Exertional rhabdomyolysis: clinical features, management, complications and prediction of acute kidney injury

Abstract

Objectives

To describe the clinical course and complications of exertional rhabdomyolysis, compare inpatient and outpatient management, and evaluate admission markers of acute kidney injury (AKI).

Methods

Prospective, multicentre study with consecutive inclusion of patients with rhabdomyolysis (creatine kinase (CK) ≥5000 U/L and/or myoglobin ≥1000 ng/mL) with exercise as the precipitating factor, presenting to the emergency departments of four hospitals in Oslo and Akershus, Norway (2019–2022). Management followed usual care. Primary outcome was AKI, defined by Kidney Disease: Improving Global Outcomes criteria.

Results

We included 136 patients with exertional rhabdomyolysis (median age 28 years, 55% male), mainly after strength training (106, 78%); 62 (46%) were managed as inpatients and 74 (54%) as outpatients. Management was similar in both groups, consisting of fluids, serial blood tests, with a median follow-up of 3 days. Complications were rare and no outpatients required escalation to inpatient care. Five (4%) patients developed AKI (median peak creatinine 217 µmol/L, 132–410). Other major complications (severe electrolyte disturbances and compartment syndrome) occurred only in those with AKI.

All patients who developed AKI had serum creatinine above the normal reference range at admission, though a few patients with elevated creatinine did not develop AKI (95% specificity). The myoglobin-to-CK ratio (≥0.48) performed best among muscle injury biomarkers and might serve as a supportive high-risk marker (99% specificity).

Conclusion

No patient with a normal admission creatinine developed AKI in this cohort, supporting outpatient management as a safe option for otherwise healthy individuals when admission creatinine is within the reference range, provided follow-up is ensured.

Trial registration number

NCT04118608.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Exertional rhabdomyolysis is usually benign.

• Still, a few patients will develop complications, most importantly acute kidney injury (AKI).

WHAT THIS STUDY ADDS

• No patient with a normal admission creatinine developed AKI or other major complications during follow-up in this cohort.

• Except for higher creatine kinase (CK) values, there were few other differences between inpatients and outpatients.

• Management was similar across inpatient and outpatient settings (fluids and serial blood tests), and no outpatients developed AKI.

• Among muscle-injury biomarkers, the myoglobin-to-CK ratio was the most accurate predictor of AKI and may support risk assessment together with other clinical findings.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• High CK activity alone can be misleading as an indicator of severity.

• Outpatient management may be considered for otherwise healthy patients with normal creatinine levels at admission, provided that appropriate clinical and laboratory follow-up is ensured.

• These findings may reduce overtreatment and hospitalisation of otherwise healthy adults.

Introduction

Elevated creatine kinase (CK) activity with few symptoms is a normal post-exercise response.¹,⁵ However, severe exercise-induced muscle injury can cause exertional rhabdomyolysis with pronounced symptoms and a risk of systemic complications.¹,^{46 7} Typical clinical features include severe muscle pain, weakness and swelling, often accompanied by dark-coloured urine,⁶,⁹ leading patients to seek medical attention. Acute kidney injury (AKI) is the key complication in rhabdomyolysis, occurring in up to 10% of exertional cases¹⁰,¹³ versus up to 50% across all causes.¹³,¹⁸

Managing exertional rhabdomyolysis and identifying those at risk of complications is a clinical challenge.^{6 7 19} It is important to identify the few who will develop AKI without admitting otherwise healthy adults. Although research evaluating outpatient follow-up is limited, clinical practice guidelines from the military suggest that most patients can be managed as outpatients with oral rehydration, restricted physical activity and follow-up every 24–72 hours.^{6 7} A consensus recommendation suggests admission only for patients at high risk of AKI, as predicted by the McMahon score.²⁰ However, this eight-item score is validated to predict dialysis or death in all-cause rhabdomyolysis.¹⁴ Military guidelines advise hospital admission if one or more high-risk markers are present: CK >20 000 U/L, suspected compartment syndrome, AKI, dark urine or confirmed myoglobinuria, electrolyte disturbances, sickle cell trait or limited compliance.^{6 7} However, fixed CK thresholds are problematic. Values exceeding 20 000 U/L have been observed in healthy volunteers after maximal eccentric exercise (>80 000 U/L)¹ and in military recruits during basic training (>35 000 U/L) without complications.² Moreover, studies show that CK is a poor predictor of AKI.^{11 14 18 21} Relying on CK alone may therefore lead to unnecessary hospitalisation and treatment of otherwise healthy patients.

