Prevalence and prognostic significance of anticardiolipin antibodies in pregnancies complicated by human immunodeficiency virus -1 infection Richard R. Viscarello, MD, CarlaJ. Williams, MD, Nancy J. DeGennaro, RN, BSN, MPH, and John C. Hobbins, MD New Haven, Connecticut OBJECTIVES: Anticardiolipin antibodies are estimated to occur in 2.2% of all pregnancies and are associated with adverse outcomes including thrombotic events, fetal wastage, intrauterine growth retardation, and preterm delivery. We studied 32 human immunodeficiency virus-seropositive gravidas (1) to determine the prevalence of anticardiolipin antibodies in pregnant women infected with human immunodeficiency virus-1 and (2) to investigate the association between the presence of anticardiolipin antibodies and pregnancy outcome, disease status, and perinatal transmission of human immunodeficiency virus-1 . STUDY DESIGN: Serum samples obtained at the first prenatal visit were analyzed for anticardiolipin immunoglobulin M and immunoglobulin G by enzyme-linked immunosorbent assay. Relevant antepartum, intrapartum, and postpartum data, including maternal CD4 + lymphocyte subsets, human immunodeficiency virus p24 antigen determinations, Venereal Disease Research Laboratory test, hematocrit, platelet counts, and placental pathologic tissue of the anticardiolipin antibody-positive and anticardiolipin antibody-negative groups were compared. RESULTS: Test results for 17 (53%) of patients were positive for anticardiolipin antibody: 4 had only immunoglobulin M, 1 had only immunoglobulin G, and the remaining 12 had both antibodies. The patients in the anticardiolipin antibody-positive group were delivered of infants with a mean gestational age of 39 weeks and mean birth weight of 2983 gm. In the anticardiolipin antibody-negative group 15 deliveries had a mean gestational age of 36.3 weeks and a mean birth weight of 2330 gm. CONCLUSIONS: We conclude that there is a high prevalence of anticardiolipin antibodies in patients who have human immunodeficiency virus, which is not associated with adverse maternal or neonatal outcome, maternal human immunodeficiency virus status, or perinatal transmission of human immunodeficiency virus-1. (AM J OSSTET GVNECOL 1992;167:1080-5.)
Key words: Anticardiolipin antibodies, human immunodeficiency virus-l infection, pregnancy, perinatal outcome
Anticardiolipin antibodies, a heterogeneous group of antibodies directed against negatively charged phospholipids, are estimated to occur in 2.2% of all pregnancies.' The prevalence of anticardiolipin antibodies is significantly higher in women with antoimmune diseases, especially systemic lupus erythematosus!' 3 A variety of adverse reproductive outcomes are associated with anticardiolipin antibody Viscarello et al. seropositivity, including recurrent pregnancy loss, thrombosis, From the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Yale University School of Medicine. Supported by grant 50034-7-PG from the Pediatric AIDS Foundation and the American Foundation for AIDS Research. Presented in part at the Twelfth Annual Meeting of the Society of Perinatal Obstetricians, Orlando, Florida, February 3-8, 1992. Reprint requests: Richard R. Viscarello, MD, Division of MaternalFetal Medicine, Department of Obstetrics and Gynecology, Yale University School of Medicine, 333 Cedar Street, P.O. Box 3333, New Haven, CT 06510. 6/6/39915
thrombocytopenia, fetal wastage, intrauterine growth retardation, and preterm delivery!' 5 Recently, significantly elevated levels of anticardiolipin antibody have been reported in human immunodeficiency virus (HIV)-seropositive homosexual and hemophiliac male subjects. 6. 8 Individuals with HIV-I do not demonstrate the same adverse clinical outcomes described for other patients with anticardiolipin antibody, nor is there a clear correlation with progression to acquired immunodeficiency syndrome (AIDS).6.'o To date there are no studies that investigate the presence and clinical significance of anticardiolipin antibodies in pregnant women with HIV-l. We conducted a prospective study of 32 pregnant women known to have HIV-I to determine the prevalence of anticardiolipin antibodies and their association with adverse outcome. In addition, the relationship between anticardiolipin antibody-seropositivity and perinatal transmission of HIV-l was investigated.
Anticardiolipin antibodies in HIV-infected pregnancies
Volume 167 Number 4, Part 1
1081
Table I. Demographic comparison of ACA-positive and ACA-negative pregnancies stratified by infant HIV status ACA-negative (N = 15) Seroreversion (n = 10) Infant
Parity Primiparous Multiparous Race Black White Hispanic Risk factor Intravenous drug abuser Heterosexual with IVDA partner Transfusion Age (yr, mean ± SEM)
J
HIV-infected (n = 5)
%
No.
3 7
30 70
8
80 10 10
No.
I
1
7 70 3 30 0 0 29.9 ± 4.6
ACA-positive (N = 17)
I
Seroreversion (n = 13)
%
No.
0 5
0 100
3 2 0
60 40 0
3 60 40 2 0 0 31.4 ± 4.2
I
HIV-infected (n = 4)
I
%
No.
4 9
30.8 69.2
1 3
25 75
8 2 3
61.5 15.4 23.1
3 1 0
75 25 0
6 46.2 6 46.2 1 7.7 26.2 ± 6.1
%
50 2 50 2 0 0 23.5 ± 4.1
ACA, Anticardiolipin antibody; IVDA, intravenous drug abuser.
