Factors associated with acute kidney injury in rhabdomyolysis: Insights from a prospective study

Abstract

BACKGROUND

Rhabdomyolysis is a common condition in intensive care units (ICUs), but data on acute kidney injury (AKI) associated with rhabdomyolysis remain limited, particularly in North Africa.

AIM

To determine the incidence of AKI during rhabdomyolysis and to identify its associated factors among ICU patients.

METHODS

We conducted a prospective, single-center correlational study from January to July 2024 in the ICU of Rabta University Hospital in Tunisia, including patients diagnosed with rhabdomyolysis. Clinical and laboratory parameters, along with the McMahon score, were assessed at the time of inclusion. Patients were then monitored during their ICU stay to detect the occurrence of AKI.

RESULTS

AKI occurred in 51.9% of the 54 patients included in the study. Univariate analysis identified several admission parameters associated with AKI, including elevated levels of creatine phosphokinase, lactate dehydrogenase (LDH), phosphorus, creatinine, and transaminases, as well as low bicarbonate and albumin levels. The optimal McMahon score threshold for predicting AKI was 4.75, with a sensitivity of 78.6% and specificity of 76.9%. In multivariate analysis, only initial LDH [P = 0.012; odds ratio (OR) = 0.993; 95%CI: 0.988-0.999] and initial bicarbonate (P = 0.025; OR = 1.553; 95%CI: 1.058-2.280) levels were independently associated with AKI occurrence.

CONCLUSION

Simple admission biomarkers such as LDH and bicarbonate should be considered for early AKI risk assessment in rhabdomyolysis, particularly in settings with limited access to advanced diagnostic tools.

Key Words: Acute kidney injury; Factors; Intensive care unit; Prospective study; Rhabdomyolysis

Core Tip: To the best of our knowledge, this is the first prospective study specifically designed to explore factors associated with rhabdomyolysis-induced acute kidney injury. Our findings provide important clinical insights for nephrologists. These results may enhance risk stratification at the time of admission, allowing for earlier implementation of targeted preventive strategies and closer monitoring of renal function, ultimately contributing to improved patient outcomes.

INTRODUCTION

Rhabdomyolysis often complicates traumatic situations, with an estimated frequency of 85%[1]. It can also occur in non-traumatic conditions such as intoxications, sepsis, and seizures[1]. It consists of the breakdown of skeletal muscle cells and the release of their contents, like myoglobin, creatinine phosphokinase (CPK), lactate dehydrogenase (LDH) and electrolytes into the systemic circulation, which causes various biological abnormalities[2].

It is a serious and potentially life-threatening condition that may result in an acute kidney injury (AKI), which can adversely affect patient prognosis[2]. AKI occurs in 10%-60% of rhabdomyolysis cases and accounts for 5%-25% of all AKI etiologies[3,4]. Its prevalence varies depending on geographic region and the characteristics of the study population.

The mechanisms underlying rhabdomyolysis-associated AKI include the nephrotoxic effects of myoglobin, renal hypoperfusion due to hypovolemia, and inflammation that exacerbates kidney injury[5-7]. Beyond its immediate clinical consequences, AKI is now considered to be part of a broader continuum of kidney damage rather than an isolated event. Emerging evidence indicates that AKI and chronic kidney disease (CKD) are interrelated conditions, with many patients failing to fully recover baseline kidney function after an AKI episode. This incomplete recovery can lead to the development of new-onset CKD or to the progression of pre-existing CKD. This transition is mediated by sustained tubular injury, maladaptive repair processes, and ongoing inflammation, particularly when the initial kidney injury is severe or prolonged[8,9]. Consequently, early recognition of individuals at risk of AKI is essential, not only to reduce short-term morbidity but also to preserve long-term renal function. In the setting of rhabdomyolysis, where kidney injury can occur abruptly and progress silently, prompt identification of high-risk patients provides a critical opportunity for early intervention and prevention of long-term renal decline.

Identifying risk factors and applying predictive tools such as the McMahon score are essential for guiding clinical management and improving patient outcomes. The McMahon score has been validated in several studies as a reliable predictor of AKI and mortality in patients with rhabdomyolysis[10,11].

