Acute Kidney Injury Biomarker Responses in Young and Older Female Adults Following Mild Hypohydration

Abstract

Background:

Hypohydration reduces kidney function and increases acute kidney injury (AKI) risk. Aging increases hypohydration-induced kidney dysfunction in males, but female aging studies are lacking. Therefore, we compared the effects of mild hypohydration on kidney function and AKI biomarkers in young (YF) and older female adults (OF).

Methods:

In a random crossover design, seventeen YF (20–35 years old) and 9 OF (55–75 years old), who were apparently healthy, completed two hydration protocols with ≥1 week washout: 1) baseline hydration, and 2) stepwise water restriction over 3 days, concluding with 16 hours water deprivation. We assessed hydration, AKI biomarkers, and the renin-angiotensin-aldosterone system (RAAS) from blood and 24-hour urine samples. The effects of age group and condition were assessed using two-way mixed-effects analysis and reported as mean±SD.

Results:

Hypohydration increased urine specific gravity and osmolality (condition effect: p<0.001) with no other main effects. Neutrophil gelatinase-associated lipocalin excretion decreased with hypohydration (p=0.04), independent of age (p=0.62). Urine cystatin C excretion increased in YF (p<0.001) but not OF (p=0.69), with a significant interaction effect (p=0.017). Nephrin excretion and urinary IGFBP7*TIMP-2 increased after hypohydration (both p<0.001), independent of age (both p≥0.35). OF exhibited lower plasma renin activity than YF (p=0.046), with no other main or interaction effects for other RAAS markers.

Conclusions:

OF did not exhibit greater AKI biomarker responses to mild hypohydration, contrasting with male studies showing age-related kidney function decline. These results highlight the need for research to clarify potential sex-based differences in age-related decline in kidney function.

Graphical Abstract

INTRODUCTION

Reduced thirst sensation, urine concentration ability, and higher baseline plasma osmolality in older adults can increase the overall vulnerability to hypohydration^1–3. Indeed, National Health and Nutrition Examination Survey (NHANES) data suggest that nearly two-thirds of older adults do not consume enough fluids⁴. This low fluid intake puts older adults in a state of body water deficit, or hypohydration, which is associated with higher rates of disability, morbidity, and mortality^5,6, including an increased risk for acute kidney injury (AKI) and chronic kidney disease (CKD)^7,8. Hypohydration is associated with a fourfold higher risk for developing AKI and a 60% increased mortality among older adults who are hospitalized⁹. While hypohydration is clinically relevant, previous research on the integrative physiological responses to hypohydration in young and older adults is primarily limited to males³. Thus, there are key knowledge gaps in understanding age-related renal responses to hypohydration in female adults.

In healthy males and females, acute hypohydration can induce sub-clinical increases in AKI biomarkers¹⁰. Chronic or repeated hypohydration may further raise the risk of developing CKD^11–13. However, a large randomized controlled trial in adults with CKD found that increasing water intake did not slow kidney function decline, suggesting that the relation between hydration and kidney outcomes is complex and may be variable between different populations¹⁴. Sensitive AKI biomarkers such as urinary neutrophil gelatinase-associated lipocalin (NGAL), urinary cystatin C, and the product of urinary insulin-like growth factor-binding protein 7 and tissue inhibitor of metalloproteinase-2 (IGFBP7*TIMP-2) increase during hypohydration, suggesting early kidney strain^10,15,16. Critically, existing research on hypohydration-induced AKI is predominantly limited to male populations, despite evidence that both sexes are susceptible¹⁰, leaving responses across the female lifespan poorly characterized.

Hypohydration stimulates the Renin-Angiotensin-Aldosterone System (RAAS) to maintain plasma volume by increasing sodium and water reabsorption^17,18. However, studies limited to male participants have linked aging to lower baseline renin activity and blunted RAAS responses during hypohydration¹⁹. Additionally, the urine concentration ability of the kidneys in response to hypohydration is reflected by free water clearance (CH₂O), which may be influenced by sex and age. Studies in young males reported reduced urine concentration ability during exercise-induced hypohydration compared with euhydration, reflected by less negative CH₂O²⁰. However, studies among females remain absent, despite established sex differences in anti-diuretic hormone (ADH) regulation^21–23. The effect of aging reveals further knowledge gaps. Aging is associated with higher ADH concentrations²⁴, especially in elderly individuals with hypernatremia^25,26, which may reflect a compensatory mechanism for reduced urine concentrating ability. Notably, no studies have compared changes in copeptin (a stable surrogate marker of ADH) or CH₂O in young females (YF) and OF following mild hypohydration.

