American Journal of Kidney Diseases
Volume 49, Issue 6 , Pages 729-731, June 2007

Pregnancy and CKD: Any Progress?

  • Marshall D. Lindheimer, MD

      Affiliations

    • University of Chicago, Chicago, Illinois
    • Corresponding Author InformationAddress correspondence to Marshall D. Lindheimer, MD, Departments of Obstetrics & Gynecology and Medicine, the University of Chicago, Chicago, IL 60637.
    • ,
  • John M. Davison, MD

      Affiliations

    • Newcastle University, Newcastle upon Tyne, England

Article Outline

 

Related Article, p. 753

A report by Imbasciati et al1 in this issue surveys renal prognosis plus pregnancy outcome in women with moderately severe chronic renal disease. The data were collected over a 27-year period, and the findings should make us reflect on what progress, if any, we have made during these years in counseling women with underlying kidney disease who wish to conceive or are already pregnant. But first a little history is in order.

An anonymous editorial published in Lancet in 1975 noted “Children of women with renal disease used to be born dangerously or not at all — not at all if their doctors had their way.”2 The author(s)’ thesis: most caregivers consider any degree of renal disease incompatible with a safe and/or successful pregnancy and recommend initial trimester terminations (and presumably sterilization or adherence to strict contraceptive practices thereafter). This approach, the author(s) continued, must be seriously reconsidered, as most of these gestations succeed if renal function is only mildly compromised. These bold views, based on limited reports, including many where the nature of the underlying parenchymal disorder was not identified, have proven largely true. Current recommendations are that if functional loss is below 50%, stated in most texts as serum creatinine concentration equal to or less than 1.4 mg/dL (124 μmol/L), the gestation should succeed and should not adversely influence the mother’s renal prognosis, though complications such as preeclampsia and premature birth are increased.3, 4 However, and as discussed further below, presence of poorly controlled hypertension, as well as greater degrees of dysfunction are more ominous, at least for maternal outcome.3, 4, 5, 6, 7

The strongest data supporting these forecasts derive from series of clinical-pathological correlation type studies, often modeled on the one8 described in 1980 in which the underlying disorders were identified by a renal biopsy, with chart evidence of the pregnancy course and outcome, and often involving years of maternal follow-up. One of us (J.M.D.) has periodically reviewed this literature, the cumulative total exceeding 2,300 women with pregnancies that were permitted to proceed beyond the first trimester (see table 44-2 in reference 3). In essence, advice is based on the degree of renal function and not the underlying parenchymal lesion, though certain disorders (mainly systemic in nature) have poorer prognoses. The latter diseases include periarteritis nodosa, scleroderma, and recently active lupus nephritis; disagreements exist as to whether reflux nephropathy, mesangioproliferative glomerulonephritis, and IgA nephropathy also have more guarded prognoses. Of interest, too, is a recent survey of 120,000 birth records in which the authors identified 82 women with chronic renal disease.9 Using logistic regression and receiver operating characteristic analysis, they confirmed that creatinine levels of 1.4 mg/dL or less (≤124 μmol/L) were associated with live outcomes similar to the pregnant population as a whole, and went on to demonstrate a cut-off value of 1.1 mg/dL (97 μmol/L), above which the premature birth rate rose significantly. Of importance is that despite decades of recording these experiences, the advice we continue to give is based almost entirely on retrospective data.

The 1975 editorial discussed other issues relating to gestation in women with renal disease. The author(s) claimed that proteinuria rose to nephrotic range levels during most of these gestations, stating this due to the “increasing glomerular filtration rather than worsening renal disease.” This claim too has been borne out, though the number of gravidas manifesting increments in protein excretion to nephrotic levels is closer to one quarter,8 and more likely caused by glomerular membrane changes physiological for pregnancy that result in greater protein filtration than due to increased GFR.10

