|Home | About | Journals | Submit | Contact Us | Français|
Acyclovir therapy in late pregnancy among women with recurrent genital herpes is effective in decreasing genital lesion frequency and subclinical viral shedding rates at delivery, thereby decreasing the need for caesarean delivery. Despite good adherence and increased dosing schedules, breakthrough lesions and viral shedding are still observed in some women at or near delivery. Anecdotal data suggests that low levels of HSV replication at delivery may result in transmission to the neonate. Therefore, defining optimal acyclovir dosing during labor and delivery is warranted. Our objectives were to determine actual acyclovir levels at delivery, and explore associations between acyclovir levels, duration of labour and time since last acyclovir dose.
Twenty-seven patients were prescribed oral acyclovir 400 mg three times daily from 36 weeks gestation. Cord blood (venous and arterial) and maternal venous blood samples were collected at delivery, and acyclovir levels measured using capillary electrophoresis. Correlations between duration of labour and time since last acyclovir dose with acyclovir blood levels were calculated.
Acyclovir levels were below the published mean steady-state trough value (180 ng/ml) in 52% of venous cord, 55% of arterial cord, and 36% of maternal samples. There was a significant inverse correlation between time since last dose and venous cord (rs19=−0.57, p<0.015), arterial cord (rs16=−0.63, p<0.01), and maternal acyclovir levels (r10=−0.69, p<0.03).
Oral dosing of acyclovir in late pregnancy may result in insufficient levels at delivery to prevent viral shedding. Alternative approaches should evaluate dosing through labor, perhaps intravenously, and its effect on viral shedding.
Neonatal herpes is one of the most serious consequences of herpes simplex virus (HSV) infection, and is a significant public health concern in many developed countries. . The estimated incidence of neonatal herpes varies by geographical region and study methodology, ranging from 1 case per 20,000 – 60,000 live births in European countries1-3, 1 case per 17, 000 live births in Canada4, 1 case per 11,000 live births in Australia5, and up to 1 case per 3,200 live births in parts of the United States.6 In at least 85% of cases, HSV infection in the newborn is due to transmission during the time of vaginal delivery from the infected mother's genital tract.7
To reduce the risk of perinatal transmission of HSV, the American College of Obstetricians and Gynecologists (ACOG) and the Society of Obstetricians and Gynaecologists of Canada (SOGC) have recommended that caesarean delivery be performed on women presenting with genital herpes lesions at the onset of labour.8,9 This management strategy relies on the clinical recognition of genital infection. The increased morbidity and mortality associated with caesarean as compared to vaginal delivery has prompted evaluation of alternate strategies, including antiviral suppressive therapy near term10-16, now an ACOG and SOGC recommended strategy.8,9
The antiviral agents acyclovir, valacyclovir, and famciclovir are effective in the suppression of recurrent genital herpes in non-pregnant women.17 Several studies have also evaluated the safety and effectiveness of oral acyclovir suppressive therapy in late pregnancy10-13,15,18 and found a significant reduction in both symptomatic genital herpes recurrence and viral shedding at delivery, as well as a reduction in cesarean rates.19 The current recommendations in women with known recurrent HSV infection are to start oral acyclovir or valacyclovir therapy at 36 weeks' gestation until delivery.9,20
While pharmacokinetics of acyclovir in late pregnancy have been assessed21, maternal and fetal levels of acyclovir at the time of delivery have not been evaluated. This information would be of value because the circumstances of labour and impending delivery may interrupt the dosing schedule of antiviral medication, and potentially interfere with drug adherence and absorption. In turn, this could contribute to escape viral replication around time of delivery. Therefore, the objectives of this study were 1) to determine the level of maternal and umbilical cord acyclovir in women who were prescribed and self-administered suppressive acyclovir from 36 weeks gestation until delivery, and 2) to explore the relationship between the level of acyclovir and the duration of labour, and time since last acyclovir dose. We hypothesized that the cord blood acyclovir concentrations will be inversely correlated with the time from last acyclovir dose and the duration of labour.