This study aimed to (1) describe clinical characteristics, management and complications of exertional rhabdomyolysis, (2) compare those admitted to hospital with those managed as outpatients and (3) evaluate whether, at admission, previously proposed risk markers identify patients who subsequently develop AKI.

Methods

Study design and setting

This was a prospective, multicentre cohort study of patients with exertional rhabdomyolysis, consecutively recruited from August 2019 to March 2022 across four hospitals in Oslo and Akershus County, Norway—Oslo University Hospital (Emergency Department and Department of Orthopaedic Emergency), Diakonhjemmet Hospital, Lovisenberg Diaconal Hospital and Akershus University Hospital. Together, these hospitals serve the Oslo–Akershus metropolitan area and approximately 1.18 million residents, about 22% of Norway’s population in 2022.²² This study included 136 patients with exertional rhabdomyolysis, drawn from a larger cohort of 310 patients with all-cause rhabdomyolysis.²³ The study was pragmatic in nature, with broad eligibility criteria, consecutive recruitment, usual-care treatment and follow-up and clinically relevant outcomes, thereby enhancing external validity and real-world applicability.²⁴ For patients managed as outpatients, follow-up included information on rhabdomyolysis, advice on oral hydration and avoidance of strenuous activity, scheduled clinical and blood test follow-up most often within 24 hours and 24/7 access to reassessment by telephone or in person if symptoms changed before the scheduled visit. Most outpatients attended at least one scheduled follow-up visit and were followed until CK began to decline and there was no suspicion of other complications. In seven patients who all presented late after the precipitating exercise, further hospital follow-up was not considered necessary due to likely declining muscle biomarkers, normal serum creatinine and minimal symptoms. Three patients were advised to follow-up with their general practitioner, while the remaining patients were considered not to require further follow-up. All seven patients were contacted by telephone to confirm follow-up and absence of further symptoms. All patients, regardless of planned follow-up, had access to reassessment in the event of worsening or changing symptoms. Management and follow-up decisions were not protocolised but were at the treating physician’s discretion, in line with good clinical practice.

Patients

All adults (≥18 years) presenting to the emergency department (ED) with exertional rhabdomyolysis and CK ≥5000 U/L¹⁴ and/or serum myoglobin ≥1000 ng/mL²¹ were eligible, with values measured at presentation or reaching these thresholds within 72 hours of the index ED visit. Patients unable to communicate in a Scandinavian language or English were excluded. Exertional rhabdomyolysis was defined as rhabdomyolysis after recent exercise,^{3 10 11} judged by the treating physician and verified by the first author as the primary cause, although contributing factors (eg, prescription drugs, ethanol, illicit substances) could be present. Participants were identified consecutively using automated alerts for elevated CK and myoglobin, as well as manual searches and clinician referrals.

Outcomes

The primary outcome was AKI, defined by the Kidney Disease: Improving Global Outcomes (KDIGO) 2012 guideline and staged by severity:²⁵

1. Serum creatinine increase ≥26.5 µmol/L or 1.5–1.9×baseline.

2. 2.0–2.9×baseline.

3. ≥353.6 µmol/L, ≥3×baseline, or requirement for dialysis.

AKI is an early, clinically relevant endpoint that informs admission and follow-up decisions, enhancing external validity and real-world applicability. Baseline renal function was defined as the most recent pre-admission measurement judged to be clinically representative or, if unavailable or uncertain, the lowest value recorded during hospitalisation or follow-up.¹⁴ Patients with pre-existing chronic kidney disease (CKD) were classified as having AKI only if renal function declined according to KDIGO criteria.

