Acute Kidney Injury in Hospitalized Patients With Exertional Rhabdomyolysis

Key Points

Question  What is the risk of acute kidney injury (AKI) in hospitalized patients with exertional rhabdomyolysis (ERM), and which factors are associated with its development?

Findings  In this cohort study of 200 hospitalized patients with ERM in Northern California, the AKI incidence was 8.5%. No association was observed between serum creatine kinase levels and AKI development; however, preadmission use of nonsteroidal anti-inflammatory drugs or dehydration was associated with a significantly increased risk of developing AKI.

Meaning  These findings suggest that an elevated creatine kinase level is not a reliable indicator of AKI in hospitalized patients with ERM, but preadmission use of nonsteroidal anti-inflammatory drugs or dehydration appear to be associated with AKI.

Importance  An association between serum creatine kinase (CK) levels and the risk of kidney failure in patients with exertional rhabdomyolysis (ERM) has been suggested. However, the actual incidence of AKI in hospitalized patients with ERM along with the contributing cofactors that may increase the risk of AKI have rarely been investigated.

Objectives  To examine the incidence of kidney injury in hospitalized patients with ERM and to identify additional cofactors that might contribute to the development of kidney injury in patients with ERM.

Design, Setting, and Participants  This retrospective cohort study was conducted in a diverse community population of patients 18 years or older with ERM who were hospitalized across Kaiser Permanente Northern California between January 1, 2009, and December 31, 2019. Patients were initially identified through electronic screening for all-cause rhabdomyolysis admissions, followed by manual medical record reviews to verify their eligibility for the study. The diagnosis of AKI and chronic kidney disease (CKD) was determined using KDIGO (Kidney Disease Improving Global Outcomes) criteria and confirmed by medical record review. Data analysis was performed from October 1, 2023, to January 31, 2024.

Exposures  History of strenuous physical exercise before hospitalization for ERM.

Main Outcome and Measures  Development of AKI, CKD, and compartment syndrome and number of deaths.

Results  Among 3790 patients hospitalized for rhabdomyolysis between 2009 and 2019 in Kaiser Permanente Northern California, 200 (mean [SD] age, 30.5 [8.5] years; 145 [72.5%] male) were confirmed to have ERM via medical record review. Seventeen patients (8.5%) developed AKI, none developed CKD, 1 (0.5%) developed compartment syndrome, and there were no fatalities. There was no association between serum CK levels and the risk of AKI. However, the risk of AKI was significantly higher in patients with ERM who used nonsteroidal anti-inflammatory drugs (NSAIDs) before admission (11 of 17 with AKI [64.7%] vs 40 of 183 without AKI [21.9%], P < .001) or experienced dehydration (9 of 183 without AKI [52.9%] vs 9 of 17 with AKI [4.9%], P < .001). This analysis suggests that eliminating preadmission NSAID use and dehydration could reduce the risk of potential AKI in patients with ERM by 92.6% (95% CI, 85.7%-96.1%) in this population.

Conclusions and Relevance  The findings of this cohort study of hospitalized patients with ERM suggest that serum CK elevation alone is insufficient as an indicator of AKI in patients with ERM. Concurrent risk factors, such as NSAID use or dehydration, may be associated with AKI development in patients with ERM.

Rhabdomyolysis, a condition characterized by muscle tissue breakdown, can result in severe complications, such as acute kidney injury (AKI) and chronic kidney disease (CKD), and can be fatal, with mortality rates up to 59%.^1,2 Studies have shown that exertional rhabdomyolysis (ERM), caused by strenuous physical exertion, had a 10-fold increase in incidence in the US from 2000 to 2019³ for unknown reasons, with incidence rates varying from 1 to 31.8 per 100 000 population across different populations.^4,5

Concurrently, research on the outcomes of ERM, both AKI and CKD, and associated risk factors remains scarce. Elevated serum creatine kinase (CK) levels within the range of 5000 to 40 000 U/L (to convert to microkatals per liter, multiply by 0.0167) are reportedly associated with an increased risk of AKI.^2,6,7 However, certain studies have reported no significant association between CK and creatinine levels in ERM.^8,9 A previous study found that the risk of AKI in rhabdomyolysis can exceed 50% and varies by etiology.² However, earlier investigations by Clarkson et al⁹ have suggested a significantly lower risk of AKI in ERM. The risk factors that lead to AKI in association with ERM are largely unknown. In this study, we aim to bridge these gaps by examining both the short-term and long-term outcomes of ERM and identifying predisposing factors to AKI within a broad, diverse cohort in Northern California, with a focus on hospitalized patients who are at higher risk of severe complications.