Material and methods
Patients. All patients were under the care of two of the investigators (R.R.Y., N.J.D.) in the Women and AIDS Program in the Women's Center, Yale-New Haven Hospital. The children are cared for by Warren A. Andiman, MD, in the Pediatric AIDS Program at our institution. Clinical disease status in the women and children was determined according to the 1987 Centers for Disease Control (CDC) classification. The 32 patients reported in this study are part of a larger, ongoing, prospective study of 70 pregnant women with HIV-l who have been tested for anticardiolipin antibodies. Seven of these patients elected to terminate their pregnancies. There were no spontaneous abortions in this group. The remaining 31 patients were either delivered of viable infants who were < 15 months old (n = 21; these patients constitute CDC group PO) or were still undelivered at this writing (n = 10). These 31 patients have been excluded from this report. Our study was approved by the Human Investigations Committee of Yale University School of Medicine, and informed consent was obtained from the women. Clinical histories and follow-up for the women and their offspring were uniformly available. Relevant antepartum, intrapartum, and postpartum data, including maternal CD4 + lymphocyte subsets, HIV p24 antigen determinations, Venereal Disease Research Laboratory test results, hematocrit, platelet counts, and placental pathologic tissue of patients with and without anticardiolipin antibodies were obtained in most cases. HIV-l seropositivity was determined by enzyme-linked immunosorbent assay (Abbott Laboratories, N. Chicago, Ill.) and confirmed by Western blot analysis in all patients. ll
Anticardiolipin antibody serologic findings. Sera were collected at the first prenatal visit from 32 pregnant women with HIV; the samples were placed on ice and centrifuged at 5000 rpm for 10 minutes. The supernatant was frozen at - 70° C and stored. Anticardiolipin antibodies were isolated by enzyme-linked immunosorbent assay, as previously described. '2 Microtiter plates (Fisher Scientific, Springfield, N.J.) coated with 1 I-lg of bovine cardiolipin in ethanol (Sigma Chemical, St. Louis) were air evaporated in a cold room overnight. Both the microtiter plates and the uncoated plates were blocked with fetal calf serum. The wells were incubated sequentially with test or control serum, "{- or j-l-chain, goat antihuman immunoglobulin (Ig) G, and "{-chain rabbit, antigoat IgG conjugated with alkaline phosphatase (Sigma Chemical). Between incubations the plates were washed with phosphate-buffered saline solution (Titertek washer, Flow Laboratories, McLean, Va.). After the antibody reactions the plates were incubated with 4-p-nitrophenyl phosphate for 1 hour in a dark room. A microtiter plate reader (Multiskan Plus, Flow Laboratories) was used to read the colorimetric reaction at 405 nm. Serial dilutions of positive control serum containing known concentrations of IgM or IgG anticardiolipin antibodies were read at an optical density of 405 nm; the readings were ,used to construct a standard linear curve. All elevated samples were repeated, and only those samples elevated in two separate assays were considered significant. Results
There were 24 multigravid women and eight primigravid women, with a mean age of 27.8 years. The racial composition of the women was 22 black, 4 His-
1082 Viscarello et al.
October 1992 Am J Obstet Gynecol
Table II. Indicators of HIV disease status in ACA-positive and ACA-negative women stratified by infant HIV status ACA-negative (N = 15)
Seroreversion (n = 10) Infant
CDC group II III
IV p24 Antigen Positive Negative NA Mean CD4+ (cells/ml) Mean hematocrit (%) Mean platelet count per 100,000
l
H1V-infected (n = 5)
%
No.
6 3 1
60 30 10
4
o
0
o
No.
5 50 5 50 271 ± 247 34.8 ± 4.3 207 ± 143
ACA -positive (N = 17)
1
o
I
Seroreversion (n = 13)
%
No.
80 20
10 3
0 0
2 40 3 60 400 ± 178 31.7 ± 2.9 224 ± 132
o
I
%
No.
76.9 23.1 0
o
7.7 61.5 4 30.8 393 ± 195 31.5 ± 4.6 249 ± 128 1
8
H1V-infected (n = 4)
I
%
2
50
2
50
2
50
2
50
o
0
0
387 ± 67 28.3 ± 2.7 205 ± 184
ACA, Anticardiolipin antibody.
panic, and 6 white. Eighteen women were intravenous drug abusers and 13 were heterosexual partners of intravenous drug abusers; one woman contracted HIV1 via blood transfusion (Table I). Twenty-two patients were in CDC group II, 7 were in CDC group III, and 3 were in CDC group IV (patients with AIDS). Neither maternal HIV-risk behavior nor disease status was correlated with anticardiolipin antibody status. The mean maternal CD4 + lymphocyte counts, hemocrits, and platelet counts stratified by maternal anticardiolipin antibody status and infant HIV outcome are given in Table II. There were no positive HIV p24 antigen determinations among the seven anticardiolipin antibody-negative women tested. There were three positive HIV p24 antigen determinations among the women with anticardiolipin antibodies, two in infected children and one in a person with seroreversion. Seventeen (53%) of the 32 patients had anticardiolipin antibodies: both IgM and IgG antibodies were detected in the sera of 12, 4 had IgM only, and 1 had IgG only. Only 4 patients had positive Venereal Disease Research Laboratory serologic test results: two patients in each of the anticardiolipin antibody-positive and anticardiolipin antibody-negative groups. Two of the positive Venereal Disease Research Laboratory test results were confirmed with a positive fluorescent titer antibody test, one in each anticardiolipin antibody group. There was no significant difference between the groups with respect to the incidence of hypertension, thrombotic events, preterm labor, premature rupture of the membranes, intrauterine growth retardation, or placental pathologic tissue. This study had a power ranging from 2% to 52% to detect a statistically significant difference between the anticardiolipin antibody-
negative and anticardiolipin antibody-positive groups with respect to these adverse pregnancy outcomes. The 17 patients in the anticardiolipin antibody-positive group delivered infants at a mean gestational age of 39 weeks with a mean birth weight of 2983 gm. Fourteen had normal spontaneous vaginal deliveries, 2 had cesarean sections, and 1 had a low-forceps delivery. In the anticardiolipin antibody-negative group, 15 newborns had a mean gestational age of 36 weeks and a mean birth weight of 2330 gm. Ten of the deliveries were normal spontaneous vaginal deliveries, 4 women had cesarean sections, and 2 had low-forceps deliveries. Table III illustrates the delivery data stratified by maternal anticardiolipin antibody status and newborn HIV outcome according to CDC guidelines. This study had a power of 2% and 61 % to detect a significant difference between the anticardiolipin antibody-negative and anticardiolipin antibody-positive groups with respect to birth weight and gestational age at delivery, respectively. Of the 17 anticardiolipin antibody-positive infants, 13 have seroreverted and 4 are CDC group P2 (symptomatic infection), including 1 who died (Table IV). Among the 15 infants of mothers without anticardiolipin antibody, 10 have seroreverted, 4 are CDC group PI (asymptomatic infection; of these 4, I is antigen positive), and 1 is CDC group P2. No significant difference was noted in antibody-banding patterns on Western blot analysis among women with and without anticardiolipin antibodies, even when mothers of infected infants were compared with mothers of uninfected infants. Comment
Elevated levels of anticardiolipin antibodies have been described in association with certain chronic ill-
Anticardiolipin antibodies in HIV-infected pregnancies
Volume 167 Number 4, Part 1
1083
Table III. Delivery data for ACA-positive and ACA-negative pregnancies stratified by infant HIV status ACA-negative (N = 15) S eroreversion (n = 10) Infant
Delivery NSVD Cesarean section Low forceps Gestational age (wk) Weight (gm)
No.