The participants in this cohort study were women with recurrent genital HSV enrolled in a randomized placebo controlled trial of acyclovir 400 mg TID versus placebo, from 36 weeks to delivery.22 The pharmacokinetics substudy was conducted at the University of British Columbia, and was approved by the UBC Clinical Research Ethics Board and the British Columbia's Children's and Women's Health Centre Research Review Committee (Approval No. C98-0076). Subjects were recruited between January 2000 and April 2003. Inclusion criteria included a normal 18-20 week obstetrical ultrasound, known recurrent symptomatic genital HSV infection, no serious medical conditions including infection with HIV, and no known sensitivity to acyclovir. All participants in this substudy were prescribed oral acyclovir 400 mg TID from 36 weeks gestation until delivery, and consented to provide additional umbilical cord and maternal blood samples at the time of delivery.
Maternal and neonatal information collected included maternal age, maternal ethnicity, gravidity, maternal weight at study onset (32-36 weeks gestational age), duration of labor, time of last acyclovir dose before delivery, length of gestation, infant's birth weight, and Apgar score. If a caesarean section was performed, duration of labour was recorded as the time of admittance to hospital until delivery.
Arterial and venous cord blood samples were collected from the clamped umbilical cord, and maternal venous samples were taken immediately after delivery. Cord and maternal samples were stored at 4°C for a maximum of 72 hours prior to centrifugation at 1200 rpm at 4°C for 10 min. The plasma was then stored at −70°C until analysis. Validation studies were performed to confirm acyclovir stability in the above conditions.23
The concentration of acyclovir in the plasma samples was determined using a validated capillary electrophoresis assay, developed in the Sacks laboratory, with a limit of quantification of 20 ng/ml.23 Briefly, plasma samples were purified using solid phase extraction and column elution. Separation and analysis of acyclovir in plasma components was performed on a Beckman MDQ automated capillary electrophoresis system, coupled with an ultraviolet (UV) detector (Beckman Instruments Inc., Missisauga, ON, Canada). The internal standard was 5-(2-hydroxyethyl)-2′-deoxyurindine (HEdU). Detection of acyclovir was monitored at the wavelength of 254 nm. Corrected peak-area ratios (ACV/HEdU) were calculated and the acyclovir concentrations were determined using a constructed calibration curve.23
Maternal weight at study onset, duration of labour, time since last acyclovir dose, and maternal acyclovir level data were approximately normal in distribution. Venous cord and arterial cord acyclovir concentration were not normally distributed; therefore non parametric tests were used in analyses of these variables. To investigate if there was a statistically significant association between plasma acyclovir concentration and maternal weight, duration of labour, and time since last acyclovir dose, correlational analyses using Pearson Correlation coefficient (parametric) or Spearman's rho statistic (nonparametric) were used. All statistical analyses were two-tailed tests. Statistical analyses were performed using SPSS release 16.0.1 for Windows (SPSS Inc., Chicago, IL, USA).
Twenty-seven women with a history of genital herpes infection who were taking suppressive acyclovir at term were included in the study (Table 1). All participating women were given a medication dosing schedule (400mg TID) and reported >95% acyclovir adherence rates in the week preceding delivery, but were not instructed regarding any specific management once labor began. All of the women were anticipating a vaginal delivery, although seven women underwent a caesarean section at delivery due to failure to progress (n = 3), breech presentation (n = 2), fetal distress (n = 1), or genital herpes prodrome (n = 1). None of the women had clinically evident genital herpes lesions at delivery, and none of the infants developed neonatal herpes infection. Additional neonatal outcome data is presented in Table 1.
In the labour and delivery wards where study participants gave birth, priority was always given to patient care and collection of samples for routine clinical diagnostics, which occasionally resulted in missed study sample collections, or collection of an insufficient volume of blood for analysis. Due to these logistical issues, it was not always possible to obtain paired cord and maternal blood samples for all of the study participants; final sample sizes are reported in Table 1. The mean concentration of acyclovir was 244.5 ng/ml in venous cord plasma, 236.0 ng/ml in arterial cord plasma, and 421.7 ng/ml in maternal plasma (Table 1). Fourteen of 27 (52.0%) venous cord acyclovir levels, 12 of 22 (54.5%) arterial cord acyclovir levels, and 4 of 11 (36.4%) of maternal acyclovir levels were below the published mean steady-state trough concentration of 0.8 μM (~180 ng/ml).21,24
Maternal weight was not correlated with venous cord (rs27 = −0.183, p = 0.4), arterial cord (rs22 = −0.013, p = 1.0), or maternal cord acyclovir levels (r11 = −0.153, p = 0.7). There was no significant correlation between duration of labour and venous cord (rs27 = 0.097, p = 0.6), arterial cord (rs22 = 0.003, p = 1.0), or maternal cord acyclovir levels (r11 = 0.0334, p = 0.3). However, in a subset of 19 women for whom time since last dose of acyclovir before delivery was known, there was a significant inverse correlation between time since last dose and venous cord (rs19 = −0.57, p < 0.015), arterial cord (rs16 = −0.63, p < 0.01), and maternal acyclovir levels (r10 = −0.69, p <0.03; Figure 1).
