University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Home > Global Health Literature Digest > Rapid HIV Testing
Rapid HIV testing and prevention of perinatal HIV transmission in high-risk maternity hospitals in St. Petersburg, Russia
Global Health Sciences Literature Digest
Published May 6, 2008
Journal Article

Kissin DM, Akatova N, Rakhmanova AG, Vinogradova EN, Voronin EE, Jamieson DJ, et al. Rapid HIV testing and prevention of perinatal HIV transmission in high-risk maternity hospitals in St. Petersburg, Russia. Am J Obstet Gynecol 2008 Feb;198(2):183.e1-7.

Objective

To evaluate the performance of a hospital-based, public health HIV rapid testing (RT) program to detect unknown HIV infections among delivering women, to provide maternal and infant antiretroviral prophylaxis, and to prevent perinatal HIV transmission among infants.

Study Design

A performance evaluation of a public health program in implementation year 1.

Setting

The program was implemented at two of the 17 maternity hospitals in St. Petersburg, Russia. In addition to serving women living in their areas, these two maternity hospitals provide care for women considered high risk because they lack a prenatal care record or because they have inadequate HIV testing documentation in their record. In case of imminent delivery, women can be admitted at any of the 17 maternity hospitals.

Participants

During the study period, 8,790 women delivered in the two high-risk maternity hospitals; 4,162 (47.4%) were eligible for RT. Of those women, 4,117 were eligible because they had no HIV tests during pregnancy (1402 women; 33.7%) or had incomplete HIV testing (2715 women; 65.2%); 45 women who were eligible for other reasons were excluded. Adequacy of testing was determined by a review of prenatal care records submitted at admission. Eligible women were tested even when they were in late labor or soon after delivery. Although most women tested with RT were in active labor, women not in active labor were also eligible for testing if, in the physician's judgment, standard EIA results would not be available before delivery. Trained hospital personnel conducted pretest counseling using an opt-out approach.

Intervention

From April 13, 2004 to April 13, 2005, pregnant women at two high-risk maternity hospitals with no or incomplete HIV testing results (negative tests at <34 weeks, none thereafter) were offered point-of-care RT, with antiretroviral prophylaxis for women with positive RT results and their infants.

Primary Outcomes

Outcomes were the percentage of women accepting RT, HIV seroprevalence, timing of the receipt of HIV test results, concordance between RT and conventional EIA, and percentages of women and exposed & infected infants receiving antiretroviral prophylaxis.

Results

A total of 4,117 women were eligible for the study, and 3,671 (89.2%) underwent RT: 98.1% of women without and 84.6% of women with prior negative test results. Of these women, 90.4% received results prior to delivery: 81.3% of women with no HIV testing during pregnancy and 95.9% of women with incomplete HIV testing. Among all women who underwent RT, HIV prevalence was 2.7% (100/3,671); among previously untested women, prevalence was 6.5% (95% CI, 5.3%-8.0%). The incidence of HIV seroconversion among women with previous negative test results during pregnancy was 0.4% (95% CI, 0.2%-0.8). Adjustment for confounders found that the only overall predictor of receiving RT results post-delivery was admission within 60 minutes of delivery, but among women with incomplete HIV testing, not being an injection drug user and being admitted during the day shift also predicted post-delivery results. Among women with positive RT results (100), 76.0% received antiretroviral prophylaxis. Among their HIV-exposed infants, 97.9% (92/94) received prophylaxis; 61.7% (58/94) had available follow-up data, and 8.6% (5/58, 95% CI, 2.9%-19.0%) met criteria for definitive or presumptive HIV infection. Four of the 5 mothers and all of the 5 infants received antiretroviral prophylaxis, and none of the infants was breastfed at the hospital.

Concordance between RT and EIA results was 99.7% (3,574/3,586 specimens). Confirmatory tests were not performed initially on RT-positive/EIA-negative specimens. However, of the 91 Western blot tests that were performed on RT specimens with positive results, 88 were positive (96.7%), and three were indeterminate (3.3%). Relatively long turnaround times for Western blot results (average, 10.5 days) made final HIV status unavailable before discharge from the hospital for all but 9% of women (9/100) who had positive RT results. Among women who were eligible for RT, median turnaround times were 15 minutes between admission and blood collection, 20 minutes between blood collection and availability of RT results, and 3.7 hours between availability of these results and delivery.

