Pregnancy after Kidney Donation: More Questions than Answers

Michelle A. Josephson


Two recent reports in the American Journal of Transplantation focus on the maternal and fetal outcomes of pregnancies in kidney donors and provide tantalizing, if somewhat worrisome, observations. The findings also leave us with several important unanswered questions.

Marginal increases in blood pressure and diminished kidney function have been noted in kidney donors.[1,2] As a result of these observations, Reisæter et al. were concerned that female kidney donors might be at risk for hypertensive disorders of pregnancy. To test their hypothesis, the authors analyzed data from the Medical Birth Registry of Norway, a database that contains information from all pregnancies in Norway that were viable at 16 weeks, from 1967 onwards.[3] The Norwegian Renal Registry provided kidney transplant records of all kidney transplantations performed in Norway between 1967 and 2002. These two data sources were linked and 326 donors were identified, with 620 pregnancies registered before donation and 106 pregnancies registered after donation. Pregnancies before donation served as controls as did 21,511 pregnancies randomly sampled from the Medical Birth Registry.

Chronic hypertension rates, gestational hypertension rates, pre-eclampsia rates, birthweights, gestational ages and infant survival were compared between groups. Comparisons were performed using the Fisher's exact test as well as the generalized linear mixed model. The latter statistical method took into account the number of pregnancies as well as maternal age, birth order and year of birth and was only used to compare the two donor groups. No differences were found between the groups with respect to chronic hypertension, gestational hypertension, birthweight, gestational age, and infant mortality. Although the Fisher's exact test did not detect a difference in the incidence of pre-eclampsia between the three groups, the generalized linear mixed model demonstrated a significantly higher incidence of pre-eclampsia in post-donation pregnancies than in pre-donation pregnancies (5.7% versus 2.6%; P = 0.026) ( Table 1 ).

In a second article in the same issue of the American Journal of Transplantation, Ibrahim and colleagues[4] from the University of Minnesota reported results of two surveys that focused on pregnancy in women who had donated kidneys. The authors sent surveys to 1,769 female donors who had donated kidneys at their center between November 1963 and 21 December 2007. The first survey focused on the outcome of the baby and the second included questions about the mother. Of 1,589 responders, 1,085 women reported at least one pregnancy: 846 donors had pre-donation pregnancies only (2,519 pregnancies in total), 141 donors had post-donation pregnancies only (317 pregnancies in total), and 98 donors had both pre-donation and post-donation pregnancies (377 pregnancies in total). According to donor responses, post-donation pregnancies were statistically less likely to have gone to term (73.7% versus 84.6%; P = 0.0004) and more likely to have resulted in fetal loss compared with pre-donation pregnancies (19.2% versus 11.3%, P <0.0001). In addition, post-donation pregnancies were more often associated with development of gestational diabetes (2.7% versus 0.7%, P = 0.0001), gestational hypertension (5.7% versus 0.6%, P <0.0001), pre-eclampsia or toxemia (5.5% versus 0.8%, P <0.0001), and proteinuria (4.3% versus 1.1%, P <0.0001).

Taken together, the results of these two studies raise the issue that although maternal outcomes for pregnant kidney donors seem to be comparable to those in the non-donating general population, donating a kidney might place a woman at higher risk for maternal complications than if she had not donated. In other words, is a woman's personal risk for maternal complications of pregnancy lower if she does not donate a kidney? Does kidney donation take an individual who is at relatively low risk of maternal complications and normalize it, but increase the risk for that individual donor? These are important questions to answer because complications such as pre-eclampsia can have negative consequences for both mother and child. Such consequences include an increased risk of remote cardiovascular and metabolic diseases in later life for the mother, and an increased risk of stroke for the child.[5,6] The transplant community cannot answer these questions. Why not? After all, Reisæter et al.'s study benefits from the availability of complete data sets for pregnancies and kidney transplantations in Norway, diagnoses made at the time of events (not retrospectively), and a large control group. The study reported by Ibrahim and colleagues boasts a large, single-center experience with long-term follow-up information including creatinine, proteinuria, and blood pressure data. So why can't we answer these questions?

The problem is that despite their strengths, neither study is definitive; each of the studies has serious limitations. In Reisæter et al.'s study, the findings are based on a retrospective analysis of registry diagnoses, not on original data. The findings therefore depend on the clinical acumen of the treating obstetricians and midwives. One could argue that the study demonstrates a difference in the incidence of the diagnosis rather than any true difference in the incidence of the disorder. In addition, the findings are based on a small number of episodes, and were only statistically significant in one analysis (see Table 1 ). Ibrahim et al.'s study should also be interpreted with caution. Their findings are based on survey results and surveys can be biased by who returns the survey and who does not. In total, 180 individuals did not respond and 333 could not be contacted. Of those who did respond, some patients described pregnancies that occurred more than 40 years before. Aside from the potential memory lapses with that long time interval, the respondents who had been pregnant ≥40 years earlier might not have been given as much information about any pregnancy complications they experienced as those who delivered more recently.

Another potential pitfall applicable to both studies is whether pre-eclampsia was accurately diagnosed. Pre-eclampsia is a clinical diagnosis, and clinical-pathological correlation studies published 30 years ago have shown that the clinical diagnosis can be erroneous in up to 15% of nulliparas and a far greater percentage of multiparas.[7] In addition, kidney donors, many of whom have mildly increased proteinuria, are undoubtedly scrutinized during pregnancy, which increases the chance that they are labeled as pre-eclamptic, whether or not they are. Until objective biomarkers are shown to be reliable indicators of pre-eclampsia, one cannot always be confident about the diagnosis.

Previously published studies on post-nephrectomy pregnancies do not help.[8–11] They yield limited data that corroborate the findings that pregnancy outcomes are no worse in kidney donors than they are in the general public. They did not address whether post-donation pregnancies differ from pre-donation pregnancies. Furthermore, as two of the four studies were from the University of Minnesota, the patients were likely captured by Ibrahim et al.'s study.[4,9,11]

Reisæter and Ibrahim's provocative observations leave us with questions, not answers. Obtaining the answers is crucial so that we can appropriately advise young women who want to donate kidneys and donors who would like to become pregnant.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.