Impact of assisted reproductive therapy on kidney and pregnancy outcomes in women with early-stage chronic kidney disease: a longitudinal analysis.

Between January 2014 and December 2023, 36 women with pre-existing CKD attended the Reproductive Medicine Center of Sun Yat-sen Memorial Hospital for ART counselling. As of the data-lock point, 19 had achieved a live birth, 12 had not conceived, and five had not yet started ovarian stimulation; consequently, the 19 live-birth cases constituted the CKD group. Fifty-seven non-CKD women who underwent ART and delivered during the same interval were selected as controls (1:3 matching) (Fig.  1 ). Fig. 1 Flowchart of the enrollment of CKD and control groups Flowchart of the enrollment of CKD and control groups Following matching for age and BMI, baseline demographics and laboratory parameters—including age, BMI, smoking status, pre-existing hypertension, diabetes mellitus, etiology of infertility, mean arterial pressure, eGFR, AMH, serum albumin, low-density lipoprotein cholesterol, alanine aminotransferase, and hemoglobin—were comparable between the CKD and control groups (all p  > 0.05, Table  1 ). One woman in the CKD cohort had antecedent hypertension; no participant in either group had pre-gestational diabetes. Table 1 Demographic and baseline characteristics in the CKD and control groups CKD group ( n  = 19) Control group ( n  = 57) P value Demographic  Age 33.8 ± 5.1 32.6 ± 3.5 0.345  BMI 21.3 ± 1.9 21.6 ± 2.8 0.573  Habit of smoking 0 (0.0) 0 (0.0) - Pre-existing disease  Hypertension 1 (5.3) 0 (0.0) 0.250  Diabetes mellitus 0 (0.0) 0 (0.0) - Etiology of infertility 0.151  Female 14 (73.7) 27 (47.4)   Tubal 7 (36.8) 14 (24.6)   Anovulatory disorders 4 (21.0) 6 (10.5)   Premature ovarian insufficiency 2 (10.5) 1 (1.8)   Advanced age 1 (5.3) 0 (0.0)   Endometriosis 0 (0.0) 5 (8.8)   Chromosome abnormality 0 (0.0) 1 (1.8)  Male 3 (15.8) 19 (33.3)  Unexplained infertility 2 (10.5) 11 (19.3) Baseline clinical characteristics  Proteinuria a 4 (21.1) 0 (0.0) 0.003  Mean arterial pressure (mmHg) 82.7 (78.7, 90.5) 84.0 (79.0, 92.0) 0.494  eGFR (ml/min/1.73m 2 ) 105.0 ± 12.6 102.9 ± 13.3 0.540  Serum anti-Müllerian hormone (ng/ml) 3.5 (1.6, 8.3) 2.9 (1.6, 6.0) 0.489  Serum albumin (g/L) 40.8 (38.3, 44.7) 41.8 (37.6, 46.2) 1.000  Serum low density lipoprotein (mmol/L) 2.9 (2.7, 3.3) 2.9 (2.4, 3.2) 0.647  Serum alanine aminotransferase (U/L) 12.9 (10.5, 15.0) 14.0 (11.0, 18.0) 0.227  Blood hemoglobin (g/L) 127.5 ± 10.8 130.5 ± 9.3 0.293  Platelet counts 276.4 ± 69.0 293.1 ± 69.6 0.369  Lactate dehydrongenase 184.9 ± 21.8 178.5 ± 24.3 0.288 Data were presented as n (%), mean ± SD or median (Q1,Q3).CKD, chronic kidney disease; BMI, body mass index; eGFR, estimated glomerular filtration rate a Proteinuria was defined as urine protein qualitative test positive or urine protein quantity more than 150 mg/day Demographic and baseline characteristics in the CKD and control groups Data were presented as n (%), mean ± SD or median (Q1,Q3).CKD, chronic kidney disease; BMI, body mass index; eGFR, estimated glomerular filtration rate a Proteinuria was defined as urine protein qualitative test positive or urine protein quantity more than 150 mg/day In the CKD cohort, isolated glomerular hematuria was the predominant presentation (15/19, 78.9%). Concomitant hematuria and proteinuria were observed in three women, and isolated proteinuria in one, yielding an overall proteinuria prevalence of 21.1% (4/19). One patient (5.3%) had pre-existing hypertension. The vast majority were CKD stage 1 (18/19, 94.7%); one woman was stage 2. Kidney biopsy was performed in four cases (21.1%), revealing IgA nephropathy( n  = 2), C3 glomerulonephritis ( n  = 1) and thin-basement-membrane nephropathy( n  = 1). With respect to ART modality, 18 women underwent IVF-ET, and one underwent AI (Table  2 and Supplementary Table 1). Table 2 Renal information of CKD women at baseline CKD group ( n  = 19) Clinical manifestation  Isolated glomerular hematuria 15 (78.9)  Concomitant hematuria and proteinuria