Obstetrics

ORLANDO, FLA. — Having a child cuts a woman's risk of unprovoked venous thromboembolism, according to results of an epidemiologic study with more than 19,000 women.

Compared with women who were never pregnant, women with a history of at least one pregnancy had a 41% reduced risk of venous thromboembolism (VTE) in a multivariate model that was adjusted for several potential confounders, Christiana Iyasere, M.D., told attendees during the annual meeting of the American College of Cardiology.

The difference in risk was statistically significant, said Dr. Iyasere, a cardiologist at Brigham and Women's Hospital in Boston.

The overall incidence of unprovoked VTE in nulliparous women was about 0.8%, compared with a rate of about 0.5% in women who had at least one pregnancy.

This means that, on average, a history of pregnancy cut the population risk for unprovoked VTE by about 3 cases per 1,000 people.

The study used data collected in the Women's Health Study, which recruited postmenopausal women in the health professions with a primary goal of assessing the safety and efficacy of both aspirin and vitamin E for the primary prevention of cardiovascular disease.

Dr. Iyasere and her associates used data collected by questionnaire from the nearly 40,000 women who participated in this study to evaluate the impact of endogenous hormones on VTE risk.

They excluded women with a history of VTE, known hypercoagulable states, a history of prolonged immobility, a history of malignancy, or recent trauma.

The analyses were done using data collected from the remaining 19,219 women.

The researchers found no detectable association between VTE risk and age of menarche, age of menopause, and total duration of menses.

There was also no significant association between VTE risk and parity when the women were subdivided into four categories based on their child-bearing history: nulliparous, one or two pregnancies, three or four pregnancies, or five or more pregnancies. But a statistically significant association existed when the comparison was made between nulliparous women and those with any pregnancy.

Among 2,635 nulliparous women, 21 reported having an unprovoked VTE.

Among 16,584 women who were pregnant at least once, 86 had an unprovoked VTE.

The fully adjusted multivariate model corrected the hazard ratio for age, history of exogenous hormone use, body mass index, history of diabetes, history of hypertension, history of smoking, and degree of exercise activity.

The researchers were surprised to find no apparent link between other measures of endogenous hormone exposure and VTE risk because exogenous hormones, in the form of oral contraceptives and hormone therapy, are proven risk factors for an increased VTE incidence, Dr. Iyasere commented at the meeting.

ORLANDO, FLA. — Having a child cuts a woman's risk of unprovoked venous thromboembolism, according to results of an epidemiologic study with more than 19,000 women.

Compared with women who were never pregnant, women with a history of at least one pregnancy had a 41% reduced risk of venous thromboembolism (VTE) in a multivariate model that was adjusted for several potential confounders, Christiana Iyasere, M.D., told attendees during the annual meeting of the American College of Cardiology.

The difference in risk was statistically significant, said Dr. Iyasere, a cardiologist at Brigham and Women's Hospital in Boston.

The overall incidence of unprovoked VTE in nulliparous women was about 0.8%, compared with a rate of about 0.5% in women who had at least one pregnancy.

This means that, on average, a history of pregnancy cut the population risk for unprovoked VTE by about 3 cases per 1,000 people.

The study used data collected in the Women's Health Study, which recruited postmenopausal women in the health professions with a primary goal of assessing the safety and efficacy of both aspirin and vitamin E for the primary prevention of cardiovascular disease.

Dr. Iyasere and her associates used data collected by questionnaire from the nearly 40,000 women who participated in this study to evaluate the impact of endogenous hormones on VTE risk.

They excluded women with a history of VTE, known hypercoagulable states, a history of prolonged immobility, a history of malignancy, or recent trauma.

The analyses were done using data collected from the remaining 19,219 women.

The researchers found no detectable association between VTE risk and age of menarche, age of menopause, and total duration of menses.

There was also no significant association between VTE risk and parity when the women were subdivided into four categories based on their child-bearing history: nulliparous, one or two pregnancies, three or four pregnancies, or five or more pregnancies. But a statistically significant association existed when the comparison was made between nulliparous women and those with any pregnancy.

Among 2,635 nulliparous women, 21 reported having an unprovoked VTE.

Among 16,584 women who were pregnant at least once, 86 had an unprovoked VTE.

The fully adjusted multivariate model corrected the hazard ratio for age, history of exogenous hormone use, body mass index, history of diabetes, history of hypertension, history of smoking, and degree of exercise activity.

The researchers were surprised to find no apparent link between other measures of endogenous hormone exposure and VTE risk because exogenous hormones, in the form of oral contraceptives and hormone therapy, are proven risk factors for an increased VTE incidence, Dr. Iyasere commented at the meeting.

ORLANDO, FLA. — Having a child cuts a woman's risk of unprovoked venous thromboembolism, according to results of an epidemiologic study with more than 19,000 women.

Compared with women who were never pregnant, women with a history of at least one pregnancy had a 41% reduced risk of venous thromboembolism (VTE) in a multivariate model that was adjusted for several potential confounders, Christiana Iyasere, M.D., told attendees during the annual meeting of the American College of Cardiology.

The difference in risk was statistically significant, said Dr. Iyasere, a cardiologist at Brigham and Women's Hospital in Boston.

The overall incidence of unprovoked VTE in nulliparous women was about 0.8%, compared with a rate of about 0.5% in women who had at least one pregnancy.

This means that, on average, a history of pregnancy cut the population risk for unprovoked VTE by about 3 cases per 1,000 people.

The study used data collected in the Women's Health Study, which recruited postmenopausal women in the health professions with a primary goal of assessing the safety and efficacy of both aspirin and vitamin E for the primary prevention of cardiovascular disease.

Dr. Iyasere and her associates used data collected by questionnaire from the nearly 40,000 women who participated in this study to evaluate the impact of endogenous hormones on VTE risk.

They excluded women with a history of VTE, known hypercoagulable states, a history of prolonged immobility, a history of malignancy, or recent trauma.

The analyses were done using data collected from the remaining 19,219 women.

The researchers found no detectable association between VTE risk and age of menarche, age of menopause, and total duration of menses.

There was also no significant association between VTE risk and parity when the women were subdivided into four categories based on their child-bearing history: nulliparous, one or two pregnancies, three or four pregnancies, or five or more pregnancies. But a statistically significant association existed when the comparison was made between nulliparous women and those with any pregnancy.

Among 2,635 nulliparous women, 21 reported having an unprovoked VTE.

Among 16,584 women who were pregnant at least once, 86 had an unprovoked VTE.

The fully adjusted multivariate model corrected the hazard ratio for age, history of exogenous hormone use, body mass index, history of diabetes, history of hypertension, history of smoking, and degree of exercise activity.

The researchers were surprised to find no apparent link between other measures of endogenous hormone exposure and VTE risk because exogenous hormones, in the form of oral contraceptives and hormone therapy, are proven risk factors for an increased VTE incidence, Dr. Iyasere commented at the meeting.

