Hypertension in the setting of pregnancy is classified as chronic hypertension, gestational hypertension, or preeclampsia (Table 1). This case highlights the management of chronic hypertension in pregnant patients with CKD. Hypertension guidelines for the general CKD, nonpregnant population recommend a target blood pressure of <140/90 mmHg and advocate for a lower target of <130/90 mmHg for patients with proteinuria [8, 9]. These guidelines are based on large randomized controlled studies in thousands of patients that addressed the risks of overall mortality, cardiovascular and renal morbidity, stroke, coronary interventions, and end-stage renal disease (ESRD). During pregnancy, the goals for blood pressure treatment differ from goals in the general population. Instead of long-term goals, the goals are short term and directed at avoiding immediate end-organ damage to the mother and fetus while limiting the risks of altered uteroplacental perfusion which might ultimately affect fetal growth and development [10].

Strict blood pressure targets in pregnancy have been extensively studied as a potential strategy to lower the risk of preeclampsia and pregnancy complications, but results have been disappointing. Several meta-analyses including a 2014 Cochrane review showed that treatment of mild to moderate hypertension during pregnancy does not decrease the risk of preeclampsia, neonatal death, preterm birth, or small-for-gestational-age babies [11].

Given the lack of definitive evidence, guidelines differ on the appropriate blood pressure target in pregnancy. The National Institute for Health and Clinical Excellence guidelines recommend a target blood pressure of <150/100 mmHg during pregnancy, whereas the American College of Obstetricians and Gynecologists recommends a target of <160/110 mmHg [12, 13]. There is strong evidence to suggest that severe hypertension ≥160/110 mmHg is associated with severe complications of maternal stroke and fetal abruption; therefore, severe hypertension should always be treated. Care of patients with hypertension and CKD adds a layer of complexity, and some advocate for slightly lower target of <140/90 mmHg although there is no data to support this recommendation [14].

In the largest study to date, 987 pregnant women with hypertension were randomized to either strict blood pressure control with a target diastolic blood pressure (DBP) of <85 mmHg vs. a target DBP <100 mmHg. Both groups had the same rates of pregnancy loss, need for high-level neonatal care, and maternal complications. The group randomized to less tight control had higher rates of severe maternal hypertension. Overall, this study showed no significant difference in the rates of serious maternal complications and major adverse perinatal outcomes with less tight versus tight control of blood pressure in pregnancy [15]. Of note, none of the patients enrolled in the study had CKD.

Ideally, hypertension management should begin before conception. Many women with CKD take angiotensin-converting enzyme inhibitors (ACE inhibitors) and angiotensin II receptor blockers (ARBs) as they are recommended for patients with proteinuric CKD to decrease proteinuria and slow progression of kidney disease [16, 17].

In pregnancy, however, ACE inhibitors cross the placenta in pharmacologically significant amounts and have a well-documented pattern of fetal risks with second- and third-trimester exposure. These fetotoxic effects include renal tubule dysplasia, renal agenesis, and other fetal abnormalities including oligohydramnios, hypoplastic lungs, hypocalvaria, and neonatal hypotension, likely stemming from a decrease in fetal angiotensin or an increase in fetal bradykinin [18]. It is generally assumed that ARBs will lead to similar complications, although published data is limited [19, 20]. The Food and Drug Administration (FDA) has labeled ACE inhibitors as contraindicated in the second and third trimesters since 1986; however, first-trimester exposure has been controversial, making prepregnancy counseling and management of young women with chronic kidney disease less straightforward. The evidence for first-trimester risk of congenital abnormalities with ACE inhibitor exposure is conflicting: a retrospective study linked first-trimester prescriptions for ACE inhibitors to a significantly greater risk of serious fetal abnormalities [21], whereas a large cohort study and systematic review did not show an increase in risk [20, 22]. It is possible that this discrepancy is related to characteristics of the hypertensive pregnant population such as undiagnosed diabetes, maternal obesity, or hypertension itself. In the general population, the decision to stop ACE inhibitors prior to conception and control blood pressures with pregnancy-safe medications is obvious. In chronic kidney disease patients, especially with significant proteinuria, the decision is more difficult since ACE inhibitors appear to slow the progression of disease in those with significant proteinuria. Increasingly, experts in the field recommend a more tailored approach to the use of ACE inhibitors in women who are considering pregnancy [23, 24]. In the patient who is able to work closely with her care team and is at high risk for progression without an ACE inhibitor, such as those with proteinuria of >1 g/day, it may be reasonable to continue the ACE inhibitor until pregnancy is confirmed in the first trimester, particularly when conception may take months to years. Yet, the decision to conceive on an ACE inhibitor requires a careful discussion with the patient regarding potential risks and benefits. If a woman chooses to conceive on an ACE inhibitor, she is instructed to discontinue the drug when she obtains a positive pregnancy test to limit exposure and arrange to be seen promptly to confirm pregnancy. At that time, if necessary, pregnancy-safe antihypertensive agents can be used and be prescribed. In patients who are less likely to identify pregnancy early or those whose blood pressure may not be readily controlled without ACE inhibitors, a preconception change in regimen is more appropriate.

