1007
function in the process leading to the initiation of parturition in human beings, possibly by shifting the balance of the production of eicosanoids in favour of those derived from n-3 rather than n-6 fatty acids.5,8 Biochemical and physiological studies are needed to clarify how and at what concentrations dietary n-3 fatty acids could affect formation of eicosanoids in human pregnancy and to compare the biological properties in the uteroplacental system of those derived from n-3 and n-6 fatty acids. Our findings suggest an easy and cheap intervention to prevent preterm delivery. However, prolongation of gestation by fish oil may not result in reduced frequencies of the complications associated with preterm delivery;l whether it will or not depends on the extent to which preterm delivery is a necessary link in the causal chain leading to the complications. Furthermore, fish-oil supplementation could lead to more post-term deliveries, which are also associated with complications?°3 A carefully monitored, controlled trial of fish oil to women at high risk of preterm delivery seems justified, and a multicentre trial is in progress.1-0 We thank Prof Jem Olsen and Prof Thorkild I. A. Sorensen for valuable discussions; Mr Jacob Hjort for his help in making the data ready for analysis; Lube Ltd for the oil capsules; and the pharmacy at Municipal Hospital of Aarhus for packing the capsules. This study was supported by the Danish Medical Research Council (J No 12-9052) and 12-9144), Sygekassemes Helsefond, Weiman’s Legat and
Michaelsen Fonden.
REFERENCES 1. Keirse MJNC. Preterm delivery. In: Chalmers I, Enkin M, Keirse MJNC, eds. Effective care in pregnancy and childbirth. Oxford: Oxford University Press, 1989; 1270-92. 2. Bakketeig LS, Bergsjø P. Post-term pregnancy: magnitude of the problem. In: Chalmers I, Enkin M, Keirse MJNC, eds. Effective care in pregnancy and childbirth. Oxford: Oxford University Press, 1989: 765-75.
3. Yudkin PL, Wood L, Redman CWG. Risk of unexplained stillbirth at different gestational ages. Lancet 1987; i: 1192-94. 4. Kramer MS. Intrauterine growth and gestational duration determinants. Pediatrics 1987; 80: 502-11. 5. Olsen SF, Hansen HS, Sørensen TIA, et al. Intake of marine fat, rich in (n-3)-PUFA, may increase birthweight by prolonging gestation. Lancet 1986; ii: 367-69. 6. Olsen SF, Joensen HD. High livebom birthweights in the Faroes: a comparison between birthweights in the Faroes and in Denmark. J Epidemiol Community Health 1985; 39: 27-32. 7. Olsen SF, Hansen HS. Marine fat, birthweight, and gestational age: a case report. Ag Act 1987; 22: 373-74. 8. Hansen HS, Olsen SF. Dietary (n-3)-fatty acids, prostaglandins, and prolonged gestation in humans. Prog Clin Biol Res 1988; 282: 305-17. 9. Leaf A, Weber PC. Cardiovascular effects of n-3 fatty acids. N Engl J Med 1988; 318: 549-57. 10. Oates JA, FitzGerald GA, Branch AR, et al. Clinical implications of prostaglandin and thromboxane A2 formation. N Engl J Med 1988; 319: 689-98. 11. Eriksen PS, Secher NJ, Weis-Bentzon M. Normal growth of the fetal biparietal diameter and the abdominal diameter in a longitudinal study. Acta Obstet Gynecol Scand 1985; 64: 65-70. 12. von Schacky C. Prophylaxis of atherosclerosis with marine omega-3 fatty acids. Ann Intern Med 1987; 107: 890-99. 13. Radack KL. Omega-3 fatty adds. Ann Intern Med 1988; 109: 81-82. 14. Haraldsdottir J, Holm L, Jensen JH, Møller A. Dietary habits in Denmark 1985, vol 2. Copenhagen: National Food Agency of Denmark. 1987. 15. Olsen SF, Hansen HS, Sommer S, et al. Gestational age in relation to marine n-3 fatty adds in maternal erythrocytes: a study of women in the Faroe Islands and Denmark: Am J Obstet Gynecol 1991; 164: 1203-09. 16. The People’s League of Health. Nutrition of expectant and nursing mothers. Lancet 1942; ii: 10-12. 17. Olsen SF, Secher NJ. A possible preventive effect of low-dose fish oil on early delivery and preeclampsia: indications from a 50 year old controlled trial. Br J Nutr 1990; 64: 599-609. 18. Secher NJ, Olsen SF. Fish-oil and pre-eclampsia. Br J Obstet Gynaecol 1990; 97: 1077-79. 19. Pipkin FB. Fish-oil and pre-eclampsia. Br J Obstet Gynaecol 1991; 98: 737. 20. Secher NJ, Olsen SF, Sørensen JD. Fish-oil and pre-eclampsia. Br J Obstet Gynaecol 1991; 98: 738-40.