Given the lack of prospective studies on this clinically significant topic, we conducted this study to estimate the incidence of AKI in patients with rhabdomyolysis and to identify associated factors among those admitted to the intensive care unit (ICU). We hypothesized that a prospective investigation might reveal a distinct risk profile compared with that identified in retrospective studies.

MATERIALS AND METHODS

Study population

This prospective correlational, single-center study was conducted from January to July 2024 in the ICU of Rabta University Hospital Center in Tunisia. The study included patients diagnosed with rhabdomyolysis who met the predefined inclusion criteria. The study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Ethics Committee of Rabta University Hospital Center. Eligible participants were adults aged ≥ 18 years with CPK level ≥ 5 times the upper limit of normal (≥ 1000 U/L). Patients were excluded if they met any of the following criteria: Stage 5 chronic kidney disease; kidney transplantation; refusal to participate; or elevated CPK levels attributable to myocardial infarction. Additional exclusion criteria included alternative causes of AKI, such as acute glomerulonephritis, hepatorenal syndrome, cardiorenal syndrome, obstructive uropathy, or vascular injury; failure to undergo the required diagnostic evaluations; or a follow-up duration of less than 5 days.

Study design

As part of the study protocol, serum CPK levels were systematically measured upon admission to the ICU and monitored daily thereafter. Patients were enrolled after confirmation of rhabdomyolysis. Demographic characteristics, medical history, lifestyle habits, and chronic medications were recorded. The etiologies of rhabdomyolysis were classified as traumatic or non-traumatic. Physical examination and laboratory evaluations were performed at inclusion and during routine ICU follow-up. The McMahon score was calculated for all patients at the time of inclusion (Table 1). Complications occurring during the study period were documented, and all therapeutic interventions and imaging studies were recorded. Upon diagnosis of rhabdomyolysis, statin therapy was discontinued. Patients received lactated Ringer’s solution and/or isotonic saline unless contraindicated by volume overload or oliguria. Intravenous furosemide was initiated when urinary output decreased to < 50 mL/hour despite adequate fluid balance. Renal replacement therapy (RRT) was initiated in cases of oligoanuria, refractory hyperkalemia, severe metabolic acidosis, or acute pulmonary edema unresponsive to loop diuretics. Serum creatinine was measured using the Jaffe colorimetric-kinetic method. Serum and urinary myoglobin are not routinely assessed at our center; therefore, these variables were not included in the study.

Table 1 Calculation of the McMahon score.

Variable	Value	Points
Age (years)	50–70	1.5
	71–80	2.5
	> 80	3
Women		1
Admission creatinine (μmol/L)	124–194	1.5
	> 194	3
Admission calcium (mmol/L)	< 1.875	2
Admission creatinine kinase (UI/L)	> 40000	2
Etiology	Not seizures, syncope, exercise, statins or myositis	3
Initial phosphate (mmol/L)	1.3-1.74	1.5
	> 1.74	3
Initial bicarbonate (mmol/L)	< 19	2

Definitions

The study period was defined as the time from rhabdomyolysis diagnosis to either the normalization of CPK levels or the last follow-up. Hyperkalemia was defined as a serum potassium level > 5 mmol/L, while levels exceeding 6.5 mmol/L were considered severe hyperkalemia[12,13].

Statistical analysis

Analyses were performed using SPSS version 24. Quantitative variables with a normal distribution were expressed as means ± SD, while those with a non-normal distribution were presented as medians with interquartile range. Normality was assessed using the Kolmogorov-Smirnov test. Qualitative variables were described as n (%). The comparison of qualitative variables was performed using the χ² test. For quantitative variables, the student’s t-test or the Mann-Whitney U test was used, depending on data distribution. A binary logistic regression was performed to identify factors independently associated with AKI occurrence. Variables were selected for inclusion in the multivariate model based on univariate analysis results (P < 0.05). A backward stepwise approach was used, starting with all preselected variables and progressively removing the least significant ones. Model fit was assessed using the Hosmer-Lemeshow test. Adjusted odds ratios (aOR) with 95% confidence intervals were reported. To avoid collinearity, the McMahon score and its constituent variables were excluded from the same multivariate model. Receiver operating characteristic (ROC) curves were generated for continuous variables significantly associated with AKI to determine optimal threshold values, maximizing sensitivity and specificity. A P-value < 0.05 was considered statistically significant for all analyses.