RENO, NEV. — Glyburide may successfully control gestational diabetes in all but about 20% of patients, and a failed trial of glyburide appears to cause no long-term harm, Meredith Rochon, M.D., of the department of obstetrics, gynecology, and reproductive science at Mount Sinai School of Medicine, New York, and colleagues reported.

Glyburide treatment has been shown to be effective and safe for gestational diabetes.

The investigators reviewed the records of all patients with class A2 gestational diabetes treated with glyburide at a diabetes clinic over a period of 2 years to ensure that there were no adverse effects when the treatment failed.

The study found no reason to avoid using glyburide instead of insulin as first-line therapy, the researchers wrote in a poster presented at the annual meeting of the Society for Maternal-Fetal Medicine.

Of 83 patients identified, 18 (22%) were patients who underwent a trial of glyburide but failed to reach the target of fasting and postprandial-glucose levels of 60-90 mg/dL and 120 mg/dL, respectively, even when treated with a dose of 20 mg daily. Consequently, those patients were switched to insulin.

Despite their initial lack of blood glucose control, the pregnancy outcomes in the patients who failed treatment—including birth weight, mode of delivery, and incidence of macrosomia—were no different from those who were successfully managed on glyburide.

The sole difference in outcome was in the patients who were successfully treated with glyburide. Those patients had more neonates who required admission to the neonatal intensive care unit (NICU) than did the women who had been switched to insulin (35% vs. 11%).

The most common reason for the admission was hypoglycemia (10 of 23 admissions).

In an interview, Dr. Rochon said the finding was a surprise and something of a mystery, since glyburide does not cross the placenta, and previous studies have not noted this potential association.

The hypoglycemia was not considered by the investigators to be a serious adverse effect because it was transient in all cases. However, if it proves to be true that glyburide treatment does produce a higher rate of hypoglycemic neonates who need NICU admission, it may have significant cost implications, Dr. Rochon said.

The mean length of stay was 8 days for all the babies that were admitted to the NICU and 4 days for those admitted with hypoglycemia.

RENO, NEV. — Glyburide may successfully control gestational diabetes in all but about 20% of patients, and a failed trial of glyburide appears to cause no long-term harm, Meredith Rochon, M.D., of the department of obstetrics, gynecology, and reproductive science at Mount Sinai School of Medicine, New York, and colleagues reported.

Glyburide treatment has been shown to be effective and safe for gestational diabetes.

The investigators reviewed the records of all patients with class A2 gestational diabetes treated with glyburide at a diabetes clinic over a period of 2 years to ensure that there were no adverse effects when the treatment failed.

The study found no reason to avoid using glyburide instead of insulin as first-line therapy, the researchers wrote in a poster presented at the annual meeting of the Society for Maternal-Fetal Medicine.

Of 83 patients identified, 18 (22%) were patients who underwent a trial of glyburide but failed to reach the target of fasting and postprandial-glucose levels of 60-90 mg/dL and 120 mg/dL, respectively, even when treated with a dose of 20 mg daily. Consequently, those patients were switched to insulin.

Despite their initial lack of blood glucose control, the pregnancy outcomes in the patients who failed treatment—including birth weight, mode of delivery, and incidence of macrosomia—were no different from those who were successfully managed on glyburide.

The sole difference in outcome was in the patients who were successfully treated with glyburide. Those patients had more neonates who required admission to the neonatal intensive care unit (NICU) than did the women who had been switched to insulin (35% vs. 11%).

The most common reason for the admission was hypoglycemia (10 of 23 admissions).

In an interview, Dr. Rochon said the finding was a surprise and something of a mystery, since glyburide does not cross the placenta, and previous studies have not noted this potential association.

The hypoglycemia was not considered by the investigators to be a serious adverse effect because it was transient in all cases. However, if it proves to be true that glyburide treatment does produce a higher rate of hypoglycemic neonates who need NICU admission, it may have significant cost implications, Dr. Rochon said.

The mean length of stay was 8 days for all the babies that were admitted to the NICU and 4 days for those admitted with hypoglycemia.

RENO, NEV. — Glyburide may successfully control gestational diabetes in all but about 20% of patients, and a failed trial of glyburide appears to cause no long-term harm, Meredith Rochon, M.D., of the department of obstetrics, gynecology, and reproductive science at Mount Sinai School of Medicine, New York, and colleagues reported.

Glyburide treatment has been shown to be effective and safe for gestational diabetes.

The investigators reviewed the records of all patients with class A2 gestational diabetes treated with glyburide at a diabetes clinic over a period of 2 years to ensure that there were no adverse effects when the treatment failed.

The study found no reason to avoid using glyburide instead of insulin as first-line therapy, the researchers wrote in a poster presented at the annual meeting of the Society for Maternal-Fetal Medicine.

Of 83 patients identified, 18 (22%) were patients who underwent a trial of glyburide but failed to reach the target of fasting and postprandial-glucose levels of 60-90 mg/dL and 120 mg/dL, respectively, even when treated with a dose of 20 mg daily. Consequently, those patients were switched to insulin.

Despite their initial lack of blood glucose control, the pregnancy outcomes in the patients who failed treatment—including birth weight, mode of delivery, and incidence of macrosomia—were no different from those who were successfully managed on glyburide.

The sole difference in outcome was in the patients who were successfully treated with glyburide. Those patients had more neonates who required admission to the neonatal intensive care unit (NICU) than did the women who had been switched to insulin (35% vs. 11%).

The most common reason for the admission was hypoglycemia (10 of 23 admissions).

In an interview, Dr. Rochon said the finding was a surprise and something of a mystery, since glyburide does not cross the placenta, and previous studies have not noted this potential association.

The hypoglycemia was not considered by the investigators to be a serious adverse effect because it was transient in all cases. However, if it proves to be true that glyburide treatment does produce a higher rate of hypoglycemic neonates who need NICU admission, it may have significant cost implications, Dr. Rochon said.

The mean length of stay was 8 days for all the babies that were admitted to the NICU and 4 days for those admitted with hypoglycemia.

WASHINGTON — Women who experienced hyperemesis gravidarum had a significantly increased risk of preeclampsia, compared with controls, wrote Michele Soltis, M.D., and colleagues in a poster presented at the annual meeting of the American College of Preventive Medicine.

Hyperemesis gravidarum was associated with several obstetric complications in a retrospective study of singleton pregnancies, reported Dr. Soltis of Madigan Army Medical Center in Tacoma, Washington, and her associates.

In this retrospective study, the investigators compared 4,808 women hospitalized for hyperemesis gravidarum with 9,616 controls and calculated the relative risks for certain obstetric outcomes.

Women with hyperemesis gravidarum had relative risks of 1.3 for preeclampsia, 1.3 for infant birth weight less than 2,500 g, and 2.1 for premature deliveries before 28 weeks' gestation. The relative risk of premature delivery at 28-32 weeks or at 33-36 weeks was 1.5.

In addition, hyperemesis gravidarum was associated with a longer hospital stay after both vaginal and cesarean deliveries.

