Exertional Rhabdomyolysis: A Case Series of 30 Hospitalized Patients

ABSTRACT

Introduction: Exertional rhabdomyolysis is a clinical entity of significant muscle breakdown in the setting of exercise. However, clinical course and discharge criteria, once hospitalized, are poorly described. We describe 30 cases of exertional rhabdomyolysis and their hospital course. Methods: Thirty hospitalized cases with ICD-9 code of 722.88 (rhabdomyolysis) as the primary diagnosis were reviewed from 2010 to 2012. We excluded those with associated trauma, toxin, and heat illnesses. Results: The average length of stay was 3.6 days (range: 1–8 days). Length of stay correlated significantly with peak creatine kinase (CK) levels. The mean admission CK was 61,391 U/L (range 697–233,180 U/L). The mean discharge CK was 23,865 U/L with a wide range (1,410–94,665 U/L). Six cases (20%) had evidence of acute kidney injury, but most had serum creatinine (Cr) <1.7 mg/dL. One had a peak Cr of 4.8 mg/dL. Higher serum Cr levels correlated significantly with lower CK levels. Twenty-nine out of 30 patients were discharged when CKs downtrended. Conclusion: Higher peak CK levels predicted longer length of stay. Higher serum Cr significantly correlated with lower CK levels. There did not appear to be any threshold CK for admission or discharge, however, all but one patient were discharged after CK downtrended.

INTRODUCTION

Exertional rhabdomyolysis (ER) is muscle breakdown associated with vigorous exercise and is well described in the military population.^1,–3 Although there is not a universally accepted definition for ER, it is often defined as a clinical syndrome associated with severe muscular pain, tea-colored urine, and elevations of serum creatine kinase (CK). Complications include acute kidney injury (AKI), electrolyte disturbances, and compartment syndrome. Although it may be associated with exertional heat stroke and heat illnesses, it is also a distinct clinical entity. Hospitalization for ER is often warranted for patients with significant CK elevations and those at risk for complications.^4,–6

Once hospitalized, typical inpatient management includes aggressive intravenous hydration and monitoring for the complications of rhabdomyolysis. Recent studies describe the natural history of CK levels with exercise and the relationship with renal function in clinically asymptomatic individuals.^7,8 However, the natural history of hospitalized patients with ER is not well described in the literature. A 1974 case series described 40 Marines hospitalized with ER after 3 days of physical training,³ but this report may not represent current management. This retrospective case series study was undertaken to describe the length of stay, CK trends, complications, and overall hospital course for active duty service members admitted with acute ER in an effort to inform clinicians of the natural history and elucidate potential discharge criteria.

METHODS

Tripler Army Medical Center is a tertiary care military hospital located in Honolulu, Hawaii, that serves as the primary inpatient facility for over 40,000 active duty personnel.⁹ A database of patients admitted to Tripler for rhabdomyolysis (ICD-9 code 728.88) from January 1, 2010 through December 31, 2012 was obtained from the Military Health System Management Analysis and Reporting Tool (M2) database. Because there is not a separate ICD-9 code for exertional versus non-ER, chart review determined if the admission was defined as ER and not because of toxin, ingestions, or trauma. Patients with rhabdomyolysis in association with heat stroke or exertional heat illness were excluded. Study investigators adjudicated all cases of ER via consensus. A total of 30 charts were included for final analysis. The study protocol was approved by the Human Use Committee at Tripler Army Medical Center. Investigators adhered to the policies for protection of human subjects as prescribed in 45 Code of Federal Regulation 46.

RESULTS

Thirty cases of ER were identified from 2010 to 2012. (See Table I for details on each case.) 94% were male and 6% were female. The average age was 25.7 years (range 21–37). The rest of the demographics are listed in Table II.

Mechanism of Injury

Military training (physical fitness training, ruck marches) was implicated as the cause of ER for 12 of the 30 cases (40%). Physical training was implicated in 7 cases (23%) and ruck marches in 5 cases (17%).

Other nonspecific training such as weight lifting, exercise of increased intensity, new fitness programs, martial arts, and exercise with a personal trainer accounted for 43% (n = 13) of the cases.

In the remaining 5 of 30 cases of ER (17%), high intensity workouts such as CrossFit (n = 4) and P90x (n = 1) were implicated.

