Predicting adverse hemodynamic events in critically ill patients.

PURPOSE OF REVIEW: The art of predicting future hemodynamic instability in the critically ill has rapidly become a science with the advent of advanced analytical processed based on computer-driven machine learning techniques. How these methods have progressed beyond severity scoring systems to interface with decision-support is summarized. RECENT FINDINGS: Data mining of large multidimensional clinical time-series databases using a variety of machine learning tools has led to our ability to identify alert artifact and filter it from bedside alarms, display real-time risk stratification at the bedside to aid in clinical decision-making and predict the subsequent development of cardiorespiratory insufficiency hours before these events occur. This fast evolving filed is primarily limited by linkage of high-quality granular to physiologic rationale across heterogeneous clinical care domains. SUMMARY: Using advanced analytic tools to glean knowledge from clinical data streams is rapidly becoming a reality whose clinical impact potential is great.

Incidence of myocardial infarction or stroke or death at 47-month follow-up in patients with diabetes and a predicted exercise capacity 85% during an exercise treadmill sestamibi stress test.

A treadmill exercise sestamibi stress test (TESST) was performed in 609 consecutive diabetic persons with a mean age of 70 years and no history of coronary artery disease (CAD) who were referred for a TESST because of chest pain or dyspnea. Of 609 patients, 301 (49%) had a predicted exercise capacity 85% (group B). Group A patients had a higher prevalence of myocardial ischemia (43% vs 30%, P=.0005), 2- or 3-vessel obstructive CAD (38% vs 18%, P=.001), myocardial infarction (17% vs 9%, P=.004), death (10% vs 4%, P=.008), and myocardial infarction or stroke or death at 47-month follow-up (21% vs 12%, P=.001). Stepwise Cox regression analysis showed that the only significant independent predictor for the time to development of myocardial infarction or stroke or death was a predicted exercise capacity >85% (hazard ratio, 0.52; 95% confidence interval, 0.34-0.78; P=.002). Diabetic persons with a predicted exercise capacity >85% had a 48% lower chance of myocardial infarction, stroke, or death than those with a predicted exercise capacity

Risk factors for major bleeding and for minor bleeding after percutaneous coronary intervention in 634 consecutive patients with acute coronary syndromes.

Of 634 consecutive patients who had percutaneous coronary intervention (PCI) for acute coronary syndromes, 34 (5%) had major bleeding after PCI, 253 (40%) had minor bleeding after PCI, and 347 (55%) had no bleeding after PCI. Significant independent risk factors for major bleeding after PCI were increased troponin I level (P = 0.004; odds ratio [OR] = 4.7), prior coronary artery disease (P = 0.029; OR = 3.7), platelet glycoprotein IIb/IIIa inhibitors (P = 0.002; OR = 9.8), glomerular filtration rate (GFR) <30 versus >or=60 mL/min/1.73 m(2) (P < 0.0001; OR = 39.7), GFR 30-59 versus >or=60 mL/min/1.73 m(2) (P = 0.0001; OR = 9.4), and clopidogrel loading dose >300 mg (P = 0.0001; OR = 8.9). Significant independent risk factors for minor bleeding after PCI were increased troponin I level (P = 0.0004; OR = 2.1), platelet glycoprotein IIb/IIIa inhibitors (P = 0.039; OR = 2.4), GFR 30-59 versus >or=60 mL/min/1.73 m(2) (P < 0.0001; OR = 2.5), thrombolytics (P = 0.01; OR = 2.7), clopidogrel loading dose >300 mg (P < 0.0001; OR = 4.2), and systolic blood pressure during PCI (P < 0.0001; OR = 1.03 per mm Hg). In-hospital deaths included 5 of 34 patients (15%) with major bleeding, none of 253 patients (0%) with minor bleeding, and none of 347 patients (0%) with no bleeding (P < 0.0001). Hospital duration was 11.0 days in patients with major bleeding, 3.4 days in patients with minor bleeding, and 1.8 days in patients with no bleeding (P < 0.0001).

Relation of bone mineral density to stress test-induced myocardial ischemia.

Dual-energy x-ray absorptiometric scans of the spine and left hip were performed before stress testing for myocardial ischemia in 629 women and 136 men (mean age 63 years) with chest pain and no previous coronary artery disease. Of the 765 patients, 254 (33%) had osteoporosis, 260 (34%) had osteopenia, and 251 (33%) had normal bone mineral density (BMD). Stress test-induced myocardial ischemia was present in 95 of 254 patients (37%) with osteoporosis, in 81 of 260 patients (31%) with osteopenia, and in 62 of 251 patients (25%) with normal BMD (p = 0.002 comparing osteoporosis with normal BMD and p = 0.007 comparing osteoporosis or osteopenia with normal BMD). Stepwise logistic regression analysis showed that patients with osteoporosis or osteopenia had a 1.7 times higher chance of stress test-induced myocardial ischemia than those with normal BMD after controlling the confounding effects of systemic hypertension, diabetes mellitus, body mass index, and age. In conclusion, patients with chest pain undergoing stress testing have a higher prevalence of stress test-induced myocardial ischemia if they have osteoporosis or osteopenia than if they have normal BMD.