WASHINGTON — Women who experienced hyperemesis gravidarum had a significantly increased risk of preeclampsia, compared with controls, wrote Michele Soltis, M.D., and colleagues in a poster presented at the annual meeting of the American College of Preventive Medicine.

Hyperemesis gravidarum was associated with several obstetric complications in a retrospective study of singleton pregnancies, reported Dr. Soltis of Madigan Army Medical Center in Tacoma, Washington, and her associates.

In this retrospective study, the investigators compared 4,808 women hospitalized for hyperemesis gravidarum with 9,616 controls and calculated the relative risks for certain obstetric outcomes.

Women with hyperemesis gravidarum had relative risks of 1.3 for preeclampsia, 1.3 for infant birth weight less than 2,500 g, and 2.1 for premature deliveries before 28 weeks' gestation. The relative risk of premature delivery at 28-32 weeks or at 33-36 weeks was 1.5.

In addition, hyperemesis gravidarum was associated with a longer hospital stay after both vaginal and cesarean deliveries.

WASHINGTON — Women who experienced hyperemesis gravidarum had a significantly increased risk of preeclampsia, compared with controls, wrote Michele Soltis, M.D., and colleagues in a poster presented at the annual meeting of the American College of Preventive Medicine.

Hyperemesis gravidarum was associated with several obstetric complications in a retrospective study of singleton pregnancies, reported Dr. Soltis of Madigan Army Medical Center in Tacoma, Washington, and her associates.

In this retrospective study, the investigators compared 4,808 women hospitalized for hyperemesis gravidarum with 9,616 controls and calculated the relative risks for certain obstetric outcomes.

Women with hyperemesis gravidarum had relative risks of 1.3 for preeclampsia, 1.3 for infant birth weight less than 2,500 g, and 2.1 for premature deliveries before 28 weeks' gestation. The relative risk of premature delivery at 28-32 weeks or at 33-36 weeks was 1.5.

In addition, hyperemesis gravidarum was associated with a longer hospital stay after both vaginal and cesarean deliveries.

RENO, NEV. — Oral glucose challenge test results appear to have different values for patients of different ethnicity, reported Tania Esakoff, M.D.

Specifically, Asians may need to meet a higher threshold cutoff than whites before undergoing a 3-hour fasting glucose tolerance test to determine whether they have gestational diabetes, and African Americans need a lower cutoff, Dr. Esakoff and her colleagues wrote in a poster presented at the annual meeting of the Society for Maternal-Fetal Medicine.

The researchers reviewed the records of 14,565 pregnancies screened for gestational diabetes with the 50-g glucose loading test in ethnically diverse San Francisco. They then computed the sensitivities and false-positive rates based on study results and known results for fasting glucose tolerance testing.

The results show Asians tend to be more sensitive to an oral glucose challenge and therefore need a higher cutoff, said Dr. Esakoff of the University of California, San Francisco.

With the currently recommended cutoff of a serum-glucose level of 140 mg/dL, which is now used for everyone regardless of ethnicity, the test would have a sensitivity of 89% and a false-positive rate of 14% for Asian women.

To reduce the false-positive rate closer to 10%, Asians need to have a cutoff threshold of 145 mg/dL, but then the sensitivity is only 80%.

In African Americans, on the other hand, the test at a cutoff of 140 mg/dL has a sensitivity of 94%, but a false-positive rate of only 6%.

The study suggests if the goal is to have the maximum sensitivity with a near 10% false-positive rate, the threshold cutoff glucose level for African Americans should be 135 mg/dL; for Hispanics and whites, 140 mg/dL; for Asians, 145 mg/dL.

If the goal is to have at least a 95% sensitivity, then the cutoff needs to be 135 mg/dL for African Americans (sensitivity 97%, false-positive rate 9%), and 130 mg/dL for the other ethnic groups.

At the 130 mg/dL cutoff, the false-positive rates are high for those groups (whites 18%, Hispanics 19%, and Asians 24%), but a higher cutoff of 135 mg/dL does not achieve a 95% sensitivity.

The researchers also found an overall prevalence of gestational diabetes in its population of 6%, with a prevalence among whites of 4%, and African Americans of 4%, and Hispanics of 7%, and a prevalence among Asians of 10%.

RENO, NEV. — Oral glucose challenge test results appear to have different values for patients of different ethnicity, reported Tania Esakoff, M.D.

Specifically, Asians may need to meet a higher threshold cutoff than whites before undergoing a 3-hour fasting glucose tolerance test to determine whether they have gestational diabetes, and African Americans need a lower cutoff, Dr. Esakoff and her colleagues wrote in a poster presented at the annual meeting of the Society for Maternal-Fetal Medicine.

The researchers reviewed the records of 14,565 pregnancies screened for gestational diabetes with the 50-g glucose loading test in ethnically diverse San Francisco. They then computed the sensitivities and false-positive rates based on study results and known results for fasting glucose tolerance testing.

The results show Asians tend to be more sensitive to an oral glucose challenge and therefore need a higher cutoff, said Dr. Esakoff of the University of California, San Francisco.

With the currently recommended cutoff of a serum-glucose level of 140 mg/dL, which is now used for everyone regardless of ethnicity, the test would have a sensitivity of 89% and a false-positive rate of 14% for Asian women.

To reduce the false-positive rate closer to 10%, Asians need to have a cutoff threshold of 145 mg/dL, but then the sensitivity is only 80%.

In African Americans, on the other hand, the test at a cutoff of 140 mg/dL has a sensitivity of 94%, but a false-positive rate of only 6%.

The study suggests if the goal is to have the maximum sensitivity with a near 10% false-positive rate, the threshold cutoff glucose level for African Americans should be 135 mg/dL; for Hispanics and whites, 140 mg/dL; for Asians, 145 mg/dL.

If the goal is to have at least a 95% sensitivity, then the cutoff needs to be 135 mg/dL for African Americans (sensitivity 97%, false-positive rate 9%), and 130 mg/dL for the other ethnic groups.

At the 130 mg/dL cutoff, the false-positive rates are high for those groups (whites 18%, Hispanics 19%, and Asians 24%), but a higher cutoff of 135 mg/dL does not achieve a 95% sensitivity.

The researchers also found an overall prevalence of gestational diabetes in its population of 6%, with a prevalence among whites of 4%, and African Americans of 4%, and Hispanics of 7%, and a prevalence among Asians of 10%.

RENO, NEV. — Oral glucose challenge test results appear to have different values for patients of different ethnicity, reported Tania Esakoff, M.D.

Specifically, Asians may need to meet a higher threshold cutoff than whites before undergoing a 3-hour fasting glucose tolerance test to determine whether they have gestational diabetes, and African Americans need a lower cutoff, Dr. Esakoff and her colleagues wrote in a poster presented at the annual meeting of the Society for Maternal-Fetal Medicine.

The researchers reviewed the records of 14,565 pregnancies screened for gestational diabetes with the 50-g glucose loading test in ethnically diverse San Francisco. They then computed the sensitivities and false-positive rates based on study results and known results for fasting glucose tolerance testing.

