In August 1999 the government announced that HIV testing would be offered and recommended routinely to pregnant women in the UK. It was planned that all health authorities (as they were then called) should be offering the test on this basis by the end of 2000.

No woman who refuses HIV testing during pregnancy should receive different antenatal services from those who are tested. Conversely, the standard antenatal care of a pregnant HIV positive woman should not be any different from any other pregnant woman. The number of visits to see her midwife should not necessarily be different. She will probably see a specialist obstetrician more frequently to discuss her mode of delivery – elective caesarean or vaginal. In some hospitals a “minimal blood loss” caesarean is offered. Her HIV care will normally be undertaken by specialist HIV physicians. Standard infection control and universal precautions are sufficient to protect health care providers from the risk of infection.

Reasons why midwives should be informed of HIV status

This is because appropriate care can be offered and certain procedures will be avoided.

A number of procedures may increase the risk of transmission, including:

• Amniocentesis.
• Use of foetal electrodes.
• Prolonged labour (4 hours +).
• Vaginal delivery.
• Breastfeeding.

A midwife will only be able to provide support for adherence to antiretroviral therapy during and after labour if she is aware of the mother's HIV status. Adherence can be particularly problematic at this time if other family members who are present at the birth are not aware of the mother's HIV status.

Adherence may also be complicated by extreme nausea experienced during labour. Once again, an informed midwife may be able to help.

Finally, mothers may need active assistance to bottle feed after delivery, both in practical terms and in terms of persuading other family members that this is good idea whilst preserving the mother's confidentiality.

HIV testing pregnant women in UK: still failing to meet targets

The latest figures from the Health Protection Agency show that during the first six months of 2002, England as a whole was still failing to reach government set targets to diagnose 80% of HIV infections in pregnant women prior to delivery.

National Unlinked Anonymous (UA) monitoring of the prevalence of HIV infection in pregnant women, by testing for maternal antibody in infant dried blood spots, began in the UK in 1988. Since 1992, the survey has covered approximately 70% of UK births. The results of UA monitoring are aligned with reports of HIV infected pregnant women made through the Royal College of Obstetricians and Gynaecologists (RCOG) to the National Study of HIV in Pregnancy and Childhood (NSHPC). This provides the best estimates of the proportions of HIV-positive pregnant women who have had their infection diagnosed prior to pregnancy or during current antenatal care.

In 1994 it became apparent that interventions implemented during pregnancy and in the perinatal period can reduce the risk of transmission of HIV from mother-to-child from one in four to less than one in 50, and the uptake of such interventions by diagnosed HIV-positive pregnant women in the UK is high. There are also direct benefits to the woman’s own health from having her HIV infection diagnosed earlier than might otherwise have happened.

Outside London, the prevalence of HIV infection among women giving birth has remained consistently low. Prevalence in the rest of England for the first half of 2002 is estimated to be 3.5/10,000, and in Scotland 4.8/10,000 women.

In 1999, national targets that offer and recommend HIV testing to all pregnant women throughout England were established by the Department of Health. It was intended that by increasing the uptake of antenatal HIV testing to 90%, and by increasing the proportion of HIV infections diagnosed prior to delivery to 80%, an 80% reduction in the proportion of children acquiring HIV infection from their mothers should be achieved by December 2002.

In London, up to the end of June 2002, 135 diagnosed maternal infections had been reported to the NSHPC by the end of the year, giving a minimum overall detection rate of 66% (135/205) and an antenatal detection rate of 53%. The data for the rest of England in the first half of 2002 show an overall detection rate of 81% (57/70), meeting the Department of Health target for the first time.

For Scotland, data show that at least 67% (8/12) of maternal HIV infections were diagnosed before delivery in the first half of 2002. The equivalent overall detection rates in the same period of 2001 for London, the rest of England, and Scotland were 73%, 33% and 75% respectively.

These interim half-year data are, however, subject to reporting delay and the estimated proportion of infections diagnosed is expected to rise further as late reports are received. In 2001, it can be estimated that there were about 560 births to HIV-positive women in the UK. Assuming a transmission rate of about 25% for undiagnosed women, and 2% for diagnosed women (allowing for late diagnoses and a small proportion of diagnosed women declining interventions, about 50 infants would have acquired HIV infection from their mothers.

