Prevention of ischemic events in patients with peripheral arterial disease design, baseline characteristics and 2-year results an observational study.

AIM: Peripheral arterial disease (PAD) is associated with frequent cardiovascular ischemic events. We followed the survival of PAD patients and tested whether PAD remains an adverse prognostic indicator in spite of treatment according to the current European guidelines on cardiovascular disease prevention. METHODS: Eight hundred eleven patients with PAD and 778 control subjects, aged 65 (SD 9) years at inclusion, with a male/female ratio of 3/2 were treated according to the European guidelines on cardiovascular disease prevention and evaluated yearly for occurrence of death, non-fatal acute coronary syndrome, stroke or critical limb ischemia (major events) and revascularization procedures (minor events). At baseline, classical risk factors were significantly more prevalent in the PAD group and protective cardiovascular medication was prescribed to patients with PAD more frequently than to control subjects. RESULTS: In the PAD group, the 2-year Kaplan-Meier survival estimate was 96.7% (CI 95.4-97.9) vs. 98.2% (CI 97.2-99.1) in the control group, P=0.059. The groups differed in the 2-year major event-free survival: 93.5% (CI 92.7-95.3) in PAD vs. 97.1% (CI 95.9-98.4) in controls, P<0.017, as well as in event-free survival: 79.9% (CI 77.1-82.9) in PAD vs.96.4% (CI 95.0-97.9) in controls, P<0.001. CONCLUSION: Patients with PAD had a borderline higher risk of all-cause death and a significantly higher risk of major and minor non-fatal cardiovascular events compared to control subjects. However, treatment according to the European guidelines on cardiovascular disease prevention resulted in encouragingly low absolute mortality and morbidity. (ClinicalTrials.gov number NCT00761969.).

Accuracy of point predictions in survival analysis.

Survival time prediction is important in many applications, particularly for patients diagnosed with terminal diseases. A measure of prediction error taken from the medical literature is advocated as a practicable method of quantifying reliability of point predictions. Optimum predictions are derived for familiar survival models and the accuracy of these predictions is investigated. We argue that poor predictive capability is inherent to standard survival models with realistic parameter values. A lung cancer example is used to illustrate difficulties in prediction in practice.

Supporting discovery in medicine by association rule mining in Medline and UMLS.

The paper presents an interactive discovery support system for the field of medicine. The intended users of the system are medical researchers. The goal of the system is: for a given starting concept of interest, discover new, potentially meaningful relations with other concepts that have not been published in the medical literature before. The known relations between the medical concepts come from the Medline bibliographic database and the UMLS. We use association rules for discovering the relationship between medical concepts. We evaluated the system by testing how successfully it predicted future discoveries (new relations between concepts). We first divided the Medline database into two segments (older and newer) using the publication date. Then we calculated how many of the new relations found by the system in the older segment become known relations in the newer segment. We found out with statistical significance that the system predicts new relations better then someone predicting randomly. The evaluation showed that our approach for supporting discovery in medicine is successful, but also that some improvements are needed, especially on limiting the number of potential discoveries the system generates.

Seasonal variation of some metabolic and haemostatic risk factors in subjects with and without coronary artery disease.

Acute myocardial infarction (AMI) is more frequent in winter months than in summer months. The aetiologic mechanisms underlying this seasonal pattern are poorly understood. We investigate whether seasonal variation of metabolic and haemostatic coronary risk factors exists, and whether this variation is more pronounced in subjects with coronary artery disease (CAD). In 82 subjects (47 free of clinical signs of CAD and in 35 survivors of AMI), measurements of body mass index (BMI), lipoproteins, glucose, insulin, plasminogen activator inhibitor-1, tissue-type plasminogen activator (t-PA), euglobulin clot lysis time, fibrinogen, and platelet count were performed twice in the cold months (December and March) and twice in the warm months (June and September). Significantly higher BMI (26.8 versus 26.2 kg/m2, P < 0.01), glucose (5.5 versus 5.1 mmol/l, P < 0.01), total cholesterol (5.61 versus 5.32 mmol/l, P < 0.05), low-density lipoprotein cholesterol (3.63 versus 3.34 mmol/l, P < 0.05), triglycerides (1.79 versus 1.61 mmol/l, P < 0.01), Lp(a) (270.7 versus 237.5 mg/l, P < 0.01), fibrinogen level (3.50 versus 2.95 g/l, P < 0.00001), platelet count (212 x 10(9) versus 173 x 10(9)/l, P < 0.01) and significantly lower high-density lipoprotein cholesterol level (1.22 versus 1.28 mmol/l, P < 0.05) were observed in the cold months compared with the warm months. Significant seasonal variation of t-PA activity (1.19 versus 0.87 IU/ml, P = 0.015) and t-PA antigen (8.5 versus 7.3 ng/ml, P = 0.05) was demonstrated only in subjects with CAD. Clustering of peak values of several metabolic and haemostatic coronary risk factors was observed in winter months. This variation might be of aetiopathogenetic importance for the seasonal pattern of acute myocardial infarction.

