The Effect of Gender-Affirming Hormone Therapy on Measures of Kidney Function

Visual Abstract

Abstract

Background and objectives

Gender-affirming hormone therapy modifies body composition and lean muscle mass in transgender persons. We sought to characterize the change in serum creatinine, other kidney function biomarkers, and GFR in transgender persons initiating masculinizing and feminizing gender-affirming hormone therapy.

Design, setting, participants, & measurements

We searched PubMed, EMBASE, the Cochrane Library, and ClinicalTrials.gov from inception to September 16, 2020 for randomized controlled trials, observational studies, and case series that evaluated the change in serum creatinine, other kidney function biomarkers, and GFR before and after the initiation of gender-affirming hormone therapy in adult transgender persons. Two reviewers independently screened and abstracted data, and disagreements were resolved by a third reviewer. A random effects meta-analysis was performed to determine the change in outcomes over follow-up of 3, 6, and 12 months.

Results

Of the 4758 eligible studies, 26 met the inclusion criteria, including nine studies that recruited 488 transgender men and 593 women in which data were meta-analyzed. There was heterogeneity in study design, populations, gender-affirming hormone therapy routes, and dosing. At 12 months after initiating gender-affirming hormone therapy, serum creatinine increased by 0.15 mg/dl (95% confidence interval, 0.00 to 0.29) in 370 transgender men and decreased by −0.05 mg/dl (95% confidence interval, −0.16 to 0.05) in 361 transgender women. No study reported the effect of gender-affirming hormone therapy on albuminuria, proteinuria, cystatin C, or measured GFR.

Conclusions

Gender-affirming hormone therapy increases serum creatinine in transgender men and does not affect serum creatinine in transgender women. The effect on gender-affirming hormone therapy on other kidney function biomarkers and measured GFR is unknown.

Clinical Trial registry name and registration number:

Change in Kidney Function Biomarkers in Transgender Persons on Gender Affirmation Hormone Therapy–A Systematic Review and Meta-Analysis, CRD42020214248

Introduction

It is estimated that 0.3%–0.6% of the adult population identifies as transgender, which amounts to approximately 150,000 transgender persons in Canada, 1 million in the United States, and 25 million worldwide (1–3). Gender-affirming hormone therapy (GAHT) shifts an individual’s sex hormone milieu to biochemically reflect their affirmed gender identity (4,5). For transgender men, GAHT typically includes injectable or transdermal testosterone therapy, and for transgender women, GAHT typically includes exogenous oral, sublingual, transdermal, or injectable estradiol and antiandrogen therapy (cyproterone acetate, GnRH agonists, mineralocorticoid receptor antagonists, 5α-reductase inhibitors, and bicalutamide) with or without progesterone. Nonbinary and gender-diverse persons may also be treated with either masculinizing or feminizing GAHT. GAHT is known to alter body composition, including lean muscle mass and body fat, in transgender persons (6,7).

Creatinine is a biomarker derived from skeletal muscle used to estimate GFR and is influenced by non-GFR determinants (i.e., age, sex/gender, and race) (8). The degree to which GAHT affects serum creatinine and other kidney function biomarkers (cystatin C, albuminuria, and proteinuria) given its effect on body composition is unclear (9). Similarly, whether GAHT affects GFR independent of changes in kidney function biomarkers related to changes in body composition is unknown. These issues are important because both sex/gender and kidney function biomarkers are incorporated into estimating GFR, which is a cornerstone in CKD detection, drug dosing, prognosis, and management.

We completed a systematic review and meta-analysis to characterize the change in serum creatinine, other kidney function biomarkers, and GFR in adult transgender persons initiating masculinizing and feminizing GAHT in order to inform the care of this patient population (10).

