex presentation. METHODS: 8,350 deliveries retrospectively from March 1998 to July 2003 were analysed. Of 419 deliveries (5.1%) in breech presentation, 58 cases were selected for the study (breech group), according to the following criteria: patients who had had one or more babies through vaginal delivery, gestational age ³37 weeks, no fetal malformation, no complications in the current pregnancy, birth weight between 2,500 and 3,750 g, and no previous cesarean section. The breech group was matched to 1,327 newborns in vertex position from pregnant women with no previous cesarean section (vertex group). Maternal age, parity, gestational age, delivery way, birth weight, meconium-stained amniotic fluid, 1- and 5-min Apgar score, need of neonatal intensive care unit, and small- and big-for-gestational age newborns were analyzed. Statistical analysis was performed by the c² test and by Student's t test, with the level of significance set at p<0.05. RESULTS: when breech and vertex groups were compared, they showed significant differences regarding the following variables: birth weight (3,091±538 g vs 3,250±497 g; p<0.01), vaginal delivery (63.8 vs 95.0%; p<0.0001), cesarean section (36.2 vs 5.0%; p<0,0001), and 1-min Apgar score (p<0.0001), respectively. CONCLUSIONS: we conclude that in term fetuses in breech position from pregnant women with previous vaginal deliveries, birth weight, delivery way, and 1-min Apgar score were different compared to fetuses in vertex position from women with the same characteristics.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(10):781-786
DOI 10.1590/S0100-72032004001000005
PURPOSE: to assess the obstetric and perinatal outcomes in cases of term newborns in breech presentation, in patients with previous vaginal deliveries, comparing them to term newborns in vertex presentation. METHODS: 8,350 deliveries retrospectively from March 1998 to July 2003 were analysed. Of 419 deliveries (5.1%) in breech presentation, 58 cases were selected for the study (breech group), according to the following criteria: patients who had had one or more babies through vaginal delivery, gestational age ³37 weeks, no fetal malformation, no complications in the current pregnancy, birth weight between 2,500 and 3,750 g, and no previous cesarean section. The breech group was matched to 1,327 newborns in vertex position from pregnant women with no previous cesarean section (vertex group). Maternal age, parity, gestational age, delivery way, birth weight, meconium-stained amniotic fluid, 1- and 5-min Apgar score, need of neonatal intensive care unit, and small- and big-for-gestational age newborns were analyzed. Statistical analysis was performed by the c² test and by Student's t test, with the level of significance set at p<0.05. RESULTS: when breech and vertex groups were compared, they showed significant differences regarding the following variables: birth weight (3,091±538 g vs 3,250±497 g; p<0.01), vaginal delivery (63.8 vs 95.0%; p<0.0001), cesarean section (36.2 vs 5.0%; p<0,0001), and 1-min Apgar score (p<0.0001), respectively. CONCLUSIONS: we conclude that in term fetuses in breech position from pregnant women with previous vaginal deliveries, birth weight, delivery way, and 1-min Apgar score were different compared to fetuses in vertex position from women with the same characteristics.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(10):791-798
DOI 10.1590/S0100-72032004001000006
PURPOSE: to find out the preference in regard to the way of delivery among primigravidae, as well as the reasons for their choice, in order to improve the quality of the doctor-patient relationship. METHODS: a qualitative-type study was conducted through analysis of the collective subject, including primigravidae attended from September to November 2003 at the emergency rooms of the hospital of the "Faculdade de Medicina de Jundiaí". A questionnaire, specially developed to accomplish the proposed objectives was applied. An informed and free consent, signed by the pregnant woman and one of the researchers in charge was obtained. This questionnaire was based on doubts of patients attended at this hospital some months before the trial. For the purpose of sample standardization, the patients' selection followed some inclusion criteria: age above 16, primigravidae that were receiving prenatal assistance and a post-informed and free consent. Mental disorders and clinical and/or obstetric pathologies that could interfere in the patient's choice were considered exclusion criteria. RESULTS: the studied population had as prevailing profile women in the third quarter of gestation, above 21 years of age, white, married and with completed school. Most of the women (90%) preferred vaginal delivery for the following main reasons: ease to be done (94%) and the fear of suffering and pain during the postpartum period caused by cesarean section. There was a relationship between older and married women and the preference for vaginal delivery, with no significant difference between races. CONCLUSION: these results show an enormous contrast between women's preference and the high cesarean section rates in Brazil. We conclude that there may be a lack of information and dialogue between the health professionals and patients about the possible difficulties, doubts and anxieties that involve the women's choice for a specific way of delivery. From an ethical point of view, we conclude that obstetricians should question every cesarean section indication and take into account the women's right to choose, without ignoring clinical criteria, when making the medical decision about the way of delivery.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(10):791-798