a 3 (15.8)  Isolated proteinuria 1 (5.3) Etiology of CKD  IgA nephropathy 2 (10.5)  C3 glomerulonephritis 1 (5.3)  Thin basement membrane nephropathy 1 (5.3)  Glomerulonephritis without renal pathology 15 (78.9) Stage of CKD  1 18 (94.7)  2 1 (5.3) Data were presented as n (%). CKD, chronic kidney disease a Proteinuria was defined as urine protein qualitative test positive or urine protein quantity more than 150 mg/day Renal information of CKD women at baseline Data were presented as n (%). CKD, chronic kidney disease a Proteinuria was defined as urine protein qualitative test positive or urine protein quantity more than 150 mg/day Urine protein, blood pressure and serum creatine were monitored during pregnancy and the perinatal period. In the CKD group, one patient had lost follow-up of kidney information, which was excluded from the relevant analysis. The remaining patients exhibited stable proteinuria and kidney function throughout pregnancy, except for one individual presented with a doubling of proteinuria in the late trimester and developed post-partum AKI. The case of AKI was attributed to inadequate effective circulating blood volume, secondary to massive proteinuria. Serum creatinine peaked at 110 μmol/L and then fell to 61 μmol/L in parallel with the remission of proteinuria following corticosteroid therapy (Table  3 and Supplementary Table 2). In the control group, all women maintained stable kidney function while one woman had new-onset proteinuria due to pre-eclampsia in the late trimester of pregnancy. Among pregnant women, new-onset hypertension attributable to preeclampsia occurred in 11.1% (2/18) of those with pre-existing CKD versus 3.5% (2/57) of controls, and antihypertensive therapy was initiated in every case( p  = 0.240). In addition, no significant difference of variation in proteinuria and kidney function was found in both groups (both p  > 0.05, Table  4 ). Table 3 Clinical details regarding the AKI case Proteinuria (g/24 h) Blood Pressure (mmHg) SCr (μmol/L) eGFR (ml/min/1.73m 2 ) Baseline 2.1 127/79 68 90.7 Third trimester 19.8 145/81 69 101.9 One week postpartum 13.6 130/82 110 58 One month postpartum 2.3 126/76 61 116.8 SCr serum creatinine, eGFR estimated glomerular filtration rate Table 4 The maternal kidney outcomes in the CKD and control groups CKD group ( n  = 18) Control group ( n  = 57) P value Proteinuria a  New-onset proteinuria 0 (0.0) 1 (1.8) 1.000  Doubling of proteinuria 1 b (5.6) 0 (0.0) 0.240 New-onset hypertension 2 (11.1) 2 (3.5) 0.242 Kidney function 0.240  Stable 17 (94.4) 57 (100.0)  Worsen 1 (5.6) a 0 (0.0) Data were presented as n (%). CKD, chronic kidney disease a In pregnancy, pathological proteinuria was defined as urinary protein ≥ 0.3 g/24 h b The patient had a doubling of proteinuria in the late trimester of pregnancy and acute kidney injury in the perinatal period Clinical details regarding the AKI case SCr serum creatinine, eGFR estimated glomerular filtration rate The maternal kidney outcomes in the CKD and control groups Data were presented as n (%). CKD, chronic kidney disease a In pregnancy, pathological proteinuria was defined as urinary protein ≥ 0.3 g/24 h b The patient had a doubling of proteinuria in the late trimester of pregnancy and acute kidney injury in the perinatal period The incidence of OHSS, gestational diabetes mellitus, gestational hypothyroidism, intrahepatic cholestasis of pregnancy, placental abnormalities, preterm delivery, caesarean section, and postpartum hemorrhage did not differ significantly between the CKD and control groups (all p  > 0.05, Table  5 ). Table 5 Maternal complications in the CKD and control groups CKD group ( n  = 19) Control group ( n  = 57) P value Eclampsia associated parameters  Platelet counts 279.1 ± 48.8 263.4 ± 49.0 0.234  Lactate dehydrongenase 190.0 ± 25.5 183.2 ± 29.2 0.341 OHSS(moderate degree or above) 1 (5.3) 5 (8.8) 1.000 Gestational diabetes 4 (22.2) 20 (35.1) 0.308 Gestational hypothyroidism 2 (11.1) 5 (8.8) 1.000 Intrahepatic cholestasis of pregnancy 1 (5.6) 0 (0.0) 0.240 