Materials and Methods

Data Sources and Searches

To identify studies, the population, intervention, comparison, outcomes, and timing criteria for the search strategy were used, and the protocol was registered at PROSPERO (11). In collaboration with a medical librarian (N.A.), original research articles were identified from the following databases: PubMed, EMBASE, the Cochrane Library, and ClinicalTrials.gov. Our search of these databases ranged from the date of their establishment until September 16, 2020. The search strategy was tailored to each database and used a combination of key terms, including “transgender,” “trans,” “gender,” “sex,” “creatinine,” “cystatin c,” “glomerular filtration rate,” “gender-affirmation,” “hormone therapy,” “estrogen,” “testosterone,” “anti-androgen,” “spironolactone,” “cyproterone acetate,” and “gonadotropin-releasing hormone agonist.” The search strategy is shown in Supplemental Table 1. The gray literature was not assessed.

Target studies included those that enrolled adult (aged 18+) transgender persons (men, women, nonbinary, or gender diverse) who had received treatment with GAHT of any type and had serum creatinine, creatinine clearance, BUN, cystatin C, eGFR, measured GFR, albuminuria, or proteinuria (spot or 24-hour values) reported at baseline (prior to the initiation of GAHT) and after 3, 6, and 12 months or longer on GAHT. We also included studies that compared outcomes between transgender persons either pre- or post-GAHT and matched cisgender controls cross-sectionally. Types of studies included randomized controlled trials and prospective, retrospective, or cross-sectional observational cohort studies, including case series but not case reports. There was no limit on study size, duration, or setting. Articles in languages other than English, French, Spanish, and German were excluded. All citations were uploaded in Covidence (12).

Study Selection

Two individuals (S.C. and E.K.) independently reviewed titles and abstracts of the studies identified by the search strategy, and those that did not match the inclusion criteria were excluded. For the remaining studies, full texts were retrieved and reviewed for inclusion. A third reviewer (D.C.) reviewed conflicts, and final decisions were made by consensus. References of included studies were also reviewed to identify relevant studies not captured by the search strategy.

Data Extraction

We created a data extraction form to capture the following information: study characteristics (first author, year of publication, country, setting, study design, number of participants, and duration), participant characteristics (age, sex/gender, race, comorbidities, GAHT including type, dose, and frequency during study), gender confirmation surgery, follow-up duration, and outcomes pre- and post-GAHT (serum creatinine; cystatin C; urine protein-creatinine ratio; urine albumin-creatinine ratio; 24-hour urine for creatinine and albuminuria/proteinuria; eGFR by the Modification of Diet in Renal Disease [13] equation, the Chronic Kidney Disease Epidemiology Collaboration [CKD-EPI] [8] equation, or other equations; measured GFR using any methodology; and eGFR bias [i.e., the difference between eGFR and measured GFR using cisgender and transgender in eGFR equations]). We contacted all corresponding authors of included studies for missing or additional outcomes in case they were not reported in the original manuscript. Creatinine in micromoles per liter was converted to milligrams per deciliter using a factor of 0.0113. If a study used multiple follow-up time points, values at baseline and those closest to 3 months, 6 months, 9 months, and 1 year were recorded. If a study reported outcomes over a time period (e.g., 3–6 and 6–18 months), only the earliest time point was used (i.e., 3 and 6 months, respectively). We did not analyze outcomes beyond 1 year given their limited reporting and variability in the timing of assessments. If the variance of an outcome of interest was not reported, it was derived from the interquartile range or standard error of the mean (SEM). If neither of these were reported, it was assumed to be equal to that of earlier time points if available or was the overall mean of the study. Two reviewers (S.C. and E.K.) independently extracted data; inconsistencies were corrected and resolved by consensus and in consultation with a third reviewer (D.C.). The National Institutes of Health Quality Assessment Tool for Before-After (Pre-Post) Studies with No Control Group was utilized to determine the quality of included studies in the meta-analysis only. This tool is composed of 12 questions used to assess the risk of types of bias, such as selection, reporting, or observer bias. The methodologic quality of included studies was assessed by two authors (S.C. and E.K.) separately. Each study was defined as poor, fair, or good.