Therefore, the purpose of this study was to compare AKI biomarkers, RAAS responses, and renal function biomarkers (CH₂O, eGFR, etc.) between YF and OF undergoing water restriction-induced hypohydration. We hypothesized that OF would have attenuated compensatory mechanisms and greater renal stress (e.g., higher concentrations of AKI biomarkers) after mild hypohydration compared to YF.

METHODS

Participants

The Institutional Review Board at the University of Delaware approved the study protocol, and the study was preregistered, which aligns with the Declaration of Helsinki. All participants provided written consent after being informed about this study’s nature, purpose, and procedures. Data reported in this manuscript are part of a larger registered trial (ClinicalTrials.gov Identifier: NCT03560869); we previously published on sympathetic regulation of blood pressure (BP) among a subset of 21 adults²⁷ and BP reactivity among a subset of 22 adults¹. The present dataset, used for a secondary analysis, includes 26 participants with novel data on RAAS, copeptin, renal function, and AKI biomarkers from blood and urine samples that have not been previously reported.

We enrolled 17 YF (20–35 years old) and 9 OF (55–75 years old). Inclusion criteria were a body mass index of <30 kg/m², systolic BP of 90–139 mmHg, and diastolic BP of 50–89 mmHg. The BP screening was conducted using oscillometric BP measurement and is reported in Table 1. Exclusion criteria were any cardiovascular, metabolic, or pulmonary diseases, previous diagnosis of hypertension, past or current use of antihypertensive medication, and current or recent (within the last 6 months) use of nicotine. All YF were tested in their early follicular phase, and OF self-reported as post-menopausal.

Table 1.

Screening Characteristics

	YF (n = 17)	OF (n = 9)	p-value	Effect Size
Age (years)	22[4]	64[12]	<0.0001	r=1.00
Race/Ethnicity	1 Asian, 4 Black, 11 White, 1 Hispanic	1 Black, 8 White
Body mass index (kg/m2)	21.3[4.4]	22.5[3.4]	0.77	r=0.08
Body fat (%)	24.8±7.1	31.3±5.8	0.03	d=1.00
Systolic BP (mmHg)	104±2	118±4	<0.01	d=4.43
Diastolic BP (mmHg)	59±6	75±7	<0.001	d=2.45
MVPA (minutes/day)	80±34	70±35	0.54	d=0.29
Step Count (steps/day)	9,053±4,268	7,293±2,788	0.31	d=0.49

Blood pressure (BP), Moderate-to-vigorous physical activity (MVPA), rank-biserial correlation (r), Cohen’s d (d). Data shown as mean±SD or median[IQR].

Experimental Trials

All participants completed two different hydration trials in a random crossover fashion, separated by at least a one-week washout. During the baseline hydration protocol, participants were instructed to consume 23 mL water/kg body mass/day for three consecutive days, as previously described^1,27,28. The baseline water intake was designed to be between participants’ typical water consumption of ~1 L/day²⁹ and the standard recommendation of 2.7 L/day³⁰. The hypohydration protocol consisted of a gradual reduction of water consumption over three consecutive days: Participants consumed 23 mL water/kg body mass on day one, 17 mL water/kg body mass on day two, and 10 mL water/kg body mass on day three, followed by a ≥16-hour water restriction (complete abstinence from water intake) before the study visit. During the 24 hours preceding both laboratory visits, participants wore an ambulatory BP monitor and collected their urine. Additionally, participants were instructed to consume the recommended daily sodium intake of 2,300 mg for both three-day protocols. Their sodium intake was assessed using three-day diet logs during both hydration conditions.

Laboratory Visit

All participants fasted for at least four hours prior to each laboratory visit and consumed 0.3 L of water before arriving at the laboratory for spot urine and venous blood samples. Upon arrival, the ambulatory BP monitor and the 24-hour urine sample were collected, and participants’ spot urine specific gravity (USG; Refractometry), body composition (Tanita Body Composition Analyzer, Model TBF-300A; Arlington Heights, IL), and rating of perceived thirst were assessed. Perceived thirst was rated using a Likert Scale consisting of a 10 cm line with anchor points of “not at all” at 0 cm on the left and “very” at 10 cm on the right. Participants rested quietly for at least 20 minutes in a temperature-controlled room (22–24°C) with dimmed lights before we collected venous blood samples in EDTA tubes for plasma, tubes with no additives for serum, and EDTA tubes pre-chilled and pre-treated with a protease inhibitor for angiotensin 2.

