Lupus (2006) 15, 148–155 www.lupus-journal.com

REVIEW

Lupus nephritis and renal disease in pregnancy S Germain and C Nelson-Piercy* Obstetric Medicine, Guy’s & St Thomas’ Hospitals, St Thomas’ Hospital, London, UK

Management of pregnant women with renal disease involves awareness of, and allowance for, physiological changes including decreased serum creatinine and increased proteinuria. For women with systemic lupus erythematosus (SLE), pregnancy increases likelihood of flare. These can occur at any stage, and are more difficult to diagnose, as symptoms overlap those of normal pregnancy. Renal involvement is no more common in pregnancy. Worsening proteinuria may be lupus flare but differential includes pre-eclampsia. In women with chronic renal disease, pregnancy may accelerate decline in renal function and worsen hypertension and proteinuria, with increased risk of maternal (eg, pre-eclampsia) and fetal (eg, IUGR, IUD) complications, strongly correlating with degree of renal impairment peri-conception. Pregnancy success rate varies from 20% to 95% depending on base-line creatinine. Best outcome is obtained if disease was quiescent for ⬎6 months pre-conception. Women on dialysis or with renal transplants can achieve successful pregnancy but have higher maternal and fetal complication rates. Acute on chronic renal failure can develop secondary to complications such as HELLP and AFLP. Management needs to be by a multidisciplinary team involving physicians and obstetricians, ideally beginning with pre-pregnancy counselling. Treatment of flares includes corticosteroids, hydroxychloroquine, azothioprine, NSAIDs and MMF. Blood pressure is controlled with methyldopa, nifedipine or hydralazine. Lupus (2006) 15, 148–155. Key words: nephritis; pregnancy; renal

Physiological changes in pregnancy During pregnancy the urinary collecting system markedly dilates, the result of a combination of progesterone-induced ureteral smooth-muscle relaxation, and compression of the ureters by the enlarging uterus or iliac vessels. A pelvicalceal diameter of up to 2 cm is usually allowed as ‘physiological hydronephrosis’ of pregnancy. There is also an increased risk of pathological hydronephrosis and of pyelonephritis. For anatomical reasons this is more frequently on the right than the left. Renal plasma flow rises from early in pregnancy, and by the second trimester has increased by 60–80%. It then falls throughout the third trimester, but at term is still 50% greater than pre-pregnancy values. The glomerular filtration rate (GFR) increases by 55%, creatinine clearance rises by about 50% to 120–160 mL/min, and there is a resultant fall in

*Correspondence: Catherine Nelson-Piercy, Consultant Obstetric Physician, Guys & St Thomas’ Hospitals, 10th Floor, North Wing, St Thomas’ Hospital, Lambeth Palace Road, London SE1 7EH, UK. © 2006 Edward Arnold (Publishers) Ltd

serum urea (2–4.5 mmol/L) and creatinine levels (25–75 ␮mol/L). Normal ranges differ in pregnancy both from non-pregnant and between trimesters. For example, with creatinine; first trimester 52–68 ␮mol/L, second trimester 44–64 ␮mol/L, and third trimester 55–73 ␮mol/L. Proteinuria is increased, due to increased excretion, and levels of up to 300 mg/24 hours are considered within the normal range in pregnancy. There is no increase in glomerular capillary pressure, and no longterm adverse effects on glomerular morphology. During pregnancy, there is physiological sodium and water retention, with decreased ability to excrete a sodium and water load, especially near term. This means up to 80% of pregnant women develop some oedema, especially in the late third trimester, so it is not usually a pathological sign. There are a number of renal vasodilators produced in pregnancy. These include prostaglandins, endothelium derived relaxing factor (EDRF)/nitric oxide (NO) and relaxin. EDRF/NO mediates both vasodilation and hyperfiltration. Relaxin is produced by the corpus luteum, stimulated by hCG, and increases in the luteal phase and more markedly after conception. When given exogenously to non-pregnant rats there is an 10.1191/0961203306lu2281rr

Renal lupus in pregnancy S Germain and C Nelson-Piercy

149

increase in GFR, and in humans in the luteal phase GFR also increases by 15%. Hormonal changes include increased levels of erythropoietin, renin and vitamin D, with reduced PTH. Uric acid falls in early pregnancy, but then increases with increasing gestation, and bicarbonate levels are reduced to 18–22 mmol/L.

