HIV detection in amniotic fluid samples

European Journal of Obstetrics & Gynecology and Reproductive Biology 108 (2003) 137–141

HIV detection in amniotic fluid samples Amniocentesis can be performed in HIV pregnant women? Vicente Maiquesa,*, Amparo Garcı´a-Tejedora, Alfredo Peralesb, Juan Co´rdobac, Roque J. Estebanc a

Obstetric Service, Hospital Maternal ‘‘La FE’’, 58 Planta, Avda. Campanar 21, 46009 Valencia, Spain b Obstetric Service, Hospital Peset, Valencia, Spain c Microbiology Service, Molecular Biology Unity, Hospital Universitario ‘‘La FE’’, Valencia, Spain Received 9 May 2002; received in revised form 18 September 2002; accepted 25 September 2002

Abstract Objectives: To assess amniotic fluid (AF) HIV contamination as a marker of intrauterine HIV infection and to evaluate amniocentesis as a risk factor for vertical HIV transmission when the mother was under antiretroviral treatment. Study design: Three hundred and sixty-six HIV pregnant women were included in the study. AF was obtained from three groups of patients: (a) genetic amniocentesis before 1997 (n ¼ 11); (b) amniocentesis a few days before the delivery day (n ¼ 18); and (c) AF collected on delivery (n ¼ 38). An univariate study was conducted to analyze amniocentesis as a risk factor of HIV transmission (groups a and b). Groups b and c were recruited after 1997; these patients were under combined antiretroviral treatment, they were studied to relate AF HIV contamination with fetal infection and maternal blood viral load at delivery (n ¼ 56). Results: From 1984 to 1996, before antiretroviral therapy use in HIV pregnant women, transmission rate was 17%. In the group of patients who underwent genetic amniocentesis (group a) it was 30% (3/10) versus 16.2% (40/247) for patients without amniocentesis. Between 1997 and 2000 transmission rate was 3%. In group b it was 0% (0/18) when amniocentesis was done versus 3.7% (3/81) if no amniocentesis was done (no statistical differences). AF virus was undetectable in all samples (n ¼ 56) and no newborn infection was observed after the follow up. Conclusions: Amniotic fluid virus load was undetectable when maternal antiretroviral therapy was used, even if the virus was detectable in maternal blood samples. This finding raises the possibility to perform amniocentesis, when it is indicated, to provide the mother with an adequate antiretroviral treatment. # 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: HIV; Amniocentesis; Amniotic fluid; Virus load; Antiretroviral treatment

1. Introduction Before antiretroviral therapy use mother-to-child HIV transmission rate ranged from 14 to 48% [1,2]. Nowadays, developed countries have achieved a transmission rate near to zero when elective caesarian section was associated with antiretroviral treatment [3,4]. Different authors have demonstrated that 75% of perinatal HIV transmission occurs intrapartum and 25% in uterus [5] (in these cases, less than 2% is in the mid-trimester [6]). Mother-to-child HIV transmission has been related to several immunological and obstetric risk factors [7,8]. Maternal virus load was included recently and was named as the most important risk factor implicated in perinatal HIV *

Corresponding author. Tel.: þ34-66-02-2-3417; fax: þ34-96-38-6-8702. E-mail address: [email protected] (V. Maiques).

transmission [9–11]. Newborn infection varied between 2 and 12% for patients below 1000 copies/ml, around 14% from 1000 to 10,000 copies/ml, 11–29% from 10,000 to 100,000 copies/ml and 40–41.7% above 100,000 copies/ml [2,12,13]. Zidovudine treatment has decreased transmission rate to 0.7% below 1000 copies/ml, 4% between 1000 and 10,000 copies/ml and 8.7% above 10,000 copies/ml [9]. Last year, highly active antiretroviral treatment (HAART) dramatically reduced maternal virus load as well as vertical transmission rate to 0.9% when it was undetectable, 2.3% below 10,000 copies/ml, 4.7% above 10,000 copies/ml and 11.6% if it was unknown [14]. Elective cesarean section was suggested as a good protective method for HIV transmission [2,15] but surgery might increase mother morbi-mortality [16], especially in the most immunocompromised women [17,18]. Amniocentesis and other invasive procedures were used to diagnose fetus infection for agents such a rubella,

