Correspondence

Maternal Viral Load and the Risk of Perinatal Transmission of HIV-1

N Engl J Med 1999; 341:1698-1700November 25, 1999

Article

To the Editor:

In the August 5 issue, Mofenson et al.1 conclude that antiretroviral therapy “should be recommended to all infected pregnant women regardless of their HIV-1 [human immunodeficiency virus type 1] RNA levels,” and Garcia et al.2 conclude that zidovudine therapy has “demonstrated efficacy in reducing the risk of transmission regardless of the maternal HIV-1 RNA levels.” Neither study provides sufficient evidence to support these statements.

We updated the results of a meta-analysis of the predictive value of viral load for perinatal HIV-1 transmission3 with data presented by Garcia et al. Among 10 studies of pregnant women in the United States and Europe, none have shown that women with RNA levels of less than 1000 copies per milliliter have a lower rate of transmission with treatment than without treatment. Pooled data show an overall transmission rate of 3.9 percent in both untreated women (6 of 153) and treated women (5 of 127) with such low RNA levels. By contrast, transmission rates obtained with pooled data are clinically and statistically significantly higher for untreated than for treated women both in the category of 1000 to 10,000 RNA copies per milliliter (16.2 percent [60 of 370 women] vs. 7.8 percent [22 of 283 women]) and in the category of more than 10,000 copies per milliliter (34.5 percent [158 of 458 women] vs. 23.3 percent [44 of 189 women]).

We drew similar inferences when we combined study-specific rates or odds ratios for transmission rates among treated as compared with untreated women with fixed- or random-effects models (data not shown). Data on individual patients according to HIV-1 RNA levels (less than 1000, 1000 to 10,000, and more than 10,000 copies per milliliter) were not presented by Mofenson et al. but would shed more light on this issue. They did not include untreated women, thus prohibiting direct comparisons. Nevertheless, if the transmission rate for women with RNA levels of less than 1000 copies per milliliter is 0 in this study (as was reported for women with levels of less than 500 copies per milliliter), the pooled transmission rate for treated women in this RNA category among all studies would be closer to 2 percent.

Lack of randomization in the comparison of treated and untreated groups unavoidably leads to selection bias. Further bias may stem from the use of different assays for measuring viral load, different ways of storing specimens, measurement error (especially with single measurements), and variable therapeutic regimens.3 More data are needed to support the conclusion that a clinically relevant difference in transmission rates results from the use of antiretroviral therapy in pregnant women with RNA levels of less than 1000 copies per milliliter. This point is important, because women who were doing very well may have deferred initiation of antiretroviral therapy before pregnancy and may have been reluctant to commit to treatment with zidovudine or combination regimens.4

John P.A. Ioannidis, M.D.
Despina G. Contopoulos-Ioannidis, M.D.
University of Ioannina School of Medicine, Ioannina 45110, Greece

4 References

1
Mofenson LM, Lambert JS, Stiehm ER, et al. Risk factors for perinatal transmission of human immunodeficiency virus type 1 in women treated with zidovudine. N Engl J Med 1999;341:385-393
Full Text | Web of Science | Medline

2
Garcia PM, Kalish LA, Pitt J, et al. Maternal levels of plasma human immunodeficiency virus type 1 RNA and the risk of perinatal transmission. N Engl J Med 1999;341:394-402
Full Text | Web of Science | Medline

3
Contopoulos-Ioannidis DG, Ioannidis JP. Maternal cell-free viremia in the natural history of perinatal HIV-1 transmission: a meta-analysis. J Acquir Immune Defic Syndr Hum Retrovirol 1998;18:126-135
CrossRef | Medline

4
Ioannidis JP, O'Brien TR, Goedert JJ. Evaluation of guidelines for initiation of highly active antiretroviral therapy in a longitudinal cohort of HIV-infected individuals. AIDS 1998;12:2417-2423
CrossRef | Web of Science | Medline

Author/Editor Response

The authors reply:

To the Editor: As we noted, current Public Health Service guidelines recommend zidovudine prophylaxis to reduce the risk of perinatal HIV-1 transmission regardless of maternal HIV-1 RNA level.1 Drs. Ioannidis and Contopoulos-Ioannidis correctly point out the lack of data that could definitively demonstrate that zidovudine lowers the rate of perinatal transmission in the subgroup of HIV-1–infected pregnant women with RNA levels of less than 1000 copies per milliliter. Data on HIV-1 RNA levels among untreated women are limited, because HIV-1 RNA assays were not generally available before 1994 and because since 1994, most HIV-1–infected women receiving prenatal care in the United States have received zidovudine prophylaxis. Furthermore, when transmission rates are low (as they appear to be in women with low HIV-1 RNA levels), large numbers of patients are required for a study to have the power to detect an effect of treatment on transmission. For example, if the transmission rate is assumed to be 5 percent among untreated women with HIV-1 RNA levels of less than 1000 copies per milliliter, approximately 1700 mother–infant pairs (850 per group) would be required for a study to establish that zidovudine reduces the transmission rate by 50 percent (with a power of 80 percent and a one-sided alpha of 0.05).

Limited data on the association of HIV-1 RNA levels, zidovudine use, and the risk of transmission are available from the original placebo-controlled trial (Pediatric AIDS Clinical Trials Group Protocol 076).2 The transmission rates for patients with HIV-1 RNA levels of less than 1730 copies per milliliter (the lowest quartile according to the results of the reverse-transcription HIV-1 RNA polymerase-chain-reaction assay) were 2.5 percent in the zidovudine group and 7.5 percent in the placebo group.2 As we also noted, our results can only be used to address transmission rates among women receiving zidovudine and cannot be used specifically to address whether zidovudine is effective in lowering the rate among women with HIV-1 RNA levels of less than 1000 copies per milliliter. In our study, 1 of 135 women with RNA levels of less than 1000 copies per milliliter (0.7 percent), 6 of 149 women with RNA levels of 1000 to 10,000 copies per milliliter (4.0 percent), and 17 of 195 women with RNA levels of more than 10,000 copies per milliliter (8.7 percent) had infected infants. If one pools our data with the data provided by Drs. Ioannidis and Contopoulos-Ioannidis, the transmission rate was 41 percent lower among zidovudine-treated women (2.3 percent among 262 treated women, as compared with 3.9 percent among 153 untreated women).

The mechanism of zidovudine efficacy is most likely multifactorial and involves both lowering of the maternal viral load and pre- and postexposure prophylaxis in the infant. The importance of pre- and postexposure prophylaxis for preventing perinatal transmission is demonstrated by the recent results of a Ugandan study showing the efficacy of a two-dose intrapartum–postpartum regimen of nevirapine in lowering the risk of perinatal transmission.3

The risk of transmission among untreated women with HIV-1 RNA levels of less than 1000 copies per milliliter is low but not zero. For women with low viral loads who wish to reduce the risk of transmission of HIV-1 infection to their infants even further, potential interventions include zidovudine prophylaxis and, for those who do not wish to expose the fetus to antiretroviral drugs during pregnancy, intrapartum–postpartum nevirapine or elective cesarean delivery.4

Lynne M. Mofenson, M.D.
National Institute of Child Health and Human Development, Rockville, MD 20852

John S. Lambert, M.D.
University of Maryland, Baltimore, MD 21201

E. Richard Stiehm, M.D.
UCLA Medical Center, Los Angeles, CA 90024

4 References

1
Public Health Service Task Force recommendations for the use of antiretroviral drugs in pregnant women infected with HIV-1 for maternal health and for reducing perinatal HIV-1 transmission in the United States. MMWR Morb Mortal Wkly Rep 1998;47:1-30[Erratum, WR Morb Mortal Wkly Rep 1998;47:287, 315.]
Medline

2
Sperling RS, Shapiro DE, Coombs RW, et al. Maternal viral load, zidovudine treatment, and the risk of transmission of human immunodeficiency virus type 1 from mother to infant. N Engl J Med 1996;335:1621-1629
Full Text | Web of Science | Medline

3
Guay LA, Musoke P, Fleming T, et al. Intrapartum and neonatal single-dose nevirapine compared with zidovudine for prevention of mother-to-child transmission of HIV-1 in Kampala, Uganda: HIVNET 012 randomised trial. Lancet 1999;354:795-802
CrossRef | Web of Science | Medline

4
The International Perinatal HIV Group. The mode of delivery and the risk of vertical transmission of human immunodeficiency virus type 1 -- a meta-analysis of 15 prospective cohort studies. N Engl J Med 1999;340:977-987
Full Text | Web of Science | Medline

Author/Editor Response

We agree with Drs. Ioannidis and Contopoulos-Ioannidis that our data do not prove that zidovudine therapy reduces the rate of perinatal transmission of HIV-1 among the select and somewhat arbitrarily defined subgroup of women with HIV-1 RNA levels below 1000 copies per milliliter. The small size of this subgroup in our study as well as in the pooled data that Ioannidis and Contopoulos-Ioannidis used precludes the detection of differences in transmission rates with respect to treatment, since the transmission rates are low. The frequencies both of the condition and the event are low, and there are ethical problems with placebo-controlled trials once the efficacy of affordable and easily administered chemoprophylaxis has been demonstrated.1,2 Hence, it is unlikely that we will ever know definitively whether any intervention is better than no intervention for pregnant women with low levels of plasma HIV-1 RNA.