RESULTS

Patients’ characteristics

We assessed 1048 patients for eligibility during the study period, and 110 patients were diagnosed with rhabdomyolysis. After exclusions, 54 patients were enrolled (Figure 1). The median follow-up time was 13 (5-136) days. Patients’ characteristics are summarized in Table 2. Rhabdomyolysis occurred after a median of 2 (0-35) days of hospitalization. The median duration of rhabdomyolysis was 6 (3-65) days. The mean maximum CPK level was 2634.5 UI/L (1062-212780), reaching its peak immediately in 37 patients (68.5%) and after 2 days (1-10) in 17 patients (31.5%).

Figure 1 Flowchart of patient selection. ICU: Intensive care unit; CKD: Chronic kidney disease; AKI: Acute kidney injury; CPK: Creatinine phosphokinase.

Table 2 Baseline demographic characteristics and comorbidities of patients with rhabdomyolysis: Comparison between those who developed acute kidney injury and those who did not, n (%)/mean ± SD.

Variables	All patients (n = 54)	AKI group (n = 28)	Non-AKI group (n = 26)	P value
Age (years)	46.1 ± 17.9	50.2 ± 17.3	41.8 ± 17.8	0.087
Gender				0.381
Male	43 (79.6)	21 (75)	22 (84.6)
Female	11 (20.4)	7 (25)	4 (15.4)
Smoking	35 (64.8)	16 (57.1)	19 (73.1)	0.221
Comorbidities
Diabetes	10 (18.5)	8 (28.6)	2 (7.7)	0.05
Dyslipidemia	14 (25.9)	11 (39.3)	3 (11.5)	0.02
Hypertension	14 (25.9)	9 (32.1)	5 (19.2)	0.279
Coronary artery disease	10 (18.5)	4 (14.3)	6 (23.1)	0.406
Chronic heart failure	4 (7.4)	3 (10.7)	1 (3.8)	0.336
Chronic kidney disease	3 (5.6)	2 (7.1)	1 (3.8)	0.597
Regular treatment before admission to the intensive care unit
Oral antidiabetics	6 (11.1)	4 (14.3)	2 (7.7)	0.44
Insulin therapy	6 (11.1)	5 (17.9)	1 (3.8)	0.114
Statin	13 (24.1)	8 (28.6)	5 (19.2)	0.422
Renin-angiotensin system blockers	5 (9.3)	2 (7.1)	3 (11.5)	0.464
Oral furosemide	6 (11.1)	3 (10.7)	3 (11.5)	0.629
Causes of rhabdomyolysis				0.001
Traumatic	23 (42.6)	6 (21.4)	17 (65.4)
Non traumatic	31 (57.4)	22 (78.6)	9 (34.6)

AKI: Acute kidney injury.

A total of 28 patients (51.9%) developed AKI after a median of 4 (2-12) days after rhabdomyolysis diagnosis. According to the KDIGO-defined stages of AKI, 4 patients (7.4%) were at stage 1, 9 patients (16.7%) at stage 2, and 15 patients (27.8%) at stage 3. The median peak creatinine level was 252.85 (115-1791) μmol/L, occurring after a median of 3 (1-24) days from the onset of AKI. RRT was indicated in 11 patients (39.3%), after a median of 4 (1-23) days from the onset of AKI. The dialysis modalities used were conventional hemodialysis in 13 cases (45%) and hemodiafiltration in 15 cases (55%). The average number of dialysis sessions per patient was 4 (2-16). Intravenous furosemide was initiated in 19 patients (35.2%), which was administered for a median duration of 7 (1-30) days, with a median maximum dose of 600 (20-1500) mg per day.

