Risk factors for exertional rhabdomyolysis with renal stress

Subjects

This is a case–control study based on a cluster of eight military recruits who were admitted to the hospital due to exertional rhabdomyolysis with gross pigmenturia. The incident occurred about 1 week after arrival in camp. All the new recruits had been divided into squads for their physical training. Importantly, the various squads had exercise programmes that differed somewhat in regard to duration, intensity and composition.

Day 0 was the day when the patients first got muscular symptoms with pigmenturia. Days −1 and −2 were before muscular injury, and days 1–24 in the time thereafter.

The control group consisted of 26 randomly selected healthy recruits from the same military camp. The healthy subjects provided blood samples while the patients still were in the hospital (day 3).

Diagnostic criteria for rhabdomyolysis with pigmenturia

There should be appropriate clinical findings and serum CK above 20 000 U/L. The urine should be positive for blood on dipstick test, but without haematuria on urine microscopy.

Treatment

The patients were treated according to current guidelines.13 On arrival, they were promptly volume-expanded with intravenous isotonic saline, combined with drinking water, causing a mean urine output of 5.0 L per day during the hospital stay. The urine was alkalinised with bicarbonate tablets and the mean urine pH was 6.9±0.8 during the stay. Patients were discharged when their condition had improved and myoglobinuria no longer could be detected with urine dipstick test. The mean duration of the hospital treatment was 4.9±0.8 days.

Clinical and demographic data collection

Both patients and controls responded to a questionnaire about their current intake of drugs, tobacco, alcohol, diet and physical training level prior to entering military service. In the same way, information was obtained about exercise schedule and medical symptoms in the days before the rhabdomyolysis cluster occurrence. Data about body weight, height and conscription board examinations were extracted from the military medical record.

Prior to military service at conscription board examination, aerobic capacity and muscular strength were assessed with a maximal treadmill running test and two maximum isometric tests for arms and legs.14 The results were expressed in a semiquantitative scale from 1 to 9, with 1 representing lowest and 9 best results.

In the period of muscular injury, the subjects reported their exercise type on the questionnaire: A: aerobic training; S: strength training; M: mixed training; No: no exercise. Exercise duration was reported in minutes. Exercise intensity was reported on a subjective scale from 1 to 10, with 1 being the lowest and 10 the highest intensity.

Laboratory methods

All biochemical analyses were performed in serum.

Cystatin C was analysed with a particle-enhanced turbidimetric immunoassay using reagents from Gentian (www.gentian.no) on a Modular P800 analyser (Roche Diagnostics, Mannheim, Germany). Coefficient of variation (CV) was ≤5%. Laboratory reference interval is 0.51–1.05 mg/L.

Myoglobin was analysed with the electrochemiluminescent immunoassay Elecsys (Roche Diagnostics) on a Modular E analyser (CV ≤7%). Laboratory reference interval is 30–70 µg/L for men.

NGAL was analysed by enzyme immunoassay (ELISA) Quantikine (R&D Systems, Stillwater, Minnesota, USA). The assay had been validated, and kit independent controls were run on every test plate. The interassay (n=5) and intra-assay (n=10) variation coefficients were 4.8% and 2.3%, respectively.

C reactive protein (CRP) was analysed by turbidimetric immunoassay using a Roche Modular kit on a Modular analyser, with CV ≤7.5%. Laboratory reference interval is 0–4 mg/L.

CK was analysed by a photometric assay using the Architect Creatine Kinase Reagent Kit (Abbott Laboratories, Abbott Park, Illinois, USA) on an Architect c16000 analyser (Abbott Laboratories), with CV ≤6.5%. Laboratory reference interval is 50–400 U/L for men between 18 and 49 years.

Creatinine concentrations were obtained by an enzymatic method using the Multigent Creatinine Assay (Sentinel CH, Milan, Italy) on an Abbott Architect c16000 analyser, with CV ≤3.6%. Laboratory reference interval is 60–105 µmol/L for men above 17 years.

Estimated glomerular filtration rate (eGFR) was estimated with the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) creatinine equation.15

Statistical analysis

Age, body mass index (BMI), exercise duration, myoglobin, NGAL, CRP, creatinine, eGFR and cystatin C were normally distributed and analysed with appropriate Student’s t-tests. Student’s t-tests were two-tailed. The Mann-Whitney U test and the Wilcoxon signed-rank test were applied for evaluating differences in preservice training level, muscular strength, aerobic capacity, exercise intensity and CK. Spearman rank correlation coefficient was used to explore the relationship between the duration of muscular pain and peak levels of CK and myoglobin. The χ² test was used for testing categorical differences such as gender, tobacco habit, dietary additives, alcohol use and exercise type. Simple, multiple and logistic regression were employed for exploring the relationship between different risk factors and rhabdomyolysis or biochemical markers of tissue injury. p≤0.05 was considered significant. The statistical calculations were performed using StatView V.5.0.