Continued improvements in detection rates before delivery have resulted in a decreasing proportion of HIV-positive women passing HIV infection on to their child and, despite an increase in the number of HIV-positive women giving birth, the number of maternally- acquired HIV infections was probably similar in 2001 to 2000.

Increase in prenatal testing rate in USA 1998-2001

The availability of increasingly effective antiretroviral drugs for prevention of transmission and maternal treatment has resulted in greater emphasis on prenatal HIV testing. In 2000, preliminary unpublished data from the Centers for Disease Control (CDC) in Atlanta indicated that 766 (93%) of 824 HIV-positive women in 25 States knew their status before delivery.

Each year an estimated 280-370 perinatal HIV transmissions occur in the USA.

In the USA and Canada three different prenatal HIV testing approaches have been implemented. The CDC reviewed prenatal HIV antibody testing rates to assess their effectiveness. Under the opt-in approach, women typically are provided pre-HIV test counselling and must consent specifically to an HIV test. Under the opt-out approach, women are notified that an HIV test will be included in a battery of prenatal tests and procedures and that they may refuse testing. Under mandatory newborn HIV testing, newborns are tested for HIV, with or without the mother’s consent, if the mother’s HIV status is unknown at delivery.

Medical record data suggest that the opt in voluntary testing approach is associated with lower testing rates than either the opt-out or the mandatory newborn HIV testing approach.

Three different approaches were used to estimate prenatal testing rates among all women who delivered, regardless of whether they received prenatal care; firstly, eight US areas that participated during 1998-1999 in CDC’s Active Bacterial Core Surveillance/Emerging Infections Program (ABC) Network assessed HIV testing during prenatal care and Secondly, public health investigators in each of the five Canadian provinces tallied the number of HIV tests among pregnant women that were submitted to provincial laboratories and divided the total by an estimate of all live and stillborn births in each province during the same year. Finally CDC analysed weighted data collected in 1999 by interviewers in nine states for CDC’s Pregnancy Risk Assessment Monitoring System (PRAMS) (an ongoing, population based survey conducted in 32 states and New York City among women who have given birth in the preceding 2-6 months) who had asked women if they had been tested for HIV during pregnancy.

HIV testing rates varied depending upon which approach was used. Rates for states using the opt in approach to prenatal HIV testing included in the ABC Network ranged from 25% to 69%, testing rates in Canada ranged from 54% to 83%, and rates derived from the PRAMS data ranged from 61% to 81%.

Two US states Arkansas and Tennessee and two Canadian provinces (Alberta, and Newfoundland and Labrador) reported using an opt-out prenatal HIV testing policy. ABC Network data indicated that Tennessee had a testing rate of 85%. Canada’s population-based data indicated a 98% testing rate in Alberta and a 94% testing rate in Newfoundland and Labrador. PRAMS interview data indicated a 71% testing rate in Arkansas, compared with a 57% testing rate in early 1997 before the law was implemented. Two states New York and Connecticut require HIV testing of newborns whose mothers were not tested during pregnancy.

Network review of medical records in the seven counties in the Rochester area indicated that the proportion of pregnant women who received a prenatal HIV test increased from 52% of 438 charts during January 1998-July 1999 to 83% of 112 charts during August-December 1999 after New York required that newborn HIV testing results be made available within 48 hours of specimen collection. PRAMS data for the state of New York in 1999 indicated that the proportion of women statewide who reported having received an HIV test during pregnancy increased from 69% of 758 women during January-July to 93% of 502 during August-December.

Among the three prenatal HIV testing approaches assessed, opt out voluntary testing and the mandatory testing of newborns appear to be associated with the highest testing rates. CDC is working with states with high HIV prevalence rates among women of childbearing age and high numbers of paediatric AIDS cases to ensure standardized monitoring of prenatal testing rates. The data suggest that jurisdictions that use an opt-in approach and that have low prenatal HIV testing rates should re-evaluate their approach.

References

HIV infection in women giving birth in the United Kingdom – trends in prevalence and proportions diagnosed to the end of June 2002. CDR Weekly, 17 March 2003.

Roome A et al. HIV testing among pregnant women-United States and Canada 1998-2001. Morbidity and Mortality Weekly Report 51: 1013-1016, 2002.