Legal Performance-enhancing Drugs Alter Course and Treatment of Rhabdomyolysis-induced Acute Kidney Injury

ABSTRACT

Introduction

Rhabdomyolysis-induced acute kidney injury (RIAKI) can interrupt physical training and increase mortality in injured warfighters. The legal performance-enhancing drugs caffeine and ibuprofen, which can cause renal injury, are widely used by service members. Whether caffeine or ibuprofen affects RIAKI is unknown. Cilastatin treatment was recently identified as an experimental treatment to prevent RIAKI at injury. To determine potential interacting factors in RIAKI treatment, we test the hypothesis that caffeine and ibuprofen worsen RIAKI and interfere with treatment.

Materials and Methods

In mice, RIAKI was induced by glycerol intramuscular injection. Simultaneously, mice received caffeine (3 mg/kg), ibuprofen (10 mg/kg), or vehicle. A second cohort received volume resuscitation (PlasmaLyte, 20 mL/kg) in addition to caffeine or ibuprofen. In a third cohort, cilastatin (200 mg/kg) was administered concurrently with drug and glycerol administration. Glomerular filtration rate (GFR), blood urea nitrogen (BUN), urine output (UOP), renal pathology, and renal immunofluorescence for kidney injury molecule 1 were quantified after 24 hours.

Results

Caffeine did not worsen RIAKI; although BUN was modestly increased by caffeine administration, 24-hour GFR, UOP, and renal histopathology were similar between vehicle-treated, caffeine-treated, and caffeine + PlasmaLyte–treated mice. Ibuprofen administration greatly worsened RIAKI (GFR 14.3 ± 19.5 vs. 577.4 ± 454.6 µL/min/100 g in control, UOP 0.5 ± 0.4 in ibuprofen-treated mice vs. 2.7 ± 1.7 mL/24 h in control, and BUN 264 ± 201 in ibuprofen-treated mice vs. 66 ± 21 mg/dL in control, P < .05 for all); PlasmaLyte treatment did not reverse this effect. Cilastatin with or without PlasmaLyte did not reverse the deleterious effect of ibuprofen in RIAKI.

Conclusions

Caffeine does not worsen RIAKI. The widely used performance-enhancing drug ibuprofen greatly worsens RIAKI in mice. Standard or experimental treatment of RIAKI including the addition of cilastatin to standard resuscitation is ineffective in mice with RIAKI exacerbated by ibuprofen. These findings may have clinical implications for the current therapy of RIAKI and for translational studies of novel treatment.

INTRODUCTION

Destruction of skeletal muscle by crush, blast, burn, excessive activity, or muscle-toxic drug administration causes the release of muscle proteins into the systemic circulation, which are toxic to the kidney, causing crush syndrome.¹ A primary cause of death because of crush syndrome is rhabdomyolysis-induced acute kidney injury (RIAKI).² Loss of renal clearance of potassium, combined with the release of potassium from injured muscle, leads to hyperkalemia and cardiac arrest. RIAKI is common in evacuated warfighters. In one study, 34% of patients with explosive, penetrating, or blunt injury developed acute kidney injury (AKI), while 25% developed rhabdomyolysis.³ Traumatic AKI is primarily an early complication, occurring within 3 days of injury and closely associates with mortality, increasing the risk of death 4-fold.⁴ Thus, RIAKI is a common, lethal complication of battlefield injury, which occurs during the expected course of prolonged field care. As RIAKI and AKI are expected to be more common in future conflicts, novel management and treatment are imperative.⁵ In addition, RIAKI is an important cause of civilian AKI—incidence increased 10-fold in the last decade among U.S. civilians, likely because of popular fitness training regimes⁶; training-induced RIAKI thus impacts both military recruits and civilian athletes. RIAKI is caused by tubular epithelial cell uptake of muscle myoglobin, which is dependent on the endocytic receptor megalin. Genetic interference with megalin completely prevents RIAKI, and treatment with the megalin inhibitor cilastatin greatly ameliorates RIAKI.⁷ To pursue early translation of this important discovery, it is critical to understand potential real-world and combat-relevant limitations. This report describes an investigation into a critical potential limitation and the effect of legal, performance-enhancing drugs on RIAKI and its treatment.