Chronic renal failure in pregnancy Effect of pregnancy on renal disease Pregnancy may accelerate the decline in renal function of 2% to 75% of women with chronic renal disease, depending on the baseline creatinine. They are also more prone to escalating hypertension and worsening proteinuria during their pregnancy, even if there is no superimposed pre-eclampsia. Up to doubling of the quantity of proteinuria is generally allowed as a physiological response to pregnancy, but greater than this needs investigation for possible pre-eclampsia or a lupus flare. Initially in all but those with very severe renal impairment, the usual increase in GFR occurs, leading to a fall in the serum creatinine level early in pregnancy. However, in those with moderate and severe renal impairment, the serum creatinine level usually begins to rise to and beyond pre-pregnancy levels during the second trimester. Effect of renal disease on pregnancy Women with chronic renal disease have an increased risk of both maternal and fetal complications during pregnancy. These include miscarriage, intrauterine growth retardation (IUGR), intrauterine death (IUD), preterm delivery (and the associated morbidity and mortality), and pre-eclampsia. The increased risk of pre-eclampsia applies even to those women whose renal function is good and BP normal on entering pregnancy. Twenty-five percent of those with a baseline creatinine ⬍125 ␮mol/L develop maternal complications, and this rises to 85% if creatinine is ⬎250 ␮mol/L.

lastly, the underlying type of renal disease. In general, women without hypertension or renal impairment prior to conception have successful pregnancies, and pregnancy does not adversely influence the progression of the renal disease. Degree of renal impairment The degree of renal impairment at conception strongly correlates with the likelihood of renal deterioration, maternal and fetal complications, and pregnancy success rates. This is demonstrated in Tables 1 and 2. Most of the studies comparing these various outcomes for the degree of renal dysfunction have based the latter on serum creatinine levels. These may be misleading if allowance is not made for the woman’s size, and further studies using GFR instead would be useful. The Cockcroft Galt equation to calculate GFR from serum creatinine is not applicable in pregnancy. Women with more severe renal impairment are more likely to have an accelerated decline and/or a permanent worsening of renal function as a result of the pregnancy (Table 1). If creatinine is ⬍125 ␮mol/L at conception then in only 2% will renal function deteriorate during pregnancy, with no significant progression to end-stage renal failure or deterioration post-partum. At the other end of the spectrum, if creatinine is ⬎170 ␮mol/L at conception, then in 65–75% renal function will worsen during pregnancy, 50–60% will have further deterioration post-partum and 33–40% will develop end-stage renal failure. Even in the middle ground, with creatinine 125–170 ␮mol/L, renal function will deteriorate further during pregnancy in 40%. Table 1

Effect of pregnancy on renal impairment

Degree of renal impairment

Mild

Moderate

Creatinine (␮mol/L) Loss of function Postpartum deterioration End-stage renal failure

⬍125 2% – –

125–170 40% 20% 2%

Severe 170–220 65% 50% 33%

⬎220 75% 60% 40%

Source: Nelson-Piercy C. Renal disease. In Handbook of Obstetric Medicine, third edition. Taylor Francis, 2006.