Secondary outcomes:

• Mortality during hospitalisation or follow-up.

• CKD at 3 months: renal function in patients without normalised renal function at discharge was reassessed ≥3 months later using serum creatinine and estimated glomerular filtration rate from hospital follow-up or the patient’s general practitioner and classified as KDIGO 2012 stages I–V.²⁶

• Other complications:

  • Compartment syndrome was confirmed by intracompartmental pressure measurement.²⁷

  • Disseminated intravascular coagulation: clinical and laboratory criteria.²⁷

  • Electrolyte disturbances were defined as values outside reference ranges.¹⁵

Variables

The primary exposure was treatment setting defined as inpatient (any overnight hospital stay) or outpatient management.

Based on previously proposed risk markers, we used the following admission risk thresholds: CK >20 000 U/L (1, 2); myoglobin ≥4489 ng/mL and myoglobin-to-CK ratio ≥0.48²³; McMahon score >5, as proposed in the original publication¹⁴; and creatinine >105 µmol/L in men or >90 µmol/L in women (ie, above the upper reference limit).

Other variables included demographics (age and sex), level of care and length of stay and comorbidities, including pre-existing CKD, regular medication use and polypharmacy (≥5 concurrent drugs) as a proxy for comorbidity.²⁸ Alcohol and illicit drug use were recorded. Laboratory variables included daily CK, myoglobin and creatinine, as well as admission, peak and minimum values of electrolytes and other laboratory values. The myoglobin-to-CK ratio and the McMahon score¹⁴ were calculated.

Exercise-related variables included clinical features, time from exertion to presentation, training type and affected muscle groups. Treatment variables included intravenous and/or oral fluids (volume and duration), electrolyte supplementation, sodium bicarbonate, diuretics, dialysis and fasciotomy.

Data sources and measurements

Data were collected prospectively from patient interviews and clinical examinations, and by review of electronic health records and laboratory databases. To reduce information bias and misclassification, site coordinators completed case report forms using predefined definitions; the first author then reviewed them for accuracy and consistency.

Study size

This was a prespecified analysis of the exertional subgroup, consisting of all consecutive cases of exertional rhabdomyolysis presenting over 2 years across the four participating hospitals.

Statistical analysis

Categorical variables were reported as n (%) and compared using the χ² or Fisher’s exact test. Continuous variables were assessed for normality by histogram; normally distributed data were presented as mean (SD) and compared with the t-test, while skewed data were presented as median (IQR) and compared using the Mann-Whitney U test. All tests were two-sided, with p<0.05 considered statistically significant. Admission risk markers for AKI were evaluated for sensitivity, specificity, positive predictive value and negative predictive value with exact (binomial) 95% CIs. Analyses were performed using Stata V.18 (StataCorp, College Station, Texas, USA).

Missing data

Patients (n>50) were contacted by telephone after discharge to obtain missing data. Myoglobin values were missing in four patients and handled by complete-case analysis. McMahon scores were incomplete for 82 patients, mainly because bicarbonate was not measured in otherwise healthy individuals. Bicarbonate was therefore assumed normal and scored as 0.

Patient and public involvement

There was no patient or public involvement in this study, as no patient organisation for rhabdomyolysis exists.

Results

Study population

Of 145 eligible patients, 136 with exertional rhabdomyolysis were included; peak CK median 27 840 U/L and mean 37 225 U/L (figure 1). Table 1 summarises their characteristics, stratified by management: inpatient (n=62, 46%) and outpatient (n=74, 54%). Strength training was the most common activity (106, 78%), followed by CrossFit (15, 11%), and most participants trained independently (101, 74%). Group classes, personal trainers and competitions were less common (19 (14%), 6 (4%) and 10 (7%), respectively). Four (3%) had suspected or confirmed muscle disease, four had a concurrent viral infection and two were dehydrated.