Blood and Urine Analysis

Plasma copeptin was used as a stable surrogate for ADH³¹. Copeptin was measured using an automated chemiluminescent sandwich immunoassay (Brahms Kryptor compact Plus, Thermo Fisher Scientific, Hennigsdorf, Germany). To assess the sensitivity of plasma copeptin to changes in plasma osmolality, we compared responses between age groups. After logarithmic transformation of plasma copeptin (to correct for non-normal distribution), we performed a linear mixed-effects model, looking at the relation between copeptin, plasma osmolality, age group, and hydration condition. Log-transformed plasma copeptin was set as the dependent variable, plasma osmolality as a covariate, and age group and hydration condition as fixed factors. Additionally, participant ID was specified as a cluster variable to account for random intercepts and between-participant variability.

Plasma samples were analyzed for circulating aldosterone and angiotensin 2 using radioimmunoassays, and plasma renin activity using enzyme-linked immunosorbent assays (ELISA). Whole blood samples were also analyzed for hemoglobin (Hb; Hb 201+; Hemocue, Lake Forest, CA, USA), hematocrit (Hct; Sorvall Legend Micro 17 Microcentrifuge, Thermo Fisher Scientific, Waltham, MA, USA), total plasma volume (TPV), and plasma volume change. TPV was estimated using Nadler’s equation³²:

$$TPV=\left(\left(0.3561\times {\text{Height}}^3\right)+\left(0.03308\times \text{Mass}\right)+0.1833\right)\times \left(1-\text{Hct}\right)$$

Plasma volume change was estimated using the equation of Dill and Costill³³:

$$\mathit{\text{Plasma Volume Change}}=\frac{{Hb}_{pre}\times \left(1-{Hct}_{\mathit{\text{post}}}\right)}{{Hb}_{\mathit{\text{post}}}\times \left(1-{Hct}_{pre}\right)}-1$$

24-hour urine samples were analyzed for USG (Goldberg Brix Refractometer, Reichert Technologies), NGAL (R&D Systems, NGAL Quantikine ELISA Kit, #QK1757), Nephrin (LSBio, Nephrin ELISA Kit, #LS-F21185), IGFBP7 (RayBiotech, Human IGFBP-RP1 ELISA Kit, #ELH-IGFBPRP1), TIMP-2 (RayBiotech, Human TIMP-2 ELISA Kit, #ELH-TIMP2), and cystatin C (R&D Systems, Human Cystatin C Quantikine ELISA Kit, #DSCTC0). Urinary AKI biomarkers (NGAL, nephrin, IGFBP7, and TIMP-2) were normalized to urine flow rate to account for physiological reductions in urine volume during fluid restriction. Urine flow rate was calculated based on the 24-hour urine collection as follows:

$$\mathit{\text{Urine Flow Rate}}\left(mL/min\right)=\frac{\text{24hr Urine Volume}\left(\text{mL}\right)}{\text{Collection Time}\left(\text{min}\right)}$$

In addition, we calculated the product of IGFBP7 and TIMP-2 concentrations as a sensitive FDA-approved AKI biomarker³⁴. A threshold value of >0.3 (ng/mL)²/1000 was used to indicate elevated AKI risk³⁵.

24-hour urine samples and venous blood samples were analyzed for electrolyte concentrations (EasyElectrolyte Analyzer; Medica, Beford, MA, USA), osmolality (3D3 Osmometer; Advanced Instruments, Norwood, MA), and creatinine concentrations using underivatized stable isotope dilution LC-MS/MS³⁶.