7 3
o
I
HIV-infected (n = 5)
%
No.
70 30
3
36.6 ± 5.1 2343 ± 970
ACA-positive (N = 17)
0
1 1
I
S eroreversion (n = 13)
%
No.
60 20 20
10
36.0 ± 4.9 2317 ± 829
I
HIV-infected (n = 4)
%
No.
76.9 15.4 1 7.7 39.0 ± 2.4 3033 ± 400
2
I
%
4
100 0 0
o o
38.8 ± 1.9 2933 ± 308
ACA, Anticardiolipin antibody; NSVD, normal spontaneous vaginal delivery.
nesses, including rheumatoid arthritis, Sjogren's syndrome, autoimmune thrombocytopenia, tuberculosis, and leprosy.7. 13 Women with recurrent pregnancy loss demonstrate a 13% to 40% anticardiolipin antibody prevalence l ; Harris et al. 13 (1985) noted a prevalence of 61 % in women with systemic lupus erythematosus.' The prevalence of the anticardiolipin antibody in our study group of pregnant women infected with HIV-l was 17 (53%) of 32, a significant increase over the 2.2% prevalence reported for the general obstetric population. I This increased rate is, however, consistent with several recent studies that report anticardiolipin antibody prevalence rates from 50% to 92% among men who are homosexual or who have hemophilia and are infected with HIV-1.6.10 To assess the possibility of a biologic false-positive anticardiolipin antibody result because of a reactive test for syphilis,14 we analyzed Venereal Disease Research Laboratory test titers that were available in 30 of the 32 women. Only four test results were positive: 2 in women without anticardiolipin antibodies and 2 in women with anticardiolipin antibodies. In one patient in each anticardiolipin antibody group, Venereal Disease Research Laboratory test reactivity was confirmed by fluorescent titer antibody, and the test results of the sera of the other two women were negative for fluorescent titer antibody. In spite of the reported increased risk for adverse pregnancy outcomes associated with the presence of anticardiolipin antibodies during pregnancy (as much as 75% in some studies I), no significant complications occurred in our cohort of women with HIV-l. Harris et al .. (1986) have shown that the IgG subtype level was predictive for fetal loss and thrombosis 4; however, we did not see a correlation with either IgG or IgM subtypes because the majority of women had both subtypes. The mean platelet count in patients with and without anticardiolipin antibodies was similar, >200,000 in all patients. Placental pathologic tissue samples were available in 15 of 32 deliveries. The in-
Table IV. Western blot analysis patterns in ACA-positive and ACA-negative mothers stratified by infant HIV status
Western blot pattern
p17 Absent Present p24 Absent Present p31 Absent Present gp41 Absent Present p51 Absent Present p55 Absent Present p66 Absent Present gp120 Absent Present gp160 Absent Present
ACA-negative (N = 15)
ACA -positive (N = 17)
~ Infected
I
Infected (n = 4)
(n=5)
Seroreversion (n = 13)
2 3
1 2
2 6
0 3
1 4
0 3
0 9
0 3
0 5
2 1
0 9
0 3
0 5
0 3
1 8
0
1 4
0 3
1 7
0 3
0
0 3
1
5
7
0 3
1 4
2
1
1 7
0 3
0
0
2 7
0 4
0 5
3
0
2 7
0 4
Seroreversion (n=10)
5
3
3
ACA, Anticardiolipin antibody.
cidence of clinically significant abruptio placentae was the same in both groups, with two of three cases associated with documented cocaine use within 24 hours of delivery. The presence of anticardiolipin antibodies in our study group of women with HIV-I did not correlate with maternal disease status as evaluated clinically or by CD4 count, HIV-l p24 antigen seropositivity, he-
