A prospective study of the impact of serial troponin measurements on the diagnosis of myocardial infarction and hospital and six-month mortality in patients admitted to ICU with non-cardiac diagnoses

Table 4 Comparison between cardiac events recognised by clinical teams and unrecognised events

	Recognised MI (definite or possible MI) (n = 12)	Not recognised MI (definite or possible MI) (n = 47)	unadjusted P-value*	adjusted P-value**
ICU mortality, n (%)	2 (17%)	15 (32%)	0.31	0.22
Hospital mortality, n (%)	3 (25%)	19 (40%)	0.33	0.24
180 days mortality, n (%)	5 (42%)	22 (47%)	0.75	0.52
ICU length of stay, median (IQR)	14.5 (8.5 to 20)	7 (3 to 14)	0.030***	0.024***
Hospital length of stay, median (IQR)	48.5 (35.5 to 63.5)	24 (11 to 42)	0.038***	0.030***

ICU, intensive care unit; IQR, inter-quartile range; MI, myocardial infarction.
*P-value from unadjusted regression model where group was recognised and not recognised.
**P-value from adjusted regression model where group was recognised and not recognised; adjusted variables were age, APACHE II score, hypertension, diabetes and any type of vascular disease (reduced to two principal components in regression models).
***estimates based on log transformed outcome.

ISSN: 1364-8535