Adams JE, 1997, USA | Patients with suspected cardiac trauma | Review article including 3 relevant papers | Serial TnT and total CK | Sensitivity 0.63 and specificity 0.71 for TnT | Excluded those with pre-existing cardiac disease |
CK and TnI over first 72 hours and ECHO | Sensitivity 1 and specificity 0.68 for TnI | Very small numbers involved |
| | | TnI, CK and CK-MB at 12 and 24 hours and ECHO | Specificity 0.72 for TnI TnI specificity better than TnT specificity | |
| | | TnT v TnI | | |
Ferjani M et al, 1997, France | 128 consecutive patients who had suffered blunt chest trauma. All patients had TnT measured at admission, 4 and 24 hours after admission. Cardiac contusion defined as abnormal echocardiography compatible with contusion, severe cardiac rhythm abnormality, severe cardiac conduction abnormality or haemopericardiumecutive patients who had suffered blunt chest trauma | Prospective observational study | Sensitivity and specificity of troponin T >0.5 ug over 1st 24 hours | ROC curve analysis performed AROC = 0.69 with 95% C.I. of 0.56 to 0.80 | Excludes patients with pre-existing coronary artery disease. |
| | | Troponin versus CKMB | TnT of >0.5 ug has specificity of 0.91 but sensitivity of 0.31, indicating that it is unreliable | Only measured TnT at admission, 4 and 24 hours. Used TnT not TnI trop of >0.5 ug is a high level. |
| | | | Trop T had greater AROC than CKMB or CKMB/CK ration | |
Mori F et al, 2001, Italy | 32 patients with clinical or radiological signs of acute blunt chest trauma. All patients had cTnI measured at 6, 12, 24, 48 and 96 hours post injury. Cardiac contusion defined as abnormal trans-oesophageal echocardiography | Prospective observational study | Sensitivity of raised troponin | All with normal or minimally raised TnI (<0.4 ng/ml) had a normal echo (mean 0.6 +/−1.4) | Excluded patients with pre-existing cardiac disease |
| | | Specificity of raised troponin | Mean TnI was higher in those with abnormal echo (mean 2.6 +/−1.6) p<0.0001 | Small numbers included in the study. |
Kaye P et al, 2002, UK | Patients with suspected myocardial contusion. ECG and ECHO used to define significant blunt cardiac injury. | Review article including 3 relevant papers | Utility of troponin to diagnose myocardial contusion | Animal studies suggest troponin may be useful. 2 Human studies show high sensitivity, 1 shows low spensitivity (but had questionable gold standard). | Papers used variable gold standards, abnormal ECG, clinically significant finding and/or ECHO. |
| | | | Authors concluded cardiac troponins seem to be highly specific and sensitive for myocardial injury. They may offer the possibility of stratification of severity and risk of adverse outcome. | 1 paper looked at ventilated trauma patients and none had an abnormal echo |
| | | | | Small numbers involved in the trials |
Velmahos G et al, 2003, USA | 333 consecutive patients with significant blunt thoracic trauma. TnI was performed on all patients at admission, 4 and 8 hours post admission. Significant blunt cardiac injury was determined by any of the following: hypotension in the absence of bleeding or a neurogenic cause, cardiac arrhythmia, echocardiagraphic abnormality, severe arrhythmi,a or shock of unexplained origin | Prospective observational study | Clinical diagnosis of significant blunt cardiac injury. Serial ECG and TnI analysis | None with normal ECG and TnI at 8 hours were felt to have significant blunt cardiac injury. TnI was considered abnormal if values were greater than 1.5 ng/mL | The diagnosis of significant blunt cardiac injury was made clinically. |
| | | | | High cut off for raised TnI (1.5 ng/ml) |
Sybrandy KC et al, 2003, Netherlands | Patients with suspected cardiac contusion | Review article including 2 further relevant papers | Utility of troponins to detect myocardial contusion | Sensitivity 100%, all with normal TnI had no problems. Specificity 83–87.5% | One paper excluded intubated and haemodynamically unstable patients. |
| | | | | Small numbers involved in the trials |