1084 Viscarello et al.
matocrit, or platelet count, nor was it associated with maternal age, parity, or HIV risk behavior. Overall, 23 (72%) of the 32 infants have had seroreversion. Thus the perinatal transmission rate is 28%, a rate consistent with previously published estimates. I5 The rate of transmission between the two groups was similar, with 23% transmission in the anticardiolipin antibody-positive and 33% in the anticardiolipin antibody-negative group. Our study had a power of 14% to detect a significant difference between the anticardiolipin antibody-negative and anticardiolipin antibody-positive groups with respect to perinatal transmission rate. It does not appear that a positive anticardiolipin antibody confers protective immunity to the fetus, even though other types of adverse outcomes that could be immune mediated do not occur. Of note, 4 of the 5 CDC group P2 infants were born to mothers with anticardiolipin antibodies, whereas 4 of 5 delivered by mothers without anticardiolipin antibodies are in CDC group PI. Whether the presence of anticardiolipin antibody could be an early marker of newborns at risk for a more aggressive clinical course remains to be determined. However, those infants born to mothers with anticardiolipin antibodies who escaped infection with HIV-l seem to have larger birth weights and longer mean lengths of gestation than infants born to women without anticardiolipin antibodies, although the difference is not statistically significant (power = 18%). Several abnormalities in the humoral component of the immune system have been described in patients with AIDS, including a polyclonal increase in serum immunoglobulins and the presence of antilymphocyte antibodies, antisperm antibodies, rheumatoid factor, and lupus anticoagulant. 9 Whether the increased prevalence of anticardiolipin antibodies in individuals with HIV-l represents a nonspecific immunostimulation to a foreign antigen,7 a cross-reaction to other phospholipids such as phosphatidic acid or phosphatidylglycerol or even to deoxyribonucleic acid is controversial. 16 However, recent work by Canoso et al. 6 supports the existence of independent classes of anticardiolipin antibody in patients with HIV-l infection. The possibility also exists that viral destruction of the cell membrane exposes or modifies phospholipids such that they become more antigenic. 1O Several studies have demonstrated that absence of antibodies to epitopes in the principal neutralizing domain (V3 loop) of gp120 is associated with increased transmission of HIV -1. 17, 18 Goedert et al. 19 (1989) reported that mothers who did not transmit HIV-l infection were shown to possess high titer-high affinity antibodies against gp120, whereas transmitting mothers had low or undetectable gp120 levels as measured by radioimmunoassay or immunoblot. 19 The humoral immune response has been suggested to be an impor-
October 1992 Am J Obstet Gynecol
tant component of protective immunity against infectious agents. Although there was no statistically significant correlation between the presence of anticardiolipin antibodies and specific HIV-l banding patterns by Western blot analysis, or between HIV -1 banding patterns by Western blot analysis and HIV-l transmission from mother to child, it is interesting to note that 3 of 4 infants infected with HIV born to mothers with anticardiolipin antibodies reacted to all bands, The fourth infant reacted only to two bands, gp120 and gp160. Our preliminary data suggest that pregnant women infected with HIV -1 exhibit a high prevalence of anticardiolipin antibodies, which are not related to maternal disease status, adverse pregnancy outcome, or perinatal transmission of HIV -1. However, future studies, including studies with larger numbers of patients, will be required to address the issue of anticardiolipin antibodies as predictors of disease progression or perinatal transmission, REFERENCES 1. Lockwood C], Romero R, Feinberg RF, Clyne LP, Coster B, Hobbins ]C, The prevalence and biologic significance of lupus anticoagulant and anticardiolipin antibodies in a general obstetric population. AM ] OBSTET GVNECOL 1989; 161 :369-73, 2, Love PE, Santoro SA, Antiphospholipid antibodies: anticardiolipin and the lupus anticoagulant in systemic lupus erythematosus (SLE) and non-SLE disorders. Ann Intern Med 1990; 112:682-98. 3. Harris EN, Gharavi AE, Hughes GR. Anticardioplin antibodies. Clin Rheumatol Dis 1985;11:591-609. 4. Harris EN, Chan ]KH, Asherton RA, Aber VR, Gharavi AE, Hughes GRY. Thrombosis, recurrent fetal loss, and thrombocytopenia: predictive value of the anticardiolipin antibody test. Arch Intern Med 1986;146:2153-6. 5. Lockshin MD, Druzin ML, Goei S, et al. Antibody to cardiolipin as a predictor of fetal distress or death in pregnant patients with systemic lupus erythematosus. N Engl ] Med 1985;313:152-6. 6. Canoso RT, Zon LI, Groopman ]E. Anticardiolipin antibodies associated with HTLV-III infection. Br] Haematol 1987;65:495-8. 7, Panzer S, Stain C, Hartl H, Dudczak R, Lechner K. Anticardiolipin antibodies are elevated in HIV-l infected haemophiliacs but do not predict for disease progression, Thromb Haemost 1989;61:81-5. 8. Mulhall BP, NaseIli G, Whittingham S. Anticardiolipin antibodies in homosexual men: prevalence and lack of association with human immunodeficiency virus (HIV) infection.] Clin Immunol 1989;9:208-13, 9. Stimmler MM, Quismorio FP, McGehee WG, Boylen T, Sharma OP. Anticardiolipin antibodies in the acquired immunodeficiency syndrome. Arch Intern Med 1989; 149: 1833-5. 10. Cohen H, Mackie I], Anagnostopoulos N, Savage GF, Machin SJ. Lupus anticoagulant, anticardiolipin antibodies, and human immunodeficiency virus in haemophilia. ] Clin Pathol 1989;42:629-33. 11. Martin K, Katz B, Miller G. AIDS and antibodies to human immunodeficiency virus (HIV) in children and their families.] Infect Dis 1987;155:54-9. 12. Cowchuck S, DeHoratius RD, Wapner R], Jackson LG. Subclinical autoimmune disease and unexplained abortion, AM] OBSTET GVNECOL 1984;150:367-71. 13, Harris EN, Asherton RR, Gharavi AE. Thrombocyto-
Volume 167 Number 4, Part 1
penia in SLE and related disorders: association with anticardiolipin antibodies. Br J Haematol 1985;59:227-30. 14. Koike T, Sueishi M, Funaki H, Tomioka H, Yoshida S. Antiphospholipid antibodies and biologic false-positive serologic test for syphilis in patients with systemic lupus erythematosus. Clin Exp Immunol 1984;56: 193-9. 15. Andiman WA, Simpson BJ, Olson B, Dember L, Saliva T, Miller G. Rate of transmission of HIV-l infection from mother to child and short-term outcome of neonatal infection: Results of a prospective cohort study. Am J Dis Child 1990:144:758-66. 16. Rauch J, Tannenbaum H, Stollar BD. Monoclonal anticardiolipin antibodies bind to DNA. Eur J Immunol 1984; 14:529-34.
