Pregnancy in Women Treated With Dialysis: Lessons From a Large Series Over 20 Years
Article Outline
It is widely stated and believed, but not documented, that the frequency of pregnancy in women treated with dialysis is increasing. Even if the increase is real, pregnancy in dialysis patients remains uncommon and difficult to study. These pregnancies occur in widely separated dialysis units and the practice of intensive dialysis makes it difficult to refer pregnant women to central dialysis units experienced in providing dialysis therapy to pregnant women and to initiate any prospective studies of various dialysis regimens, approaches to treating anemia, or the host of other management questions that arise. For the same reason, most nephrologists will encounter only 1 or 2 pregnant dialysis patients during their time in practice.
In this issue of the American Journal of Kidney Diseases, Luders and colleagues at São Paolo University Medical Center describe the largest number of pregnancies in dialysis patients at a single institution.1 The 52 pregnancies occurred over a 20-year period from 1988 to 2008. Despite the interval over which the pregnancies occurred and the range of residual kidney function in the patients, the treating physicians evolved a consistent and coherent approach to caring for these women.
This is the first report of systematic adjustment of the dialysis dose based on the patient's weight and residual kidney function, as indicated by time on dialysis and residual urine output. All 28 women who conceived before starting dialysis were treated less intensively (mean dialysis time 12 h/wk) than the 24 women who conceived after starting dialysis (mean dialysis time 15 h/wk). No differences in outcome were noted in women who conceived before or after starting dialysis, including mean birth weight and gestational age, percentage of preterm births, and percentage of surviving infants. However, the outcome in these pregnancies was so good that the number of adverse events was low and we may not see a difference until the authors report even more pregnancies.
This report by Luders and colleagues is also the first to describe a number of pregnant women treated with short daily dialysis. Most previous recommendations about dialysis regimens have been firm in the belief that the most important variable in Kt/V as applied to pregnant women is time (t). The observation that women dialyzed more than 20 h/wk have better outcomes than women dialyzed for shorter times was statistically significant, but was applied only to women that conceived after starting dialysis and that reached the second trimester.2 No analysis was done on the basis of residual kidney function. Barua and colleagues in Toronto reported 6 pregnancies in women on dialysis treated with an average of 36 hours of nocturnal dialysis at the time of conception and an average of 48 h/wk during pregnancy, which resulted in 6 surviving infants with only 1 born at less than 36 weeks. These unprecedented results reinforce our views that time on dialysis is an important factor in outcomes.3 Our usual approach is to prescribe 4 hours of dialysis 6 d/wk in an attempt to be sure of reaching 20 h/wk even with access problems and snow days. The schedule is onerous for patients, whose motivation may fade as the pregnancy progresses. In the United States, it is difficult to offer even the most motivated pregnant woman 48 h/wk of dialysis since nocturnal dialysis programs (home or in center) are not yet widely available. The overall success rate of 87% in the current report compares favorably even to the outcomes of pregnancy in transplant recipients, although the mean gestational age of 32.7 weeks is not as good. The success rate of 79% infant survival for pregnancies in women who conceived after starting dialysis is dramatically better than the <50% we reported in 1998.4 Nonetheless, the spectacular success of a handful of patients treated with nocturnal dialysis in Toronto will make it difficult to pry us away from the idea that longer dialysis time is better.
Four of the fetal losses in this series were stillbirths (7.6%) similar to 8.1% of pregnancies recorded in the American National Registry of Pregnancy in Dialysis Patients. In our report, no stillbirths occurred in women receiving ≥20 hours of dialysis. Stillbirths have also occurred in peritoneal dialysis patients.5 In addition, fetal losses have occurred in women receiving short daily dialysis.
The biggest risk factor for adverse outcomes noted in this series was preeclampsia, which occurred in 10 of the 52 pregnancies. Only 6 of these 10 babies survived. All were premature and the mean gestational age and birth weight were lower than in patients without preeclampsia. The diagnosis of preeclampsia is very difficult to make in women with kidney disease. Luders and colleagues were careful not to classify every woman with hypertension, even severe hypertension, as having preeclampsia. The difficulty for most investigators addressing the problems of pregnancy in women with kidney disease is that the diagnosis of preeclampsia can be made with confidence only if the preeclampsia is severe. We are beginning to understand the role of angiogenic and antiangiogenic factors in the pathophysiology of preeclampsia. It would not be surprising if levels of soluble vascular endothelial growth factor, placental growth factor, and soluble endoglin were different in women with kidney disease than in healthy women, and as the measurement of these factors becomes more widespread, we may have another tool for sorting out the diagnostic dilemma of preeclampsia in pregnant women with kidney disease.
Absent from the report by Luders and colleagues is a discussion of what has always been one of the greatest hurdles to successful pregnancy, extremely premature labor. The severely premature deliveries were almost entirely confined to women with preeclampsia, with only 3 of 42 pregnancies without preeclampsia resulting in birth before 30 weeks' gestation.
Luders and colleagues provide several important insights. Women who start dialysis after conception need not be on a dialysis regimen of 20 or 48 h/wk. Over 75% of women who start dialysis after conception have a surviving infant, although those who have cared for even 1 woman with a late fetal loss know that 75% is not good enough. We have concentrated on outcomes in women who conceive after starting dialysis because their outcomes have been so poor. There may also be benefit to carefully measuring residual kidney function in women who conceive after starting dialysis. It is clear that the prevention of preeclampsia is urgent in women with kidney disease as it is in all areas of obstetrics. We are indebted to Luders and colleagues for providing information on a large number of pregnancies in women treated with dialysis over 20 years and we look forward to their next report. I am not quite ready to give up the idea that time is the key to successful pregnancy in women with no residual kidney function.
Acknowledgements
Financial Disclosure: The author declares she has no relevant financial interests.
References
- Obstetric outcome in pregnant women on long-term dialysis: a case series. Am J Kidney Dis. 2010;56(1):77–85
- . Daily dialysis in pregnancy. Hemodial Int. 2004;8(2):167–171
- Successful pregnancies on nocturnal home hemodialysis. Cln J Am Soc Nephol. 2008;3(2):392–396
- . Registry of pregnancy in dialysis patients. Am J Kidney Dis. 1998;31(5):766–773
- . Pregnancy in a continuous ambulatory peritoneal dialysis patient. Perit Dial Int. 1983;3(1):13–14
PII: S0272-6386(10)00835-8
doi:10.1053/j.ajkd.2010.05.002
© 2010 National Kidney Foundation, Inc. Published by Elsevier Inc All rights reserved.
Refers to article:
- Obstetric Outcome in Pregnant Women on Long-term Dialysis: A Case Series , 12 April 2010