The results show Asians tend to be more sensitive to an oral glucose challenge and therefore need a higher cutoff, said Dr. Esakoff of the University of California, San Francisco.

With the currently recommended cutoff of a serum-glucose level of 140 mg/dL, which is now used for everyone regardless of ethnicity, the test would have a sensitivity of 89% and a false-positive rate of 14% for Asian women.

To reduce the false-positive rate closer to 10%, Asians need to have a cutoff threshold of 145 mg/dL, but then the sensitivity is only 80%.

In African Americans, on the other hand, the test at a cutoff of 140 mg/dL has a sensitivity of 94%, but a false-positive rate of only 6%.

The study suggests if the goal is to have the maximum sensitivity with a near 10% false-positive rate, the threshold cutoff glucose level for African Americans should be 135 mg/dL; for Hispanics and whites, 140 mg/dL; for Asians, 145 mg/dL.

If the goal is to have at least a 95% sensitivity, then the cutoff needs to be 135 mg/dL for African Americans (sensitivity 97%, false-positive rate 9%), and 130 mg/dL for the other ethnic groups.

At the 130 mg/dL cutoff, the false-positive rates are high for those groups (whites 18%, Hispanics 19%, and Asians 24%), but a higher cutoff of 135 mg/dL does not achieve a 95% sensitivity.

The researchers also found an overall prevalence of gestational diabetes in its population of 6%, with a prevalence among whites of 4%, and African Americans of 4%, and Hispanics of 7%, and a prevalence among Asians of 10%.

KEVIN FOLEY, RESEARCH/LOUISE KOENIG, DESIGN

KEVIN FOLEY, RESEARCH/LOUISE KOENIG, DESIGN

KEVIN FOLEY, RESEARCH/LOUISE KOENIG, DESIGN

RENO, NEV. — Newborn infants with neurologic injury resulting in death are twice as likely to have had a nighttime delivery as healthy infants, according to a study presented by Adam C. Urato, M.D., at the annual meeting of the Society for Maternal-Fetal Medicine.

The causes of this association are unclear, but may include prolonged labor, differences in staffing at night, and errors by medical personnel due to fatigue, said Dr. Urato of Tufts University, Boston.

The study made use of a Florida state database of perinatal neurologic injuries, created as a result of a law providing for no-fault compensation of families whose children suffer a birth-related neurologic injury.

Of the 447 cases of neurologic injury recorded from 1989 to 2002, there were 80 that resulted in the death of an infant. The database is limited to live-born children with brain or spinal cord injury whose birth weight was greater than 2,500 g. The neurologic problems were caused by oxygen deprivation or mechanical injury occurring in a hospital during labor, birth, delivery, or post delivery.

The 80 deaths were compared with a control group of 999 randomly selected births in Florida from 1996, the midpoint of the time span covered in the neurologic injury database, according to Dr. Urato.

Among the control group, 28% were born during the hours of 11 p.m. to 8 a.m. But 45% of the births with neurologic injury resulting in death occurred during those nighttime hours. Even after correcting the odds ratio for repeat cesarean deliveries, the deaths were 1.95 times more likely to occur at night, a statistically significant increase in risk.

Index cases were significantly less likely than controls to have been born via a normal spontaneous vaginal delivery (12.3% vs. 70.7%), and significantly more likely to have been born via cesarean section (71.7% vs. 21.4%), vacuum delivery (12.3% vs. 6.4%), or forceps delivery (3.7% vs. 1.5%).

Prolonged labor is one possible explanation for the excess of nighttime neurologic injuries resulting in death. Spontaneous labor often leads to an afternoon delivery, but most inductions start during the day. Labors that continue into the night may represent a group at higher risk. But a review of the 80 deaths showed that most did not follow prolonged labor.

Nighttime differences in staffing might be another explanation. All hospital departments, including anesthesia, nursing, pediatrics, and obstetrics, have lower staffing levels at night. But of the 80 deaths, a staffing issue was noted in only a single chart. In that case, the obstetrician was forced to wait for an operating room to perform a cesarean. Other staffing delays might not have been mentioned in the charts, Dr. Urato said.

The third explanation involves fatigue by medical personnel. At some point from 11 p.m. to 8 a.m. most people are at their circadian nadir, with an increased sleep propensity. Many studies of workers have documented an increased risk of accidents and errors at night. No mention of fatigue was made in any of the charts, however.

“Perhaps more than any other field of medicine, obstetrics is associated with nighttime work,” he said. “The image of the obstetrician awakened in the middle of the night to deliver a baby is embedded in our collective consciousness. This finding of increased risk in our field of obstetrics should not surprise us given the preponderance of evidence for other fields that demonstrate that night is indeed a time of increased risk.”

RENO, NEV. — Newborn infants with neurologic injury resulting in death are twice as likely to have had a nighttime delivery as healthy infants, according to a study presented by Adam C. Urato, M.D., at the annual meeting of the Society for Maternal-Fetal Medicine.

The causes of this association are unclear, but may include prolonged labor, differences in staffing at night, and errors by medical personnel due to fatigue, said Dr. Urato of Tufts University, Boston.

The study made use of a Florida state database of perinatal neurologic injuries, created as a result of a law providing for no-fault compensation of families whose children suffer a birth-related neurologic injury.

Of the 447 cases of neurologic injury recorded from 1989 to 2002, there were 80 that resulted in the death of an infant. The database is limited to live-born children with brain or spinal cord injury whose birth weight was greater than 2,500 g. The neurologic problems were caused by oxygen deprivation or mechanical injury occurring in a hospital during labor, birth, delivery, or post delivery.

The 80 deaths were compared with a control group of 999 randomly selected births in Florida from 1996, the midpoint of the time span covered in the neurologic injury database, according to Dr. Urato.

Among the control group, 28% were born during the hours of 11 p.m. to 8 a.m. But 45% of the births with neurologic injury resulting in death occurred during those nighttime hours. Even after correcting the odds ratio for repeat cesarean deliveries, the deaths were 1.95 times more likely to occur at night, a statistically significant increase in risk.

Index cases were significantly less likely than controls to have been born via a normal spontaneous vaginal delivery (12.3% vs. 70.7%), and significantly more likely to have been born via cesarean section (71.7% vs. 21.4%), vacuum delivery (12.3% vs. 6.4%), or forceps delivery (3.7% vs. 1.5%).

Prolonged labor is one possible explanation for the excess of nighttime neurologic injuries resulting in death. Spontaneous labor often leads to an afternoon delivery, but most inductions start during the day. Labors that continue into the night may represent a group at higher risk. But a review of the 80 deaths showed that most did not follow prolonged labor.

Nighttime differences in staffing might be another explanation. All hospital departments, including anesthesia, nursing, pediatrics, and obstetrics, have lower staffing levels at night. But of the 80 deaths, a staffing issue was noted in only a single chart. In that case, the obstetrician was forced to wait for an operating room to perform a cesarean. Other staffing delays might not have been mentioned in the charts, Dr. Urato said.