Table 2 Effect of degree of renal impairment on pregnancy outcome

Factors affecting outcome

Degree of renal impairment

Mild

Moderate

Severe

At the time of conception, there are a number of important factors that will influence the likelihood of a successful pregnancy outcome and any adverse effect of pregnancy on the underlying renal disease. First, the presence and degree of renal impairment; second, the presence and severity of hypertension; third, the presence and degree of proteinuria; and

Creatinine (␮mol/L) Maternal complications eg, pre-eclampsia IUGR Preterm delivery Pregnancy success

⬍125 25%

125–250 50%

⬎250 85%

– – 85–95%

30% 55% 60–90%

60% 70% 20–30%

Source: Nelson-Piercy C. Renal disease. In Handbook of Obstetric Medicine, third edition. Taylor Francis, 2006. Lupus

Renal lupus in pregnancy S Germain and C Nelson-Piercy

150

The severity of renal impairment also determines the risk of maternal and fetal complications during pregnancy. If creatinine is ⬍125 ␮mol/L there is a 85–95% pregnancy success rate, although 25% will develop maternal complications such as pre-eclampsia. Whereas, with creatinine ⬎250 ␮mol/L, pregnancy success rate is only 20–30%, 85% develop maternal complications, 60% of babies are growth restricted, and 70% preterm. Another important factor in determining development of fetal complications is maternal urea level. If urea ⬎10 mmol/L, polyhydramnios may develop, and the accompanying risks of preterm rupture of the membranes and cord prolapse. This is because the high maternal urea level leads to an osmotic load on the fetus, and resultant fetal polyuria. Once the maternal urea level is ⬎20–25 mmol/L, there is also a risk of fetal death. Dialysis and pregnancy Fertility is markedly reduced in women on haemodialysis or chronic ambulatory peritoneal dialysis (CAPD), and the pregnancy rate is only about one in 200 women per year. The chance of a successful pregnancy outcome is also reduced to about 30%. Poor prognostic features for pregnancy in dialysis patients include age ⬎35 years, more than five years on dialysis, and delayed diagnosis of pregnancy (leading to late increase in dialysis times). If a woman is able to become pregnant, then her dialysis requirements will markedly increase. Both the duration and frequency of dialysis usually need to be increased, to a total of ⬎20 hours per week, with sessions on five to six days per week. Continued adherence to fluid restriction is important to avoid large fluid shifts during dialysis, as fluctuations in blood pressure and fluid balance are already more common in pregnancy. Dietary restrictions can usually be lifted. The aim should be to maintain the predialysis urea at ⬍15–20 mmol/L, to avoid the fetal complications discussed previously. The procoagulable state of pregnancy may also mean increased heparin doses to prevent clotting of dialysis lines. Anaemia is exacerbated by pregnancy, and transfusion requirements increase. Erythropoietin and intravenous iron can be used safely in pregnancy, and doses or frequency may also need to be increased. Conversely, doses of vitamin D and calcium may need to be reduced. Pregnant women on dialysis have an increased risk of hypertension and pre-eclampsia, miscarriage, intrauterine death, preterm labour, preterm rupture of membranes, polyhydramnios related to uraemia, and placental abruption. The increased heparinization Lupus

requirements during haemodialysis also increase the risk of bleeding. Women on CAPD are at risk of peritonitis. Renal transplants and pregnancy Fertility usually returns to normal post-transplant, and it is important to educate these women regarding the need for contraception, as they may have been anovulatory and oligo/amenorrhoeic previously and not required contraception. If women wish to get pregnant, they should be advised to wait about one to two years after transplantation, with no recent episodes of rejection, to allow graft function to stabilize and maintenance levels of immunosuppressive drugs to be reached, to minimize any risk to the fetus. Successful pregnancy outcome for those transplant recipients who become pregnant and do not miscarry before 12 weeks is now 95%. Renal allografts adapt to pregnancy in the same way as normal kidneys, with an initial increase in GFR and dilatation of the collecting system, and then a fall in GFR in the third trimester. As discussed for chronic renal disease in general, any adverse effects of pregnancy on the renal allograft will depend on the baseline serum creatinine. Poor graft function at conception, higher baseline serum creatinine, and the presence of hypertension, increases the risk of deterioration in graft function. Pregnancy usually has no adverse long-term effect on renal allograft function or survival in women with baseline creatinine levels ⬍100 ␮mol/L. Conversely, renal graft survival is only 65% at three years, if women enter pregnancy with serum creatinine ⬎130 ␮mol/L. More than 10% of women are likely to develop new long-term problems following pregnancy, but it is difficult to ascertain whether this is actually as a direct result of pregnancy. The risk of long-term problems is higher in women developing pregnancy complications prior to 28 weeks’ gestation. About 10% of women will die within one to seven years after pregnancy, and about 50% within 15 years. Pregnancy outcome is best in those without hypertension, proteinuria, recent episodes of graft rejection, and in those with normal or near-normal renal function (serum creatinine ⬍125 ␮mol/L). For a baseline creatinine ⬍125 ␮mol/L the chance of a successful pregnancy outcome beyond 12 weeks is 97%, but this is reduced to 75% for creatinine ⬎125 ␮mol/L. Women with diabetes, and those with poor graft function, have a higher complication rate. The incidence of problems overall is about 50%, including graft rejection (10%), hypertension/pre-eclampsia (30%), IUGR (20–40%), preterm delivery (45–60%), and infection (especially urinary tract infection).

