Pregnancy in Chronic Kidney Disease: Need for Higher Awareness. A Pragmatic Review Focused on What Could Be Improved in the Different CKD Stages and Phases

Pregnancy is possible in all phases of chronic kidney disease (CKD), but its management may be difficult and the outcomes are not the same as in the overall population. The prevalence of CKD in pregnancy is estimated at about 3%, as high as that of pre-eclampsia (PE), a better-acknowledged risk for adverse pregnancy outcomes. When CKD is known, pregnancy should be considered as high risk and followed accordingly; furthermore, since CKD is often asymptomatic, pregnant women should be screened for the presence of CKD, allowing better management of pregnancy, and timely treatment after pregnancy. The differential diagnosis between CKD and PE is sometimes difficult, but making it may be important for pregnancy management. Pregnancy is possible, even if at high risk for complications, including preterm delivery and intrauterine growth restriction, superimposed PE, and pregnancy-induced hypertension. Results in all phases are strictly dependent upon the socio-sanitary system and the availability of renal and obstetric care and, especially for preterm children, of intensive care units. Women on dialysis should be aware of the possibility of conceiving and having a successful pregnancy, and intensive dialysis (up to daily, long-hours dialysis) is the clinical choice allowing the best results. Such a choice may, however, need adaptation where access to dialysis is limited or distances are prohibitive. After kidney transplantation, pregnancies should be followed up with great attention, to minimize the risks for mother, child, and for the graft. A research agenda supporting international comparisons is highly needed to ameliorate or provide knowledge on specific kidney diseases and to develop context-adapted treatment strategies to improve pregnancy outcomes in CKD women.


Introduction
Chronic kidney disease (CKD) is a well-acknowledged risk factor for adverse pregnancy outcomes [1][2][3][4][5][6][7]. The literature on this issue is rapidly accumulating and the term "obstetric nephrology" has been proposed to identify this important clinical and research field [8].
However, awareness of the importance of identifying CKD in pregnancy is still insufficient and the experience is mainly limited to some large, but still few, referral centers [3,[9][10][11][12].
This narrative review, in association with a review on pregnancy and acute kidney injury (p-AKI), has been based on the theme of the World Kidney Day 2018, which highlighted women's health and kidney disease [13]. Importantly, the focus of this narrative is on what could be done to improve CKD care throughout and after pregnancy.
The review follows the classic, even if not necessarily sequential, phases of renal diseases: chronic kidney disease, dialysis, and transplantation, and focuses on the knowledge gaps, on the delay of application of what is known into the clinical practice, and on the potential interventions that could improve the care of mother and child during and after pregnancy.

