Outdoor workers frequently exposed to heat stress are at risk of developing chronic kidney disease (CKD). This kidney pathology is not associated with traditional risk factors (e.g., hypertension, diabetes, obesity), but is elicited following a single or recurrent episodes of acute kidney injury (AKI) caused by exertional hyperthermia and/or dehydration stemming from occupational heat exposures typically experienced in the agriculture, construction, and manufacturing sectors (1, 2). As such, this CKD is of non-traditional origin (CKDnt) or is often referred to as heat stress nephropathy (1, 3–5). Several field (6, 7) and laboratory-based studies (8–13) have identified that the risk of developing AKI is largely modified by the magnitude of heat strain and/or dehydration, duration of exposure, and nature of the physical work (1, 2, 14).

Federal and non-federal agencies issue exposure limit recommendations that aim to mitigate hyperthermia and ultimately, heat-related injury/illness. In the US, the National Institute for Occupational Safety and Health (NIOSH) heat stress recommendations (15) are designed to prevent peak core temperature during a workday from exceeding 38.0°C in unacclimatized workers. We have previously reported that adherence to the NIOSH heat stress recommendations were largely effective at accomplishing this objective when work-rest ratios were prescribed on the basis of wet-bulb globe temperature (WBGT) at a fixed work intensity (i.e., metabolic heat production) (11). Importantly, this translated to a maintenance of glomerular filtration rate (i.e., kidney function) (16, 17) and attenuated AKI risk compared to a work-rest matched positive control scenario that violated the recommendations (11). Whether these observations remain consistent across work intensities has rarely been explored. In fact, to our knowledge, only one study has examined the implications of work intensity on AKI risk during physical work in the heat. Houck et al. (18) reported that high-intensity intermittent work in the heat exacerbated AKI risk compared to work-matched, moderate intensity continuous work. That said, whether this conclusion translates to occupational scenarios is unclear. For example, the prescribed work in Houch et al. (18) was likely not reflective of manual labor such that work intensity oscillated between 80% and 30% of each participant’s peak oxygen uptake every 5 minutes. Therefore, the purpose of the present study was to examine AKI risk during simulated occupational heat stress that adhered to the NIOSH heat stress recommendations by fixing the work-rest ratio and adjusting the work intensity (metabolic heat production) as a function of WBGT. Importantly, we examined the effects of work intensity under steady-state conditions. To do so, we compared two scenarios that were compliant with the NIOSH recommendations to a positive control scenario that violated the recommendations by exceeding the allowable work intensity for a given WBGT at a given work-rest ratio. We tested the hypothesis that violating the recommendations would augment hyperthermia resulting in elevated AKI risk during simulated occupational heat stress.

Core temperature and hemodynamic responses to simulated occupational heat stress are presented in Figure 2A-F. Briefly, core temperature increased throughout the exposure in all trials (p<0.0001), but non-compliance with the recommendations (NC_high) resulted in a greater increase in core temperature starting at minute 30 (p≤0.0059). This resulted in a greater mean (C_high : 37.6±0.2°C; C_low : 37.5±0.3°C; NC_high : 37.8±0.2°C; p=0.0009; d: 0.99) and peak (C_high : 38.0±0.3°C; C_low : 37.8±0.4°C; NC_high : 38.3±0.4°C; p=0.0002; d: 1.25) core temperature in the non-compliant vs. both compliant trials. Perceptual data are presented in Figure 3A-E. Indices of hydration and fluid intake are presented in Figure 4A-D. Percentage change in body weight was not different between the non-compliant and compliant trials (C_high : −0.1±0.8%; C_low : 0.1±1.0%; NC_high : 0.0±1.2%; p=0.6784) despite differences in whole-body sweat rate (C_high : 0.4±0.1 L/hour; C_low : 0.3±0.1 L/hour; NC_high : 0.5±0.1 L/hour; p<0.0001 d: 0.99). The augmented fluid loses via sweating were offset by increased ad libitum fluid intake (p=0.0016).

Fluid and electrolytes are presented in Table 1 and creatinine clearance and indices of fluid conservation are presented in Table 2. Briefly, plasma volume increased from preexposure in the C_high trial (p≤0.0036), but did not change in the C_low (p≥0.0513) and NC_high (p≥0.8125) trials. Serum creatinine increased from pre- to postexposure in all trials (C_high : +0.06±0.05 mg/dL; C_low : 0.05±0.04 mg/dL; NC_high : 0.11±0.06 mg/dL; p≤0.0213; d: 0.90) and remained elevated at recovery in the NC_high trial (+0.09±0.05 mg/dL; p<0.0001). Creatinine clearance was reduced from pre- to postexposure in the NC_high trial (p=0.0053; d: 0.86). Urine flow rate was not changed from pre- to postexposure and following recovery or different between trials (Time × Trial: p=0.5891), but other indices of fluid conservation (i.e., osmolar clearance, fractional excretion of sodium and potassium) were reduced from preexposure in all trials (p≤0.0104).

