Background

Through the technical advancement of dialysis therapy, long-term prognosis and quality of life in patients with end-stage renal failure have drastically improved in recent decades. The average age of dialysis initiation has also been increasing year after year. However, many young females require chronic dialysis therapy; therefore, nephrologists are challenged by the important issue of pregnancy in such patients. In general, successful pregnancy is uncommon in dialysis patients. After the first report of successful delivery in 1971 [1], a number of case reports and reviews have been published about pregnancy and delivery in patients on dialysis [2,3,4,5,6,7]. Holley and Reddy suggested the ideal condition of hemodialysis for having a healthy child [8]. However, there are still many problems that hinder successful full-term pregnancy in patients on dialysis, partly due to complications including spontaneous miscarriage, premature delivery, intrauterine growth inhibition, polyhydramnios, and pregnancy-induced hypertension [8].

One of the important issues is the determination of dry weight (DW), which should be changed according to fetal growth. In nonpregnant patients, the cardiothoracic ratio obtained by X-ray photography is used together with the blood pressure (BP) and edematous condition to determine the ideal DW. Recently, the plasma levels of atrial and brain natriuretic peptides (ANP and BNP) have been recognized as useful markers for not only the assessment of cardiovascular risk but also the determination of the appropriate extracellular fluid volume status [9,10,11]. X-ray exposure is not preferable for pregnant women especially during the early pregnancy period. Furthermore, most Japanese people became extremely concerned about radiation after the serious accident that occurred at the Fukushima Daiichi Nuclear Power Plant. This report describes a successful pregnancy and full-term delivery of an infant in a dialysis patient using mainly ANP and BNP, instead of X-ray examination, to determine DW.

Case presentation

A 38-year-old female receiving maintenance hemodialysis was referred to our hospital for control during pregnancy. She was diagnosed with IgA nephropathy by renal biopsy at 20 years of age. At ages 26 and 29, she delivered full-term infants with normal birth weights without complications. Then, she discontinued attending the hospital regularly and her renal function deteriorated with time. Four years later, hemodialysis was initiated in our hospital due to uremia. When hospitalized, she was diagnosed as having heparin-induced thrombocytopenia type II (HIT type II); at this time, nafamostat mesilate was chosen as an anticoagulant agent. After discharge, she continued maintenance hemodialysis at a satellite facility during which time her case remained uneventful. Five years later, she was diagnosed as 5-week pregnant at the maternity division in our hospital. She was referred to our division for the maintenance of dialysis therapy.

Blood data before dialysis and echocardiographic findings after dialysis are shown in Table 1. Hemoglobin (Hb) and electrolyte levels were well controlled, but the intact parathyroid hormone (iPTH) level was high. The blood urea nitrogen (BUN) concentration was below 50 mg/dL. The postdialysis ANP level was slightly high, but the BNP level was maintained low. Slight pericardial effusion was observed, but cardiac function was well preserved. According to previous reports on management guidance [8, 12], we performed sufficient dialysis over 20 h/week. Because she had HIT, nafamostat mesilate was used as an anticoagulant agent.

Table 1 Physical findings and laboratory data upon admission
Full size table

The changes in the patient’s body weight (BW) and BP are shown in Fig. 1. DW started to increase after the 12th week of gestation by 0.3–0.5 kg/week, in accordance with the guide and from recommendations by obstetricians of our hospital. Weight gain between each dialysis session remained almost stable, within a 5% DW increase between each session. For determining the ideal DW, we measured the plasma levels of ANP and BNP (Fig. 1) rather than performing the X-ray examination to avoid radiation exposure. Echocardiographic data were also followed each month after the last dialysis session of the week to evaluate the diameter of inferior vena cava (IVC), as well as cardiac function. During the observation period, cardiac function remained normal and well preserved. Measurement of the IVC diameter was considered useful at an early stage of pregnancy. At a later stage of pregnancy, however, the IVC became collapsed even before the dialysis session due to the fetal growth, making it difficult to evaluate body fluid status echocardiographically. Target BUN and Hb levels before the dialysis session were below 50 mg/dL and over 10 g/dL, respectively. The hemodialysis dose was increased to 4 times/week after 7 weeks of gestation and was further increased to 5 times/week after 26 weeks of gestation.

Fig. 1
figure 1

Clinical course of the patient during pregnancy. HD, hemodialysis; BW, body weight; DW, dry weight; BP, blood pressure; BUN, blood urea nitrogen; ANP, atrial natriuretic peptide; BNP, brain natriuretic peptide

Full size image

At 29 weeks of gestation, the patient suffered from impending preterm delivery and was admitted to our hospital for the remainder of her pregnancy. She started to receive 50 μg/kg/min of ritodrine hydrochloride. Because her weight gain between dialysis sessions became larger and the BUN levels tended to increase, the hemodialysis dose was increased to 6 days/week (4.5 h/session) starting at 31 weeks of gestation. Thereafter, the BUN levels remained mostly below 50 mg/dL until full-term (Fig. 1). During the gestation period, the single-pool Kt/V ranged from 1.3 to 2.2 (average, 1.8 ± 0.05). The Hb levels were maintained between 10 and 12 g/dL with the administration of an erythropoietin analog and iron. Calcium, phosphate, and PTH levels were controlled within recommended ranges throughout the course with vitamin D and phosphate-binding agents.

