Clinical predictors of fever in stroke patients: relevance of nasogastric tube.

OBJECTIVES: Fever frequently occurs in stroke patients and worsens their prognosis. However, only few studies have assessed the determinants of fever in acute stroke, and no study has specifically addressed the possible prediction of the development of fever. MATERIALS AND METHODS: This investigation included 536 patients with acute stroke and a body temperature <=37°C during the first 24 h of stay. Ninety-two of them (17.2%) subsequently developed fever (defined as a temperature >=37.5°C starting after 24 h). Among the clinical variables available during the first 24 h from admission, those predictive of the subsequent appearance of fever were searched for. One hundred further patients had a temperature >37°C during the first 24 h. RESULTS: In univariate analysis, many variables were predictive of the subsequent development of fever, but in multivariate analysis, only the following four predictors remained significant (odds ratio [95% confidence interval], P value): nasogastric tube (4.0 [2.2-7.4], <0.0001), atrial fibrillation (2.3 [1.4-3.8], 0.001), total anterior circulation syndrome (2.0 [1.2-3.5], 0.01), and urinary catheter (1.9 [1.1-3.3], 0.01). Among the 52 (9.7%) patients with three or four predictors, 31 (59.6%) subsequently developed fever. In addition, the factors independently associated with a temperature >37°C during the first 24 h were as follows: National Institutes of Health Stroke Scale (P < 0.0001), hemorrhagic stroke (P = 0.0008), atrial fibrillation (P = 0.002), and total parenteral nutrition (P = 0.03). CONCLUSIONS: In patients with acute stroke, four clinical variables were found to be independently associated with the risk of developing fever and, of them, nasogastric tube was the strongest and most significant one.

A comparison of risk factors as predictors of cardiovascular and non-cardiovascular mortality in the elderly people--relevance of N-terminal pro-B-type natriuretic peptide and low systolic blood pressure.

BACKGROUND: Many risk factors are known to predict ischaemic events and mortality in the elderly people, but their ranking of importance remains uncertain. This study was designed to identify and compare the main predictors of total mortality (TM), cardiovascular mortality (CVM) and non-cardiovascular mortality (NCVM) in older adults. METHODS: Nine hundred and seventy-nine community resident adults aged ≥ 65 years, free of previous heart failure and cardiovascular events, participated in the study. The univariate and multivariate (Cox regression) relationships of baseline cardiovascular risk factors, treatments and laboratory data with TM, CVM and NCVM were assessed after a median follow up of 6.7 years. RESULTS: Overall, there were 104 deaths (30 because of CVM and 74 to NCVM). In multivariate analysis, the following factors remained independently associated with mortality: NT pro-B-type natriuretic peptide (NT-proBNP) upper quintile (≥ 237 pg/ml for men, ≥ 280 pg/ml for women): hazard ratio (HR) vs. the rest of the population (95% confidence interval) 2.34 (1.52-3.60), p < 0.001 for TM; HR 5.41 (2.32-12.65), p < 0.001 for CVM; systolic blood pressure lower quintile (≤ 130 mmHg): HR 3.06 (1.80-5.21), p < 0.001 for NCVM; diabetes: HR 2.46 (1.29-4.72), p = 0.007 for NCVM; erythrocyte sedimentation rate (ESR) upper decile (≥ 41 mm/h): HR 2.33 (1.16-4.69), p = 0.02 for NCVM; platelet count lower quintile (≤ 177 × 10(9) /l): HR 2.09 (1.20-3.64), p = 0.009 for NCVM; ever-smoker status: HR 2.08 (1.23-3.52), p = 0.007 for NCVM. CONCLUSIONS: In elderly community dwellers, NT-proBNP was the strongest predictor of TM and CVM, while especially low systolic blood pressure, together with diabetes, ESR, reduced platelet count and ever-smoker status, were the main predictors of NCVM.

A simple scoring system for outcome prediction of ischemic stroke.

