A retrospective analysis of 358 pregnant women with CKD in the United Kingdom found an association between diastolic blood pressure (BP) above 90 mm Hg (treated or untreated) and neonatal death. This relatively mild degree of hypertension also compounded the risk of preterm birth resulting from renal impairment alone.³

Progression of Chronic Kidney Disease

The risk of progression of renal disease during the pregnancy likewise depends less on the specific renal disorder and more on baseline serum creatinine, control of hypertension, and increasing proteinuria or onset of superimposed preeclampsia. The natural history of most mild CKD is that accelerated deterioration during a pregnancy is unlikely; the main exception is lupus nephritis, which can flare, leading to worsening renal disease during the pregnancy. However, 50% of women with moderate baseline renal impairment (serum creatinine >1.5 mg/dl; 130 µmol/l) have a significant rise in serum creatinine in the third trimester or early postpartum. If this occurs, almost one in five progress to end-stage renal disease (ESRD) within 6 months after delivery.

Prematurity or Fetal Growth Restriction

Prematurity, fetal growth restriction, and stillbirth are the major concerns for pregnancies of women with renal impairment, particularly if superimposed preeclampsia develops. Uncontrolled hypertension at conception is a poor prognostic feature for fetal outcome (Box 45-2). A new reason to avoid prematurity, besides the immediate effects on the baby, is the association with higher mortality in young adulthood.⁴

Pre-pregnancy Counseling

It has long been known that any woman with CKD stage 3 to 5 should receive pre-pregnancy counseling. We now know that this should also apply to women with CKD stage 1 or 2, particularly if they have hypertension, significant proteinuria, a poor obstetric history, recurrent urinary tract infection (UTI), inheritable renal disease, or other disease (e.g., SLE) likely to worsen during pregnancy. Studies from Italy have found that preterm delivery, cesarean rates, and need for neonatal intensive care were higher than in the general population, even in women with stage 1 CKD; this was particularly true if baseline proteinuria was more than 1 g/day.^5,6 Another meta-analysis included older studies but also a large number of women in stage 1 CKD; both maternal and fetal outcomes were worse in women with any form of CKD.⁷ A study from Arabian Gulf countries examined outcomes of women with stage 1 and 2 CKD and also showed unexpectedly high rates of preterm delivery, fetal growth restriction, and fetal death in women with eGFR as high as 60 to 89 ml/min.⁸ Although this may not be a worldwide experience, it now seems prudent to follow all women with any CKD more closely during pregnancy, not just those with impaired GFR or proteinuria above 1 g/day. This is more relevant because CKD stage 1 or 2 affects 3% of women of childbearing age, whereas CKD stage 3 to 5 affects about 1 in 150 of these women.⁹ Box 45-4 summarizes issues that should be covered in counseling.

During normal pregnancy, GFR rises by about 50%, typically apparent by the end of the first trimester (see Chapter 43). Serum creatinine greater than 1 mg/dL (88 µmol/l) in a pregnant woman generally indicates reduced GFR. The Modification of Diet in Renal Disease (MDRD) or Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations or other formulas that estimate GFR are not valid for pregnancy; whereas it is reasonable to assess pregnancy risks based on pre-pregnancy eGFR, these formulas should not be applied in pregnancy. Measurement of creatinine clearance is the only clinical way of truly assessing GFR in pregnancy but requires 24-hour urine collection, which is cumbersome, and even when conducted diligently, may be inaccurate because of ureteral dilation, which results in pooling of urine and an incomplete collection. Serum cystatin C is being evaluated but also appears inaccurate. Serum creatinine remains the clinical standard for assessing GFR during pregnancy.

Blood Urea Nitrogen

Increments in blood urea nitrogen (BUN) and creatinine may indicate deteriorating GFR and the need for initiation of dialysis if the pregnancy is to succeed (see Dialysis in Pregnancy). However, a rising BUN, particularly when accompanied by rising hemoglobin or hematocrit, may represent intravascular contraction, typically seen in preeclampsia. The main role for BUN in pregnancy is for monitoring the effectiveness of dialysis.