Anticardiolipin antibodies in HIV-infected pregnancies
1085
17. Rossi P, Moschese V, Broliden PA, et al. Presence of maternal antibodies to human immunodeficiency virus 1 envelope glycoprotein gp120 epitopes correlates with the uninfected status of children born to seropositive mothers. Proc Nat! Acad Sci USA 1989;86:8055-8. 18. Devash Y, Calvelli TA, Wood DG, Reagan KJ, Rubenstein A. Vertical transmission of human immunodeficiency virus is correlated with the absence of high-affinity / avidity maternal antibodies to the gp120 principal neutralizing domain. Proc Nat! Acad Sci USA 1990;87:3445-9. 19. Goedert J, Drummond J, Minkoff H, et al. Mother to infant transmission of human immunodeficiency virus type 1: association with prematurity or low anti-gpI20. Lancet 1989;2:1351-4.
Key words: Anticardiolipin antibodies, human immunodeficiency virus-l infection, pregnancy, perinatal outcome
Anticardiolipin antibodies, a heterogeneous group of antibodies directed against negatively charged phospholipids, are estimated to occur in 2.2% of all pregnancies.' The prevalence of anticardiolipin antibodies is significantly higher in women with antoimmune diseases, especially systemic lupus erythematosus!' 3 A variety of adverse reproductive outcomes are associated with anticardiolipin antibody Viscarello et al. seropositivity, including recurrent pregnancy loss, thrombosis, From the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Yale University School of Medicine. Supported by grant 50034-7-PG from the Pediatric AIDS Foundation and the American Foundation for AIDS Research. Presented in part at the Twelfth Annual Meeting of the Society of Perinatal Obstetricians, Orlando, Florida, February 3-8, 1992. Reprint requests: Richard R. Viscarello, MD, Division of MaternalFetal Medicine, Department of Obstetrics and Gynecology, Yale University School of Medicine, 333 Cedar Street, P.O. Box 3333, New Haven, CT 06510. 6/6/39915
thrombocytopenia, fetal wastage, intrauterine growth retardation, and preterm delivery!' 5 Recently, significantly elevated levels of anticardiolipin antibody have been reported in human immunodeficiency virus (HIV)-seropositive homosexual and hemophiliac male subjects. 6. 8 Individuals with HIV-I do not demonstrate the same adverse clinical outcomes described for other patients with anticardiolipin antibody, nor is there a clear correlation with progression to acquired immunodeficiency syndrome (AIDS).6.'o To date there are no studies that investigate the presence and clinical significance of anticardiolipin antibodies in pregnant women with HIV-l. We conducted a prospective study of 32 pregnant women known to have HIV-I to determine the prevalence of anticardiolipin antibodies and their association with adverse outcome. In addition, the relationship between anticardiolipin antibody-seropositivity and perinatal transmission of HIV-l was investigated.
Anticardiolipin antibodies in HIV-infected pregnancies
Volume 167 Number 4, Part 1
1081
Table I. Demographic comparison of ACA-positive and ACA-negative pregnancies stratified by infant HIV status ACA-negative (N = 15) Seroreversion (n = 10) Infant
Parity Primiparous Multiparous Race Black White Hispanic Risk factor Intravenous drug abuser Heterosexual with IVDA partner Transfusion Age (yr, mean ± SEM)
J
HIV-infected (n = 5)
%
No.
3 7
30 70
8
80 10 10
No.
I
1
7 70 3 30 0 0 29.9 ± 4.6
ACA-positive (N = 17)
I
Seroreversion (n = 13)
%
No.
0 5
0 100
3 2 0
60 40 0
3 60 40 2 0 0 31.4 ± 4.2
I
HIV-infected (n = 4)
I
%
No.
4 9
30.8 69.2
1 3
25 75
8 2 3
61.5 15.4 23.1
3 1 0
75 25 0
6 46.2 6 46.2 1 7.7 26.2 ± 6.1
%
50 2 50 2 0 0 23.5 ± 4.1
ACA, Anticardiolipin antibody; IVDA, intravenous drug abuser.
Material and methods
Patients. All patients were under the care of two of the investigators (R.R.Y., N.J.D.) in the Women and AIDS Program in the Women's Center, Yale-New Haven Hospital. The children are cared for by Warren A. Andiman, MD, in the Pediatric AIDS Program at our institution. Clinical disease status in the women and children was determined according to the 1987 Centers for Disease Control (CDC) classification. The 32 patients reported in this study are part of a larger, ongoing, prospective study of 70 pregnant women with HIV-l who have been tested for anticardiolipin antibodies. Seven of these patients elected to terminate their pregnancies. There were no spontaneous abortions in this group. The remaining 31 patients were either delivered of viable infants who were < 15 months old (n = 21; these patients constitute CDC group PO) or were still undelivered at this writing (n = 10). These 31 patients have been excluded from this report. Our study was approved by the Human Investigations Committee of Yale University School of Medicine, and informed consent was obtained from the women. Clinical histories and follow-up for the women and their offspring were uniformly available. Relevant antepartum, intrapartum, and postpartum data, including maternal CD4 + lymphocyte subsets, HIV p24 antigen determinations, Venereal Disease Research Laboratory test results, hematocrit, platelet counts, and placental pathologic tissue of patients with and without anticardiolipin antibodies were obtained in most cases. HIV-l seropositivity was determined by enzyme-linked immunosorbent assay (Abbott Laboratories, N. Chicago, Ill.) and confirmed by Western blot analysis in all patients. ll
Anticardiolipin antibody serologic findings. Sera were collected at the first prenatal visit from 32 pregnant women with HIV; the samples were placed on ice and centrifuged at 5000 rpm for 10 minutes. The supernatant was frozen at - 70° C and stored. Anticardiolipin antibodies were isolated by enzyme-linked immunosorbent assay, as previously described. '2 Microtiter plates (Fisher Scientific, Springfield, N.J.) coated with 1 I-lg of bovine cardiolipin in ethanol (Sigma Chemical, St. Louis) were air evaporated in a cold room overnight. Both the microtiter plates and the uncoated plates were blocked with fetal calf serum. The wells were incubated sequentially with test or control serum, "{- or j-l-chain, goat antihuman immunoglobulin (Ig) G, and "{-chain rabbit, antigoat IgG conjugated with alkaline phosphatase (Sigma Chemical). Between incubations the plates were washed with phosphate-buffered saline solution (Titertek washer, Flow Laboratories, McLean, Va.). After the antibody reactions the plates were incubated with 4-p-nitrophenyl phosphate for 1 hour in a dark room. A microtiter plate reader (Multiskan Plus, Flow Laboratories) was used to read the colorimetric reaction at 405 nm. Serial dilutions of positive control serum containing known concentrations of IgM or IgG anticardiolipin antibodies were read at an optical density of 405 nm; the readings were ,used to construct a standard linear curve. All elevated samples were repeated, and only those samples elevated in two separate assays were considered significant. Results
There were 24 multigravid women and eight primigravid women, with a mean age of 27.8 years. The racial composition of the women was 22 black, 4 His-