OBJECTIVES: According to most existing models, a computer is usually needed for predicting stroke outcome. Our purpose was to construct a simple and reliable prognostic scale not requiring the use of a calculating machine. MATERIALS AND METHODS: The scale [the Bologna Outcome Algorithm for Stroke (BOAS)] was obtained in 221 patients with ischemic stroke not undergoing thrombolysis and was then validated in a test group of 100 different patients. Outcome was assessed at 9 months as the number of dependent or dead patients (modified Rankin scale - mRS > 2). RESULTS: By a preliminary systematic univariate analysis, 25 of 415 baseline variables were found to be associated with a mRS > 2 independently of stroke severity and age. Subsequent multivariable analyses led to a final model based on five dichotomous risk factors (RF): National Institutes of Health Stroke Scale score ≥10, age ≥78, need of urinary catheter, oxygen administration, and persistence of upper limb paralysis at discharge from stroke unit. The patients with two or more RF (53%) had a mRS > 2 in 91% of cases and were dead in 42% of cases. With 0-1 RF, the two percentages were 24% and 2%, respectively (overall accuracy of prediction 83.9%, area under ROC curve [AUC] 0.891). In the test group, the accuracy was 79.0% and the AUC was 0.839. CONCLUSIONS: The need of urinary catheter, oxygen administration, and persistence of upper limb paralysis, together with stroke severity and advanced age, allow a simple and accurate prediction of dependency or death after ischemic stroke.

The relationship of mean platelet volume with the risk and prognosis of cardiovascular diseases.

BACKGROUND: Mean platelet volume (MPV) is arousing increasing interest as a new independent cardiovascular risk factor. AIM: To provide a comprehensive review on the biological significance, the main determinants and the prognostic implications of MPV. METHODS: A literature search was performed using key terms, such as 'MPV' or 'mean platelet volume', together with 'stroke', 'myocardial infarction' and 'diabetes and 'obesity'. RESULTS: Large platelets are likely more reactive: elevated MPV values are associated with a shortened bleeding time and increased thromboxane B2 plasma levels. Thus, MPV could be considered an indicator of platelet function. Platelet size is mainly determined in the bone marrow during megakaryocytopoiesis, and subsequently does not substantially change. MPV is only partially regulated by thrombopoietin: in fact, growth factors and cytokines may also elicit the production of larger and more reactive platelets in the bone marrow, in the presence of conditions capable of increasing their concentrations, such as obesity, endothelial dysfunction and possibly myocardial and cerebral ischaemia. This phenomenon could play an important role in vascular diseases. In fact MPV is predictive of stroke, acute myocardial infarction (AMI) and restenosis of coronary angioplasty, is increased in the presence of obesity, diabetes mellitus, metabolic syndrome, AMI and stroke and has been shown to have a prognostic significance in patients with stroke and AMI. CONCLUSION: In assessing whole blood count, MPV should not be undervalued, as its increase should suggest a careful assessment of cardiovascular risk.

Physical activity is negatively associated with the metabolic syndrome in the elderly.

BACKGROUND: An inverse association between physical activity and metabolic syndrome has been reported in several cohorts, but very few specific studies are available in the elderly, in whom neurological and musculo-skeletal diseases are expected to lead to a remarkable age-related decline of physical activity. AIM AND DESIGN: The relationships among physical activity, insulin resistance and metabolic syndrome were assessed in a cross-sectional study concerning 1144 subjects aged 65-91 years resident in Pianoro (northern Italy). Household and leisure-time activities were assessed by a self-administered questionnaire (Physical Activity Scale for Elderly--PASE). Routine clinical and biochemical data (including fasting insulin) were used to assess insulin resistance [Homeostasis Model Assessment (HOMA) method] and the prevalence of metabolic syndrome. RESULTS: All PASE scores were inversely correlated with waist circumference, triglycerides and HOMA index, with highest significance for leisure-time activities (P < or = 0.005). The PASE score for household activities was also correlated inversely with blood glucose (P < 0.05), and directly with HDL cholesterol (P < 0.001). In logistic regression analysis, the metabolic syndrome was more prevalent among sedentary subjects (corresponding to the low tertile of leisure-time activities) than in the remaining more active population (odds ratio 1.51, 95% confidence interval 1.12-2.03, P = 0.007), independently of possible confounders. CONCLUSION: Physical activity is inversely associated with insulin resistance and the metabolic syndrome even in the elderly. Community programs favoring physical activity are expected to significantly improve the health status in these subjects.

Genes and atherosclerosis: at the origin of the predisposition.

