Hyperhomocysteinemia: a risk factor in women with unexplained recurrent early pregnancy loss*t

FERTILITY AND STERILITY Copyright" 1993 The American Fertility Society Vol. 50, No.5, November 1993 Printed on acid-free paper in U. S. A. Hyperhomocysteinemia: a risk factor in women with unexplained recurrent early pregnancy loss*t Maurice G. A. J. Wouters, M.D.:!: Godfried H. J. Boers, M.D.II Henk J. Blom, Ph.D.~ Frans J. M. Trijbels, Ph.D.~ Chris M. G. Thomas, Ph.D.:!: George F. Borm, Ph.D.** Regine P. M. Steegers-Theunissen, M.D.:!: Tom K. A. B. Eskes, M.D.:!: University Hospital Nijmegen St Radboud, Nijmegen, The Netherlands Objective: To establish the prevalence of hyperhomocysteinemia in women with unexplained recurrent.early pregnancy loss. Design: In a patient-control study, the methionine-homocysteine metabolism was investigated by a standardized oral methionine-loading test. Setting: Gynecologic outpatient department of university hospital. Patients: One-hundred and two women who had been referred to the hospital because they suffered from at least two consecutive unexplained spontaneous abortions (study group) as well as 41 controls who were recruited by public advertisement were selected. Interventions: Blood samples were collected just before and 6 hours after oral methionine administration to determine plasma total homocysteine concentrations. Main Outcome Measure: Plasma total homocysteine concentrations 6 hours after methionine loading. Hyperhomocysteinemia was defined as total homocysteine concentration at 6 hours exceeding the 97.5 percentile level of the controls. Results: Hyperhomocysteinemia was diagnosed in 21 women of the study group (21%). In the parous women of the study group, the prevalence of hyperhomocysteinemia was more than two times greater compared with the nulliparous subjects (33% and 14%, respectively). Conclusion: Hyperhomocysteinemia is a risk factor in women with unexplained recurrent early pregnancy loss. Fertil Steril 1993;60:820-5 Key Words: Homocysteine, habitual abortion, folate, vitamin B12 Recurrent early pregnancy loss is an intriguing problem in obstetric practice. Over the last decade, Received May 11, 1993; revised and accepted July 26, 1993. * Supported by grant number 28.1006.1 from Praeventiefonds, The Hague, The Netherlands. t Presented at the 40th Annual Meeting of the Society for Gynecologic Investigation, Toronto, Canada, March 31 to April 3,1993. :j: Department of Obstetrics and Gynecology, University Hospital Nijmegen St Radboud. Reprint requests: Maurice G. A. J. Wouters, M.D., Department of Obstetrics and Gynecology, University Hospital Nijmegen St Radboud, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands. II Department of Medicine, University Hospital Nijmegen St Radboud. many etiologic factors and several treatment regimens have been reported. However, there is little evidence for most causes of recurrent miscarriage and therapeutic interventions of any kind can be considered as merely empirical (1). Recently, we have presented preliminary data on hyperhomocysteinemia as a possible risk factor in women with recurrent spontaneous abortion or placental abruption (2). The present study reports the 1"[ Department of Pediatrics, University Hospital Nijmegen St Radboud. ** Department of Medical Statistics, University Hospital Nijmegen St Radboud. 820 Wouters et al. Homocysteine and habitual abortion Fertility and Sterility

~I METHIONINE S-ADENOSYLMETHIONINE 1 BETAINE TETRAlY:ROFOLATE TETRAHYDROFOLATE + 5.10-METHYLENE- 5-METHYL TETRAHYDROFOLATE ~-synthase deficiency, is also associated with an increased risk of premature vascular disease (6-10). Methionine-homocysteine metabolism was investigated in 102 women with unexplained recurrent early pregnancy failure and in 41 controls to establish the prevalence of hyperhomocysteinemia. MATERIALS AND METHODS CYSTATHIONINE PYRIDOXAL-5-PHOSPHATE CYSTEINE Figure 1 Simplified scheme of the methionine-homocysteine metabolism and enzymes and vitamins involved. (1) Cystathionine {3-synthase; (2) 5-methyltetrahydrofolate-homocysteine methyltransferase (methionine synthase); (3) 5,1O-methylenetetrahydrofolate reductase. results of our extended investigation in women