Risk factors for mother-to-child transmission of HIV-1 EUROPEAN COLLABORATIVE STUDY
Children born to women known to be infected with human immunodeficiency virus type 1 (HIV-1) before delivery were followed prospectively from birth in nineteen European centres. This analysis, encompassing the period end-December, 1984, to beginning-August, 1991, focuses on risk factors for mother-to-child transmission of HIV-1 infection. Rate of vertical transmission, based on 721 children born to 701 mothers more than 18 months before the time of analysis, was 14·4% (95% Cl 12·0-17·1%). Transmission was associated with maternal p24-antigenaemia and a CD4 count of less than 700/µl. In a multivariate analysis, odds ratios of transmission were: 2·25 (95% Cl 0·97-5·23) in breastfed children vs never-breastfed children; 3·80 (1·62-8·91) in children born before 34 weeks’ gestation; and 0·56 (030-1·04) in children delivered by caesarean section. Transmission was higher with vaginal deliveries in which episiotomy,
scalp electrodes, forceps, or vacuum extractors were used, but only in centres where these procedures were not
routine.
On the basis of these results, HIV-infected women contemplating pregnancy should be counselled according to their immunological findings and, if they have p24-antigenaemia or a low CD4 count, warned of an increased risk of viral transmission. Caesarean deliveries may have a protective effect, although it is premature to recommend routine operative delivery. The mechanism for the higher infection rate in children born before 34 weeks’ gestation is unclear, but could reflect inadequate passive or active immunity at that age, combined with substantial transmission during labour or delivery. The balance of evidence suggests that mothers with established infection can transmit HIV infection through breastmilk, although the relative importance of this route remains to be defined. Prepared by M. L Newell, PhD, Mr D. Dunn, MSc, Prof C. S. Peckham, MD, A E. Ades, PhD (Coordinating Centre, London), with Prof G. Pardi, MD, and A. E. Semprini, MD (Ospedale San Paolo, Milan, Italy). Correspondence to Prof C. S. Peckham, Department of Epidemiology and Biostatistics, Institute of Child Health, 30 Guilford Street, London WC1N 1 EH, UK Collaborators are listed at the end of the article.
1008
Introduction
TABLE I-RATE OFTRANSMISSION BY SELECTED RISK FACTORS
Reported rates of transmission of human immunodeficiency virus type 1 (HIV-1) infection from mother to child range from 7 to 39%." It is unclear to what extent this variation is due to methodological differences or to different distributions of risk factors in the populations studied. There is some evidence that stage of maternal HIV infection, as indicated by clinical and immunological status, is associated with increased transmission.2,4,9,lo Some researchers" 12 have suggested that most HIV infection is acquired around the time of delivery. Mode of delivery has not been shown to play a part in HIV transmission,1,13 but there is no information about the role of obstetric procedures or events in the intrapartum period. Although breastfeeding during primary infection poses a risk, 14 the additional risk of breastfeeding in established maternal infection is poorly
quantified. 13.15 In this report we analyse data collected by participants in the European Collaborative Study from end-December, 1984, to beginning-August, 1991.5 Our aim was to investigate the role of factors that may be associated with vertical transmission of HIV-1 infection-for example, clinical and immunological status of the mother during pregnancy, length of gestation, mode of delivery, and