0301-2115/02/$ – see front matter # 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 1 - 2 1 1 5 ( 0 2 ) 0 0 4 0 5 - 0

138

V. Maiques et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 108 (2003) 137–141

cytomegalovirus, parvovirus or toxoplasma gondii and to correlate the findings with the occurrence of fetal damage [19]. HIV has been found in fluid and cells of amniocentesis specimens of infected women [20]. Therefore, amniocentesis could help to ascertain amniotic fluid (AF) HIV contamination and the likelihood of fetal infection. The aim of our study was to assess AF HIV contamination as a marker of intrauterine HIV infection and to evaluate amniocentesis as a risk factor for vertical HIV transmission.

2. Material and methods The study was done in the Maternal Hospital ‘‘La Fe’’, Valencia (SPAIN). Between January 1984 and December 2000, 369 newborns from 366 HIV positive pregnant women were followed (three twin pairs were included in the sample). We excluded 13 children whose serostatus was unknown (eight stillborns, two perinatal dead not related with HIV infection and three lost to follow up), therefore the study group comprised 356 mother–child pairs. Women’s serostatus was determined by enzyme-linked immunosorbent assays (ELISA) and confirmed by Westernblot. Infant’s HIV infection was determined according to the international consensus definition [21]. Virus load for each sample (1 ml) was determined measuring HIV RNA using a isothermal nucleic acid amplification assay (NASBA HIV RNA QT kit; Organon-Teknica Ltd., Cambridge, UK) adapted for use with HIV [22]. There were three stages: isolation of nucleic acids from patients specimens’ and the addition of three internal RNA standards, then amplification of all RNAs with an enzyme solution containing avian myeloblastosis virus-reverse transcriptase, RNase H, and T-7 polymerase, followed by detection using the NASBA QR system electrochemiluminescence principle. The NASBA QT system calculates the amount of RNA in the original sample based on data generated from native RNA and the three synthetic internal RNA viral calibrators. The assay was carried out blinded to all other test results according to the manufacturer’s specifications. The threshold of maternal blood virus load detection was 80 copies/ml. AF samples were obtained from three well-differentiated groups of patients: (a) genetic amniocentesis (n ¼ 11), (b) amniocentesis a few days before delivery day (n ¼ 18), and (c) AF collected at delivery (n ¼ 38). The study was approved by the local ethical committee and maternal consent was obtained for every mother. Every amniocentesis technique was practiced under ultrasound control. Twenty milliliters of AF were removed for the analysis. None of the 38 amniocentesis carried out in the third trimester was made through the placenta, but we do not know if it was possible in the first trimester. Before 1997, genetic amniocentesis was performed in 11 HIV pregnant women between the 16th and 18th week of gestation (group ‘‘a’’). Two of them were under oral zidovudine treatment because of maternal requirements (one of

