Objective
The objective of the study was to determine whether interpregnancy human immunodeficiency virus (HIV) viral load suppression affects outcomes in subsequent pregnancies.
Study Design
This is a retrospective review of all women who delivered 2 consecutive pregnancies while diagnosed with HIV from Jan. 1, 1984, until Jan. 1, 2012. Medical records were reviewed for maternal, infant, and delivery data. Pregnancies were divided into index and subsequent pregnancy and analyzed for outcomes.
Results
During the study period, 172 HIV-infected women who delivered 2 pregnancies at our institution were identified. There was no difference in median HIV viral load at presentation or delivery between the index and subsequent pregnancies. During the subsequent pregnancy, more women presented on antiretroviral therapy (ART) and more often remained compliant with ART; however, there was no difference in vertical transmission risk between the pregnancies. Of those with a viral load less than 1000 copies/mL at the end of their index pregnancy (n = 103), 57 (55%) presented for their subsequent pregnancy with a viral load still less than 1000 copies/mL. Those women who maintained the viral load suppression between pregnancies were more likely to present for their subsequent pregnancy on ART, maintained a greater viral load suppression and CD4 counts during the pregnancy, and had fewer vertical transmissions compared with those who presented with higher viral loads in their subsequent pregnancy (0% vs 9%, P = .02).
Conclusion
Maintaining an HIV viral load suppression between pregnancies is associated with improved HIV disease status at delivery in subsequent pregnancies. Interpregnancy HIV viral load suppression is associated with less vertical transmission, emphasizing the importance of maintaining HIV disease control between pregnancies.
Currently more than 1.1 million people are living with human immunodeficiency virus (HIV) in the United States, with approximately 50,000 new cases diagnosed every year. Females account for approximately 20% of these newly diagnosed cases. Concurrently, antiretroviral therapy (ART) has led to an increasing number of people living with chronic HIV infection and therefore an increasing number of women living with HIV who are giving birth.
The current guidelines involving the treatment of nonpregnant adults living with HIV are becoming more aggressive, with higher levels of CD4 cell count being recommended to initiate ART. The World Health Organization currently recommends initiating therapy on any HIV-infected person with a CD4 cell count of 500 cells/μL or less. The rationale for this more aggressive approach is that early initiation and continuation of therapy improves long-term outcomes and reduces disease progression. The Department of Health and Human Services perinatal guidelines recommend treating all pregnant women, irrespective of CD4 cell count or prior treatment, for the benefit of reducing perinatal transmission.
In contrast to this, there are no consistent recommendations regarding the use of ART in the preconceptual period. The only currently recommended indication for preconceptual initiation of therapy is for the maternal medical benefit according to the guidelines for ART initiation in the nonpregnant adult population. This is despite the fact that there is some evidence, although limited, that treatment of HIV-infected women prior to conception may decrease the rate of vertical transmission. Furthermore, there are no data as to the optimal management of women with HIV of reproductive age between pregnancies. Although there have been previous studies of repeat pregnancies in women living with HIV, these have been limited to epidemiological studies and the effects of pregnancy on HIV progression. None of these studies have examined the effect of interpregnancy HIV viral load suppression on subsequent pregnancy outcomes.
Considering that current recommendations for the treatment of the nonpregnant adult with HIV are becoming more aggressive and preconceptual and interpregnancy guidelines regarding the initiation of treatment are limited, we believed this cohort of patients required further investigation. We hypothesized that the interpregnancy HIV viral load suppression would positively affect pregnancy outcomes in subsequent pregnancies.
Materials and Methods
This is a retrospective cohort study of all women living with HIV who delivered 2 consecutive live births at our institution from Jan. 1, 1984, until Jan. 1, 2012. This study was approved by the Institutional Review Board of The University of Texas Southwestern Medical Center and Parkland Hospital.
All women at our institution with a known diagnosis of HIV or newly diagnosed with HIV during pregnancy are cared for at a single prenatal clinic by dedicated obstetricians, nurse practitioners, and case managers. Our clinical practice is dictated by current national guidelines for the care of HIV-infected pregnant women. During the study period, as national guidelines changed, our clinical practice changed accordingly. In conjunction with our high-risk obstetrics clinic, nonpregnant adults living with HIV are cared for at dedicated clinics within our institution that maintain communication with our high-risk obstetrics clinic. A dedicated woman’s health nurse practitioner cares for HIV-infected women at both locations, thereby maintaining continuity of care between the nonpregnant, prenatal, and postpartum periods.
