Pediatric Nephrology

, Volume 27, Issue 8, pp 1213–1219

The influence of gender and sexual hormones on incidence and outcome of chronic kidney disease

Authors

    • Department of General PediatricsUniversity Children’s Hospital
  • Gero von Gersdorff
    • Renal Division, Department of MedicineUniversity Hospital Cologne
  • Markus J. Kemper
    • Department of Pediatric NephrologyUniversity Medical Center Hamburg-Eppendorf
  • Jun Oh
    • Department of Pediatric NephrologyUniversity Medical Center Hamburg-Eppendorf
Review

DOI: 10.1007/s00467-011-1963-1

Cite this article as:
Kummer, S., von Gersdorff, G., Kemper, M.J. et al. Pediatr Nephrol (2012) 27: 1213. doi:10.1007/s00467-011-1963-1

Abstract

It has long been known that the female sex is associated with a better clinical outcome in chronic renal diseases. Although many experimental, clinical, and epidemiological studies in adults have attempted to explain the difference in disease progression between females and males, a definitive understanding of the underlying mechanisms is still lacking. Hormone-modulating therapies are being increasingly used for various indications (such as post-menopausal hormone replacement, estrogen- or androgen-receptor antagonists for cancer therapy). Therefore, a deeper knowledge of the interaction between sexual hormones and progression of kidney disease is important, as hormone-modulating therapy for non-renal indication may influence both kidney structure and function. In addition, specific modulation of the sexual hormone system, such as the use of selective estrogen receptor modulators, may represent a therapeutic option for patients with renal diseases. Although conclusive data on this topic in the pediatric population are still lacking, the aim of this review is to familiarize pediatric nephrologists with gender-specific differences in renal physiology, pathophysiology, and the progression of kidney diseases. Experimental models that analyze the effects of sexual hormones on renal structure and function are discussed. It is hoped that this review will stimulate researchers to focus on pediatric studies that will provide a deeper insight into the interaction of gender hormones and the kidney both before and during puberty.

Keywords

SexDisease progressionRisk factorFSGSRenal disease

Epidemiology: gender-dependent incidence and progression of chronic renal diseases

In patients with chronic renal diseases, the prevalence of end-stage renal disease (ESRD) is higher in males than in females (Fig. 1). These gender differences show remarkable variations with age: when the male-to-female ratio of ESRD is plotted as a function of age (Fig. 2), there are two peaks. The first peak appears during infancy and early childhood and most likely represents renal failure due to congenital urologic anomalies, which are predominantly seen in boys. A second peak appears between ages 40–50 years, when the activity of gender hormones in women begins to decline. These gender-linked differences disappear after the beginning of the menopausal years.
https://static-content.springer.com/image/art%3A10.1007%2Fs00467-011-1963-1/MediaObjects/467_2011_1963_Fig1_HTML.gif
Fig. 1

Prevalence of end-stage renal disease in the USA depending on age and gender (data are taken from the U.S. Renal Data System 2000 Annual Data Report. The National Institutes of Health, NIDDK, Bethesda, MD)

https://static-content.springer.com/image/art%3A10.1007%2Fs00467-011-1963-1/MediaObjects/467_2011_1963_Fig2_HTML.gif
Fig. 2

Male-to-female ratio of end-stage renal disease as a function of age. The first peak during early infancy is caused by a male predominance in congenital anomalies of the kidney and urinary tract (CAKUT), representing the most frequent cause for end-stage disease in infancy. The second peak is located between menarche and menopause, when hormonal influences are the most pronounced (data taken from the U.S. Renal Data System 2000 Annual Data Report. The National Institutes of Health, NIDDK, Bethesda, MD)

Thus, two different mechanisms can be distinguished, which may cause the gender differences in renal diseases:
  1. a)

    Gender-specific incidence of congenital anomalies of the kidney and urinary tract (CAKUT). These differences are presumably due to genetic factors. Hormonal influences are of minor importance since maternal sex hormones cross the placental barrier easily, thus overriding gender differences in fetal hormone production. Therefore, this group of patients is of minor relevance for the subject of this review.

     
  2. b)

    Influences of gender hormones on renal structure and function. These may either directly affect renal structures (e.g., via gender hormone receptors on renal cells) or indirectly via secondary influence factors (e.g. cardiovascular system).