Figure 1. Flowchart of patient inclusion.

Table 1. Patient characteristics and clinical findings in 136 patients with exertional rhabdomyolysis (inpatients and outpatients).

	Total (n=136)	Inpatients (n=62)	Outpatients (n=74)	P value
Male, n (%)	75 (55.2)	35 (56.5)	40 (54.1)	0.779
Age, median (IQR), years	28 (24–32)	28 (24–32)	28 (24–32)	0.943
Laboratory variables
Peak CK, U/L—median (IQR)	27 840 (15 116–51 968)	51 689 (19 976–78 815)	19 371 (12 875–34 253)	<0.001
Peak myoglobin, ng/mL—median (IQR)	2484 (1262–4528)	3799 (1749–7193)	1771 (1078–3194)	<0.001
Peak myoglobin/CK ratio—median (IQR)	0.082 (0.049–0.139)	0.082 (0.049–0.141)	0.081 (0.050–0.123)	0.816
Peak creatinine, μmol/L—median (IQR)	81 (71–89)	82 (70–89)	79 (73–87)	0.824
History
Alcohol daily/weekly, n (%)	59 (43.4)	27 (43.6)	32 (43.2)	0.971
Drug of abuse daily/weekly, n (%)	4 (2.9)	2 (3.2)	2 (2.7)	0.857
Pre-existing CKD, n (%)	2 (1.5)	1 (1.6)	1 (1.4)	0.900
Regular medication, n (%)*	53 (39.0)	29 (46.8)	24 (32.4)	0.088
Polypharmacy (≥5 drugs), n (%)	3 (2.2)	3 (4.8)	0 (0.0)	0.056
Clinical features
Muscle pain, n (%)	118 (86.8)	51 (82.3)	67 (90.5)	0.156
Swelling, n (%)	86 (63.2)	28 (45.2)	58 (78.4)	<0.001
Muscle weakness, n (%)	30 (22.1)	14 (22.6)	16 (21.6)	0.893
Dark urine, n (%)	50 (36.8)	29 (46.8)	21 (28.4)	0.027
Reduced general condition, n (%)	31 (22.8)	13 (21.0)	18 (24.3)	0.642
Others, n (%)	20 (14.7)	14 (22.6)	6 (8.1)	0.018
Time from exertion to admission
<1 day, n (%)	7 (5.2)	5 (8.1)	2 (2.7)	0.159
1–2 days, n (%)	46 (33.8)	18 (29.0)	28 (37.8)	0.280
≥3 days, n (%)	83 (61.0)	39 (62.9)	44 (59.5)	0.682
Type of exercise
Strength training, n (%)	106 (77.9)	47 (75.8)	59 (79.7)	0.583
CrossFit, n (%)	15 (11.0)	7 (11.3)	8 (10.8)	0.929
Cycling, n (%)	2 (1.5)	1 (1.6)	1 (1.4)	0.900
Running, n (%)	5 (3.7)	4 (6.5)	1 (1.4)	0.115
Other training, n (%)	8 (5.9)	4 (6.5)	4 (5.4)	0.796
Muscle groups affected
Upper extremity, n (%)	103 (75.7)	43 (69.4)	60 (81.1)	0.112
Lower extremity, n (%)	25 (18.4)	13 (21.0)	12 (16.2)	0.476
Other muscle groups, n (%)	8 (5.9)	6 (9.7)	2 (2.7)	0.085

^*Regular medications mainly included oral contraceptives and antihistamines. In addition, 2 patients (1.5%) used renin-angiotensin-aldosterone system inhibitors, 2 (1.5%) statins, 2 (1.5%) antipsychotics and 1 (0.7%) NSAIDs.

CK, creatine kinase; CKD, chronic kidney disease; NSAIDs, non-steroidal anti-inflammatory drugs.

Management

Management was similar between inpatients and outpatients—all received fluids, either intravenous (53%, mostly inpatients) or oral (47%, mostly outpatients), with clinical and laboratory monitoring. All outpatients attended follow-up appointments.