We used a retrospective cohort design for comparative analyses to identify factors that differentiate patients with AKI from those without AKI in a cohort of adult patients aged 18 years from multiethnic backgrounds¹⁰ who, after presentation at an emergency department (ED), were subsequently admitted with ERM from January 1, 2009, to December 31, 2019, to 21 Kaiser Foundation Hospitals within Kaiser Permanente Northern California, a private, not-for-profit, integrated health system with 4.5 million members. Additional inclusion criteria included having engaged in strenuous physical activity, defined based on patient-reported information as any substantial physical activity occurring within 48 hours before hospitalization, which was verified with meticulous manual medical record review. The study was approved by the Kaiser Permanente Northern California Institutional Review Board with a waiver of the requirement for informed consent because risk was minimal. We adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. Cases were identified using a combination of International Classification of Diseases, Ninth Revision (ICD-9) and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) diagnosis codes to query electronic medical records. In our system, acute outpatient conditions are referred to the ED before hospital admission. We used regular expressions in SAS software, version 9.4 (SAS Institute Inc) and a natural language processing tool to identify clinical notes with key exercise-related phrases. Search terms combined exercise severity (eg, intense and extreme) and types (eg, running and weightlifting). We used ICD-9 and ICD-10 codes instead of CK levels to define ERM due to the lack of a universally accepted CK threshold, higher CK levels associated with AKI, and variability in CK levels by ethnicity. All cases identified for inclusion received manual medical record review and were verified for eligibility, history of dark urine, dehydration, and nonsteroidal anti-inflammatory drugs (NSAIDs) and acetaminophen exposure before hospitalization. Dark urine was defined based on the documentation of discolored urine, such as dark, brown, cola-colored, or red. Dehydration was defined based on the documentation of any findings, such as dehydration, nausea, vomiting, diarrhea, presyncope, and orthostatic hypotension, in the medical records, extracted through a manual review of all cases. Manual medical record reviews showed high interrater reliability (κ = 0.95; 95% CI, 0.84-1.00), indicating strong consensus among evaluators. We analyzed NSAID and illicit drug exposure within 48 hours before hospital admission in the ERM cohort, including ibuprofen, naproxen, diclofenac, celecoxib, mefenamic acid, etoricoxib, indomethacin, aspirin, cocaine, methamphetamine, amphetamine, methadone, heroin, and acetaminophen. Acute kidney injury was defined according to KDIGO (Kidney Disease Improving Global Outcomes) guidelines, and CKD was identified by a glomerular filtration rate (GFR) below 60 mL/min/1.73 m². Patients with a history of AKI or end-stage kidney disease were excluded. Patients were also excluded if elevated CK levels could be attributed to trauma (defined as any recent physical injury to the body), significant hyperthermia (characterized by recorded body temperatures >38 °C), electrolyte imbalances, infections, or exposure to illicit drugs that could be myotoxic or nephrotoxic.

Bivariate comparisons involving categorical variables were performed using χ² or Fisher exact tests. Normally distributed continuous variables were compared using 2-sample t tests. Comparisons of nonnormally distributed continuous variables were conducted using Wilcoxon rank sum tests. The small number of AKI cases made a multivariable analysis not feasible. Data analysis was performed from October 1, 2023, to January 31, 2024, using SAS software, version 9.4. Two-sided P ≤ .05 was considered statistically significant. The prevented fraction of the disease was the proportion of incidents in the unexposed group that could be prevented by exposure. It was calculated using the conventional method.¹¹

This retrospective cohort study included patients who presented to the ED and were hospitalized with the primary diagnosis of ERM between January 1, 2009, and December 31, 2019, at Kaiser Permanente Northern California medical centers (21 hospitals), with a mean (SD) follow-up duration of 5.0 (3.4) years. A total of 200 patients (mean [SD] age, 30.9 [8.8] years; 145 [72.5%] male and 55 [27.5%] female; 45 [22.5%] Asian or Pacific Islander, 32 [16.0%] Black or African American, 30 [15.0%] Hispanic or Latino, 86 [43.0%] White, and 7 [3.5%] other, including American Indian or Alaska Native, declined to state, unknown, and any other race) were identified with rhabdomyolysis that was unequivocally attributed to physical exertion and not associated with any other causes (Figure 1). During a 10-year period, the incidence of hospitalized patients with ERM increased from 0.38 to 0.97 cases per 100 000 population, with an overall mean (SD) of 0.63 (2.51) cases per 100 000 population.