Serum and urinary biomarkers of oxidative stress, inflammation, and acute kidney injury risk are presented in Table 3 and Figure 4A-F. Serum TRX-1, IGFBP7, and TIMP-2 did not change over time or differ between trials (p≥0.1657). In contrast, serum MCP-1 increased from pre- to postexposure in the NC_high trial (p<0.0468; d: 0.77), was higher than C_low at postexposure (p=0.0017; d: 0.71) and C_high at recovery (p=0.0332; d: 0.67). All urinary biomarkers were elevated from pre- to postexposure in each trial (p≤0.0009) when normalized to urine flow rate. However, there were only main effects of time for urinary MCP-1, IGFBP7, and TIMP-2 when normalized to urine specific gravity (p≤0.0461), but posthoc analyses did not reveal any within trial changes. There were no changes in the fractional excretion of TRX-1, MCP-1, IGFBP7, or TIMP-2 (p≥0.0673). Finally, acute kidney injury risk ([IGFBP7•TIMP-2]_USG) is presented in Figure 5. There was a main effect of time (p=0.0451), but no differences between trials over time (Figure 5A) or as a peak change (Figure 5B).

The purpose of the present study was to examine the influence of work intensity (i.e., H_prod) on AKI risk and indices of kidney function/fluid conservation during simulated occupational heat stress that was either compliant (C_high and C_low) or non-compliant (NC_high) with the NIOSH heat stress recommendations. We tested the hypothesis that exceeding the allowable work intensity at a given WBGT and fixed work-rest ratio would augment AKI risk. In the present study, non-compliance with the NIOSH recommendations (NC_high) resulted in a marginally higher magnitude of hyperthermia (Figure 2A-D) compared to NIOSH compliant trials. In contrast with our hypothesis, this resulted in no difference in AKI risk between the three experimental trials (Figures 5A-H, 6A-B). In disagreement with the first study (11, 16) as part of our registered clinical trial (NCT04767347), in the present study we observed that modest non-compliance with the NIOSH recommendations (i.e., exceeding allowable H_prod by ~120 W or environmental limits by ~4°C WBGT at 30 minutes work-rest per hour) did not augment AKI risk quantified via urinary biomarkers. This highlights subtle differences with regards to the severity of non-compliance and when rest is issued in compensable (present study) versus uncompensable conditions (11) (discussed in more detail below).

The NIOSH heat stress recommendations aim to keep peak core temperature below 38.0°C for non-acclimated workers (15). ~62% (15/24) of compliant trials maintained a peak core temperature below 38.0°C, whereas non-compliance resulted in 100% of trials exceeding a peak core temperature of 38.0°C (Figure 2B). In the current study, this was paralleled by a greater mean core temperature and time above 38.0°C in the non-compliant versus compliant trials (Figures 2C and 2D). These findings are consistent with our previous study (11, 16). Also in agreement with our previous study (17), hydration status was maintained in both compliant and non-compliant trials with ad libitum fluid intake despite differences in environmental conditions, work intensity (metabolic heat production), and sweat rates (Figures 4A-D). This is important to note, as AKI risk and kidney function are modulated by hydration status and fluid intake (2, 10). Thus, any anticipated differences in kidney function and/or AKI risk between compliant and non-compliant trials (although not observed in the present study) would have been reflective of differences in hyperthermia, but unlikely total work (11) completed during the simulated occupational heat stress.

Urine specific gravity and serum osmolality did not change nor was different between trials (Table 1). However, plasma volume was expanded in all trials (i.e., a main effect of trial was observed). Serum creatinine was elevated from pre- to postexposure in each trial and remained elevated following recovery during the C_low and NC_high trials. Furthermore, serum creatinine was higher in the non-compliant trial compared to C_low. Our previous study revealed that the change in serum creatinine is related to peak core temperature (11), which likely explains the higher serum creatinine in the non-compliant trial in the present study given the greater hyperthermia. An interesting finding in the present study is that several indices of fluid conservation remained unchanged from pre- to postexposure and following a one-hour recovery (Table 2). Urine flow rate and free water clearance were unchanged in all trials, which is likely due to the magnitude of fluid intake during the four-hour exposure. This contrasts with our previous studies that have reliably shown a reduction in urine flow rate and free water clearance reflective of fluid conservation during heat strain and/or dehydration (9–13, 16, 34). However, the fractional excretion of sodium and osmolar clearance was reduced from pre- to postexposure in all trials and remained reduced from preexposure following a one-hour recovery, but was not different between trials despite differences in heat strain (Figure 2A-D).

