Interpretable Machine Learning Model for Early Prediction of Mortality in ICU Patients with Rhabdomyolysis.

PURPOSE: Rhabdomyolysis (RM) is a complex set of clinical syndromes that involves the rapid dissolution of skeletal muscles. Mortality from RM is approximately 10%. This study aimed to develop an interpretable and generalizable model for early mortality prediction in RM patients. METHOD: Retrospective analyses were performed on two electronic medical record databases: the eICU Collaborative Research Database and the Medical Information Mart for Intensive Care III database. We extracted data from the first 24 h after patient ICU admission. Data from the two data sets were merged for further analysis. The merged data sets were randomly divided, with 70% used for training and 30% for validation. We used the machine learning model extreme gradient boosting (XGBoost) with the Shapley additive explanation method to conduct early and interpretable predictions of patient mortality. Five typical evaluation indexes were adopted to develop a generalizable model. RESULTS: In total, 938 patients with RM were eligible for this analysis. The area under the receiver operating characteristic curve (AUC) of the XGBoost model in predicting hospital mortality was 0.871, the sensitivity was 0.885, the specificity was 0.816, the accuracy was 0.915, and the F1 score was 0.624. The XGBoost model performance was superior to that of other models (logistic regression, AUC = 0.862; support vector machine, AUC = 0.843; random forest, AUC = 0.825; and naive Bayesian, AUC = 0.805) and clinical scores (Sequential Organ Failure Assessment, AUC = 0.747; Acute Physiology Score III, AUC = 0.721). CONCLUSIONS: Although the XGBoost model is still not great from an absolute performance perspective, it provides better predictive performance than other models for estimating the mortality of patients with RM based on patient characteristics in the first 24 h of admission to the ICU.

Rhabdomyolysis is Associated with In-Hospital Mortality in Patients with COVID-19.

PURPOSE: Rhabdomyolysis (RM) has been associated with many viral infectious diseases, and associated with poor outcomes. We aim to evaluate the clinical features and outcomes of RM in patients with coronavirus disease 2019 (COVID-19). METHOD: This was a single-center, retrospective, cohort study of 1,014 consecutive hospitalized patients with confirmed COVID-19 at the Huoshenshan Hospital in Wuhan, China, between February 17 and April 12, 2020. RESULTS: The overall incidence of RM was 2.2%. Compared with patients without RM, those with RM tended to have a higher risk of deterioration. Patients with RM also constituted a greater percentage of patients admitted to the intensive care unit (90.9% vs. 5.3%, P < 0.001) and a greater percentage of patients undergoing mechanical ventilation (86.4% vs. 2.7% P < 0.001). Moreover, patients with RM had laboratory test abnormalities, including the presence of markers of inflammation, activation of coagulation, and kidney injury. Patients with RM also had a higher risk of in-hospital death (P < 0.001). Cox's proportional hazard regression model analysis confirmed that RM indicators, including peak creatine kinase levels > 1,000 IU/L (HR = 6.46, 95% CI: 3.02-13.86) and peak serum myoglobin concentrations > 1,000 ng/mL (HR = 9.85, 95% CI: 5.04-19.28), were independent risk factors for in-hospital death. Additionally, patients with COVID-19 that developed RM tended to have delayed viral clearance. CONCLUSION: RM might be an important contributing factor to adverse outcomes in COVID-19 patients. The early detection and effective intervention of RM may help reduce mortality among COVID-19 patients.

A MELAS Patient Developing Fatal Acute Renal Failure with Lactic Acidosis and Rhabdomyolysis.

We herein present a patient with mitochondrial myopathy, encephalopathy, lactic acidosis, and stroke-like episodes (MELAS), who developed serious acute renal failure with lactic acidosis, followed by rhabdomyolysis. Despite receiving intensive care, he suffered multiple cardiopulmonary arrests and died 10 days after presentation due to a sudden deterioration of his symptoms. Renal pathology revealed diffuse tubular necrosis with interstitial edema and tubular dilatation on light microscopy, and a severe degeneration of intracellular organelles on electron microscopy. These pathological findings could have resulted from multiple cardiopulmonary arrests; however, we must be aware of the extremely rare but sudden occurrence of these fatal conditions in MELAS patients.