Creatinine clearance was calculated from urine and serum creatinine, urine volume, and collection time based on the 24-hour urine collection³⁷ and normalized to 1.73m² body surface area using the Du Bios formula³⁸:

$$\mathit{\text{Creatinine Clearance}}\left(mL/min/1.73m^2\right)=\frac{\mathit{\text{Urine Creatinine}}(mg/dL)\times \mathit{\text{Urine Volume}}(mL)}{\mathit{\text{Serum Creatinine}}(mg/dL)\times \mathit{\text{Collection Time}}(\text{min})}\times \frac{1.73}{0.007184\times {\mathit{\text{Body Mass}}\left(kg\right)}^{0.425}\times {\mathit{\text{Height}}\left(cm\right)}^{0.725}}$$

eGFR was calculated from serum creatinine (Scr) and age³⁹:

$$\mathit{\text{eGFR}}\left(mL/min/1.73m^2\right)=142\times min{\left(\frac{Scr}{0.7},1\right)}^{-0.241}\times \text{max}{\left(\frac{Scr}{0.7},1\right)}^{-1.2}\times {0.9938}^{Age}\times 1.012$$

CH₂O was calculated from urine flow rate (V), and urine and plasma osmolality⁴⁰:

$$C{H}_{2}O\left(mL/min\right)=\text{V}-\left(\frac{U_{Osm}\times V}{P_{Osm}}\right)$$

Statistical Analysis

Normality of all variables was assessed using the Shapiro-Wilk test and QQ plots. Age, body mass index, body mass, CH₂O, urine creatinine, urine sodium, serum potassium, plasma renin activity, aldosterone, urinary NGAL, and urinary nephrin were not normally distributed. Screening characteristics and plasma volume between conditions were compared between YF and OF using unpaired, two-tailed t-tests when data were normally distributed or Mann-Whitney U tests when data were not normally distributed. The proportion of self-reported race/ethnicity was compared between age groups using Fisher’s exact test. To examine the effects of hydration condition and age group on hydration-related variables, two-way mixed-effects analyses (age group [independent factor] × condition [repeated factor]) were conducted. For copeptin, a linear mixed-effects model was applied after log-transformation to compare osmolality-mediated copeptin responses between YF and OF. A post hoc Tukey test for multiple comparisons was used when a significant interaction effect was found. Additionally, we compared the proportion of participants who passed the threshold of increased risk for AKI (IGFBP7*TIMP-2 above 0.3 (ng/mL)²/1000)³⁵ between age groups using Fisher’s exact test. All data were analyzed using GraphPad Prism 10.4 (GraphPad Software Inc., La Jolla, CA). Significance was set a priori at α < 0.05. All data are presented as mean±SD or median[IQR].

RESULTS

Participant screening characteristics are provided in Table 1. By design, YF participants were significantly younger than OF. YF exhibited lower body fat percentage, systolic BP, and diastolic BP (all p≤0.03). In contrast, racial and ethnic composition, body mass index, moderate-to-vigorous intensity physical activity, and step counts were not different between groups (all p≥0.31).

Hydration Responses

Changes in hydration-related variables are presented in Table 2. Hypohydration led to significant increases in thirst, USG, urine osmolality, plasma osmolality, and decreased CH₂O (condition effect: all p<0.01). Plasma osmolality was significantly higher in OF (age effect: p<0.0001) with no significant interaction effect (p=0.95). No other age or interaction effects were observed in these variables (all p≥0.06).

Table 2.

Changes in Hydration-Related Variables

Variables	YF (n = 17)		OF (n = 9)		Main effects
	BH	MH	BH	MH	Age	Cond	Age*Cond
Thirst	2.4±2.5	6.6±2.5	2.4±1.5	6.6±2.3	0.98	<0.001	0.95
Body Mass (kg)	59.1±10.3	59.0±10.2	58.6±8.1	58.0±8.4	0.86	0.13	0.16
CH2O (mL/min)	−0.53±0.50	−1.29±0.97	−0.48±0.29	−0.73±0.14	0.20	0.001	0.08
Spot USG	1.015±0.006	1.022±0.004	1.012±0.006	1.022±0.004	0.54	<0.0001	0.41
24-hr USG	1.012±0.004	1.018±0.003	1.011±0.002	1.017±0.004	0.46	<0.0001	0.92
24-hr Urine Osmolality (mOsm/kg H2O)	462±132	681±127	436±88	669±158	0.71	<0.0001	0.75
Plasma Osmolality (mOsm/kg H2O)	287±5	290±5	297±4	300±4	<0.0001	<0.01	0.95
Urine Creatinine (mg/dL)	98±52	127±45	117±83	117±52	0.78	0.36	0.35
Serum Creatinine (mg/dL)	0.80±0.24	0.86±0.18	0.74±0.15	0.72±0.33	0.29	0.86	0.59
Urine Na+ (mmol/24hr)	149±103	121±49	111±35	98±28	0.16	0.28	0.71
Serum Na+ (mmol/L)	141.5±1.5	141.7±1.4	142.9±2.4	144.1±1.7	<0.01	0.18	0.35
Serum K+ (mmol/L)	4.0±0.3	4.0±0.3	4.2±0.5	4.0±0.4	0.43	0.46	0.42
Serum Cl− (mmol/L)	103.9±2.5	106.2±2.2a	106.7±1.8b	106.6±2.3	0.06	0.07	0.045

BH, baseline hydration condition; MH, mild hypohydration condition; CH₂O, free water clearance; USG, urine specific gravity.