Baseline characteristics, laboratory values, treatments, and risks associated with AKI in hospitalized patients with ERM are summarized in Table 1. The CK levels ranged from 1140 to 360 000 U/L (mean [SD], 82 074 [62 978] U/L). In the study cohort, 17 patients developed AKI during hospitalization. The incidence of AKI was 8.5% (95% CI, 5.0%-13.3%). Among the 17 patients with AKI, 7 (41.2%) had stage 1 disease, 1 (5.9%) had stage 2 disease, and 9 (52.9%) had stage 3 disease. Of the patients with stage 3 AKI, 4 did not require dialysis, whereas 5 did. Two of 200 patients (1.0%) were classified as not having a return to baseline creatinine levels (Table 2).

There were no significant differences between patients with ERM who developed AKI and those who did not regarding demographic or clinical variables, including patient’s age, gender, race and ethnicity, baseline body mass index, presence of baseline diabetes, hypertension, smoking, year of disease onset, and season or month of disease occurrence (Table 1). We compared serum CK levels between patients with ERM who developed AKI during hospitalization and those who did not. No significant differences were observed in initial or peak CK levels (Table 1 and Figure 2). Given the reported increased AKI risk with NSAIDs,¹² we investigated the association between preadmission NSAID and acetaminophen use and AKI development in our cohort. We found a statistically significant association between a history of NSAID exposure and the development of AKI in patients with ERM (11 of 17 patients with AKI [64.7%] vs 40 of 183 patients without AKI [21.9%], P < .001) (Table 1). We did not find a statistically significant association between the preadmission use of acetaminophen and the development of AKI. One patient with AKI had a maximum CK level of 4180.0 U/L, which is considered low. However, this patient was exposed to NSAIDs before hospitalization (Table 2), highlighting that CK levels alone may not determine AKI risk. We next examined the association between a history of preadmission dehydration and the risk of AKI development in the cohort. Our findings revealed a significant association between a history of preadmission dehydration and the risk of AKI development (9 of 17 [52.9%] vs 9 of 183 [4.9%], P < .001) (Table 1). The findings suggested that if NSAIDs are eliminated, the risk of AKI in patients with ERM could be significantly reduced, with a preventable fraction¹¹ of 81.3% (95% CI, 52.1%-92.7%). Furthermore, the elimination of dehydration could reduce the risk by 91.2% (95% CI, 80.0%-96.1%), and addressing both NSAIDs and dehydration could lead to a 92.6% (95% CI, 85.7%-96.1%) preventable fraction. Male patients exhibited a significantly higher prevalence of dark urine than female patients (87 of 145 [60.0%] vs 21 of 55 [38.2%]; P = .007) (eTable 1 in Supplement 1). Further analysis revealed that dark urine was associated with significantly higher CK levels across the entire cohort and within both genders (eFigure, eTable 2, and eTable 3 in Supplement 1). We had limited data available for myoglobinuria analysis (15 of 200); however, within the first 24 hours of hospitalization, there was no significant association between dark urine discoloration and urine myoglobin levels, regardless of AKI development or differences in gender.

We did not find significant differences in the types of exercise leading to ERM between 183 patients with ERM who did not develop AKI vs the 17 patients who developed AKI (Table 1). Additionally, we did not observe any significant difference in the length of hospital stay or the rate of daily intravenous fluid use during the index hospitalization between patients with ERM who developed AKI and those who did not, both in the initial 24 hours and in the mean daily intravenous fluid intake throughout the entire hospitalization period. All of the 17 patients with ERM who had AKI in this cohort presented with AKI from the initial laboratory results. There was no statistically significant difference in the intravenous fluid administered within the first 24 hours or during the entire hospitalization between patients with ERM who developed AKI and those who did not. Patients with ERM who developed AKI were significantly more likely to receive mannitol (AKI: 2 [11.8%] vs no AKI: 1 [0.5%]; P = .02) and bicarbonate (AKI: 12 [70.6%] vs no AKI: 77 [42.1%]; P = .04), but not furosemide, compared with those who did not develop AKI. One patient (0.5%) with ERM developed compartment syndrome and required fasciotomy; however, his kidney function remained normal. There were no fatalities associated with the ERM cases in our cohort. Among the 17 patients who developed AKI, 5 (29.4%) required hemodialysis. Characteristics of the 17 patients with AKI are summarized in Table 2. None of the 17 patients with AKI had prior ERM evaluations. One patient developed ERM without AKI after resuming exercise in 3 months, with a maximum CK level of 143 481 U/L initially (with NSAID exposure) and 40 000 U/L on the second occurrence (without NSAIDs) (Table 2). No patient presented with anemia or thrombocytopenia or had evidence of disseminated intravascular coagulation. We identified only 1 patient with the sickle cell trait, who did not develop AKI. There was no significant difference in the incidence of AKI between patients with ERM admitted in the summer months compared with patients with ERM admitted in other months (Table 1). During our 10-year cohort study, 8 of the 200 patients (4%) were rehospitalized due to ERM but did not develop AKI during either the initial hospitalization or the subsequent episode. There was no difference in the initial or peak CK levels between the patients with ERM who were rehospitalized and those who were not. None of these patients reported a family history of rhabdomyolysis or hereditary myopathy. Genetic testing, encompassing whole exome sequencing and next-generation sequencing, was conducted on 5 of 200 patients with ERM (2.5%), including 3 rehospitalized individuals, to screen for hereditary myopathies; however, no pathogenic mutations were identified.