these children was infected and the other one was uninfected). To ascertain if there was a transmission rate increase related to the invasive procedure (amniocentesis), a retrospective cohort study was done comparing this group versus the one of patients without amniocentesis in this period of time (n ¼ 247). After 1997, a prospective study was done with ‘‘b’’ and ‘‘c’’ groups in order to identify AF HIV contamination:  Amniocentesis was performed at 37–38 weeks of gestation a few days before the planned delivery date in mothers under HAART therapy (group ‘‘b’’). Amniocentesis in the third trimester was made to determine delivery route (vaginally or cesarean section). When AF was free of virus we made a caesarian section to avoid HIV transmission, but if AF was contaminated vaginal delivery was permitted because the fetus would have already been infected. Eighteen patients were enrolled between 1997 and 1999 after obtaining an informed consent. To control if the invasive procedure was increased HIV transmission rate, we compared this group ‘‘b’’ with the patients without amniocentesis in this period of time (n ¼ 81).  Group ‘‘c’’ was performed after European Studies adopted cesarean section as a convenient route of delivery for HIV pregnant women (1999) [2,23]. AF specimens were collected during caesarian section before membrane rupture under direct vision with a syringe. Relative risk (RR) of transmission and 95% confidence interval (95% CI) were calculated for the statistical analysis. A univariate study to evaluate amniocentesis as a risk factor of HIV transmission (groups ‘‘a’’ and ‘‘b’’) was made with a non-parametric test (Chi-square or Fisher’s exact test if expected value was below 5). The significance was set P < 0:05. Groups ‘‘b’’ and ‘‘c’’ (n ¼ 56) were pooled to study the relationship between AF HIV contamination, fetal infection and maternal blood virus load at delivery (n ¼ 53). There were three maternal blood virus loads unknown because of technical problems. Furthermore, we analyzed the other perinatal risk factors for vertical HIV transmission as a gestational age at birth (above or below 37 weeks), membrane rupture duration (above or below 6 h), weight of newborn (above or below 2500 g), maternal immunity (CD4 lymphocytes above or below 500 cell/ml), and antiretroviral treatment from mother and child. Women from these two groups had been treated by antiretroviral therapy (HAART) during pregnancy and intravenous zidovudine at delivery, newborns have been received zidovudine prophylaxis the next 6 weeks after birth.

3. Results Global transmission rate in our sample was 12.9% (46 infected children from 356). Between 1984 and 1996 it was 17% (43 infected from 257). Genetic amniocentesis was

V. Maiques et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 108 (2003) 137–141

139

Fig. 1. Maternal blood virus load at amniotic fluid recollection time (‘‘b’’ and ‘‘c’’ groups), n ¼ 56 (number of cases), AF virus load was undetectable in all samples.

performed in 11 pregnant women (group ‘‘a’’) but one of the fetus died before delivery, so transmission rate in this group was 30% (3 infected from 10) versus 16.2% (40 infected from 247) in patients without amniocentesis: RR (95% CI) ¼ 1.85 (0.69–4.98) (not significant differences). The three infected children in group ‘‘a’’ were associated to other important risk factors: all were delivered vaginally, two of them without any antiretroviral treatment and with membrane rupture duration of longer than 12 h. The mother of the other one was deeply immunocompromised with 120 CD4 cell/ml. Between 1997 and 2000, transmission rate was 3% (3 infected from 99). The three infected children were born from the only three mothers without medical pregnancy control. There were no statistical differences between group ‘‘b’’ (no child infected from 18) and the group of patients who did not suffer amniocentesis (3 infected from 81). AF virus load was undetectable in every sample (n ¼ 56), and no newborn was infected. In contrast, maternal blood virus load in groups ‘‘b’’ and ‘‘c’’ (Fig. 1) ranked from undetectable to 85,000 copies/ml (40 cases below 1000 copies/ml and 12 cases above 1000 copies/ml). Maternal and obstetrical risk factors from these pairs of mother-to-

Table 1 Relevant risk factors in vertical HIV transmission (‘‘b’’ and ‘‘c’’ groups) Risk factor

n/total

Percentage

Gestational age <37 weeks Maternal CD4 <500 cell/ml Ruptured of membranes >6 h No antiretroviral treatment Newborn weight <2500 g

1/55 37/56 1/56 0/56 5/55

2 66 2 0 9

Here, n: number of cases with the risk factor.

child are showed in Table 1. Furthermore, every mother carried HAART therapy during pregnancy and delivered by elective caesarian section under intravenous zidovudine treatment.

4. Discussion Nowadays, vertical HIV transmission rate is decreasing as a result of avoiding risk factors and improving antiretroviral therapies. After 1997, it has been reduced more than 4 times in our hospital, with zero transmission rate in well-controlled mothers. Viscarello and co-workers performed different studies on AF HIV contamination as a marker of intrauterine HIV infection. First, they were looking for the p24 antigen in AF samples from 26 amniocentesis practiced in HIV pregnant women, between the 29th and the 41st week. But they found the p24 antibody, concluding that only HIV antibody was present in AF samples [24]. Afterwards, in 22 samples (between the 16th and the 24th week of gestation) they detected p24 antibody in every AF specimen and p24 antigen in 5 (38%) with a total concordance with maternal serum p24 antigen. When AF p24 antigen was positive every child was infected, but in the negative ones 11% were also infected [25]. In experimental studies with simians, after AF infection by direct virus inoculation with 200 animal infection dose of simian immunodeficiency virus (SIVmac251) five mother– infant infection pairs and two discordance pairs were found (one infected mother without child infection and conversely one infected child without mother infection) [26]. SIV mother status was determined by co-cultivation and PCR, whereas congenital infection was also documented