State of the Art: What We Know on the CKD-Pregnancy Relationship: CKD Stages
Kidney function is of crucial importance in healthy pregnancy [8]. Several changes in kidney function occur in the pregnant woman, affecting the vascular, glomerular, and tubular components, ultimately resulting in increased renal clearances and "physiological" proteinuria, decrease in blood pressure, and expansion of the intravascular volume [8,[14][15][16][17].
The kidney is the target and the central player in the hypertensive disorders of pregnancy, an umbrella term that gathers the most common pregnancy-induced disorders: isolated hypertension (usually identified by the acronym PIH: pregnancy induced hypertension), pre-eclampsia (PE), in which hypertension is usually associated with proteinuria and may be associated with acute and transient reduction of the kidney function (now considered a hallmark of the PE syndrome, even in the absence of proteinuria), HELLP syndrome, an acronym for haemolysis, elevated liver enzymes, low platelets, a severe, occasionally life-threatening, endothelial disorder [8,[14][15][16][17][18][19][20]. Isolated proteinuria may also transiently appear in pregnancy and is usually indicated as "pregnancy-induced proteinuria". Proteinuria may precede PE, but even when isolated, it heralds a risk of adverse pregnancy outcomes, including growth restriction and preterm delivery; the differential diagnosis between pregnancy-induced and pre-existent proteinuria may not be easy [20][21][22][23].
Due to the central role of the kidney, target, and actor in the pathogenesis of the hypertensive disorders of pregnancy, it is not surprising that a reduction of the kidney function may affect pregnancy outcomes [1][2][3][4][5][6][7][8][9]24]. What may be surprising is that kidney diseases are associated with a significant increase in the risk of adverse pregnancy outcomes even in the absence of kidney function reduction [4,5,[25][26][27][28][29][30].
In this regard, interesting insights come from the analysis of pregnancy after kidney donation, which shows that this condition of "healthy" reduction of the kidney parenchyma is associated with a higher risk of pre-eclampsia and hypertensive disorders of pregnancy [25,26,[31][32][33].
In each CKD stage, hypertension and proteinuria are important modulators of the entity of pregnancy-related risks; however, the specific role of each element (kidney disease, stage, hypertension, and proteinuria) is not fully known, thus limiting the information available for counseling [41][42][43][44] ( Table 1). Furthermore, perinatal outcomes depend also upon the setting of care, an issue that has to be taken into account both in interpreting results and in planning treatment strategies. As it will be further discussed, our knowledge on the specific risks linked with the different kidney diseases is limited; overall, we know more about glomerular diseases, and the most common ones, such as IgA nephropathy, are extensively studied; conversely, the specific risks associated with interstitial nephropathies or polycystic kidney disease (ADPKD) are not fully appreciated [7,9,[45][46][47][48][49][50][51]. Table 2 resumes the main pregnancy-related risks in CKD patients: overall, malformations are not increased with respect to the overall population (out of the context of inherited diseases and of diabetic nephropathy); maternal death is exceptional, at least in highly resourced countries.
Conversely, the incidence of preterm delivery and of small babies ( Figure 1) is increased already in stage 1 CKD, with respect to the overall population, and rises along with the increase of CKD stages [1][2][3][4][5][6][7]24,27,28]. Likewise, the effect of pregnancy on CKD progression is debated, also on account of the different study designs, obstetric policies, and duration of follow-up of CKD women after pregnancy. Overall, short-and long-term decrease in the kidney function is exceptional in the first CKD stages, but rapid decrease of the kidney function may be an issue in late CKD stages [1][2][3][4][5][6][7].
Once more, studies are heterogeneous and evidence is limited; as a consequence, entity of the risk of CKD progression has been variously estimated ( Table 2).
The information on the outcomes of pregnancy in the last CKD phases is still scant. Fertility is usually reported as reduced in the last CKD stages, but it is possible that the attitude of discouraging pregnancy in advanced CKD has induced a selection bias [52,53]. Table 2. Adverse pregnancy outcomes in CKD patients and in their offspring [13].

A Particular Case: Systemic Immunologic Diseases
Kidney disease in SLE is a critical concern for pregnancy. Kidney involvement includes lupus nephritis (LN), characterized by glomerular damage and interstitial and vascular lesions. Beyond risk factors associated with CKD (proteinuria, hypertension, and impaired kidney function, Tables 1 and 2), SLE is associated with a specific increase in miscarriages and perinatal death. All risks are higher in active SLE, and adverse outcomes are associated in particular with LN class 3 and 4, history of renal flares, longer disease, hypocomplementemia, presence of antiphospholipid antibodies (aPL), and antiphospholipid syndrome (APS) [5][6][7]54,55].
Conversely, pregnancy carries a risk of SLE flares: high estrogen level may act as triggers, by mediating transcription activity of the intracellular estrogen receptors and interacting with regulatory T cells, key modulators of maternal-foetal tolerance [56]. Upregulation of IFN-α may also play a role in SLE and LN: this cytokine, highly expressed by the placenta, contributes to placentation and increases susceptibility to SLE [57][58][59]. Low C3 and high anti-DNA antibodies predict renal flares, whereas high anti-C1q antibodies and low C4 predict early flares.
Pregnancy complications, along with vascular thrombosis, are the main clinical criteria for antiphospholipid syndrome, which may be isolated (primary forms) or associated to SLE or other autoimmune disorders. The disease directly affects placental vasculature, ultimately resulting in placental dysfunction: antiphospholipid antibodies affect the cytotrophoblast via thrombosis, inflammation, apoptosis, and immunomodulatory impairment; direct damage of endometrial cells has also been described [60,61]. APS-related complications in pregnancy include the whole list of pregnancy complications: recurrent miscarriage, preterm delivery, IUGR, stillbirth, fetal distress, fetal or neonatal thrombosis, PE, eclampsia, HELLP syndrome, arterial or venous thrombosis, and placental insufficiency. High titers and triple positivity for aPL (usually defined as positivity for LLAC and for anti-cardiolipin (aCL) and anti-β2GPI antibodies of the same isotype by the same method) are associated with mother and foetal complications, including miscarriage [60][61][62][63][64][65][66][67][68][69]. A particularly severe syndrome, named catastrophic APS (CAPS), combines all these damages into multiorgan failure [60][61][62][63][64][65][66][67][68]. The occurrence of HELLP syndrome in a patient with APS should raise the suspicion of CAPS, and defines a permanent risk for further pregnancies [69].
The link between pregnancy-induced thrombocytopenia or endothelial damage and other causes, including thrombotic thrombocytopenic purpura (TTP) and hemolytic uremic syndrome (HUS), now often referred to as thrombotic microangiopathies (TMA), is not fully clear; the diseases may overlap and the differential diagnosis may become evident only after pregnancy (Table 3) [70]. The different "small babies": Growth curve of small, small for gestational age (SGA), intrauterine growth restricted (IUGR) babies (Y axis: weight and reference curves; X axis: gestational weeks). Legend: (A) is a very small, early preterm normal for gestational age child; (B) is a very small, early preterm, SGA but harmoniously grown, preterm child (mother and father also of small body size); child (C) is a small, SGA, preterm child with a flattening of the growth curve. Although (B,C) are identified also as IUGR, flattening of the growth curve may have a different (unfavorable) meaning as for life-long complications. (Courtesy of R. Attini and P. Gaiotti).