In this study, we observed an annual ERM incidence of 0.63 per 100 000 individuals, with 8.5% developing AKI. We suggest the actual ERM incidence may be higher due to our focus on hospitalized patients, as subclinical cases go undetected. Conversely, the AKI incidence may be lower than 8.5% because individuals with highly elevated CK levels from strenuous exercise may not develop clinical ERM symptoms.^4,9,13 Our study suggests that the risk of AKI in ERM is significantly lower compared with other common causes of rhabdomyolysis, such as trauma or sepsis.² These conditions may involve intravascular volume depletion and a plethora of cytokines, which could contribute to kidney injury in the setting of medullary hypoperfusion and elevated serum CK levels.

Our findings suggest that elevated serum CK levels alone may not be a sufficient risk factor for AKI in patients with ERM. For example, cases with CK levels of 360 000 U/L did not result in AKI, whereas levels at 4180 U/L did (Table 2). This finding indicates that AKI development in ERM is influenced by a combination of elevated CK levels and additional risk factors, such as NSAID exposure. Dark urine was associated with male sex in our cohort; however, it was not associated with increased risk of AKI in the entire cohort or for either sex (eTables 1 and 2 in Supplement 1). However, it was associated with significantly higher CK levels across the entire cohort and within each sex (eFigure in Supplement 1). The observed phenomenon can be attributed to potential differences between genders in reporting bias, clearance rates, and symptom onset thresholds, which may be different across genders. Myoglobin, filtered by the kidneys, appears in the urine when plasma concentrations exceed 1.5 mg/dL.¹⁴ Given the lack of association between dark urine, which could be considered indicative of myoglobinuria, and AKI, this observation may suggest that, similar to serum CK, myoglobinuria alone may not suffice to cause AKI in patients with ERM.

Our study suggests that eliminating key risk factors, such as NSAID use and dehydration, may potentially reduce the incidence of AKI in patients with ERM. Animal studies offer a parallel perspective that elevated CK levels alone may be insufficient to cause AKI in ERM, emphasizing the crucial role of NSAIDs and dehydration.^15,16 The association of prior dehydration with the development of AKI from rhabdomyolysis was also supported in these models.¹⁵ Interestingly, a previous study examined the effects of acetaminophen and ibuprofen on kidney function during exercise and dehydration. Ibuprofen was found to have a small but significant association with GFR in a sodium-depleted state, whereas acetaminophen had no such effect.¹² These findings suggest the synergistic effects of NSAIDs and dehydration during strenuous exercise on kidney function and underscore the importance of choosing acetaminophen over NSAIDs for patients experiencing muscle pain.

The reduced GFR observed with NSAID use is primarily due to the inhibition of kidney prostaglandins, which regulate kidney hemodynamics. NSAIDs inhibit cyclooxygenase, reducing prostaglandin synthesis by 50% to 60%, affecting arteriole diameters and kidney filtration.^2,17 In 2 of 17 (11.8%) of our patients with ERM and AKI, no history of NSAID ingestion or dehydration was identified. This finding may be due to unreported NSAIDs, other nephrotoxic agents, dehydration, or unknown risk factors. Our findings suggest that straightforward interventions, such as avoiding NSAIDs and prioritizing hydration, may potentially lower AKI risk in those with intense exercise-induced muscle pain. A previous study suggests that more than 50% of healthy volunteers exhibited CK levels in the rhabdomyolysis range after strenuous exercise,⁹ underscoring the significant clinical relevance of our findings.