     

A multitude of epidemiologic studies have analyzed the interplay of gender and chronic renal diseases. Berg et al. demonstrated that the physiological decline of renal function during aging is significantly slower in healthy females than in males [loss of glomerular filtration rate (GFR) of 1.4 ml/min/1.73 m2 per decade in healthy female kidney donors compared to 8.7 ml/min/1.73 m2 per decade in males, as measured by inulin clearance] [1]. These differences were detectable through the whole age spectrum (21.5–67.1 years, median 38.5 years).

The largest meta-analysis investigating gender differences in chronic renal diseases compared data collected on 11,345 patients participating in a total of 68 studies [2]. Female patients with polycystic kidney diseases, membranous glomerulonephritis (MGN), and nephropathies of unknown etiology had a favorable renal outcome compared to their male counterparts. The clinical outcomes were ESRD requiring renal replacement therapy, age of initiation of renal replacement therapy, and changes in measured GFR. The influence of menopause was not addressed in this study due to the heterogeneous patient population.

Cattran and colleagues further showed a significant benefit of female gender in 395 patients with MGN and 370 patients with focal-segmental glomerulosclerosis (FSGS) [3]. They measured the decrease in creatinine-based clearance and the overall survival of renal function. In particular, the negative impact of proteinuria on the rate of disease progression was clearly lower in women, and this interaction was independent of blood pressure.

Renal involvement in several systemic diseases also shows distinct gender-dependent features. For example, renal involvement occurs more frequently and more severely in males with systemic lupus erythematodes (SLE), although SLE incidence is considerably higher in females [4]. Data on renal involvement in diabetes mellitus (DM) type 2 are conflicting, likely due to the lack of multivariate analyses of blood pressure and pre-/postmenopausal age influencing gender-dependent factors. Therefore, consistent evidence is still missing on potential gender differences in the renal involvement of DM type 2 patients.

A large meta-analysis of 27,805 adult patients with type I DM showed an association of male gender with an increased risk for an early development of diabetic nephropathy [5]. Renal involvement was further dependent on pre- or postpubertal onset of diabetes in children. Prepubertal diagnosis of diabetes leads to significantly prolonged latency until the first manifestation of nephropathy. However, gender influences were not studied in this trial [6].

Epidemiologic studies published to date which have examined gender differences in renal diseases did not analyze different racial groups separately. In turn, studies demonstrating an increased risk of renal disease for several ethnic groups, such as African Americans, Hispanics, Native Americans, or Asians, have not specifically addressed gender differences in these populations [7]. Therefore, it remains unclear whether gender differences in renal diseases are also dependent on racial background.

Interestingly, the female advantage largely disappears after the initiation of dialysis therapy, with women having as poor survival as men despite the continuing worse risk profile of men also during dialysis therapy. The reasons for this observation have not as yet been fully elucidated, but may be due to—among others—a more severe impact of several risk factors on the overall prognosis of women despite a less frequent occurrence (see [8] for an overview of this topic).

Gender differences associated with renal transplantation

It has long been noted that the outcome and survival of renal grafts is much better in female than in male recipients [9], suggesting that the influence of the female recipient’s ‘environment’ is protective of the transplanted graft.

In contrast, renal grafts from female donors have been reported to have an inferior survival rate compared to those from male donors [10]. Interestingly, in this study, the effect was observed to be even more pronounced for younger donors (16–45 years) than for older donors (>45 years). As an explanation for this finding, the rapid loss of high premenopausal estrogen levels of the female donor organism on the removed kidney may compromise graft function and survival after transplantation into a male recipient. Another explanation for the inferior survival of grafts from female donors is provided by the concept of ‘nephron underdosing’: fewer and smaller nephrons may compromise organ function and survival by overloading the capacity of individual nephrons. In contrast to early anatomic studies suggesting larger kidney weight in men, more recent data do not show any significant differences in kidney weight when corrected for body weight [11]. Furthermore, the number of glomeruli in females seems to be equal to that in males [12]. Thus, ‘nephron underdosing’ seems to be of minor importance in terms of gender differences in renal transplantation. This conclusion is also supported by a study showing that other transplanted organs (e.g. heart) show similar drawbacks when taken from female donors [10]. The same study demonstrated that the frequency of rejection treatments 1 year after transplantation was higher when the kidneys were taken from female donors. Consistent with these findings, Vereerstraten and coworkers saw a higher incidence of acute rejection episodes in female kidneys [13]. Both findings point to an immunological effect, for example, increased immunogenicity of female grafts by the higher expression of HLA antigens [14].