140

V. Maiques et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 108 (2003) 137–141

by different anti-SIV antibody response. After inoculation, six out of seven infant simians were infected, therefore AF contamination does not imply fetal-newborn infection. Fetal infection was 100% (6/6) provided that fetal SIV exposure occurs at least 19 days prior the day of birth. We have applied the new techniques for viral load detection in AF samples. None of our samples was HIV contaminated, which seems to agree with the finding that most fetal infections occur peripartum [5,27]. HAART therapy effect reducing viral load in every biological fluid should determine our result and elective caesarian section should avoid vertical transmission. Maternal viral load in plasma was undetectable in the majority of cases because of HAART therapy, but sometimes it was near to 100,000 copies/ml and undetectable in AF so they must be independent compartments. There are few studies focusing on amniocentesis in the literature, since Mundy et al. in 1987 isolated the virus from an AF sample after amniocentesis in a patient with Rh isoimmunization, and he could not rule out the possibility of a maternal blood contamination during amniocentesis [20]. But even without accurate proof, most of the authors recommended avoiding this technique because they were concerned about their theoretical HIV transmission risk. Mandelbrot et al. suggested that certain invasive procedures increase risk of transmission: amniocentesis, needling procedures, cerclage, laser therapy and amnioscopy. They reported 13 cases including amniocentesis and needling procedures [8], but unfortunately they did not provide more detailed data on this subset of patients. Other studies have estimated that a needle-stick injury of an adult carries about a 1/200 risk of subsequent seroconversion [19]. Only Tess et al. [28] had linked third trimester amniocentesis as a risk factor for vertical HIV transmission with a OR (95% IC) of 4.1(1.2–13.5), but they excluded mothers under antiretroviral treatment during pregnancy because the sample was collected between 1988 and 1993 and they had not differentiated if the placenta was perforated when the technique was performed. Viscarello et al. In a study made before the routine use of antiretroviral therapies did not find a transmission increase when amniocentesis was carried out [29,30]. Amniocentesis was evaluated taking into account the antiretroviral treatment by Shapiro et al. They had verified only a transmission rate increase when no antiretroviral therapy was used: five infected children from nine (56%) versus zero infected from five (0%) when the mother was treated [10]. In our experience, amniocentesis does not seem to increase vertical HIV transmission, especially when antiretroviral treatment is used. Before 1997, previous to routine use of HAART therapy, transmission was higher in mothers who suffered this invasive procedure as Shapiro et al. studies have pointed out [10]. We collected only 11 patients and Shapiro et al. even less, so this number is not enough to determine if other confounding risk factors were the cause of this result. After 1997, none of the children from mothers

Table 2 HIV transmission rate in women with amniocentesis versus without amniocentesis, according to different authors References

Transmission rate (infected/total)

RR (CI 95%)

Amniocentesis No amniocentesis Viscarello et al. [30] Tess et al. [28] Shapiro et al. [10] Our study (2001)

23% 40% 36% 11%

(11/48) (6/15) (5/14) (3/28)

Total

24% (25/105)

29% 15% 14% 13%

(34/117) (63/420) (54/379) (43/323)

16% (194/1239)

NS * *

NS 1.52 (1.05–2.19)*

RR (CI 95%): relative risk (confidence interval 95%). NS: not statistically significant. * Statistically significant.

