Low haptoglobin and a positive direct antiglobulin test without haemolysis in pregnancy

Abstract

Haemolysis is typically associated with low haptoglobin and elevated reticulocyte count, lactate dehydrogenase and indirect bilirubin. Positive direct antiglobulin testing is consistent with autoimmune haemolysis. A case of anaemia in pregnancy with low haptoglobin levels and positive direct antiglobulin testing in a woman with systemic lupus erythematosus is presented. The lack of response to intravenous immune globulin and absence of other markers of haemolysis prompted further investigation. In the setting of mild renal dysfunction, the woman’s serum erythropoietin was inappropriately low consistent with a failure of erythropoietin response to anaemia, and the woman’s haemoglobin improved rapidly with darbopoietin therapy. darbepoetin Health professionals need to be aware of the possibility of low haptoglobin and positive direct antiglobulin testing in the absence of haemolysis with autoimmune disease and anticardiolipin antibodies, and the possibility of anaemia due to failure of erythropoietin response with mild renal dysfunction in pregnancy.

Case

A 29-year-old woman presented at seven weeks’ gestation for antenatal care. Systemic lupus erythematosus (SLE) had been diagnosed at age 18 when she presented with lupus nephritis (membranous nephropathy with focal segmental proliferation on biopsy). Other complications of SLE included mitral valvulopathy requiring mechanical valve replacement, dilated cardiomyopathy without significant coronary artery disease on angiography, and positive anticardiolipin antibodies and lupus anticoagulant. Her medications included cyclosporine, azathioprine, hydroxychloroquine, bisoprolol, aspirin, enoxaparin, prednisone, aspirin, folic acid and calcium. The woman had not been treated with biologic agents.

Between 6 and 12 weeks’ gestation, the woman’s haemoglobin (Hb) fell from 103 to 79 g/L (Figure 1). The results for concurrent investigations are listed in Table 1. There were no clinical features suggestive of a lupus flare. Serum creatinine (se Cr) was 100 µmol/L, renal function having been stable for the previous two years. Serum complement levels had also been unchanged for the previous two years. Liver function tests were normal. The blood film revealed a left shift with moderate toxic granulation and anisocytosis; no schistocytes, acanthocytes, spherocytes or bite cells were seen.

Figure 1.

Change in haemoglobin during pregnancy.

Table 1.

Investigations at 12 weeks’ gestation.

Investigation	Result	Normal	Investigation	Result	Normal
Haemoglobin	79 g/L	110–145	Creatinine	100 µmol/L	30–76
Mean cell volume	103 fL	81–99	Urine PCR	29 mg/mmol	0–30
White cell count	4.6 × 109/L	5.9–16.9
Platelets	142 × 109/L	150–400	IgG anticardiolipin	17 U/ml	<10
			IgM anticardiolipin	Negative
Ferritin	327 µg/L	30–150	β2 glycoprotein	20 U/ml	<20
Transferrin saturation	38%	15–45	Lupus anticoagulant	Positive
B12	451 pmol/L	89–664	C3	0.4 g/L	0.9–2.0
Folate	39.1 nmol/L	>11.8	C4	0.07 g/L	0.15–0.45
Haptoglobin	<0.2 g/L	0.25–1.8	ANA titre	1/2560
Reticulocyte	55.6 × 109/L	10–100	SSA, Sm	Positive
LDH	215 U/L	120–250	ds DNA	Negative
Bilirubin	6 µmol/L	<20
DAT	Positive
Parvovirus	IgM negative

LDH: lactate dehydrogenase; DAT: direct antiglobulin test; PCR: protein:creatinine ratio.

Two years earlier while not pregnant, a positive direct antiglobulin test (DAT) and undetectable haptoglobin were noted when the woman presented with iron deficiency anaemia in the setting of severe menorrhagia. Reticulocyte count, lactate dehydrogenase (LDH) and serum bilirubin were normal, and there was no evidence on haemolysis on peripheral blood smear. The woman’s anaemia resolved with intravenous iron carboxymaltose and further investigation was not undertaken.