with at least two consecutive unexplained early pregnancy losses. Homocysteine is the demethylated derivative of the essential amino acid methionine. It is either irreversibly trans-sulphurated via cystathionine into cysteine, in which the first reaction is catalyzed by a pyridoxal-5-phosphate-(plp, a form of vitamin B6) dependent enzyme (cystathionine ~-synthase), or it is remethylated to methionine by either of two reactions. One of these reactions is dependent on folate and vitamin B12 (Fig. 1) (3). Homocystinuria due to cystathionine ~-synthase deficiency is an inborn error of metabolism with autosomal recessive inheritance and is characterized by massive urinary excretion of homocysteine and severe elevation of blood homocysteine and methionine (4). Premature arteriosclerosis and thromboembolism are life-threatening complications of homo cystinuria. This is explained by homocysteine-induced endothelial cell injury, which may initiate the development of typical arteriosclerotic lesions (5). Notably, fetal loss has been observed in 25 of 52 pregnancies of untreated homozygous cystathionine ~-synthase-deficient women, but these data were strongly influenced by the high number of stillbirths and spontaneous abortions in a few individuals (4). Several studies have demonstrated that hyperhomocysteinemia, in which the mildly elevated blood homocysteine concentrations are comparable to those in obligate heterozygosity for cystathionine Subjects Spontaneous abortion was defined as early pregnancy loss within 16 weeks of menstrual age, not including ectopic pregnancy and elective abortion. We studied 102 women (aged from 22 to 47 years) who had been referred to our hospital because they suffered from at least two consecutive unexplained spontaneous abortions (range 2 to 10) after conceiving from the same partner (study group). Seventy-seven subjects (75%) had three consecutive miscarriages or more. In total, all women (n = 102) experienced 378 spontaneous abortions. Sixtythree percent of all early pregnancy losses (n = 238) were histologically confirmed. In the remaining 37% (n = 140), pregnancy was diagnosed by a positive routine urinary heg test (>50 IU/L) and/or ultrasound imaging of at least one intrauterine gestational sac, with or without positive heartbeat. Thirty-six women (35%) had experienced one or more pregnancies (range 1 to 4) that had developed past 16 weeks amenorrhea. In all women of the study group, chromosomal rearrangements, severe uterine anomalies, and immunologic disorders were excluded by routine investigation procedures. These consisted of parental chromosome karyotyping, uterine evaluation by either hysterosalpingogram and/or hysteroscopy (84% of cases), or ultrasound alone, and tests for both anticardiolipin antibodies (standardized ELISA) and lupus anticoagulant (kaolin clotting time, 1:5 dilution of patients' samples in phospholipid-free substrate plasma). As controls, we studied 41 women (aged 30 to 44 years) who had delivered at least one live child (range 1 to 4) and had not suffered from spontaneous abortion, fetal death, or placental abruption. They were recruited by public advertisement. All women of both the study and the control group (n = 143) were healthy and had not suffered from renal, liver, or vascular disease. Serum creatinine, range 0.6 to 1.0 mg/dl (52 to 91 ~mol/l), serum alanine aminotransferase, range 5 to 29 IU /L (0.08 to 0.48 ~mol/s per L), and serum aspartate Vol. 60, No.5, November 1993 Wouters et al. Homocysteine and habitual abortion 821

aminotransferase, range 4 to 21 IU/L (0.07 to 0.35 JLmol/s per L) were within the normal range. The study was approved by the Ethical Committee of the University Hospital Nijmegen St Radboud, Nijmegen, The Netherlands. Before participation, informed consent was obtained from all subjects. Investigation Procedure All women were instructed not to become pregnant before the investigation. They were not allowed to take oral contraceptives, hormonal and vitamin supplements, and other medication possibly interfering with methionine-homocysteine metabolism for at least 3 months before the study. Methionine-homocysteine metabolism was investigated by a standardized oral methionine-loading