Univariable analysis

Dyslipidemia and non-traumatic causes of rhabdomyolysis were significantly associated with the occurrence of AKI during the study period, while gender, smoking history, regular treatment, diabetes, and arterial hypertension were not significantly associated with AKI (Table 2). Physical examination findings at inclusion were not significantly associated with AKI occurrence during the study period (Table 3). Statistical analysis of baseline laboratory parameters revealed significant associations between AKI occurrence and several biochemical markers. Patients who developed AKI had higher levels of CPK (2868 UI/L vs 2004 UI/L; P = 0.04), LDH (877.5 U/L vs 321.5 U/L; P < 0.001), phosphorus (1.2 mmol/L vs 0.8 mmol/L; P = 0.024), urea (10 mmol/L vs 4.9 mmol/L; P = 0.001), creatinine (149.4 μmol/L vs 66 μmol/L; P < 0.001), ALT (51.5 U/L vs 25 U/L; P = 0.038), and AST (139 U/L vs 58.5 U/L; P < 0.001). In contrast, serum bicarbonate levels were significantly lower in the AKI group (20.1 mmol/L vs 24.5 mmol/L; P < 0.001), as were albumin levels (26.3 g/L vs 29.7 g/L; P = 0.017). Prothrombin time was significantly longer in patients who developed AKI (60.6 ± 16.8 vs 74.2 ± 18; P = 0.006). Additionally, the McMahon score was significantly higher in patients who developed AKI (7.29 vs 4.15; P < 0.001) (Table 4). The optimal McMahon score threshold for predicting AKI occurrence was 4.75, with a sensitivity of 78.6% and specificity of 76.9% (Figure 2A). Hydration was administered to 48 patients (88.9%) for a median duration of 6 days (1-16). Isotonic saline solution was used in 47 cases (87%), lactated Ringer’s solution in 13 cases (24.1%), and the two solutions were combined in 35 patients (27%). Hydration was identified as a protective factor against AKI development (P = 0.012). During the study period, infections complicated the ICU stay in 38 patients (70.4%). Among the 28 patients who developed AKI during follow-up, 23 (82.1%) had an infection before AKI diagnosis.

Figure 2 Receiver operating characteristic curve. A: Receiver operating characteristic (ROC) curve showing the performance of the McMahon score in predicting acute kidney injury. The optimal cut-off value was 4.75, with a sensitivity of 78.6% and a specificity of 76.9%, as determined by the Youden index. The area under the curve (AUC) was 0.791 (95%CI: 0.662-0.919; P < 0.001); B: For lactate dehydrogenase as a predictor of acute kidney injury in patients with rhabdomyolysis. The optimal cut-off value was 537.5 IU/L, with a sensitivity of 85.2% and a specificity of 80.8%, as determined by the Youden index. The AUC was 0.853 (95%CI: 0.748–0.958; P < 0.001); C: For bicarbonate level as a predictor of acute kidney injury in patients with rhabdomyolysis, the optimal cut-off value was 21.35 mmol/L, yielding a sensitivity of 83.3% and a specificity of 67%, as determined by the Youden index. The AUC was 0.798 (95%CI: 0.677-0.919; P < 0.001). AUC: Area under the curve; LDH: Lactate dehydrogenase; ROC: Receiver operating characteristic; AKI: Acute kidney injury.

Table 3 Baseline clinical data of patients with rhabdomyolysis: Comparison between those who developed acute kidney injury and those who did not, mean ± SD.

Variables	All patients (n = 54)	AKI group (n = 28)	Non-AKI group (n = 26)	P value
SBP (mmHg)	120.2 ± 23.5	121.2 ± 25.4	119.1 ± 21.9	0.752
DBP (mmHg)	69.2 ± 14.1	68.2 ± 13.9	70.23 ± 14.5	0.605
MAP (mmHg)	86.2 ± 12.2	85.8 ± 12.5	86.5 ± 12.1	0.21
Heart rate (beats per minute)	91.5 ± 24.8	90.2 ± 21.1	93 ± 28.6	0.688
Diuresis (mL/hour)	66 ± 28.1	62 ± 35	70.5 ± 17.7	0.254

SBP: Systolic blood pressure; DBP: Diastolic blood pressure; MAP: Mean arterial pressure; AKI: Acute kidney injury.

Table 4 Baseline laboratory data and McMahon score of patients with rhabdomyolysis: Comparison between those who developed acute kidney injury and those who did not, mean ± SD/median (25^th-75^th percentiles).