DOI 10.1590/S0100-72032004001000006
PURPOSE: to find out the preference in regard to the way of delivery among primigravidae, as well as the reasons for their choice, in order to improve the quality of the doctor-patient relationship. METHODS: a qualitative-type study was conducted through analysis of the collective subject, including primigravidae attended from September to November 2003 at the emergency rooms of the hospital of the "Faculdade de Medicina de Jundiaí". A questionnaire, specially developed to accomplish the proposed objectives was applied. An informed and free consent, signed by the pregnant woman and one of the researchers in charge was obtained. This questionnaire was based on doubts of patients attended at this hospital some months before the trial. For the purpose of sample standardization, the patients' selection followed some inclusion criteria: age above 16, primigravidae that were receiving prenatal assistance and a post-informed and free consent. Mental disorders and clinical and/or obstetric pathologies that could interfere in the patient's choice were considered exclusion criteria. RESULTS: the studied population had as prevailing profile women in the third quarter of gestation, above 21 years of age, white, married and with completed school. Most of the women (90%) preferred vaginal delivery for the following main reasons: ease to be done (94%) and the fear of suffering and pain during the postpartum period caused by cesarean section. There was a relationship between older and married women and the preference for vaginal delivery, with no significant difference between races. CONCLUSION: these results show an enormous contrast between women's preference and the high cesarean section rates in Brazil. We conclude that there may be a lack of information and dialogue between the health professionals and patients about the possible difficulties, doubts and anxieties that involve the women's choice for a specific way of delivery. From an ethical point of view, we conclude that obstetricians should question every cesarean section indication and take into account the women's right to choose, without ignoring clinical criteria, when making the medical decision about the way of delivery.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(10):799-805
DOI 10.1590/S0100-72032004001000007
PURPOSE: to assess risk factors for low Apgar score. METHODS: this was a cross-sectional study preformed in a random sample of patients admitted to a level III maternity hospital in 2001. The outcome was low Apgar score defined as an Apgar score 1-6 (study group) versus Apgar score 7-10 (control group) in the first minute of life. The first step was the evaluation of the association of each possible risk factor with low Apgar score. The second step was multivariate analysis with the backward stepwise logistic regression model. RESULTS: there were 39 (14%) depressed newborns which were compared to 238 (86%) not depressed babies. The final analysis (multivariate) showed an association between low Apgar score and previous case of stillbirth (OR=52.6), preterm labor threat (OR=33.8), low birth weight, less than 2,500 g body weight (OR=11.2) and previous cesarean section (OR=7.4). Some factors acted as a protection, including birth weight, in grams (OR=0.9), female sex of the newborn (OR=0.1), medical complications (OR=0.4) and prematurity (gestational age < 37 weeks, OR=0.1). CONCLUSION: the study may help in the identification of fetuses at great risk of asphyxia, allowing proper reference within the health system and planning of effective assistance in neonatal intensive care units.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(10):799-805
DOI 10.1590/S0100-72032004001000007
PURPOSE: to assess risk factors for low Apgar score. METHODS: this was a cross-sectional study preformed in a random sample of patients admitted to a level III maternity hospital in 2001. The outcome was low Apgar score defined as an Apgar score 1-6 (study group) versus Apgar score 7-10 (control group) in the first minute of life. The first step was the evaluation of the association of each possible risk factor with low Apgar score. The second step was multivariate analysis with the backward stepwise logistic regression model. RESULTS: there were 39 (14%) depressed newborns which were compared to 238 (86%) not depressed babies. The final analysis (multivariate) showed an association between low Apgar score and previous case of stillbirth (OR=52.6), preterm labor threat (OR=33.8), low birth weight, less than 2,500 g body weight (OR=11.2) and previous cesarean section (OR=7.4). Some factors acted as a protection, including birth weight, in grams (OR=0.9), female sex of the newborn (OR=0.1), medical complications (OR=0.4) and prematurity (gestational age < 37 weeks, OR=0.1). CONCLUSION: the study may help in the identification of fetuses at great risk of asphyxia, allowing proper reference within the health system and planning of effective assistance in neonatal intensive care units.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(10):807-812
