Validation of cardiologists' decisions to initiate reperfusion therapy foracute myocardial infarction with electrocardiograms viewed on liquid crystal displays of cellular telephones

Background The transmission of 12-lead electrocardiograms from remote locat ions to hand-held computers of cardiologists is now possible with the devel opment of wireless technology and computer software. This investigation det ermined whether the cardiologist's decisions regarding reperfusion therapy for patients with symptoms suggestive of an acute myocardial infarction are the same when given electrocardiograms displayed on a cellular telephone a s on a standard paper recording. Methods Cardiologists were given 20 electrocardiograms of patients with acu te chest pain suggestive of acute myocardial infarction to test the diagnos tic reliability of the Nokia 9000i cellular telephone liquid crystal displa y (LCD) screen. The cardiologists made their decision to initiate or not in itiate reperfusion therapy for the patients after viewing their electrocard iograms displayed on both 5-mm and l-mm formatted grids and twice on tradit ional printout electrocardiograms. The control level of intraobserver agree ment between the responses from the 2 sets of paper display electrocardiogr ams was compared with the experimental level of intraobserver agreement bet ween the l-mm LCD electrocardiograms and both sets of paper display electro cardiograms to determine whether the viewing medium affected the cardiologi st's decisions. The l-mm and 5-mm ICD screen electrocardiograms were compar ed to determine if the grid size affected the cardiologist's decisions. Results Ninety-three percent of the 2 sets of paper-guided decisions were i n agreement. When comparing the l-mm LCD-guided decisions with both sets of paper-guided decisions, 94% and 89% of the decisions, respectively, were i n agreement The differences between the control and experimental degrees of intraobserver agreement of 1% and 4% were not statistically significant (P 1 =.81, P2 =.29). Ninety-one percent of the l-mm LCD-guided decisions were in agreement with the 5-mm LCD-guided decisions. Conclusions Cardiologists' decisions did not vary significantly when viewin g either traditional paper electrocardiograms or LCD screen electrocardiogr ams. Even though there was not a significant difference in the cardiologist s' decisions when they viewed electrocardiograms displayed on both the l-mm and 5-mm grid, it is recommended that the l-mm grid be used for clinical i mplementation of the ICD screen.