Data Synthesis and Analyses

A random effects meta-analysis for longitudinal effect estimates with 95% confidence intervals (95% CIs) of GAHT on change in outcomes at 3, 6, and 12 months was performed using estimated within-study and between-study correlations using PROC MIXED with autoregressive within-study and between-study covariance matrices (14). Modeling were performed separately for transgender men on GAHT and transgender women on GAHT. All analyses were done in SAS version 9.4 (Cary, NC). In order to estimate heterogeneity as measured by τ and I², we also performed a random effects meta-analysis at 3, 6, and 12 months using the DerSimonian and Laird method in RevMan, acknowledging that this meta-analytic approach is less credible than the primary meta-analytic approach described above due to the presence of repeated correlated measures. We did not perform any sensitivity analyses.

Results

Search Results and Study Characteristics

From 4758 eligible studies, 26 met the inclusion criteria. The studies are summarized in Table 1 (18,19,24–47), Supplemental Table 2 (transgender men on GAHT), and Supplemental Table 3 (transgender women on GAHT). Included studies were published between 1998 and 2019 and included three randomized controlled trials, 12 prospective studies, five retrospective studies, and six cross-sectional studies. Studies were mostly from the United States and Europe, with heterogeneity across designs, populations, GAHT route, dose, and follow-up duration. No study included nonbinary or gender-diverse individuals on GAHT. Participants were typically younger without any evidence of CKD. Twenty-three studies reported serum creatinine. BUN was reported infrequently (seven studies), and no study reported cystatin C. eGFR and creatinine clearance (Cockcroft Gault) were reported in two and three studies, respectively; 24-hour urine creatinine was reported in two studies, and creatinine clearance calculated by a 24-hour urine collection was reported in one study. Urinalysis and urine creatinine were reported in one study each. Albuminuria, proteinuria, measured GFR, and eGFR performance were not reported in any study.

Table 1.

Characteristics of included studies (pre/post and cross-sectional)