A Particular Case: Systemic Immunologic Diseases
Kidney disease in SLE is a critical concern for pregnancy. Kidney involvement includes lupus nephritis (LN), characterized by glomerular damage and interstitial and vascular lesions. Beyond risk factors associated with CKD (proteinuria, hypertension, and impaired kidney function, Tables 1 and 2), SLE is associated with a specific increase in miscarriages and perinatal death. All risks are higher in active SLE, and adverse outcomes are associated in particular with LN class 3 and 4, history of renal flares, longer disease, hypocomplementemia, presence of antiphospholipid antibodies (aPL), and antiphospholipid syndrome (APS) [5][6][7]54,55].
Conversely, pregnancy carries a risk of SLE flares: high estrogen level may act as triggers, by mediating transcription activity of the intracellular estrogen receptors and interacting with regulatory T cells, key modulators of maternal-foetal tolerance [56]. Upregulation of IFN-α may also play a role in SLE and LN: this cytokine, highly expressed by the placenta, contributes to placentation and increases susceptibility to SLE [57][58][59]. Low C3 and high anti-DNA antibodies predict renal flares, whereas high anti-C1q antibodies and low C4 predict early flares.
Pregnancy complications, along with vascular thrombosis, are the main clinical criteria for antiphospholipid syndrome, which may be isolated (primary forms) or associated to SLE or other autoimmune disorders. The disease directly affects placental vasculature, ultimately resulting in placental dysfunction: antiphospholipid antibodies affect the cytotrophoblast via thrombosis, inflammation, apoptosis, and immunomodulatory impairment; direct damage of endometrial cells has also been described [60,61]. APS-related complications in pregnancy include the whole list of pregnancy complications: recurrent miscarriage, preterm delivery, IUGR, stillbirth, fetal distress, fetal or neonatal thrombosis, PE, eclampsia, HELLP syndrome, arterial or venous thrombosis, and placental insufficiency. High titers and triple positivity for aPL (usually defined as positivity for LLAC and for anti-cardiolipin (aCL) and anti-β2GPI antibodies of the same isotype by the same method) are associated with mother and foetal complications, including miscarriage [60][61][62][63][64][65][66][67][68][69]. A particularly severe syndrome, named catastrophic APS (CAPS), combines all these damages into multiorgan failure [60][61][62][63][64][65][66][67][68]. The occurrence of HELLP syndrome in a patient with APS should raise the suspicion of CAPS, and defines a permanent risk for further pregnancies [69].