Variables	All patients (n = 54)	AKI group (n = 28)	Non-AKI group (n = 26)	P value
CPK (UI/L)	2151 (1340-3729)	2868 (1410.7-9178.2)	2004 (1216-2426.7)	0.04
LDH (U/L)	580 (315-95)	877.5 (665.5-1793)	321.5 (286-406.7)	< 0.001
Corrected serum calcium (mmol/L)	2.3 ± 0.1	2.32 ± 0.2	2.3 ± 0.1	0.562
Phosphorus level (mmol/L)	1.07 ± 0.6	1.2 ± 0.8	0.82 ± 0.2	0.024
Serum bicarbonate (mmol/L)	22.25 ± 4.3	20.1 ± 4.2	24.5 ± 3.3	< 0.001
Urea (mmol/L)	6.8 (5-11.6)	10 (6.8-15.2)	4.9 (3.2-6.5)	0.001
Serum creatinine, (μmol/L)	82.5 (63.8-152.8)	149.4 (97.5-189)	66 (50.7-81.4)	< 0.001
Potassium level (mmol/L)	4.15 ± 0.8	4.3 ± 0.8	3.9 ± 0.6	0.147
Albumin level (g/L)	27.92 ± 5.1	26.3 ± 5.8	29.7 ± 3.6	0.017
AST (UI/L)	80 (55-178)	139 (69.5-202)	58.5 (44-94)	< 0.001
ALT (UI/L)	38.5 (23-76.5)	51.5 (27-91)	25 (20-44)	0.038
Total cholesterol (mmol/L)	3.65 ± 1.2	3.7 ± 1.5	3.5 ± 0.7	0.439
LDLc (mmol/L)	2.4 ± 1.1	2.4 ± 1.4	2.3 ± 0.7	0.724
Glycemia (mmol/L)	7.34 ± 3.7	8.1 ± 4.2	6.5 ± 2.8	0.126
C-reactive protein (mg/L)	150.61 ± 113.1	167.6 ± 128.7	132.3 ± 92.3	0.255
Hemoglobin (g/dL)	10.1 ± 2.2	9.9 ± 2.2	10.2 ± 2.05	0.674
White blood cells per liter	14776.8 ± 5970.2	14935.4 ± 6865.4	14606.2 ± 4960	0.842
Prothrombin time (%)	67.17 ± 18.5	60.6 ± 16.8	74.2 ± 18	0.006
McMahon score	5.78 ± 3	7.2 ± 3.9	4.1 ± 1.39	< 0.001

CPK: Creatinine phosphokinase; LDH: Lactate dehydrogenase; AST: Aspartate transaminase; ALT: Alanine transaminase; LDLc: Low-density lipoprotein cholesterol; AKI: Acute kidney injury.

Shock occurred in 17 patients (31.5%), with a median onset of 4 (2-21) days after inclusion. Univariate analysis showed a statistically significant association between AKI occurrence and infection (23 cases vs 15 cases, P = 0.049).

Antibiotic therapy was administered to 40 patients (74.1%). Among those who developed AKI, 21 patients (75%) received antibiotics before its onset, with a median duration of 6 (4-23) days. Of the various agents used, only imipenem and colistin were significantly associated with AKI occurrence (P = 0.034 and P = 0.023, respectively).

A contrast agent was administered to 35 patients (64.8%). Among patients who developed AKI, the diagnosis was made at a median of 3 (2-5) days after contrast agent exposure. Exposure to contrast agents was significantly associated with AKI occurrence (P = 0.006) (Table 5).

Table 5 Medical management and intensive care unit stay-related complications of patients with rhabdomyolysis: Comparison between those who developed acute kidney injury and those who did not, n (%).