DOI 10.1590/S0100-72032004001000008
PURPOSE: placental villitis has been correlated with perinatal infection, although a percentage of cases remains etiologically unknown. The present study was aimed at the systematic morphological study of placentas for imunohistochemical characterization of villitis and assessment of its possible correlation with maternal and fetal outcome. METHODS: a hundred and twenty-eight placentas were studied. Gross examination was performed and all collected fragments were analyzed microscopically by the hematoxylin-eosin method. Villits was classified according to the inflammatory degree in to mild, moderate and severe. The immunohistochemical study to identify infectious agents was performed using monoclonal antibodies against Toxoplasma gondii and Cytomegalovirus. For inflammatory cell phenotype identification monoclonal antibodies against CD68, CD57, CD3, and CD20 were used. Statistical analysis was performed with the variables: maternal age and fetal gestational age, fetal and placental weight, and fetal and maternal outcomes. To compare the two groups we used the Mann-Whitney test and for proportions we used the chi2 test. The differences in the mean values between the treatment groups were considered statistically significant when p<0.05 (5%). RESULTS: villitis was identified in 11.7% of the cases. In 40% of the cases the children were stillborn (p=0.003). One case showed positive staining for toxoplasmosis while the remaining cases were negative. Imunohistochemical staining showed CD68+ cells, PanT+ cells and negative CD57 and PanB cells. CONCLUSION: we concluded that the intensity of the inflammatory process in the placenta was correlated with the severity of the fetal disease. The inflammatory cells in the villitis focus were macrophages; however, we could not identify infectious agents correlated with the villitis.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(10):807-812
DOI 10.1590/S0100-72032004001000008
PURPOSE: placental villitis has been correlated with perinatal infection, although a percentage of cases remains etiologically unknown. The present study was aimed at the systematic morphological study of placentas for imunohistochemical characterization of villitis and assessment of its possible correlation with maternal and fetal outcome. METHODS: a hundred and twenty-eight placentas were studied. Gross examination was performed and all collected fragments were analyzed microscopically by the hematoxylin-eosin method. Villits was classified according to the inflammatory degree in to mild, moderate and severe. The immunohistochemical study to identify infectious agents was performed using monoclonal antibodies against Toxoplasma gondii and Cytomegalovirus. For inflammatory cell phenotype identification monoclonal antibodies against CD68, CD57, CD3, and CD20 were used. Statistical analysis was performed with the variables: maternal age and fetal gestational age, fetal and placental weight, and fetal and maternal outcomes. To compare the two groups we used the Mann-Whitney test and for proportions we used the chi2 test. The differences in the mean values between the treatment groups were considered statistically significant when p<0.05 (5%). RESULTS: villitis was identified in 11.7% of the cases. In 40% of the cases the children were stillborn (p=0.003). One case showed positive staining for toxoplasmosis while the remaining cases were negative. Imunohistochemical staining showed CD68+ cells, PanT+ cells and negative CD57 and PanB cells. CONCLUSION: we concluded that the intensity of the inflammatory process in the placenta was correlated with the severity of the fetal disease. The inflammatory cells in the villitis focus were macrophages; however, we could not identify infectious agents correlated with the villitis.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(9):685-690
DOI 10.1590/S0100-72032004000900002