test. After an overnight fast, venous blood samples were drawn to measure the total homocysteine concentration in plasma at baseline and the concentrations of vitamins (PLP in whole blood, vitamin B12 in serum, and folate, both in serum and red cells). Then L-methionine, 0.1 g (0.7 mmol)/kg body wt, was administered orally in 200 ml of orange juice. All women used a standardized methionine-restricted breakfast and luncheon (total content of 14 mg methionine in 2 g protein, 95 g carbohydrate, and 31 g fat; total energy load of 2.8 kj). No drinks, except for coffee and tea without milk were allowed during the test period. Six hours after methionine loading, venous blood was collected for the measurement of the plasma total homocysteine concentration. To minimize possible hormonal influences on homocysteine metabolism, testing procedures were routinely arranged about 1 week before the expected first day of the next menstrual period (11). The majority of tests (73%) were performed within 6 months after completion of the last pregnancy. Hyperhomocysteinemia was diagnosed if total homocysteine concentration at 6 hours exceeded the 97.5 percentile level of the controls, independent of total homocysteine concentration at baseline and the fasting concentrations of PLP, vitamin B12, and folate. Sample Preparation and Analysis Blood samples for measurements of total homocysteine concentrations in plasma were drawn in ethylenediamine tetra-acetate vacutainer tubes of 4 ml and centrifuged within 30 minutes at 3,000 X g for 10 minutes. The plasma was separated and immediately stored at -20 C. Total homocysteine concentrations were measured by high-perfor- mance liquid chromatography (HPLC) technique and fluorometric detection (detection limit, 0.5 JLmol/L; intra-assay and interassay coefficients of variation, both < 5 %), according to Fiskerstrand et a1. (12). Dry and heparinized vacutainer tubes of 10 ml were used for collecting venous blood samples to assay PLP (whole blood), vitamin B12 (serum), and folate (serum and red cells) concentrations. Determination of PLP was performed by HPLC technique (11). Vitamin B12 and folate concentrations were measured simultaneously with Dualcount SPB (solid phase boil) Radioassay (Diagnostic Products Corporation, Los Angeles, CA), as previously described (13). Statistical Analysis Results are given as means ± SD. In assessing statistical significance, the Wilcoxon rank sum test and Spearman's rank correlation were used. P values were two tailed, and P < 0.05 was considered significant. RESULTS Total homocysteine concentration at baseline in the control group ranged from 6 to 19 JLmol/L whereas in women of the study group it ranged from 3 to 81 JLmol/L. The mean levels of total homocysteine concentration at baseline were significantly different between both groups (Table 1). In the control women, total homocysteine concentration at 6 hours ranged from 20 to 55 JLmol/L (Fig. 2), and the 97.5 percentile level of total homocysteine concentration at 6 hours was calculated as 51 JLmol/L. In the study group, total homocysteine concentration at 6 hours varied from 19 to 116 JLmol/L (Fig. 2). Mean total homocysteine concentration at 6 hours ofthe control and the study group were significantly different (Table 1). Hyperhomocysteinemia, i.e., total homocysteine concentration at 6 hours> 51 JLmol/L, was diagnosed in 21 of 102 women (21 %) with unexplained recurrent early pregnancy loss. The mean concentrations of serum vitamin B12, serum folate, and red cell folate were not significantly different between the control and the study group (Table 1). Severe cobalamin deficiency, i.e., serum vitamin B12 < 100 pmol/l, was found in six women of the study group, four of whom were hyperhomocysteinemic, but not in the controls. Low serum folate, i.e., 4 to 6 nmol/l, was detected in three hyperhomocysteinemic women (one of whom 822 Wouters et al. Homocysteine and habitual abortion Fertility and Sterility