Eight (6%) patients received alkalinisation with sodium bicarbonate: three with AKI and five without AKI, all treated in the intensive care unit (ICU). Those without AKI had very high peak CK (median 100 000 U/L, 96 078–112 460) but no other risk factors, normal creatinine and a low myoglobin-to-CK ratio. The median length of stay or follow-up was three (2–4) days for both groups (table 2).

Table 2. Outcomes, follow-up and treatment in 136 patients with exertional rhabdomyolysis, by treatment setting.

Variable	Total (n=136)	Inpatients (n=62)	Outpatients (n=74)	P value
Primary outcome
AKI, n (%)	5 (3.7)	5 (8.1)	0	0.018
Stage I, n (%)	2 (1.5)	2 (3.2)	0	0.206
Stage II, n (%)	1 (0.7)	1 (1.6)	0	0.456
Stage III, n (%)	2 (1.5)	2 (3.2)	0	0.206
Secondary outcomes*
Electrolyte disturbances, n (%)	20 (14.7)	16 (25.8)	4 (5.4)	0.001
Compartment syndrome, n (%)	1 (0.7)	1 (1.6)	0	0.456
Treatment level
ICU, n (%)	8 (5.9)	8 (12.9)	0	–
General ward, n (%)	54 (39.7)	54 (87.1)	0	–
Outpatient follow-up, n (%)	74 (54.4)	0	74 (100)	–
Length of stay/follow-up, days
Overall, median (IQR)	3 (2–4)	3 (2–4)	3 (2–4)	0.579
ICU stay, median (IQR)	5 (4–6.5)	5 (4–6.5)	–	–
Hospital admission, median (IQR)	3 (2–4)	3 (2–4)	–	–
Outpatient follow-up, median (IQR)	3 (2–4)	–	3 (2–4)	–
Treatment†
Intravenous fluids, n (%)	72 (52.9)	60 (96.8)	12 (16.2)	<0.001
Days of intravenous fluids, median (IQR)	1 (1–2)	2 (2–2)	1 (1–1)	<0.001
Volume/day (mL), median (IQR)	2302 (1959–3604)	2806 (2000–4000)	1500 (1000–2000)	0.027
Only oral fluids, n (%)	64 (47.1)	2 (3.2)	62 (83.8)	<0.001
Treated electrolyte disturbances, n (%)	6 (4.4)	5 (8.1)	1 (1.4)	0.092
Sodium bicarbonate, n (%)	8 (5.9)	8 (12.9)	0	0.001
Loop diuretics, n (%)	5 (3.7)	5 (8.1)	0	0.018

^*No patients developed DIC, CKD after 3 months and no patients died.

^†No patients required dialysis.

AKI, acute kidney injury; CKD, chronic kidney disease; DIC, disseminated intravascular coagulation; ICU, intensive care unit.

Complications

Complications were rare and no escalation of care from outpatient to inpatient occurred after the initial assessment (table 2). AKI developed in five patients (4%), and compared with the non-AKI group, these patients were older, had higher myoglobin levels, a higher myoglobin-to-CK ratio and lower CK (online supplemental table S1). Three patients had stage II–III AKI; all three had hyperkalaemia. Among these, one had hyperthermia and one developed compartment syndrome requiring bilateral lower-limb fasciotomy. The latter had markedly increased strength training with anabolic steroid use (peak myoglobin 25 224 µg/L; CK 18 701 U/L) and had ‘drop foot’ at discharge. All three received ICU care, none required dialysis and renal function was normal after 3 months. All clinically important complications occurred in patients with AKI, while transient electrolyte disturbances in some without AKI were mild and not clinically relevant.

Risk markers of AKI at admission

All who developed AKI had admission creatinine above the reference range (table 3). Thus, a normal creatinine was associated with a very low risk of AKI in this cohort, but specificity (95%) was slightly reduced because 7 (5%) patients with creatinine above the reference range did not develop AKI. The myoglobin-to-CK ratio (≥0.48) performed best among the muscle injury biomarkers, with 60% sensitivity and the highest specificity (99%) of all markers, supporting its use as a high-risk marker alongside creatinine.

