Rhabdomyolysis of infectious and noninfectious causes. (Review Articles).

ABSTRACT

Background. This study was done to determine variables associated with infectious rhabdomyolysis (IRM).

Methods. In this retrospective case-control study, rhabdomyolysis (RM) was defined as a fivefold or greater elevation in creatine kinase (CK) levels with a muscle/brain (MB) fraction <5%. Patients with myocardial infarction or cerebrovascular accident or a recent history of surgery, trauma, or immobilization were excluded.

Results. We analyzed 52 cases of RM seen at our institution between 1992 and 2000; IRM was the most frequent cause (31%), most commonly respiratory tract infections (38%). When a microorganism could be identified (50%), it was more often gram-negative (63%). Patients with IRM were elderly and had fever and lower CK levels. Infectious rhabdomyolysis was associated with a higher morbidity but not with a higher risk of death.

Conclusions. Infectious rhabdomyolysis is the main cause of RM and must be suspected in elderly patients with fever and low levels of CK.

**********

RHABDOMYOLYSIS (RM) is a potentially lethal disorder that occurs as a primary disease or as a complication of a broad spectrum of other diseases. (1) Rhabdomyolysis is characterized by elevated serum concentrations of creatine kinase (CK) due to skeletal muscle injury. Infectious diseases are among the numerous factors precipitating RM. (2,3) The aims of this study were to determine the causes of IRM and to analyze the clinical and biologic characteristics of IRM as opposed to noninfectious rhabdomyolysis (NIRM).

PATIENTS AND METHODS

Using hospital discharge data, we retrospectively reviewed all cases of RM diagnosed between 1992 and 2000. Patients were divided into infectious and noninfectious groups. The laboratory marker used to define RM was that proposed by Gabow et al (1) (a fivefold or greater elevation in CK levels with a MB fraction <5%). Patients with evidence of myocardial infarction or cerebrovascular accident and those with a recent history of surgery, trauma, immobilization, or intramuscular injection were excluded from the study.

The following covariates were included: age, sex, etiologic factors (infectious vs noninfectious RM), clinical findings (fever, myalgias, weakness), peak of the analytic parameters (CK, CK-MB, leukocyte count, creatinine, aspartate aminotransferase [AST], alanine aminotransferase [ALT], and lactate dehydrogenase [LDH]), morbidity (dialysis) and mortality, and microorganisms identified. Patients could be classified in only one category (IRM or NIRM), so charts that listed both causes were excluded from review. Death was attributed to the infectious process if the patient died within 10 days of the bacteremic episode with a clinical course suggesting persistent infection or if death occurred during the phase of acute infection.

The data were entered and categorized with EXCEL software (Microsoft). Mean values are presented throughout with standard deviation. Student's t test was used to compare group means. Descriptive data were compared by chi-square testing. Statistical analysis was done using SPSS software (SPSS Inc, Chicago, Ill). Statistical significance was defined as P < .05.

RESULTS

Of the 52 cases of RM included in this study, 16 (31%) were due to infection. Table 1 lists all etiologic factors. Table 2 shows patient characteristics and clinical findings.

The overall mean age was 60.1 years (range, 16 to 93 years) and was significantly higher among patients with IRM (70.8 vs 55.3; P = .03). When we analyzed cases by sex, we found a strong predominance of IRM in men (69% of IRM vs 92% of NIRM; P = .04). Among the clinical findings (fever, myalgias, and weakness), fever was the only one associated with IRM (P < .001). The only analytic parameter associated with IRM was CK, which was lower in patients with IRM (P < .01). Table 3 shows serum chemistry values for both groups. None of the other parameters analyzed were associated with IRM (CK-MB, leukocyte count, creatinine, AST, ALT, and LDH).

In patients with IRM, the main origin was respiratory tract (38%), followed by urinary tract (25%). Microorganisms were identified in 8 cases (50%), and in 5 of these the isolate was gram-negative. There were no polymicrobial infections. Table 4 shows the microorganisms isolated from those 8 patients.

Finally, IRM was associated with a higher morbidity (dialysis) (13% vs 0%; P = .03) but not with a higher risk of death.