Rhabdomyolysis: a 10-year retrospective study of patients treated in a medical department.

BACKGROUND: Rhabdomyolysis is a common and potentially life-threatening syndrome, and acute kidney injury (AKI) is a serious complication. We performed a 10-year retrospective study that included all patients treated for rhabdomyolysis in a medical clinic. We examined the relationships between the levels of creatine kinase (CK), myoglobin, and creatinine (as a marker of renal function and thereby AKI), and whether the myoglobin/CK ratio could be a valuable tool in the clinical evaluation of this patient group. Clinical characteristics were noted. PATIENTS AND METHODS: The study included all patients treated for rhabdomyolysis in the Department of Medicine, Oslo University Hospital Ulleval, from 2003 to 2012. Rhabdomyolysis was defined as a serum CK activity more than five times the upper reference limit. RESULTS: A total of 341 patients were included in the study; 51% developed AKI, and 20% of those required dialysis. Logistic regression showed that myoglobin concentration [P < 0.001, odds ratio (OR) = 6.24] was a better predictor than CK activity (P = 0.001, OR = 3.45) of the development of AKI. The myoglobin/CK ratio was a good predictor of AKI (P < 0.001, OR = 5.97). The risk of developing AKI increased with increasing myoglobin/CK ratio (P < 0.001); a ratio more than 0.2 was associated with an increased likelihood of developing AKI. CONCLUSION: Serum myoglobin concentration was a better predictor of AKI than was serum CK activity. The myoglobin/CK ratio may be useful for assessing the likelihood of developing AKI.

Rhabdomyolysis observed at forensic autopsy: a series of 52 cases.

Rhabdomyolysis is characterized by skeletal muscle injury resulting in the release of intracellular proteins (such as myoglobin) and electrolytes into the blood circulation, which cause acute kidney injury, myoglobinuria and electrolyte imbalances. Clinical diagnosis of rhabdomyolysis is made on the basis of biochemical analysis; however, for forensic autopsies, biochemical data are often not available, and it is necessary to diagnose rhabdomyolysis via histopathological examinations. This study analyzed 52 cases with rhabdomyolysis and applied myoglobin immunohistochemistry to kidney, urine and blood samples. We found that blunt force injuries were the most common cause of rhabdomyolysis across all age groups, and drugs were the second most common cause. The drugs included ketamines, amphetamines, synthetic cathinones, entheogens, benzodiazepines, opioid analgesics, and anesthesia. Less than 60% of our cases had biochemical data, including myoglobin (92.5~416,978 ng/mL), creatine kinase (220~774,015 U/L), potassium (1.6~10.3 meq/L), calcium (2.7~29.2 mg/dL), and phosphorus (2.6~14.2 mg/dL). In the kidney tissue sections, we found that 95% of the rhabdomyolysis cases were positive for myoglobin immunohistochemistry and that 96% were associated with acute tubular necrosis. Our findings describe the features of fatal rhabdomyolysis in a large series and suggest that myoglobin immunohistochemistry can be used in post-mortem blood and urine cell blocks to detect myoglobin.

Variables clínicas asociadas al desarrollo de rabdomiólisis y sus complicaciones en ancianos: un estudio de cohorte retrospectivo / Clinical variables associated with rhabdomyolysis and its complications in elderly patients. A retrospective cohort study

No disponible

Rhabdomyolysis in a Tertiary PICU: A 10-Year Study.

