Clinical effectiveness of influenza vaccination in Manitoba.

OBJECTIVE: To assess the clinical effectiveness of influenza vaccination in preventing influenza-associated hospitalization and death. DESIGN: Case-control study. SETTING AND PATIENTS: Noninstitutionalized persons aged 45 years or older living in Manitoba, on December 1, 1982, and December 1, 1985. METHODS: Linked records of the Manitoba population registry, hospital-discharge abstracts, physician claims for ambulatory-patient visits and influenza vaccination, and vital statistics were used. A matched-set analysis estimated the clinical effectiveness of influenza vaccination in preventing hospital admissions and deaths from influenza-associated conditions during influenza A (H3N2) outbreak periods in 1982 to 1983 (12 weeks) and 1985 to 1986 (10 weeks). The analysis adjusted for hospital discharge and ambulatory care for high-risk conditions within the previous 15 months and 3 months, respectively. RESULTS: Influenza vaccination prevented 32% to 39% of hospital admissions with pneumonia and influenza and 15% to 34% of admissions with all respiratory conditions. Vaccination was 43% to 65% effective in preventing hospital deaths with these conditions (all listed diagnoses) and 27% to 30% effective in preventing deaths from all causes. CONCLUSION: Influenza vaccination has substantial clinical effectiveness in preventing hospital admission and death from influenza-associated conditions in noninstitutionalized individuals.

Healthy patients who perceive poor health and their use of primary care services.

OBJECTIVE: To determine what proportion of patients who have poor health perceptions are physically healthy and to explore why some patients perceive a healthy state while others perceive illness. DESIGN: A prospective consecutive series of office patients completed the Rand Corporation's General Health Perceptions Questionnaire, and their physicians rated their physical health. Their use of health care services was determined for the following 12 months. SETTING: A rural teaching office practice. PATIENTS: Of 243 adult patients asked to complete the questionnaire, 32 were excluded, for dementia (8), illiteracy (4), illness (8), incomplete questionnaires (6), and other reasons (6). 208 patients (86%) formed the final study group. MEASUREMENTS AND MAIN RESULTS: 62 of 208 patients had poor health perception scores. 39 of the 62 were rated by physicians as physically healthy and were not statistically different in physical health ratings or numbers of prescribed medications from the 146 patients who had higher health perception scores. However, these 39 patients had significantly more health-related worry, acute pain, and depression than did the other 146 patients. They also made more office visits and telephone calls and had higher total primary care charges. CONCLUSIONS: This study suggests that 21% of adult primary care patients (39 of 208) have health perceptions lower than expected for their levels of physical health. These low health perceptions are correlated with increased emotional distress and higher utilization of health care resources. Strategies to identify these patients and interventions to improve their views of their health could reduce utilization.

Glipizide treatment of pancreas autotransplantation: effects on alterations in glucose-insulin relationships.

Pancreas transplantation has been proven effective in supplying an endogenous insulin supply in diabetics. However, alterations in glucose metabolism after transplantation suggest a possible "insensitivity" to its action in the periphery. We hypothesized that sulfonylurea treatment of canines who had received segmental pancreas autotransplants would correct these alterations by altering peripheral insulin sensitivity. Glipizide therapy (5 mg p.o. b.i.d.) did appear, in fact, to enhance basal insulin sensitivity by lowering fasting glucose (100 +/- 3 to 81 +/- 11 mg/dl pre-treatment to post-treatment) while not affecting basal insulin levels. However, glipizide therapy was associated with decreased insulin response to challenge by either oral glucose (2 gm/kg) or sustained intravenous hyperglycemia (150 mg/dl above basal). We conclude that our model of pancreas autotransplantation documents alterations in glucose metabolism which are devoid of the effect of immunosuppression. Glipizide treatment appears to affect fasting sensitivity to insulin, but results in a decrement of insulin response to oral or intravenous glucose challenge.

Hospital-based pneumococcal immunization. Epidemiologic rationale from the Shenandoah study.

To explore the potential usefulness of a strategy of hospital-based pneumococcal immunization, we studied a population-based linked record of hospital discharges for Medicare enrollees living in the Shenandoah region of Virginia. A retrospective study of 1633 persons discharged with pneumonia in 1983 showed that 61% to 62% had been discharged within the previous 4 years. Among these patients, 87% had had one or more high-risk conditions recognized during previous hospital admissions. A cohort study demonstrated that discharged patients had a 6% to 9% probability of readmission with pneumonia within 5 years. Each such readmission could be prevented by immunizing few (approximately 100) discharged patients with pneumococcal vaccine. Furthermore, the costs of vaccination would be approximately one-third the costs of hospital care for unvaccinated discharged patients readmitted with pneumonia. These results provide an epidemiologic rationale for current recommendations that elderly patients discharged from hospitals should be immunized with pneumococcal vaccine.