Table 3. Diagnostic performance of predefined admission cut-offs for AKI among patients with exertional rhabdomyolysis (n=136).

Admission risk marker (cut-off)	Sensitivity n/N (%) (95% CI)	Specificity n/N (%) (95% CI)	PPV n/N (%) (95% CI)	NPV n/N (%) (95% CI)
CK >20 000 U/L	1/5 (20.0) (0.5 to 71.6)	60/131 (45.8) (37.1 to 54.7)	1/72 (1.4) (0.04 to 7.5)	60/64 (93.8) (84.8 to 98.3)
Myoglobin ≥4489 ng/mL	3/5 (60.0) (14.7 to 94.7)	98/127 (77.2) (68.9 to 84.1)	3/32 (9.4) (2.0 to 25.0)	98/100 (98.0) (93.0 to 99.8)
Myoglobin/CK ratio ≥0.48	3/5 (60.0) (14.7 to 94.7)	126/127 (99.2) (95.7 to 100.0)	3/4 (75.0) (19.4 to 99.4)	126/128 (98.4) (94.5 to 99.8)
McMahon score >5	2/5 (40.0) (5.3 to 85.3)	129/131 (98.5) (94.6 to 99.8)	2/4 (50.0) (6.8 to 93.2)	129/132 (97.7) (93.5 to 99.5)
Creatinine >105 (M)/>90 (F) µmol/L	5/5 (100.0) (47.8 to 100.0)	124/131 (94.7) (89.3 to 97.8)	5/12 (41.7) (15.2 to 72.3)	124/124 (100.0) (97.1 to 100.0)

Sensitivity, specificity, positive and negative predictive values were calculated with exact (binomial) 95% CIs. Estimates of 100% are reported with one-sided 97.5% CIs. Myoglobin and myoglobin/CK ratio were available for n=132 patients. The number of AKI events was small (n=5), and estimates should be interpreted accordingly.

AKI, acute kidney injury; CK, creatine kinase; NPV, negative predictive value; PPV, positive predictive value.

In contrast, CK >20 000 U/L was common—72/136 (53%) overall and 28/74 (38%) above this threshold were managed as outpatients without complications or later admission. As an AKI predictor, this CK cut-off would have identified only 1/5 (20%) cases.

Discussion

In this study, complications from exertional rhabdomyolysis were uncommon. No patients with a normal admission serum creatinine developed AKI or other adverse outcomes during follow-up. These findings support outpatient care as a safe option for otherwise healthy patients when admission creatinine is normal and follow-up is ensured.

Our cohort was comparable to previous exertional rhabdomyolysis studies, typically involving young men^{8 10 11 13 29} presenting 2–3 days after exercise,¹⁰ most often with upper-limb involvement.^{8 13 29 30} Mean peak CK was comparable (37 225 vs 38 552 U/L⁷), and strength and high-intensity functional training (eg, CrossFit) were the main triggers, consistent with larger cohorts and reviews.^{10 11 29 31} As in previous reports, most patients had muscle pain, often with swelling and weakness.^{3 29 30}

Inpatients and outpatients in our study had similar characteristics and a median follow-up of 3 days, comparable to hospital stays of 2–4 days in other studies.^{312 13 30},³² In our cohort, outpatient management appeared safe, with high adherence to oral rehydration, scheduled follow-up and 24/7 access to reassessment if symptoms changed. In seven otherwise healthy patients, no further hospital follow-up was arranged based on the initial clinical assessment, including delayed presentation after exercise, normal serum creatinine and minimal symptoms. Three patients were advised to follow-up with their general practitioner, and declining muscle biomarker levels were confirmed. No additional symptoms were reported, and none required reassessment or further medical care. All remaining outpatients attended their scheduled hospital follow-up. This high level of compliance may reflect a publicly funded healthcare system with universal coverage and no-cost access to follow-up.