Our recommendation that women be offered zidovudine prophylaxis regardless of their HIV-1 RNA level, in accordance with Public Health Service guidelines,3 is supported by several observations. Perinatal transmission is multifactorial, and our multivariate analysis clearly demonstrates the independent effect of zidovudine therapy on transmission. Sperling et al.4 point out that the protection against transmission offered by zidovudine appears to be due only in part to its ability to decrease viral load. The recently demonstrated efficacy of peripartum antiretroviral prophylaxis2 suggests a mechanism more closely related to prophylactic treatment of the fetus than to a reduction of maternal viral load. Despite our finding that there was no transmission among women with HIV-1 RNA levels of less than 1000 copies per milliliter, there are multiple reports of transmission at this level, including one study in which 16 of 132 women with such RNA levels transmitted HIV-1 infection to their infants.5 We would therefore caution readers that translating population-level risks to the assessment of individual risks for purposes of clinical management is potentially hazardous. Measurement of maternal plasma HIV-1 RNA has inherent biologic and technical variability that makes establishing an individual threshold value difficult, not to mention the practical issues associated with monitoring and management decisions when time is so critical.

We concur that further study is needed to determine the specific benefits and risks of prophylactic interventions such as antiretroviral therapy or cesarean delivery for women with low viral loads and their infants. The existence of clinically significant negative maternal consequences of brief exposure to zidovudine or nevirapine is not sufficiently well established to justify deferring perinatal prophylaxis; the risks of such treatment thus do not outweigh its potential benefits.

Patricia M. Garcia, M.D., M.P.H.
Northwestern University Medical School, Chicago, IL 60611-3095

Jack Moye, Jr., M.D.
National Institute of Child Health and Human Development, Bethesda, MD 20892-7510

Judy F. Lew, M.D.
University of Florida, Gainesville, FL 32610-0296

5 References

1
Shaffer N, Chuachoowong R, Mock PA, et al. Short-course zidovudine for perinatal HIV-1 transmission in Bangkok, Thailand: a randomised controlled trial. Lancet 1999;353:773-780
CrossRef | Web of Science | Medline

2
Guay LA, Musoke P, Fleming T, et al. Intrapartum and neonatal single-dose nevirapine compared with zidovudine for prevention of mother-to-child transmission of HIV-1 in Kampala, Uganda: HIVNET 012 randomised trial. Lancet 1999;354:795-802
CrossRef | Web of Science | Medline

3
Public Health Service Task Force recommendations for the use of antiretroviral drugs in pregnant women infected with HIV-1 for maternal health and for reducing perinatal HIV-1 transmission in the United States. MMWR Morb Mortal Wkly Rep 1998;47:1-30[Erratum, WR Morb Mortal Wkly Rep 1998;47:287, 315.]
Medline

4
Sperling RS, Shapiro DE, Coombs RW, et al. Maternal viral load, zidovudine treatment, and the risk of transmission of human immunodeficiency virus type 1 from mother to infant. N Engl J Med 1996;335:1621-1629
Full Text | Web of Science | Medline

5
Mayaux MJ, Dussaix E, Isopet J, et al. Maternal virus load during pregnancy and mother-to-child transmission of human immunodeficiency virus type 1: the French perinatal cohort studies. J Infect Dis 1997;175:172-175
CrossRef | Web of Science | Medline

Citing Articles (2)

Citing Articles

1
Mark R. Schleiss, Janna C. Patterson. 2012. Viral Infections of the Fetus and Newborn and Human Immunodeficiency Virus Infection during Pregnancy. , 468-512.
CrossRef
2
Jocelyn Wertz, Jason Cesario, Jennifer Sackrison, Sean Kim, Chi Dola. (2011) Acute HIV Infection in Pregnancy: The Case for Third Trimester Rescreening. Case Reports in Infectious Diseases 2011, 1-5
CrossRef