1082 Viscarello et al.
October 1992 Am J Obstet Gynecol
Table II. Indicators of HIV disease status in ACA-positive and ACA-negative women stratified by infant HIV status ACA-negative (N = 15)
Seroreversion (n = 10) Infant
CDC group II III
IV p24 Antigen Positive Negative NA Mean CD4+ (cells/ml) Mean hematocrit (%) Mean platelet count per 100,000
l
H1V-infected (n = 5)
%
No.
6 3 1
60 30 10
4
o
0
o
No.
5 50 5 50 271 ± 247 34.8 ± 4.3 207 ± 143
ACA -positive (N = 17)
1
o
I
Seroreversion (n = 13)
%
No.
80 20
10 3
0 0
2 40 3 60 400 ± 178 31.7 ± 2.9 224 ± 132
o
I
%
No.
76.9 23.1 0
o
7.7 61.5 4 30.8 393 ± 195 31.5 ± 4.6 249 ± 128 1
8
H1V-infected (n = 4)
I
%
2
50
2
50
2
50
2
50
o
0
0
387 ± 67 28.3 ± 2.7 205 ± 184
ACA, Anticardiolipin antibody.
panic, and 6 white. Eighteen women were intravenous drug abusers and 13 were heterosexual partners of intravenous drug abusers; one woman contracted HIV1 via blood transfusion (Table I). Twenty-two patients were in CDC group II, 7 were in CDC group III, and 3 were in CDC group IV (patients with AIDS). Neither maternal HIV-risk behavior nor disease status was correlated with anticardiolipin antibody status. The mean maternal CD4 + lymphocyte counts, hemocrits, and platelet counts stratified by maternal anticardiolipin antibody status and infant HIV outcome are given in Table II. There were no positive HIV p24 antigen determinations among the seven anticardiolipin antibody-negative women tested. There were three positive HIV p24 antigen determinations among the women with anticardiolipin antibodies, two in infected children and one in a person with seroreversion. Seventeen (53%) of the 32 patients had anticardiolipin antibodies: both IgM and IgG antibodies were detected in the sera of 12, 4 had IgM only, and 1 had IgG only. Only 4 patients had positive Venereal Disease Research Laboratory serologic test results: two patients in each of the anticardiolipin antibody-positive and anticardiolipin antibody-negative groups. Two of the positive Venereal Disease Research Laboratory test results were confirmed with a positive fluorescent titer antibody test, one in each anticardiolipin antibody group. There was no significant difference between the groups with respect to the incidence of hypertension, thrombotic events, preterm labor, premature rupture of the membranes, intrauterine growth retardation, or placental pathologic tissue. This study had a power ranging from 2% to 52% to detect a statistically significant difference between the anticardiolipin antibody-
negative and anticardiolipin antibody-positive groups with respect to these adverse pregnancy outcomes. The 17 patients in the anticardiolipin antibody-positive group delivered infants at a mean gestational age of 39 weeks with a mean birth weight of 2983 gm. Fourteen had normal spontaneous vaginal deliveries, 2 had cesarean sections, and 1 had a low-forceps delivery. In the anticardiolipin antibody-negative group, 15 newborns had a mean gestational age of 36 weeks and a mean birth weight of 2330 gm. Ten of the deliveries were normal spontaneous vaginal deliveries, 4 women had cesarean sections, and 2 had low-forceps deliveries. Table III illustrates the delivery data stratified by maternal anticardiolipin antibody status and newborn HIV outcome according to CDC guidelines. This study had a power of 2% and 61 % to detect a significant difference between the anticardiolipin antibody-negative and anticardiolipin antibody-positive groups with respect to birth weight and gestational age at delivery, respectively. Of the 17 anticardiolipin antibody-positive infants, 13 have seroreverted and 4 are CDC group P2 (symptomatic infection), including 1 who died (Table IV). Among the 15 infants of mothers without anticardiolipin antibody, 10 have seroreverted, 4 are CDC group PI (asymptomatic infection; of these 4, I is antigen positive), and 1 is CDC group P2. No significant difference was noted in antibody-banding patterns on Western blot analysis among women with and without anticardiolipin antibodies, even when mothers of infected infants were compared with mothers of uninfected infants. Comment
Elevated levels of anticardiolipin antibodies have been described in association with certain chronic ill-
Anticardiolipin antibodies in HIV-infected pregnancies
Volume 167 Number 4, Part 1
1083
Table III. Delivery data for ACA-positive and ACA-negative pregnancies stratified by infant HIV status ACA-negative (N = 15) S eroreversion (n = 10) Infant
Delivery NSVD Cesarean section Low forceps Gestational age (wk) Weight (gm)
No.