PURPOSE: to identify the responses of heart rate (HR), blood pressure (BP), and hydrostatic weight (HW) in pregnant women immersed up to different anatomic points as far as the xiphoid process. METHODS: eleven pregnant women underwent the following experimental procedure: 10 minutes in recumbent position for evaluation of HR and BP at rest; 2 minutes in standing position for evaluation of initial measures of HR, BP and mass, and one minute for each immersion depth. HR, BP and HW were measured after immersion up to the level of the ankle, knee, hip, navel, and xiphoid process, respectively. Descriptive statistics, test of normality (Shapiro-Wilks), homogeneity of variance test (Levene), one-way ANOVA and the Bonferroni test (SPSS version 8.0) were used, with significance at p<0.05. RESULTS: significant differences were found for HR, diastolic BP and mean BP starting from the xiphoid process (79.1±5.1 bpm; 53.3±6.7 mmHg and 63.9±6.2 mmHg, respectively) and for the systolic BP starting from the navel (92.7±11.1 mmHg). Significant differences were seen in all measurements of percent HW reduction, as in previous studies carried out with non-pregnant women. CONCLUSION: the obtained results showed a decrease in HR and BP on water immersion when compared non-immersion, as well as decreases in HW, which were proportional to the depth of immersion. The decrease in HW was found to influence the decrease in mechanical load imposed on the lower limb joints, since the mechanical load depends on both the vertical force (hydrostatic weight) and on the acceleration with which the body touches the ground. As a result, it is concluded that water is a healthy environment for the population under study, and may be adequate for the practice of physical activities.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(9):685-690
DOI 10.1590/S0100-72032004000900002
PURPOSE: to identify the responses of heart rate (HR), blood pressure (BP), and hydrostatic weight (HW) in pregnant women immersed up to different anatomic points as far as the xiphoid process. METHODS: eleven pregnant women underwent the following experimental procedure: 10 minutes in recumbent position for evaluation of HR and BP at rest; 2 minutes in standing position for evaluation of initial measures of HR, BP and mass, and one minute for each immersion depth. HR, BP and HW were measured after immersion up to the level of the ankle, knee, hip, navel, and xiphoid process, respectively. Descriptive statistics, test of normality (Shapiro-Wilks), homogeneity of variance test (Levene), one-way ANOVA and the Bonferroni test (SPSS version 8.0) were used, with significance at p<0.05. RESULTS: significant differences were found for HR, diastolic BP and mean BP starting from the xiphoid process (79.1±5.1 bpm; 53.3±6.7 mmHg and 63.9±6.2 mmHg, respectively) and for the systolic BP starting from the navel (92.7±11.1 mmHg). Significant differences were seen in all measurements of percent HW reduction, as in previous studies carried out with non-pregnant women. CONCLUSION: the obtained results showed a decrease in HR and BP on water immersion when compared non-immersion, as well as decreases in HW, which were proportional to the depth of immersion. The decrease in HW was found to influence the decrease in mechanical load imposed on the lower limb joints, since the mechanical load depends on both the vertical force (hydrostatic weight) and on the acceleration with which the body touches the ground. As a result, it is concluded that water is a healthy environment for the population under study, and may be adequate for the practice of physical activities.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(9):691-695
DOI 10.1590/S0100-72032004000900003
PURPOSE: to evaluate the importance of circulating maternal and fetal leptin in the healthy gestation, using its association with maternal, placental and fetal anthropometric variables, obtained at birth, and the relationship between the evaluated compartments. METHODS: in a transversal study a population of 33 single, healthy and term gestations was studied. The evaluated variables were maternal age, maternal weight, body mass index (BMF), weight of the newborn, placental weight, and placental index. Samples of maternal blood were immediately obtained before birth and from fetal umbilical cord blood at birth. Determination of serum leptin was performed using conventional radioimmunoassay. The relationships between serum leptin concentrations in maternal blood, umbilical artery and vein and the studied variables were assessed through linear regression. RESULTS: leptin levels were detected in the blood of all 33 pregnant women and their respective newborns, with maternal blood concentration (17.1±1.77 ng/mL) higher than that of umbilical vessels (vein: 9.0±1.16 ng/mL; artery: 8.23±1.02 ng/mL), p<0.0001. Leptin concentrations in the maternal blood were correlated with leptin concentrations in fetal blood (artery: coef. 0.63, p=0.037; vein: coef. 0.72, p=0.006). Regarding the anthropometric variables, leptin measured in the maternal blood was associated with initial and final maternal BMF (coef. 1.13; p=0.002; coef. 1,18, p=0.001) and cord leptin levels were correlated with the fetal weight at birth (vein: coef. 0.007, p=0.02; artery: coef. 0.006, p=0.02). CONCLUSION: there was a correlation between maternal and fetal leptin production and probably by the action of similar stimuli during gestation. Serum leptin was associated with the weight of the compartment where it circulates.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(9):691-695
DOI 10.1590/S0100-72032004000900003