Methyldopa, labetalol, and nifedipine are considered first-line agents in pregnant patients or patients with hypertension who are trying to conceive as they have been studied in pregnancy and appear to be the safest [25–27], though methyldopa is used less often because it commonly causes nausea and fatigue and may increase liver enzymes. Thiazide diuretics are considered second-line treatment but are likely to be safe [28]. Patients with chronic hypertension should be seen in the office every 2–4 weeks in the first two trimesters, and visit frequency should be increased as needed, depending on the clinical course. Some suggest that women with chronic hypertension also monitor their blood pressure at home. In patients whose measurements correlate with office measurements, this can be a helpful adjunct to care but would not replace office evaluation of blood pressure. During office visits, urine protein should also be measured.

In this vignette, blood pressure variations mirrored the normal trends in blood pressure during pregnancy. The patient’s blood pressure medication requirement decreased during the first half of pregnancy, and then her blood pressure increased toward her baseline during the third trimester. This normal trend is worth noting so that it is not mistaken for the onset of preeclampsia.

The classification of CKD in pregnancy has traditionally differed from the classification of CKD in the general population. This is a reflection of the fact that these considerations preceded the newer “stages” of CKD that have been embraced by the National Kidney Foundation. The latter uses estimated GFR (eGFR) rather than the serum creatinine, whereas, in pregnancy, risk is separated into three categories: mild, moderate, and severe based solely on the value of the serum creatinine prior to conception. These categories are also defined slightly differently in various publications, which make it more difficult to counsel patients. More recent publications have begun to consider pregnancy risk in the context of the estimated GFR and also consider proteinuria [29]. See Fig. 2 for a summary of risks by classification.

For the general population, formulas such as the Cockroft-Gault or the Modification of Diet in Renal Disease (MDRD) equation are used to estimate GFR. These formulas are used when the renal function is stable and include the serum creatinine and additional factors such as age, weight, gender, ethnicity, blood urea nitrogen, and serum albumin. None of these formulas perform well when the renal function is close to normal or when there is acute kidney injury. Importantly, these formulas also have not been validated for pregnant women [30]. The performance of MDRD equation in pregnancy may substantially underestimate GFR in pregnancy compared to GFR measured by inulin clearance [31]. Reliance on the MDRD formula during pregnancy is not recommended because of this discrepancy. Nevertheless, these estimates have become the standard for measurement of eGFR in the nonpregnant population and therefore are readily available and can be useful in preconception counseling. For comparison, a 30-year-old nonpregnant woman whose serum creatinine is 1.4 mg/dL would have an eGFR of approximately 50 ml/min/1.73 m² using several different formulas.

The degree of proteinuria plays a very important role in determining renal prognosis for patients with CKD who are not pregnant. In fact, this characteristic is so important that the Kidney Disease: Improving Global Outcomes guidelines recommend classifying patients based on both eGFR and level of albuminuria [32]. Those with the most albuminuria are the most likely to have a further decline in renal function. It is likely that significant proteinuria also increases the risk of adverse pregnancy outcomes for women [29, 33]. The gold standard for measuring proteinuria remains the 24-h urine collection (with measurement of the urinary creatinine to assure a complete collection). Significant proteinuria in pregnancy is defined as proteinuria ≥0.3 g/day. Yet this test is difficult to do repeatedly because the sample sometimes requires refrigeration and its collection is cumbersome and time consuming. Instead, a random urine specimen that measures protein and creatinine can be used to calculate the urine protein-to-creatinine (urine PrCr) ratio. This value approximates the number of grams per day of urinary protein. Normal urine PrCr is < 0.2 mg/mg (20 mg/mmol) so a ratio of >0.3 mg/mg (0.30 mg/mmol) represents significant proteinuria in singleton pregnancy [34]; a threshold up to 0.4 mg/mg (40 mg/mmol) may be more appropriate in multiple pregnancy [35, 36].