Study	Author and Year	Country	Study Design	Population	Previous Gender-Affirming Hormone Therapy	Gender Confirmation Surgery	Follow-Up	Outcomes	Meta-Analysis
1	Bunck et al. (25) 2006	The Netherlands	Randomized controlled trial, AI versus placebo	30 TM	18–24 mo	Yes	12 wk	Creatinine, urinary creatinine	No
2	Fernandez et al. (26) 2016	United States	Retrospective	19 TM, 33 TW	N/A	No	3–6, 6–18 mo	Creatinine	Yes
3	Giltay et al. (27) 1998	The Netherlands	Prospective	17 TM, 17 TW	N/A	No	4 mo	Creatinine, 24-h urine creatinine	Yes
4	Giltay et al. (28) 2008	The Netherlands	Prospective	14 TM, 15 TW	N/A	No	4, 12 mo	Creatinine, 24-h urine creatinine	No (duplicate data)
5	Humble et al. (29) 2019	United States	Retrospective	150 TM, 152 TW	N/A	24% TM, 19% TW	6, 12 mo	BUN, creatinine, eGFR	Yes
6	Hiransuthikul et al. (30) 2019	Thailand	Prospective	20 TW	N/A	No	15 wk	Creatinine clearance	No
7	Jain et al. (31) 2019	United States	Retrospective	92 TW	N/A	No	3 mo, annually, mean 3.4±1.7 yr	BUN, creatinine	No
8	Kurahashi et al. (32) 2013	Japan	Prospective	160 TM	N/A	No	3, 6, 12 mo	Urinalysis, creatinine	No
9	Meriggiola et al. (33) 2008	Italy	Randomized, open-label, uncontrolled safety study	15 TM	71.4±30.5 mo	Yes	6, 18, 30, 42, 54 wk	BUN, creatinine	No
10	Nicholls et al. (34) 1973	United Kingdom	Prospective	22 TM	N/A	No	2 wk to 10 mo (mean 10 wk)	24-h urine for creatinine clearance	No
11	Shieh et al. (35) 2019	United States	Prospective	8 CisM, 8 TW	1–27 yr	No	7 d	Creatinine, eGFR, creatinine clearance	No
12	SoRelle et al. (36) 2019	United States	Retrospective	Baseline: 62 TM, 87 TW; post-GAHT: 89 TM, 133 TW	≥6 mo	Table 1	N/A	BUN, creatinine	Yes
13	van Kesteren et al. (37) 1998	The Netherlands	Prospective	19 TM, 20 TW	N/A; surgery 20.3±4.1 after GAHT	Yes	12 mo, 25.2±10.4 mo after surgery	Creatinine	Yes
14	van Velzen et al. (17) 2019	ENIGI, The Netherlands, Belgium, Norway, Italy	Prospective	188 TM, 242 TW	N/A	No	12 mo	Creatinine	No (duplicate data)
15	Vita et al. (38) 2018	Italy	Retrospective	11 TM, 21 TW	N/A	No	Mean 30 (25.9) mo	Creatinine	Yes
16	Vlot et al. (18) 2019	ENIGI, The Netherlands, Belgium, Norway, Italy	Prospective	132 TM, 121 TW	N/A	No	q3,12 mo	Creatinine	No (duplicate data)
17	Wierckx et al. (19) 2014	ENIGI, Norway, Belgium	Prospective	53 TM, 53 TW	N/A	No	q3 mo,12 mo	Creatinine	No (duplicate data)
18	Yahyaoui et al. (39) 2008	Spain	Prospective	47 TM, 22 TW	N/A	No	1, 2 yr	Creatinine	Yes
19	Angus et al. (40) 2019	Australia	Cross-sectional	80 TW	≥6 mo, median 1.5 yr (0.9–2.6)	No	N/A	BUN, creatinine	No
20	Becerra Fernández et al. (41) 1999	Spain	Cross-sectional	26 TM, 31 TW	Minimum 6 mo, maximum 10 yr	No	N/A	BUN, creatinine	No
21	Cottrell et al. (42) 2018	United States	Cross-sectional	4 TW, 4 CisM, 4 CisW	1–9 yr	No	N/A	Creatinine clearance	No
22	Roberts et al. (43) 2014	United States	Cross-sectional	55 TW, 20 CisM, 20 CisW	≥6 mo to ≤10 yr follow-up	No	Median 4 yr	BUN, creatinine	No
23	Van Caenegem et al. (44) 2012	Belgium	Cross-sectional	50 TM GAHT, 16 TM no GAHT, 50 CisW, 16 CisW	9.9 yr (range, 3.2–27.5)	Yes	N/A	Creatinine	No
24	Wierckx et al. (45) 2012	Belgium	Cross-sectional	50 TM, 50 TW	TM=10 yr, TW=9.2 yr	Yes	N/A	Creatinine	No
25	Giltay et al. (47) 2003	The Netherlands	Open-label, randomized trial	40 TW	N/A	No	2, 4 mo	Creatinine	Yes
26	Scharff et al. (46) 2019	ENIGI, The Netherlands, Belgium, Norway	Prospective	278 TM, 248 TW	N/A	No	3, 12 mo	Creatinine	Yes

AI, aromatase inhibitor; TM, transgender man; TW, transgender woman; N/A, not applicable; CisM, cisgender man; GAHT, gender-affirming hormone therapy; ENIGI, European Network for the Investigation of Gender Incongruence; q3, every; CisW, cisgender woman.

The Effect of Gender-Affirming Hormone Therapy on Serum Creatinine, Other Markers of Kidney Function, and Glomerular Filtration Rate

There were nine studies with data available for meta-analysis without duplicate data. The quality of included studies is shown in Supplemental Figure 1. Seven studies were deemed to be of fair quality, and two studies were deemed to be of poor quality. No study was deemed to be of good quality. No study reported a sample size calculation, and no study was blinded; also, there were issues regarding the generalizability of cohorts (as most were single center with younger, healthier populations); the lack of reporting or use of standardized, calibrated creatinine assays (Supplemental Table 4); and loss to follow-up. There were a total of 488 transgender men, of which 33 (two studies), 92 (three studies), and 370 (five studies) contributed data to the 3-, 6-, and 12-month post-GAHT time points. In transgender men, the mean baseline creatinine (SD) prior to the initiation of GAHT was 0.77 mg/dl (0.49). Serum creatinine increased after initiating GAHT by 0.14 mg/dl (95% CI, −0.07 to 0.35) at 3 months, 0.13 mg/dl (95% CI, −0.14 to 0.41) at 6 months, and 0.15 mg/dl (95% CI, 0.00 to 0.29) at 12 months (Table 2).