PURPOSE: to evaluate the importance of circulating maternal and fetal leptin in the healthy gestation, using its association with maternal, placental and fetal anthropometric variables, obtained at birth, and the relationship between the evaluated compartments. METHODS: in a transversal study a population of 33 single, healthy and term gestations was studied. The evaluated variables were maternal age, maternal weight, body mass index (BMF), weight of the newborn, placental weight, and placental index. Samples of maternal blood were immediately obtained before birth and from fetal umbilical cord blood at birth. Determination of serum leptin was performed using conventional radioimmunoassay. The relationships between serum leptin concentrations in maternal blood, umbilical artery and vein and the studied variables were assessed through linear regression. RESULTS: leptin levels were detected in the blood of all 33 pregnant women and their respective newborns, with maternal blood concentration (17.1±1.77 ng/mL) higher than that of umbilical vessels (vein: 9.0±1.16 ng/mL; artery: 8.23±1.02 ng/mL), p<0.0001. Leptin concentrations in the maternal blood were correlated with leptin concentrations in fetal blood (artery: coef. 0.63, p=0.037; vein: coef. 0.72, p=0.006). Regarding the anthropometric variables, leptin measured in the maternal blood was associated with initial and final maternal BMF (coef. 1.13; p=0.002; coef. 1,18, p=0.001) and cord leptin levels were correlated with the fetal weight at birth (vein: coef. 0.007, p=0.02; artery: coef. 0.006, p=0.02). CONCLUSION: there was a correlation between maternal and fetal leptin production and probably by the action of similar stimuli during gestation. Serum leptin was associated with the weight of the compartment where it circulates.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(9):697-701
DOI 10.1590/S0100-72032004000900004
PURPOSE: to evaluate perinatal results in pregnant women over 35 years old and to check differences between two groups: 35 to 39-year-old women and women older than 40. METHODS: a retrospective survey was made during the period between January/2000 and July/2003, through the analysis of obstetric charts of 3,093 pregnant women who delivered in the "Hospital do Servidor Público Estadual - Francisco Morato de Oliveira", excluding 933 patients. The patients were divided into 3 groups: 18 to 29 years old (control group), 35 to 39 years old, and over 40 years old. Data collection was done with standardized forms, and the data were transferred to an electronic spreadsheet (Excel - Microsoft Office 2000). Statistical analysis was performed using the chi2 test and the Fisher test. The alpha risk was less or equal to 5% and the confidence interval 95%. RESULTS: cesarean section was the most used method not only in the 35 to 39-year-old group (438/792; 55.3%) but also in the group of women over 40 (153/236; 64.8%). The rates of prematurity (39/236; 16.5%), low weight (37/236; 15.7%), and restriction of fetal growth (38/236; 16.1%) were significantly higher in the group of women over 40, when compared to the other groups. Concerning fetal death, a five times higher incidence was observed in the group over 40 years old, as compared to the other groups, a statistically significant difference. CONCLUSION: the only difference between the 35 to 39-year-old group and the control group was the cesarean section rate. This allows us to suggest a differentiated prenatal attendance for pregnant women over 40.
Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2004;26(9):697-701
DOI 10.1590/S0100-72032004000900004
PURPOSE: to evaluate perinatal results in pregnant women over 35 years old and to check differences between two groups: 35 to 39-year-old women and women older than 40. METHODS: a retrospective survey was made during the period between January/2000 and July/2003, through the analysis of obstetric charts of 3,093 pregnant women who delivered in the "Hospital do Servidor Público Estadual - Francisco Morato de Oliveira", excluding 933 patients. The patients were divided into 3 groups: 18 to 29 years old (control group), 35 to 39 years old, and over 40 years old. Data collection was done with standardized forms, and the data were transferred to an electronic spreadsheet (Excel - Microsoft Office 2000). Statistical analysis was performed using the chi2 test and the Fisher test. The alpha risk was less or equal to 5% and the confidence interval 95%. RESULTS: cesarean section was the most used method not only in the 35 to 39-year-old group (438/792; 55.3%) but also in the group of women over 40 (153/236; 64.8%). The rates of prematurity (39/236; 16.5%), low weight (37/236; 15.7%), and restriction of fetal growth (38/236; 16.1%) were significantly higher in the group of women over 40, when compared to the other groups. Concerning fetal death, a five times higher incidence was observed in the group over 40 years old, as compared to the other groups, a statistically significant difference. CONCLUSION: the only difference between the 35 to 39-year-old group and the control group was the cesarean section rate. This allows us to suggest a differentiated prenatal attendance for pregnant women over 40.