A urinary albumin-to-creatinine ratio (ACR) can also be used since albumin is the major protein in blood and, therefore, the major protein in urine when there is proteinuria from altered glomerular hemodynamics or glomerular disease. The ACR was initially popularized to detect minute amounts of urinary albumin, too small to be detected by the urinary dipstick, and the units of measurement are three orders of magnitude smaller than the urine protein-to-creatinine ratio. Normal ACR is <30 mcg/mg creatinine. The urinary dipstick typically turns positive at approximately 300 mcg/mg creatinine. The sensitivity and specificity for both the ACR and urine PrCr are excellent when the urinary protein is very low [37]. Although the urinary ACR test has not been validated in pregnancy, it is recommended for screening nonpregnant diabetic patients and routinely used instead of 24-h urine collections to assess proteinuria in patients with CKD. Thus, either the urine PrCr or the urinary ACR can be used to follow patients who have CKD and are pregnant. Although these “spot” tests are less accurate than the 24-h collection, they are clearly more accurate than the urinary dipstick alone since the latter is dependent on the concentration of the urine.

All women with underlying kidney disease are at increased risk of both maternal and fetal complications in pregnancy [38]. A systematic review of pregnancy outcomes in chronic kidney disease revealed that women with CKD appear to have at least a twofold higher risk of developing adverse maternal outcomes (gestational hypertension, preeclampsia, eclampsia, and maternal mortality) compared with women without CKD. Similarly, premature births occurred at least twice as often in women with CKD compared with women without CKD. Other fetal outcomes such as IUGR, SGA, neonatal mortality, stillbirths, and low birth weight were all higher in women with CKD compared to women without CKD; however, the rates vary depending on the study [39].

The risk of complications in pregnancy depends on many factors including baseline kidney function, proteinuria, type of renal disease, disease activity, scarring on kidney biopsy, and hypertension. The literature suggests that the strongest predictors of complications are baseline kidney function and severity of hypertension.

Mild CKD has traditionally been classified as a prepregnancy serum creatinine less than 1.5 mg/dL(133umol/l), which would represent an estimated GFR of 50 ml/min/1.73 m² or greater for women over 21 years of age. These patients typically do relatively well during pregnancy. Most of the available studies suggest that renal function is generally preserved among patients with mild CKD. Less than 10 % of women with mild CKD who also have minimal proteinuria (<1 g/24 h) and well-controlled blood pressure will develop permanent, significant renal impairment, and greater than 90 % of patients will have successful pregnancy outcomes defined as having a live birth in the absence of preeclampsia, premature delivery, and IUGR [40, 41]. Hypertension appears to be the main predictor of pregnancy outcome for patients with mild CKD, as uncontrolled hypertension (defined as MAP >105 mmHg) in this population carries a higher risk for pregnancy complications [41].

Moderate CKD is traditionally classified as a prepregnancy serum creatinine range of 1.5–2.5 mg/dL (132–221umol/L). This includes a wide range of renal function that would reflect a preconception range of estimated GFR from 25 to 50 ml/min/1.73 m^2..

In patients with moderate CKD, up to 30 % will experience significant deterioration of renal function that will persist postpartum [42, 43]. In addition, there is a 10 % risk of progression to ESRD by 12 months postpartum; this risk is likely to be greater with higher preconception creatinine values and significant proteinuria [42, 44]. Among patients with moderate CKD, although the fetal survival rate exceeds 90 %, greater than 50 % of fetuses will experience IUGR or prematurity (often related to preeclampsia) [33, 42, 43]. Furthermore, a longitudinal multicenter cohort study of 49 pregnant patients with moderate-to-severe kidney disease showed that the combination of prepregnant proteinuria (>1 g/day) and Cr > 2.0 mg/dL predicted postpartum renal decline and worse fetal outcomes more than reduced GFR or proteinuria alone [33].

