Structural changes in the kidney:
Increase in kidney size by 1–1.5 cm (and up to 30% in volume)
Dilatation of the collecting system, more prominent on the right
Systemic hemodynamic changes:
Increase in cardiac output by 40–50% of normal
Increase in plasma volume by 40–50% of normal
Drop in SBP by approx. 9 mm of Hg and DBP by 17 mm of Hg (prominent in second trimester)
Hormonal changes: 10–20-fold increase in aldosterone, 8-fold increase in renin, and 4-fold increase in angiotensin
Resistance to pressor effect of angiotensin
Increased production of prostacyclin and nitric oxide
Renal hemodynamic changes:
Increase in GFR and RPF by 50% above normal
Decrease in glomerular capillary oncotic pressure
Decrease in BUN (<13 mg/dl) and serum creatinine (0.4–0.5 mg/dl)
Increase in proteinuria but generally less than 300 mg/day
Increase in total body water by 6–8 l
Net retention of approx. 900 meq of sodium
Decrease in plasma osmolality by 10 mOsm/L due to reset osmostat
Decrease in serum sodium by 4–5 meq/L
Mild respiratory alkalosis with compensatory metabolic acidosis
(Serum bicarbonate of 18–22 meq/L)
Decrease in serum uric acid levels (2.5–4 mg/dl)
Glucosuria irrespective of blood sugar levels
Prevalence of CKD in Women of Childbearing Age
In women of childbearing age, the estimated prevalence of CKD is reported to be around 4% (where CKD is defined as serum creatinine >1.2 mg/dl or estimated glomerular filtration rate, eGFR <90 ml/min)  with 1 in 150 being affected by stage 3–5 CKD (defined as eGFR <60 ml/min) . The prevalence of kidney disease in pregnancy or the incidence of pregnancy in women with non-dialysis-dependent CKD is not known.
Pregnancy Outcomes in CKD
Stages of chronic kidney disease based on estimated GFR
GFR in ml/m/m2
Kidney damage with normal or raised GFR
Kidney damage with mildly lower GFR
Moderately lower GFR
Severely low GFR
<15 or dialysis
Pregnancy Outcomes Based on Severity of Renal Disease
Mild Kidney Disease
Mild kidney disease in pregnancy literature is generally defined as serum creatinine <1.4 mg/dl with varying degree of proteinuria. Not long ago, renal disease of any degree was considered to be incompatible with safe and successful pregnancy, and termination of pregnancy in the first trimester was routinely recommended. We now have reasonable evidence to suggest good pregnancy outcomes in mild CKD in the absence of hypertension, proteinuria, or underlying systemic disease. Though adverse pregnancy-related outcomes are higher when compared to general population, pregnancy is generally considered safe.
Fertility and conception rates are generally considered to be normal in women with mild kidney disease although no data are available to confirm this long-standing theory.
Most frequent maternal complications noted are new-onset hypertension, worsening proteinuria, and preeclampsia. Renal function is generally unaltered during pregnancy. A recent study involving a large cohort of pregnant Italian women reported progressive worsening of maternal outcomes from stage I to stage V CKD . Tubulo-interstitial disorders accounted for the majority (38%) of the underlying CKD followed by glomerular pathology (16%). A 7.9% increased risk of new-onset hypertension, 20.5% increased risk of new-onset or worsening proteinuria, and 7.6% increased risk of worsening renal function were reported in stage I CKD and 12.6% increased risk of worsening renal function in stage II CKD. Further clinical characteristics of women with stage I and stage II CKD who had pregnancy-related decline in renal function (e.g., presence or absence of hypertension and proteinuria) are not known.
Fetal outcomes are generally good in mild kidney disease with fetal mortality being relatively uncommon. There is increased risk of low birth weight, preterm delivery, and caesarian deliveries when compared to general population . Hence, while women with mild CKD can be counseled about good pregnancy outcomes, close monitoring is needed for early detection and management of pregnancy- related complications.