Atherosclerosis (ATS) is a multifactorial disease caused by the interaction of established or emerging risk factors with multiple predisposing genes that regulate ATS-related processes. This review will discuss the current knowledge concerning the potential role of the genetic variations that could promote and/or accelerate ATS, in both animal models and humans. Allelic polymorphisms or variations of distinct genes that enhance the risk of ATS frequently occur in the general population, but only adequate gene-environment interactions will lead to the disease. The main genes so far studied are involved in the regulation of processes such as endothelial function, antioxidant potential, coagulation, inflammatory response, and lipid, protein and carbohydrate metabolism. The detection of candidate genes associated with ATS could allow, in the near future, earlier interventions in genetically susceptible individuals. Further, large-scale population studies are needed to obtain more information on the specific gene-environment and drug-gene interactions capable of influencing ATS progression.

Peripheral blood markers of inflammation and functional impairment in elderly community-dwellers.

The cross-sectional association of functional impairment with several peripheral blood inflammatory markers (increased C-reactive protein (CRP), fibrinogen and leucocyte count, decreased cholesterol and albumin) was studied in 739 elderly community-dwellers. Functional measures included Tinetti test for gait and balance, and basic and Instrumental Activities of Daily Living. When considering each marker individually, only increased CRP was inversely associated with all functional measures independently of demographics, lifestyle, and comorbidity (P < 0.05). When considering the sum of positive markers, having more than one marker was also inversely associated with all functional measures (P < 0.05), but no clear gradient of impairment was found across increasing numbers of markers. When considering specific combinations of markers, having both increased CRP and at least another positive marker had a stronger association with functional impairment (P < 0.01 for all measures) than increased CRP alone (P > 0.05), or other positive markers alone or in combination (P < 0.05). In conclusion, in elderly individuals, peripheral blood markers of inflammation are associated with functional impairment independently of potential confounders. A specific combination of CRP with other markers provides a better correlate of functional impairment than both individual markers or a simple count of positive markers.

The clock-drawing test in elderly Italian community dwellers: associations with sociodemographic status and risk factors for vascular cognitive impairment.

To evaluate whether performance in the clock-drawing test (CDT) is associated with sociodemographic status and risk factors for cognitive impairment, we examined 744 Italian community dwellers aged > or =65 (73 +/- 6) years scoring > or =24 on the MMSE. CDT was scored by two different methods, the Sunderland and the Wolf-Klein methods. Sociodemographic, lifestyle, clinical and biochemical variables were also recorded. For both scoring methods, lower CDT scores were associated with age, poor education, increased serum C-reactive protein and history of cancer. Associations of lower CDT scores with increased serum glucose and history of cerebrovascular disease were also found for the method with the highest sensitivity to cognitive impairment. This study shows that in elderly community dwellers, CDT may be not totally free from sociodemographic biases, and that it is associated with risk factors for cognitive impairment and frailty.

A comparison of acute phase proteins and traditional risk factors as markers of combined plaque and intima-media thickness and plaque density in carotid and femoral arteries.

OBJECTIVES: To test the hypothesis that some acute phase proteins may be better independent predictors of objective measures of arterial wall impairment than traditional risk factors. DESIGN: Cross-sectional study. MATERIALS AND METHODS: C-reactive protein (CRP), fibrinogen, C3 complement and traditional risk factors were measured in 288 men aged 55-64 years, randomly chosen from the local registry lists. By ultrasound assessment of the bifurcations of carotid and femoral arteries, maximum combined plaque/intima-media thickness (CPIMTmax) and mean plaque density (MPD, in a grey scale from 0 to 255) were also measured. RESULTS: In multivariate analysis only traditional risk factors remained associated with the overall CPIMTmax: smoking (r = 0.35, p < 0.0001), cholesterol (r = 0.23, p = 0.0001), age (r = 0.22, p = 0.0002), glucose (r = 0.18, p = 0.002) and systolic blood pressure (r = 0.13, p = 0.02). However, with regard to carotid disease only, fibrinogen was the strongest covariate of CPIMT (r = 0.18, p = 0.002). The overall MPD was independently associated with CRP (r = 0.25, p = 0.0008), physical activity (r = 0.19, p = 0.009), triglycerides (r = -0.18, p = 0.02) and body mass index (r = 0.15, p = 0.04). CRP was mainly associated with femoral MPD, while triglycerides were the major (inverse) covariate of carotid MPD. CONCLUSIONS: Traditional risk factors are the main determinants of CPIMTmax, although fibrinogen seems to play a role in carotids. CRP was associated with high density femoral plaques. Finally, no acute phase protein was independently associated with low density, potentially vulnerable, plaques.