OBJECTIVES: Rhabdomyolysis is a disorder of muscle breakdown. The aim of this study was to describe the epidemiology of rhabdomyolysis in children admitted to a PICU and to assess the relationship between peak creatinine kinase and mortality. DESIGN: Retrospective cohort study in children admitted to the PICU with rhabdomyolysis between January 1, 2005, and December 31, 2014. Demographic, clinical, and outcome data were recorded. Outcomes were analyzed by level of peak creatinine kinase value (0-10,000, 10,001-50,000, > 50,000IU/L). Long-term renal outcomes were reported for PICU survivors. SETTING: A single-centre academic tertiary PICU. PATIENTS: Children admitted to the PICU with serum creatinine kinase level greater than 1,000 IU/L. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: There were 182 children with rhabdomyolysis. The median peak creatinine kinase value was 3,583 IU/L (1,554-9,608). The primary diagnostic categories included sepsis, trauma, and cardiac arrest. Mortality for peak creatinine kinase values 0-10,000, 10,001-50,000, and > 50,000 IU/L were 24/138 (17%), 6/28 (21%), and 3/16 (19%), respectively (p = 0.87). Children with a peak creatinine kinase greater than 10,000 IU/L had a longer duration of mechanical ventilation and ICU length of stay than children with peak creatinine kinase less than 10,000. Renal replacement therapy was administered in 29/182 (16%). There was longer duration of mechanical ventilation (273 [141-548] vs. 73 [17-206] hr [p < 0.001]) and ICU length of stay (334 [147-618] vs. 100 [37-232] hr (p < 0.001)] in children receiving renal replacement therapy. Continuous veno-venous hemofiltration was the most common modality 23/29 (79%). Only one child required renal replacement therapy postintensive care stay, and adverse long-term renal outcomes were uncommon. CONCLUSIONS: In children with rhabdomyolysis requiring intensive care, peak creatinine kinase was not associated with mortality but is associated with greater use of intensive care resources. Chronic kidney disease is an uncommon sequelae of rhabdomyolysis in children requiring intensive care.

Rhabdomyolysis after co-administration of a statin and fusidic acid: an analysis of the literature and of the WHO database of adverse drug reactions.

Following a severe case of rhabdomyolysis in our University Hospital after a co-administration of atorvastatin and fusidic acid, we describe this interaction as this combination is not clearly contraindicated in some countries, particularly for long-term treatment by fusidic acid. All cases of rhabdomyolysis during a co-administration of a statin and fusidic acid were identified in the literature and in the World and Health Organization database, VigiBase® . In the literature, 29 cases of rhabdomyolysis were identified; mean age was 66 years, median duration of co-administration before rhabdomyolysis occurrence was 21 days, 28% of cases were fatal. In the VigiBase® , 182 cases were retrieved; mean age was 68 years, median duration of co-administration before rhabdomyolysis was 31 days and 24% of cases were fatal. Owing to the high fatality associated with this co-administration and the long duration of treatment before rhabdomyolysis occurrence, fusidic acid should be used if there is no appropriate alternative, as long as statin therapy is interrupted for the duration of fusidic acid therapy, and perhaps a week longer. Rarely will interruption of this sort have adverse consequences for the patient.

Causes of Troponin Elevation and Associated Mortality in Young Patients.

BACKGROUND: While increased serum troponin levels are often due to myocardial infarction, increased levels may also be found in a variety of other clinical scenarios. Although these causes of troponin elevation have been characterized in several studies in older adults, they have not been well characterized in younger individuals. METHODS: We conducted a retrospective review of patients 50 years of age or younger who presented with elevated serum troponin levels to 2 large tertiary care centers between January 2000 and April 2016. Patients with prior known coronary artery disease were excluded. The cause of troponin elevation was adjudicated via review of electronic medical records. All-cause death was determined using the Social Security Administration's death master file. RESULTS: Of the 6081 cases meeting inclusion criteria, 3574 (58.8%) patients had a myocardial infarction, while 2507 (41.2%) had another cause of troponin elevation. Over a median follow-up of 8.7 years, all-cause mortality was higher in patients with nonmyocardial infarction causes of troponin elevation compared with those with myocardial infarction (adjusted hazard ratio [HR] 1.30; 95% confidence interval [CI], 1.15-1.46; P < .001). Specifically, mortality was higher in those with central nervous system pathologies (adjusted HR 2.21; 95% CI, 1.85-2.63; P < .001), nonischemic cardiomyopathies (adjusted HR 1.66; 95% CI, 1.37-2.02; P < .001), and end-stage renal disease (adjusted HR 1.36; 95% CI, 1.07-1.73; P = .013). However, mortality was lower in patients with myocarditis compared with those with an acute myocardial infarction (adjusted HR 0.43; 95% CI:, 0.31-0.59; P < .001). CONCLUSION: There is a broad differential for troponin elevation in young patients, which differs based on demographic features. Most nonmyocardial infarction causes of troponin elevation are associated with higher all-cause mortality compared with acute myocardial infarction.