DISCUSSION

Although alcohol, trauma, and seizures are the most common disorders associated with RM, (1) the relationship between infectious diseases and muscular lesions has been proved. Influenza virus, human immunodeficiency virus, and coxsackievirus are the most common viral etiologies, and Legionella sp, Francisella tularensis, Streptococcus sp, and Salmonella sp are the most common bacterial precipitants. (2)

The reported incidence of IRM is from 5% to 19.4%, (1,3,4) respiratory tract infections being the main cause. In our study, infection was the cause of 16 of the 52 cases of RM (31%), and the respiratory tract was the most common site of infection. There are several explanations for the higher percentage of IRM in our study. First, our selection criteria excluded trauma and other surgical complications. Second, CK is one of the analytic parameters included in the evaluation protocol for all patients admitted to our hospital, thus leading to increased detection of RM. Third, an improved rate of culture and identification of microorganisms could be a factor, though microbiologic identification was accomplished in only half of our 16 cases of IRM. Other possible explanations could be geographic distribution and individual susceptibility to the microorganisms. Q fever is a zoonosis associated with RM, and to date this association has been reported only in Spain. (5-7)

We found an increased incidence of IRM related to advanced age and male sex. Previous investigators have also reported a higher risk of RM among males, (1) so it is possible that men have a predisposition to the disease. Studies of trends in infectious diseases have shown higher morbidity and mortality rates in males. (8,9) In contrast, other authors have found a higher incidence of IRM among females. (3) Younger patients usually have NIRM, possibly because of their greater physical activity or because of alcohol or drug abuse, (1,3) findings that are more common among those less than 35 years old. Since exertion and alcohol or drug use can produce greater muscular destruction, this could explain why patients with IRM had lower levels of CK (P < .01). During the initial evaluation of RM, muscular symptoms and signs are not definitive, (1) and fever is usually the only sign that make us suspect IRM.

It is known that morbidity (dialysis) and mortality from RM is independent of the presence of acute renal failure. (1) Renal failure associated with RM is due to several factors. Myoglobin obstructs tubules and is a direct renal toxin. Cortical ischemia and decreased glomerular filtration are also injuries, as well as the concomitant hypovolemia. (2) Neither peak CK levels nor initial creatinine values correlated with the development of renal failure in one study, (1) but Betrosian et al (3) reported acute renal failure in patients with higher CK levels. In our study, patients with IRM had lower levels of CK (P < .01) but a higher morbidity (dialysis) (P = .03). Betrosian et al (3) observed also that patients with IRM had acute renal failure more frequently than others with infection but not RM.

Acute renal failure is usually a reversible complication when managed by urine alkalinization and forced diuresis, in addition to antibiotics when an infectious disease is suspected. In patients with IRM, acute renal failure is associated with bacterial (57%) more than viral causes (34%). (2) Some authors have found a higher incidence of IRM due to gram-positive organisms (Staphylococcus aureus) vs gram-negative (Pseudomonas aeruginosa) in patients admitted to the ICU, though the majority of microorganisms in the ICU are gram-negative. (3) These authors have also found a higher CK level in the gram-positive group. (3) In our study, the CK level was also higher' in the gram-positive group (3,045 vs 2,878 IU/L), but more of our cases (5/8) were due to gram-negative organisms (Table 4). Perhaps the. absence of microbiologic identification in half of the cases of IRM could explain this difference.

During the past decade, mortality associated with RM has decreased because of expanded knowledge of the disease and its treatment. In 1982, the overall mortality associated with RM was reported as 9.6%. (1) More recently, a higher mortality rate (38%) has been reported among cases of IRM. (2) In our study, we found a higher overall mortality associated with RM (14 patients, 27%) but a similar mortality associated with IRM (6 patients, 38%). We found no differences when we analyzed mortality due to IRM (38%) and NIRM (22%). Betrosian et al (3) considered that RM is not a risk factor for death.

CONCLUSION

Infection is an important cause of RM. Early recognition of IRM is important to avoid severe and potentially lethal complications. Physicians should be especially alert to the possibility of an infectious process in middle-aged and elderly male patients with fever and RM but low CK levels.

TABLE 1.

Etiologic Factors in 52 Cases of Rhabdomyolysis

                                          No. of Cases
                                              (%)

Infections (n = 16)
  Respiratory                                6 (38)
  Urinary                                    4 (25)
  Catheter-related sepsis                    2 (13)
  Others (1 Q fever and 1 leptospirosis)     2 (13)
  Without focus                              2 (13)
Excessive muscular activity (n = 12)
  Sport                                      6 (50)
  Seizures                                   3 (25)
  Delirium tremens                           2 (17)
  Parkinson's disease                        1 (8)
Drugs (n = 10)
  Alcohol                                    5 (50)
  Heroin or cocaine                          3 (30)
  Others (autolitic)                         2 (20)
Metabolic disorders (n = 7)
  Myxedema                                   3 (43)
  Electrolytic misbalance                    4 (57)
Ischemia (n = 5)                             5 (100)
Immunologic disease (n = 1)
  Dermatomyositis                            1 (100)
Genetic disease (n = 1)
  Steinert's disease                         1 (100)
TABLE 2.