Renal lupus in pregnancy S Germain and C Nelson-Piercy

151

Renal lupus and pregnancy Flares during pregnancy Pregnancy increases the likelihood of a lupus flare, from about 40% to about 60%. These are not usually more severe than in non-pregnant women, and recent evidence does not suggest that flares are more likely immediately postpartum. Lupus flares may occur at any stage of pregnancy or the puerperium. It is not possible to predict when, or if, an individual patient will flare, although flare is more likely if disease has been active within six months of conception. The type of flare can to some extent be predicted by a woman’s previous disease patterns. Flares are not prevented with prophylactic steroids or routine increases of dose, so such prophylactic therapy is not recommended either ante- or postpartum. One difficulty diagnosing a flare during pregnancy is that many of the features also occur in normal pregnancy, such as oedema, palmar and facial erythema, fatigue, anaemia, raised ESR, musculoskeletal pain and hair loss. Another difficulty is that active lupus has overlapping features with other pregnancy complications. Lupus nephritis For a woman with SLE, renal involvement is no more common in pregnancy, although lupus nephropathy may manifest for the first time in pregnancy. As already discussed for chronic renal disease, pregnancy does not seem to jeopardize renal function in the long term in general, but this will depend on factors such as baseline creatinine. The presence of anti-phospholipid antibodies (anticardiolipin antibodies or lupus anticoagulant, APLs) does have a potential deleterious impact on renal outcome, as well as increasing the risk of thrombosis and adverse obstetric outcome. Women should be advised to delay pregnancy until at least six months after a lupus nephritis flare, to reduce their risk of a further flare during pregnancy. Worsening proteinuria in pregnancy could herald a lupus flare, but the differential diagnosis would also include the physiological response to pregnancy, the effect of discontinuing an ACEi during pregnancy, and pre-eclampsia. Up to a doubling of baseline proteinuria may be expected in pregnancy but more than this would be indicative of either worsening lupus nephritis or pre-eclampsia. Distinguishing between active renal lupus and pre-eclampsia is notoriously difficult, and the two conditions may be superimposed. Hypertension, proteinuria, thrombocytopenia and renal impairment are all signs they have in common, so

diagnosis of lupus flare requires other features, such as increasing anti-dsDNA titre, fall in complement levels C3 and C4, red blood cells or cellular casts in the urinary sediment, and other symptoms of flare such as arthralgia and skin rash. Conversely, hyperuricaemia and abnormal liver function tests are unusual in a lupus flare, and point more towards pre-eclampsia. The only definitive investigation to reliably differentiate between the two is a renal biopsy, but this is rarely undertaken in pregnancy. It may be indicated though prior to fetal viability, since confirmation of active lupus nephritis allows immunosuppressive treatment of the SLE, and delaying of delivery. Beyond 24–28 weeks’ gestation, when the fetus is viable, delivery may be the most appropriate course if the mother or her fetus is at risk. This will both cure pre-eclampsia and allow administration of drugs such as cyclophosphamide for a renal flare. Effect on pregnancy Overall, SLE pregnancies have an increased risk of spontaneous miscarriage, pre-eclampsia, IUGR, fetal death, and preterm delivery. The degree of risk depends on a number of factors including the presence of lupus nephritis, hypertension, APLs, active disease at the time of conception, and first presentation of SLE during pregnancy. Pregnancy outcome is particularly affected by renal disease, and even quiescent renal lupus is associated with increased risk of fetal loss, pre-eclampsia and IUGR, particularly if there is hypertension or proteinuria. Anti-phospholipid syndrome (APS) is also associated with a worse prognosis. Conversely, for women in remission, or without the risk factors mentioned above, the risk of pregnancy loss and pre-eclampsia is probably no higher than in the general population. Another complication of pregnancy in women with SLE or APS, although very rare, is chorea.