What is Missing, and What We Still Need to Know
Indirect estimates of the prevalence of CKD in pregnancy come from the assumption that fertility is not affected in early CKD and that the prevalence of early kidney disease is the same as in the overall population [1,85]. Based upon this assumption, it has been suggested that early-stage CKD is present in 3:100 pregnancies and later-stage CKD in 1:750 pregnancies [1,85]. Interestingly, the reported prevalence of CKD is strikingly similar to that of PE, much better acknowledged as an important risk in pregnancy [1,85,86]. Furthermore, the estimation of the prevalence of CKD in pregnancy, in the classic paper by Williams and Davison, is based upon an evaluation performed in western, highly resourced countries, and may be underestimated in medium-and low-income countries, in which the prevalence of CKD is usually higher [1,[85][86][87][88][89][90][91] Assessing the actual prevalence of CKD in pregnancy should therefore be a health care priority, also since pregnancy may represent an underutilised, but highly valuable, occasion for the diagnosis of CKD, in particular in disadvantaged populations in which the opportunities for early CKD diagnosis are few [90][91][92][93][94]. In fact, pregnancy is often the first occasion in which an apparently healthy woman undergoes a clinical and laboratory evaluation, which has the potential to timely reveal the presence of underlying CKD. However, the only assessment systematically done and regarding the kidney is the urinalysis, which is usually interpreted only with regard to the presence of proteinuria, or of urinary tract infection [95][96][97].
Simple and inexpensive measures, such as including the assessment of serum creatinine in the routine pregnancy tests, could improve early diagnosis of kidney diseases, in particular if the physiological decrease of serum creatinine levels, due to physiological changes in pregnancy, is acknowledged. In this regard, there is also a need to determine reference values for creatinine during pregnancy, by race and pregnancy trimester. Addition of serum creatinine will not solve all problems, but may allow identifying at least cases with kidney function impairment. Indeed, many kidney diseases are fully asymptomatic; not all kidney diseases manifest with hypertension and proteinuria, the symptoms on which attention is more focused; haematuria or electrolyte derangements may be overlooked in pregnancy and there is a consistent overlap of signs and symptoms between PE and CKD in pregnancy [98][99][100][101]. These diagnostic challenges are summarised in Table 4.
Furthermore, signs and symptoms of CKD and PE may overlap, and there is a need for establishing common lines for systematically considering the differential diagnosis between CKD and PE in pregnancy; while a combination of Doppler flows and the biomarkers employed in the assessment of PE may support diagnosis, large prospective studies are needed to refine the procedures and validate their use [101][102][103].

The State of the Art
Fertility is reduced in end-stage kidney disease; Australian and European data suggest that pregnancy occurs ten times less frequently in patients with kidney transplantation than in the general population and that pregnancy occurs ten times less frequently in dialysis patients compared with kidney transplant patients. In other words, a woman on dialysis has a probability of having a baby that is 1% of the overall population [104][105][106].
While the first sporadic cases of successful pregnancy on dialysis were described in the seventies, it was only in the new millennium that pregnancy on dialysis became an acknowledged clinical possibility [104][105][106][107][108][109][110]. So far, more than 1000 pregnancies have been reported in dialysis patients, with an increasing trend worldwide [109].
The most important advance in this field has been the demonstration of a strict relationship between the intensity (frequency and duration) of the dialysis sessions and pregnancy results, thus leading to intensify dialysis up to daily, favoring also long-hours treatment as compared to standard schedules [107][108][109][110].
The improvement in results recorded with daily, extended-hours dialysis not only allowed a more permissive attitude towards pregnancy in dialysis, but is also leading to a more positive attitude towards pregnancy in advanced CKD, often previously discouraged for the fear of needing to start dialysis during gestation [107][108][109][110][111][112][113][114].
While the "best" (if any) dialysis prescription is still not agreed upon, some common lines are emerging, specifically from the Canadian experience: in patients without residual renal function, at least 36 h of dialysis per week should be prescribed, and daily frequency should be chosen; weight loss during the session should be very slow, hypotension should be avoided, and hypertension not hypercorrected; heparin anticoagulation is safe (more experience with unfractioned heparin); biocompatible membranes should be used; multivitamin supplementation and high-protein diets should be prescribed to compensate for intradialytic nutrient loss; phosphate supplements may also be needed. Low blood and dialysate flows should allow a "soft" dialysis; Kt/V is not the ideal marker of dialysis efficiency and the target is set at "near normal" predialysis urea (after the day break) (Table 4) [10,109,111]. Table 4. Challenges for the diagnosis of kidney diseases in pregnancy.