As in other studies, oral or intravenous fluids were the cornerstone of management in our cohort.^{8 10} Although guidelines recommend rehydration alone and discourage routine alkalinisation,³³,³⁵ some patients (6%) received bicarbonate, more than half without AKI. Bicarbonate was often given for high CK alone, despite normal creatinine and no other risk factors and likely represented unnecessary therapy.

Complications were rare and consistent with reports that exertional rhabdomyolysis has fewer complications than other causes,^{13 14 17 31} with AKI as the main complication.^{13 14 17} Most studies report AKI in fewer than 10% of such patients.¹⁰,^{1330 31 36} By contrast, military cohorts reported AKI in up to 28% of cases, with up to 8% mortality in one systematic review.³⁷ However, all four deaths were from two case reports and attributed to hypothermia/extreme conditions and exertional sickling in individuals with sickle cell trait.³⁷ Most studies find that renal function normalises by discharge.^{9 13 30} In our cohort, three out of five patients had not normalised renal function at discharge, but all had normal renal function at 3 months.

In our study, major complications including severe electrolyte disturbances and compartment syndrome occurred only in patients with AKI. However, another study has reported compartment syndrome without AKI,¹¹ highlighting the need to specifically assess for this complication, although seldom reported in exertional rhabdomyolysis (0–4%).¹¹,^{1330 32 36} Mild electrolyte disturbances are common and often of little relevance,^{3 13 30} consistent with our findings.

Other recognised risk factors were uncommon (eg, dehydration, illicit drugs, non-steroidal anti-inflammatory drugs, viral illness^{1 6 7 11}), but in particular hyperthermia and sickle cell trait (rare in Norway) require closer monitoring, including hospital admission, because of the risk of multiorgan failure.^{6 7 37}

All patients who developed AKI during follow-up already had creatinine above the reference range at admission, although not all yet met the KDIGO criteria. This is consistent with a previous retrospective cohort.¹¹ The high sensitivity of creatinine should be interpreted in light of its role as a defining component of the KDIGO criteria, and therefore partly reflects outcome definition rather than independent prediction.²⁵ Accordingly, admission creatinine may be best viewed as a screening or triage marker for early risk stratification rather than an independent predictor of AKI. The role of admission creatinine may be particularly relevant in exertional rhabdomyolysis, because most patients present after 2–3 days,¹⁰ by which time serum myoglobin has already peaked^{21 38 39} and any impending kidney injury is likely to be apparent.

Creatinine, however, had a lower specificity than the myoglobin-to-CK ratio, an acceptable trade-off in emergency care to avoid missed AKI, but the ratio can help identify high-risk patients and guide admission and monitoring decisions when interpreted alongside creatinine and clinical assessment. The ratio outperformed both CK and myoglobin alone, consistent with previous results.^{18 23} Using a lower ratio cut-off derived from quartile analysis, ≥0.2, as reported in previous studies,^{18 23} would have identified all five AKI cases (5/5) in this exertional cohort, although this finding should be interpreted with caution given the small number of outcome events.

Admission decisions appeared most influenced by CK. Many had high CK, yet a CK >20 000 U/L, cited as high risk in military guidance,^{6 7} would have identified only 1/5 AKI cases if used alone. This aligns with studies showing CK is poorly associated with AKI.^{11 14 17 18 21 32} Nonetheless, higher CK correlates with more intensive treatment and longer stay,^{11 12 32} and exertional cases typically have higher CK than other causes but lower complication risk,^{9 11 17 31} highlighting CK’s limited value for risk stratification and the need for better markers. The myoglobin-to-CK ratio did not differ between inpatients and outpatients or among patients without AKI who received alkalinisation, supporting that it better reflects risk and may help avoid unnecessary admissions. One possible biological explanation for the performance of the ratio is that myoglobin rises rapidly and earlier than CK after muscle injury and is a main cause of AKI, thereby better reflecting the nephrotoxic burden.^{21 40} Myoglobin may also better reflect the severity of muscle injury and necrosis.⁴¹ However, the ratio appears to perform better than myoglobin alone,^{18 23} possibly because higher CK values are more common in younger individuals with low comorbidity,^{5 9 13} likely reflecting greater muscle mass rather than injury severity alone, and younger individuals often have better renal reserve, which may attenuate AKI risk despite high myoglobin levels.