Hyperkalemia in Combat Casualties: Implications for Delayed Evacuation.

OBJECTIVE: Fixed facilities and rapid global evacuation ensured that delayed complications of trauma, such as hyperkalemia, occurred late in the evacuation chain where renal replacement therapies were available. However, future conflicts or humanitarian disasters may involve prolonged evacuation times. We sought to quantify one potential risk of delayed evacuation by assessing hyperkalemia in combat casualties. METHODS: Retrospective study of military members admitted to intensive care units in Iraq and Afghanistan from February 1, 2002, to February 1, 2011. This study was approved by the U.S. Army Medical Research and Materiel Command Institutional Review Board. Demographics, injury severity score, burn injury, mechanism of injury, vital signs, creatinine, and potassium were collected. Logistic regression models were used to identify incidence and risk factors for hyperkalemia. RESULTS: Of 6,011 patient records, 1,472 had sufficient data to be included for analysis. Hyperkalemia occurred in 5.8% of patients. Those with hyperkalemia had higher injury severity scores, higher shock index, were more likely to have acute kidney injury, and were more likely to die. On multivariate analysis, acute kidney injury and shock index were significantly associated with the development of hyperkalemia. In a subgroup of patients with data on creatine kinase, rhabdomyolysis was associated with hyperkalemia in the univariate model, but was not significant after adjustment. CONCLUSION: Hyperkalemia occurred in 5.8% of patients in our cohort of critically injured combat casualties. The development of hyperkalemia was independently associated with acute kidney injury and shock index. In future conflicts, with prolonged evacuation times, mitigation strategies should be developed to treat hyperkalemia in casualties before arrival at definitive care.

Risk Factors for Rhabdomyolysis in the U.S. Army.

The standardized mortality rate of rhabdomyolysis (RM) in Active Duty U.S. Army Soldiers is considerably higher than in the civilian population. RM occurs when large amounts of intracellular contents from damaged skeletal muscle escape into circulation, leading to serious sequelae (e.g., acute renal failure, hyperkalemia, compartment syndrome). Extended physical exertion, especially in hot environments, and trauma can precipitate RM. The aim of this study was to identify RM risk factors among U.S. Active Duty Army (ADA) Soldiers. METHODS: This nested case-control study used data from the Total Army Injury and Health Outcomes Database (years 2004-2006) to examine RM among ADA male Soldiers. Demographic and occupational variables were identified as potential risk factors. Each RM case was age and date-matched to 4 controls. Adjusted odds ratios (OR) were computed using conditional logistic regression analyses. RESULTS: From years 2004 to 2006, 1,086 Soldiers (0.19%) met the study criteria for clinically diagnosed RM. Three variables were found to increase the odds of acquiring RM: (1) prior heat stroke, OR 4.95 (95% confidence interval [CI] 1.1-21.7); (2) self-reported Black race, OR 2.56 (95% CI 2.2-3.0); and (3) length of service (0-90 days), OR 2.05 (95% CI 1.6-2.7). CONCLUSION: There is a substantially greater likelihood for male U.S. Army Soldiers to develop RM who: (1) have had a prior heat injury, (2) self-report in the Black racial category, and (3) who are within the initial 90 days of service. Greater awareness of the risk factors associated with RM may improve force health protection and readiness through targeted mitigation strategies.

Accidental hypothermia: factors related to long-term hospitalization. A retrospective study from northern Finland.