Sign or Symptom Interference in Pregnancy Potential Correction
Hypertension Blood pressure is physiologically reduced, particularly in the 1st trimester; mild-moderate hypertension may be masked by the physiologic changes in early pregnancy.
Attention to "high normal" BP; strict monitoring in presence of risk factors (obesity, diabetes, family history, age, CKD); if associated with proteinuria, interference in PE diagnosis.

Kidney function reduction
Kidney function physiologically increased, in particular in 1st-2nd trimester; CKD stages 2-3 may be missed, in particular if reference eGFR of nonpregnant women is employed.
Lack of hyperfiltration is a potential sign of initial reduction of kidney function; "normal" values outside pregnancy should be interpreted with caution in pregnancy.

Hematuria
Presence of contaminant RBCs in the urine is common and microhematuria may be misinterpreted as of gynaecologic origin. Possible underestimation in polyuria.
Microscopic examination, search for urinary casts, and attention to correct sampling avoid missing this sign pointing to a glomerular origin (most frequent, IgA nephropathy).

Proteinuria
The physiologic limit is increased to 300 mg/day. Proteinuria may show day-to-day and circadian variations. If assessed on spot urine collection, mild proteinuria may be missed in patients with polyuria.
Repeated tests and 24 h urine assessment in cases with trace proteinuria may avoid missing low-grade (albeit clinically relevant) proteinuria.

Tubular derangements
The usual workout of physiologic pregnancy includes the major ions (Na, K, Calcium). Mild hypokalaemia and hyponatremia are common, due to physiologic hemodilution. In settings where dialysis is available without restrictions, pregnancy in dialysis is becoming more common, though still a rare occurrence, underlining the importance of a network of care for sharing opinions, gathering data, and optimizing results [110,114].
Conversely, in poorly resourced countries, pregnancy is a common precipitating event of severe CKD, often not previously diagnosed; furthermore, in these settings, p-AKI is more common and may not be reversible in 5-30% of the cases, further underlining the complex link between PE, p-AKI, and CKD in pregnancy [115][116][117][118].

What Is Missing and Could Be Done in the Clinical Practice
The evidence on pregnancy and dialysis is heterogeneous and several questions are still unanswered. Two of them are of pivotal importance: when to start dialysis in pregnancy and how we can mediate between the excellent results obtained by Hladunewich in Canada with long-hours daily dialysis and the limited access to dialysis in many developing countries.
The issue of dialysis start in pregnancy is complex; the Canadian data, which indicate that a target of predialysis "near normal" urea level is associated with better outcomes, are often taken as an indication to start dialysis early. While this attitude may be reasonable, it has not been proven, and is not in keeping with the recent indications to start dialysis within an "intent to delay" policy in all other categories of patients [111,[119][120][121][122].
As a consequence of this uncertainty, the residual kidney function at dialysis start in pregnancy ranges from 20 mL/min, a level that would not have supported the indication to start RRT outside of pregnancy, to the usual levels of less than 10 mL/min [123,124].
There are very few position statements from scientific societies on these issues; the most recent Italian one leaves the question open, advocating an individualised approach [6,111]. Furthermore, the role of nutritional support to delay dialysis start in pregnancy has been insufficiently studied, and the promising results obtained with plant-based diets need large-scale validation [125][126][127]. There is an urgent need for gathering and exchanging data on this crucial issue to ensure the best timing of start of dialysis in pregnancy, in the interest of the mothers and the babies.
The second issue, regarding the difficulty in implementing the long, daily dialysis schedules, is also open; this is of utmost importance in developing countries, such as Mexico, where intensive dialysis may lead to a competition for a lifesaving treatment [92]. Shorter dialysis schedules and peritoneal dialysis are reported as alternatives, but publication bias is an important limit to assess the real equivalence of these more easily manageable treatments [128][129][130][131][132] Low income should not become synonymous with low quality, and data coming from medium-low-income countries show that very good results can be obtained in all settings; the populations may, however, be different, as well as the kidney diseases. There is a need for establishing a common language, allowing exchanging on dialysis approaches and detailed results to allow tailoring dialysis to patients, considering also the available resources. Some open questions regarding the management of the dialysis schedules are summarized in Table 5.

