J. clin. Path., 1970, 23, 514-516 J Clin Pathol: first published as 10.1136/jcp.23.6.514 on 1 September 1970. Downloaded from Serum alpha2-macroglobulin, transferrin, albumin, and IgG levels in preeclampsia

C. H. W. HORNE, P. W. HOWIE, AND R. B. GOUDIE From the University Departments ofPathology and Obstetrics and Gynaecology, Western Infirmary, Glasgow

SYNOPSIS A radial immunodiffusion technique has been used to measure levels of four serum proteins in preeclampsia with or without proteinuria and in normal pregnant and non-pregnant controls. In preeclampsia unaccompanied by proteinuria, albumin and transferrin levels are similar to those found in the normal pregnant controls, but there are significant falls in 0x2-macroglobulin and IgG. When preeclampsia is accompanied by proteinuria there is a marked fall in albumin and an increase in o'2-macroglobulin. Since oU2-macroglobulin has antiplasmin activity it is possible that increased levels of this protein in preeclampsia accom-copyright. panied by proteinuria contribute to the intravascular coagulation which has been described in this disorder.

Both in pregnancy and the nephrotic syndrome tension (bloodpressurehigher than 140/90mm Hg) increased levels of serum x2-macroglobulin have on two or more separate occasions after 28 weeks been reported (Schumacher and Schlumberger, of pregnancy in patients whose blood pressurehttp://jcp.bmj.com/ 1963; Schultze and Schwick, 1959). We therefore was less than 140/90 m-m Hg in the first trimester. thought it would be of interest to determine the Most of the patients had oedema. Preeclampsia serum oI2-macroglobulin levels in preeclampsia, a with proteinuria was diagnosed when proteinuria complication of pregnancy which bears a certain was detected for the first time after 28 weeks of similarity to the nephrotic syndrome. For pregnancy and was not due to urinary tract comparison the serum levels of transferrin, IgG, infection, preexisting renal disease, or contamin- and albumin have also been measured to demon- ation with vaginal discharge. For comparison on September 28, 2021 by guest. Protected strate changes due to factors such as haemo- 13 sera were selected from 37 obtained from pre- dilution. eclamptic patients without proteinuria: the cases We report here the serum levels of c2-macro- were chosen to match the first group as closely globulin, transferrin, IgG, and albumin, assayed as possible (see Table) with respect to age (within by a radial immunodiffusion technique, in pre- two years), parity (primiparous or multiparous), eclampsia and in matched normal pregnant and and duration of pregnancy (within five weeks non-pregnant controls. except one in which the difference was 10). For each preeclamptic case a control serum was obtained from a normal pregnant woman matched for age (within two years), parity, and Subjects and Methods duration of pregnancy (within four weeks except for two where the difference was seven and eight Sera were obtained from four groups of patients: weeks). An additional control group consisted (1) preeclampsia with proteinuria (13 cases); of 26 patients attending a gynaecological out- (2) preeclampsia without proteinuria (13 cases); patient clinic and matched for age with the pre- (3) normal pregnant (26 cases); (4) non-pregnant eclamptic patients. The normal pregnant and (26 cases). Preeclampsia was defined as hyper- non-pregnant controls used in this experiment Received for publication 27 January 1970. were selected from a group of 37 described in a 515 Serum alpha2-macroglobulin, transferrin, albumin, and IgG levels in preeclampsia J Clin Pathol: first published as 10.1136/jcp.23.6.514 on 1 September 1970. Downloaded from

previous communication (Horne, Howie, Weir, 1000 r 0 and Goudie, 1970). The sera were stored at -20°C for up to six months before assay. 0

The assay of the serum proteins was carried 0 out using a radial immunodiffusion technique (Horne et al, 1970) similar to that described Soo F previously. To minimize the effects of interplate

variation (Thompson, Home, Steele, and Goudie, A 1969) matched cases of preeclampsia, together with the pregnant and non-pregnant controls, E 600 * * 0

were in on same assay z always tested duplicate the y

plate and the results read by an observer who did .0 0 I. not know the arrangement of the sera within each 0 0 plate. The arithmetic mean ofthe two readings was 400 used in all calculations. The 'absolute values' of 1 80 the measured proteins were determined from the 1C4 calibration curves with solutions of a freeze-dried, vd 05:0 t0 LO reconstituted serum *& pooled human containing 3, IS * 6, 12, and 18 g protein per 100 ml and standard- 200o ized with reference to a serum(BehringwerkeAG) containing specified amounts of the particular protein.