Those women who gave birth to 2 consecutive live births at our institution while diagnosed with HIV were identified, and their medical records were reviewed. In women who had more than 2 pregnancies during the study period, only the first 2 pregnancies were included in this analysis. Information on maternal HIV disease status at the initiation of prenatal care was obtained including HIV viral load, length of diagnosis, and previous antiretroviral treatment. Prenatal records were reviewed for ART use during pregnancy, number of prenatal visits, and markers of HIV disease progression.
For the purposes of this report, antiretroviral therapy was defined as either monotherapy or multidrug therapy. Patient compliance with therapy was determined by the patient’s self-report and was reassessed at each prenatal visit as well as at presentation for delivery. Delivery information including route of delivery, intrapartum zidovudine administration, length of labor, and evidence of intrapartum or postpartum febrile illness was also obtained.
The infant record was reviewed to determine HIV status. All infants at our institution received standardized care according to national guidelines in our neonatal intensive care unit that was appropriate for the year in which they delivered. Also, all women cared for in our institution are instructed not to breast-feed after delivery.
For the purposes of this report, consecutive pregnancies will be referred to as an index pregnancy and subsequent pregnancy. These pregnancies serve as the basis for this analysis. The interpregnancy interval was defined as the time from the delivery of the index pregnancy until the time of presentation for prenatal care of the subsequent pregnancy.
After analysis of outcomes comparing the index pregnancy with the subsequent pregnancy, women with an HIV viral load less than 1000 copies/mL at the time of delivery of the index pregnancy were identified and divided into cohorts defined by their viral load at presentation for their subsequent pregnancy. This threshold was chosen because of its current use as a clinically relevant threshold above which cesarean delivery is indicated for the reduction of vertical transmission. Those women who at presentation for their subsequent pregnancy had HIV viral load less than 1000 copies/mL were determined to have maintained HIV viral load suppression during the interpregnancy interval, whereas those who presented for their subsequent pregnancy with an HIV viral load greater than 1000 copies/mL were determined to have not maintained suppression during the interpregnancy interval, without regard to the reported antiretroviral therapy use. These cohorts were then compared for the selected outcomes.
Statistical analysis included a Pearson χ 2 , Student t test, Cochran-Mantel-Haenszel, χ 2 for trend, and Wilcoxon rank sum. After the initial analysis, we adjusted our results to take into consideration the introduction of multidrug therapy, which occurred in 1996. This was done by subgrouping the patients based on the year of delivery of the index and subsequent pregnancy, resulting in 3 groups: those with both pregnancies prior to 1996, those with the index pregnancy prior to 1996 and the subsequent pregnancy after, and those with both pregnancies after 1996. For categorical data the analysis was conducted with a Cochran-Mantel-Haenszel test and for continuous measures analysis of covariance. Values of P < .05 were considered significant. The analysis was performed using SAS version 9.2 (SAS Institute Inc, Cary, NC).
Results
During the study period, 172 women who delivered 2 consecutive pregnancies were identified. The mean interpregnancy interval was 2.75 years (1004 days; 95% confidence interval [CI], 889–1118 days). Selected demographic characteristics of these repeat pregnancies are shown in Table 1 .
| Characteristic | Index pregnancy (n = 172) | Subsequent pregnancy (n = 172) | P value |
|---|---|---|---|
| Age, y | 23.9 ± 5.5 | 26.6 ± 5.6 | NA |
| Race | |||
| Black | 122 (71) | — | NA |
| White | 21 (12) | — | NA |
| Hispanic | 29 (17) | — | NA |
| Parity | 1 [1, 2] | 2 [2, 3] | NA |
| Gestational age at first prenatal visit, wks | 14 [9, 23] | 12 [8, 20] | .12 |
| Number of prenatal visits | 9 [6, 12] | 10 [6, 12] | .79 |
There was no difference in HIV viral load at presentation or at delivery between the index and subsequent pregnancies, with the median viral load at delivery of both the index and subsequent pregnancy being nondetectable ( Table 2 ). More women presented to their subsequent pregnancy on ART (56% vs 35%, P < .001), and more women remained on ART at delivery (87% vs 78%, P = .02).