The preconception serum creatinine that characterizes severe CKD in pregnancy varies somewhat in the literature, and this is related to the limited number of patients in this category who conceive. Certainly, a serum creatinine greater than or equal to 2.5 mg/dL would be considered severe CKD by most. This value is likely to correlate with an eGFR below 25 mL/min/1.73 m². Since preparations for dialysis or evaluation for renal transplant are appropriate for patients with eGFR below 25 ml/min/1.73 m², the prospect of pregnancy in this setting complicates care considerably. These patients tend to have a difficult time conceiving, and they have the highest risk of maternal and fetal complications. Indeed, combined data from several sources suggest that nearly all women with severely reduced renal function will have a complication during pregnancy and half will experience permanent loss of renal function and may need to start dialysis during pregnancy or postpartum [29]. The risk of prematurity is significant and estimated to be as high as 70–90 % in some studies and IUGR risk of up to 50–65 % [29, 45, 46]. Despite the high risk of complications, approximately 75–90 % of patients will have a live birth. Patients must be counseled about the potential long-term consequences associated with prematurity and the additional stress of potentially caring for a sick child while addressing their own health issues such as initiation of dialysis. Counseling patients about these complex issues is important so that patients can make educated decisions.

Although much of the literature on CKD in pregnancy categorizes the clinical course by the preconception creatinine, the underlying renal disease and the disease activity prior to conception can greatly affect outcome. For example, in patients with systemic lupus erythematosus or systemic sclerosis, adverse maternal and fetal outcomes are more likely than in women who have a history of reflux nephropathy. Patients with a history of nephrolithiasis or recurrent pyelonephritis may have specific challenges but tend to do well with pregnancy. Similarly, women who have previously donated a kidney and therefore have reduced renal mass tend to have favorable outcomes during pregnancy but may have a greater risk of gestational hypertension and preeclampsia compared to the general population [47]. In general, renal disease should be treated and well controlled prior to pregnancy to improve outcomes; this is particularly true for lupus nephritis, diabetic nephropathy, and glomerulonephritis.

In the first case, the urine protein-to-creatinine ratio increased during pregnancy from 0.4 mg/mg prepartum to 0.8 mg/mg at delivery. Many patients with CKD have proteinuria and hypertension at baseline, and during pregnancy the proteinuria often increases. This can make it challenging to differentiate between chronic kidney disease and preeclampsia particularly since hypertension and proteinuria are two of the criteria used to make the diagnosis of preeclampsia. If hypertension and proteinuria are preexisting, other clinical signs should be used to diagnose preeclampsia such as placental dysfunction, elevated liver enzymes, low platelets, or clinical symptoms. It is critical that the distinction between CKD and preeclampsia is made since an incorrect diagnosis of preeclampsia could lead to iatrogenic prematurity. To complicate matters, patients with CKD have up to four times higher risk of developing preeclampsia, compared to those without CKD [38]. An evolving understanding of the pathogenesis of preeclampsia may lead to the identification of serum biomarkers to help distinguish between the two entities. Biomarkers such as soluble fms-like tyrosine kinase 1 (sFlt1), an anti-angiogenic factor, are elevated in patients with preeclampsia and lead to a reduction in placental growth factor (PIGF) and vascular endothelial growth factor (VEGF). The dysregulation of these growth factors appears to result in placental endothelial dysfunction and the subsequent clinical manifestations of preeclampsia [48]. Patients with preeclampsia have marked elevations in sFlt1, reductions in PIGF, and a higher sFlt1/PIGF ratio compared to those who do not have preeclampsia but have CKD [49, 50]. Although prospective and longitudinal studies are still needed to define the role of these markers in clinical management, in the future, some combination of measurements of angiogenic factors may provide unique tools for caregivers to predict preeclampsia and differentiate preeclampsia from a change in renal function among pregnant patients with CKD.

There are no treatments that can reliably prevent the development of preeclampsia. In several large randomized, multicenter trials and meta-analyses, low-dose aspirin appeared to confer some benefit in the general pregnant population though these benefits have not been confirmed in subgroup analyses of high-risk patients or those with CKD [51]. Nevertheless, low-dose aspirin at doses of 75–120 mg taken at bedtime and begun by the 16th week of gestation is broadly recommended by major relevant societies including the International Society of Hypertension in Pregnancy, the United Kingdom’s National Institute of Health Care Excellence (NICE), and the American Congress of Obstetricians and Gynecologists (ACOG) [52, 53].