A retrospective cohort study of nosocomial diarrhea as a risk factor for nosocomial infection.

Because most nosocomial infections are caused by enteric flora, it was undertaken to determine whether diarrhea predisposes to nosocomial infections by conducting a retrospective cohort study involving patients with and without nosocomial diarrhea who were hospitalized during a 12-month period. The rate of urinary tract infection per person-day after onset of diarrhea was significantly greater in patients with diarrhea (relative risk = 10.3; 95% confidence interval = 1.7-63.1, P = .006). When indwelling bladder catheter and duration of catheter placement were considered, diarrhea remained an independent risk factor for urinary tract infection in logistic regression analysis (P = .009). These data suggest that diarrhea and consequent urethral meatal contamination may be a cause of nosocomial urinary tract infection, especially in patients with a catheter, and that removal of the catheter should be considered whenever possible in patients with diarrhea.

Prediction of microbial aetiology at admission to hospital for pneumonia from the presenting clinical features. British Thoracic Society Pneumonia Research Subcommittee.

The selection of initial antimicrobial treatment in a patient with community acquired pneumonia is an important clinical decision. Because this decision is usually made before the results of specific microbiological tests are available, we attempted to determine how well the presenting clinical features would allow prediction of microbial aetiology in 441 adults admitted to hospital with pneumonia. Five of 90 variable available on admission were selected for inclusion in a multivariate discriminant function analysis because of their strong association with one or more of the major aetiological subsets (Mycoplasma pneumoniae, Streptococcus pneumoniae, "other," and undetermined). These variables were age, number of days ill before admission, presence or absence of bloody sputum and of lobar infiltration on chest radiograph, and white blood cell count. The microbial aetiology was correctly predicted by this discriminant function analysis in only 42% of cases, which gives a quantitative estimate of the degree of difficulty encountered in determining the microbial aetiology at the time of admission for pneumonia. When a similar discriminant function analysis was applied to the third of patients in whom the microbial aetiology was never determined, most of these cases were predicted to be due to Streptococcus pneumoniae.

Hospital stay and mortality attributed to nosocomial enterococcal bacteremia: a controlled study.

A retrospective cohort study of 97 patients identified by prospective hospital-wide surveillance was conducted to determine the length of hospital stay and mortality attributed to hospital-acquired enterococcal bacteremia. The mean duration of hospitalization for cases was 83 days compared with 44 days for matched controls (p = 0.0001). The mortality rate during the study period was 43% among cases and 12% in matched controls (p less than 0.001). Thus the mortality rate attributable to enterococcal bacteremia was 31% and the risk ratio was 4.75. Stepwise discriminant function analysis indicated that the use of vascular catheters and renal dialysis and the presence of immune deficiency were predictors of fatal outcome in cases. Enterococcal bacteremia has become a prominent nosocomial pathogen and is associated with mortality rates well above those expected from the underlying disease.

Improving preventive care at a medical clinic: how can the patient help?

We developed a comprehensive individualized preventive care reminder system and then tested the hypothesis that directly involving patients in the reminder process would lead to greater use of preventive services than involving physicians only. There were three experimental groups of 350 patients each: in group 1 physicians and patients received the reminder; in group 2 physicians only received the reminder; in group 3 neither physicians nor patients received the reminder. Nine preventive care services were studied: blood pressure measurement; dental exam; ocular pressure measurement; stool exam for occult blood; influenza, pneumococcal, and tetanus vaccinations; mammography; and Papanicolaou smears. Need for these services was determined by telephone interview and chart review. To determine whether services were obtained, charts were reviewed after four to eight months of follow-up. For overall compliance with preventive recommendations and for several individual services (stool exam for occult blood, tetanus vaccination, mammography), group 1 patients received significantly more preventive care than group 2. Likewise, group 2 patients received more preventive care than group 3. These data show that involving patients in reminder efforts is an effective means of raising the level of preventive services.

Progenitor cell numbers (CFU-GM, CFU-D, and CFU-Mix) and hemopoietic recovery following autologous marrow transplantation.