Dialysis efficiency
The most common target is "near normal" urea (10-15 umol/L) following the day break

Dialysis membrane Biocompatible membranes
No study addressed to different membranes; low-flux membranes limit nutrient loss; high-flux membranes increase middle molecules clearance, but increase also nutrient loss and may induce back-filtration.

Kidney Transplantation
Fertility is at least partly restored after kidney transplantation, and the possibility to undertake a successful pregnancy is usually considered an added value of a functioning kidney graft. Reports on pregnancies after kidney transplantation rapidly followed the development of kidney transplant programs and thousands of pregnancies have now been reported all over the world [133][134][135][136][137][138][139][140].
However, even in an ideal situation, the risk of complications is higher than in the general population and was recently described as corresponding to stage-1 CKD in native kidneys, in patients with potentially progressive CKD [140,141]. If teratogen drugs are avoided (Appendix A Tables A1-A4), pregnancy after kidney transplantation shares the same risk factors with CKD pregnancies, suggesting that kidney function, hypertension, and proteinuria matter more than treatments [141]. Indeed, the reduced nephron mass of a solitary transplanted kidney may not be resilient enough to the stressors of hyperfiltration of pregnancy, even in the presence of normal kidney function.
The profile of the "ideal" candidate for pregnancy after transplantation is well defined: a young, nonobese, normotensive woman, with normal kidney function, in the absence of proteinuria, without any rejection episode, or at least any recent rejection episode, with good compliance, and at least two years from transplantation [133,134,[140][141][142][143][144]. However, a univocal definition of age, kidney function, and interval after the last rejection episode is missing, and the grading of the risks is not clear, in particular in women with signs of kidney function impairment. This "ideal" situation is indeed not always the rule, in particular in deceased donor transplantation. In fact, expanded donor policies may lead to suboptimal kidney function; higher age at transplantation and reduced fertility are not infrequent; conversely, the good results of pregnancy after kidney transplantation are somehow smoothing the contraindications, with a widespread agreement on a shorter stabilization time, and a permissive attitude towards pregnancy with less-than-optimal kidney function [133][134][135][136][137][138][139][140][141][142][143][144].
The literature is rich with reports on extreme situations, including kidney transplantation during pregnancy; while single cases cannot lead to potentially risky changes in daily practices, such cases do warn against fully negative attitudes and suggest an individually based decision [145][146][147][148][149][150][151].
In particular in western countries, where age of the transplant recipients is often older, assisted fertilization is increasingly popular. Besides the ethical challenges, whose discussion is beyond the scope of this review, assisted fertilization techniques, in particular those in vitro, are associated with an increase of pregnancy complications and of hypertensive disorders of pregnancy, with a potential negative effect on the kidney function [152][153][154][155][156]. The scattered data on assisted fertilization in kidney transplant recipients are, however, encouraging; once more, the series are very small, most of them regarding single cases, and, in such a context, it is highly probable that reports are influenced by a publication bias [152][153][154][155][156].

What Is Missing and Could Be Done in the Clinical Practice
Most of the available data on kidney transplantation regard pregnancies in ideal or "almost ideal" clinical conditions; there is very little evidence available on pregnancy in patients with reduced kidney function. In these cases, in which counseling is particularly important, the data we now have are scant and we can rely only on personal experience or indirect evidence.
There is a need for gathering data on these situations, to improve counseling and to increase awareness of the challenges of pregnancy in patients with a failing graft.
Pregnancy represents an immunologic challenge and is a potential cause of hyperimmunization, a well-known problem for transplantation. Living unrelated grafts increasingly offer a clinical solution within couples: the eventual role of pregnancy as an immunologic trigger leading to rejection of the grafted kidney should be clarified to optimize the outcomes of transplantation between husband and wife [142].
Not least, long-term data on children born from a transplanted mother are still lacking, and attention should be focused on this important issue [157].