The third explanation involves fatigue by medical personnel. At some point from 11 p.m. to 8 a.m. most people are at their circadian nadir, with an increased sleep propensity. Many studies of workers have documented an increased risk of accidents and errors at night. No mention of fatigue was made in any of the charts, however.

“Perhaps more than any other field of medicine, obstetrics is associated with nighttime work,” he said. “The image of the obstetrician awakened in the middle of the night to deliver a baby is embedded in our collective consciousness. This finding of increased risk in our field of obstetrics should not surprise us given the preponderance of evidence for other fields that demonstrate that night is indeed a time of increased risk.”

RENO, NEV. — Newborn infants with neurologic injury resulting in death are twice as likely to have had a nighttime delivery as healthy infants, according to a study presented by Adam C. Urato, M.D., at the annual meeting of the Society for Maternal-Fetal Medicine.

The causes of this association are unclear, but may include prolonged labor, differences in staffing at night, and errors by medical personnel due to fatigue, said Dr. Urato of Tufts University, Boston.

The study made use of a Florida state database of perinatal neurologic injuries, created as a result of a law providing for no-fault compensation of families whose children suffer a birth-related neurologic injury.

Of the 447 cases of neurologic injury recorded from 1989 to 2002, there were 80 that resulted in the death of an infant. The database is limited to live-born children with brain or spinal cord injury whose birth weight was greater than 2,500 g. The neurologic problems were caused by oxygen deprivation or mechanical injury occurring in a hospital during labor, birth, delivery, or post delivery.

The 80 deaths were compared with a control group of 999 randomly selected births in Florida from 1996, the midpoint of the time span covered in the neurologic injury database, according to Dr. Urato.

Among the control group, 28% were born during the hours of 11 p.m. to 8 a.m. But 45% of the births with neurologic injury resulting in death occurred during those nighttime hours. Even after correcting the odds ratio for repeat cesarean deliveries, the deaths were 1.95 times more likely to occur at night, a statistically significant increase in risk.

Index cases were significantly less likely than controls to have been born via a normal spontaneous vaginal delivery (12.3% vs. 70.7%), and significantly more likely to have been born via cesarean section (71.7% vs. 21.4%), vacuum delivery (12.3% vs. 6.4%), or forceps delivery (3.7% vs. 1.5%).

Prolonged labor is one possible explanation for the excess of nighttime neurologic injuries resulting in death. Spontaneous labor often leads to an afternoon delivery, but most inductions start during the day. Labors that continue into the night may represent a group at higher risk. But a review of the 80 deaths showed that most did not follow prolonged labor.

Nighttime differences in staffing might be another explanation. All hospital departments, including anesthesia, nursing, pediatrics, and obstetrics, have lower staffing levels at night. But of the 80 deaths, a staffing issue was noted in only a single chart. In that case, the obstetrician was forced to wait for an operating room to perform a cesarean. Other staffing delays might not have been mentioned in the charts, Dr. Urato said.

The third explanation involves fatigue by medical personnel. At some point from 11 p.m. to 8 a.m. most people are at their circadian nadir, with an increased sleep propensity. Many studies of workers have documented an increased risk of accidents and errors at night. No mention of fatigue was made in any of the charts, however.

“Perhaps more than any other field of medicine, obstetrics is associated with nighttime work,” he said. “The image of the obstetrician awakened in the middle of the night to deliver a baby is embedded in our collective consciousness. This finding of increased risk in our field of obstetrics should not surprise us given the preponderance of evidence for other fields that demonstrate that night is indeed a time of increased risk.”

WASHINGTON — Patients who received emergent cerclage had higher rates of premature birth, spontaneous premature rupture of membranes, and chorioamnionitis, compared with those receiving elective cerclages, Chi P. Dola, M.D., reported in a poster presentation at the annual meeting of District VII of the American College of Obstetricians and Gynecologists.

Dr. Dola, of Tulane University, New Orleans, and her colleagues performed a retrospective chart review of data from 140 pregnant women who presented with an incompetent cervix.

The women were separated into three groups: 91 asymptomatic patients who received an elective cerclage based on a history suggestive of cervical incompetence; 29 asymptomatic patients who received an urgent cerclage after an ultrasound exam brought to light an abnormal finding, and 20 patients who received an emergent cerclage after presenting with typical symptoms of cervical incompetence.

The percentage of patients whose pregnancy lasted beyond 36 weeks was the highest, at 73%, in those who received an elective cerclage, followed by 62% in the asymptomatic patients who received urgent cerclage and 21% in the symptomatic patients who had an urgent cerclage.

Spontaneous premature rupture of membranes was 21% in the asymptomatic patients who received elective cerclage, versus 40% and 71% in the other two groups, respectively.

The incidence of chorioamnionitis was only 2% in the asymptomatic patients who received elective cerclage, compared with 9% and 50% in the other two groups, respectively.

All differences between groups were significant.

The researchers' findings show that patients with urgent cerclages had poor outcomes compared with those receiving elective cerclages. This finding led the researchers to conclude “that by awaiting abnormal cervical findings on ultrasound exam prior to placement of an urgent cerclage, the potential for poor pregnancy outcomes increases.”

WASHINGTON — Patients who received emergent cerclage had higher rates of premature birth, spontaneous premature rupture of membranes, and chorioamnionitis, compared with those receiving elective cerclages, Chi P. Dola, M.D., reported in a poster presentation at the annual meeting of District VII of the American College of Obstetricians and Gynecologists.

Dr. Dola, of Tulane University, New Orleans, and her colleagues performed a retrospective chart review of data from 140 pregnant women who presented with an incompetent cervix.

The women were separated into three groups: 91 asymptomatic patients who received an elective cerclage based on a history suggestive of cervical incompetence; 29 asymptomatic patients who received an urgent cerclage after an ultrasound exam brought to light an abnormal finding, and 20 patients who received an emergent cerclage after presenting with typical symptoms of cervical incompetence.

The percentage of patients whose pregnancy lasted beyond 36 weeks was the highest, at 73%, in those who received an elective cerclage, followed by 62% in the asymptomatic patients who received urgent cerclage and 21% in the symptomatic patients who had an urgent cerclage.

Spontaneous premature rupture of membranes was 21% in the asymptomatic patients who received elective cerclage, versus 40% and 71% in the other two groups, respectively.

The incidence of chorioamnionitis was only 2% in the asymptomatic patients who received elective cerclage, compared with 9% and 50% in the other two groups, respectively.

All differences between groups were significant.

The researchers' findings show that patients with urgent cerclages had poor outcomes compared with those receiving elective cerclages. This finding led the researchers to conclude “that by awaiting abnormal cervical findings on ultrasound exam prior to placement of an urgent cerclage, the potential for poor pregnancy outcomes increases.”