Accidental hypothermia has a low incidence, but is associated with a high mortality rate. Knowledge about concomitant factors, complications, and length of hospital stay is limited. A retrospective cohort study on patients with accidental hypothermia admitted to Oulu University Hospital in Finland, over a 5-year period. Patients were categorized as short-stay patients (7 days or less) and long-stay patients (more than 7 days) according to their length of stay in hospital. From a total of 105 patients, 67 patients were included in the analyses. Alcohol abuse was the most common concomitant factor (54 %). Median length of hospital stay was 4 days, and 16 patients (24 %) stayed in hospital over 7 days (median 15 days). Thirty-day mortality was low (14/105, 13 %). Patients with long-term hospitalization had a lower initial temperature (28.4 versus 31.2 °C, p = 0.011), a lower level of consciousness (GCS score 8.4 versus 12.8, p = 0.003), more severe acidosis (pH 7.08 versus 7.28, p = 0.005, and lactate 7.2 versus 3.9, p = 0.043), and a lower level of platelets (183 versus 242, p = 0.041) on admission compared with short-stay patients. Thirty-six patients (54 %) had at least one complication, and this prolonged median hospital treatment for 2.5 days (p < 0.001). Alcohol is the most common concomitant factor and every fourth patient spends more than 7 days in hospital. Long-term hospitalization is related to a lower core temperature, lower consciousness, more severe lactic acidosis, lower platelet level and infections, rhabdomyolysis, and renal failure.

Rapid Diagnosis of Rhabdomyolysis with Point-of-Care Ultrasound.

It is important to rapidly diagnosis and treat rhabdomyolysis in order to decrease morbidity and mortality. To date there are no reports in the emergency medicine literature on the use of point-of-care ultrasound in the diagnosis of rhabdomyolysis. This unique case describes how ultrasound was used in the emergency department (ED) to quickly diagnose and treat rhabdomyolysis prior to confirmation with an elevated serum creatine kinase. When coupled with a high index of suspicion, ultrasound can be one of the most portable, readily available, low cost, and minimally invasive techniques for making a rapid diagnosis of rhabdomyolysis in the ED.

Earlier diagnosis and strict diets improve the survival rate and clinical course of long-chain 3-hydroxyacyl-CoA dehydrogenase deficiency.

AIM: Long-chain 3-hydroxyacyl-CoA dehydrogenase deficiency (LCHADD) is a severe metabolic disease that, without treatment, often leads to premature death or serious handicap. The aim of this study was to evaluate the clinical course of LCHADD with the homozygous 1528G>C (E510Q) mutation when patients underwent strict dietary treatment. METHODS: From 1997 to 2010, 16 patients with LCHADD were diagnosed in Finland. They were followed up, and data were prospectively collected as they emerged. Clinical data before diagnosis were retrospectively collected from hospital records. This cohort was compared with an earlier cohort of patients diagnosed from 1976 to 1996. RESULTS: The disease presented from birth to five months of age with failure to thrive, hypotonia, hepatomegaly, metabolic acidosis, cardiomyopathy and hypoketotic hypoglycaemia. In this cohort, the therapeutic delay was 0-30 days and the survival rate at the end of the study was 62.5% compared with 10-year survival rate of 14.3% for the earlier cohort. The survivors were in good overall condition, but some of them had developed mild retinopathy or mild neuropathy. CONCLUSION: Earlier diagnosis and stricter dietary regimes improved the survival rates and clinical course of patients with LCHADD in Finland. However, improvements in therapy are still needed to prevent the development of long-term complications, such as retinopathy and neuropathy.

The Contemporary Incidence and Sequelae of Rhabdomyolysis Following Extirpative Renal Surgery: A Population Based Analysis.