Acute kidney injury is understood to be exacerbated or accelerated by legal performance-enhancing drugs commonly used by warfighters. For example, 50 to 70% of deployed active duty service members report using caffeine-based, performance-enhancing, or weight-loss supplements.^8,9 Caffeine supplements cause dehydration¹⁰ and can cause RIAKI when used in excess.¹¹ Similarly, ibuprofen, a non-steroidal anti-inflammatory drug (NSAID) with analgesic properties, is widely used by service members¹²; ibuprofen and other NSAIDS are well understood to cause and/or exacerbate AKI.¹³ As some accelerants will be present in many injured warfighters who might be eligible for cilastatin treatment, it is critical to understand whether mitigating the effect of these drugs is necessary for cilastatin-mediated protection from myoglobin-induced kidney injury. Therefore, this investigation tests the hypothesis that caffeine and NSAIDs worsen RIAKI and interfere with both conventional and experimental cilastatin treatment.

METHODS

All animal procedures were approved by the Portland VA Medical Center Institutional Animal Care and Use Committee (protocol no. 4514).

Mice

Eight-week-old male C57BL/6J mice were purchased from Jackson Laboratories (Bar Harbor, ME) and were housed in standard conditions for 7 days before experimentation.

Glycerol Model of Rhabdomyolysis in Mice

Glycerol injection to the hindlimb of mice reliably induces myocyte necrolysis and elevates plasma myoglobin, leading to RIAKI; this model was extensively described in the 1960s¹⁴ and is considered rigorous, repeatable, and well characterized. Mice were deprived of water for 4 hours before injection. Under brief inhalant isoflurane anesthesia (1.5 L/min flow rate and 2.5% isoflurane), the hindlimb was depilated, and the major anterior muscle group of the hindlimb was injected with a glass syringe and a 26-gauge needle (Hamilton [Reno, NV] 7638-01, 7804-03 [custom length 2.54cm, 12°, point 4]) with a total dose of 8 mL/kg of sterile-filtered 50% glycerol in water solution, half of the dose in each leg (total volume <250 µL).

Experimental Groups

All mice were subjected to glycerol injection. Mice were randomly assigned to receive one of the two legal performance-enhancing drugs (ibuprofen or caffeine) alone, ibuprofen or caffeine with standard treatment for RIAKI, or standard treatment for RIAKI alone. After assigned treatments, mice were housed in metabolic cages (Hatteras Instruments, Cary, NC) for 24 hours for urine collection before glomerular filtration rate (GFR) and perfusion procedures, as outlined later, were conducted. Urine was collected for 24 hours, the total 24-hour volume was recorded, and samples were frozen at −80 °C.

Preparation and Administration of Legal Performance-enhancing Drugs With and Without Standard Treatment

Ibuprofen and caffeine were purchased from Sigma-Aldrich (St. Louis, MO). Doses of caffeine and ibuprofen were selected to be relevant to use by warfighters.^8,9 The caffeine dose was 3 mg/kg, which is midrange for dietary supplements, approximately equivalent to 1.5 cups of strong coffee. The ibuprofen dose was 10 mg/kg, approximating the commonly used 600 to 800 mg human dose. Drug injections were performed immediately before intramuscular glycerol injection. Caffeine was dissolved in PlasmaLyte (2B2543Q, pH 7.4, Baxter International, Deerfield, IL) at 3 or 0.15 mg/mL. Since PlasmaLyte was the vehicle for caffeine administration, in experiments testing the effect of caffeine on RIAKI, mice were injected retroorbitally with vehicle (PlasmaLyte, 1 mL/kg), caffeine alone (3 mg/kg in PlasmaLyte 1 mL/kg, delivered as 1 μL/g of the 3 mg/mL caffeine solution), or caffeine + standard treatment (caffeine 3 mg/kg in 20 μL/g of the 0.15 mg/mL caffeine solution, delivering PlasmaLyte 20 mL/kg). Because of the poor water solubility, ibuprofen (10 mg/kg) was mixed in corn oil and injected intraperitoneally. The vehicle group received an identical volume of corn oil intraperitoneally and PlasmaLyte 1 mL/kg; the third group received ibuprofen intraperitoneally with standard treatment, PlasmaLyte 20 mL/kg. In experiments testing cilastatin treatment, cilastatin sodium was purchased from Sigma-Aldrich and administered (200 mg/kg) by retroorbital injection immediately after glycerol injection.