WASHINGTON — Patients who received emergent cerclage had higher rates of premature birth, spontaneous premature rupture of membranes, and chorioamnionitis, compared with those receiving elective cerclages, Chi P. Dola, M.D., reported in a poster presentation at the annual meeting of District VII of the American College of Obstetricians and Gynecologists.

Dr. Dola, of Tulane University, New Orleans, and her colleagues performed a retrospective chart review of data from 140 pregnant women who presented with an incompetent cervix.

The women were separated into three groups: 91 asymptomatic patients who received an elective cerclage based on a history suggestive of cervical incompetence; 29 asymptomatic patients who received an urgent cerclage after an ultrasound exam brought to light an abnormal finding, and 20 patients who received an emergent cerclage after presenting with typical symptoms of cervical incompetence.

The percentage of patients whose pregnancy lasted beyond 36 weeks was the highest, at 73%, in those who received an elective cerclage, followed by 62% in the asymptomatic patients who received urgent cerclage and 21% in the symptomatic patients who had an urgent cerclage.

Spontaneous premature rupture of membranes was 21% in the asymptomatic patients who received elective cerclage, versus 40% and 71% in the other two groups, respectively.

The incidence of chorioamnionitis was only 2% in the asymptomatic patients who received elective cerclage, compared with 9% and 50% in the other two groups, respectively.

All differences between groups were significant.

The researchers' findings show that patients with urgent cerclages had poor outcomes compared with those receiving elective cerclages. This finding led the researchers to conclude “that by awaiting abnormal cervical findings on ultrasound exam prior to placement of an urgent cerclage, the potential for poor pregnancy outcomes increases.”

RENO, NEV. — Fetal echocardiography before 16 weeks of gestation is feasible and can detect a substantial proportion of cardiac lesions, investigators reported in a poster presentation at the annual meeting of the Society for Maternal-Fetal Medicine.

The technique does have limitations regarding accurate visualization of the great artery relationship and the crux of the heart. It may therefore be best to reserve early echocardiography for cases at the greatest risk for cardiac defects, as judged by increased nuchal translucency and the presence of extracardiac lesions. Second-trimester echocardiograms remain the gold standard, concluded Fionnuala McAuliffe, M.D., of University College Dublin (Ireland) and colleagues.

The study involved 160 fetal echocardiograms performed before the 16th week, with an average gestation time of 13.5 weeks. Investigators used the transabdominal approach for 100 cases, and the transvaginal approach in 60 cases in which the transabdominal approach yielded poor visualization.

Of the 160 patients, 100 were referred because of nuchal translucency greater than the 95th percentile, 51 because of a family history of congenital cardiac defects, and 9 because of the presence of extracardiac lesions.

Adequate cardiac examinations were possible in 152 cases, and pregnancy outcome was available in 137 cases. Of those, there were 20 cardiac defects. Fourteen (70%) showed an abnormality on the early echocardiogram, and 6 (30%) were passed as normal.

The early echocardiogram identified two cases of ectopia cordis, two cases of atrioventricular septal defect, two cases of hypoplastic left heart syndrome, two cases of ventricular septal defect, two cases of left atrial isomerism, two cases of hypoplastic right ventricle, and one case each of double outlet right ventricle and cardiac diverticulum.

On the other hand, the early echocardiogram failed to detect three cases of ventricular septal defect, two cases of dextro-looped transposition of the great arteries, and one case of hypertrophic cardiomyopathy.

A four-chamber view of the heart was obtained in all of the cases. The atrioventricular valves could be visualized 96% of the time, the aorta and pulmonary artery 95% of the time, and the inferior and superior vena cava 76% of the time.

Early fetal echocardiography was less effective in visualizing the aortic and ductal arches (45% of the time), branch pulmonary arteries (37% of the time), and pulmonary veins (19% of the time).

RENO, NEV. — Fetal echocardiography before 16 weeks of gestation is feasible and can detect a substantial proportion of cardiac lesions, investigators reported in a poster presentation at the annual meeting of the Society for Maternal-Fetal Medicine.

The technique does have limitations regarding accurate visualization of the great artery relationship and the crux of the heart. It may therefore be best to reserve early echocardiography for cases at the greatest risk for cardiac defects, as judged by increased nuchal translucency and the presence of extracardiac lesions. Second-trimester echocardiograms remain the gold standard, concluded Fionnuala McAuliffe, M.D., of University College Dublin (Ireland) and colleagues.

The study involved 160 fetal echocardiograms performed before the 16th week, with an average gestation time of 13.5 weeks. Investigators used the transabdominal approach for 100 cases, and the transvaginal approach in 60 cases in which the transabdominal approach yielded poor visualization.

Of the 160 patients, 100 were referred because of nuchal translucency greater than the 95th percentile, 51 because of a family history of congenital cardiac defects, and 9 because of the presence of extracardiac lesions.

Adequate cardiac examinations were possible in 152 cases, and pregnancy outcome was available in 137 cases. Of those, there were 20 cardiac defects. Fourteen (70%) showed an abnormality on the early echocardiogram, and 6 (30%) were passed as normal.

The early echocardiogram identified two cases of ectopia cordis, two cases of atrioventricular septal defect, two cases of hypoplastic left heart syndrome, two cases of ventricular septal defect, two cases of left atrial isomerism, two cases of hypoplastic right ventricle, and one case each of double outlet right ventricle and cardiac diverticulum.

On the other hand, the early echocardiogram failed to detect three cases of ventricular septal defect, two cases of dextro-looped transposition of the great arteries, and one case of hypertrophic cardiomyopathy.

A four-chamber view of the heart was obtained in all of the cases. The atrioventricular valves could be visualized 96% of the time, the aorta and pulmonary artery 95% of the time, and the inferior and superior vena cava 76% of the time.

Early fetal echocardiography was less effective in visualizing the aortic and ductal arches (45% of the time), branch pulmonary arteries (37% of the time), and pulmonary veins (19% of the time).

RENO, NEV. — Fetal echocardiography before 16 weeks of gestation is feasible and can detect a substantial proportion of cardiac lesions, investigators reported in a poster presentation at the annual meeting of the Society for Maternal-Fetal Medicine.

The technique does have limitations regarding accurate visualization of the great artery relationship and the crux of the heart. It may therefore be best to reserve early echocardiography for cases at the greatest risk for cardiac defects, as judged by increased nuchal translucency and the presence of extracardiac lesions. Second-trimester echocardiograms remain the gold standard, concluded Fionnuala McAuliffe, M.D., of University College Dublin (Ireland) and colleagues.

The study involved 160 fetal echocardiograms performed before the 16th week, with an average gestation time of 13.5 weeks. Investigators used the transabdominal approach for 100 cases, and the transvaginal approach in 60 cases in which the transabdominal approach yielded poor visualization.

Of the 160 patients, 100 were referred because of nuchal translucency greater than the 95th percentile, 51 because of a family history of congenital cardiac defects, and 9 because of the presence of extracardiac lesions.