Serum Electrolytes, Albumin, and Volume Homeostasis

Serum sodium and bicarbonate are slightly reduced, potassium is at the lower end of the normal range, and albumin and uric acid are lower than in the nonpregnant state. Increases in serum sodium to those of nonpregnant women should suggest (reversible) pregnancy-specific diabetes insipidus (from excess placental vasopressinase). In general, this is a mild disorder, but DDAVP should be given if serum sodium rises above 150 mmol/l with low urine osmolality.

Adequate intravascular volume is essential to preservation of GFR and good pregnancy outcome for mother and baby. Clinically, it is difficult to assess maternal volume homeostasis. Edema is an unhelpful sign in pregnancy, so hematocrit should be measured in women with underlying CKD at the initial first trimester visit, along with serum albumin. Both measures should fall slightly as pregnancy progresses. A rise in either value strongly suggests intravascular volume contraction, although there is no absolute discriminant value. Conversely, a significant fall in either value does not by itself diagnose excessive volume expansion, because the hematocrit depends on other factors, and serum albumin may fall in patients with nephrotic syndrome, who in turn may have reduced intravascular volume.

In practice, provided there is no respiratory compromise and BP can be controlled, volume excess is more favorable for maternal renal function and fetal growth than volume depletion. When there is concern about fetal growth or deteriorating GFR in women with CKD, and reduced intravascular volume is suggested by the change in hematocrit and albumin from baseline, a trial of intravenous normal saline (no more than 1 liter) under observation in hospital is a reasonable clinical approach.

Urinalysis

A dipstick test showing negative or trace protein in most (but not all) cases excludes true proteinuria. Any woman with 1+ (0.3 g/L) proteinuria or above should have formal quantification of protein excretion either with a spot urine protein-creatinine (PC) ratio or 24-hour urine collection. Glycosuria may be normal and is not itself diagnostic of diabetes mellitus, but when glycosuria is detected in early pregnancy in a woman with CKD, a 75-g oral glucose tolerance test should be performed. Dipstick hematuria during pregnancy is common and often resolves after delivery. Provided there is no UTI, urine sediment is inactive, and serum creatinine normal, this is not associated with adverse maternal or fetal outcomes during pregnancy and can be investigated if persistent postpartum.¹⁰ Pregnancy may be the woman’s first urinalysis, so previously undetected renal disorders are diagnosed at this point.

Proteinuria and Nephrotic Syndrome

The upper limit of normal protein excretion during pregnancy is defined as 300 mg/day. In a midstream specimen of urine, a PC ratio greater than 30 mg/mmol (0.27 mg/mg) correlates with more than 300 mg/day proteinuria.¹¹

Hypertension

Hypertension in pregnancy is defined as blood pressure above 140/90 mm Hg, and this is generally the threshold for treatment, with an acceptable range on treatment of 110 to 140 systolic and 80 to 90 mm Hg diastolic. This range is not based on solid pregnancy outcome data but is thought to be the range that reduces maternal risk from severe hypertension while providing sufficient systemic BP to maintain placental perfusion. Although the target BP for nonpregnant women with CKD is also less than140/90 mm Hg, it is probably lower (≤125/75 mm Hg) if proteinuria is more than 1 g/day; therefore a pregnant woman with proteinuric CKD may have about 40 weeks when BP is above usual target, which may contribute to progressive renal impairment postpartum. Nevertheless, if a pregnant woman develops BP below 110/80 mm Hg, my practice is to reduce antihypertensive drugs and avoid the risk of fetal hypoperfusion.

Most pregnant women with CKD will not exhibit the usual first-trimester fall in BP, and in many, BP increases as the pregnancy progresses. The reason is not clear. Normal pregnancy is accompanied by significant volume expansion, which does not usually induce hypertension. In CKD, however, there is often inability to excrete a sodium load with accompanying hypertension, and this mechanism likely contributes in pregnancy. Other factors almost certainly playing a role in this hypertension include stimulation of the renin-angiotensin and sympathetic nervous systems, alterations in endothelial factors such as prostacyclin, nitric oxide, and endothelin, and particularly in transplant patients, the drugs used, such as calcineurin inhibitors and corticosteroids. Regardless of its cause, persistence of hypertension is an adverse factor in pregnancy outcome.¹⁴ Concern about using antihypertensives in pregnancy has been associated with poorer pregnancy outcomes, at least in women with renal transplant.¹⁵

Blood pressure will often rise significantly soon after delivery. Therefore, BP measurement and treatment should be undertaken diligently in the early postpartum period.