| Variable | Index pregnancy (n = 172) | Subsequent pregnancy (n = 172) | P value | Adjusted OR (95% CI) |
|---|---|---|---|---|
| Viral load at presentation, copies/mL | 1860 [0, 21,700] | 1120 [0, 14,500] | .59 | NA |
| CD4 cell count at presentation, cells/μL | 456 [343, 612] | 479 [321, 632] | .63 | 0.07 a |
| ART at presentation | 61 (35) | 97 (56) | < .001 | 2.52 (1.60–3.96) |
| Viral load at delivery, copies/mL | 0 [0, 119] | 0 [0, 436] | .22 | NA |
| CD4 cell count at delivery, cells/μL | 532 [361, 717] | 520 [349, 674] | .14 | 0.17 a |
| ART at delivery | 134 (78) | 150 (87) | .02 | 2.25 (1.19–4.26) |
| Viral load <1000 copies/mL at delivery | 111 (65) | 120 (70) | .30 | 1.36 (0.81–2.28) |
| Vertical transmission | 5 (3) | 12 (7) | .08 | 2.98 (0.91–9.75) |
The rate of vertical transmission was similar between the index and subsequent pregnancy. When the data were adjusted for the introduction of highly active antiretroviral therapy in 1996, there was no difference in the significance of our findings. More women presented to their subsequent pregnancy on ART (adjusted odds ratio [OR], 2.52; 95% CI, 1.60–3.96) and remained on ART at delivery (adjusted OR, 2.25; 95% CI, 1.19–4.26).
Analysis of covariance for CD4 cell count at presentation and delivery also showed no change when adjusted for the year of delivery as described. Viral load at presentation and delivery was unable to be adjusted because the HIV viral load was not routinely obtained prior to 1996 because of routine clinical practice.
To evaluate the interpregnancy interval, we compared CD4 cell count and HIV viral load at the time of delivery of the index pregnancy with CD4 cell count and HIV viral load at presentation of the subsequent pregnancy. During this interpregnancy interval, the overall cohort HIV viral load significantly increased by a median of 182 copies/mL ( P < .001), and the median CD4 cell count decreased during this time by 54 cells/μL ( P = .01).
Of the original 172 HIV-infected women, we identified 111 women with viral load less than 1000 copies/mL at the time of delivery of their index pregnancy. Of those 111 women, 103 women had a viral load determined upon presentation for prenatal care of the subsequent pregnancy. Fifty-seven of the women with HIV viral load less than 1000 copies/mL at the end of the index pregnancy (55%) maintained HIV viral load suppression, with the HIV viral load still less than 1000 copies/mL at presentation for care for their subsequent pregnancy.
There were no differences in demographic characteristics between those women who maintained the HIV viral load suppression between pregnancies and those who did not as shown in Table 3 . The interpregnancy interval was not different between the cohorts (2.7 vs 2.6 years, P = .93). There was also no difference in the gestational age at which they presented for care in the subsequent pregnancy (10 vs 11 weeks, P = .69) or in the number of prenatal visits they received in the subsequent pregnancy (10 vs 11 visits, P = .87).
| Characteristic | Maintained viral load suppression (n = 57) | Did not maintain viral load suppression (n = 46) | P value |
|---|---|---|---|
| Age, y | 27.8 ± 5.4 | 26.8 ± 5.9 | .37 |
| Race | .71 | ||
| Black | 37 (65) | 31 (67) | |
| Hispanic | 12 (21) | 7 (15) | |
| White | 8 (14) | 8 (17) | |
| Parity | 2 [2, 3] | 2 [2, 3] | .79 |
| Gestational age at first presentation, wks | 10 [8, 21] | 11 [7, 17] | .69 |
| Number of prenatal visits | 10 [6, 14] | 11 [8, 12] | .87 |
| Length of HIV diagnosis, y | 3 [2, 5.5] | 3 [2, 6] | .93 |
| Interpregnancy interval, y | 2.7 ± 1.7 | 2.6 ± 1.6 | .93 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