In addition, the World Health Organization endorses calcium supplementation before 20-week gestation in populations where calcium intake is low with a goal of 1.5–2 g of elemental calcium intake daily. This recommendation stems from randomized controlled trials that demonstrated a decreased risk of preeclampsia and preterm delivery with calcium supplementation, particularly in those women who had low-calcium diets [54, 55].

A 39-year-old nulliparous woman with polycystic kidney disease (PKD) complicated by hypertension presents for prepregnancy evaluation. She is otherwise healthy. She has never had visible hematuria, kidney stones, or urinary tract infections. Her father had PKD and developed ESRD in his late 40s and then died of a heart attack at the age of 61. Her blood pressure has been well controlled on metoprolol sustained release 100 mg daily for several years. Her menstrual cycles are normal, and she is using condoms for contraception. Her blood pressure was 132/87 mmHg. Abdominal examination was notable for palpable kidneys bilaterally. Her creatinine is 2.6 mg/dL (eGFR 22 mL/min/1.73 m²). Old records reveal a creatinine of 1.9 mg/dL 3 years prior and 2.1 mg/dL the year prior to referral.

Despite minimal proteinuria and good blood pressure control, her renal function has declined relatively rapidly. She is counseled about the high risk of pregnancy complications and decides against pregnancy. She has no suitable donors for transplantation and is counseled regarding options for dialysis. She begins oral contraception to prevent an unplanned pregnancy and decides to pursue adoption.

The following year, she begins dialysis for symptoms of nausea and vomiting when her creatinine reaches 5.6 mg/dL, but the symptoms persist. Abdominal ultrasound reveals that she is pregnant, and the approximate age of the fetus is 21-week gestation. Her hemodialysis regimen is intensified and she receives 4 h, 6 days per week (total 24 h per week). Her residual urine output is 1.5 L/day so minimal fluid removal is required with dialysis. Her blood pressures remain normal throughout pregnancy without medication. The fetus has mild fetal growth restriction, but ultrasounds demonstrate normal uterine Doppler flow and amniotic fluid throughout pregnancy. At 35-week gestation, spontaneous labor begins and she delivers a 2619 g baby boy (50–70th percentile). Apgar scores are 8 and 9. The infant is monitored in the neonatal intensive care unit for 1 day given prematurity, but is then discharged and achieves age-appropriate milestones.

A 36-year-old woman is seen for preconception counseling. She has ESRD secondary to focal and segmental glomerulosclerosis and required dialysis for 1 year before she had a living-related kidney transplant from her brother. Her medications include an extended release formulation of tacrolimus 7 mg daily, prednisone 5 mg daily, and mycophenolate mofetil (MMF) 1000 mg twice daily. She has had stable renal function, since her transplant with a creatinine of 1.0 mg/dL (88 umol/L), a hemoglobin of 11.1 g/dL, and no proteinuria. She has had no complications, rejections, or infections. She never had hypertension.

She is counseled about the risks of pregnancy. She and her partner accept the risks and proceeded with preparation. Her mycophenolate mofetil is switched to azathioprine 125 mg daily and she begins prenatal vitamins. Her renal function remains stable and after 3 months, they try to conceive. Six months later, she conceives spontaneously and begins aspirin 81 mg daily; a high-calcium diet or calcium supplementation 1500 mg daily is also recommended.

She is followed jointly with a high-risk obstetrician and nephrologist. She is seen monthly with regular blood work, and during this time, her serum creatinine and blood pressure decrease with a nadir at 24 weeks. Her hemoglobin and iron studies decrease and she takes iron supplements as recommended. At 30 weeks, her blood pressure increases to 148/98 mmHg consistently, and she begins a long-acting formulation of nifedipine at 30 mg daily. Fetal ultrasound reveals mild fetal growth restriction, but normal umbilical artery Doppler studies were normal throughout the pregnancy. At 38 weeks, spontaneous labor begins, and she vaginally delivers a healthy baby girl who weighs 2922 g (70–90th percentile). Immediately postpartum the creatinine is noted to be higher at 1.29 mg/dL (114 umol/L), maternal blood pressure is normal, and she starts breastfeeding without difficulty. At 3 days postpartum, her blood pressure rises to 150/100 mmHg, and the nifedipine is increased. Eight weeks postpartum, her blood pressure returns to normal, and she no longer requires antihypertensive medication. At 1 year postpartum, the blood pressure is normal, and the creatinine remains stable at 1.2 mg/dL (106umol/L), without proteinuria.