PURPOSE: We evaluate the contemporary incidence and consequences of postoperative rhabdomyolysis after extirpative renal surgery. MATERIALS AND METHODS: We conducted a population based, retrospective cohort study of patients who underwent extirpative renal surgery with a diagnosis of a renal mass or renal cell carcinoma in the United States between 2004 and 2013. Regression analysis was performed to evaluate 90-day mortality (Clavien grade V), nonfatal major complications (Clavien grade III-IV), hospital readmission rates, direct costs and length of stay. RESULTS: The final weighted cohort included 310,880 open, 174,283 laparoscopic and 69,880 robotic extirpative renal surgery cases during the 10-year study period, with 745 (0.001%) experiencing postoperative rhabdomyolysis. The presence of postoperative rhabdomyolysis led to a significantly higher incidence of 90-day nonfatal major complications (34.7% vs 7.3%, p <0.05) and higher 90-day mortality (4.4% vs 1.02%, p <0.05). Length of stay was twice as long for patients with postoperative rhabdomyolysis (incidence risk ratio 1.83, 95% CI 1.56-2.15, p <0.001). The robotic approach was associated with a higher likelihood of postoperative rhabdomyolysis (vs laparoscopic approach, OR 2.43, p <0.05). Adjusted 90-day median direct hospital costs were USD 7,515 higher for patients with postoperative rhabdomyolysis (p <0.001). Our model revealed that the combination of obesity and prolonged surgery (more than 5 hours) was associated with a higher likelihood of postoperative rhabdomyolysis developing. CONCLUSIONS: Our study confirms that postoperative rhabdomyolysis is an uncommon complication among patients undergoing extirpative renal surgery, but has a potentially detrimental impact on surgical morbidity, mortality and costs. Male gender, comorbidities, obesity, prolonged surgery (more than 5 hours) and a robotic approach appear to place patients at higher risk for postoperative rhabdomyolysis.

Neuroleptic Malignant Syndrome: Complications, Outcomes, and Mortality.

OBJECTIVE: Our study objective was to identify real-world rates of complications, mortality, and outcomes in patients with neuroleptic malignant syndrome (NMS) over the last decade in the United States. METHODS: A total of 1346 patients were obtained from the nationwide inpatient sample for the years 2002-2011. Common complications known to be associated with NMS were identified. Multivariable regression analyses were used to identify predictors of mortality. RESULTS: The most prevalent complication was rhabdomyolysis (30.1%). Other common complications were acute respiratory failure (16.1%), acute kidney injury (17.7%), sepsis (6.2%), and other systemic infections. Unadjusted mortality rate was 5.6%. Older age, acute respiratory failure, acute kidney injury, sepsis, and comorbid congestive heart failure were significant predictors of mortality. Acute respiratory failure was the strongest independent mortality predictor (p < 0.001). CONCLUSION: In our large sample population-based study on NMS, we were able to identify the rates of several preselected complications and the mortality. The identification of independent mortality predictors in this study can guide physicians in the management and prognostication of this rare syndrome.

Rhabdomyolysis among critically ill combat casualties: Associations with acute kidney injury and mortality.

BACKGROUND: Rhabdomyolysis has been associated with poor outcomes in patients with traumatic injury, especially in the setting of acute kidney injury (AKI). However, rhabdomyolysis has not been systematically examined in a large cohort of combat casualties injured in the wars in Iraq and Afghanistan. METHODS: We conducted a retrospective study of casualties injured during combat operations in Iraq and Afghanistan who were initially admitted to the intensive care unit from February 1, 2002, to February 1, 2011. Information on age, sex, Abbreviated Injury Scale (AIS) score, Injury Severity Score (ISS), mechanism of injury, shock index, creatine kinase, and serum creatinine were collected. These variables were examined via multivariate logistic and Cox regression analyses to determine factors independently associated with rhabdomyolysis, AKI, and death. RESULTS: Of 6,011 admissions identified, a total of 2,109 patients met inclusion criteria and were included for analysis. Rhabdomyolysis, defined as creatine kinase greater than 5,000 U/L, was present in 656 subjects (31.1%). Risk factors for rhabdomyolysis identified on multivariable analysis included injuries to the abdomen and extremities, increased ISS, male sex, explosive mechanism of injury, and shock index greater than 0.9. After adjustment, patients with rhabdomyolysis had a greater than twofold increase in the odds of AKI. In the analysis for mortality, rhabdomyolysis was significantly associated with death until AKI was added, at which point it lost statistical significance. CONCLUSION: We found that rhabdomyolysis is associated with the development of AKI in combat casualties. While rhabdomyolysis was strongly associated with mortality on the univariate model and in conjunction with both ISS and age, it was not associated with mortality after the inclusion of AKI. This suggests that the effect of rhabdomyolysis on mortality may be mediated by AKI. LEVEL OF EVIDENCE: Prognostic and epidemiologic study, level III.