During the dialysis session, the BP levels were targeted to remain below 140/90 mmHg with minimal BP fluctuations. After being referred to our hospital, oral administration of 750 mg/d methyldopa hydrate was started. Nifedipine controlled-release (CR) was used up to 80 mg/d from 31 gestation weeks for the gradual increase in BP. The ANP and BNP levels remained almost stable despite BW increases near the delivery date (Fig. 1).

At 37 weeks of gestation, she had genital bleeding and underwent emergency cesarean section. After a 45-min surgery, she delivered a 2372 g (− 0.6SD) healthy girl who had an appropriate-for-date birth weight. Apgar scores were 8 and 9 at 1 and 5 min, respectively. After the delivery, the patient returned to regular hemodialysis (3 times/week) without peripartum complications and was discharged with her baby 10 days after delivery.

Discussion and conclusions

In dialysis patients, the chance of pregnancy is rare, partly due to anovulation, hyperprolactinemia, and decreased libido caused by altered human chorionic gonadotropin pulses and reduced renal leptin clearance [13]. The registry revealed that only 2% of women on chronic dialysis became pregnant over a 4-year period [14]. Levy et al. proposed the management guide to optimize pregnancy outcomes in dialysis patients [12]. First, the dialysis dose must be increased to keep the levels of uremic toxins low. Second, the BP should be stabilized for maintaining stable placental circulation. Most importantly, DW should be increased carefully according to the gestation week and fetal growth. Many evaluations such as the edematous condition, BP levels, X-ray exam, Crit-Line measurement, IVC diameter, and plasma ANP or BNP levels are usually employed to determine the ideal DW. Among them, we decided to use mainly the ANP and BNP levels, along with the edematous condition, BP, and IVC diameter to avoid X-ray exposure.

Increases in ANP and BNP, cardiac hormones secreted predominantly from the atria and ventricles, respectively, are well known to be increased in congestive heart failure [15,16,17,18]. However, plasma levels of these peptides are also elevated in patients undergoing dialysis in response to hypervolemia, as well as latent or overt ventricular dysfunction [19]. In our case, we mainly used the plasma levels of ANP and BNP to control appropriate hydration status and evaluate cardiac function. Eventually, these peptide levels remained stable throughout the gestation period, and BP and amniotic fluid volume were kept quite stable until the full-term delivery of a healthy baby. We used an X-ray exam only once at a stable period (29th gestation week; CTR 46.0%). There are a few clinical studies regarding delivery in patients on hemodialysis. A retrospective, multicentric study consisting of 100 pregnancies in patients on hemodialysis from 1985 to 2015 in France revealed a mean gestational age of 33.2 ± 3.9 weeks and a mean birth weight of 1719 ± 730 g [20]. Another retrospective cohort study consisting of 93 pregnancies in patients on hemodialysis from 2000 to 2017 in Brazil revealed a median gestational age of 35 (12–39) and a mean birth weight of 1698 ± 719 g; in addition, preeclampsia, primigravida, average midweek BUN, polyhydramnios, and residual diuresis were associated with birth weight [21]. In a nationwide survey on pregnancy in patients with end-stage renal disease carried out in Japan, neonates who died were younger in gestational age (26.2 weeks) with lower birth weight, and all infants who died were born less than 1000 g [22]. These authors concluded that premature delivery is a major problem in patients on chronic dialysis in spite of the improved rate of pregnancy, impeding successful outcomes without complications. As case reports have reported for the last several years (Table 2), premature delivery is still a problem that should be considered. To this end, the present case suggests that strict volume control is crucial. Thus, we propose monitoring plasma ANP and BNP levels as a useful strategy for managing volume during pregnancy and achieving successful full-term delivery. In addition, rigorous managements of anemia and mineral and bone disorder, which are inevitable complications in dialysis patients, also should be considered to keep pregnancy in good condition and to reach full-term delivery.

Table 2 Case reports on delivery with maintenance hemodialysis or peritoneal dialysis
Full size table

Future studies are required to further explore the regulation and monitoring of hydration status and the best management in dialysis patients during pregnancy for a successful and safe delivery. In our case, ANP and BNP levels were stable and relatively low, reflecting good control of hydration status. Stability of ANP and BNP levels could lead to full-term delivery. Appropriate ranges of ANP and BNP need to be examined.