Control of BP is imperative to successful pregnancy outcome in women with underlying renal disease. Diuretics are not recommended during pregnancy because any reduction in maternal plasma volume may have adverse effects on uteroplacental or renal per­fusion. Angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs) must be discontinued, preferably before pregnancy but certainly as soon as pregnancy is diagnosed, because of increased risks of fetal growth restriction, oligohydramnios, neonatal renal failure, and probably cardiac and neurologic developmental abnormalities. Aldosterone antagonists should also be avoided, and atenolol has been associated with fetal growth restriction. Suitable antihypertensives include methyldopa, labetalol, oxprenolol, hydralazine, prazosin, and nifedipine, and all may be used in conventional doses. Diltiazem may also have a small benefit on reducing proteinuria, but this has not been addressed specifically for pregnancy.

Superimposed Preeclampsia

Preeclampsia is a placental disorder of unknown etiology that has several predisposing risk factors, one of which is CKD (see Chapter 44). The maternal renal effects of preeclampsia include a reduction in renal blood flow, increased sodium and uric acid reabsorption, reduced circulating renin and aldosterone concentrations, proteinuria, and impaired GFR. Superimposed preeclampsia in a woman with underlying renal impairment will lead to worsening renal function, exaggerated hypertension and proteinuria with risks of nephrotic syndrome, short-term and long-term risks to maternal renal function, as well as increased risks for fetal growth restriction, prematurity, and perinatal mortality.

However, it is difficult to diagnose superimposed preeclampsia in a woman who begins her pregnancy with renal impairment or proteinuria. An increase in BP, decline in GFR, increased plasma uric acid, or increasing protein excretion can all be caused by progression of the underlying renal disorder rather than superimposed preeclampsia, and no definitive diagnostic test yet exists. However, when these features are accompanied by neurologic symptoms, such as hyperreflexia with clonus, or by abnormal liver transaminases or new-onset thrombocytopenia (except in SLE), it is highly likely that superimposed preeclampsia has developed. Typically, these changes occur after 20 weeks’ gestation. It has been proposed that an increased ratio of soluble fms-like tyrosine kinase 1 (sFlt-1) to placental growth factor (PlGF) is diagnostic of preeclampsia and distinguishes this condition from pregnant women with CKD.¹⁶ Although prom­ising, incorporated this into routine clinical practice is not yet recommended.

Diagnosing superimposed preeclampsia in women with CKD is somewhat unnecessary. Clinicians managing the woman with underlying renal disease should be vigilant for changes in maternal and fetal condition in all cases, and the indications for delivery in women with preeclampsia are broadly the same as those in women with progressive underlying renal disease (Box 45-5). On the other hand, it is now well recognized that preeclampsia is associated with long-term cardiovascular, metabolic, and renal risks for the mother so it is practical to error on the side of caution and “overdiagnose” this disorder because this may prompt better long-term follow-up.¹⁷

Limited studies suggest that aspirin is of benefit in reducing superimposed preeclampsia and perinatal death in women with underlying renal disease. A review found that the number needed to treat was 9 to 57 women for prevention of preeclampsia and 42 to 357 to prevent perinatal death.¹⁸ The PARIS study also found that aspirin reduced superimposed preeclampsia in women with underlying renal disease, but only 450 women were included in this analysis.¹⁹ My practice is to use low-dose aspirin (100 to 150 mg/day) for all women with CKD and creatinine greater than 1.5 mg/dl (130 µmol/l) or for those with lower creatinine who have had early-onset or severe preeclampsia or fetal loss in a previous pregnancy. The effects of this dose of aspirin on renal function are minimal, and generally this is a safe approach. I also use calcium (1200 mg/day) as prophylaxis against preeclampsia when there is no concern about hypercalcemia. Data for calcium as prophylaxis are less convincing than for aspirin, but when used carefully, calcium has minimal effects and may have benefit.