Forty-one consecutive patients were treated with high-dose chemotherapy with or without total body irradiation followed by autologous marrow transplantation. Four treatment regimens of varying intensity were used. Every patient's harvested marrow was evaluable for nucleated cell and progenitor cell loss during the cell separation and cryopreservation process. Of the 41 patients, 38 were evaluable for peripheral blood count recovery. Multivariate analysis of colony-forming cell assays and recovery of neutrophils and platelets showed a significant association with absolute numbers of post-thaw mixed colony-forming units (CFU-Mix) infused (p less than 0.002). Prefreeze CFU-Mix also correlated with recovery to a lesser degree, as did absolute numbers of nucleated cells. The number of diffusion chamber colony-forming units (CFU-D) prefreeze, but not post-thaw infused into the patient, was associated with recovery of neutrophils (p = 0.0001), but not platelets. When the precursor cell numbers were adjusted for body weight, post-thaw CFU-Mix showed the best correlation with recovery of both platelets and neutrophils. Prefreeze CFU-D per kg was also associated with recovery of neutrophils (p = 0.02). To some extent nucleated cells per kg predicted for recovery with neutrophils and platelets (p less than 0.05). When analyzed according to treatment regimen, cyclophosphamide-BCNU-VP16 (CBV) or cyclophosphamide-total body irradiation (CY/TBI) was associated with prolonged recovery compared to cyclophosphamide-adriamycin-vinblastine (CAV) or etoposide-cyclophosphamide (EC). In this setting only CFU-D number predicted neutrophil recovery (p less than 0.002). We conclude that determination of the number of total nucleated cells, CFU-D, and CFU-Mix, before cryopreservation of the sample is important in predicting hemopoietic reconstitution in autologous bone marrow transplantation.

Insulin-dependent and insulin-independent effects after surgical alterations of the pancreas.

Anatomic alterations of the pancreas result in physiologic alterations that have not been completely analyzed. Insulin plays a major role in carbohydrate metabolism; nevertheless, as much as 50% of a hyperglycemic load may be metabolized independent of insulin. We analyzed the effects of surgical alterations of the pancreas on postoperative glucose metabolism, including insulin-independent effects. Mongrel female dogs underwent one of three procedures: proximal partial pancreatectomy (PPx), PPx plus diversion of pancreatic venous effluent to the systemic circulation (SC), or PPx plus segmental pancreatic autotransplantation (PAT). Intravenous glucose tolerance tests, with or without a background infusion of somatostatin (SST; 400 ng/kg/min) were performed on all animals preoperatively and postoperatively. SST completely suppressed secretion of assayable peripheral insulin. The rate of glucose disposal during SST suppression approximates the rate of insulin-independent glucose disposal (IIGD). Although there was a significant decrease in the rate of glucose disposal during SST infusion when compared with the rate without SST, no differences in IIGD were found between postoperative groups. IIGD was calculated at 50% to 55% for control, PPx, and SC groups and at 67% for PAT. Peripheral sensitivity to an exogenous insulin infusion (euglycemic clamp) was unchanged by any of the procedures. We conclude that surgical alteration of the pancreas, including pancreas transplantation, results in altered glucose handling in the face of "normal" peripheral levels of insulin. Changes in IIGD and analysis of peripheral sensitivity to insulin do not explain these alterations completely.

Plasma renin activity and albumin excretion in teenage type I diabetic subjects. A prospective study.

Plasma renin activity (PRA) may be high among teenage and young adult insulin-dependent diabetic subjects. Supine PRA and stimulated PRA were therefore measured in 50 female and 50 male diabetic subjects, 13-20 years old, diagnosed before the age of 16. Fifty percent have been restudied after 4.6 +/- 0.2 (mean +/- SEM) years. Initially, 43% had high PRA (supine 4.0 +/- 0.37, stimulated 12.02 +/- 0.8 ng/ml/hr angiotensin I), 45% had normal activity (supine 2.89 +/- 0.26, stimulated 6.47 +/- 0.34 ng/ml/hr/angiotensin I), and 12% had low activity (supine 1.57 +/- 0.05, stimulated 3.09 +/- 0.08 ng/ml/hr/angiotensin I). Levels were directly associated with prepubertal duration of diabetes and were inversely associated with duration of diabetes after onset of puberty but not with total duration or patient age. Within 4.6 +/- 0.2 years the percentage of subjects with high PRA fell to 13%, and the percentage of those with low PRA rose to 35%. Initially 51% of the cohort had normal albumin excretion rates (AER) at rest and during exercise equal to or less than 10 micrograms/min/m2; 32% had elevated rates only during exercise of 39 +/- 5 micrograms/min/m2; 13% had elevated rates at rest of 41 +/- 8 micrograms/min/m2 and during exercise of 116 +/- 21 micrograms/min/m2; and 4% had clinical proteinuria at rest and during each exercise period equal to or greater than 150 micrograms/min/m2. After 5 years, 58% continued to have normal AER, or their AER improved.(ABSTRACT TRUNCATED AT 250 WORDS)

Hospital-acquired pneumonia. Attributable mortality and morbidity.