Different associations of C-reactive protein, fibrinogen and C3 with traditional risk factors in middle-aged men.

BACKGROUND: Some acute phase proteins are associated with both ischemic events and traditional risk factors. Since they are strongly interrelated, each of them partly reflects the characteristics of other proteins. This study was carried out to ascertain the specific preferential associations of some acute phase proteins with traditional risk factors for atherosclerotic disease. METHODS: High-sensitivity C-reactive protein, fibrinogen and C3-complement were assessed in 288 unselected men aged 55-64 years. Three multiple linear regression analyses were performed, in which each of the three acute phase proteins was considered the dependent variable of both traditional risk factors and the other two proteins. RESULTS: The three acute phase proteins strongly correlated with each other. Moreover, C-reactive protein was independently associated with triglycerides (P<0.0001), age (P=0.0130), body mass index (P=0.0179), and acute (P=0.0280) and chronic (P=0.0582) inflammations (R2=0.17). Fibrinogen was associated with alcohol consumption (inversely, P=0.0001) and smoking (P=0.0598) (R2=0.06). Finally, C3 was associated with insulin (P<0.0001), cholesterol (P=0.0001), sedentarity (P=0.0028), glucose (P=0.0077), and systolic blood pressure (P=0.0124) (R2=0.28). CONCLUSIONS: When simultaneously studied in multivariate analysis, acute phase proteins have different preferential associations with traditional risk factors, a probable consequence of their involvement in different cellular activations and metabolic processes.

Conditional risk factors in men with previous myocardial infarction: relevance of C3 and homocysteine.

OBJECTIVE: To establish which traditional and conditional risk factors were effectively treated, and which remained active, in patients with previous myocardial infarction (PMI). METHODS AND RESULTS: In 47 PMI patients recently submitted to cardiological assessment and in 42 controls (50-70 years old men), traditional risk factors (total cholesterol, high-density lipoprotein cholesterol, blood glucose, blood pressure, cigarette smoking and body mass index) and the following variables were measured: fibrinogen, plasminogen activator inhibitor-1 (PAI-1), lipoprotein(a) [Lp(a)], total homocysteine, plasma folates, vitamin B12, high sensitivity C-reactive protein and C3 complement. Most patients were taking beta-blockers, ACE inhibitors and statins. Accordingly, patients had lower blood pressure and cholesterol values than controls. Moreover, they consumed less alcohol and coffee and did not differ from controls in cigarette smoking and body mass index. Conversely, patients had higher levels of homocysteine, fibrinogen, C3 complement and Lp(a), although of these factors only C3 and homocysteine remained significantly associated with PMI in multivariate analysis. C-reactive protein, PAI-1 and especially C3 often correlated with traditional risk factors in controls, but these correlations tended to disappear or reverse in PMI patients. Fibrinogen inversely correlated with alcohol consumption. Homocysteine correlated (inversely) with plasma folates only. Lp(a) did not correlate with any variable. CONCLUSIONS: Forty-seven patients with previous myocardial infarction displayed an excellent control of traditional risk factors, but they had higher mean C3 and homocysteine levels than the control group.

Current thinking in statin therapy.

The HMG-CoA reductase inhibitors (statins) are now considered the most potent lipid-lowering drugs. Treatment with statins reduces both morbidity and mortality rates due to coronary artery disease. There is now increasing evidence that the clinical benefits obtained with statins cannot be solely attributed to a decrease in low-density lipoprotein (LDL) level. These drugs may also have beneficial effects on endothelial dysfunction, LDL oxidation, rheological and thrombogenic factors, cellular inflammation and plaque formation and stability. Further, there are differences among the various statins on these non-lipid variables. The biochemical effects of statins, as well as their clinical benefits, should be taken into consideration.

HMG-CoA reductase inhibitors: Is the endothelium the main target?

Endothelial dysfunction is an early event in atherosclerosis and could be considered a response to the injury induced by major risk factors. There is evidence that endothelial dysfunction is intimately involved in the onset and the progression of cardiovascular disease through abnormalities in the production, release or degradation of endothelium-derived factors, mainly nitric oxide and endothelin 1. Several reports have shown that drugs of the statin class could have multiple beneficial effects related to endothelium-mediated vasoactive, antithrombotic, antiproliferative and anti-inflammatory actions. Thus, the question arises of whether endothelial cells are the main target of statin therapy, in the setting of both hypercholesterolemia and normocholesterolemia. Experimental and clinical studies are reported that could support this hypothesis.