breastfeeding. Methods Mothers and infants In ten participating centres infants born to women known to be HIV-1seropositive at or before delivery were enrolled at birth and followed-up regularly according to a standard protocol.5,16 At the time of birth, we obtained information about maternal clinical status, mode of delivery, length of gestation, and birthweight for these children, and we sought further information about maternal immunological variables and obstetric procedures from the obstetricians in these centres. In 1991, obstetricians in an additional nine centres provided similar information which they had already collected systematically on all HIV-1 seropositive mothers who were known to be infected before delivery, and on the subsequent infection status of the children. All infants bom 18 months or more before the date of analysis are included. HIV-1 infection in the child was defmed as: AIDS;I’ HIV-related death; persistence of antibody beyond 18 months; or detection of virus or p24-antigenaemia in at least two samples. 4 children who became seronegative but who were repeatedly virus positive were excluded from the analysis. The remaining seronegative children are judged to be uninfected. All laboratory assessments were carried out locally; lymphocyte subsets were determined by immunofluorescence with cytofluorometric readout.18 The presence of p24-antigenaemia was determined by commercially available enzyme-linked immunosorbent assays (Abbott, Du Pont, Pasteur, Organon) or, in one centre, by an in-house system.19 The result closest to delivery was used in the analysis provided the maternal sample was taken during pregnancy or less than 1 month after delivery. 86% of the p24 antigen tests were done during the third trimester or at the time of delivery, as were 76% of the lymphocyte subset estimations. CD4 counts were divided into three groups of approximately equal numbers by means of cut-off values of 400 and 700 per ul. The CD4/CD8 ratio was divided similarly, the values being 06 and 0-9. This approach was found to be more appropriate than conventional categories such as CD4 counts of less than 200/til since only 9 women in this largely symptom-free cohort fell into this category. Women were included as HIV-symptomatic (Centers for Disease Control [CDC] stage III or IVA) if any HIV-related symptom or sign was recorded by the clinician on two or more occasions up to 3 months post partum. AIDS was defmed according to the CDC classification 2’ and a mother was regarded as having AIDS if the diagnosis was made no later than 1 year post partum, to allow for increased infectiousness before onset of AIDS.21
Numbers do not always add up to total because of missing data *Significance assessed by test for heterogeneity, tSigmficance assessed by test for < 005; p < 0.01 trend, tp
function in the process leading to the initiation of parturition in human beings, possibly by shifting the balance of the production of eicosanoids in favour of those derived from n-3 rather than n-6 fatty acids.5,8 Biochemical and physiological studies are needed to clarify how and at what concentrations dietary n-3 fatty acids could affect formation of eicosanoids in human pregnancy and to compare the biological properties in the uteroplacental system of those derived from n-3 and n-6 fatty acids. Our findings suggest an easy and cheap intervention to prevent preterm delivery. However, prolongation of gestation by fish oil may not result in reduced frequencies of the complications associated with preterm delivery;l whether it will or not depends on the extent to which preterm delivery is a necessary link in the causal chain leading to the complications. Furthermore, fish-oil supplementation could lead to more post-term deliveries, which are also associated with complications?°3 A carefully monitored, controlled trial of fish oil to women at high risk of preterm delivery seems justified, and a multicentre trial is in progress.1-0 We thank Prof Jem Olsen and Prof Thorkild I. A. Sorensen for valuable discussions; Mr Jacob Hjort for his help in making the data ready for analysis; Lube Ltd for the oil capsules; and the pharmacy at Municipal Hospital of Aarhus for packing the capsules. This study was supported by the Danish Medical Research Council (J No 12-9052) and 12-9144), Sygekassemes Helsefond, Weiman’s Legat and
Michaelsen Fonden.
REFERENCES 1. Keirse MJNC. Preterm delivery. In: Chalmers I, Enkin M, Keirse MJNC, eds. Effective care in pregnancy and childbirth. Oxford: Oxford University Press, 1989; 1270-92. 2. Bakketeig LS, Bergsjø P. Post-term pregnancy: magnitude of the problem. In: Chalmers I, Enkin M, Keirse MJNC, eds. Effective care in pregnancy and childbirth. Oxford: Oxford University Press, 1989: 765-75.