Gender differences in pediatric renal disease

In children, the prevalence of glomerular diseases, such as minimal-change glomerulonephritis (MCGN), is slightly higher in boys than in girls [15]. Furthermore, girls with steroid-sensitive nephrotic syndrome show a trend towards less frequent post-pubertal recurrences than males (43 vs. 83%) [16]. However, the general frequency of MCGN relapses reported in this study was higher than that found in other studies in which relapses were reported in only 30% of patients after the initiation of puberty [17].

Kyrieleis et al. demonstrated that therapy with cyclophosphamide of steroid-dependent and frequently relapsing MCGN in prepubertal girls achieved definite remission more often compared with boys, although this trend did not reach statistical significance [18]. Examination of the same patients again after puberty revealed that, interestingly, of the 29% patients who developed a relapse, nine were boys and only three were girls [19]. These data suggest a similar trend towards a clinical advantage of female gender in pediatric glomerular diseases.

A retrospective analysis of 4,166 children with chronic kidney disease (CKD) of different etiologies identified an association between CKD progression (among others) and male gender [20]. However, this finding did not show statistical significance in the multivariate analysis, possibly due to the heterogeneous spectrum of disease etiologies. Surprisingly, to the best of our knowledge, there have not been any subsequent studies specifically focusing on gender differences in pediatric patients with chronic glomerular diseases.

Taken together, the available epidemiological data show a significant benefit of female gender for the prognosis of renal diseases. In the following part of this review, different experimental models explaining these gender differences will be discussed.

Influences of hormones on experimental models of renal function and disease

Various animal models of renal disease (hypertensive, age-related, polycystic, and renal mass-reduction) have shown that the progression of renal injury is faster in male animals than in their female littermates: as early as 1975 Elma and colleagues were able to show that male rats spontaneously develop proteinuria and glomerulosclerosis during aging, whereas female animals were remarkably resistant against these changes [21]. In another experiment, disease progression in male rats could be slowed by estrogen substitution or orchiectomy [22], suggesting a protective influence of estrogens and adverse effects of androgens. Similarly, experimental castration of 5/6 nephrectomized rats protected male animals against proteinuria, tubulointerstitial damage, and renal accumulation of fibronectin, whereas female animals did not show any protective effect after removal of estrogen influence by ovariectomy [23].

In several other studies reviewed by Neugarten et al., estrogen application reduced proteinuria and glomerular fibrosis after experimental renal damage in different animal models [24]. In addition, the expression of glomerular damage markers, such as desmin, in spontaneously hypertensive rats and rats after puromycin treatment could be attenuated by estrogen treatment [25, 26].

Thus, female sexual hormones generally exhibit protective effects, while androgens often accelerate disease progress in animal models of renal disease.

Possible mechanisms explaining gender differences in the progression of renal disease

Lifestyle factors and nutritional profile

Factors such as physical and psychological stress, smoking, and especially nutritional factors seem to have a significant impact on the progression of renal diseases [27]. In particular, an excessive consumption of protein, sodium, and phosphate has been shown to be harmful. Dyslipidemia with high low-density lipoprotein (LDL), low high-density lipoprotein (HDL), and hypertriglyceridemia [28], as well as obesity in general [29], further facilitate disease progression. Compared to men, women have a more favorable nutritional profile, with higher intakes of fruit, vegetables, and dietary fibers and lower intakes of fat and protein [30], and a beneficial serum lipid profile [31, 32]. This may provide significant benefit for the prognosis of renal diseases.

Uric acid has recently turned out to be another risk factor for accelerated kidney disease progression—probably mediated by increasing the incidence of arterial hypertension [33]. As uric acid levels are known to be higher in males than in females [34], this may further contribute to a faster progression of CKD in men

Blood pressure and intrarenal hemodynamics

Arterial hypertension is significantly associated with a worse prognosis in CKD [35]. Various studies have shown that women have a lower mean arterial blood pressure [36] and a lower incidence of arterial hypertension than age- and weight-matched men [37]. Interestingly, these differences in blood pressure level between men and women disappear after the onset of the menopause [37]. Therefore, hormonal influences on blood pressure level may substantially affect the outcome of renal diseases [15].