Adequate cardiac examinations were possible in 152 cases, and pregnancy outcome was available in 137 cases. Of those, there were 20 cardiac defects. Fourteen (70%) showed an abnormality on the early echocardiogram, and 6 (30%) were passed as normal.

The early echocardiogram identified two cases of ectopia cordis, two cases of atrioventricular septal defect, two cases of hypoplastic left heart syndrome, two cases of ventricular septal defect, two cases of left atrial isomerism, two cases of hypoplastic right ventricle, and one case each of double outlet right ventricle and cardiac diverticulum.

On the other hand, the early echocardiogram failed to detect three cases of ventricular septal defect, two cases of dextro-looped transposition of the great arteries, and one case of hypertrophic cardiomyopathy.

A four-chamber view of the heart was obtained in all of the cases. The atrioventricular valves could be visualized 96% of the time, the aorta and pulmonary artery 95% of the time, and the inferior and superior vena cava 76% of the time.

Early fetal echocardiography was less effective in visualizing the aortic and ductal arches (45% of the time), branch pulmonary arteries (37% of the time), and pulmonary veins (19% of the time).

RENO, NEV. — Plasma volume expansion, once thought of as a possible treatment for severe preeclampsia, appears to confer no benefit even after a 1-year follow-up, results of a randomized controlled trial suggest.

The trial, which compared plasma volume expansion with temporizing treatment, showed that there were no differences in pregnancy outcomes and no differences in mental or psychomotor scores of the children at 1 year of age, reported Wessel Ganzevoort, M.D., and colleagues of Vrije University Medical Center, Amsterdam.

The study involved 216 women with severe preeclampsia; hemolysis, elevated liver enzymes, and low platelet count (HELLP) syndrome; or fetal growth restriction related to pregnancy-induced hypertension.

Gestational ages ranged between 24 and 33 weeks at the start of the study, the investigators wrote in a poster that was presented at the annual meeting of the Society for Maternal-Fetal Medicine.

Women randomized to the plasma volume expansion group received 500 mL of 6% hexaethyl starch daily. Patients in both groups received antihypertensive treatment and magnesium sulfate when appropriate, they said.

Of the 111 women in the treatment group, 89 (80%) delivered live infants. Similarly, of the 105 women in the control group, 91 (87%) delivered live infants. The difference between the two groups was not statistically significant.

At 1 year of age, children from both groups showed lower scores than would be expected in the normal population on the Bayley mental developmental index (MDI) and the Bayley psychomotor developmental index (PDI).

But a comparison between the two groups demonstrated no significant differences in the proportion of children with abnormal or suspect MDI or PDI scores.

RENO, NEV. — Plasma volume expansion, once thought of as a possible treatment for severe preeclampsia, appears to confer no benefit even after a 1-year follow-up, results of a randomized controlled trial suggest.

The trial, which compared plasma volume expansion with temporizing treatment, showed that there were no differences in pregnancy outcomes and no differences in mental or psychomotor scores of the children at 1 year of age, reported Wessel Ganzevoort, M.D., and colleagues of Vrije University Medical Center, Amsterdam.

The study involved 216 women with severe preeclampsia; hemolysis, elevated liver enzymes, and low platelet count (HELLP) syndrome; or fetal growth restriction related to pregnancy-induced hypertension.

Gestational ages ranged between 24 and 33 weeks at the start of the study, the investigators wrote in a poster that was presented at the annual meeting of the Society for Maternal-Fetal Medicine.

Women randomized to the plasma volume expansion group received 500 mL of 6% hexaethyl starch daily. Patients in both groups received antihypertensive treatment and magnesium sulfate when appropriate, they said.

Of the 111 women in the treatment group, 89 (80%) delivered live infants. Similarly, of the 105 women in the control group, 91 (87%) delivered live infants. The difference between the two groups was not statistically significant.

At 1 year of age, children from both groups showed lower scores than would be expected in the normal population on the Bayley mental developmental index (MDI) and the Bayley psychomotor developmental index (PDI).

But a comparison between the two groups demonstrated no significant differences in the proportion of children with abnormal or suspect MDI or PDI scores.

RENO, NEV. — Plasma volume expansion, once thought of as a possible treatment for severe preeclampsia, appears to confer no benefit even after a 1-year follow-up, results of a randomized controlled trial suggest.

The trial, which compared plasma volume expansion with temporizing treatment, showed that there were no differences in pregnancy outcomes and no differences in mental or psychomotor scores of the children at 1 year of age, reported Wessel Ganzevoort, M.D., and colleagues of Vrije University Medical Center, Amsterdam.

The study involved 216 women with severe preeclampsia; hemolysis, elevated liver enzymes, and low platelet count (HELLP) syndrome; or fetal growth restriction related to pregnancy-induced hypertension.

Gestational ages ranged between 24 and 33 weeks at the start of the study, the investigators wrote in a poster that was presented at the annual meeting of the Society for Maternal-Fetal Medicine.

Women randomized to the plasma volume expansion group received 500 mL of 6% hexaethyl starch daily. Patients in both groups received antihypertensive treatment and magnesium sulfate when appropriate, they said.

Of the 111 women in the treatment group, 89 (80%) delivered live infants. Similarly, of the 105 women in the control group, 91 (87%) delivered live infants. The difference between the two groups was not statistically significant.

At 1 year of age, children from both groups showed lower scores than would be expected in the normal population on the Bayley mental developmental index (MDI) and the Bayley psychomotor developmental index (PDI).

But a comparison between the two groups demonstrated no significant differences in the proportion of children with abnormal or suspect MDI or PDI scores.

HARROGATE, ENGLAND — A woman's vitamin D status in late pregnancy is predictive of her offspring's lumbar spine volumetric bone density at age 9, a prospective study has shown.

The findings add to the growing body of evidence confirming that a woman's diet while pregnant can influence her child's later bone mass, said Nicholas W. Harvey, B.Chir.

The results of the population-based investigation also point to the potential efficacy of preventive measures to protect children's bone health, Dr. Harvey said in a presentation at the annual conference of the National Osteoporosis Society. “Vitamin D supplementation in pregnant women who are deficient may optimize peak accrual of bone mineral in their offspring,” he stated.

The investigation included 210 offspring of mothers enrolled in a larger cohort study of maternal nutrition and fetal development conducted by Dr. Harvey and his colleagues at the MRC environmental epidemiology unit of the University of Southampton (England). The mothers completed a questionnaire regarding their diet and lifestyle beginning from early pregnancy.

Anthropometric measures were recorded, including mid-upper arm circumference, which is a potential indicator of maternal nutritional status. The mothers gave venous blood samples in late pregnancy for the measurement of 25-hydroxyvitamin D levels and other nutrients. Concentrated umbilical cord blood was collected at birth to measure calcium, albumin, and phosphate.

The investigators recorded the size and weight of the offspring at birth. When the children reached age 9 years, they underwent dual-energy x-ray absorptiometry (DXA) for bone mass measurement. Because bone mineral density measured by DXA represents the areal density (grams per square centimeter) rather than the volumetric density (grams per cubic centimeter) of bone, the investigators generated mathematical estimates of volumetric bone density from the DXA measurements of bone mineral content and bone area.