7 3
o
I
HIV-infected (n = 5)
%
No.
70 30
3
36.6 ± 5.1 2343 ± 970
ACA-positive (N = 17)
0
1 1
I
S eroreversion (n = 13)
%
No.
60 20 20
10
36.0 ± 4.9 2317 ± 829
I
HIV-infected (n = 4)
%
No.
76.9 15.4 1 7.7 39.0 ± 2.4 3033 ± 400
2
I
%
4
100 0 0
o o
38.8 ± 1.9 2933 ± 308
ACA, Anticardiolipin antibody; NSVD, normal spontaneous vaginal delivery.
nesses, including rheumatoid arthritis, Sjogren's syndrome, autoimmune thrombocytopenia, tuberculosis, and leprosy.7. 13 Women with recurrent pregnancy loss demonstrate a 13% to 40% anticardiolipin antibody prevalence l ; Harris et al. 13 (1985) noted a prevalence of 61 % in women with systemic lupus erythematosus.' The prevalence of the anticardiolipin antibody in our study group of pregnant women infected with HIV-l was 17 (53%) of 32, a significant increase over the 2.2% prevalence reported for the general obstetric population. I This increased rate is, however, consistent with several recent studies that report anticardiolipin antibody prevalence rates from 50% to 92% among men who are homosexual or who have hemophilia and are infected with HIV-1.6.10 To assess the possibility of a biologic false-positive anticardiolipin antibody result because of a reactive test for syphilis,14 we analyzed Venereal Disease Research Laboratory test titers that were available in 30 of the 32 women. Only four test results were positive: 2 in women without anticardiolipin antibodies and 2 in women with anticardiolipin antibodies. In one patient in each anticardiolipin antibody group, Venereal Disease Research Laboratory test reactivity was confirmed by fluorescent titer antibody, and the test results of the sera of the other two women were negative for fluorescent titer antibody. In spite of the reported increased risk for adverse pregnancy outcomes associated with the presence of anticardiolipin antibodies during pregnancy (as much as 75% in some studies I), no significant complications occurred in our cohort of women with HIV-l. Harris et al .. (1986) have shown that the IgG subtype level was predictive for fetal loss and thrombosis 4; however, we did not see a correlation with either IgG or IgM subtypes because the majority of women had both subtypes. The mean platelet count in patients with and without anticardiolipin antibodies was similar, >200,000 in all patients. Placental pathologic tissue samples were available in 15 of 32 deliveries. The in-
Table IV. Western blot analysis patterns in ACA-positive and ACA-negative mothers stratified by infant HIV status
Western blot pattern
p17 Absent Present p24 Absent Present p31 Absent Present gp41 Absent Present p51 Absent Present p55 Absent Present p66 Absent Present gp120 Absent Present gp160 Absent Present
ACA-negative (N = 15)
ACA -positive (N = 17)
~ Infected
I
Infected (n = 4)
(n=5)
Seroreversion (n = 13)
2 3
1 2
2 6
0 3
1 4
0 3
0 9
0 3
0 5
2 1
0 9
0 3
0 5
0 3
1 8
0
1 4
0 3
1 7
0 3
0
0 3
1
5
7
0 3
1 4
2
1
1 7
0 3
0
0
2 7
0 4
0 5
3
0
2 7
0 4
Seroreversion (n=10)
5
3
3
ACA, Anticardiolipin antibody.
cidence of clinically significant abruptio placentae was the same in both groups, with two of three cases associated with documented cocaine use within 24 hours of delivery. The presence of anticardiolipin antibodies in our study group of women with HIV-I did not correlate with maternal disease status as evaluated clinically or by CD4 count, HIV-l p24 antigen seropositivity, he-
1084 Viscarello et al.
matocrit, or platelet count, nor was it associated with maternal age, parity, or HIV risk behavior. Overall, 23 (72%) of the 32 infants have had seroreversion. Thus the perinatal transmission rate is 28%, a rate consistent with previously published estimates. I5 The rate of transmission between the two groups was similar, with 23% transmission in the anticardiolipin antibody-positive and 33% in the anticardiolipin antibody-negative group. Our study had a power of 14% to detect a significant difference between the anticardiolipin antibody-negative and anticardiolipin antibody-positive groups with respect to perinatal transmission rate. It does not appear that a positive anticardiolipin antibody confers protective immunity to the fetus, even though other types of adverse outcomes that could be immune mediated do not occur. Of note, 4 of the 5 CDC group P2 infants were born to mothers with anticardiolipin antibodies, whereas 4 of 5 delivered by mothers without anticardiolipin antibodies are in CDC group PI. Whether the presence of anticardiolipin antibody could be an early marker of newborns at risk for a more aggressive clinical course remains to be determined. However, those infants born to mothers with anticardiolipin antibodies who escaped infection with HIV-l seem to have larger birth weights and longer mean lengths of gestation than infants born to women without anticardiolipin antibodies, although the difference is not statistically significant (power = 18%). Several abnormalities in the humoral component of the immune system have been described in patients with AIDS, including a polyclonal increase in serum immunoglobulins and the presence of antilymphocyte antibodies, antisperm antibodies, rheumatoid factor, and lupus anticoagulant. 9 Whether the increased prevalence of anticardiolipin antibodies in individuals with HIV-l represents a nonspecific immunostimulation to a foreign antigen,7 a cross-reaction to other phospholipids such as phosphatidic acid or phosphatidylglycerol or even to deoxyribonucleic acid is controversial. 16 However, recent work by Canoso et al. 6 supports the existence of independent classes of anticardiolipin antibody in patients with HIV-l infection. The possibility also exists that viral destruction of the cell membrane exposes or modifies phospholipids such that they become more antigenic. 1O Several studies have demonstrated that absence of antibodies to epitopes in the principal neutralizing domain (V3 loop) of gp120 is associated with increased transmission of HIV -1. 17, 18 Goedert et al. 19 (1989) reported that mothers who did not transmit HIV-l infection were shown to possess high titer-high affinity antibodies against gp120, whereas transmitting mothers had low or undetectable gp120 levels as measured by radioimmunoassay or immunoblot. 19 The humoral immune response has been suggested to be an impor-