lacental abnormality 3 (16.7) 5 (8.8) 0.611  Placenta previa 1 (5.6) 3 (5.3)  Low-lying placenta 0 (0.0) 2(3.5)  Placenta accreta 1 (5.6) 0 (0.0)  Placental abruption 1 (5.6) 0 (0.0)  Placenta increta 0 (0.0) 1 (1.8) Preterm 3 (15.8) 7 (12.3) 1.000 Caesarean section 12 (63.2) 35 (61.4) 0.892 Postpartum hemorrhage 2 (11.1) 5 (8.8) 1.000 Data were presented as n (%). CKD, chronic kidney disease; OHSS, ovarian hyperstimulation syndrome Maternal complications in the CKD and control groups Data were presented as n (%). CKD, chronic kidney disease; OHSS, ovarian hyperstimulation syndrome There were 21 live-born neonates in the CKD group and 66 in the control group, with no stillbirths recorded. Gestational age at delivery, birth weight, rates of NICU admission, preterm birth, small- or large-for-gestational-age status, and congenital anomalies did not differ significantly between the two groups (all p  > 0.05, Table  6 ). Table 6 Neonatal outcomes in the CKD and control groups CKD group ( n  = 21) Control group ( n  = 66) P value Gestational week 39.0 (37.9, 39.4) 38.3 (37.3, 39.6) 0.937 Birthweight (g) 3200.0 (2700.0, 3600.0) 2995.0 (2572.5, 3292.5) 0.179 Admition in NICU 7 (33.3) 33 (50.0) 0.182 Preterm 5 (23.8) 10 (15.2) 0.560  Very preterm ( 4000 g) 1 (4.8) 0 (0.0)  Low birthweight (< 2500 g) 2 (9.5) 14 (21.2)  Very low birthweight (< 1500 g) 2 (9.5) 0 (0.0) Congenital malformations 3 (15.0) 10 a (15.2) 1.000  Congenital heart disease 2 (10.0) a 6 (9.1)  Chromosome abnormality 0 (0.0) 2 (3.0)  Biliary atresia 1 (5.0) 0 (0.0)  Retinopathy of prematurity 0 (0.0) 1 (1.5)  Indirect inguinal hernia 0 (0.0) 2 (3.0)  Syndactyly 0 (0.0) 1 (1.5) Data were presented as n (%) or median (Q1, Q3). CKD, chronic kidney disease; NICU, neonatal intensive care unit a One infant had congenital heart disease along with chromosome abnormality. One infant had congenital heart disease comorbid with indirect inguinal hernia Neonatal outcomes in the CKD and control groups Data were presented as n (%) or median (Q1, Q3). CKD, chronic kidney disease; NICU, neonatal intensive care unit a One infant had congenital heart disease along with chromosome abnormality. One infant had congenital heart disease comorbid with indirect inguinal hernia

Between January 2014 and December 2023, women with pre-existing CKD who conceived via ART at Sun Yat-sen Memorial Hospital and achieved a live birth were consecutively enrolled and designated as the CKD group. Patients with any of the following pre-existing diagnoses documented before ART initiation were excluded: 1) hypertension not attributable to glomerular or interstitial nephritis; 2) diabetes mellitus; 3) heart diseases; 4) hepatitis or liver cirrhosis; 5) hematological system diseases; 6) tumor; 7) rheumatic diseases, such as systemic lupus erythematosus, Sjogren's syndrome, rheumatoid arthritis, vasculitis, and so on. During the same interval, non-CKD women who conceived via ART and achieved a live birth at our institution were serially screened for inclusion as controls, with identical exclusion criteria applied. An optimal full-matching of propensity score matching was then performed at a 1:3 ratio, with maternal age and BMI included as covariates to ensure comparability between groups. Only women with early-stage CKD (stages 1–2) were included, as patients with advanced CKD or higher risk of renal progression were not eligible for ART in our center due to ethical and safety considerations. After matching, the final cases were included in the control group. The study had received approval from the Ethics Review Board of Sun Yat-sen Memorial Hospital (No. SYSKY-2023–008-01). Hematuria was defined as ≥ 3 red blood cells per high-power field on urine microscopy; cases in which dysmorphic erythrocytes predominated were classified as glomerular hematuria [ 10 ]. Proteinuria was defined as urine protein qualitative test positive or urine protein quantity more than 150 mg/24 h [ 11 ]. For pregnant women, pathological proteinuria was defined as urinary protein ≥ 0.3 g/24 h [ 12 ]. All abnormal