Variables	All patients (n = 54)	Non-AKI group (n = 26)	AKI group (n = 28)	P value
Hydration	48 (88.9)	26 (100)	22 (78.6)	0.012
Infection	38 (70.4)	15 (57.7)	23 (82.1)	0.049
Antibiotics	40 (74.1)	17 (65.4)	21 (75)	0.439
Amoxicillin-clavulanic acid	19 (35.2)	11 (42.3)	8 (28.6)	0.352
Piperacillin/tazobactam	17 (31.5)	9 (34.6)	8 (28.6)	0.753
Cefazolin	2 (3.7)	0 (0)	2 (7.1)	0.244
Cefotaxime	7 (13)	1 (3.8)	6 (21.4)	0.053
Imipenem	13 (24.1)	3 (11.5)	10 (35.7)	0.034
Gentamicin	4 (7.4)	1 (3.8)	3 (10.7)	0.606
Amikacin	17 (31.5)	10 (38.5)	7 (25)	0.242
Vancomycin	1 (1.9)	0 (0)	1 (3.6)	0.523
Teicoplanin	5 (9.3)	1 (3.8)	4 (14.3)	0.202
Metronidazole	3 (5.6)	1 (3.8)	2 (7.1)	0.535
Colistin	5 (9.3)	0 (0)	5 (17.9)	0.023
Fluoroquinolone	4 (7.4)	1 (3.8)	3 (10.7)	0.303
Quinolone	2 (3.7)	0 (0)	2 (7.1)	0.267
States of shock	17 (31.5)	7 (26.9)	10 (35.7)	0.487
Contrast agent use	35 (64.8)	12 (46.2)	23 (82.1)	0.006

AKI: Acute kidney injury.

Multivariable analysis

Multivariate logistic regression showed that the variables independently associated with AKI occurrence were initial LDH level (P = 0.025; OR = 1.553; 95%CI: 1.058-2.280) and initial bicarbonate level (P = 0.012; OR = 0.993; 95%CI: 0.988-0.999). The AUC, sensitivity, specificity, and optimal threshold values for LDH and bicarbonate levels in predicting AKI are presented in Figure 2B and C.

DISCUSSION

AKI is the most frequent and severe systemic complication of rhabdomyolysis[6]. In this context, AKI pathophysiology is multifactorial, involving myoglobin-induced tubular toxicity, intratubular obstruction, renal vasoconstriction, and oxidative stress. Immune cells activated by muscle cell injury release cytokines that amplify tissue damage, leading to inflammation and necrosis. The immune response is primarily mediated by the innate immune system; however, further research is needed to fully elucidate its role[3,7].