Supplement Use

Only 17% (n = 5) reported any supplement use. Protein powders and creatine were the most commonly recorded.

Hospital Data

Length of Stay and Complications

The average length of stay was 3.6 days with a range of 1 to 8 days. There were 6 cases (20%) of AKI (AKI, defined by peak creatinine (Cr) > 0.3 mg/dL compared to the discharge Cr), according to the Acute Kidney Injury Network criteria.¹⁰ There was one patient who had a peak Cr of 4.8, whose hospitalization was complicated by pulmonary edema (see Table I, case no. 8). The other 5 cases of AKI (17%) had a peak Cr < 2.0 mg/dL (average 1.50 md/dL). There were no episodes of compartment syndrome.

CK Levels

See the hospital data reported in Table III for details on average CK levels for admission, peak, discharge, and at the first clinic follow-up. All but one patient (case no. 19) were discharged after demonstrating a decrease in CK from the peak. Of the 30 cases, 25 had a CK documented in an outpatient setting after being discharged. There were no complications noted at posthospitalization follow-up except for one episode of intravenous (IV) site associated cellulitis (case no. 4). There were no readmissions for ER noted within the study dates.

Exploratory Correlations

Correlations were explored to identify a relationship between length of stay, CK levels, and renal function (BUN, Cr).

Relationship Between Length of Stay and CK levels

On exploratory correlation, there was a strong positive correlation between length of stay and peak CK levels (Spearman's ρ = 0.637; p < 0.0001) and moderate correlation with admit CK levels (Spearman's ρ = 0.458; p = 0.011)

Relationship Between Cr and CK

Admit Cr was strongly negatively correlated with admit CK (Spearman's ρ = −0.700; p < 0.0001) and moderately negatively correlated with peak CK (Spearman's ρ = −0.575; p = 0.001) and discharge CK (Spearman's ρ = −0.460; p = 0.011)

Peak Cr was strongly negatively correlated with admit CK (Spearman's ρ = −0.658; p < 0.001) and moderately negatively correlated with peak CK (Spearman's ρ = −0.545; p = 0.002) and discharge CK (Spearman's ρ = −0.415; p = 0.023).

Discharge Cr was moderately negatively correlated with admit CK (Spearman's ρ = −0.515; p = 0.004) and peak CK (Spearman's ρ = −0.476; p = 0.008) but not with discharge CK (Spearman's ρ = −0.289; p = 0.121).

DISCUSSION

In this case series we found that clinicians were admitting patients diagnosed with ER for an average of 3.6 days with a mean admission CK of 61,391 U/L (range 697–233,180 U/L). Clinicians did not appear to utilize a standard CK level to determine discharge. The mean discharge CK was 23,865 U/L with a wide range (1,410–94,665 U/L). However, on exploratory analysis, clinicians appear to be influenced by CK levels, as length of stay was strongly correlated with peak CK levels. More importantly, it appears that a downtrending CK was an important discharge marker, as 29 of 30 patients were discharged after their CK downtrended. There did not appear to be a significant trend in a specific cause of ER. However, it is notable that 40% (n = 12) of the cases were attributed to routine Army training such as physical training and ruck marches.

There is a paucity of studies in the current literature to help guide the clinician once a patient is hospitalized for ER, and there is little consensus guiding the clinician on appropriate discharge timing. In general, patients are admitted to prevent and monitor for the complications of AKI and compartment syndrome. Hospital treatment generally consists of IV fluid hydration to keep urine output > 200 mL/h. There are some authorities that recommend alkalinization of the urine to help prevent AKI in rhabdomyolysis, but in our study none of the patients were managed with mannitol or bicarbonate during their hospitalization. In 1974, Demos described 40 Marines who were admitted for ER.³ In that study, the mean peak CK level was only 4,600 U/L with an average length of stay of 18 days and therefore may not be consistent with current knowledge and management of ER. In another early report, Demos also described 11 Marines hospitalized for 14 to 28 days with ER, whose mean CK on the second day of admission was 76,803 U/L (range of 26,708–109,609 U/L).¹¹ However, this study is limited since there were no reports of admission, peak, and discharge CK, or data on renal function. Expert opinions vary widely regarding CK levels and discharge. Several sources recommend hospitalization for rhabdomyolysis until CK drops to <1,000 U/L.^12,13 Other experts have argued that discharge at higher CK thresholds of up to 20,000 to 50,000 U/L can be safely done from the emergency room.^4,5 Case reports describe patients with ER and markedly elevated CK levels without the need for hospital admission. One study described a 16-year-old boy with ER and a CK of 181,690 U/L who was safely managed with oral hydration alone.¹⁴ Another case report described a 37-year-old emergency medicine physician who had ER with peak CK of 70,158 U/L treated on an outpatient basis without complication.¹⁵ In our study, the mean discharge CK was 23,865 U/L, with a wide range (1,410–94,665 U/L). In follow-up, all 25 cases that had a follow-up CK drawn showed decreased levels after discharge. There were no readmissions for ER and there were no complications noted, other than a report of cellulitis over the IV site, which is more a complication related to hospitalization versus a complication attributed specifically to ER. Overall, our study suggests that there does not appear to be a threshold CK level for discharge.