The most pessimistic parts of the 1975 Lancet editorial related to prognosis in women with moderate renal dysfunction or worse prior to conception, especially when hypertension is present (which is often the case). The author(s) noted, though, that “nature takes a helping hand by blunting fertility as renal function fails, thus the worse risks are rarely run.” (As prescient as the editorial was, the author(s) were probably not prepared to speculate on the emerging era of fertility drugs, assisted reproduction techniques, and improvements in renal replacement therapy.) But was this pessimism warranted? Series of women with moderate dysfunction prior to conception (serum creatinine >1.4 mg/dL [>124 μmol/L]) seem harder to find; the total numbers of patients in our own periodic reviews consist of fewer than 200 cases, again almost completely retrospective.4 Still these limited series have led to the current views that it is the mother’s disease that suffers, but fetal survival is quite good (≥90% over 30 years and in excess of 95% more recently). However, approximately one third of these mothers experience functional deterioration during pregnancy or in the immediate puerperium, ∼10% of which may not revert, and occasionally functional deterioration accelerates after the pregnancy.3, 4, 5, 6, 7 The statistics for both mother and fetus, of course, get worse as creatinine levels rise, with those for women treated by dialysis the most dismal (and yet despite the relative infertility, such women can conceive and therefore require contraceptive counseling). Of note, though, the live birth successes have been increasing, even in dialysis units.11, 12

The report by Imbasciati et al contains several new nuances, one being its primarily prospective observational design, and a second the authors’ use of the chronic kidney disease (CKD) stages to describe the severity of renal dysfunction and to classify outcomes in their patients. However, this latter nuance may be confusing, especially to our obstetric colleagues, for two reasons: the literature to date has classified renal dysfunction in pregnancy solely by absolute creatinine levels, (≤1.4 mg/dL [≤124 μmol/L] as mild; 1.4-2.5 mg/dL [124-221 μmol/L], moderate; and above these values, severe), and the formulas used here to calculate GFR before and after gestation are not reliable for monitoring renal function during pregnancy.13

Imbasciati and colleagues evaluated the pregnancies of 49 women with chronic disease whose prepregnancy GFR was less than 60 mL/min/1.73 m2 (<1.00 mL/s/1.73 m2; corresponding to a serum creatinine concentration equal to or greater than 1.4 mg/dL [≥124 μmol/L] in a 20 to 30-year-old woman) and for whom sufficient data had been obtained to calculate the rate of functional loss prior to conception. Function was also evaluated during gestation, in the immediate puerperium, and for an average of 39 months postpartum. Despite the study’s inadequate power and a failure to reveal the delivery pools surveyed or their recruitment success, the authors’ main observation is encouraging: while the group as a whole lost function during gestation, the rate of functional loss was not affected by the pregnancy, being similar prepartum and postpartum. Also, 95% of the fetuses survived beyond the neonatal period (albeit many were prematurely born and/or growth restricted). The best news, however, was focused on women whose prepregnancy function was between 60 and 40 mL/min/1.73 m2 (1.00 and 0.67 mL/s/1.73 m2; corresponding to serum creatinine concentrations between 1.4 and 1.6-1.7 mg/dL [124 and 141-150 μmol/L] or proteinuria <1 g/d). On the other hand, women with function below 40 mL/min/1.73 m2 (0.67 mL/s/1.73 m2) and proteinuria exceeding 1 g/d had poorer outcomes, the combination resulting in worse outcomes than either factor alone. Although these latter subjects appeared to develop renal failure faster than the other groups, it was not possible to determine whether the gestation was a causal factor.

These authors “labored” 23 years to meticulously describe 49 pregnancies, the 42 prospective observations being a tour de force for research on the subject. One must realize such patience is necessary as the incidence of chronic renal disease in pregnancy is quite low, ranging between 0.002% and 0.01% in various surveys,9, 14, 15, 16 and the vast majority of those conceiving having but mild dysfunction. However, has knowledge advanced much from that anonymous Lancet voice, and the series of clinical-pathological studies that have followed? Unfortunately, in our opinion, only minimally.

The cut-offs for optimistic advice when function is mildly impaired are still about the same, with little new data forthcoming to change them. The cautions to be given to those with moderate dysfunction, including warnings relating to the possibility of accelerating functional loss, the ominous features of concomitant poorly controlled hypertension, and the increased propensity for all patients with chronic renal disease to develop superimposed preeclampsia (not always easy to differentiate from an exacerbation of their primary renal disorder17) are also fairly well described. Even the focus of proteinuria as a specific risk factor was, as the authors acknowledge, commented on by others. In the latter respect, suggestions in the literature that women whose functional loss may be due to interstitial disease may do better than those with chronic glomerulonephritis may be the confounding variable here as such patients usually have less proteinuria, and this too is suggested in the authors’ data. Again, all this is in literature preceding the current millennium. However, one thing still stands out here: the authors’ contribution in this issue is the result of “laboring” over two decades; their persistence has given us prospective data, and we are aware of few who have done this successfully before.