The incidence of AKI during rhabdomyolysis varies from 10% to 60% according to the literature[3,4]. Differences in AKI definitions, study populations, and inclusion criteria may account for this variability[14,15]. More recent studies report even higher rates, reaching up to 94% across all causes of rhabdomyolysis[3,16,17]. This apparent rise may reflect an increased awareness and reporting of the condition. In our study, AKI occurred in 51.9% of patients, with most cases classified as KDIGO stage 3, followed by stages 2 and 1. Burgess et al[2] reported a 60% incidence of rhabdomyolysis-induced AKI in a cohort of 733 patients from the Intensive Care National Audit and Research Centre, with 52% requiring RRT. de Fallois et al[17] observed a substantially higher incidence (83.5%), along with an in-hospital mortality rate of 55.6% among AKI cases. Similarly, Candela et al[1] documented a 44.1% incidence of KDIGO-defined stage 3 AKI attributable to rhabdomyolysis in their multicenter retrospective analysis. These findings highlight the heterogeneity of AKI incidence across studies, likely influenced by differences in study design, patient severity, and timing of AKI assessment. Compared with reports in the literature, our incidence falls within the expected range but supports the notion that critically ill patients with rhabdomyolysis remain at high risk of severe AKI. Our results suggest that AKI in this setting is frequently advanced at diagnosis, underlining the need for earlier identification and targeted preventive strategies. Further multicenter prospective studies are warranted to better characterize at-risk subgroups and optimize early management. Various epidemiological, clinical, and biological parameters have been extensively studied to identify risk factors for AKI in patients with rhabdomyolysis. Age has consistently been reported as a significant predictor, with older individuals showing a higher risk of rhabdomyolysis-related AKI[1,18]. In our study, patients who developed AKI were indeed older than those who did not; however, this association did not reach statistical significance. This finding may be explained by the relatively young age of our study population (median age: 46.2 years), which differs from most previously published studies where the risk of AKI increased substantially in patients over 50[1,18]. It is plausible that in younger populations, age-related variation in renal vulnerability is less pronounced, limiting its discriminatory power in predicting AKI. This underscores the importance of adapting risk assessment models to the demographic and clinical characteristics of the target population rather than applying generalized predictors across all settings. In a retrospective cohort study, Seo et al[19] identified several factors independently associated with AKI development, including older age, statin therapy, elevated baseline serum creatinine (> 1.3 mg/dL), high lactate levels (> 2.25 mmol/L), and elevated initial CPK. Similarly, Yang et al[20] reported multiple predictors of moderate-to-severe AKI in patients with rhabdomyolysis, such as hypertension, leukocytosis, hypertriglyceridemia, low HDL cholesterol, elevated serum phosphorus, and increased CPK. These findings are largely consistent with our results, particularly regarding the association of elevated CPK and phosphorus levels with AKI. However, other variables such as statin use or lipid profile abnormalities were not significant in our study, possibly due to differences in patient selection or underlying comorbidities. This variability underlines the importance of context-specific validation of risk factors across different clinical settings. The McMahon score is a clinical tool used to assess the risk of AKI in patients with rhabdomyolysis[10,11]. It incorporates demographic factors (age and gender), biological parameters (serum creatinine, calcium, phosphate, and bicarbonate levels), and the underlying etiology of rhabdomyolysis[11]. Few studies have specifically focused on validating its predictive value and identifying an optimal cutoff with acceptable sensitivity and specificity. Notably, all of these studies were retrospective. In a cohort of 151 victims of the recent earthquake in Turkey, a McMahon score threshold of 6 was associated with AKI occurrence, with a sensitivity of 80% and specificity of 64.5%[21]. Ahmad et al[4] reported that a cutoff of 7.8 offered a sensitivity of 71.4% and specificity of 77.8% for predicting AKI. In the study conducted by McMahon et al[11], a score of 5 was identified as the threshold for moderate risk, and a score of 10 was considered high risk of AKI and mortality. In our study, the optimal McMahon score threshold for predicting AKI was 4.75, which appears lower than thresholds reported in previous studies[11,21]. This value was determined using the Youden index derived from the ROC curve, aiming to identify the score that best balances sensitivity (78.6%) and specificity (76.9%) for AKI prediction. Several methodological and clinical factors may explain this lower threshold. First, the prospective design of our study allowed for systematic and early identification of AKI through predefined monitoring protocols. Second, all patients were managed in an ICU, where continuous hemodynamic and biochemical surveillance is standard practice. This likely facilitated the early recognition of renal dysfunction, potentially lowering the score required to detect AKI. Third, the clinically homogeneous nature of our study population, composed exclusively of ICU patients, may have reduced interindividual variability in AKI risk. This uniformity could have influenced the distribution and predictive performance of the McMahon score in our population. Finally, the relatively small sample size and single-center design may have contributed to variability in threshold estimation. Nonetheless, our findings suggest that a McMahon score threshold < 5 may be helpful for early risk stratification and clinical decision-making in patients with rhabdomyolysis admitted to the ICU. In our multivariate analysis, the McMahon score was not identified as an independent predictor of AKI. Instead, elevated LDH and low serum bicarbonate levels emerged as independent predictors of AKI in the context of rhabdomyolysis. These findings raise two important issues that deserve to be addressed. The first concerns the initial serum CPK level, which is considered relevant in the McMahon score only when it exceeds 40000 IU/L. In our study, none of the patients reached this threshold, and as a result, CPK did not contribute to the score calculation. However, we observed that even moderate elevations in CPK were significantly associated with AKI occurrence. This suggests that CPK may retain prognostic value below the 40000 IU/L threshold in critically ill patients. Moreover, it is well established that CPK levels can continue to rise several hours after rhabdomyolysis onset, potentially limiting their early predictive value when the score is calculated at admission[22]. Clinicians should therefore be cautious when interpreting low McMahon scores in patients with moderate CPK elevations, as these may still indicate a substantial risk of renal complications. These considerations support the idea that modeling CPK as a continuous variable could enhance the predictive accuracy of risk scores in this context. The second point concerns the potential utility of LDH as a predictive biomarker for AKI in the setting of rhabdomyolysis. Although its role has been relatively underexplored, we believe that LDH merits greater attention. LDH is a nonspecific yet sensitive marker of tissue and cellular injury, and its elevation may reflect early renal involvement. Recent studies have begun to highlight a possible association between elevated LDH and AKI in rhabdomyolysis[23-26], suggesting that it could serve as an early indicator of renal stress in this context. However, data remain limited, and the heterogeneity in study designs and definitions of AKI complicates direct comparisons. This underscores the need for prospective, large-scale investigations to validate the prognostic value of LDH and to define its added predictive utility alongside established clinical scores. Rhabdomyolysis was historically associated primarily with traumatic injuries, but recent studies have highlighted a wide range of non-traumatic causes, many of which confer a comparable risk of AKI[2,19,27]. In the present study, we observed a higher frequency of non-traumatic etiologies among patients who developed AKI. However, this association did not remain statistically significant in multivariate analysis, likely due to the limited sample size. The literature presents conflicting findings regarding the relationship between rhabdomyolysis etiology and AKI risk, with variations largely influenced by study design and population characteristics[19,20]. Although our study is limited by its sample size and single-center design, it reinforces the importance of early recognition of rhabdomyolysis in patients with non-traumatic risk factors, such as sepsis, metabolic disturbances, or drug-induced muscle injury. These patients should undergo prompt laboratory evaluation for rhabdomyolysis and be closely monitored for early signs of AKI. Further multicenter studies are needed to clarify whether specific etiologies are independently associated with a higher risk of AKI and to assess their contribution to the predictive performance of risk models.