Management of renal disease in pregnancy Pre-pregnancy counselling Management should begin with pre-pregnancy counselling. Assessment of pre-conceptual renal function, proteinuria, blood pressure, anti-Ro/La antibodies and APLs enables accurate counselling and provides a baseline with which to compare trends in pregnancy. In view of the increased risk of pre-eclampsia, treatment with low-dose aspirin should be considered, especially in those with hypertension and renal impairment or a previous poor obstetric history, and folic acid should be started. Timing of pregnancy should be discussed, inparticular, stressing that outcome is improved if Lupus

Renal lupus in pregnancy S Germain and C Nelson-Piercy

152

conception occurs during disease remission and at least six months after a flare. Realistic, evidence based, estimates should be given for likely success and chance of complications, and the possibilities of prematurity and handicap discussed. If renal impairment is severe (baseline creatinine ⬎250 ␮mol/L), then women should usually be advised against pregnancy, because of the very high risks of maternal and fetal complications and low chance of success, and appropriate contraception discussed.

Table 3

Use of drugs in pregnancy

Yes

No/caution

Cyclosporin Tacrolimus Azathioprine Prednisolone Penicillins Gentamicin Alpha-blockers Calcium antagonists

MMF Cyclophosphamide Rapamycin ACE inhibitors and A2RB (OK in first trimester) Beta-blockers (OK in later pregnancy) Diuretics Statins

ACE: angiotensin converting enzyme; A2RB: angiotensin II receptor blockers.

General management Pregnancy care is best undertaken by a multidisciplinary team in combined clinics, involving both physicians and obstetricians with expertise in the care of renal disease in pregnancy. It is important to establish baseline values in early pregnancy for Fbc, U and E, serum creatinine, uric acid, liver function, anti -dsDNA and complement titres and to quantify any proteinuria. The woman should have regular assessment of renal function by serum creatinine and urea, as well as creatinine clearance and 24-hour protein excretion or protein creatinine ratio. It may be useful to give the woman urine testing strips so she can monitor the presence and severity of any proteinuria or haematuria. The woman should be assessed regularly for evidence of disease activity, and other markers, such as dsDNA and complement titres, repeated as indicated. Anaemia is common and haematinics should be prescribed. Maternal hypocalcaemia and hypercalcaemia are both potential problems, and calcium status should be carefully monitored. Doses of calcium and vitamin D may need to be altered in pregnancy. The fetus should be monitored with regular ultrasound scans for growth and liqor volume. Doppler assessment of uterine artery blood flow at 20–24 weeks is useful in predicting pre-eclampsia and IUGR, and assesment of the umbilical flow is helpful in the presence of IUGR. As already discussed, the indications for renal biopsy during pregnancy are mostly limited to situations where a delay before delivery is desirable (ie, before 32 weeks’ gestation) and a diagnosis of a steroid or chemotherapy-responsive lesion is suspected. See Table 3 for a summary of the safety of the various immunosuppressant, anti-hypertensive, and other drugs in pregnancy. They are discussed in more details in the following sections. Management of lupus flares and medication Hydroxychloroquine is often used to prevent flares. It should not be stopped in early pregnancy, as this Lupus