Moderate-Severe Kidney Disease
Moderate kidney disease is arbitrarily defined as a serum creatinine of 1.4–2.5 mg/dl in pregnancy and severe kidney disease when the serum creatinine is >2.5 mg/dl. The most compelling factor that differentiates pregnant women with moderate to severe CKD from any other pregnant population is the risk of rapid decline in renal function. The mechanism by which pregnancy accelerates progressive decline of renal function is unknown. In nonpregnant patients, prolonged periods of renal vasodilatation in some types of chronic progressive kidney disease can lead to an increase in intra-glomerular pressure (PGC) contributing to a further decline in renal function . There is a parallel reduction in tone of both afferent and efferent arterioles with no increase in PGC in normal pregnancy. Whether these parallel changes in both the afferent and efferent systems are disrupted in pregnant women with CKD is not known. Whatever the underlying mechanisms, pregnancy exerts adverse effects only after a critical amount of renal function has been lost regardless of the cause of the CKD. Once a decline in renal function has occurred, it cannot be predictably reversed even upon termination of pregnancy.
Fertility rate in women with non-dialysis-dependent CKD is not known. While there is a general agreement that fertility rates decline with increasing severity of renal insufficiency, the stage of CKD at which the predilection starts is not known. Hyperprolactinemia occurs infrequently in CKD and only when renal failure is fairly advanced . There may be several possible reasons for lack of such data. Pregnancy is less common in women with moderate to severe kidney disease, and many women with moderate to severe kidney disease are generally beyond childbearing age . Restoration of fertility in women with moderate to severe non-dialysis-dependent CKD is often discouraged due to probable risk of progression of kidney disease especially in women who have underlying proteinuria and hypertension. While progression is not an absolute rule (20% risk reported in a recent meta-analysis) , preconception counseling with a multidisciplinary team o f high-risk obstetricians, neonatologist, and nephrologists is necessary. Conception may occur and birth control should also be discussed in these patients.
A prospective study comparing the rate of loss of renal function before and after conception in women with stage III to V CKD reported no significant difference, but in subgroup analyses, women with GFR <40 ml/min/1.73 m2 and 24-h proteinuria >1 gm had an accelerated rate of deterioration of GFR . In another report of 82 pregnancies in 67 women with a mean serum creatinine of 1.9 mg/dl, worsening renal function was more common in women with creatinine >2.0 gm/dl . In the Italian cohort of pregnant women that compared outcomes of 504 pregnancies in women with CKD stages I to V to 836 low-risk pregnancies in women without CKD, there was an increasing trend of shift in CKD stage (7.6% in stage I, 12.6% in stage II, 16.2% in stage III, and 20% in stage IV–V CKD, P = 0.12) . Only 47 of the 504 pregnancies were in women with stage III to V CKD, probably contributing to the statistically insignificant result. The risk of preeclampsia, failure of pregnancy (including stillbirth, fetal death, and neonatal death), and caesarian delivery is also higher in this patient population. In a recent meta-analysis that included 23 studies, pregnancy with CKD had greater odds of preeclampsia (odds ratio, OR 10.36), caesarian section (OR 2.67), and failure of pregnancy (OR 1.80) compared to those without CKD. Risk of preterm delivery (<37 weeks of gestation) is also higher with progressive increase in risk with severity of renal function (23.5% with stage I, 50.6% in stage II, 78.4% in stage III, and 88.9% in stage IV–V CKD) . Increased risk of placental abnormalities and polyhydramnios has also been reported.
The most common fetal complications noted in moderate-severe CKD are prematurity, intrauterine growth retardation (IUGR), and intrauterine deaths. Not surprisingly, risk of fetal complications increases with severity of the kidney dysfunction with highest risk seen in stage IV–V CKD. The overall possibility of a pregnancy resulting in a live infant ranged from 50% to 100% in various reports. Of the 90 conceptions included in 78 women with CKD, 14 stillbirth/neonatal deaths and 5 spontaneous abortions were reported . Fetal outcomes should be addressed while providing comprehensive counseling to women with moderate to severe CKD.