^aDifferent from BL,

^bDifferent from YF

There were no significant age, condition, or interaction effects in urine and serum creatinine concentrations, sodium excretion over 24 hours, or serum potassium (all p≥0.16). However, serum sodium was significantly higher in OF (age effect: p<0.01) with no significant condition or interaction effects (all p≥0.18), whereas serum chloride exhibited a significant interaction effect (p=0.045). In YF, serum chloride increased significantly from baseline to mild hypohydration (p=0.01), but not in OF (p=0.90).

In addition, we noted YF had lower awake systolic BP (baseline: 114±7 & mild hypohydration: 113±5 mmHg) compared with OF (baseline: 126±14 & mild hypohydration: 128±19 mmHg; age effect: p=0.01) with no significant condition (p=0.43) or interaction effect (p=0.34). Similarly, awake diastolic BP was significantly lower in YF (baseline: 67±3 vs. mild hypohydration: 66±3 mmHg) compared with OF (baseline: 74±11 vs. mild hypohydration: 74±12 mmHg, age effect: p=0.04) with no significant condition (p=0.91) or interaction effects (p=0.57). Self-reported sodium intake was not different between age groups during baseline hydration (YF: 2,085±401 vs. OF: 2,036±309 mg/day) nor during mild hypohydration (YF: 2,073±447 vs. OF: 2,039±371 mg/day), with no significant age (p=0.78), condition (p=0.95), or interaction effect (p=0.92).

The hematocrit in OF (baseline: 41±2 vs. mild hypohydration: 41±2%) was significantly higher compared with YF (baseline: 38±3 vs. mild hypohydration: 38±4%; age effect: p=0.03), but there were no significant condition or interaction effects (both p≥0.74). Similarly, hemoglobin was significantly higher in OF (baseline: 13.2±0.9 vs. mild hypohydration: 13.1±1.0 g/dL) compared with YF (baseline: 12.2±1.0 vs. mild hypohydration: 12.4±0.8 g/dL; age effect: p=0.02), without significant condition (p=0.23) or interaction (p=0.77) effects. Further, plasma volume change was not significantly different (YF: −2.2±7.9% vs. OF: −0.4±5.8%, p=0.59). The estimated total plasma volume also did not show significant main effects (all p≥0.24). However, total body water (absolute volume in Liters) in OF (baseline: 29±2 L vs. mild hypohydration: 28±6 L) was significantly lower compared with YF (baseline: 33±4 L vs. mild hypohydration: 32±3 L; age effect: p=0.006), without significant condition (p=0.20) or interaction effects (p=0.73).

As depicted in Figure 1A, the urine flow rate decreased from baseline hydration to the mild hypohydration condition (condition effect: p<0.0001). However, there were no age or interaction effects (all p≥0.62). There were no main or interaction effects (all p≥0.07) for creatinine clearance (Figure 1B) and eGFR (Figure 1C).

Figure 1. Kidney Function during Baseline Hydration and Mild Hypohydration.

We assessed urine flow rate (A), creatinine clearance (B), and estimated glomerular filtration rate (eGFR; C) during baseline and mild hypohydration in young females (YF) and older females (OF). The tables above each panel display the p-values derived from two-way mixed-effects models.

Hormonal Responses

Changes in the RAAS are presented in Figure 2. Plasma renin activity was lower in OF compared with YF (age effect: p=0.046), with no condition or interaction effects (all p≥0.83). Angiotensin 2 and aldosterone did not have any significant main effects (all p≥0.053).

Figure 2. Renin-Angiotensin-Aldosterone System.

We measured plasma renin activity (A), angiotensin II (B), and aldosterone (C) during baseline and mild hypohydration conditions in young females (YF) and older females (OF). The tables above each panel display the p-values derived from two-way mixed-effects models.

Serum sodium was inversely related to log-transformed plasma renin activity (p=0.02), independent of age group and hydration condition (p≥0.16). The linear mixed-effects model explained 77% of the variance in plasma renin activity, accounting for both fixed and random effects (conditional R²=0.77).