Table 1 Concentrations of Plasma Total Homocysteine and Blood Vitamins in the Control and the Study Group* Total homocysteine concentration at baseline, plasma (ILmol/L) Total homocysteine concentration at 6 hr, plasma (ILmol/L) PLP, whole blood (nmol/l) Vitamin B12, serum (pmol/l) Folate, serum (nmol/l) Folate, red cells (nmol/l) Control group Study group (n = 41) (n = 102) 10 ± 3 31 ± 9 51 ± 12 293 ± 109 14 ± 4 512 ± 151 13 ± lot 42 ± 19t 46 ± 18t 261 ± 117 14 ± 7 584 ± 302 * Values are means ± SD. t Statistically significant difference (P < 0.05) compared with the control group. was also severely cobalamin deficient) and in two normohomocysteinemic subjects of the study group but in none of the controls. Severe cobalamin deficiency' low serum folate, and low red cell folate concentrations were simultaneously present in one hyperhomocysteinemic woman of the study group (serum vitamin B12, 84 pmol/l; serum folate, 4.8 nmol/l; red cell folate, 170 nmol/l; and total homocysteine concentration at 6 hours, 83 ~mol/l). The mean concentration of PLP was significantly lower in the women of the study group as compared with the controls (Table 1). However, PLP was normal in all subjects of the study group and in all controls. In the control group, only serum folate was negatively and significantly correlated to both total homocysteine concentration at baseline (r = -0.4) and total homocysteine concentration at 6 hours (r = -0.5). In the study group, serum and red cell folate were negatively and significantly correlated with total homocysteine concentration at baseline (r = -0.3 and -0.4, respectively), and total homocysteine concentration at 6 hours (r = -0.4 and -0.3, respectively), whereas serum vitamin B12 was negatively and significantly correlated to total homocysteine concentration at baseline (r = -0.5), but not to total homocysteine concentration at 6 hours. Neither the number of spontaneous abortions nor the mean gestational age at which pregnancy losses occurred were significantly correlated to total homocysteine concentration at 6 hours. In the parous women of the study group, mean total homocysteine concentration at 6 hours was significantly higher compared with the nulliparous subjects (48 ± 21 versus 38 ± 16 ~mol/l; means ± SD). Hyperhomocysteinemia was found in 12 of 36 (33%) parous and in 9 of 66 (14%) nulliparous women of the study group. The means of total homocysteine concentration at baseline and all studied vitamins, however, did not differ significantly between these subgroups (data not shown). Two parous women, in whom hyperhomocysteinemia was identified, had received vitamin B supplements with five previous pregnancies, but these had all aborted. DISCUSSION The main finding of the present study is the observation of hyperhomocysteinemia in 21 % of women with unexplained recurrent early pregnancy loss. This result confirms our preliminary report (2). The question arises by which pathogenetic mechanism elevated concentrations of plasma homocysteine may cause early pregnancy loss. There is convincing evidence that even mild elevation of THC-6 (J.Imol/l) 125 100 75 50 25 I',.:. I u:: :n control group :i:, study group Figure 2 Individual total homocysteine concentration at 6 hours of all control women (n = 41) and of all women of the study group (n = 102). Vol. 60, No.5, November 1993 Wouters et al. Homocysteine and habitual abortion 823

blood homocysteine concentrations leads to premature vascular disease (6-10). One could speculate that increased levels of maternal plasma homocysteine may cause early damage of decidual or chori- 0nic vessels, leading to disturbed implantation of the conceptus. Recently, Meegdes et al. (14) provided histologic evidence of a higher prevalence of avascular villi and a decreased vascular density of vascularized villi in cases of spontaneous abortion compared with controls. On the other hand, maternal hyperhomocysteinemia may account for other toxic effects in some cases of early pregnancy loss. In this respect, specific embryotoxicity of L-homocysteine has recently been demonstrated in vitro (15). It is also interesting that neural tube defects (NTDs) have been identified in about 1 % of spontaneously aborted fetuses and that the development of NTD in the offspring has been associated with hyperhomocysteinemia in the mother (16, 17). The etiology of hyperhomocysteinemia in patients with premature vascular disease is still unknown but may be due to decreased