3. Yudkin PL, Wood L, Redman CWG. Risk of unexplained stillbirth at different gestational ages. Lancet 1987; i: 1192-94. 4. Kramer MS. Intrauterine growth and gestational duration determinants. Pediatrics 1987; 80: 502-11. 5. Olsen SF, Hansen HS, Sørensen TIA, et al. Intake of marine fat, rich in (n-3)-PUFA, may increase birthweight by prolonging gestation. Lancet 1986; ii: 367-69. 6. Olsen SF, Joensen HD. High livebom birthweights in the Faroes: a comparison between birthweights in the Faroes and in Denmark. J Epidemiol Community Health 1985; 39: 27-32. 7. Olsen SF, Hansen HS. Marine fat, birthweight, and gestational age: a case report. Ag Act 1987; 22: 373-74. 8. Hansen HS, Olsen SF. Dietary (n-3)-fatty acids, prostaglandins, and prolonged gestation in humans. Prog Clin Biol Res 1988; 282: 305-17. 9. Leaf A, Weber PC. Cardiovascular effects of n-3 fatty acids. N Engl J Med 1988; 318: 549-57. 10. Oates JA, FitzGerald GA, Branch AR, et al. Clinical implications of prostaglandin and thromboxane A2 formation. N Engl J Med 1988; 319: 689-98. 11. Eriksen PS, Secher NJ, Weis-Bentzon M. Normal growth of the fetal biparietal diameter and the abdominal diameter in a longitudinal study. Acta Obstet Gynecol Scand 1985; 64: 65-70. 12. von Schacky C. Prophylaxis of atherosclerosis with marine omega-3 fatty acids. Ann Intern Med 1987; 107: 890-99. 13. Radack KL. Omega-3 fatty adds. Ann Intern Med 1988; 109: 81-82. 14. Haraldsdottir J, Holm L, Jensen JH, Møller A. Dietary habits in Denmark 1985, vol 2. Copenhagen: National Food Agency of Denmark. 1987. 15. Olsen SF, Hansen HS, Sommer S, et al. Gestational age in relation to marine n-3 fatty adds in maternal erythrocytes: a study of women in the Faroe Islands and Denmark: Am J Obstet Gynecol 1991; 164: 1203-09. 16. The People’s League of Health. Nutrition of expectant and nursing mothers. Lancet 1942; ii: 10-12. 17. Olsen SF, Secher NJ. A possible preventive effect of low-dose fish oil on early delivery and preeclampsia: indications from a 50 year old controlled trial. Br J Nutr 1990; 64: 599-609. 18. Secher NJ, Olsen SF. Fish-oil and pre-eclampsia. Br J Obstet Gynaecol 1990; 97: 1077-79. 19. Pipkin FB. Fish-oil and pre-eclampsia. Br J Obstet Gynaecol 1991; 98: 737. 20. Secher NJ, Olsen SF, Sørensen JD. Fish-oil and pre-eclampsia. Br J Obstet Gynaecol 1991; 98: 738-40.
Risk factors for mother-to-child transmission of HIV-1 EUROPEAN COLLABORATIVE STUDY
Children born to women known to be infected with human immunodeficiency virus type 1 (HIV-1) before delivery were followed prospectively from birth in nineteen European centres. This analysis, encompassing the period end-December, 1984, to beginning-August, 1991, focuses on risk factors for mother-to-child transmission of HIV-1 infection. Rate of vertical transmission, based on 721 children born to 701 mothers more than 18 months before the time of analysis, was 14·4% (95% Cl 12·0-17·1%). Transmission was associated with maternal p24-antigenaemia and a CD4 count of less than 700/µl. In a multivariate analysis, odds ratios of transmission were: 2·25 (95% Cl 0·97-5·23) in breastfed children vs never-breastfed children; 3·80 (1·62-8·91) in children born before 34 weeks’ gestation; and 0·56 (030-1·04) in children delivered by caesarean section. Transmission was higher with vaginal deliveries in which episiotomy,
scalp electrodes, forceps, or vacuum extractors were used, but only in centres where these procedures were not
routine.