Short term effect of atorvastatin and vitamin E on serum levels of C3, a sensitive marker of the risk of myocardial infarction in men.

C3 complement is produced in response to macrophage activation and is a reliable marker of the risk of myocardial infarction in men. This study was designed to ascertain whether the treatment with atorvastatin, a powerful cholesterol lowering drug, and/or vitamin E, a natural antioxidant, may induce a short term decrease in serum C3 in subjects with persistently elevated levels. From an initial random sample of 1100 men aged 55-64 years, 140 subjects with 3 consecutive C3 measurements in the high tertile (>1.19 g/l) were selected. Those with total cholesterol <5.56 mmol/l were double blindly randomized in groups 1 (placebo, N = 28, G1) and 2 (vitamin E 600 IU/day, N= 30, G2). The subjects with total cholesterol values >5.56 mmol/l were randomized in groups 3 (placebo, N= 30, G3), 4 (atorvastatin 10 mg/day, N = 27, G4) and 5 (atorvastatin 10 mg/day + vitamin E 600 IU/day, N = 25, G5). After 3 months C3 levels were substantially unchanged in the first 4 groups, while in G5 a very significant decrement occurred: -0.070 g/l (5.2%); 95% CI 0.043-0.098; p <0.0001. "Normal" levels of C3 (< 1.19 g/l) were reached by 28% of G5 subjects. In G2 and G5 vitamin E levels increased by 60 and 36%, while in G4 they decreased by 23% (p < 0.0001), paralleling cholesterol and triglyceride fall. In all groups a progressive decrease in HDL cholesterol occurred (-17%, p < 0.0001). In conclusion, treatment with atorvastatin plus vitamin E for three months can lower persistently elevated C3 levels.

Mortality rates and number of deaths from cardiovascular diseases in Italy from 1982 to 1993. A comparison with total and tumour data.

In this study the recent Italian trends in cardiovascular deaths and mortality are described and compared with the trends regarding total and tumour deaths and mortality. The data, collected from the National Institute of Statistics, are presented as total (T), tumour (TU), cardiovascular (CV), cerebrovascular (CVD), ischemic heart disease (IHD) standardized mortality (sm), non-standardized mortality (nsm) and absolute number of deaths (d), according to sex, age, and geographical area. Data on sm were available only for the age group <75 years old. In males, from 1982 to 1993, T-sm fell by 18%, TU-sm by 4%, CV-sm by 30%, CVD-sm by 38% and IHD-sm by 24%. In females, the decrements were generally greater: T-sm 20%, TU-sm 4%, CV-sm 35%, CVD-sm 39% and IHD-sm 28%. Since 1985/87, tumours have been the leading cause of mortality, in both sexes. By 1991/93, the highest rates of CV, CVD, IHD-sm were reported mostly in the South of Italy. Non-standardized mortality rates for tumours increased, and for cardiovascular diseases decreased, in both sexes and age groups (<75 and >/=75 years old). As for sm, in the group <75 years, old tumours have been the leading cause of mortality since 1985/87, but in the older age group CV-nsm has been more than twice TU-nsm. By 1991/93 in comparison with 1982/84, CV deaths have fallen by 6% (-28% in the age group <75 years, +3% in the age group >/=75 years), while TU deaths have grown by 17% (+3% in the age group <75 years, +45% in the age group >/=75 years). Considering all age groups, by 1991/93 the absolute number of CV-d (239.241) was much greater than the number of TU-d (151.908); overall, almost 70% of CV-d and 40% of TU-d took place in the older age group. For the near future, the rapid aging of the Italian population (from 1982/84 to 1991/93 there was a 40% increment in the population older than 75 years) is a relevant variable to take into account. Thus, despite the 'reassuring' fall in CV-sm and nsm, cardiovascular diseases are expected to remain the major cause of death and physical disability in adults.

Endothelial dysfunction in hypertension.