A total of 1,001 consecutive episodes of nosocomial pneumonia in 901 patients was identified by routine surveillance at the University of Virginia Medical Center between 1979 and 1983 (8.6 episodes/1,000 admissions). When only initial episodes were examined, 890 patients comprised the study sample. The overall case fatality rate was 30%. Stepwise logistic regression indicated that time from admission to pneumonia (p = 0.0006), age (p less than 0.0001), prior use of mechanical ventilation (p = 0.0032), and neoplastic disease (p = 0.0062) were associated with mortality. Multiple regression analysis indicated that the factors associated with increased length of hospitalization included posttracheostomy status (p = 0.0001), prior mechanical ventilation (p = 0.0001), immunosuppressive or leukopenic status (p = 0.0009), nasogastric intubation (p = 0.0003), and prior bacteremia (p = 0.0127). A sampled, individually matched cohort study (n = 74 pairs) was conducted to determine the proportion of mortality in cases that was attributable to infections (33%) and to determine excess hospital stay (seven days) among the patients with nosocomial pneumonia. Excess stay was statistically significant (p less than 0.0001), but proportional mortality was only marginally significant (p = 0.0892). Our findings suggest that nosocomial pneumonia accounts for approximately 33% of the crude mortality and contributes significantly to the economic burden associated with prolonged hospitalization.

Health perceptions of primary care patients and the influence on health care utilization.

This prospective study was conducted to determine the influence of primary care patients' health perceptions on their utilization of health care services. Patients' health perceptions were measured using the RAND Corporation's General Health Perceptions Questionnaire. Physicians provided scores of how they thought the patients perceived their health and of actual physical and emotional health. Utilization data (number of office visits, number of telephone calls to the physician, and ambulatory charges) were evaluated for a 12-month period after completion of the questionnaire. Of 208 patients, 62 (30%) patients with health perceptions scores less than 50 had greater degrees of anxiety (P less than .001), depression (P less than .001), health-related worry (P less than .001), and felt less able to resist illness (P less than .001) than patients with higher health perception scores. Analysis of covariance was used to control for differences in physical health among groups of patients with varying health perceptions. These analyses revealed that patients with low health perceptions made more office visits (P = .002), more telephone calls to the physician (P = .01), and had more office charges (P = .05) than patients with higher scores. Physicians accurately predicted the patients' health perceptions in 49% of the cases. In 37%, they thought patients would score their health perceptions higher than they did; in 14% they thought patients would score their health perceptions lower. Health perceptions are an important factor contributing to the use of health care by primary care patients, regardless of the patient's actual physical health. Persons with low health perceptions account for approximately 5% of office visits, a clinically important fraction, especially when compared to the 9% of office visits for hypertension, the most common disease treated in the medical office.

Lack of in vivo somatotroph desensitization or depletion after 14 days of continuous growth hormone (GH)-releasing hormone administration in normal men and a GH-deficient boy.