This study has some limitations that should be acknowledged. First, its single-center design and relatively small sample size may limit the generalizability of the findings to broader populations. Second, because the study was conducted exclusively in an ICU setting, it excluded patients with severe rhabdomyolysis who did not require intensive care, thereby narrowing the applicability of our results to less critically ill populations. Additionally, serum and urinary myoglobin levels were not measured, and renal biopsies were not performed, which limits the ability to definitively confirm rhabdomyolysis as the direct cause of AKI. Although the diagnosis was established based on elevated CPK levels and a compatible clinical context, the absence of myoglobin data may have affected diagnostic precision. Furthermore, given the well-recognized nephrotoxic potential of myoglobin, this limitation may have restricted our ability to investigate mechanistic pathways or assess its prognostic relevance. Future studies should aim to include myoglobin measurements to enhance diagnostic accuracy and refine risk prediction in rhabdomyolysis-associated AKI. Nonetheless, the prospective design and comprehensive, systematic assessment of potential contributing factors strengthen the internal validity of the study and help mitigate the impact of these limitations. Such data are often underreported or missing in retrospective studies, making them difficult to capture. Additionally, the prospective design enabled standardized clinical management and follow-up. To the best of our knowledge, this is the first prospective investigation specifically designed to identify factors associated with rhabdomyolysis-related AKI.

CONCLUSION

In this prospective study of ICU patients from a North African center, routinely available biomarkers, including LDH and serum bicarbonate, were significantly associated with the risk of rhabdomyolysis-associated AKI. Defining optimal threshold values for these parameters may enhance early risk stratification. These findings suggest that simple, inexpensive tests performed at admission may assist nephrologists and ICU physicians in identifying high-risk patients early. This is particularly valuable in resource-limited countries where advanced biomarkers such as cystatin C or NGAL are not routinely available. Clinical protocols for patients with rhabdomyolysis should incorporate early assessment of AKI risk using such accessible markers. Our results also underscore the importance of validating the McMahon score in diverse populations and clinical settings. In addition, the emerging role of LDH as a potential predictor of AKI warrants further investigation. Future multicenter studies with larger cohorts are needed to refine predictive models and determine whether integrating less conventional biomarkers can improve risk stratification and outcomes in rhabdomyolysis.

ACKNOWLEDGEMENTS

We would like to express our sincere gratitude to the medical and nursing staff of the Nephrology and Intensive Care Units at Rabta University Hospital for their valuable collaboration and continuous support throughout the study.