The assessment of contact stress in the hip joint after operative treatment for severe slipped capital femoral epiphysis.

We determined contact stress on the articular surface of the hip joint in a group of patients who underwent operative treatment for severe slipped capital femoral epiphysis. Two different procedures were considered: the modified osteotomy of Dunn-Fish and the osteotomy of Imhäuser. In order to determine the stress distribution, a three-dimensional mathematical model was used taking into account the geometrical parameters of the pelvis and hip, which were measured from standard antero-posterior radiographs. We found that the Dunn-Fish procedure produced lower peak stress than the Imhäuser procedure.

BJ: an S-Plus program to fit linear regression models to censored data using the Buckley-James method.

Most researchers are familiar with ordinary multiple regression models, most commonly fitted using the method of least squares. The method of Buckley and James (J. Buckley, I. James, Linear regression with censored data, Biometrika 66 (1979) 429-436.) is an extension of least squares for fitting multiple regression models when the response variable is right-censored as in the analysis of survival time data. The Buckley-James method has been shown to have good statistical properties under usual regularity conditions (T.L. Lai, Z. Ying, Large sample theory of a modified Buckley-James estimator for regression analysis with censored data, Ann. Stat. 19 (1991) 1370-1402.). Nevertheless, even after 20 years of its existence, it is almost never used in practice. We believe that this is mainly due to lack of software and we describe here an S-Plus program that through its inclusion in a public domain function library fully exploits the power of the S-Plus programming environment. This environment provides multiple facilities for model specification, diagnostics, statistical inference, and graphical depiction of the model fit.

Mild hyperhomocysteinemia and fibrinolytic factors in patients with history of venous thromboembolism.

Mild hyperhomocysteinemia is recognized as a risk factor for venous thromboembolism (VTE), though its role in the thrombogenic processes is not understood. Its possible association with impaired fibrinolysis was investigated in 157 patients (61 women, 96 men) below the age of 60 years (43+/-11, mean+/-SD) with a history of objectively confirmed VTE. Patients had significantly higher fasting total plasma homocysteine (tHcy) levels than 138 apparently healthy subjects (8.0, 6.6-9.9 micromol/L vs. 7.2, 5.9-8.6 micromol/L, P=0. 001; median, range between first and third quartile). In 17 of 157 patients (12%) hyperhomocysteinemia (tHcy>11.4 micromol/L for women and tHcy>12.6 micromol/L for men) was established. The adjusted odds ratio as an estimate of relative risk for VTE was 2.3 (0.8-7.0; 95% confidence interval). When patients with hyperhomocysteinemia were compared to patients without hyperhomocysteinemia, no significant differences in t-PA (antigen 9.2+/-5.5 microg/L and 9.7+/-4.7 microg/L, respectively; activity 1.3+/-0.5 IU/mL and 1.3+/-0.7 IU/mL, respectively) and PAI-1 (antigen 19.3+/-17.5 microg/L and 22.6+/-20. 4 microg/L, respectively; activity 15.0+/-12.6 and 15.8+/-13.3 IU/mL, respectively) were observed. In conclusion, this study showed an association between mild hyperhomocysteinemia and VTE, but provided no evidence for an independent association between hyperhomocysteinemia and alterations in fibrinolytic proteins.

The 4G/5G sequence polymorphism in the promoter of plasminogen activator inhibitor-1 (PAI-1) gene: relationship to plasma PAI-1 level in venous thromboembolism.

Impaired fibrinolysis due to increased plasminogen activator inhibitor-1 (PAI-1) is observed in up to 40% of patients with venous thromboembolism and might be causally related to the disease. There is evidence that genetic variations in the promoter of the PAI-1 gene and metabolic factors contribute to increased plasma PAI-1 levels. A single nucleotide insertion/deletion (4G/5G) polymorphism in the promoter region of the PAI-1 gene and metabolic factors were studied in 158 unrelated patients below the age of 61 years (43 +/- 11 years, mean +/- standard deviation) with history of objectively confirmed venous thromboembolism and in 145 apparently healthy controls. Patients had on average two times higher PAI activity (11.9 vs. 6.1 IU/ml) and by 40% higher PAI-1 antigen (14.8 vs. 10.7 ng/ml) than healthy controls, and also higher body mass index, lipid levels, fasting glucose and insulin. Patients differed significantly from healthy controls neither in the frequency of the 4G and 5G alleles (0.57/0.43 in patients and 0.52/0.48 in controls) nor in the distribution of the 4G/5G genotypes. Possession of the 4G/4G or the 4G/5G genotype did not increase relative risk for venous thromboembolic disease and the distribution of the 4G/5G genotypes was neither associated with recurrent nor with spontaneous disease. In patients association between the 4G/5G genotypes and PAI activity (adjusted for body mass index, triglyceride and glucose level) was observed, with the highest PAI activity values in the 4G/4G genotype (14.6 IU/ml), intermediate in the 4G/5G genotype (13.3 IU/ml) and the lowest in the 5G/5G genotype (5.2 IU/ml, all values means). Association between PAI activity and triglyceride level was the strongest in the 4G/4G genotype (correlation coefficient r = 0.47, p < 0.01) and the weakest in the 5G/5G genotype (r = -0.04, not significant). In conclusion, the present case-control study shows an association between the 4G/5G polymorphism in the promoter of the PAI-1 gene and plasma PAI-1 levels in patients with venous thromboembolism. Similar distribution of the 4G/5G genotypes in patients and healthy controls suggests that this genetic variation by itself is not a major risk factor for venous thromboembolism.