Demographics

The patients and the controls were of the same age. One control subject was of Asian origin; the others were Caucasian. Two controls were female; the others were men (table 1). We found no significant differences between patients and controls in regard to age, diet, tobacco, alcohol and training level before entering military service (table 1). During the first 2 weeks of military service, the recruits were not allowed to leave the camp and their intake of alcoholic beverages was considered moderate.

Biochemical markers of muscular injury

The eight patients had a peak serum CK concentration of 126 788±66 649 U/L (mean±SD), range 48 861–202 522 U/L. Their peak serum myoglobin concentration was 8811±6360 µg/L (mean±SD), range 1698–21 298 µg/L.

The patients had significantly higher concentrations of CK and myoglobin compared with the controls on day 3 (table 2, figure 1). There was a close correlation between CK and myoglobin concentrations (Spearman’s r 0.93, p<0.0001). On day 24, the CK and myoglobin concentrations were significantly reduced from peak levels (table 2, figure 1).

Request permission

On day 3, the 26 controls had a mean CK concentration of 1553 U/L, which was above the laboratory reference interval, that is, 40–400 U/L (table 2). Of these, seven controls (27%) had CK below 400 U/L, and three had CK above 5000 U/L. The mean myoglobin concentration in controls was 133 µg/L, which was above the laboratory reference interval, that is, 30–80 µg/L (table 2). Of these, 11 controls (42%) had myoglobin concentrations within the laboratory reference interval.

Muscular pain

The patients reported muscular pain and tenderness for a duration of 7–9 days (mean 8.4±0.7 days), while the control subjects had significantly shorter duration of pain: 0–9 days (mean 2.1±3.4 days, p<0.0001). Seventeen controls (65%) did not report muscular pain. Moreover, the patients reported higher pain intensity than controls. There was a significant, positive correlation between the peak concentrations of CK and myoglobin and the number of days with muscular pain (Spearman’s r 0.5, p=0.004 and Spearman’s r 0.6, p=0.001, respectively).

Risk factors for rhabdomyolysis with myoglobinuria

Prior to military enrolment, patients and controls had similar aerobic capacity; however, the patients had less isometric muscular strength than the control subjects (table 1). The patients reported significantly higher exercise intensity during the 3 days before the injury cluster occurred—this was most evident on day −1 (Monday). Furthermore, the patients reported significantly longer exercise duration on day −1, and more strength training on day 0 (Tuesday) (table 3). The strength training involved a significant amount of eccentric exercises. Importantly, all patients reported that they had kept on with the high-intensity exercises despite significant muscular pain.

On multivariate analysis (logistic multiple regression), a maximal level of exercise intensity on day −1 was the only independent predictor of rhabdomyolysis with myoglobinuria (p=0.02).

Biochemical markers of kidney injury

The patients had stable and normal serum creatinine of 85±12, 77±7 and 82±8 µmol/L (mean±SD) on day 1, day 3 and day 24, respectively. There was no difference in creatinine, cystatin C and eGFR between patients and controls (table 2). On day 24, all patients had a normal albumin/creatinine ratio in their urine (mean 0.5±0.3 mg/mmol).

On day 3, the patients had significantly higher serum concentrations of NGAL than the controls. Three weeks later, NGAL concentrations had decreased significantly from peak levels (table 2, figure 1).

Causes of exertional rhabdomyolysis

Previous reports have been somewhat conflicting on the relative importance of work intensity, duration and volume.16 In the present study, exercise intensity, duration and type were all significant determinants of exertional rhabdomyolysis in univariate models. The patients reported higher intensity at all time-points during the 3-day period prior to hospitalisation for rhabdomyolysis. However, in a multivariate model, exercise intensity on day −1 appeared as the only independent predictor of exertional rhabdomyolysis. At this time-point all of the patients reported an intensity score of 9 or 10. This may indicate that the risk of rhabdomyolysis is present primarily at the highest exercise intensity levels, whereas lower levels carry less risk of exertional rhabdomyolysis. Moreover, it appears to be possible to quantify the risk of rhabdomyolysis by a simple questionnaire.