Table 2.

The change in serum creatinine over time in transgender men treated with gender-affirming hormone therapy

Time, mo	Change in Serum Creatinine, mg/dl	SEM	P Value	95% Confidence Interval
3	0.14	0.11	0.19	−0.07 to 0.35
6	0.13	0.14	0.34	−0.14 to 0.41
12	0.15	0.07	0.05	0.00 to 0.29

Random effects using estimated within-study and between-study correlations using PROC MIXED with autoregressive within-study and between-study covariance matrices (14).

There were a total of 529 transgender women, of which 77 (four studies), 167 (three studies), and 361 (five studies) contributed data to the 3-, 6-, and 12-month post-GAHT time points. In transgender women, the mean baseline serum creatinine (SD) prior to the initiation of GAHT was 0.93 (0.58). Serum creatinine decreased after initiating GAHT by 0.00 mg/dl (95% CI, −0.12 to 0.13) at 3 months, −0.05 mg/dl (95% CI, −0.24 to 0.13) at 6 months, and −0.05 (95% CI, −0.16 to 0.05) at 12 months (Table 3). Significant heterogeneity was not detected in estimates for either transgender men or women at any time point (Figures 1–3). We did not attempt to perform a meta-regression or multivariable regression to identify study-level or patient-level factors independently associated with changes in serum creatinine (i.e., age, race, GAHT type/route/dosing, achieved hormone levels, and baseline biomarker values) due to the number of studies and the lack of individual patient-level data.

Table 3.

The change in serum creatinine over time in transgender women treated with gender-affirming hormone therapy

Time, mo	Change in Serum Creatinine, mg/dl	SEM	P Value	95% Confidence Interval
3	0.00	0.06	0.94	−0.12 to 0.13
6	−0.05	0.10	0.57	−0.24 to 0.13
12	−0.05	0.05	0.29	−0.16 to 0.05

Random effects using estimated within-study and between-study correlations using PROC MIXED with autoregressive within-study and between-study covariance matrices (14).

Figure 1.

PRISMA flow diagram. PRISMA, preferred reporting items for systematic reviews and meta-analyses.

Figure 3.

Changes in serum creatinine in transgender women on GAHT at 3, 6, 12 months: nine studies, n=593, random effects, DerSimonian and Laird method. 95% CI, 95% confidence interval; IV, weighted mean difference SE, standard error.

Figure 2.

Changes in serum creatinine in transgender men on gender-affirming hormone therapy (GAHT) at 3, 6, and 12 months: eight studies, n=488, random effects, DerSimonian and Laird method. 95% CI, 95% confidence interval; IV, weighted mean difference; SE, standard error.

There were a total of 94 transgender men, of which 79 (two studies) and 15 (one study) contributed data to the 6- and 12-month post-GAHT time points for BUN. In transgender men, the mean baseline BUN (SD) prior to the initiation of GAHT was 10.94 mg/dl (0.02). BUN did not significantly change after initiating GAHT at 6 months (−0.28 mg/dl; 95% CI, −3.53 to 2.98) or 12 months (0.60 mg/dl; 95% CI, −2.61 to 3.81) (Table 4). There were a total of 198 transgender women, of which 126 (two studies) and 47 (one study) contributed data to the 6- and 12-month post-GAHT time points for BUN. In transgender women, the mean baseline BUN (SD) prior to the initiation of GAHT was 12.18 mg/dl (0.10). BUN did not significantly change after initiating GAHT at 6 months (0.21 mg/dl; 95% CI, −1.63 to 2.05) or 12 months (0.92 mg/dl; 95% CI, −1.14 to 2.98) (Table 5).