“When studying bone mineral density during growth, the differences [between volumetric and areal BMD] have to be taken into consideration,” Dr. Harvey explained. As bones grow, the volume increases at a faster rate than the area, so the areal bone density will increase even if the volumetric density remains stable, he said.

At 9 years, the boys in the study group (112) were significantly taller than the girls, and had higher age-adjusted lumbar spine bone mineral content and bone area but lower volumetric bone mineral density. After adjustment for child age and gender, maternal vitamin D was positively correlated with childhood volumetric BMD. “There was a threshold in the relationship, such that mothers in the lowest fifth of the [vitamin D] distribution had children with significantly lower volumetric bone mineral density at age 9 than those in the remaining four fifths,” Dr. Harvey said.

Maternal mid-upper arm circumference and vitamin D supplementation in late pregnancy both had significant positive associations with volumetric BMD, while social class, maternal smoking, and umbilical cord phosphate, calcium, and albumin levels did not.

Calcium from the cord blood was predictive of increased bone mass, but not volumetric BMD, Dr. Harvey noted.

In a multivariate model, both maternal mid-upper arm circumference and low serum vitamin D remained significant predictors of childhood volumetric BMD status, he said.

HARROGATE, ENGLAND — A woman's vitamin D status in late pregnancy is predictive of her offspring's lumbar spine volumetric bone density at age 9, a prospective study has shown.

The findings add to the growing body of evidence confirming that a woman's diet while pregnant can influence her child's later bone mass, said Nicholas W. Harvey, B.Chir.

The results of the population-based investigation also point to the potential efficacy of preventive measures to protect children's bone health, Dr. Harvey said in a presentation at the annual conference of the National Osteoporosis Society. “Vitamin D supplementation in pregnant women who are deficient may optimize peak accrual of bone mineral in their offspring,” he stated.

The investigation included 210 offspring of mothers enrolled in a larger cohort study of maternal nutrition and fetal development conducted by Dr. Harvey and his colleagues at the MRC environmental epidemiology unit of the University of Southampton (England). The mothers completed a questionnaire regarding their diet and lifestyle beginning from early pregnancy.

Anthropometric measures were recorded, including mid-upper arm circumference, which is a potential indicator of maternal nutritional status. The mothers gave venous blood samples in late pregnancy for the measurement of 25-hydroxyvitamin D levels and other nutrients. Concentrated umbilical cord blood was collected at birth to measure calcium, albumin, and phosphate.

The investigators recorded the size and weight of the offspring at birth. When the children reached age 9 years, they underwent dual-energy x-ray absorptiometry (DXA) for bone mass measurement. Because bone mineral density measured by DXA represents the areal density (grams per square centimeter) rather than the volumetric density (grams per cubic centimeter) of bone, the investigators generated mathematical estimates of volumetric bone density from the DXA measurements of bone mineral content and bone area.

“When studying bone mineral density during growth, the differences [between volumetric and areal BMD] have to be taken into consideration,” Dr. Harvey explained. As bones grow, the volume increases at a faster rate than the area, so the areal bone density will increase even if the volumetric density remains stable, he said.

At 9 years, the boys in the study group (112) were significantly taller than the girls, and had higher age-adjusted lumbar spine bone mineral content and bone area but lower volumetric bone mineral density. After adjustment for child age and gender, maternal vitamin D was positively correlated with childhood volumetric BMD. “There was a threshold in the relationship, such that mothers in the lowest fifth of the [vitamin D] distribution had children with significantly lower volumetric bone mineral density at age 9 than those in the remaining four fifths,” Dr. Harvey said.

Maternal mid-upper arm circumference and vitamin D supplementation in late pregnancy both had significant positive associations with volumetric BMD, while social class, maternal smoking, and umbilical cord phosphate, calcium, and albumin levels did not.

Calcium from the cord blood was predictive of increased bone mass, but not volumetric BMD, Dr. Harvey noted.

In a multivariate model, both maternal mid-upper arm circumference and low serum vitamin D remained significant predictors of childhood volumetric BMD status, he said.

HARROGATE, ENGLAND — A woman's vitamin D status in late pregnancy is predictive of her offspring's lumbar spine volumetric bone density at age 9, a prospective study has shown.

The findings add to the growing body of evidence confirming that a woman's diet while pregnant can influence her child's later bone mass, said Nicholas W. Harvey, B.Chir.

The results of the population-based investigation also point to the potential efficacy of preventive measures to protect children's bone health, Dr. Harvey said in a presentation at the annual conference of the National Osteoporosis Society. “Vitamin D supplementation in pregnant women who are deficient may optimize peak accrual of bone mineral in their offspring,” he stated.

The investigation included 210 offspring of mothers enrolled in a larger cohort study of maternal nutrition and fetal development conducted by Dr. Harvey and his colleagues at the MRC environmental epidemiology unit of the University of Southampton (England). The mothers completed a questionnaire regarding their diet and lifestyle beginning from early pregnancy.

Anthropometric measures were recorded, including mid-upper arm circumference, which is a potential indicator of maternal nutritional status. The mothers gave venous blood samples in late pregnancy for the measurement of 25-hydroxyvitamin D levels and other nutrients. Concentrated umbilical cord blood was collected at birth to measure calcium, albumin, and phosphate.

The investigators recorded the size and weight of the offspring at birth. When the children reached age 9 years, they underwent dual-energy x-ray absorptiometry (DXA) for bone mass measurement. Because bone mineral density measured by DXA represents the areal density (grams per square centimeter) rather than the volumetric density (grams per cubic centimeter) of bone, the investigators generated mathematical estimates of volumetric bone density from the DXA measurements of bone mineral content and bone area.

“When studying bone mineral density during growth, the differences [between volumetric and areal BMD] have to be taken into consideration,” Dr. Harvey explained. As bones grow, the volume increases at a faster rate than the area, so the areal bone density will increase even if the volumetric density remains stable, he said.

At 9 years, the boys in the study group (112) were significantly taller than the girls, and had higher age-adjusted lumbar spine bone mineral content and bone area but lower volumetric bone mineral density. After adjustment for child age and gender, maternal vitamin D was positively correlated with childhood volumetric BMD. “There was a threshold in the relationship, such that mothers in the lowest fifth of the [vitamin D] distribution had children with significantly lower volumetric bone mineral density at age 9 than those in the remaining four fifths,” Dr. Harvey said.

Maternal mid-upper arm circumference and vitamin D supplementation in late pregnancy both had significant positive associations with volumetric BMD, while social class, maternal smoking, and umbilical cord phosphate, calcium, and albumin levels did not.

Calcium from the cord blood was predictive of increased bone mass, but not volumetric BMD, Dr. Harvey noted.

In a multivariate model, both maternal mid-upper arm circumference and low serum vitamin D remained significant predictors of childhood volumetric BMD status, he said.