Conclusions

The authors conclude that this RT program provided timely detection of HIV infection among women in or entering labor and facilitated appropriate delivery of antiretroviral prophylaxis. This result thus demonstrated that a model RT program can be adapted to local conditions and successfully implemented. The authors think that a point-of-care testing model was critical to program success because it allowed maternity hospital personnel to receive results promptly and to provide perinatal prophylaxis based on test results rather than on risk assessment. The perinatal HIV surveillance system made it possible to identify areas for program improvement. Late admission to the hospital (within minutes before delivery) limited the availability of RT results before delivery and the provision of maternal antiretroviral prophylaxis. Very late admission may reflect the same poor health care-seeking behavior that leads to failure to receive prenatal care. Although early presentation to the hospital is essential, hospital personnel must be aware that the opportunity for learning HIV status before delivery can be missed among women admitted before the onset of labor because they are often tested with standard (rather than rapid) tests. Although RT was an effective means of identifying HIV infection, many women did not receive a final HIV diagnosis before discharge from the hospital because of the relatively long turnaround times of confirmatory tests.

Quality Rating

There is no instrument for assessing the quality of this evaluation study, but the authors note several limitations. There was a low follow-up rate of HIV-infected women and their infants and no out-of-hospital information on breastfeeding. Among 94 infants born to HIV-positive mothers, more than one third (38%) still had undetermined HIV status 12 to 24 months after delivery, a delay possibly leading to an inaccurate estimate of perinatal HIV transmission rates. Starting January 2006, program staff in St. Petersburg implemented home-visit referrals for HIV-exposed infants who needed testing and care, which substantially increased follow-up rates. Although HIV incidence may be overestimated by the greater chance of being tested among injection drug users, the authors believe that the extent of this overestimation would be minimal because RT was performed in >84% of non-injection drug users who were eligible for such testing. No data were provided on the number of women giving birth who were taken to the other 15 hospitals for emergency delivery.

In Context

This study confirms the demonstration by the U.S. 16-hospital Mother-Infant Rapid Intervention at Delivery (MIRIAD) program that RT testing can be effective in rapidly identifying women with previously unknown HIV infection.(4) It is less clear how well the specific results in St. Petersburg, Russia can be generalized to other settings. The designation of two hospitals of 17 for "high-risk" women without known infections does not reflect the situation in many settings. The study found a high rate of HIV seroprevalence (6.5%) among women without previous HIV testing (6 times higher than that in the general population) and also high HIV incidence among those with prior negative test results. Lacking the date of the prior negative test result, the incidence had to be estimated with a range of 0.8 to 2.3 per 100 person-years. Even the lower bound estimate of 0.8 per 100 person-years is higher than that reported during pregnancy by Gray in Rakai, Uganda.(5) The estimate of 8.6% of infants infected is comparable to reductions in perinatal HIV transmission for non-breastfeeding women reported elsewhere.(4,6)

Programmatic Implications

RT programs at labor and delivery should be viewed as the last opportunity for testing women with unknown HIV status to prevent perinatal HIV transmission. Short courses of antiretroviral drugs, such as nevirapine, are clearly not the most efficacious prophylactic regimens for PMTCT. Therefore, more efforts must be directed toward the provision of early HIV testing, preferably before conception, to allow earlier initiation of antiretroviral regimens and reduction of perinatal HIV transmission. Another approach that can facilitate linkage of newly diagnosed HIV-infected women to care is the development of a rapid HIV diagnostic algorithm(1,2,3) that will allow hard-to-reach women to learn their confirmed HIV status. This approach possibly could improve their care-seeking behavior before they are discharged from the maternity hospital. Improving follow-up evaluations of HIV-infected women also will lead to improved follow-up evaluation of their infants. Every effort must be made to ensure that those who are the most vulnerable are assured access to follow-up testing, treatment, care and support.

References

  1. World Health Organization (WHO). Rapid HIV tests: guidelines for use in HIV testing and counseling services in resource-constrained settings. Geneva, Switzerland: WHO; 2004.
  2. Calero E, Malia J, Sawyer R, et al. Rapid HIV-1 diagnostic algorithms for use in HIV infection screening. 14th International AIDS Conference; 7-12 July 2002; Barcelona, Spain; 2005.
  3. Downing RG, Otten RA, Marum E, Biryahwaho B, Alwano-Edyegu MG, Sempala SD, et al. Optimizing the delivery of HIV counseling and testing services: the Uganda experience using rapid HIV antibody test algorithms. J Acquir Immune Defic Syndr Hum Retrovirol 1998;18:384-8
  4. Bulterys M, Jamieson DJ, O'Sullivan MJ, Cohen MH, Maupin R, Nesheim S, et al. Rapid HIV-1 testing during labor: a multicenter study. JAMA 2004;292:219-23.
  5. Gray RH, Li X, Kigozi G, Serwadda D, Brahmbhatt H, Wabwire-Mangen F, et al. Increased risk of incident HIV during pregnancy in Rakai, Uganda: a prospective study. Lancet 2005; 366:1182-8.
  6. Ayouba A, Tene G, Cunin P, Foupouapouognigni Y, Menu E, Kfutwa A, et al. Low rate of mother-to-child transmission of HIV-1 after nevirapine intervention in a pilot public health program in Yaounde, Cameroon. J Acquir Immune Defic Syndr. 2003;34:274-80.