Table 4.

The change in BUN over time in transgender men treated with gender-affirming hormone therapy

Time, mo	Change in BUN, mg/dl	SEM	P Value	95% Confidence Interval
6	−0.28	1.66	0.87	−3.53 to 2.98
12	0.60	1.64	0.72	−2.61 to 3.81

Random effects using estimated within-study and between-study correlations using PROC MIXED with autoregressive within-study and between-study covariance matrices (14).

Table 5.

The change in BUN over time in transgender women treated with gender-affirming hormone therapy

Time, mo	Change in BUN, mg/dl	SEM	P Value	95% Confidence Interval
6	0.21	0.94	0.83	−1.63 to 2.05
12	0.92	1.05	0.38	−1.14 to 2.98

Random effects using estimated within-study and between-study correlations using PROC MIXED with autoregressive within-study and between-study covariance matrices (14).

The limited number of studies for kidney function biomarkers other than serum creatinine and BUN precluded a meta-analysis for other outcomes of interest. However, in the two studies that reported 24-hour urine creatinine excretion, there were statistically significant changes in transgender men but not transgender women at 4 and 12 months of follow-up. Giltay et al. (27) reported that the urinary creatinine excretion rate (millimoles per 24 hours) did not change in 17 transgender women treated with ethinyl estradiol at 100 µg/d in combination with cyproterone acetate at 100 mg/d from baseline to 4 months (15.0 [SD 4.1] to 15.3 [SD 3.5]; P=0.62). However, it did increase from 10.3 (SD 2.8) to 12.1 (SD 3.7; P=0.008) from baseline to 4 months in 17 transgender men treated with testosterone esters at 250 mg every 2 weeks. Giltay et al. (28) also reported that the urinary creatinine excretion rate (millimoles per 24 hours) did not change in 15 transgender women treated with the same GAHT described above from baseline (14.7 [SD 4.1]) to 12 months (13.9 [SD 3.2]; P=0.19). However, it did increase from 10.8 (SD 2.7) at baseline to 14.0 (SD 3.8) at 12 months (P=0.01) in 14 transgender men treated with 250-mg intramuscular testosterone esters every 2 weeks. Lastly, in the one study that reported creatinine clearance measured by 24-hour urine, Nicholls et al. (34) showed that there was no difference (103 ml/min [SD 23] before treatment and 116 ml/min [SD 20] after treatment) in 22 transgender women (21 received stilboestrol at a mean of 20.25 mg/d, whereas one received ethinylestradiol at 0.4 mg daily) over a mean follow-up of 10 weeks (range, 2 weeks to 10 months).

Discussion

In this systematic review of 26 studies and meta-analysis of nine studies including 488 adult transgender men and 529 adult transgender women on masculinizing and feminizing GAHT, serum creatinine increased in transgender men and did not significantly change in transgender women initiating GAHT. BUN did not change after the initiation of GAHT in either transgender men or women. No study reported on the effect of GAHT on other kidney function biomarkers, including cystatin C, albuminuria, proteinuria, or measured GFR.