End-Stage Renal Disease (ESRD)
Successful outcomes in pregnant women on dialysis are increasingly being reported since the first report of a successful pregnancy in a hemodialysis (HD) patient in 1965 . Fetal survival has improved by almost 25% per decade, reaching over 85% in patients receiving nocturnal hemodialysis. The improved pregnancy outcomes in ESRD patients are due to improved dialysis techniques, better understanding of clearance and nutritional needs in pregnancy, and better management of anemia. Regardless, pregnancy in women on dialysis is challenging for both the patient and physicians involved. Most of available literature is based on single center experience, and as pregnancy does not lend itself to randomization, there are no clinical trials done nor evidence-based guidelines available for management of this patient population.
Fertility in ESRD
Conception is rare in ESRD. While the exact mechanisms are unclear, impaired hypothalamic-pituitary axis from uremia, underlying comorbid conditions (anemia, depression), and concomitant medication usage are all thought to be contributing factors . Amenorrhea, often associated with anovulatory cycles, is common in ESRD women. In the early days of dialysis, less than 10% of premenopausal women on dialysis reported regular menstruation. In the late 1990s, 42% of premenopausal women on dialysis reported regular menses . Sexual dysfunction is common and often ignored in ESRD population. In a multinational cross-sectional study on women undergoing hemodialysis, 84% of the study population reported sexual dysfunction . Sexual dysfunction was independently associated with age, depressive symptoms, less education, menopause, diabetes, and diuretic therapy. In another study, 80% of the women reported sexual dysfunction, but only one-fifth of them discussed this with their healthcare provider .
Reported fertility rates in ESRD women of childbearing age have been variable. Estimates of the frequency of conception in ESRD vary from 0.3% to 1.5% per year from across the world [19–21]. The frequency of conception in women on nocturnal dialysis is much higher than any other group, 15.6% . It has been postulated that enhanced uremia control may have restored the normal reproductive endocrine functions leading to increased fertility rates in this study cohort. Successful conception can in fact serve as a surrogate marker for adequacy of dialysis. Fertility rates are lower in women on peritoneal dialysis (PD) compared to HD. In a report that included a total of 6230 women of reproductive age receiving chronic dialysis (1699 on PD and 4531 on HD), only 4 conceptions per year were reported in women receiving PD compared to 27 per year for HD patients. The conception rate was 2.2% for the whole population, 2.4% for HD patients, and 1.1% for PD patients (P < 0.01) . Similar results were reported from the Australian and New Zealand Dialysis and Transplant (ANZDATA) Registry . Reasons for such a difference are not clear. There are no studies comparing reproductive cycles in women treated with these two modalities. One proposed hypothesis is that the presence of a large volume of hypertonic fluid in the peritoneal cavity interferes with the transport of the ovum to the fallopian tube .
Reported maternal mortality in pregnant ESRD women is rare. Three maternal deaths were reported to the US Registry for Pregnancy in Dialysis Patients (RPDP) with one death from lupus cerebritis in a woman who started dialysis after conception . Hypertension is the most common maternal complication, occurring in 80% of the pregnancies in ESRD . The risk of preeclampsia in ESRD women varied between 19% to as high as 66% [27, 28] with successful delivery rate being lower in women with preeclampsia (60% vs. 92.9%).
The common fetal complications are IUGR, premature birth, preterm delivery, and intrauterine death. Infant mortality has significantly declined over the last few decades. According to registry data from 1998, only approximately 50% of pregnancies in women who conceived after starting dialysis resulted in surviving infants . Those who reached the second trimester had a high likelihood of neonatal death. By 2002, there was enough data to suggest that 75% of infants would survive if dialysis was increased to 20 or more hours per week . With more intense dialysis, stillbirth became rare, but premature labor remained a major issue. By 2008, a high at term delivery rate was reported with nocturnal dialysis with dialysis times averaging 36 hours per week (5 of 6 pregnancies delivered after 36 weeks) . Even better results were achieved in women on nocturnal dialysis dialyzing an average of 48 h per week with 18 of 21 pregnancies (85.7%) resulted in surviving infants. Sixty percent were born at greater than 36 weeks . Available evidence points to better outcomes with longer dialysis sessions, and nocturnal HD may be suggested to pregnant women who are willing to try.