Immunological differentiation between tickborne encephalitis with and without concomitant neuroborreliosis.

Cerebrospinal fluid lymphocyte subsets in patients with tickborne encephalitis (TBE) and in patients with TBE with concomitant neuroborreliosis (double infection) were analysed by flow cytometry. In the TBE group, higher percentages of CD4+DR+ T cells (p = 0.02) and CD25+ T cells (p = 0.0002) were observed, while in the group with double infection, higher percentages of CD19+ cells (p = 0.007), CD8+DR- T cells (p = 0.04), and CD3+CD71 + T cells (p = 0.0002) were found. It was concluded that several differences in immune cell parameters are present between the two groups of patients. Three variables (CD19+ cells, CD3+CD25+ T cells, CD3+CD71+ T cells) were included in the logistic regression model for calculation of probability for double infection. Flow cytometric characterisation of lymphocyte subsets in CSF can further substantiate the diagnosis of concomitant neuroborreliosis in patients with TBE.

Explained variation in survival analysis.

Several measures of explained variation have been suggested for the Cox proportional hazards regression model. We have categorized these measures into three classes which correspond to three different definitions of multiple R2 of the general linear model. In an empirical study we compared the performance of these measures and classified them by their adherence to a set of criteria which we think should be met by a measure of explained variation for survival data. We suggest that currently there is no uniformly superior measure, particularly as the concepts of either uncensored or censored populations may lead to different choices. For uncensored populations, a measure by Kent and O'Quigley and the squared rank correlation between survival time and the predictor from a Cox regression model appear recommendable choices. For the latter, censored survival times are terminated using a very recent data augmentation algorithm for multiple imputation under proportional hazards. With censored populations, Schemper's measure, V2, could be considered. We give an introductory example, discuss aspects of application and stress the desirability of routinely evaluating explained variation in studies of survival.

Late sequelae in children treated for brain tumors and leukemia.

Forty-two survivors treated at an age of 2-16 years for brain tumors or leukemia were, 4-21 years after treatment, subjected to an extensive follow-up investigation, including physical examination and interview; 35 of them also had endocrinological and 33 psychological evaluation. Hormonal deficiencies were found in about two-thirds of patients and were most common in those treated for brain tumors. The great majority had verbal intelligence quotient (VIQ) within normal range. Also, the performance intelligence quotients (PIQ) were normal in most patients. However, the results suggested that the primary intellectual capacity in children treated for cancer was not being fully utilized, their PIQ being on the average higher than their VIQ; this tendency was especially pronounced in the leukemia patients.

A case-control study of breast cancer in relation to oral contraceptive use in Slovenia.

With the aim to investigate a possible association between oral contraceptive (OC) use and breast cancer occurrence, 534 women aged 24--54 years with newly diagnosed breast cancer and 1989 individually matched hospital controls were interviewed during 1980--1983. The overall risk for ever-users vs. never-users estimated by logistic regression and adjusted for several possible confounding factors was 1.62 (p less than 0.05). The analysis of potential biases indicated that this risk may be overestimated, especially because the controls might not be fully representative of the basic population. The risk was increasing with total duration of OC use, reaching the highest value by more than 7 years of use. As to the latency, the risk was the highest for women starting pill use 4--8 years before diagnosis, thus suggesting that OCs might act as promoters rather than initiators of tumor growth. There was no substantial difference in risk between women starting pill use before 25 years of age and those starting it later. The number of users before first term pregnancy was too small to warrant relative risk estimation. Interaction (significant) was found between OC use and family history of breast cancer; there was no such evidence in other subgroups of women being at baseline breast cancer risk. There were no significant differences in the distribution of cases and controls classified by individual OC formulations used. The increased relative risk for users was concentrated in early stages of breast cancer, most likely owing to detection bias. Considering the indicated biases, the results of the study may not be quite conclusive as to the adverse effect of OCs on the breast, but they call for further investigation of this problem.