could precipitate a flare, and its long half-life means that the fetus would continue to be exposed to the drug for several weeks even after discontinuation. Doses at the level used for malarial prophylaxis appear to be safe for the fetus, but there have been concerns that higher doses used for rheumatic disorders, could result in fetal retinopathy. Evidence from a randomized control trial of pregnancies in women exposed to hydroxychloroquine though, have shown the congenital abnormality rate is no higher than the background population, and there were fewer flares, less prednisolone used, and a decreased SLEPDAI (SLE in Pregnancy Disease Activity Index). Disease flares must be actively managed. Corticosteroids are the drugs of choice, usually increased oral prednisolone or pulsed intravenous methyl predisolone. Pregnant women receiving steroids are at increased risk of gestational diabetes mellitus and of premature rupture of membranes. If they are on a dose of ⬎7.5 mg prednisolone for ⬎2 weeks, then parenteral steroids will be required to cover the stress of labour and delivery, regardless of the mode of delivery. Azathioprine can be added as a steroid – sparing agent. It is safe to use in pregnancy, with no adverse fetal effects reported despite many years experience with its use. The fetus lacks the enzyme to convert azathioprine to its active metabolites. There are also only very low levels of the active metabolites of azathioprine in breast milk, and some authorities would argue that the benefits of breast feeding outweigh any potential risk of neonatal immunosuppression. NSAIDs can be used as second line treatment, in the first and second trimesters, but are usually stopped in the third trimester by 32–34 weeks, because they can cause oligohydramnios (due to effects on the fetal kidney), premature closure of the ductus arteriosus (due to inhibition of prostaglandin synthetase), and neonatal haemorrhage (due to inhibition of platelet function).

Renal lupus in pregnancy S Germain and C Nelson-Piercy

153

Like azathioprine, mycophenolate mofetil (MMF) is an antiproliferative immunosuppressant. It is however more selective than azothioprine and is gaining popularity for use in SLE as well as transplantation. It will be discussed in more detail in the section regarding management of renal transplants (see below). Cyclophosphamide carries an increased risk of congenital defects (16–22%) and it should be discontinued at least three months pre-conception. Rarely, it is indicated later in pregnancy for severe maternal disease. Admission should be considered if the woman develops worsening hypertension, deteriorating renal function or proteinuria, superimposed pre-eclampsia or polyhydramnios. Anti-hypertensives Careful monitoring and control of blood pressure both pre-pregnancy and antenatally is important. Treatment for hypertension is no different from the management of pregnant women without renal disease, however the threshold for treatment may be lower, since good control of hypertension is important to preserve renal function. For control of hypertension, the drug of choice is methyldopa, with nifedipine or hydralazine as secondline agents. Long-term use of hydralazine and methyldopa may rarely induce a SLE-like syndrome, but they are not contraindicated in SLE. Beta-blockers have been associated with impaired fetal growth when use long-term throughout pregnancy and started in the first trimester, although labetalol may be used as third line treatment in the second or third trimesters, or intra-partum parenterally for acute severe hypertension. The angiotensin-converting enzyme (ACE) inhibitors should not be used in pregnancy because they may cause oligohydramnios, hypocalvaria, renal tubular dysgenesis, renal failure, hypotension, decreased skull ossification in the fetus and an increased risk of intrauterine death. Women on ACE inhibitors should be swapped to an alternative agent, usually methyldopa, when pregnancy is confirmed. This does not need to be done preconception as ACE inhibitors are not associated with structural malformations in the first trimester. There is less evidence available for the use of angiotensin II receptor blockers in pregnancy, but as they are similar to the ACE inhibitors, should also be avoided. Management of women with renal transplants Most of the management is the same as detailed above. Additional points concern the increased risk of infection, side-effects of drugs and rejection.