cystathionine (j-synthase activity (6, 7). In the present study, cystathionine (j-synthase activity has not been investigated. In our preliminary study of hyperhomocysteinemia in women with recurrent spontaneous abortion, however, we have reported that the cystathionine (j-synthase activities, as determined in cultured fibroblasts from a skin biopsy, were within the normal range (2). These data suggest that hyperhomocysteinemia in these women is not caused by a blockade of the homocysteine trans-sulphuration pathway but is due to a defective remethylation of homocysteine into methionine. Malnutrition and malabsorption of folate and vitamin B12 or inherited enzyme defects involved in the metabolism of these vitamins, such as methylenetetrahydrofolate reductase deficiency (see Fig. 1), may also result in hyperhomocysteinemia (18, 19). In 1964, Hibbard (20) suggested a higher prevalence of defective folate metabolism in women with two or more consecutive abortions, compared with controls, as indicated by their excessive formiminoglutamic acid excretion after histidine loading. He has also measured lower serum and red cell folate levels in women with recurrent abortion compared with subjects with normal early pregnancies (Hibbard BM, abstract). In the present study, however, we observed no differences between women with unexplained recurrent early pregnancy loss and the controls with regard to the mean blood concentrations of vitamin B12 and folate. Nevertheless, severe cobalamin deficiency and low serum folate concentrations were detected in a small number of women of the study group (about 9%, the half of which was hyperhomocysteinemic) but in none of the controls. We cannot rule out the possibility that spontaneous abortion, at least in some cases, results from a primary nutritional deficit or metabolic derangement of vitamin B12 and/or folate, of which hyperhomocysteinemia is merely a concomitant finding. Apart from maternal hyperhomocysteinemia, inherited disorders of methionine-homocysteine metabolism of the conceptus itself may contribute to its death. Studies in trophoblast cells should provide data on this option. Remarkably, the prevalence of hyperhomocysteinemia in parous women of the study group was observed to be greater than two times that of nulliparous subjects. A negative association between folate metabolism and parity has been suggested in the literature (21). In the present study, however, we found no significant differences between the blood folate concentrations of parous and nulliparous women of the study group. At the moment, we have no explanation for this intriguing observation. Pyridoxine administration, with or without folic acid, has been shown to reduce plasma homocysteine in vascular patients with hyperhomocysteinemia (22, 23). According to our experience in nonpregnant hyperhomocysteinemic women, normohomocysteinemia may well be achieved by daily oral administration of only folic acid in a dose as low as 1 mg. The question remains whether biochemical normalization of hyperhomocysteinemia by periconceptional folate administration will favor the pregnancy outcome in women with unexplained recurrent miscarriage. A randomized double-blind prevention trial, which should provide the answer to this important question, is being initiated at our institute. In conclusion, hyperhomocysteinemia is a risk factor in women with unexplained recurrent early pregnancy loss. Spontaneous abortion may result from vascular-damaging effects of this condition. Acknowledgments. We thank all of the women who participated in the study. We also acknowledge the following individuals from University Hospital Nijmegen St Radboud, Nijmegen: Louis A. Schellekens, M. D. and Martijn F. G. Segers, B. S. for their critical and stimulating support; Marie-Therese E. C. Moorrees, M. D. and Nelleke J. Hamel-van Bruggen, research nurse, for their outstanding assistance in data collection; and technicians Ms. Maria T. W. B. te Poele-Pothoff and Mr. Loek 824 Wouters et al. Homocysteine and habitual abortion Fertility and Sterility