Plasma copeptin concentrations increased in YF (baseline: 3.3±1.5 vs. mild hypohydration: 7.0±5.8 pmol/L) and in OF (baseline: 4.1±1.7 vs. mild hypohydration: 7.6±4.9 pmol/L, condition effect: p<0.001), without age or interaction effects (all p≥0.61). Increases in copeptin corresponding to increases in plasma osmolality are shown in Figure 3. The relation between plasma osmolality and copeptin concentrations was best represented when copeptin was log-transformed (YF: R²=0.27 & OF: R²=0.41). Plasma osmolality (p=0.02) and hydration condition (p<0.01) were significant predictors of log-transformed plasma copeptin, with a significant interaction effect (p=0.03). However, age group was not a significant predictor (p=0.62). We found no significant interaction between the hydration condition and age group (p=0.06) or between age group and plasma osmolality (p=0.88). The linear mixed-effects model explained 87% of the variance in plasma copeptin, accounting for both fixed and random effects (conditional R²=0.87).

Figure 3. Plasma Osmolality and Copeptin.

We assessed plasma osmolality and copeptin in young females (YF) and older females (OF) during baseline and mild hypohydration conditions. Increases in plasma copeptin across both age groups and hydration conditions are shown in Panel A. The table above the panel displays the p-values derived from two-way mixed-effects model. Linear mixed-effects model was applied after log-transformation of the copeptin data to assess their relation and compare osmolality-mediated copeptin responses between YF and OF, shown in Panel B.

Acute Kidney Injury Biomarkers

NGAL excretion significantly decreased from baseline to mild hypohydration (p=0.04) with no effect of age or interaction (all p≥0.62; Figure 4A). Cystatin C excretion elicited condition and interaction effects, where cystatin C excretion increased after mild hypohydration in YF (p<0.001) but not in OF (p=0.69; Figure 4B). Nephrin excretion (Figure 4C) and IGFBP7*TIMP-2 excretion (Figure 4D) elicited increases from baseline to mild hypohydration (condition effect: all p<0.001), but no age or interaction effects (all p≥0.07). The proportion of all individuals exceeding the IGFBP7*TIMP-2 concentration threshold of 0.3 (ng/mL)²/1000 during mild hypohydration was significantly higher compared to the baseline hydration condition (baseline: 21% vs. mild hypohydration: 77%, p<0.001). However, there were no differences between age groups in the proportion of individuals exceeding the threshold during mild hypohydration (YF: 73% vs. OF: 88%; p=0.62).

Figure 4. Acute Kidney Injury Biomarkers.

We measured concentrations and excretion rate of various urinary biomarkers that can be associated with acute kidney injury (AKI), including Urinary Neutrophil Gelatinase-associated Lipocalin (NGAL; A), Urinary Cystatin C (B), Urinary Nephrin (C), and the product of Urinary Insulin-like Growth Factor-binding Protein 7 and Tissue Inhibitor of Metalloproteinases-2 (IGFBP7*TIMP-2; D). The tables above each panel display the p-values derived from two-way mixed-effects models.

DISCUSSION

The primary novel findings of this study are 1) neither YF or OF had increases in RAAS biomarkers following mild hypohydration, 2) neither YF nor OF had changes in eGFR following mild hypohydration, 3) NGAL excretion decreased among YF and OF following mild hypohydration, 4) cystatin C excretion increased in YF, but not OF, following mild hypohydration, 5) hypohydration-induced increases in nephrin and IGFBP7*TIMP-2 excretion did not differ between YF and OF. Together, these data represent the first comprehensive characterization of how female aging affects RAAS, renal function, and AKI biomarker responses during mild hypohydration.

Hydration Responses

Increases in thirst, USG, and urine osmolality confirmed that our experimental protocol induced mild hypohydration without detectable changes in total body water or body mass. In both groups, spot USG exceeded the hypohydration threshold of 1.020, with urine osmolality values falling within the typical range for mild hypohydration (500 and 800 mOsm/kg H₂O)^41,42. The preserved plasma volume alongside elevated plasma osmolality suggests intracellular, euvolemic hypohydration rather than extracellular hypohydration⁴³. This is attributable to the induced water-intake deficit, rather than a combined loss of water and sodium (e.g., through excessive sweating). Such euvolemic hyperosmolality is expected to increase ADH and its surrogate copeptin. Chronically higher ADH levels are associated with adverse cardiometabolic outcomes, including higher risks of coronary artery disease^44,45, type 2 diabetes⁴⁶, and CKD⁴⁷. In contrast to more severe hypohydration protocols that include intense exercise, our mild hypohydration condition has strong ecological validity, as many individuals are in a state of chronic mild hypohydration due to inadequate habitual water intake rather than severe hypohydration following exercise in the heat. Importantly, baseline thirst and copeptin concentrations did not differ significantly between age groups, indicating that these variables did not differ by age.