Non-pregnont Nofmal presnont Preeclzmpsia Pmeciamp ia without proteinur;a with proteinuria Results Figure Serum a2-macroglobulin levels in the four groups studied. The serum protein levels in preeclampsia and in matched normal pregnant and non-pregnant in the albumin level (-21 %) associated with the controls is given in the Table which also shows presence of proteinuria. No significant differencecopyright. where statistically significant differences between between the transferrin or IgG levels is observed. groups have been found by the Student t test. A more detailed comparison of %2-macro- globulin levels is shown in the Figure. In normal pregnancy compared with non-pregnant controls Discussion there are significant increases in the levels of cx2-macroglobulin (+ 37%) and transferrin Our findings in preeclampsia relating to albumin http://jcp.bmj.com/ (+ 53 %) and falls in the levels of albumin are similar to those of other authors (Spetz and (-31 %) and IgG (-20 %). Brody, 1967; Honger, 1966), although we do not Comparing preeclampsia without proteinuria know of a previous demonstration that protein- with normal pregnancy there are significant falls in uria in this condition is associated with more the levels of (x2-macroglobulin (-19%) and in profound hypoalbuminaemia. Honger (1966) has IgG (-16 %), but no significant difference in the found increased catabolism of albumin in pre- transferrin and albumin levels. However, com- eclampsia; presumably the resulting fall in parison of the two groups of preeclamptic plasma albumin is further accentuated by protein- on September 28, 2021 by guest. Protected patients shows a significant increase (+55%) in uria when present. The fall in IgG in preeclampsia the U2-macroglobulin level and a significant fall is in keeping with the electrophoretic findings of

Group No. of Mean Age (yr) Duration of Mean (J SD) Serum Levels (mg/100 ml) Cases Pregnancy (wk) a,-Macro- Transferrin Albumin IgG globulin 1 Preeclampsia with 13 27-7 35 2 ab fg i proteinuria (range 18-38) (range 30-40) 597 ± 216 388 + 253 2,319 ± 401 747 ± 223 2 Preeclampsia without 13 27-9 37 0 ac f i proteinuria (range 18-41) (range 34-40) 385 ± 144 381 ± 104 2,938 ± 543 811 ± 182 3 Normal pregnancy 26 27-9 36-8 bcd e hg ijk (range 18-41) (range 32-40) 475 + 146 427 i 159 3,221 + 830 961 i 281 4 Non-pregnancy 26 27-6 d e h k (range 18-43) - 347 ± 122 280 ± 102 4,646 ± 1,320 1,199 ± 351

Table Serum protein levels in preeclampsia and in matched normalpregnant and non-pregnant controls a P < 0 005 b P < 0 025 c P < 0-05 d P < 0 0025 e P < 0-0005 f P < 0 0025 g p < 0 0005 h P < 0 0005 i p < 0-01 j p < 0 05 k P < 0 005 516 C. H. W. Horne, P. W. Howie, and R. B. Goudie J Clin Pathol: first published as 10.1136/jcp.23.6.514 on 1 September 1970. Downloaded from some (Mack, 1960) but not all authors (Spetz activity of Ux2-macroglobulin (Williams, 1969; et al, 1967). Schultze, Heimburger, Heide, Haupt, Storiko, High c%2-macroglobulin levels in preeclampsia and Schwick, 1963; Steinbuch, Quentin, and with proteinuria do not appear to have been Pejaudier, 1965; Ganrot, 1967) it would be of encountered previously. Mack (1960) and Spetz interest to know whether intravascular thrombosis and Brody (1967) found normal o2-globulin is especially found in preeclamptic patients who levels in preeclampsia but they used the less have proteinuria or raised aU2-macroglobulin discriminating technique of filter paper electro- levels. If raised 0x2-macroglobulin levels really phoresis and did not separate preeclampsia into contribute to intravascular thrombosis in pre- two groups according to the presence or absence eclampsia they might be expected to cause similar of proteinuria. Wardle and Menon (1969) did effects in the nephrotic syndrome. The only well not find any increase in CX2-macroglobulin levels known thrombotic association of the nephrotic in preeclampsia accompanied by proteinuria syndrome is renal vein thrombosis, but the compared with normal pregnancy, but details of relationship of this to membranous glomerulo- their radial immunodiffusion technique are not nephritis (Lancet, 1969) and amyloid disease of given. Since our findings are at variance with those kidney (Heptinstall, 1967) is not fully under- of the above authors they require further confirm- stood. ation in carefully controlled studies. In a previous communication (Home et al, 1970) we showed that administration of combined oestrogen and progestogen oral contraceptives results in Our thanks are due to Mr P. E. Hall, of the increases in oa2-macroglobulin, transferrin, and Steroid Biochemistry Department, GlasgowRoyal IgG, and suggested that the raised levels of Infirmary, for advice on the hormonal changes in x2-macroglobulin and transferrin in normal preeclampsia. This study was supported by pregnancy might be due to raised blood levels of grants from Action for the Crippled Child, the these hormones. Our findings in preeclampsia Secretary of State for Scotland, and research without proteinuria of lower levels of cx2-macro- funds of Glasgow University. globulin, transferrin, and IgG than expected in normal pregnancy are consistent with the effects of diminished production of oestrogen and/or References copyright. progestogen, and there is some indication o Bonnar, J., McNicol, G. P., and Douglas, A. S. (1969). Fibrino- lytic enzyme system and pregnancy. Brit. med.J., 3, 387-389. decreased production of oestrogens in pre- Ganrot, P. 0. (1967). Inhibition of plasmin activity by a2-macro- eclampsia according to the severity of the disease globulin. Clin. chim. Acta, 16, 328-329. Heptinstall, R. H. (1967). Pathology of the Kidney, p. 584. (Klopper, 1968). Churchill, London. Severe preeclampsia resembles the nephro- Honger, P. E. (1966). Albumin metabolism in pre-eclampsia. tic syndrome in showing hypoalbuminaemia, Acta Obstet. gynec. scand., 45, Suppl., 9, 104-105. proteinuria, and oedema, and this resemblance Horne, C. H. W., Howie, P. W., Weir, R. J., and Goudie, R. B.