Our results suggest that plasma acyclovir levels at the time of delivery are often suboptimal among women receiving oral acyclovir during late pregnancy. Previous studies have reported that the mean steady state trough level is 180 ng/ml for oral acyclovir 400 mg TID.21,24 Over 50% of the cord blood samples and 35% of maternal blood samples collected in the study were below this value. As expected, maternal and cord blood acyclovir concentrations were inversely correlated with time since last dose of acyclovir. Although we did not detect any significant correlation between duration of labour and acyclovir levels as hypothesized, this may be due to the higher degree of variability between individuals in the duration of labour, or that several women (9/19) did take their last dose of acyclovir during the first phase of labour. As such, time since last dose was a more accurate baseline from which to explore temporal changes in acyclovir levels.
The primary concern with the utilization of acyclovir suppression in any clinical setting is clinical breakthrough and subclinical viral shedding. In the context of pregnancy, the goal has been both clinical and virologic suppression in order to obviate the need for caesarean delivery, but also to minimize risk of exposure of the virus to the infant. However, these approaches may not specifically incorporate maintenance of the acyclovir dosing schedule throughout labour, or recommend alternate modes of delivery of acyclovir for labouring women unable to take oral medications.
This study has demonstrated that without specifically addressing the administration of acyclovir in labour, an increased proportion of women and their infants may be exposed to subtherapeutic acyclovir levels. This could potentially result in reactivation of genital herpes in labor at a time when virologic suppression is most critical. While the minimum infectious titre of HSV required for transmission is unknown, case reports indicate that even low levels of virus in maternal secretions can result in transmission to the infant.25 Our data indicate that further studies are needed to evaluate acyclovir dosing during labour, either through oral or intravenous delivery.
Although it is difficult to determine whether the suboptimal concentrations are due to non-compliance or to the physiological circumstances of labor, the inverse relationship with last dose and level suggests typical pharmacokinetics in response to doses taken in labor. Physiological changes during pregnancy can cause alterations in pharmacokinetics, such as the increase in renal clearance.26 However, no notable differences in plasma levels and drug clearance compared to non-pregnant adults were observed in previously reported pharmacokinetic studies of acyclovir in pregnancy, if increased dose and frequency is used.21,24,27 Nevertheless, the controlled environment of a pharmacokinetic study may not accurately depict real-life conditions during labor and delivery, where issues such as vomiting and decreased gastric emptying may present as significant barriers to optimal drug bioavailability.28 Due to the short elimination half-life of acyclovir (2-3 h)29, frequent dosing is required when using oral therapy.
This study provides an important exploration of the levels of acyclovir at delivery and issues related to acyclovir administration around labour of varying duration. External generalizability would be strengthened by a larger sample; nevertheless, this study suggests the currently recommended antiviral regimen may result in subtherapeutic levels at the time of delivery. This highlights the need for further research to examine the clinical significance of neonatal exposure to sub-therapeutic acyclovir levels, and explore alternative dosing regimens that will ensure optimal acyclovir levels throughout the labour process.
Women with genital herpes who choose to take acyclovir to reduce the risk of cesarean section should continue to take oral acyclovir through labour until delivery to minimize risk for genital tract viral replication at the time of birth.
This study was supported by grants from the National Institutes of Health (P01-AI30731), and the Natural Sciences and Engineering Research Council of Canada. D.T.L. was supported by a GREAT Scholarship from the Science Council of British Columbia. One of the authors of this paper, Dr. Stephen L. Sacks, died in November 2003 but contributed throughout the study and through the editing of the manuscript.