Age-Related Differences in Plasma Osmolality

OF exhibited higher baseline plasma osmolality, exceeding the common hypohydration threshold of 295 mOsm/kg H₂O⁴⁸ during the baseline hydration condition. Because both age groups completed the same 3-day standardized baseline hydration protocol, the higher baseline osmolality in OF is unlikely to be protocol-related and may reflect a greater tendency toward underhydration in this group⁴⁹. Age-related changes in sodium handling and other osmoles may also contribute⁵⁰, as serum sodium, but not urine sodium, was higher in OF. Increased brain tonicity can stimulate sympathetic outflow and vasoconstriction⁵¹, but the small changes in osmolarity observed in our experiment are on the cusp of the one percent change needed to trigger this mechanism. Awake BP was consistently higher in OF than YF, with no further changes during mild hypohydration, suggesting that resting BP regulation remains intact during mild hypohydration in OF. Prior research supports this observation, reporting higher plasma osmolality in older compared with young adults. This may stem from an age-related reduction in total body water and plasma volume, as shown in males⁵², females with obesity⁵³, and healthy sedentary females⁵⁴, aligning with our observations. However, Jones et al.⁵⁴ found that physically active females preserved their total plasma volume, suggesting a potential protective role of physical activity. Our OF participants averaged 70 minutes per day of MVPA, which is more than three times the recommended amount of activity⁵⁵, and had preserved total plasma volume, supporting past findings⁵⁴.

Hormonal Responses

Neither group experienced activation of the RAAS, demonstrated by unchanged renin activity or aldosterone and serum sodium concentrations. These findings align with the mild hypohydration stimulus and the absence of significant extracellular fluid loss. However, OF exhibited higher serum sodium, which may have suppressed baseline renin release^56,57, as further indicated by the inverse association between serum sodium and plasma renin activity observed in our cohort. Lower baseline plasma renin activity in OF is consistent with prior studies showing an age-related decline of renin secretion in mixed populations and males only^19,58. This can further be influenced by an age-related decline in juxtaglomerular cells, leading to attenuated renin production and release⁵⁹.

Despite higher baseline plasma osmolality in OF, plasma copeptin concentrations were not different between age groups and increased similarly during hypohydration, suggesting that ADH-mediated water conservation remains effective with age. Prior studies on osmoreceptor sensitivity in aging are conflicting; some report enhanced ADH secretion in response to hypertonic saline in older versus younger adults^3,60, while others show reduced ADH release after water deprivation⁶¹. In the present study, age did not predict the relation between plasma copeptin and osmolality, indicating no differences in osmoreceptor sensitivity. This contrasts with findings from mixed-sex cohorts, which show greater ADH responses in older adults⁶², likely reflecting methodological or stimulus-specific differences across studies.

Renal Water Handling

Higher plasma osmolality in OF was primarily driven by higher serum sodium, the key determinant of plasma osmolality, which approached but did not exceed the clinical threshold for hypernatremia (≥145 mmol/L)⁶³. Because sodium excretion did not differ between age groups or hydration conditions, this pattern is consistent with a lower volume of total body water relative to sodium content in OF, rather than increased sodium retention. These findings suggest that OF maintain normal sodium handling but may operate closer to the upper limit of osmotic tolerance, increasing their vulnerability to hyperosmolar stress with further fluid loss or more severe hypohydration.

Renal water-conservation mechanisms appear to be preserved with age, as indicated by comparable declines in CH₂O across groups. This supports an intact capacity for water reabsorption despite age-related changes in body water distribution. However, prior work in males reported a paradoxical increase in CH₂O during exercise-induced hypohydration compared with baseline hydration²⁰. This discrepancy highlights potential sex- or context-specific differences in renal water handling. Further studies should investigate how physiological stressors such as exercise or heat interact with age and sex to influence renal water conservation.