Although discharge CK levels did not appear to be related to complications, this study is small and generalizations are limited. There was only one complication on discharge follow-up, but it was not related to a known complication of rhabdomyolysis. There were no readmissions for rhabdomyolysis or complications from rhabdomyolysis that we know of, however there were 5 patients lost to follow-up.

On exploratory correlation, we found strong correlations with peak CK and overall length of stay. It is likely that in patients with higher CKs, the levels of CK took longer to peak and thus increased the length of stay. Also, clinicians may have felt comfortable discharging a patient when a CK reached a certain threshold. In all cases except one, the CK was tracked until it downtrended before a patient was discharged. We are not able to ascertain, however, what threshold, if any, determined safe discharge. 16 out of 22 patients (72%) who had peak CK > 32,000 U/L were discharged once CK reached below the 32,000 U/L threshold. In our institution, CK levels > 32,000 U/L are not diluted for absolute CK count unless specifically requested. Although not written in any specific hospital protocol, some physicians waited for the CK level to drop under 32,000 U/L, so when outpatient follow-up occurred, the receiving physicians would be able to interpret the follow-up CKs without incurring another dilution step.

Prevention or treatment of AKI is one of the main reasons that patients are hospitalized with ER. Although it has been reported that higher CK levels do not necessarily confer a higher risk for AKI, the concern continues to exist. In our study up to 20% had evidence of AKI. Of those 6 cases, there was one complication of significant AKI, with a peak Cr of 4.8 mg/dL. This case had CK levels on admission, peak, and discharge of 1,166 U/L, 8,268 U/L, and 3,905 U/L, respectively. Although there were 5 other cases that met the definition of AKI, none had any rise of serum Cr > 1.67 mg/dL and 4 of 5 patients had peak CK levels of <30,000 U/L. On exploratory correlation, it appears that the higher CK levels (both admission and at peak) were not associated with worsening Cr levels. In fact, in our study, lower CK levels were correlated with higher Cr levels. This may suggest that the higher the CK, the more the aggressive physicians were in clearing CK with IV hydration. In addition, it may be that patients were more likely admitted for mild AKI rather than the concern for ER, which may have skewed admission for patients with lower CKs. Although this relationship appears paradoxical, it is reassuring that peak CK did not portend AKI.

There are significant limitations to this study. First, this case series of ER, although the largest recent study that we know of, consists of associations. Second, since this was done at one institution, institutional practice likely guided similar treatment plans among clinicians. Although there are no official guidelines for the treatment of ER, the treatments are likely similar and therefore may not be generalizable to other institutions. For example, in our institution, myoglobin is a send-out test and is not useful to obtain clinically. When it is ordered, the patients are often discharged before receiving the results. Also looking at the raw CK levels, our laboratory limits the dilution to 2×, and therefore reports of CK >32,000 U/L are not uncommon. Further dilution is done at request of the clinician, but our data is limited by this factor. However, it likely underestimates the overall CK levels.

In conclusion, we described 30 hospitalized cases of ER from a large military tertiary care center. We hope this data can provide the clinician useful information about the natural history and course of patients admitted with ER. We did not find an appropriate CK level for discharge, and there appears to be no significant relationship between peak CK and worsening renal function. This case series adds evidence that there may be no appropriate CK levels to predict or prevent complications of ER. Paradoxically, CK levels were inversely related to serum Cr levels and this relationship should be explored further. Further research should examine a larger sample of patients admitted for ER.

REFERENCES

Footnotes

Author notes