The optimal timing of pregnancy depends on the individual circumstances of the transplant recipient. Pregnancy should be planned and teratogenic medications discontinued before conception. Mycophenolate and sirolimus should be discontinued 6 and 12 weeks prior to conception, respectively. The consensus opinion of the American Transplant Society is that pregnancy can be planned after 1–2 years of transplantation as long as the patient had no rejection episodes in the past year, adequate renal function (serum creatinine of <1.5 gm/dl and minimal proteinuria) and no recent infections that could jeopardize fetal survival or health . Optimal contraception is important to initiate during the peritransplant period in women of reproductive age as pregnancy has been reported during this period. The choice of contraception is individualized according to medications the woman is on as well as concomitant comorbidities.
Sexual dysfunction and infertility which are common in the ESRD population generally reverse posttransplantation resulting in normal ovulatory function and regular menstrual cycles. In one study, LH, FSH, and estradiol levels normalized within 3–4 months of successful transplant, and 17 of the 21 women were able to conceive within 3 years .
The major concern is the effect of pregnancy on maternal allograft function and development of any graft-independent comorbidities. Acute rejection during pregnancy remains a concern as well as allograft failure. A recent study reviewed the information provided by USRDS (United States Renal Data System) and Medicare claims in 21,814 women who received a first kidney-only transplant between 1990 and 2010 to determine the risk of allograft failure with pregnancy in the first 3 years posttransplant. In a multivariate analysis, they demonstrated a higher incidence of renal allograft failure from any cause including death after pregnancy in the first and second years posttransplant. No such association was noted with pregnancy in the third posttransplant year. Data was obtained from pregnancy-related insurance claims with no information available on immunosuppressive medications or dosing or degree of prenatal care these women received . In a review of 2412 pregnant transplant recipients where acute rejection was studied, 102 (4.2%) experienced an episode of acute rejection during pregnancy . Whether pregnancy played any direct role is not known. Detection of rejection can be difficult because of pregnancy-related changes in renal physiology, and when in doubt, biopsy can be performed safely for definitive diagnosis. Observational studies have shown that a serum creatinine of >1.5 mg/dl and proteinuria >500 mg/24 h significantly increase the risk for irreversible graft loss post-pregnancy. The overall incidence of preeclampsia is reported to be high, 27% in kidney transplant recipients, a 3.8% higher rate when compared to general US population .
Fetal outcomes are generally good in kidney transplant recipients. In a meta-analysis of pregnancy outcomes in kidney transplant recipients, live birth rate was higher (73.5% vs. 66.7%), and miscarriage rates were lower than that of the US general population (14.0% vs. 17.1%) . Both premature delivery and caesarian deliveries were higher in transplant recipients compared to the general population. Similar trends were noted across the globe.
A major concern to the fetus is the potential side effects of immunosuppressive medications on organogenesis and fetal growth. All immunosuppressive medications cross the maternal-fetal circulation, and there is no choice other than to expose the fetus to the least offensive medications. There is much to be learned about the pharmacokinetics and pharmacodynamics of immunosuppressive medications during pregnancy. Maintenance immunosuppression in pregnant transplant patients generally includes cyclosporine or tacrolimus, azathioprine, and prednisone. There is mounting evidence for teratogenicity associated with the use of mycophenolate. Malformations described include microtia, cleft lip and palate, auditory canal atresia, hypertelorism, micrognathia, ocular coloboma, short fingers, and hypoplastic nails . Physicians prescribing the drug are expected to educate women in the reproductive age group about contraception and planning of pregnancy. Antibodies used to treat rejection such as thymoglobulin and alemtezumab cross the placenta, but there is not enough experience with them to know the effects on the fetus. Their use would seem preferable to losing the kidney, but they should be used only if high-dose steroids are ineffective and there is a reasonable chance that the kidney can be salvaged.