On the basis of these results, HIV-infected women contemplating pregnancy should be counselled according to their immunological findings and, if they have p24-antigenaemia or a low CD4 count, warned of an increased risk of viral transmission. Caesarean deliveries may have a protective effect, although it is premature to recommend routine operative delivery. The mechanism for the higher infection rate in children born before 34 weeks’ gestation is unclear, but could reflect inadequate passive or active immunity at that age, combined with substantial transmission during labour or delivery. The balance of evidence suggests that mothers with established infection can transmit HIV infection through breastmilk, although the relative importance of this route remains to be defined. Prepared by M. L Newell, PhD, Mr D. Dunn, MSc, Prof C. S. Peckham, MD, A E. Ades, PhD (Coordinating Centre, London), with Prof G. Pardi, MD, and A. E. Semprini, MD (Ospedale San Paolo, Milan, Italy). Correspondence to Prof C. S. Peckham, Department of Epidemiology and Biostatistics, Institute of Child Health, 30 Guilford Street, London WC1N 1 EH, UK Collaborators are listed at the end of the article.
1008
Introduction
TABLE I-RATE OFTRANSMISSION BY SELECTED RISK FACTORS
Reported rates of transmission of human immunodeficiency virus type 1 (HIV-1) infection from mother to child range from 7 to 39%." It is unclear to what extent this variation is due to methodological differences or to different distributions of risk factors in the populations studied. There is some evidence that stage of maternal HIV infection, as indicated by clinical and immunological status, is associated with increased transmission.2,4,9,lo Some researchers" 12 have suggested that most HIV infection is acquired around the time of delivery. Mode of delivery has not been shown to play a part in HIV transmission,1,13 but there is no information about the role of obstetric procedures or events in the intrapartum period. Although breastfeeding during primary infection poses a risk, 14 the additional risk of breastfeeding in established maternal infection is poorly
quantified. 13.15 In this report we analyse data collected by participants in the European Collaborative Study from end-December, 1984, to beginning-August, 1991.5 Our aim was to investigate the role of factors that may be associated with vertical transmission of HIV-1 infection-for example, clinical and immunological status of the mother during pregnancy, length of gestation, mode of delivery, and
breastfeeding. Methods Mothers and infants In ten participating centres infants born to women known to be HIV-1seropositive at or before delivery were enrolled at birth and followed-up regularly according to a standard protocol.5,16 At the time of birth, we obtained information about maternal clinical status, mode of delivery, length of gestation, and birthweight for these children, and we sought further information about maternal immunological variables and obstetric procedures from the obstetricians in these centres. In 1991, obstetricians in an additional nine centres provided similar information which they had already collected systematically on all HIV-1 seropositive mothers who were known to be infected before delivery, and on the subsequent infection status of the children. All infants bom 18 months or more before the date of analysis are included. HIV-1 infection in the child was defmed as: AIDS;I’ HIV-related death; persistence of antibody beyond 18 months; or detection of virus or p24-antigenaemia in at least two samples. 4 children who became seronegative but who were repeatedly virus positive were excluded from the analysis. The remaining seronegative children are judged to be uninfected. All laboratory assessments were carried out locally; lymphocyte subsets were determined by immunofluorescence with cytofluorometric readout.18 The presence of p24-antigenaemia was determined by commercially available enzyme-linked immunosorbent assays (Abbott, Du Pont, Pasteur, Organon) or, in one centre, by an in-house system.19 The result closest to delivery was used in the analysis provided the maternal sample was taken during pregnancy or less than 1 month after delivery. 86% of the p24 antigen tests were done during the third trimester or at the time of delivery, as were 76% of the lymphocyte subset estimations. CD4 counts were divided into three groups of approximately equal numbers by means of cut-off values of 400 and 700 per ul. The CD4/CD8 ratio was divided similarly, the values being 06 and 0-9. This approach was found to be more appropriate than conventional categories such as CD4 counts of less than 200/til since only 9 women in this largely symptom-free cohort fell into this category. Women were included as HIV-symptomatic (Centers for Disease Control [CDC] stage III or IVA) if any HIV-related symptom or sign was recorded by the clinician on two or more occasions up to 3 months post partum. AIDS was defmed according to the CDC classification 2’ and a mother was regarded as having AIDS if the diagnosis was made no later than 1 year post partum, to allow for increased infectiousness before onset of AIDS.21
Numbers do not always add up to total because of missing data *Significance assessed by test for heterogeneity, tSigmficance assessed by test for < 005; p < 0.01 trend, tp