In vitro and in vivo studies of somatotroph responsivity to GHRH stimulation indicate that partial loss of GH responsiveness occurs during constant GHRH stimulation. To determine if these observations reflect either a short term effect of GHRH or if the absence of somatostatin effects somatotroph desensitization (as occurred in in vitro studies), we administered GHRH-40 (10 ng/kg.min) by continuous iv infusion for 14 days to five normal men and one GH-deficient boy. Serum insulin-like growth factor I (IGF-I) concentrations were measured at frequent intervals to assess the biological effect of GHRH on GH secretion. The GH secretory profiles were assessed by measuring serum GH levels every 20 min for 24 h before (day 0), on the 14th GHRH infusion day, and 14 days after discontinuation of the GHRH infusion in the normal men. The GH-deficient boy was studied before and during the 14th GHRH infusion day. A supramaximal iv GHRH dose was administered at the end of the 24-h sampling period, and the GH responses were compared. Serum IGF-I concentrations increased on the 14th day of GHRH infusion in the normal men [day 0 mean, 0.84 +/- 0.14 (+/- SE) X 10(3); day 14, 1.74 +/- 0.20 X 10(3) U/L; P less than 0.05] and from 0.20 X 10(3) on day 0 to a maximum of 0.67 X 10(3) U/L on day 3 in the GH-deficient boy; they declined to pretreatment levels after discontinuation of GDRH. The mean integrated serum GH concentrations in the normal men were 1.44 +/- 0.10 micrograms/L.h on day 0 and 3.11 +/- 0.95 on day 14 of GHRH infusion. The integrated GH concentration in the GH-deficient boy was 1.53 micrograms/L.h on day 0 and 4.23 on day 14 of GHRH infusion. Pulsatile GH secretion, assessed by cluster analysis, was preserved in the normal men and occurred de novo in the GH-deficient boy on the 14th GHRH infusion day. The GH response to bolus GHRH administration was also preserved; no attenuation of the response occurred in the normal men or the GH-deficient boy after 14 days of GHRH infusion. The increase in IGF-I concentrations during 14 days of continuous GHRH administration, the persistence of pulsatile GH release in normal men, the de novo appearance of GH pulses in the GH-deficient boy, and the preservation of the response to a supramaximal GHRH dose indicates that the somatotrophs remain responsive to prolonged constant stimulation.(ABSTRACT TRUNCATED AT 400 WORDS)

Effect of glipizide on the surgically altered pancreas.

Surgical alterations of the pancreas affect peripheral glucose, insulin, and glucagon levels with accompanying changes in carbohydrate metabolism. The sulfonylurea glipizide has been used to treat insulin-deficient states; however, its mechanism is not completely known. We hypothesized that glipizide would correct postoperative changes in glucose handling in a way that would allow more complete understanding of the drug's action. Two surgical groups (Group 1:80 percent proximal pancreatectomy; Group 2: proximal pancreatectomy plus splenocaval diversion) were compared with a healthy control group (Group 3). We have concluded that glipizide may have affected basal insulin sensitivity in the control group and Group 2 animals without affecting insulin secretion in response to oral or intravenous glucose stimulation. Glipizide does not correct the alterations in glucose handling or insulin secretion after reduction in beta-cell mass.

A dynamic model of angiotensin II infusion experiments.

Applications of control theory in studies of biological system dynamics have come to be called compartmental modelling. A second order, nonlinear, compartmental model is developed which describes the dynamics of the hormone angiotensin II (AII) and arterial blood pressure (BP) during AII infusion experiments. The model is partially identified using dose response data for constant infusion rates between 0.01 and 0.10 microgram/kg/min over a period of several minutes. This study represents a first step in understanding the dynamics of regulation of arterial blood pressure by the renin-angiotensin system. AII is a vasoconstrictor and is known to participate in the natural regulation of BP. AII is also believed to be an agent in the development of hypertension and atherosclerosis. The model is used to identify causal mechanisms which are consistent both with the established correlation between plasma AII concentration and arterial BP and with current physiological knowledge. The study demonstrates how a simple state variable model can be used to provide guidance concerning the design of future infusion experiments.

Random and Non-Random Coincidence Between Luteinizing Hormone Peaks and Follicle-Stimulating Hormone, Alpha Subunit, Prolactin and Gonadotropin-Releasing Hormone ulsations.

Abstract We have examined the co-pulsatility of luteinizing hormone (LH) and prolactin, LH and follicle-stimulating hormone (FSH), and LH and alpha subunit in normal men. We tested whether the degree of physiologically observed co-pulsatility (peak coincidence) significantly exceeded expected random concordance between independently pulsating hormone series. To this end, computer simulations were used to create synthetic endocrine time series pulsating randomly and independently at known frequencies. Resultant predictions of the mean, variance and probability distribution of the number of randomly coincident peaks permitted us to test the null hypothesis that physiologically observed hormone co-pulsatility was due to chance peak associations alone. Physiological observations were made in 33 normal men and in six ovariectomized ewes subjected to combined hypothalamo-pituitary and jugular venous catheterization. The following salient results were obtained: 1) random peak coincidence rates between independently pulsating hormone series were substantial at high pulse frequencies, but such random rates were significantly exceeded in the case of gonadotropin-releasing hormone and LH peaks (P< 0.0001); 2) random coincidence was further increased when coincidence was defined as peak maxima occurring not only simultaneously but also within some defined time window (e.g. +/-10 min, as commonly done in the literature); 3) significant co-pulsatility could be demonstrated for simultaneous LH and FSH pulsations in normal men (P< 0.0001); 4) coincidence rates for 10-min lagged (but not for simultaneous) LH and prolactin pulses were significantly more likely than chance associations; 5) observed coincidence between LH and a subunit pulses significantly exceeded expected (random) peak overlap (P<0.001); and 6) in contrast, hormone peaks in different men were only randomly associated. We conclude that based upon the means, variances and probability distributions calculated here, available reports on peak coincidence between pulsatile neuroendocrine time series must be re-examined in the light of high rates of random coincidence observed between independently pulsating hormone series.