Previous studies addressing the effect of GAHT on kidney function were, individually, small with limited power to precisely quantify differences in biomarkers of kidney function. Our finding that an increase in serum creatinine in transgender men occurs by 12 months on GAHT and encompasses a clinically meaningful change requires awareness by health care providers, including primary care physicians, endocrinologists, and nephrologists. For example, a 25-year-old transgender man whose baseline serum creatinine is 0.77 mg/dl (eGFR=127 ml/min per 1.73 m² by CKD-EPI) and whose serum creatinine 12 months post-GAHT is either 0.92 mg/dl (+0.15 mg/dl or the mean changes in serum creatinine as above, which corresponds to an eGFR of 118 ml/min per 1.73 m²) or 1.06 mg/dl (0.29 mg/dl or the 95% upper CI as above, which corresponds to an eGFR of 100 ml/min per 1.73 m²) represents an approximately 10%–15% decline in eGFR (8). This presumably reflects the corresponding increase in muscle mass due to the use of testosterone rather than a true change in GFR (6,7) and would not require further investigation in the absence of other signs of kidney disease. In transgender women, on the other hand, serum creatinine did not change significantly over 3, 6, and 12 months on GAHT. However, it should be acknowledged that across both transgender men and women, there is a spectrum of change in serum creatinine likely due to changes in lean muscle mass as a result of GAHT type, dose, adherence, and achieved target hormone levels as well as potentially other non-GFR determinants of serum creatinine, such as diet (21). Lastly, this interpretation assumes that changes in serum creatinine are due to changes in lean muscle mass due to GAHT and not GFR. This is currently unknown and possibly erroneous given the association of female sex/gender with adverse outcomes in CKD (22–24).

This study has important clinical, research, and health policy implications. From a clinical perspective, physicians should consider measuring GFR when considering important treatment decisions in patients undergoing GAHT given its potential effect on serum creatinine, especially in transgender men. These decisions could include referral to nephrology, initiation or discontinuation of disease-modifying therapy (e.g., renin-angiotensin-aldosterone system inhibitors, sodium-glucose cotransport 2 inhibitors, and mineralocorticoid receptor antagonists), dosing for medications with narrow therapeutic indices, and/or referral for kidney transplantation and dialysis initiation. From a research perspective, measured GFR studies and studies on other established kidney biomarkers (cystatin C, β-trace protein, and β2-microglobulin) are needed in transgender populations treated with and without GAHT. From a health policy perspective, education efforts on the metabolic changes that occur with GAHT are needed for all health care providers (i.e., primary care providers, endocrinologists, surgeons, and subspecialists) who provide care to transgender patients who may not be aware of the specialized literature on kidney biomarkers.

To our knowledge, this is the first systematic review and meta-analysis to examine the effect of GAHT on kidney function biomarkers and GFR in any transgender population. It used a liberal search strategy with limited exclusion criteria to capture all relevant studies and meta-analyzed outcomes over a series of clinically relevant time points after the initiation of GAHT. Limitations include the low certainty of evidence given the inclusion of pre/post studies mostly of fair but not good quality due to a lack of blinding; concerns regarding generalizability and loss to follow-up; the lack of inclusion of pediatrics, adolescent, and nonbinary/gender-diverse populations; its inclusion of mostly studies from the United States and Europe; and a lack of information regarding creatinine assay standardization and calibration across studies. Studies also focused mostly on creatinine and did not include other kidney function biomarkers such as cystatin C, albuminuria/proteinuria, eGFR, and measured GFR, so the effect on GAHT on these outcomes is uncertain as well as the influence of GAHT type, route of administration, and dose (we were unable to perform any sensitivity analyses due to the inability to properly categorize exposures, and further subgroup analyses would be underpowered). Included studies were predominantly in younger, healthier populations without CKD, and therefore, how GAHT affects kidney function biomarkers in those with established CKD is unknown (15,16). Studies in transgender women who used mineralocorticoid receptor antagonists as antiandrogen therapy universally did not account for its known antihypertensive effects, which decrease GFR and subsequently influence serum levels of kidney biomarkers and proteinuria. Lastly, we are missing some patients from the multicenter ENIGI (17–19) cohort as we conservatively only included one study (46) with the most comprehensive data as suggested by the study’s authors to remove the possibility of any duplicate data.

In summary, our systematic review and meta-analysis show that GAHT increases serum creatinine to a clinically significant degree in transgender men but does not significantly affect serum creatinine in transgender women; however, patients with CKD were not represented. Awareness of the serum creatinine trajectory post-GAHT is important to frame the need for diagnostic testing for AKI or CKD and nephrology consultation. Additional research is needed to determine the effect of GAHT on cystatin C, albuminuria, proteinuria, and measured GFR in addition to how to accurately and precisely estimate GFR in transgender populations treated with and without GAHT (9). Specifically, high-quality, adequately powered prospective observational studies with repeated measurements are needed in diverse populations (adults, pediatrics, and the elderly) evaluating different types of GAHT (routes and dosing) in both transgender men and women across the spectrum of health and CKD.