Aspects Unique to Underlying Etiology of CKD
Lupus Nephritis (LN)
The rule of renal failure not progressing when serum creatinine is <1.4 mg/dl does not apply to women with LN. Despite significant improvements in fetal outcomes (fetal loss decreased from 40% in 1960–1965 to 17% 2000–2003), maternal mortality is still unacceptably high, a 20% higher maternal mortality in women with SLE . Pregnancy-induced immunological and hormonal changes are associated with flares of systemic lupus erythematosus (SLE) and LN in particular. Increased risk of a LN flare during pregnancy was reported in multiple studies that used a patient’s own nonpregnant course for comparison. Women with class III and IV LN are at increased risk of hypertension during pregnancy and renal flares. Lupus nephritis patients with antiphospholipid syndrome (APS) face a further increase in risk of thromboembolic events during pregnancy. A 54% reduction in pregnancy loss is reported with combination therapy of low-dose aspirin and heparin compared to aspirin alone in women with APS . Aspirin is generally prescribed when conception is attempted and prophylactic dose of heparin (low molecular weight heparin is a safe alternative) at confirmation of intrauterine pregnancy.
New-onset lupus with rapid deterioration of renal function may be an indication for kidney biopsy during pregnancy since diffuse proliferative lupus nephritis requires prompt treatment and first-line treatments are teratogenic. Cyclophosphamide and mycophenolate are teratogenic in the first trimester. Cyclosporine, azathioprine, and prednisone are considered safe in pregnancy, with cyclosporine probably being the most effective in lupus nephritis. Fertility in women with lupus nephritis is similar to that of the general population if renal function is normal unless they were treated with large doses of cyclophosphamide. In a study done reviewing the effects of cyclophosphamide on ovarian function in women with breast cancer, the cumulative dose exposure before the onset of amenorrhea was 5.2 g for women in their 40s and 9.3 g for women in their 30s . Both age at the time of exposure and cumulative dose of cyclophosphamide are the major determinants of risk of ovarian failure. In women with preexisting lupus nephritis, pregnancy is safest if the disease has been in remission on less than 10 mg daily of prednisone for at least 6 months, serum creatinine is less than 1.5 mg/dL, and blood pressure is well controlled.
Glomerulonephritis (GN) Other than Lupus
Pregnancy does not adversely affect the course of maternal renal disease in women with underlying primary GN as long as renal function is near normal. In women with IgA nephropathy, higher prenatal mortality rate is seen in women with GFR of <70 ml/min or when pre-pregnancy BP is consistently >140/90 mm of Hg . In the largest published cohort of 360 women with various histological forms of GN, overall end-stage renal failure (ESRF)-free survival did not differ significantly in women who became pregnant (n = 171) after clinical onset of kidney disease and those that did not conceive (n = 189) . Hypertension and the histological form of GN were predictive factors for subsequent development of ESRD, unrelated to pregnancy-induced decline in renal function.
The prevalence of diabetic kidney disease (DKD) in pregnant women with type I diabetes is reported to be around 0.5–5% and 0.8–2.6% for type II diabetes. Severe congenital anomalies have been reported in children born to women with diabetic nephropathy. Seven of 35 children that were born to women with diabetic nephropathy were noted to have psychomotor retardation at 4.5 years of follow-up . Patients with poor glycemic control or diabetic nephropathy before pregnancy are at increased risk of developing preeclampsia. A lower threshold for starting BP medications (BP goal of <135/85 mm Hg) has been advocated by few studies that report improved pregnancy outcomes with such a strategy; however this should be done with caution with an intention to avoid placental hypoperfusion through hypotension .
Autosomal Dominant Polycystic Kidney Disease (ADPKD)
ADPKD does not affect fertility in women with normal renal function. Reports of ectopic pregnancies have not been validated in large studies. Women with ADPKD are at higher risk of developing hypertensive disease of pregnancy including preeclampsia. Normotensive women with ADPKD who developed preeclampsia or hypertension during pregnancy are more likely to develop chronic hypertension subsequently . Repeated pregnancies may cause symptomatic enlargement of liver cysts.