Screening and characterization of antiepileptic drugs with rapidly recurring hippocampal seizures in rats.

A method to efficiently screen antiepileptic drugs (AED) for their actions against complex partial and secondarily generalized seizures is presented. The procedure relies on rapidly recurring hippocampal seizures (RRHS) in rats which are first used to bring epileptic responses to a stable, fully kindled state and then to test 3 parameters--behavioral seizures, electrographic seizures, and afterdischarge thresholds--before and after drug administration. With the methods described, the effects of a given drug treatment can be thoroughly determined in a single study period. Quantitative determinations of dose-response, time-action and relative potency characteristics are readily ascertained. A battery of known AED, encompassing those in common clinical use, was studied with this system. Kindled motor seizures (classes 4 and 5) were more readily suppressed than limbic behavioral seizures (classes 1-3). Electrographic seizures were usually, but not always, shortened concurrently with suppression of behavioral seizures. Under the conditions of this study, afterdischarge thresholds were not elevated, indicating that a critical role of AED is to counteract seizure spread and prolongation. The overall behavior of the RRHS test system with AED was identical to that with traditional amygdala kindled seizures and results were in good agreement with the clinical responsiveness of the kinds of seizures that these experimental systems model. The features of RRHS make it a useful system for screening new agents for antiepileptic effects, even in circumstances where little or no information about the drug under study is available.

A comprehensive analysis of myocardial infarction due to left circumflex artery occlusion: comparison with infarction due to right coronary artery and left anterior descending artery occlusion.

Forty consecutive patients with creatine kinase-MB confirmed myocardial infarction due to circumflex artery occlusion (Group 1) were prospectively evaluated and compared with 107 patients with infarction due to right coronary artery occlusion (Group 2) and 94 with left anterior descending artery occlusion (Group 3). All 241 patients underwent exercise thallium-201 scintigraphy, radionuclide ventriculography, 24 h Holter electrocardiographic (ECG) monitoring and coronary arteriography before hospital discharge and were followed up for 39 +/- 18 months. There were no significant differences among the three infarct groups in age, gender, number of risk factors, prevalence and type of prior infarction, Norris index, Killip class and frequency of in-hospital complications. Acute ST segment elevation was present in only 48% of patients in Group 1 versus 71 and 72% in Groups 2 and 3, respectively (p = 0.012), and 38% of patients with a circumflex artery-related infarct had no significant ST changes (that is, elevation or depression) on admission (versus 21 and 20% for patients in Groups 2 and 3, respectively) (p = 0.001). Abnormal R waves in lead V1 were more common in Group 1 than in Group 2 (p less than 0.003) as was ST elevation in leads I, aVL and V4 to V6 (p less than or equal to 0.048). These differences in ECG findings between Group 1 and 2 patients correlated with a significantly higher prevalence of posterior and lateral wall asynergy in the group with a circumflex artery-related infarct. Infarct size based on peak creatine kinase levels and multiple radionuclide variables was intermediate in Group 1 compared with that in Group 2 (smallest) and Group 3 (largest). During long-term follow-up, the probability of recurrent cardiac events was similar in the three infarct groups. When patients with a circumflex artery-related infarct were stratified according to the presence or absence of abnormal R waves in lead V1 or V2, the abnormal R wave group had more admission ST elevation (p = 0.025), a larger infarct (p less than 0.05) and more extensive coronary artery disease (p = 0.027). In fact, all patients with a circumflex artery-related infarct and an abnormal R wave in lead V1 had multivessel disease. An abnormal R wave in lead V1 had a 96% specificity for circumflex versus right coronary artery-related infarction but a sensitivity of only 21%. Discriminate function analysis of all admission historical and ECG variables identified inferior and lateral ST elevation as independent predictors of circumflex artery-related infarction...