Patients and controls had the same aerobic capacity at enrolment for military service, but patients had lower muscular strength. This finding suggests that patients were at a good fitness level, but had less experience with muscular training, and therefore more prone to overuse injury. The present study provides no definite answers, but it is possible that a combination of personal ambition, external pressure and insufficient understanding of the risks paved the way for exertional muscular injury.

The findings need to be interpreted with caution due to small sample size. However, there are few confounding factors: Both patients and controls were previously healthy and had similar background in regard to age, gender, BMI, fitness and ethnicity. We did not test for the asymptomatic sickle cell trait or genetic muscle diseases that may predispose for rhabdomyolysis.17 18 However, the patients did not report previous episodes of rhabdomyolysis and had muscular strength within normal limits. Therefore, we think that underlying genetic disorders were of no importance in the present study. Furthermore, there was little negative influence from ambient temperature, alcohol, legal or illicit drugs, toxins or metabolic disorders.13 Admittedly, recall bias may be possible.

Prevention of exertional rhabdomyolysis

The present findings suggest some possible targets for the prevention of exertional rhabdomyolysis. However, it is important to emphasise that these measures need to be tested in future prospective studies.

First, prevention of exertional rhabdomyolysis requires increased awareness and education, both directed against individuals who participate in physical exercise and others who supervise physical exercise.

Second, the use of a self-perceived muscular intensity scale may be a useful aid to guide the intensity level of physical exercise.

Third, complementary use of CK concentrations may be of help in case of uncertainty. A mildly elevated CK level in serum is common after physical exercise and may be regarded as a normal response. Mougios 19reported an adjusted laboratory reference interval for male athletes during training and competition: 82–1083 U/L. Serum CK concentrations rising above 5000 U/L indicate significant muscle injury with risk of acute kidney injury.2 20 In the present study, the patients had peak CK levels 10–50 times higher than this limit.

Biomarkers of acute kidney injury

Myoglobin is released from injured muscle and filtered by the kidneys, imparting a dark red-brown colour to the urine.21 The iron-containing myoglobin has toxic effects in the kidney, such as vasoconstriction, cast formation and proximal tubular injury.1 High levels of serum myoglobin appear to predict acute kidney injury, especially with concentrations over 15 000 µg/L.22

In the present study, the eight patients had high serum myoglobin levels with potential risk of kidney injury, but maintained a stable kidney function. It appears that myoglobinuria seldom causes acute kidney failure in patients without volume depletion.1

In clinical practice, kidney function is evaluated by means of serial creatinine levels. However, serum creatinine may be regarded a slow and insensitive biomarker as concentrations do not rise above normal until a significant amount of kidney function has been lost.23 Cystatin C may be somewhat better for detecting changes in glomerular filtration rate.24 In the present study, all the patients had stable and normal GFR, as estimated by both creatinine and cystatin C.

The new biomarker NGAL appears to be related to tubular injury or stress.25 After acute renal injury the tubular cells undergo a complex sequence of events. As a part of this, NGAL is rapidly upregulated and concentrations increase markedly in urine and plasma.26 Therefore, NGAL has been advocated as an early and more sensitive marker of acute kidney injury.27 28 NGAL concentrations may be influenced by other factors such as pre-existing renal disease and infections. The best results for NGAL as a marker of acute kidney injury have been demonstrated in young populations without comorbidities and with a well-defined timing of the pathogenic event, such as in the present study.27 NGAL has diverse functions in the kidney; it may modulate cellular responses, have iron-binding properties and may serve to limit tubular damage.29 We observed elevated serum levels of NGAL in our patients, and this probably indicates stress and/or structural damage of the tubular cells. Three weeks after the injury, NGAL concentrations had decreased from peak levels and there was no evidence of permanent kidney injury.

In conclusion, sustained high-intensity physical activity, exceeding muscular functional capacity, is a crucial risk factor for exertional rhabdomyolysis. During physical exercise it may be difficult to distinguish between ordinary muscular training soreness and alarming muscular pain at the initial stages. However, the present findings indicate that it may be possible to quantify the risk of rhabdomyolysis by a simple questionnaire.

Although traditional markers of renal function often remain within normal limits in exertional rhabdomyolysis with gross pigmenturia, elevated serum NGAL may indicate the presence of renal stress or subclinical injury.

In the future, measures should be taken to prevent exertional rhabdomyolysis by providing information and awareness of the condition to persons who participate in or supervise physical exercise.

Limitations

A rhabdomyolysis questionnaire will need to be validated in a larger prospective study with calculation of sensitivity and specificity and receiver operator characteristic analysis. Furthermore, specific questions should be added to exclude underlying genetic diseases.