with amniocentesis was infected, everyone were under HAART treatment during pregnancy and were delivered by elective caesarian section. Antiretroviral treatment must decrease viral load and virus inoculation at puncture site should be low, therefore fetal infection will be more difficult. In spite of these results about invasive procedures applied during pregnancy, in a previous study we observed an HIV transmission rate increase when intrapartum invasive procedures were used, like internal electrode or scalp fetal sample [31]. Since papers on amniocentesis in HIV pregnant women are scant and their series are limited, we have summarized their results and pooled all available data in a contingency table showing the perinatal transmission rate for women with and without amniocentesis (Table 2) [26,29,31]. Perinatal transmission rate was significantly higher in the group with amniocentesis (24%) than in the non-amniocentesis group (16%). To determine the role of antiretroviral treatment role in perinatal HIV transmission of those women who undergo an invasive procedure, again we have analyzed the cases according to the therapy used (Table 3). When amniocentesis was practised under antiretroviral treatment transmission rate was zero in this review. A significant difference seems to be between the group of mothers receiving antiretroviral treatment and the mothers without treatment. It could demonstrate that when amniocentesis is performed mother-to-child transmission rate under antiretroviral regime is lower than expected in the non-treated cases. Experimental studies have shown that fetal infection

Table 3 Mother-to-child transmission rate in patients who underwent amniocentesis as a function to be under antiretroviral treatment or not (data obtained from literature studies)

Antiretroviral treatment No treatment Fisher2a ¼ 0.001.

Infected children

Non-infected children

Transmission rate (%)

0 25

23 57

0 30.5

V. Maiques et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 108 (2003) 137–141

takes at least 19 days after virus inoculation in AF [26]. If it is the same in human fetuses, our results should be limited by the sample size (18 cases) and the time between the amniocentesis procedure and delivery day (2 days). In summary, when HAART therapies were administered amniotic fluid virus load was undetectable in our patients, even if it was detectable in maternal blood samples. This finding suggests some degree of placental barrier, and raises the possibility of performing amniocentesis when it is indicated under antiretroviral treatment for the mother.

[15]

[16]

[17]

[18]

References [19] [1] John GC, Kreiss J. Mother-to-child transmission of human immunodeficiency virus type 1. Epidemiol Rev 1996;18:149–57. [2] The European Collaborative Study. Maternal virus load and vertical transmission of HIV-1: an important factor but not the only one, AIDS 1999;13:1377–85. [3] Kind C, Rudin C, Siegrist CN, et al. Prevention of vertical HIV transmission: additive protective effect of elective Cesarean section and zidovudine prophylaxis. AIDS 1998;12:205–10. [4] Mandelbrot L, Le Chenadec J, Berrebi A, Bongain A, et al. For the French perinatal cohort. Perinatal HIV-1 transmission: interaction between zidovudine prophylaxis and mode of delivery in the French perinatal cohort. JAMA 1998;280:55–60. [5] Mock PA, Shaffer N, Bhadrakom C, et al. Maternal viral load and timing of mother-to-child HIV transmission, Bangkok, Thailand. AIDS 1999;13:407–14. [6] Brossard Y, Aubin JT, Mandelbrot L, et al. Frequency of early in utero HIV-1 infection: a blind DNA polymerase chain reaction study on 100 fetal thymuses. AIDS 1995;7:337–41. [7] European Collaborative Study. Vertical transmission of HIV-1: maternal immune status and obstetric factors, AIDS 1996;10:1675–81. [8] Mandelbrot L, Mayaux MJ, Borgain A, et al. Obstetric factors and mother-to-child transmission of human immunodeficiency virus type 1: the French perinatal cohorts. Am J Obstet Gynecol 1996;175:661–7. [9] Mofenson LM, Lambert JS, Stiehm ER. Maternal viral load and the risk of perinatal transmission of HIV-1. N Engl J Med 1999;341(22):1699 [correspondence]. [10] Shapiro DE, Sperling RS, Mandelbrot L, Britto P, Cunningham BE. For the Pediatric AIDS Clinical Trials Group Protocol 076 Study Group. Risk factors for perinatal human immunodeficiency virus transmission in patients receiving zidovudine prophylaxix. Obstet Gynecol 1999;94:897–908. [11] Shaffer N, Roongpisuthipong A, Siriwasin W, et al. Maternal viral load and perinatal HIV-1 subtype E transmission, Thailand. J Infect Dis 1999;179:590–9. [12] O’Shea S, Newel ML, Dunn DT, Maiques V, Cervera J, Bellver J, Abad A, Serra V. Maternal viral load, CD4 cell count and vertical transmission of HIV-1. J Med Virol 1998;54:113–7. [13] 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. SEROGEST Cohort Group. J Infect Dis 1997;175:172–5. [14] Samelson R, Shapiro D, Tuomala R, et al., HIV vertical transmission rates according to antiretroviral therapy and viral load during