Renal Function and Acute Kidney Injury Markers

Renal filtration capacity remained stable across age groups, as indicated by unchanged eGFR, serum creatinine, and urinary creatinine. Both YF and OF reduced their urine flow rate appropriately during mild hypohydration, demonstrating preserved renal water conservation and excretion mechanisms.

Excretion of urinary cystatin C, typically reabsorbed and degraded in the proximal tubules, increased in YF but not OF. Urinary cystatin C concentrations were well below the threshold associated with AKI in intensive care populations (120 ng/mL), suggesting a more transient tubular adjustment rather than pathology⁶⁴. The significant increase in YF during mild hypohydration may be due to greater sensitivity to fluid shifts in YF, whereas unchanged values in OF may reflect a reduced responsiveness rather than impairment.

Excretion of urinary nephrin, a marker of glomerular permeability⁶⁵, significantly increased during mild hypohydration in both age groups. However, nephrin concentration remained within the normal physiological range (85 to 850 ng/mL)⁶⁶, indicating only mild glomerular stress rather than injury. Excretion of urinary NGAL, mainly expressed in the loop of Henle and the collecting ducts⁶⁷, decreased after mild hypohydration in this study. In contrast, previous studies reported increases in NGAL concentration during hypohydration associated with illness, strenuous exercise, or more severe fluid restriction, where tubular stress or injury is more likely to occur^16,68,69. The decrease in NGAL excretion observed in our study suggests a lack of tubular injury. This downward pattern is likely driven by the reduction in urine flow rate rather than altered synthesis.

Finally, IGFBP7 and TIMP-2 excretion, which reflect tubular stress⁷⁰, increased in both age groups. On average, the product of IGFBP7*TIMP-2 concentrations exceeded the threshold of 0.3 (ng/mL)²/1000 linked to increased clinical AKI risk³⁵. This pattern of increased IGFBP7*TIMP-2 concentration without increases in NGAL concentration indicates a mild and reversible response to kidney stress, rather than structural kidney injury¹⁶. These results collectively suggest that mild hypohydration triggers early stress signaling in renal tissue in females, providing insight into the adaptive mechanisms that occur during hydration challenges.

Limitations

In this study, we evaluated the impact of short-term mild hypohydration on kidney function in female adults. Consequently, our findings are limited in that they are not generalizable to moderate or severe hypohydration, or long-term or repeated exposures to hypohydration. Because recovery following rehydration was not assessed, the reversibility of these responses could not be confirmed. Future research should determine how these markers change with the resumption of adequate fluid intake. Additionally, because only female participants were included, we cannot directly compare sex differences in hypohydration responses. Thus, any interpretation regarding sex differences should be made cautiously. Future studies should include both sexes to confirm whether age-related patterns differ by sex.

Given that the current literature is limited to young male populations, future studies should aim to explore a broader range of hypohydration intensities and durations, as well as include both female and male participants. Despite our modest sample size, we observed distinct changes in kidney function markers. Our cohort’s limited diversity in terms of racial, ethnic, and health status backgrounds limits the broader generalizability of our results. Future studies should prioritize the inclusion of diverse populations and populations with comorbidities who are at an increased risk of hypohydration and its complications.

Additionally, some kidney biomarkers used to assess kidney stress and kidney injury lack well-established thresholds for healthy populations, making it difficult to interpret their clinical significance in this context. Future research is needed to determine their sensitivity, specificity, and optimal thresholds, and would contribute to an improved early detection and evaluation of kidney stress, AKI, and CKD.

Conclusion

Water deprivation resulted in subclinical, mild kidney stress in YF and OF. OF exhibited higher baseline plasma osmolality, but hormonal (RAAS) and renal responses to mild hypohydration were largely preserved. Urinary AKI biomarker excretion indicated early renal stress responses at both glomerular and tubular levels, highlighting the importance of sensitive AKI biomarkers like IGFBP7*TIMP-2 in detecting mild renal stress to prevent renal injury.

New & Noteworthy.

This is the first study to assess the effects of mild hypohydration on acute kidney injury biomarkers in young and older female adults. Mild hypohydration increased acute kidney injury biomarker concentrations in both age groups, with no observed age-related differences, contrasting with male studies that show an age-related decline. These findings highlight the need for targeted research on potential sex-related differences in kidney function decline with aging.

Data Sharing Statement

Data are available upon reasonable request to the principal investigator after institutional data transfer approvals.