Conclusions
Pregnancy is now possible in all phases of chronic kidney disease, but its management may be difficult and the outcomes are not the same as in the overall population.
There is a lot to do to improve pregnancy outcomes in CKD, whose prevalence in pregnancy is probably as high as that of PE, a better-acknowledged risk for adverse outcomes. When CKD is known, pregnancy should be considered as at high risk and followed accordingly, since the earliest stages. Furthermore, since CKD is often asymptomatic, pregnant women should be screened for the presence of CKD, allowing better management of pregnancy and timely treatment after pregnancy. The differential diagnosis between CKD and pre-eclampsia may be difficult but is important for pregnancy management.
Women on dialysis should be aware of the possibility of conceiving and having a successful pregnancy, and intensive, long-hours, quotidian dialysis is the choice allowing the best results; these indications have, however, to be adapted to poorly resourced countries, where daily dialysis may not be feasible for clinical, economical, or logistic issues.
After kidney transplantation, pregnancies should be followed up with great attention to minimize the risks for mother, child, and for graft.
Specialized and committed teams are crucial for optimizing the care of pregnant CKD patients, and efforts should be done to organise such teams.
The advances in our knowledge are strictly linked to the progress in research, and international comparisons are highly needed to ameliorate our understanding and to define treatment strategies. The research agenda is long, from epidemiology of CKD to differential diagnosis between CKD and PE, evaluation of target blood pressure in CKD pregnancies, or the role of the nutritional follow-up. Furthermore, dialysis start, dialysis policy, and modulation of posttransplant therapies should be discussed on a global basis, to identify the best context-sensible policies to allow more CKD women to attain successful pregnancy.

Conflicts of Interest:
The authors declare no conflicts of interest.
Appendix A Table A1. Main immunosuppressive drugs for CKD patients in pregnancy.

Drug
Main Features FDA Usually considered as relatively safe, when absolutely needed

Azathioprine (AZA)
It is teratogenic in animal models, at high doses, but not in humans, possibly because the foetal liver is not able to activate the drug. KDIGO and European Best Practice Guidelines suggest switch from Mycophenolate to AZA before pregnancy.

D Cyclosporine A (CyA)
Hypertension, hyperglycaemia and nephrotoxicity may be relevant in pregnancy. CyA has not been associated with teratogenicity; SGA babies and preterm delivery have been reported, with an unclear link with maternal disease; levels vary in pregnancy and need strict monitoring.

C Tacrolimus
The drug has similar effects and side effects as CyA; the experience is more limited than with CyA. C

Steroids
The most frequently used short-acting corticosteroids are prednisone, methylprednisolone and prednisolone; among long-acting betamethasone and desamethasone. No major malformation reported, increase of labio-palatoschisis is debated. A risk of premature rupture of membranes has been reported, along with increased risk of infection and gestational diabetes.

Usually considered as first choice
Alpha-methyldopa Widely used in pregnancy, with no reported negative effects. May not be sufficient to correct severe hypertension in CKD. B

Niphedipine
The long acting form is commonly used. In CKD the side effect of increasing peripheral oedema may be relevant. C

Labetalol
Usually well tolerated, should be avoided in asthma. In a RCT it was comparable to alpha-methyldopa. C

Beta-blockers
The main negative effect in older studies was foetal growth restriction, possibly as an effect of overzealous correction of BP. Beta1 selective beta-blockers (atenolol) are more often in cause. May be more effective than alpha-methyldopa, alone or in combined therapy. At delivery they may induce hypoglycaemia, hypotension and bradycardia (usually mild and transient).
The effect is similar to alpha-methyldopa; side effects may be more common and hypertensive rebounds at discontinuation are common; slowing fetal growth is occasionally reported.   Malformation risk is unlikely in humans. In some patients, permanent hypercalcemia resulted in adverse effects on the fetus. No available data in humans. Low intake is associated with adverse pregnancy events.

Not assigned
To be avoided

Warfarin
Evidence of teratogenicity in animal studies. In humans, exposure during the first trimester caused congenital malformations in about 5% of the exposed. Mental retardation, blindness, schizencephaly, microcephaly, hydrocephalus, and other adverse pregnancy outcomes have been reported following exposure in the second and third trimesters. Fatal fetal hemorrhage and an increased risk of spontaneous abortion and fetal mortality are also reported. Case by case evaluation in the case of cardiac indications.