Finally, our study revealed that none of the patients with ERM who were without AKI at admission developed AKI subsequently; in other words, all patients with ERM diagnosed with AKI on admission exhibited laboratory findings consistent with this diagnosis in their initial evaluation. Similar findings were reported elsewhere by Delaney and Vohra¹⁸ that a normal first creatinine level was a strong indicator of a normal second creatinine level. The current literature lacks sufficient studies to inform clinicians about treating patients hospitalized for ERM, with minimal consensus on the appropriate timing for discharge.¹⁹ The median hospital stay for patients without AKI was 3 days, with all treated with fluids. Our findings suggest that hospitalization solely for hydration therapy and serum CK monitoring in most ERM cases may be unnecessary, potentially reducing hospitalizations. Our study concurs with prior research suggesting a lack of a clear discharge CK level threshold for hospitalized patients with ERM and challenges the common practice of extended hospital stays solely for serum CK monitoring until reaching certain low-risk levels.^14,19 Future studies should emphasize identifying patients at lower risk of developing AKI.

To our knowledge, our study represents the most extensive research to date on patients with ERM, and it has widespread implications across various disciplines. The findings have broad implications, from recreational activities to professional, institutional, and health care settings. These implications encompass physical health–oriented training activities, such as aerobic exercises and indoor cycling sessions. Moreover, these findings are relevant to multiple sectors and disciplines, including fire and police academies as well as military training. Importantly, this study’s implications are broad within the medical field and applicable to multiple specialties. Lastly, the results of this study are crucial for public health educators, broadening our comprehension of the considerable effects that over-the-counter medications may exert on public health.

A major strength of our study is the focus on a community-based population with a diverse demographic and wide range of ethnic backgrounds,¹⁰ in contrast to the predominantly homogenous cohorts featured in the existing literature. Distinguished by its longitudinal design, this study is a pioneer in exploring long-term ERM outcomes, offering insights into the progression of AKI to CKD and its implications.

Our retrospective study faced common challenges, including missing data on NSAID dosage²⁰ and urine myoglobin. A notable limitation is potential ascertainment bias, as patients with higher creatinine levels were more likely to be questioned about NSAID exposure and dehydration. Another limitation of our study is defining dehydration through patient-reported symptoms and health care professional examinations, and the lack of a thorough assessment of objective values, such as serum or urine osmolality, due to missing data. We also acknowledge the potential influence of nephrotoxic cofactors, environmental variables, and genetic factors such as sickle cell on AKI development in patients with ERM.²¹ We also acknowledge that because we did not include patients with ERM discharged from the ED in the study, the incidence of AKI-induced ERM may be overestimated. Several questions remain unanswered, including the identification of additional risk factors for AKI in patients with ERM, such as environmental factors, patient behaviors, and genetic predispositions. This study suggests that rhabdomyolysis, as a heterogeneous disease, likely has vastly different outcomes and prognoses, depending on its underlying cause, the target population, and environmental factors. Future research should aim to clarify outcomes related to ERM and focus on the etiologic factors specific to rhabdomyolysis rather than consolidating diverse groups into a single category.

This large, community-based study significantly advances our understanding of AKI in patients with ERM. It demonstrates a lower AKI risk compared with other causes of rhabdomyolysis and challenges existing assumptions by finding no direct association between elevated CK levels and AKI risk. Additionally, it highlights potential risks associated with NSAID use and dehydration in AKI development and questions the necessity for lengthy hospitalization and associated costs in ERM.

Accepted for Publication: June 17, 2024.

Published: August 13, 2024. doi:10.1001/jamanetworkopen.2024.27464

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2024 Sabouri AH et al. JAMA Network Open.

Corresponding Author: Amir H. Sabouri, MD, PhD, Department of Neurology, Kaiser Permanente, Walnut Creek Medical Center, Walnut Creek, CA 94596 (amir.sabouri@ucsf.edu).

Author Contributions: Dr Sabouri had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Sabouri, Durant, Kafaie, Klingman, Nemazie.

Acquisition, analysis, or interpretation of data: Sabouri, Yurgionas, Khorasani, Durant, Hung, Nemazie.

Drafting of the manuscript: Sabouri, Durant, Hung.

Critical review of the manuscript for important intellectual content: Sabouri, Yurgionas, Khorasani, Durant, Kafaie, Klingman, Nemazie.

Statistical analysis: Sabouri, Hung, Nemazie.

Obtained funding: Sabouri.

Administrative, technical, or material support: Sabouri, Durant, Klingman.

Supervision: Sabouri, Kafaie, Klingman, Nemazie.

Conflict of Interest Disclosures: None reported.

Funding/Support: This work was supported by the Kaiser Permanente Northern California Community Health Program (Dr Sabouri).

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Data Sharing Statement: See Supplement 2.