October 1992 Am J Obstet Gynecol
tant component of protective immunity against infectious agents. Although there was no statistically significant correlation between the presence of anticardiolipin antibodies and specific HIV-l banding patterns by Western blot analysis, or between HIV -1 banding patterns by Western blot analysis and HIV-l transmission from mother to child, it is interesting to note that 3 of 4 infants infected with HIV born to mothers with anticardiolipin antibodies reacted to all bands, The fourth infant reacted only to two bands, gp120 and gp160. Our preliminary data suggest that pregnant women infected with HIV -1 exhibit a high prevalence of anticardiolipin antibodies, which are not related to maternal disease status, adverse pregnancy outcome, or perinatal transmission of HIV -1. However, future studies, including studies with larger numbers of patients, will be required to address the issue of anticardiolipin antibodies as predictors of disease progression or perinatal transmission, REFERENCES 1. Lockwood C], Romero R, Feinberg RF, Clyne LP, Coster B, Hobbins ]C, The prevalence and biologic significance of lupus anticoagulant and anticardiolipin antibodies in a general obstetric population. AM ] OBSTET GVNECOL 1989; 161 :369-73, 2, Love PE, Santoro SA, Antiphospholipid antibodies: anticardiolipin and the lupus anticoagulant in systemic lupus erythematosus (SLE) and non-SLE disorders. Ann Intern Med 1990; 112:682-98. 3. Harris EN, Gharavi AE, Hughes GR. Anticardioplin antibodies. Clin Rheumatol Dis 1985;11:591-609. 4. Harris EN, Chan ]KH, Asherton RA, Aber VR, Gharavi AE, Hughes GRY. Thrombosis, recurrent fetal loss, and thrombocytopenia: predictive value of the anticardiolipin antibody test. Arch Intern Med 1986;146:2153-6. 5. Lockshin MD, Druzin ML, Goei S, et al. Antibody to cardiolipin as a predictor of fetal distress or death in pregnant patients with systemic lupus erythematosus. N Engl ] Med 1985;313:152-6. 6. Canoso RT, Zon LI, Groopman ]E. Anticardiolipin antibodies associated with HTLV-III infection. Br] Haematol 1987;65:495-8. 7, Panzer S, Stain C, Hartl H, Dudczak R, Lechner K. Anticardiolipin antibodies are elevated in HIV-l infected haemophiliacs but do not predict for disease progression, Thromb Haemost 1989;61:81-5. 8. Mulhall BP, NaseIli G, Whittingham S. Anticardiolipin antibodies in homosexual men: prevalence and lack of association with human immunodeficiency virus (HIV) infection.] Clin Immunol 1989;9:208-13, 9. Stimmler MM, Quismorio FP, McGehee WG, Boylen T, Sharma OP. Anticardiolipin antibodies in the acquired immunodeficiency syndrome. Arch Intern Med 1989; 149: 1833-5. 10. Cohen H, Mackie I], Anagnostopoulos N, Savage GF, Machin SJ. Lupus anticoagulant, anticardiolipin antibodies, and human immunodeficiency virus in haemophilia. ] Clin Pathol 1989;42:629-33. 11. Martin K, Katz B, Miller G. AIDS and antibodies to human immunodeficiency virus (HIV) in children and their families.] Infect Dis 1987;155:54-9. 12. Cowchuck S, DeHoratius RD, Wapner R], Jackson LG. Subclinical autoimmune disease and unexplained abortion, AM] OBSTET GVNECOL 1984;150:367-71. 13, Harris EN, Asherton RR, Gharavi AE. Thrombocyto-
Volume 167 Number 4, Part 1
penia in SLE and related disorders: association with anticardiolipin antibodies. Br J Haematol 1985;59:227-30. 14. Koike T, Sueishi M, Funaki H, Tomioka H, Yoshida S. Antiphospholipid antibodies and biologic false-positive serologic test for syphilis in patients with systemic lupus erythematosus. Clin Exp Immunol 1984;56: 193-9. 15. Andiman WA, Simpson BJ, Olson B, Dember L, Saliva T, Miller G. Rate of transmission of HIV-l infection from mother to child and short-term outcome of neonatal infection: Results of a prospective cohort study. Am J Dis Child 1990:144:758-66. 16. Rauch J, Tannenbaum H, Stollar BD. Monoclonal anticardiolipin antibodies bind to DNA. Eur J Immunol 1984; 14:529-34.
Anticardiolipin antibodies in HIV-infected pregnancies
1085
17. Rossi P, Moschese V, Broliden PA, et al. Presence of maternal antibodies to human immunodeficiency virus 1 envelope glycoprotein gp120 epitopes correlates with the uninfected status of children born to seropositive mothers. Proc Nat! Acad Sci USA 1989;86:8055-8. 18. Devash Y, Calvelli TA, Wood DG, Reagan KJ, Rubenstein A. Vertical transmission of human immunodeficiency virus is correlated with the absence of high-affinity / avidity maternal antibodies to the gp120 principal neutralizing domain. Proc Nat! Acad Sci USA 1990;87:3445-9. 19. Goedert J, Drummond J, Minkoff H, et al. Mother to infant transmission of human immunodeficiency virus type 1: association with prematurity or low anti-gpI20. Lancet 1989;2:1351-4.