Finally, one hopes that future collaborative studies can better these statistics, perhaps utilizing the potential of several established and well functioning obstetrical networks. One, the Maternal Fetal Medicine Trials Network of the National Institute of Child Health and Human Development (NICHD), gathers data on approximately 75,000 live births per year. Though developed primarily to perform clinical trials, centers such as these with well trained support staff would be good sources for nephrologists to focus their prospective observational studies with the power to perhaps arrive at better or more secure advice, and of course, more quickly. But this is an editorial, not a sermon!

Back to Article Outline

References 

  1. Imbasciati E, Gregorini G, Cabiddu G, et al. Pregnancy in CKD stages 3 to 5: fetal and maternal outcomes. Am J Kidney Dis. 2007;49:753–762
  2. Pregnancy and renal disease. Anonymous Lancet. 1975;2:801–802
  3. Davison JM, Lindheimer MD. Renal Disorders. In:  Creasy RK,  Resnik R,  Iams J editor. Maternal-Fetal Medicine: Principles and Practice. (ed 5).. Philadelphia, PA: Saunders; 2004;p. 901–924
  4. Davison JM, Baylis C. Pregnancy in patients with underlying renal disease. In:  Davison AM,  Cameron JS,  Grunfeld J-P, et al. editor. Oxford Textbook of Clinical Nephrology. Oxford, England: Oxford University Press; 2005;p. 2243–2260
  5. Hou SH, Grossman SD, Madias NE. Pregnancy in women with renal disease and moderate renal insufficiency. Am J Med. 1985;78:185–194
  6. Imbasciati E, Paidi G, Capetta P. Pregnancy in women with chronic renal failure. Am J Nephrol. 1986;6:193–198
  7. Jones DC, Hayslett JP. Outcome of pregnancy in women with moderate or severe renal insufficiency. N Eng J Med. 1996;335:226–232
  8. Katz AI, Davison JM, Hayslett JP, Singson E, Lindheimer MD. Pregnancy in women with kidney disease. Kidney Int. 1980;18:192–206
  9. Modena A, Hoffman M, Tolosa JE. Chronic renal disease in pregnancy: A modern approach to predicting outcome. Am J Obstet Gynecol. 2006;193:S86
  10. Moran P, Lindheimer MD, Davison JM. The renal response to preeclampsia. Sem Nephrol. 2004;24:586–593
  11. Holley JL, Reddy SS. Pregnancy in dialysis patients: A review of outcomes, complications, and management. Semin Dial. 2003;316:384–388
  12. Goody AJ, Umans JG. Pregnancy and dialysis. In:  Horl WH,  Koch KM,  Lindsay RM, et al. editor. Replacement of renal function by dialysis (ed 5). London, UK: Kluwer Academic; 2004;p. 1249–1257
  13. Smith MC, Moran P, Ward MK, Davison JM. The MDRD2 formula should not be used for prediction GFR in healthy and preeclamptic pregnancies. J Obstet Gynecol. 2006;26:S47–S48
  14. Cunningham FG, Cox SM, Harstad TW, Mason RA, Pritchard JA. Chronic renal disease and pregnancy outcome. Am J Obstet Gynecol. 1990;163:453–459
  15. Trevisan G, Ramos JG, Martins-Costa S, Barros EJ. Pregnancy in patients with chronic renal disease at Hospital de Clinicas of Posto Alegre (Brazil). Renal Fail. 2004;26:29–34
  16. Fischer MJ, Lehnerz SD, Herbert JR, Parikh CR. Kidney disease is an independent risk factor for adverse fetal and maternal outcomes in pregnancy. Am J Kid Dis. 2004;43:415–423
  17. Fisher KA, Luger A, Spargo BH, Lindheimer MD. Hypertension in pregnancy: Clinical pathological correlations and remote prognosis. Medicine. 1981;60:267–276

 Support: None. Potential conflicts of interest: None.

PII: S0272-6386(07)00729-9

doi:10.1053/j.ajkd.2007.04.008

Refers to article:

  • Pregnancy in CKD Stages 3 to 5: Fetal and Maternal Outcomes

    Enrico Imbasciati, Gina Gregorini, Gianfranca Cabiddu, Linda Gammaro, Giancarlo Ambroso, Antonio Del Giudice, Pietro Ravani, Collaborative Group “Rene e Gravidanza” of the Società Italiana di Nefrologia
    American Journal of Kidney Diseases June 2007 (Vol. 49, Issue 6, Pages 753-762)

American Journal of Kidney Diseases
Volume 49, Issue 6 , Pages 729-731, June 2007

Article Tools