(1970). Effect of combined oestrogen-progestogen oralhttp://jcp.bmj.com/ is further emphasized by high oz2-macroglobulin contraceptives on serum levels of a,-macroglobulin, transferrin, albumin and IgG. Lancet, 1, 49-51. levels often found in both conditions. The factors Klopper, A. (1968). The assessment of feto-placental function by causing raised a2-macroglobulin levels in these estriol assay. Obstet. gynec. Sure., 23, 813-838. conditions are obscure. Lancet (1969). Membranous glomerulonephritis. (Leadingarticle.) Hypoalbuminaemia does Lancet, 2, 626-627. not appear to be a cause since we failed to Mack, H. C. (1960). The plasma proteins. Clin. Obstet. Gynec., demonstrate an inverse correlation between serum 3, 336-348. Morris, R. H., Vassalli, P., Beller, F. K., and McCluskey, R. T. albumin and 02-macroglobulin in the cases (1964). Immunofluorescent studies of renal biopsies in the presently under study. The lack of a comparable diagnosis of toxaemia of pregnancy. Obstet. Gynec., 24, on September 28, 2021 by guest. Protected 32-46. rise in transferrin is against an oestrogen and/or Robertson, W. B., Brosens, I., and Dixon, H. G. (1967). The progestogen effect on 0x2-macroglobulin in pre- pathological response of the vessels of the placental bed to with and there is no evi- hypertensive pregnancy. J. Path. Bact., 93, 581-592. eclampsia proteinuria Schultze, H. E., Heimburger, N., Heide, K., Haupt, H., Storiko, dence that these hormones are increased in severe K., and Schwick, G. (1963). Preparation and characteriza- preeclampsia. tion of a,-trypsin inhibitor and ag-plasmin inhibitor of human serum. In Proceedings ofNinth Congress ofEuropean Intravascular thrombosis in the glomerulus Society for Haematology, pp. 1315-1320. Karger, Basle. (Morris, Vassalli, Beller, and McCluskey, 1964) Schultze, H. E., and Schwick, G. (1959). Quantitative immuno- logische Bestimmung von Plasmaproteinen. Clin. chim. and in the spiral arteries of the placental bed Acta., 4, 15-25. (Robertson, Brosens, and Dixon, 1967) are Schumacher, G., and Schlumbarger, H. D. (1963). tJber Verander- features of ungen der a,-Globuline des Serums. Dtsch. med. Wchschr., preeclampsia. Impaired fibrinolysis 88, 645-651. due to diminished plasminogen activator is well Steinbuch, M., Quentin, M., and Pejaudier, L. (1965). Protides known in normal pregnancy (Wardle et al, 1969; Biol. Fluids, 12, 375-378. Spetz, S., and Brody, S. (1967). Serum proteins in pregnancy Bonnar, McNicol, and Douglas, 1969) but there complicated by toxaemia. Acta obstet. gynec. scand., 46, is no apparent difference in fibrinolytic activity 151-167. Thompson, A., Horne, C. H. W., Steele, H., and Goudie, R. B. between normal pregnancy and preeclampsia (1969). Concentrations of serum a,-macroglobulin in twins. (Wardle et al, 1969). In view of the antiplasmin Nature (Lond.), 221, 289-290.