An MSU specimen should be taken and sent at each visit and any infection treated promptly. Some women require prophylactic antibiotics. Cytomegalovirus (CMV) titres should be checked in each trimester if the woman is CMV negative at the onset of pregnancy. If renal function is deteriorating, then additional differential diagnoses are cyclosporin nephrotoxicity, and acute and/or chronic rejection. Features of acute rejection include fever, oliguria, graft swelling and tenderness, and altered echogenicity of renal parenchyma and blurring of corticomedullary junction on ultrasound. The definitive diagnosis of rejection is only possible with renal biopsy. The levels of immunosuppressive drugs are usually maintained at pre-pregnancy levels, although sometimes doses of cyclosporin need increasing. Women will require reassurance regarding the relative safety of their drugs, as reduction or cessation of immunosuppressive therapy may provoke rejection. Prednisolone and azothioprine have already been discussed, and the dose of the latter may be monitored via maternal white-cell count. Other drugs include cyclosporin and tacrolimus, both of which appear to be safe for use in pregnancy, although plasma levels should be measured regularly. Pregnancy success rates are similar in women taking azathioprine and cyclosporin, but the incidence of IUGR is higher (30–40% versus 20%) in women taking cyclosporin. MMF is generally contraindicated in pregnancy, as it is toxic in animals, and the limited data available from human pregnancies suggest an increased risk of malformations. However, there is no pattern to the reported malformations and very few reported cases are available for analysis. In women where MMF is used because of an episode of rejection and it is deemed to be the only drug to adequately control disease or rejection, changing to a safer alternative such as azathioprine in preparation for pregnancy may not be appropriate. After counselling, women may opt to go ahead with pregnancy despite the unknown risk of teratogenesis, when balanced against the risk of deterioration in renal function if MMF is stopped. Caesarean section is only required for obstetric indications, although the overall section rate is increased compared to background rates. The renal allograft does not obstruct vaginal delivery. Prophylactic antibiotics should be given to cover any surgical procedure, including episiotomy. Neonatal problems are largely related to prematurity, but also include thymic atrophy, transient leukopenia or thrombocytopenia, septicaemia, and CMV and hepatitis B infection. Congenital abnormalities are no more common in the offspring of mothers taking anti-rejection doses of the earlier mentioned immunosuppressive drugs, apart from possibly MMF. Lupus

Renal lupus in pregnancy S Germain and C Nelson-Piercy

154

Acute renal failure in pregnancy Incidence Acute renal failure (ARF), on the background of normal renal function, is rare in pregnancy in the West (⬍0.005%), although remains a common cause of maternal mortality in the developing world. Transient mild-to-moderate renal impairment is more common. ARF usually presents post-partum. Pregnant women with underlying renal disease, such as lupus nephritis, are at increased risk of a number of complications that can lead to acute on chronic renal failure.

Management This obviously depends on the underlying cause, and is generally the same as for the management of ARF in non-pregnant individuals, although there are a few points specific to pregnancy and pregnancy-related conditions. Fluid overload must be prevented, especially in pre-eclampsia, because of the susceptibility of these women to pulmonary oedema. Plasmapheresis is not needed for HELLP syndrome, which usually improves with conservative therapy.

Conclusion Diagnosis The causes of ARF in pregnancy are listed in Table 4. Many are associated with a coagulopathy. The commonest cause of ARF in the context of pre-eclampsia is HELLP (Haemolysis, Elevated Liver enzymes, and Low Platelets) syndrome (about 50%). The underlying cause of ARF may be obvious, for example in the case of abruption and postpartum haemorrhage, although abruption occurs in 16% of women with HELLP syndrome and this may be the true underlying cause. Blood loss may not be recognized or may be underestimated, and hypotension may be absent or masked by co-existent pre-eclampsia. If ARF develops, especially post-partum, in a woman with features of pre-eclampsia, combined with a microangiopathic haemolytic anaemia and thrombocytopenia, it may be difficult to differentiate between thrombotic thrombocytopenic purpura (TTP)/haemolytic uraemic syndrome (HUS), HELLP syndrome and acute fatty liver of pregnancy (AFLP). Indeed the conditions are closely related and HUS or AFLP may evolve from HELLP. HELLP syndrome is far more common, and is characterized by abnormal liver function, a coaguloapthy (not seen in HUS) and a lower grade haemolysis. Pointers to HUS are profound thrombocytopenia, and florid microangiopathic haemolytic anaemia. Features of AFLP are abormal liver function, hypoglycaemia, high uric acid, and profound coagulopathy.