Renal Function Quantification

The GFR was measured using fluorescein isothiocyanate (FITC)-fluorimetry (MediBeacon, St. Louis, MO). A data-logging transdermal FITC-fluorimeter (MediBeacon 9027300) was attached with adhesive to the back of the mouse after the skin was depilated. FITC-sinistrin (MediBeacon FTC-FS001, 35 mg/mL, 50 µL) was retroorbitally injected. The FITC-fluorimeter records three skin fluorescence readings per second. After 2 hours of data recording, FITC-sinistrin clearance (GFR) was derived from fluorescence data using a single-compartment model. Blood urea nitrogen (BUN) was quantified using a commercially available kit (EIABUN, Invitrogen, Carlsbad, CA) from blood drawn via cardiac puncture of the left ventricular apex at the time of euthanasia.

Kidney Sectioning, Histologic, and Immunofluorescence Measures of Kidney Injury

At the time of euthanasia, mouse kidneys were perfused in situ via the cardiac left ventricular apex and then removed, paraffin embedded, and cut into 6 μm sections. Kidney sections were stained with periodic acid–Schiff (PAS) stain and evaluated for morphology by a treatment-blind renal pathologist (N.K.A.). Immunofluorescence for kidney injury molecule 1 (KIM-1, 1:100, AF1817, R&D Systems, Minneapolis, MN; AlexaFluor 647, 1:1,000, A21447, Invitrogen) was quantitatively evaluated using automated unbiased stereology on scanned slides as previously described.¹⁵ Briefly, random superimposed grids on cut sections were used to create a ratio of positive grid points to the area, which translates into the volume of positive tissues with reference to the overall tissue volume. This is repeated with KIM-1-positive cells, and the results are expressed as the volume fraction of KIM-1-positive tubular epithelium per total volume (V_KIM-1/V_Kidney).

Statistical Analysis

Means are presented in all figures ±  SEM. Experimental numbers were determined by an a priori power analysis performed using R (package stats, power.anova.test). Statistical analysis was performed using grouped analysis in Prism (v9.4, San Diego, CA). One-way analysis of variance was performed with Dunnett’s T3 test for multiple comparisons. P-values <.05 were considered statistically significant.

DISCUSSION

The most important finding of this study is that the widely used performance-enhancing drug ibuprofen greatly worsens RIAKI in mice. Administration of ibuprofen in a by-weight similar dose to that used in humans at the same time as induction of rhabdomyolysis resulted in drastic worsening of AKI as indicated by GFR, BUN, and UOP 24 hours later. Moreover, the standard treatment for RIAKI was less effective in the presence of ibuprofen. Histologic features of RIAKI with concomitant administration of ibuprofen suggested additional, perhaps mechanistically distinct, renal injury (hyaline droplets are not seen in RIAKI with vehicle). The magnitude of the increase in functional injury conferred by ibuprofen administration further suggests an additional mechanism. As NSAIDs including ibuprofen can cause AKI, this additional effect might be considered unsurprising; however, a single dose of ibuprofen would not be expected to cause this degree of AKI in isolation.¹⁶ Therefore, this finding bears further examination. A secondary, but important, clinical implication is that muscle pain induced by exercise is often treated with NSAIDs at precisely the time at which excessive exercise may induce RIAKI.^17–20 The possibility that exercise-induced RIAKI is exacerbated by NSAID use deserves further examination as treatment with alternative analgesics might prevent or ameliorate RIAKI in this setting.

A second finding of this study is that a relevant and commonly used dose of caffeine does not worsen RIAKI. GFR, UOP, and renal histology were not altered in RIAKI by caffeine administration, indicating similar renal function and histologic injury. Caffeine did induce an increase in 24-hour BUN. This likely reflects the mild diuretic effect of caffeine. This hypothesis is supported by the partial reversal of increased BUN in animals receiving additional fluid resuscitation. Together these findings may reinforce the importance of the initial volume resuscitation in warfighters with crush injury.