urine results were confirmed by at least two repeat tests. The definitions and staging of CKD and AKI referred to the KDIGO guidelines [ 13 ]. Estimated glomerular filtration rate(eGFR) was calculated by CKD-EPI equation [ 14 ]. Kidney function change was assessed by comparing eGFR at three predefined time points: within one month before ART treatment (baseline), during pregnancy, and within one week postpartum. Kidney function was then classified as follows: worsen with an eGFR reduction ≥ 25% from baseline; stable with an eGFR fluctuation < 25% of baseline [ 15 ]. The following data were collected: demographic information[age, body mass index (BMI), smoking-habit], pre-existing disease, cause of infertility, baseline clinical parameters[blood pressure, urine erythrocyte, urine protein, serum creatinine (SCr), estimated glomerular filtration rate (eGFR), serum anti-Müllerian hormone (AMH), serum albumin, serum low density lipoprotein, serum alanine aminotransferase, blood hemoglobin], data of the patients when follow-up[blood pressure, urine protein, SCr, eGFR, gestational age at delivery, gestational and perinatal complications], neonatal information[neonatal birthweight, congenital malformations, admission to the NICU]. An optimal full-matching propensity score analysis was performed to select the optimal patients of control at a ratio of 1 to 3. Given the relatively small sample size, the continuous variables were expressed as mean ± standard deviation(SD) and median (interquartile range), and compared by t-test or Mann–Whitney U test. The categorical variables were presented as numbers and percentages, and compared using the chi-square (χ 2 ) test (with or without Yates correction) or Fisher exact test. Statistical analyses were performed using SPSS (version 26.0) and R Studio (version 2023.06.1) software. All tests were two-tailed and statistical significance was considered with a p -value < 0.05.

To our knowledge, comparative data on ART outcomes in women with pre-existing CKD remained scarce. We therefore undertook a prospective study to evaluate kidney trajectory and obstetric/neonatal outcomes in CKD stages 1–2 after ART conception. The results indicate that, when compared with a matched non-CKD ART population, women with early-stage CKD did not exhibit a higher incidence of OHSS, kidney-function deterioration, adverse maternal events, or neonatal complications. OHSS was a well-recognized complication of controlled ovarian stimulation; established risk factors include polycystic ovary syndrome (PCOS) and elevated AMH levels [ 16 ]. Previous work reported moderate OHSS in 3–6% of CKD patients undergoing ART [ 9 ]. In the present cohort, moderate OHSS occurred in 5.3% (1/19) of women with CKD, a figure consistent with earlier estimates [ 9 , 16 ]. The comparable risk between groups was attributable to similar baseline rates of PCOS (10.5% vs 8.8%; data not shown) and AMH concentrations (both p  > 0.05). To date, evidence on ART in CKD derived almost exclusively from small descriptive series. Among 13 kidney-transplant recipients, 23% exhibited a ≥ 30% rise in serum creatinine after ART [ 17 ]. Another report documented progression to dialysis-requiring AKI in a woman with stage 2 CKD who conceived via ART [ 18 ]. A systematic review pooling 68 CKD patients observed AKI in 10 (14.7%) pregnancies achieved with ART [ 9 ], and contemporary cohorts confirmed that stages 3–5 CKD frequently precipitated irreversible loss of maternal kidney function [ 7 ]. Hypertension and proteinuria have consistently been identified as independent predictors of accelerated CKD progression [ 19 ].All CKD-patients in our study were in early CKD stages with preserved kidney function; 18 out of 19 (94.7%) were normotensive and 15 out of 19 (78.9%) had no proteinuria prior to ART. In addition, patients at higher risk (such as those with advanced-stage CKD or kidney transplantation) were not included, thus careful patient selection at our center might contribute to the favorable outcomes in CKD group. In our study, one