When adjusted for body weight and body surface, GFR is identical for both genders in humans and rats [38]. However, intrarenal hemodynamics show gender-specific reactions on numerous variables. The administration of angiotensin II induced an increase of GFR in men, but not in women [39]. This difference was explained by the authors in terms of an enhanced glomerular capillary pressure as a response to angiotensin II in men, which was reduced or even absent in women. One possible molecular mechanism might be that estrogens and androgens both modulate the expression of angiotensinogen, angiotensin II, angiotensin II receptor, renin, and angiotensinogen converting enzyme (ACE) in multiple different cell types, as shown in cell cultures and animal models [40, 41]. Correspondingly, human studies show an impact of estrogens and oral contraceptives on the regulation of the renin–angiontensin–aldosterone system (RAAS) [42, 43]. In contrast, endogenous increases in estrogen levels (e.g., during pregnancy) are associated with a decrease in arterial blood pressure despite the activation of RAAS [44]. Taken together, these data show gender-specific influences on both the regulation of systemic blood pressure and intrarenal hemodynamics.

Other influences of the hormonal milieu

Premenopausal application of oral contraceptives as well as postmenopausal hormone replacement therapy (HRT) increases microalbuminuria, a marker of early glomerular damage [45]. In 5,845 postmenopausal women older than 65 years, HRT was also associated with a significantly faster decline of renal function compared to women without any HRT [46]. In a subgroup analysis, this effect was limited to the estrogen monotherapy (additional decline of 1.21 ml/min/1.73 m2 compared to non-substituted individuals). Progesterone alone or in combination with estrogens did not induce an accelerated decline of kidney function. Moreover, the effect was detectable only for oral administration, not for vaginal application, and was independent of any other co-medications (such as non-steroidal antiphlogistics, ACE inhibitors, angiotensin receptor blockers, glucocorticoids, cyclosporine, etc.). Unfortunately, it was not documented whether the therapy was initiated before or during menopause.

In contrast, other studies have detected a reduced risk for albuminuria in postmenopausal women on HRT without further differentiation for the subtype of medication [47]. In postmenopausal women with DM, HRT also did not show any effect on albuminuria compared to women without HRT [48], whereas in premenopausal patients with DM, HRT was clearly associated with albuminuria [43]. Experimental data reveal similar discrepancies as epidemiological studies: in animal models, only continuous application of estrogen throughout a 9-month period was able to reduce experimentally induced glomerulosclerosis, while intermittent application every 3 months had no effect [49].

One possible explanation for this inconsistency in experimental and clinical data is the high degree of heterogeneity in the therapeutic approach to HRT, with a variety of different hormone subtypes, application forms, and timing of therapy initiation.

Cellular pathways of sexual hormones and selective estrogen receptor modulators (SERM)