Disclosures

S.B. Ahmed reports serving as an advisory board member of the Canadian Institutes of Health Research Institute of Gender and Health (volunteer position), as a Canadian Medical Association Journal Governance Council Member (elected position), and as the Education Chair of the Organization for the Study of Sex Differences (volunteer position). D. Collister reports employment with Alberta Health Services; consultancy agreements with Akebia; research funding from Boehringer Ingelheim/Research Manitoba, the Canadian Institutes of Health Research, and the Kidney Foundation of Canada; and serving in an advisory or leadership role for the Canadian Nephrology Trials Network (Executive Committee and Scientific Operations Committee). S. Curtis and R. Whitlock report employment with the Chronic Disease Innovation Centre. T. Ferguson reports employment with Chronic Disease Innovation Centre; consultancy agreements with Clinpredict, Quanta Dialysis Technologies, and Strategic Health Resources (Tricida Inc., Protagonist Therapeutics); ownership interest in Klinrisk Inc.; and honoraria from Baxter Canada. A.C. Millar reports consultancy agreements with Novo Nordisk Canada, honoraria from Dexcom Canada, and speakers bureau for Dexcom Canada. N. Tangri reports consultancy agreements with Marizyme, Mesentech Inc., PulseData Inc., Renibus, and Tricida Inc.; ownership interest in Clinpredict Ltd., Klinrisk, Marizyme, Mesentech Inc., PulseData Inc., Quanta, Renibus, and Tricida Inc.; research funding from AstraZeneca Inc., Bayer, BI-Lilly, Janssen, Otsuka, and Tricida Inc.; honoraria from AstraZeneca Inc., Bayer, BI-Lilly, Janssen, Otsuka Pharmaceuticals, and Pfizer; patents or royalties from Klinrisk and Marizyme; serving in an advisory or leadership role for Clinpredict, Klinrisk, and Tricida Inc.; other interests or relationships with Clinpredict and the National Kidney Foundation; and is the founder of Klinrisk. M. Walsh reports employment with the Ontario Renal Network; research funding from the British Heart Foundation, the Canadian Institutes of Health Research, the Health Research Council, the National Health and Medical Research Council, the National Institutes of Health Research, and Vifor (no salary support received through any research funding); serving in an advisory or leadership role for Bayer (steering committee, payment to institution) and Otsuka (national leader, payment to institution); and other interests or relationships with Novo Nordisk (event adjudication, payment to institution). All remaining authors have nothing to disclose.

Funding

None.

Author Contributions

D. Collister conceptualized the study; D. Collister, S. Curtis, and E. Krupka were responsible for data curation; D. Collister was responsible for investigation; D. Collister, T. Ferguson, and R. Whitlock were responsible for formal analysis; D. Collister was responsible for methodology; D. Collister was responsible for validation; D. Collister was responsible for visualization; D. Collister provided supervision; D. Collister wrote the original draft; and S.B. Ahmed, N. Askin, D. Collister, S. Curtis, M. Dahl, T. Ferguson, R. Fung, E. Krupka, A.C. Millar, N. Tangri, M. Walsh, and R. Whitlock reviewed and edited the manuscript.

Data Sharing Statement

All data are included in the manuscript and/or supporting information.

Supplemental Material

This article contains the following supplemental material online at http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.01890222/-/DCSupplemental.

Supplemental Table 1. Search strategy.

Supplemental Table 2. Studies of transgender men on GAHT and creatinine at baseline and follow-up (n=18 studies).

Supplemental Table 3. Studies of transgender women on GAHT and creatinine at baseline and follow-up (n=23 studies).

Supplemental Table 4. Characteristics of creatinine assays of studies included in the meta-analysis.

Supplemental Figure 1. Quality assessment.