Endothelial cells release both relaxing and contracting factors that modulate vascular smooth muscle tone and also participate in the pathophysiology of essential hypertension. Endothelium-dependent vasodilation is regulated primarily by nitric oxide but also by an unidentified endothelium-derived hyperpolarizing factor and by prostacyclin. Endothelium-derived contracting factors include endothelin-1, vasoconscrictor prostanoids, angiotensin II and superoxide anions. Under physiological conditions, there is a balanced release of relaxing and contracting factors. The balance can be altered in cardiovascular diseases such as hypertension, atherosclerosis, diabetes and other conditions, thereby contributing to further progression of vascular and end-organ damage. In particular, endothelial dysfunction leading to decreased bioavailability of nitric oxide impairs endothelium-dependent vasodilation in patients with essential hypertension and may also be a determinant for the premature development of atherosclerosis. Different mechanisms of reduced nitric oxide activity have been shown both in hypertensive states and several cardiovascular diseases, and endothelial dysfunction is likely to occur prior to vascular dysfunction. Thus, the strategies currently used to improve endothelial dysfunction may result in decreased morbidity and mortality in hypertensive patients.

Relationship of serum C3 to fasting insulin, risk factors and previous ischaemic events in middle-aged men.

AIMS: Serum C3 is a powerful indicator of the risk of myocardial infarction, which correlates with body mass index, serum lipids and blood pressure. This study was performed to ascertain whether such correlations may be explained by an association of C3 with fasting insulin, and to assess comparatively the relationships of C3 and traditional risk factors to previous myocardial infarction. METHODS AND RESULTS: The fasting levels of C3, insulin, and the main risk factors were evaluated in 1090 unselected men aged 55-64 years, including 129 cases of previous ischaemic events (51 myocardial infarctions). In multivariate analysis C3 was associated with insulin (r=0.27, P<0.0001), cholesterol (r=0.18, P<0.0001), body mass index (r=0.13, P<0.0001), glucose (r=0.12, P=0.0001), systolic blood pressure (r=0.10, P<0.001), triglycerides (r=0.09, P<0.01) and HDL-cholesterol (r=-0.06, P<0.05). These variables explained 31% of the total C3 variance. Alcohol consumption and physical activity correlated inversely with C3, while no correlation was found with smoking and family history of myocardial infarction. C3 was associated with previous myocardial infarction and stroke, but not with angina pectoris and peripheral arterial disease. In logistic regression the variables associated with previous myocardial infarction were C3 (P=0.011), family history of myocardial infarction (P=0.018), ex-smoker status (P=0.020), age (P=0.025), glucose (P=0.028) and HDL-cholesterol (P=0.051, inverse relationship). CONCLUSIONS: The association of C3 with myocardial infarction persists retrospectively, and is more significant than any other association of traditional risk factors with previous myocardial infarction. Of the many variables associated with C3, fasting insulin is its main covariate, which suggests that C3 is a marker of a pro-atherogenic metabolic imbalance partly coinciding with insulin resistance.

The shape of enzymatic curves during acute myocardial infarction: relationship to the progression of necrosis and implications for thrombolysis.

BACKGROUND: During acute myocardial infarction, the ascending branch of creatine kinase curves has a sigmoidal course whose inflection point marks the maximum rate of enzymatic increase in serum. This study was performed to assess the relationship between these morphologic characteristics of creatine kinase curves and the progression of myocardial necrosis. METHODS AND RESULTS: In isolated rat hearts exposed to different degrees of ischemia (coronary flow of 0.6 or 0.2 ml/g/min), the total quantity of creatine kinase released in the effluent had a sigmoidal course similar to the ascending branch of the curves from patients with acute myocardial infarction. Other rat hearts were frozen (which causes maximum damage to cell structures), thawed and then perfused. The resulting enzymatic curves had a downward concave ascending trend, similar to the portion beyond the inflection point of sigmoidal curves (the rate of creatine kinase release was maximum at the onset of perfusion and then decreased progressively). Finally, in some experiments ischemic rat hearts were further damaged by the perfusion, at different times, with highly concentrated catecholamines and without oxygen and substrates. This damaging perfusate was able to increase the rate of creatine kinase release (p = 0.0001) only when it was started before the inflection point of enzymatic curves. In 25 creatine kinase curves from patients with acute myocardial infarction (19 men and 6 women, age range 42 to 68 years), who were not treated with thrombolysis, the time of inflection varied from 1 to 12 hours from the onset of symptoms, with a maximum frequency between the 7th and the 8th hour. CONCLUSIONS: Based on these data, a biological model with 3 compartments has been suggested to explain the shape of creatine kinase curves, according to which the inflection point would occur after the completion of myocardial necrosis. The variability of the time of inflection might account for the cases of beneficial late thrombolysis reported in literature.