Characteristics and Clinical Findings in Infectious Rhabdomyolysis (IRM)
and Noninfectious Rhabdomyolysis (NIRM)

               IRM           NIRM
Variable     No. (%)       No. (%)     P Value

Age (SD)   70.81 (20.2)  55.31 (24.4)    .03
Male sex    11 (68.7)     33 (91.7)      .04
Fever       11 (68.7)      5 (13.9)     .001
Myalgias     8 (50.0)     13 (36.1)       NS
Weakness     5 (31.2)     10 (27.8)       NS
Dialysis     2 (12.5)      0 (0)         .03
Mortality    6 (37.5)      8 (22.2)       NS
TABLE 3.

Peak of Serum Chemistry Concentrations in Patients With Infectious
Rhabdomyolysis (IRM) and Noninfectious Rhabdomyolysis (NIRM)

Variable              IRM       NIRM    P Value

Leukocyte count     12,664.6  11,854.1    NS
Creatinine (mg/dL)       3.0       2.2    NS
CK (IU/L)            3,710.1  19,785.4   .01
CK-MB (%)                1.9       2.2    NS
LDH (IU/L)           1,192.1   2,051.9    NS
AST (IU/L)             291.0     711.6    NS
ALT (LU/L)             228.8     570.0    NS

CK = Creatine kinase, LDH = lactate dehydrogenase, AST = aspartate
aminotransferase, ALT = alanine aminotransferase.
TABLE 4.

Microorganisms Identified in Patients With Infectious Rhabdomyolysis


Gram-negative organisms (n = 5)
 Escherichia coli                2
 Haemophilus sp                  1
 Leptospira sp                   1
 Coxiella burnelii               1
Gram-positive organisms (n = 3)
 Staphylococcus aureus           2
 Streptococcus pneumoniae        1

References

(1.) Gabow PA, Kaehny WD, Kelleher SP: The spectrum of rhabdomyolysis. Medicine 1982; 61:141-152

(2.) Singh U, Scheld WM: Infectious etiologies of rhabdomyolysis: three case reports and review. Clin Infect Dis 1996; 22:642-649

(3.) Betrosian A, Thireos E, Kofinas G, et al: Bacterial sepsis-induced rhabdomyolysis. Intensive Care Med 1999; 265:469-474

(4.) Fernandez-Funez A, Segura Luque JC, Tirado Miranda R, et al: Rabdomiolisis medica en el anciano. Rev Clin Esp 1997; 197:745-751

(5.) Carrascosa M, Pascual F, Borobio MV, et al: Rhabdomyolysis associated with acute Q fever. Clin Infect Dis 1997; 25:1243-1244

(6.) Salcedo J, Blanco JR: Rabdomiolisis aguda por fiebre Q. Enferm Infecc Microbiol Clin 1999; 17:250-251

(7.) Daz F, Collazos J: Rhabdomyolysis in acute Q fever. Clin Microbiol Infect 1999; 5:770-771

(8.) Pinner RW, Teutsch SM, Simonsen L, et al: Trends in infectious diseases mortality in the United States. JAMA 1996; 275:189-193

(9.) Shohat T, Harari G, Green MS: Mortality from infectious diseases in Israel, 1979-1992, based on revised ICD-9 codes: implications for international comparisons. Am J Public Health 1999; 89:1855-1857

RELATED ARTICLE: KEY POINTS

* Infectious rhabdomyolysis (IRM) is the main cause of RM and respiratory infections its main etiology. The main microorganism implicated is gram-negative.

* Among the clinical findings of fever, myalgias, and weakness, fever is the only one associated with IRM.

* In IRM, levels of CK are lower than in other causes of rhabdomyolysis.

* Although IRM is associated with a higher morbidity (dialysis), there is not a higher risk of death.

From the Servicio de Medicina Interna y Documentacion Clinica Complejo Hospitalario San Millan-San Pedro, Lardero, Spain.

Reprint requests to Jose Ramon Blanco, MD, PhD, Avd Juan Carlos I, 90--Bloque C-6[degrees] A, ES-26140 Lardero (La Rioja), Spain.