At 12 weeks’ gestation, the woman was reviewed by the haematology service who felt autoimmune haemolysis was the most likely cause for the fall in haemoglobin based upon the positive DAT and low haptoglobin, despite the normal reticulocyte count, LDH and bilirubin, and the absence of features of haemolysis on peripheral blood smear. Intravenous immune globulin (IVIG) 0.4 g/kg was administered. In the absence of an improvement in Hb, the dose of IVIG was increased to 0.8 g/kg at 16 weeks’ gestation. The failure of improvement in haemoglobin at 20 weeks’ gestation, and the absence of other features of haemolysis caused the diagnosis of haemolytic anaemia to be questioned, and other aetiologies were sought. The woman’s erythropoietin level at 14 weeks’ gestation had been inappropriately low at 18.4 miU/L (non-pregnant reference range 4–32) when Hb was 79 g/L. At 20 weeks’ gestation, darbopoietin darbepoetin 20 mcg weekly was commenced. Her haemoglobin improved to 100 g/L by 24 weeks’ gestation, and progressively rose for the remainder of the pregnancy (Figure 1). The woman required caesarean section at 35 weeks’ gestation because of fetal growth restriction and lag in interval growth velocity, delivering and delivered a well infant with birthweight 1690 g.

Discussion

This case illustrates that low haptoglobin levels and positive DAT may occur with autoimmune disease in the absence of haemolysis, and that the erythropoietin response to anaemia may be blunted in pregnancy with mild renal dysfunction.

Haptoglobin levels usually remain relatively stable during pregnancy. ¹ Aside from haemolysis, the most common cause of low haptoglobin is hepatocellular disease.

Low haptoglobin levels without haemolysis have been described in a variety of autoimmune conditions including SLE, autoimmune thyroid disease, adrenal insufficiency, multiple sclerosis and hypogammaglobulinaemia.^2,3 Positive DAT has been described in 16% of individuals with antiphospholipid antibodies and 4–13% of women with SLE.^4,5 Only 54% of the women with SLE and a positive DAT demonstrated haemolytic anaemia. ⁵ Positive DAT have also been demonstrated in healthy blood donors with antiphospholipid antibodies. ⁶ The woman described demonstrated an undetectable haptoglobin and positive DAT in the absence of blood film features or other sensitive markers of haemolysis.

The woman exhibited an absence of erythropoietin response to anaemia in the setting of relatively mild renal dysfunction. Pregnancy is associated with a significant increase in erythropoietin (EPO) requirements, levels typically rising two to four fold above preconception levels by 20 weeks’ gestation in the absence of anaemia. ⁷ Studies of EPO concentrations in medical and surgical non-pregnant women with haemoglobin less than 80 g/L and normal renal function demonstrated serum EPO of 62–366 iu/L (20th–80th centiles). ⁸

Several studies have examined regulation of the EPO response to anaemia in non-pregnant individuals with chronic kidney disease (CKD). Fehr et al. demonstrated a significant linear inverse correlation between EPO and anaemia above a creatinine clearance of 40 ml/min. ⁹ Below this level of renal function, no correlation was found. Artunc et al. found a strong correlation between severity of anaemia and increase in EPO in individuals with normal renal function; however, this was gradually attenuated with increasing stages of CKD, and completely lost with CKD stage 4–5. ¹⁰ Korte et al. described altered regulation of EPO in response to anaemia in non-pregnant individuals with serum creatinine over 100 µmol/L. ¹¹

Increased renal blood flow and glomerular filtration rate result in a fall in se Cr in pregnancy. A systematic review of se Cr in pregnancy found that se Cr values greater than 76 µmol/L in first trimester, 72 µmol/L in second trimester, and 77 µmol/L in third trimester should be considered to be outside the upper limit of normal for pregnancy. ¹²

A recent study found that approximately 30% of women with stage 2–3 CKD received treatment with erythropoietin stimulating agents for anaemia during pregnancy. ¹³

Multiple factors may have contributed to fetal growth restriction in this case including maternal anaemia, SLE, antiphospholipid syndrome, CKD and therapy with a β-blocker and immunosuppressives (azathioprine, cyclosporin).

While not relevant to this case, a scenario in later pregnancy in which physicians need to consider that low haptoglobin and positive DAT may occur in the absence of haemolysis relates to the diagnosis of preeclampsia with haemolysis, elevated liver enzymes and low platelets (HELLP) syndrome. The risk of preeclampsia and HELLP syndrome is increased in women with SLE, particularly in women with lupus nephritis and antiphospholipid syndrome. Differentiation of preeclampsia from lupus flares, and from the physiological rise in blood pressure and proteinuria in third trimester in women with lupus nephritis may be difficult. The diagnosis of HELLP syndrome is further complicated by the presence of thrombocytopenia and abnormal liver function tests in up to 25–50% and 60% of non-pregnant women with SLE, respectively. ^14,15 Autoimmune haemolysis may also occur in up to 10% of women with SLE.

In conclusion, a low haptoglobin and/or positive DAT alone should not be regarded as diagnostic of haemolysis in pregnancy in the setting of autoimmune disease. Mild kidney dysfunction in pregnancy may be associated with renal anaemia and may benefit from erythropoietin stimulating agents.