patient (5.6%) in CKD group experienced a doubling of proteinuria level and developed AKI. However, no significant between-group difference in kidney outcomes was detected between participants with and without CKD, likely attributable to the comparable eGFRs and similar distributions of major risk factors-hypertension and proteinuria. Prior work indicated that maternal–fetal risk escalates with advancing CKD: stages 3–5 were associated with significantly higher rates of pre-eclampsia, spontaneous preterm delivery, and NICU admission than stages 1–2 [ 20 ]. In the present study, the frequencies of these events among women with stage 1–2 CKD were indistinguishable from those reported for the general obstetric population conceiving via ART [ 21 – 23 ]. Compared with matched controls, we observed no increase in preterm birth, placental disorders, post-partum hemorrhage, caesarean delivery, low birth-weight, congenital malformations, or NICU admission (all p  > 0.05). Thus, early-stage CKD (stages 1–2) did not confer additional risk for obstetric or neonatal complications in the setting of ART. Several limitations need to be clarified. First, the sample size was limited by the intrinsic rarity of CKD in women seeking ART. Secondly, given the ethical and safety considerations, only women with early-stage CKD (stages 1–2), preserved kidney function, infrequent hypertension, and minimal or no proteinuria were enrolled. Consequently, our findings primarily characterized this low-risk cohort and may not be generalizable to women with more advanced CKD. Moreover, multi-center prospective cohort studies and long-term follow-up were required in the future. In conclusion, among women with early-stage(stages 1–2) CKD, ART appears to be as safe as in the general population. In addition, neither OHSS, kidney-function decline, nor obstetric/neonatal complications were significantly increased.

Chronic kidney disease (CKD) was prevalent increasingly and considered the fifth cause of death by 2040 worldwide [ 1 ], which affected up to 6% of childbearing-aged women in developed countries [ 2 ] and could cause infertility mainly through the hypothalamic-pituitary–gonadal axis [ 3 ]. In women with CKD, impaired pulsatile gonadotropin-releasing hormone secretion disrupted luteinizing hormone/follicle-stimulating hormone cycling and estrogen production, leading to anovulation via diminished hypothalamic–pituitary–ovarian axis feedback [ 3 , 4 ].As reported, the pregnancy rates were only approximately 10% among kidney transplant recipients and 1% in dialysis patients [ 4 ]. Giving these challenges, assisted reproductive technology (ART), including artificial insemination (AI) and in vitro fertilization with embryo transfer (IVF-ET), had become an important option for CKD women seeking for pregnancy. However, ART might be associated with complications such as ovarian hyperstimulation syndrome (OHSS), which could induce acute kidney injury (AKI) [ 5 ]. In addition, pregnancy itself imposed physiological changes such as glomerular hyperfiltration, which might further increase the burden on the kidneys [ 4 , 6 ]. As reported, pregnancy in women with CKD Stages 3–5 advanced the need for dialysis or transplantation by 2.5 years [ 7 ]. Meanwhile, CKD was reported to be related with higher risks of adverse maternal and neonatal outcomes, including preterm delivery, pre-eclampsia, small-for-gestational-age neonates, fetal growth restriction, admission to neonatal intensive care units (NICU), and so on. It was revealed that poorer renal function was associated with a correspondingly higher risk of the aforementioned outcomes [ 2 , 4 , 8 ]. Until now, limited comparative clinical study had focused on CKD patients receiving ART. Only case reports and small-sample descriptive studies had been published [ 9 ]. Therefore, we performed a longitudinal cohort study to evaluate kidney outcomes and pregnancy-related adverse events in women with early-stage CKD who conceived through ART.

Supplementary Material 1. Supplementary Material 1.