Numerous studies have shown that gender hormones, in particular estrogens, have pleiotropic effects on different cell types through specific cellular receptors. Extensive investigations in this field have been performed, especially for neuronal cells and skeletal muscle cells. For example, the transcription of genes encoding mitochondrial proteins is modified by estrogen receptor (ER) activation, representing a strong link between ER signaling and intact mitochondrial function [50]. One specific example is the estrogen-induced expression of nuclear respiratory factor-1 (NRF-1), a transcriptional factor that regulates the expression of nuclear-encoded mitochondrial genes, such as mitochondrial transcription factor A (TFAM) [51]. This pathway leads to a coordinated increase in the expression of the mitochondrial genome and thereby the stimulation of mitochondrial biogenesis and function. This cellular pathway could explain why skeletal muscle cells are stabilized against oxidative damage by estrogens [52] and why degenerative diseases of muscle cells are associated with a decrease of estrogen levels [53]. The stabilization of mitochondrial function has also been shown in ocular lens epithelium, where estrogen was found to reduce destabilization of mitochondrial membrane potential by oxidative stress [54]. In addition to these genomic ER effects, ‘non-classical’ actions via membrane-associated ER are also able to activate cytoplasmic signaling pathways in a G-protein-coupled manner. For example, activation of mitogen-activated protein kinase (MAPK) or phosphatidylinositol 3-OH kinase (PI3K) pathways appears within minutes after estrogen administration [55, 56]. The MAPK pathway involves a group of intracellular signaling molecules comprising three major families with different downstream effects: extracellular signal-regulated kinase (ERK1/2), p38 MAPK, and c-Jun N-terminal kinases. These pathways involve a group of intracellular signaling molecules with impacts on cell proliferation, survival, apoptosis, differentiation, motility, among others. One example of protective estrogen action is the reduction of apoptotic cell death of neurons following ischemia by an estrogen-induced activation of the Raf–MEK–ERK–p90RSK cascade [56]. Interestingly, similar pathways have recently been shown to be active in podocytes: podocytes isolated, immortalized, and subsequently analyzed in vitro were observed to show estrogen-induced modulation of TGFβ- and MAPK-pathways [57]. Doublier and colleagues demonstrated that this mechanism also leads to a protection of podocytes against transforming growth factor beta 1 (TGFβ1)- or tumor necrosis factor alpha (TNFα)-induced apoptosis in vitro and in vivo, while testosterone induces podocyte apoptosis by signaling via androgen receptors [58]. Unpublished data from our group show a corresponding estrogen-induced protection of podocytes in the model of puromycin-induced apoptosis. Studies on metastatic cancer have implicated a novel set of signaling molecules as potential mediators of estrogens, namely, focal adhesion kinase (FAK) and paxillin. Both regulate cell adhesion and the interaction of cytoskeletal components with the extracellular matrix. In podocytes, these proteins have been shown to be critical for the maintenance of cellular structures: The interactions of podocytes with the glomerular basement membrane result in increased stability against proteinuria in FAK knockout mice [59]. Recent data also show a complex interaction of estrogen signaling with the activity of FAK, thereby modulating cell motility via control of the focal adhesion complex turnover in endothelial cells, breast cancer cells, and endometrial cells. Further protective influences of estrogens are an inhibited production of matrix proteins, such as collagen type I and IV, and stimulation of matrix-degrading enzymes, thereby reducing glomeruloslerosis [60, 61]. In TGF-β-transgenic mice, which have an increased susceptibility to extensive glomerulosclerosis, estrogen substitution significantly reduces sclerosis [62]. Protective effects have also been shown for SERM such as tamoxifen and raloxifene. These substances bind to estrogen receptors with tissue-specific effects and different binding specificities for ERα or ERβ: for example, raloxifene has a fourfold higher affinity for ERα than for ERβ and does not induce endometrial hyperplasia. In vitro, SERM are able to reduce collagen synthesis, similar to estrogen [60]. In vivo, renal prognosis is improved by SERM: raloxifene has very recently been found to be significantly renoprotective in postmenopausal women that were treated with this medication for reduction of bone fractures. In this study, raloxifene-treated women developed a significantly slower annual increase in serum creatinine and a lower decrease in GFR [63]. Several estrogenic effects can also be induced by metabolites of estradiol, such as catecholestradioles (e.g. 2-hydroxyestradiol and 4-hydroxyestradiol) [25]. Interestingly, these effects are independent of ER activation, indicating that additional independent mechanisms must be involved.

Summary/conclusion

In terms of the incidence and progression of chronic renal diseases, epidemiological data demonstrate that the female gender endows patients with significant benefits compared to the male gender.

In this review, we have presented several factors that have been evaluated extensively (lifestyle, blood pressure, and hemodynamics). More recent data also show direct actions of sexual hormones in the kidney at a cellular level (e.g., regulation of apoptosis, cytoskeletal dynamics, mitochondrial function, nitric oxide signaling). Taken together, a complex interaction of estrogen effects in combination with partly adverse androgen effects seems likely, with the balance of both contributing decisively to the overall renal prognosis.

However, the clinically most relevant question still remains to be answered: will this knowledge find any therapeutic application in the near future? A first attempt was published in 1955 [64]. Three patients with severe nephrotic syndrome had complete resolution of symptoms when treated with an initial course of estrogen application, but proteinuria recurred after the discontinuation of the medication. Two women had another course of therapy and achieved remission again, while the male patient refused to take these hormones again and died in nephrotic crisis.

Although the substitution of 17β-estradiol does not provide a reasonable therapeutic option due to significant side-effects (such as possible tumor induction and feminization), SERM or non-receptor mediated estrogen metabolites may offer promising options for a targeted therapy of renal diseases. Whether such an approach might also be extended to the pediatric population with a reasonable balanced risk profile regarding specific problems of prepubertal/pubertal patients remains to be determined. However, increasing awareness of gender aspects may encourage pediatric nephrologists to consider them as risk factors and prognostic indicators for renal diseases in the pediatric population.

Copyright information

© IPNA 2011