[20]

[21]

[22] [23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

141

pregnancy among 347 mother–child pairs 1998–99 (PACTG 367), Am J Obstet Gynecol 2000;182 [abstract 276]. 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;340:977–87. Grubert TA, Reindell D, Ka¨ stner R, Lutz-Friedrich R, Belohradsky BH, Dathe O. Complications after caesarian section in HIV-1 infected women not taking antiretroviral treatment. Lancet 1999;354:1612–3. Semprini AE, Castagna C, Ravizza M, et al. The incidence of complications after caesarian section in 156 HIV-positive women. AIDS 1995;9:913–7. Maiques V, Cervera, Bellver J, Abad A, Serra V. Post-cesarean section morbidity in HIV-1 positive women, Acta Obstet Gynecol Scand 1999;78:789–92. Valente P, Sever JL. In utero diagnosis of congenital infections by direct fetal sampling. Isr J Med Sci 1994;30:414–20. Mundy DC, Schinazi RF, Russell A, Nahmias AJ, Randall HW. Human Immunodeficiency virus isolated from amniotic fluid. Lancet 1987;22:459–60. Cadwell MB, Oxtoby MJ, Simonds RJ, Lindegren ML, Rogers MF. revised classification system for human immunodeficiency virus infection in children less than 13 years of age. MMWR Morb Mortal Wkly Rep 1994;43:1–10. Compton J. Nucleic acid sequence-based amplification. Nature 1991;350:91–2. The European Mode of Delivery Collaboration. Elective caesarian section versus vaginal delivery in prevention of vertical HIV-1 transmission: a randomised clinical trial, Lancet 1999;353: 1035–39. Viscarello R.R., Prospects for prenatal diagnosis of congenital infection with HIV. In: Melica F, editor. AIDS and reproduction. Basel: Karger; 1992. p. 92–101. Viscarello RR, Cullen MT, DeGennaro NJ, Hobbins JC. Fetal blood sampling in human immunodeficiency virus-seropositive women before elective mid-trimester termination of pregnancy. Am J Obstet Gynecol 1992;167(4):1075–9. Fazely F, Sharma P, Fratazzi C, et al. Simian immunodeficiency virus infection via amniotic fluid: a model to study fetal immunopathogenesis and prophylaxis. J Acquired Immune Defic Syndr 1997;6:107– 14. Rouzioux C, Costagliola D, Burgard M, et al. The HIV infection in Newborns French Collaborative Study Group. Timing of mother-tochild transmission depends on maternal status. AIDS 1993;7(Suppl 2):s49–52. Tess BH, Rodrigues LC, Newell ML, Dunn D. Lago TDG and the Sao Paulo Collaborative Study for Vertical Transmission of HIV-1. Breastfeeding, genetic, obstetric and other risk factors associated with mother-to-child transmission of HIV-1 in Sao Paulo State, Brazil. AIDS 1998;12:513–20. Viscarello RR, DeGenaro NJ, Griffith SM, et al. Does an antepartum invasive procedure increase the risk of perinatal transmission of HIV? Am J Obstet Gynecol 1992;166:384 [abstract]. Viscarello RR, Cullen MT, DeGenaro NJ, et al. Is the rate of perinatal transmission of HIV-1 increased by amniocentesis, Am J Obstet Gynecol 1994;170 [abstract no. 1]. Maiques V, Garcı´a-Tejedor A, Perales A, Navarro C. Intrapartum fetal invasive procedures and perinatal transmission of HIV. Eur J Obstet Gynecol Reprod Biol 1999;87:63–7.