Lastly, cilastatin therapy does not reverse RIAKI when RIAKI is compounded by ibuprofen administration. This finding may have important translational implications, as ibuprofen is commonly used in settings in which there is a high risk for RIAKI, such as fitness training, military training, and warfighting. Cilastatin is a heptanoic acid–coupled cysteine derivative, originally developed in the 1980s by Merck to inhibit renal dipeptidase-induced metabolism of the antibiotic imipenem; it was approved by the Food and Drug Administration in combination with imipenem in 1989. In the last decade, increasing data suggest that cilastatin ameliorates several forms of AKI,^7,21–29 through several mechanisms, including inhibition of Fas-ligand binding–dependent apoptosis, alteration of lipid rafts, inhibition of proinflammatory effects of renal dipeptidase, and inhibition of renal megalin. Cilastatin administration is highly protective in murine RIAKI, an effect that was megalin dependent and associated with greatly increased urinary myoglobin clearance.⁷ The loss of this effect in the presence of ibuprofen may have mechanistic implications. Ibuprofen has several relevant effects on renal physiology—inhibition of cyclooxygenase conversion of arachidonic acid could interact with the cilastatin target, renal dipeptidase, which regulates the leukotriene arachidonic acid degradation pathway.³⁰ Alternatively, the direct reduction of GFR by prostaglandin-mediated afferent arteriolar vasoconstriction could reduce cilastatin delivery or (by increasing sodium retention) worsen tubular cell edema already present in RIAKI, further reducing GFR.³¹ Additional mechanistic studies are required to understand this likely clinically important finding.

Our study has important limitations. Experiments were performed in a mouse model, and the results may not extend to humans, including humans functioning in extreme conditions. Ibuprofen dosing in humans varies widely, and the dose used in these studies is relatively high, although well within the Food and Drug Administration–approved range when considered on a weight-equivalent basis.

In conclusion, common performance-enhancing drugs worsen RIAKI, ibuprofen especially so. Both standard and investigative-specific treatments of RIAKI are impacted by concomitant ibuprofen use.

RESULTS

A summary of our findings is outlined in Table I.

TABLE I.

Sample Groups with Raw Data

	GFR (μL/min/100 g)	UOP (mL/24 h)	BUN (mg/dL)	Urine myoglobin (ng/mL)	Histopathology score	VKIM-1/VKidney
Vehicle (n = 12, s = 12)	717 ± 74	3.0 ± 0.4	68 ± 7	1698 ± 433	19 ± 2	N/A
Caffeine (n = 12, s = 11)	420 ± 119	2.1 ± 0.3	148 ± 24**	1809 ± 314	23 ± 4	N/A
Caffeine + PlasmaLyte (n = 12, s = 12)	562 ± 105	3.0 ± 0.3	108 ± 20	1659 ± 347	30 ± 6	N/A
Vehicle (n = 10, s = 9)	650 ± 427	2.7 ± 0.5	66 ± 8	1703 ± 373	19 ± 4	0.036 ± 0.004
Ibuprofen (n = 10, s = 10)	225 ± 460*	0.5 ± 0.1*,#	263 ± 67*	1398 ± 262	21 ± 6	0.031 ± 0.003
Ibuprofen + PlasmaLyte (n = 10, s = 10)	290 ± 400	2.4 ± 0.6	189 ± 38	916 ± 231	28 ± 5	0.042 ± 0.004
Ibuprofen + vehicle (n = 12, s = 12)	396 ± 162	1.2 ± 0.4	134 ± 58	N/A	8 ± 5	0.039 + 0.004
Ibuprofen + cilastatin (n = 8, s = 6)	105 ± 88	1.4 ± 0.4	110 ± 45	N/A	34 ± 17	0.045± 0.005
Ibuprofen + cilastatin + PlasmaLyte (n = 10, s = 9)	265 ± 109	1.5 ± 0.8	84 ± 47	N/A	6 ± 3	0.038± 0.003

Variation was reported as SEM. *P < .05, **P < .01 relative to vehicle, #P < .05 relative to PlasmaLyte resuscitation. n: number of mice at the start of the experiment; s: surviving mice in analysis. Unbiased stereology values are unavailable for caffeine-treated groups as there was no evidence of GFR alteration and tubular damage to pursue. Myoglobin values are unavailable for ibuprofen/cilastatin experiments: Cilastatin did not ameliorate AKI and urine myoglobin was not altered by caffeine or ibuprofen.