M. F. Geelen for their excellent assistance III laboratory analysis. REFERENCES 1. Stirrat GM. Recurrent miscarriage II: clinical associations, causes, and management. Lancet 1990; 336:728-33. 2. Steegers-Theunissen RPM, Boers GHJ, Blom HJ, Trijbels JMF, Eskes TKAB. Hyperhomocysteinaemia and recurrent spontaneous abortion or abruptio placentae [letter]. Lancet 1992;339:1122-3. 3. Mudd SH, Levy HL. Disorders of transsulfuration. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic basis of inherited diseases. New York: McGraw-Hill, 1989:693-734. 4. Mudd SH, Skovby F, Levy HL, Pettigrew KD, Wilcken B, Pyeritz RE, et al. The natural history of homocystinuria due to cystathionine /'i-synthase deficiency. Am J Hum Genet 1985;37:1-31. 5. Harker LA, Ross R, Slichter SJ, Scott CR. Homocysteineinduced arteriosclerosis: the role of endothelial injury and platelet response in its genesis. J Clin Invest 1976;58:731-41. 6. Boers G HJ, Smals AG H, Trijbels F JM, Fowler B, Bakkeren JAJM, Schoonderwaldt HC, et al. Heterozygosity for homocystinuria in premature peripheral and cerebral occlusive arterial disease. N Engl J Med 1985;313:709-15. 7. Clarke R, Daly L, Robinson K, Naughten E, Cahalane S, Fowler B, et al. Hyperhomocysteinemia: an independent risk factor for vascular disease. N Engl J Med 1991;324: 1149-55. 8. Kang S-S, Wong PWK, Malinow MR. Hyperhomocyst (e)inemia as a risk factor for occlusive vascular disease. Annu Rev Nutr 1992; 12:279-98. 9. Stampfer MJ, Malinow MR, Willett WC, Newcomer LM, Upson B, Ullmann D, et al. A prospective study of plasma homocyst(e)ine and risk of myocardial infarction in US physicians. JAMA 1992; 268:877-81. 10. Ueland PM, Refsum H, Brattstrom L. Plasma homocysteine and cardiovascular disease. In: Francis RB, ed. Atherosclerotic cardiovascular disease, hemostasis, and endothelial function. New York: Marcel Dekker, 1992:183-236. 11. Steegers-Theunissen RPM, Boers GHJ, Steegers EAP, Trijbels JMF, Thomas CMG, Eskes TKAB. Effects of sub- 50 oral contraceptives on homocysteine metabolism. Contraception 1992;45:129-39. 12. Fiskerstrand T, Refsum H, Kvalheim G, Ueland PM. Homocysteine and other thiols determined in plasma and urine: automation and sample stability. Clin Chern 1993; 39: 263-71. 13. Mooij PNM, Thomas CMG, Doesburg WH, Eskes TKAB. Multivitamin supplementation in oral contraceptive users. Contraception 1991;44:277-88. 14. Meegdes BHLM, Ingenhoes R, Peeters LLH, Exalto N. Early pregnancy wastage: relationship between chorionic vascularization and embryonic development. Fertil Steril 1988;49:216-20. 15. Aerts LAGJM van, Klaasboer HH, Postma NS, Pertijs JCLM, Copius Peereboom JHJ, Eskes TKAB, et al. Stereospecific in vitro embryotoxity of L-homocysteine in pre- and post-implantation rodent embryos. Toxicol In Vitro. In press. 16. Byrne J, Warburton D. Neural tube defects in spontaneous abortions. Am J Med Genet 1986;25:327-33. 17. Steegers-Theunissen RPM, Boers GHJ, Trijbels JMF, Eskes TKAB. Neural-tube defects and derangement of homocysteine metabolism [letter]. N Engl J Med 1990;324: 199-200. 18. Rosenblatt DS. Inherited disorders of folate transport and metabolism. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic basis of inherited diseases. New York: McGraw-Hill, 1989:2049-64. 19. Fenton W A, Rosenberg LE. Inherited disorders of cobalamin transport and metabolism. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic basis of inherited diseases. New York: McGraw-Hill, 1989:2065-82. 20. Hibbard BM. The role of folic acid in pregnancy - with particular reference to anaemia, abruption and abortion. J Obstet Gynaecol Br Commonw 1964; 71:529-42. 21. Stone ML, Luhby AL, Feldman R, Gordon M, Cooperman JM. Folic acid metabolism in pregnancy. Am J Obstet Gynecol 1967; 99:638-48. 22. Boers GHJ. The clinical usefulness of homocysteine determinations. In: Smith U, Eriksson S, Lindgarde S, eds. Genetic susceptibility to environmental factors - a challenge for public intervention. Stockholm: Almquist & Wiksell International, 1988:35-42. 23. Brattstrom L, Israelsson B, Norrving B, Bergqvist D, Thorne J, Hultberg B, et al. Impaired homocysteine metabolism in early-onset cerebral and peripheral occlusive arterial disease - effects of pyridoxine and folic acid treatment. Atherosclerosis 1990;81:51-60. Vol. 60, No.5, November 1993 Wouters et al. Homocysteine and habitual abortion 825