Important physiological adaptations to pregnancy present a challenge to the clinician managing women with renal disease in pregnancy. Minor degrees of renal impairment will be missed unless pregnancyspecific normal ranges for serum creatinine are used. Lupus flares are more common in pregnancy and the puerperium but they are harder to diagnose because many of the symptoms, signs and laboratory abnormalities may occur in normal pregnancy. A particular difficulty is the diagosis of pre-eclampsia which is more common in lupus nephritis, when there is preexisting hypertension and worsening proteinuria. Antiphospholipid antibodies increase the risk of capillary thrombotic microangiopathy in the kidney, renal artery stenosis and adverse pregnancy outcome. Most of the drugs used to manage SLE and after renal transplantation are safe in pregnancy. Pre-pregnancy counseling ensures thorough assessment of disease activity and complications, and outcome for women with lupus nephritis is optimal when disease has been quiescent for at least six months at conception. Pregnancy in these complicated women should be managed with multidisciplinary care in combined clinics with involvement from obstetricians, rheumatologists, obstetric physicians, nephrologists, fetal medicine practitioners, haematologists, neonatologists and anaesthetists.

References Table 4 Causes of acute renal failure in pregnancy Infection Blood loss Volume contraction Post-renal failure Drugs Other

Septic abortion; puerperal sepsis; pyelonephritis Postpartum haemorrhage; placental abruption Pre-eclampsia; eclampsia (6%); HELLP syndrome (7%); hyperemesis gravidarum Ureteric damage or obstruction NSAIDs; antibiotics HUS/TTP; AFLP (60%); glomerulonephritis

HELLP: haemolysis, elevated liver enzymes, and low platelets; NSAIDs: non-steroidal anti-inflammatory drugs; HUS: haemolytic uraemic syndrome; TTP: thrombotic thrombocytopenic purpura; AFLP: acute fatty liver of pregnancy. Lupus

1 Armenti VT, Radomski JS, Moritz MJ, Gaughan WJ, McGrory CH, Coscia LA. Report from the National Transplantation Pregnancy Registry (NTPR): outcomes of pregnancy after transplantation. Clin Transpl 2003; 131–141. 2 Davison J, Baylis C. Renal disease. In de Swiet M ed. Medical Disorders in Obstetric Practice, 4th edn. Blackwell Science, 2002: 198–266. 3 Epstein FH. Pregnancy and renal disease. N Engl J Med 1996; 335: 277–278. 4 Hou SH. Pregnancy in women with chronic renal insufficiency and end stage renal disease. Am J Kid Dis 1999; 33: 235–252. 5 Jones DC. Hayslett JP. Outcome of pregnancy in women with moderate or severe renal insufficiency. N Engl J Med 1996; 335: 226–232.

Renal lupus in pregnancy S Germain and C Nelson-Piercy

155 6 Jungers P, Houllier P, Forget D et al. Influence of pregnancy on the course of primary chronic glomerulonephritis. Lancet 1995; 346: 1122–1124. 7 Jungers P, Chauveau D. Pregnancy in renal disease. Kidney Int 1997; 52: 871–885. 8 Moroni G, Ventura D, Riva P et al. Antiphospholipid antibodies are associated with an increased risk for chronic renal insufficiency in patients with lupus nephritis. Am J Kidney Dis 2004; 43: 28–36.

9 Moroni G, Ponticelli C. The risk of pregnancy in patients with lupus nephritis. J Nephrol 2003; 16: 161–167. 10 Ostenson M. Disease specific problems related to drug therapy in pregnancy. Lupus 2004; 13: 746–750. 11 Oviasu E, Hicks J, Cameron JS. The outcome of pregnancy in women with lupus nephritis. Lupus 1991; 1: 19–25.

Lupus