Creatinine clearance (our marker of glomerular filtration rate) was unchanged in the compliant trials (Table 2), which is consistent with the compliant trials in our previous study (16). However, creatinine clearance was reduced from preexposure in the NC_high trial but was not different from the two compliant trials at any timepoint (Table 2). This is an interesting finding given that non-compliance in the present study only resulted in a modest difference in heat strain and no difference in hydration status compared to the compliant trials. However, this is consistent with previous work from our lab (9, 10) despite showing in our previous NIOSH recommendation study that creatinine clearance is maintained even under non-compliant conditions as long as fluids are freely available (11).

Perhaps the most interesting finding in the present study, AKI risk (quantified via [IGFBP7•TIMP-2]) was not different between the compliant and non-compliant trials (Figures 5 and 6). [IGFBP7•TIMP-2]_USG increased from pre- to postexposure and returned to preexposure levels following a one-hour recovery in all trials. Both IGFBP7 (a marker of proximal tubular injury) and TIMP-2 (a marker of distal tubular injury) increased similarly indicating the both the proximal and distal tubules were involved in the elevation in [IGFBP7•TIMP-2]_USG (2). Interestingly, we observed more muted responses when normalizing to concentration (urine specific gravity), which only revealed a time main effect, compared to when normalizing to urine flow rate in which all trials had a significant increase from pre- to postexposure. Combined with the findings from our previous study (11), there appears to be a threshold in core temperature from which AKI risk is augmented under non-compliant scenarios. We previously reported augmented AKI risk in participants that reached/exceeded 38.0°C. However, it is important to note that this was due to a ~+1.0°C difference in peak core temperature between the group of participants that did or did not reach/exceed 38.0°C. In the present study, the difference in peak core temperature between compliant and non-compliant trials was ~+0.3°C. Thus, the differences in work intensity (metabolic heat production) had marginal impact on our markers of AKI risk. Notably, while there was a statistical difference in workload between the high and low intensity trials (high: 5.4±0.7 METs vs. low: 3.5±0.9 METs; p<0.001), both were relatively moderate intensity workloads (35). Thus, greater differences in workload may be necessary to elicit differential elevations in AKI risk. This is supported by previous field work by Moyce et al. in which it was reported that AKI (quantified by KDIGO criteria for elevations in serum creatinine) across a single day (average daily maximum WBGT: ~30°C) in agricultural workers in California was associated with increased workload, but AKI risk was particularly marked when the workload exceeded moderate levels (36). Further, it is important to note that duration of exposure is an integral component to AKI risk when completing physical work in the heat (12–14). The combination of identical exposure times and small differences in core temperature responses among trials resulted in similar total heat loads among trials. Additionally, core temperature across each 30-minute rest period decreased, which was not observed in our previous investigation in which rest occurred in a hotter environment (WBGT: ~36°C (11)). These findings highlight the importance of rest to reduce heat strain, particularly when exceeding the heat stress recommendations for work intensity, findings that support interventions put in place in the field that reduce the renal burden of heat stress (37). Together, we speculate that there is a (yet undefined) core temperature threshold above 38.0°C, particularly when sustained, upon which AKI risk disproportionately increases. It is also important to note that these findings are representative of the primary effects of hyperthermia, as hydration status was well maintained via ad libitum drinking.

In the present study, we measured urinary markers of inflammation and oxidative stress. In all trials, regardless of NIOSH recommendation compliance, urinary markers of inflammation (MCP-1) and oxidative stress (TRX-1) were elevated immediately postexposure but returned to preexposure levels at recovery (Figures 5A-D). Additionally, we measured all markers in the serum (Table 3) to capture any changes in systemic inflammation (serum MCP-1) and/or oxidative stress (serum TRX-1). Serum TRX-1 was unchanged in all trials, while serum MCP-1 increased from pre- to postexposure in the NC_high trial and was higher than the C_low trial at postexposure and C_high trial at recovery. This could indicate that the greater heat strain, albeit marginal, resulted in an elevation in systemic inflammation under non-complaint conditions but this did not seemingly modify markers of AKI risk. These findings suggest that the elevated AKI risk in the present study could be contributed to by the systemic inflammatory response to hyperthermia, as is known to occur with exertional heat injury (38). This speculation is supported by our fractional excretion of MCP-1 and TRX-1 data, which were unchanged, indicating that the increases in urinary MCP-1 and TRX-1 likely reflect increases in systemic (serum) MCP-1 and TRX-1. We would have anticipated an increase in the fractional excretion of these markers if they were endogenous to the kidneys (32). Future work is required to elucidate the role of systemic inflammation in the pathophysiology of exertional heat injury, particularly as it pertains to the development of AKI.