Abbreviations: AKI, acute kidney injury; BUN, blood urea nitrogen; GFR, glomerular filtration rate; KIM-1, kidney injury molecule 1; UOP, urine output; N/A, not applicable.

Effect of Caffeine on RIAKI

Following glycerol injection, mice received standard treatment alone, caffeine alone, or caffeine with standard treatment. Glycerol injection and standard treatment still resulted in AKI in all mice, reducing GFR by 50% (1,307 ± 72 in sham and 717 ± 74 μL/min/100 g bodyweight in glycerol-injected mice, P < .001), which was not worsened by caffeine, either with or without standard treatment (420 ± 119 in caffeine-treated and 562 ± 105 μL/min/100 g in caffeine- and PlasmaLyte-treated mice) (Fig. 1A). Caffeine treatment raised BUN by almost 2-fold; this was not ameliorated by PlasmaLyte infusion (BUN 108 ± 64 in caffeine-treated mice vs. 68 ± 25 mg/dL in respective control, P < .01, Fig. 1B). Neither urine output (UOP) nor urinary myoglobin content was affected by caffeine or subsequent treatment with PlasmaLyte (Fig. 1C, D). Periodic acid–Schiff staining of renal cross-sections confirmed tubular injury: Caffeine-treated animals demonstrated patchy tubular necrosis, attenuation of tubular cytoplasm, and tubular casts, but there were no significant quantitative or qualitative differences between treatment groups (Fig. 1E). In conclusion, caffeine only minimally alters RIAKI.

FIGURE 1.

Caffeine increases blood urea nitrogen (BUN) but does not affect the glomerular filtration rate (GFR) following rhabdomyolysis-induced acute kidney injury (RIAKI). (A) An experimental model for caffeine and resuscitative treatment following RIAKI. Caffeine did not change GFR (B) but increased BUN (C). Urine output (D) and urine myoglobin concentration (E) remained unchanged. Periodic acid–Schiff (PAS)–stained sections from caffeine-injected animals showed acute tubular injury with patchy tubular necrosis, attenuation of the tubular epithelium, loss of brush border, and focal casts consistent with acute kidney injury (AKI); caffeine did not worsen this condition (F). The quantitative histopathologic scoring by a treatment-blinded renal pathologist (N.K.A.) confirmed observations. Scale bars: 100 μm. Means ± SEM, significance: one-way analysis of variance (ANOVA). **P < .01. ns: not significant.

Effect of Ibuprofen on RIAKI

Unlike caffeine, ibuprofen administration greatly worsened RIAKI, reducing GFR by 98% and causing significant oliguria and elevation of BUN (GFR 14.3 ± 19.5 vs. 577.4 ± 454.6 µL/min/100 g in control, UOP 0.5 ± 0.4 in IBU vs. 2.7 ± 1.7 mL/24 h in control, and BUN 264 ± 201 in ibuprofen-treated mice vs. 66 ± 21 mg/dL in control, P < .05 for all). Treatment with PlasmaLyte ameliorated oliguria but did not alter GFR or BUN (Fig. 2A, B), indicating that standard RIAKI treatment efficacy is mitigated in the presence of ibuprofen (PlasmaLyte treatment: GFR 169.5 ± 190.1 µL/min/100 g, P = .08, compared with ibuprofen alone; BUN 189 ± 120, P = .45, compared with ibuprofen alone; and UOP 2.4 mL/24 h, P = .03, compared with ibuprofen alone; Fig. 2A–C). There was no significant difference in the urinary myoglobin content between groups. Histologically, ibuprofen-receiving animals demonstrated tubular necrosis as seen in previous RIAKI pathology, but in addition, occasional, prominent protein resorption droplets in the deep cortex were observed (Fig. 2E). Although both PAS and KIM-1 (Fig. 2F) staining revealed the severe tubular epithelial injury, there was no evidence of a significant quantitative difference between treatment groups. From these data, it can be inferred that ibuprofen administered in a relevant dose at the time of injury greatly worsens RIAKI.

FIGURE 2.