The 276th European Neuromuscular Centre (ENMC) International Workshop recommends admission for patients at high risk of AKI, defined by a McMahon score >5.^{14 20} In our cohort, the score identified only two of five AKI cases, suggesting limited value in exertional rhabdomyolysis. The score was developed to predict dialysis or death in all-cause rhabdomyolysis, outcomes rare in exertional disease and absent in our study, which may explain its poor performance for AKI-focused triage. It also uses eight variables and requires bicarbonate from blood gas analysis. Admission creatinine, supported by the myoglobin-to-CK ratio, may therefore be a better and simpler marker for early AKI risk assessment in exertional rhabdomyolysis.

Strengths and limitations

Strengths include prospective, consecutive recruitment across four hospitals over 2 years, broad eligibility and a pragmatic design that reflects real-world EDs, supporting a representative cohort. These features, along with uniform biochemical thresholds for case identification across sites, helped minimise selection bias. Although the literature is limited and mainly retrospective, our findings are consistent with previous studies, supporting generalisability to an unselected cohort of exertional rhabdomyolysis patients presenting to the ED.

The major limitation is the small number of AKI events (n=5), which, while reassuring and consistent with previous studies, leads to uncertainty and wide CIs, particularly for sensitivity. Therefore, the predictive performance of the models and the proposed cut-off values should be validated in larger cohorts. Additional limitations include assuming normal bicarbonate values for McMahon scores in otherwise healthy individuals. This assumption may have led to underestimation of the score and biased results against its performance. Complete-case analyses showed slightly improved performance, largely explained by exclusion of one patient with AKI stage 1 (creatinine normalising the following day) due to missing bicarbonate, without altering the overall findings. Given the clinical context of young, otherwise healthy patients with mostly normal renal function at presentation, this assumption is unlikely to have materially influenced the results.

The cohort was relatively homogeneous in terms of age and comorbidity, and because of the small number of outcome events, we did not perform adjusted analyses. Residual bias cannot be excluded. A limitation of this study is that follow-up was not protocolised and that a small number of low-risk patients did not receive hospital follow-up. However, these patients were contacted by telephone, declining muscle biomarkers were confirmed in those advised primary care follow-up, and none reported further symptoms or required additional medical care. Outpatient management and follow-up in this study were supported by a publicly funded healthcare system with universal access and reliable follow-up, and these findings should be extrapolated with caution to healthcare settings without similar access or follow-up. Generalisability is greatest to otherwise healthy young and middle-aged adults.

Clinical implications

Exertional rhabdomyolysis in otherwise healthy adults is generally benign and should not deter exercise. Admission creatinine is useful for early AKI risk stratification, with a normal value indicating a very low risk of AKI and an elevated value indicating higher risk. This supports outpatient management of otherwise healthy patients with normal creatinine, with oral hydration, follow-up and attention to renal function and compartment syndrome. CK remains important to confirm rhabdomyolysis, but CK alone should not guide admission decisions. The myoglobin-to-CK ratio is a useful supportive rule-in marker that, when positive, strengthens the case for admission or closer monitoring together with other clinical findings. These findings may help avoid overtreatment and hospitalisation of otherwise healthy adults.

Conclusions

In this study of exertional rhabdomyolysis, complications were rare and no patient with normal admission creatinine developed AKI or other severe complications in this cohort during follow-up. Management was similar for inpatients and outpatients, mainly fluids and serial blood tests. These findings support outpatient management as a safe option for otherwise healthy individuals when admission creatinine is normal and follow-up is provided, although validation in larger cohorts is warranted.

Data availability statement

No data are available.