Ibuprofen greatly worsens rhabdomyolysis-induced acute kidney injury (RIAKI). (A) An experimental model of ibuprofen and resuscitative treatment following RIAKI. Rhabdomyolysis-induced acute kidney injury was more severe in animals pretreated with ibuprofen, reducing glomerular filtration rate (GFR) (B) and increasing blood urea nitrogen (BUN) (C). Ibuprofen worsens oliguria (D) but not urinary myoglobin (E). Damaged tubular epithelial cells were (F) periodic acid–Schiff (PAS) stained and scored and (G) kidney injury molecule 1 (KIM-1) stained and quantified by unbiased stereology. Scale bars: 100 μm. Means ± SEM, significance: one-way analysis of variance (ANOVA). *P < .05. ns: not significant.

Efficacy of Cilastatin Treatment for RIAKI in the Presence of Ibuprofen

As cilastatin treatment is a potential treatment for RIAKI, the efficacy of this treatment for RIAKI alone and with standard fluid resuscitation treatment was tested in the compounded conditions of RIAKI with ibuprofen. Mice treated with cilastatin demonstrated renal function similar to mice treated with vehicle 24 hours after glycerol and ibuprofen injection (105.3 ± 88.1 vs. 396.2 ± 162.0 µL/min/100 g) (Fig. 3A). Accordingly, both BUN and UOP were not different between vehicle- and cilastatin-treated mice (BUN, 133.7 ± 57.5 in vehicle vs. 110 ± 451 mg/dL in cilastatin-treated mice; UOP, 1.2 in vehicle vs. 1.4 mL/24 h in cilastatin-treated mice) (Fig. 3B, C). Resuscitative fluid administered with cilastatin was insufficient to improve the parameters to baseline. Although varying degrees of tubular injury were observed in both PAS-stained (Fig. 3D) and KIM-1-stained (Fig. 3E) sections, particularly with cilastatin combined with PlasmaLyte, cellular necrosis was limited, and no statistical differences were observed when quantifying the effect from histological sections. These results demonstrate that cilastatin may not ameliorate RIAKI in the presence of concomitant ibuprofen.

FIGURE 3.

Resuscitative care of rhabdomyolysis-induced acute kidney injury (RIAKI) is insufficient in ibuprofen-treated mice. (A) An experimental model of ibuprofen, resuscitative, and cilastatin treatment following RIAKI. RIAKI in ibuprofen-treated animals was irreversible with cilastatin with or without resuscitation, with no change in (B) glomerular filtration rate (GFR), (C) blood urea nitrogen (BUN), or (D) urinary output. Cilastatin-treated animals had tubular injury as measured by periodic acid–Schiff (PAS) (E) and kidney injury molecule 1 (KIM-1) (F) staining, consisting of epithelial attenuation with limited overt cellular necrosis. Scale bars: 100 μm. Means ± SEM. ns: not significant.

SUPPLEMENT SPONSORSHIP

This article appears as part of the supplement “Proceedings of the 2022 Military Health System Research Symposium,” sponsored by the Assistant Secretary of Defense for Health Affairs.

FUNDING

This study was supported by a grant from the U.S. Department of Defense (W81XWH-20-1-0196) (to M.P.H.) and by the National Center for Advancing Translational Sciences of the National Institutes of Health (award number TL1TR002371) (to J.F.H.).

CONFLICT OF INTEREST STATEMENT

None declared.

CLINICAL TRIAL REGISTRATION

Does not apply.

INSTITUTIONAL REVIEW BOARD (HUMAN SUBJECTS)

Does not apply.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC)

This study was approved by the VA Portland Health Care System (VAPORHCS) Institutional Animal Care and Use Committee and the VAPORHCS Research & Development Committee.

INDIVIDUAL AUTHOR CONTRIBUTION STATEMENT

M.B.E., M.N.N., A.C.M., A.A.P., T.G., and N.K.A. collected and analyzed the data and drafted the original manuscript. M.P.H. and J.F.H. designed this research and reviewed and edited the manuscript. All authors read and approved the final manuscript.

DATA AVAILABILITY

The data underlying this article will be shared on reasonable request to the corresponding author.

INSTITUTIONAL CLEARANCE

Does not apply.