Kevin McGeechan

Meta-analysis: retinal vessel caliber and risk for coronary heart disease.

Coronary heart disease (CHD) remains the leading cause of death in the United States despite advances in prevention, diagnosis, and therapy. Further improvement in outcomes can be achieved through more accurate identification of persons at risk, enhanced understanding of pathogenesis, novel interventions, and better implementation of existing preventive and therapeutic strategies. Coronary microvascular dysfunction is increasingly recognized as an important contributor to CHD, particularly among women (1), and interest in noninvasive methods of assessing the coronary microcirculation is considerable (2). The coronary and retinal vessels undergo similar changes (such as sclerosis) in hypertension (3, 4), and assessment of retinal vessels may provide an indication of coronary microvascular damage (5). With the advent of computer-assisted methods for measuring retinal vessel caliber from retinal photographs, retinal vascular imaging has been found to independently predict increased risk for CHD in prospective epidemiologic studies (610), which raises the possibility of using retinal vessel assessment as a novel risk marker. However, the results reported thus far have not been consistent. The ARIC (Atherosclerosis Risk in Communities) study (6), the first large epidemiologic study to report associations of retinal vessel caliber with incident CHD, suggested that these associations were only present in middle-age women. Subsequent studies have produced conflicting results. The CHS (Cardiovascular Health Study) (9) reported associations of narrower retinal arterioles and wider venules with incident CHD in both older women and men, but other studies found associations mainly in younger women and men, with weak or no association in older persons (10). Differences in study populations and inclusion criteria may account for the varying findings. For example, participants with diabetes or prevalent CHD were included in some studies (10) but not others (9, 11), and analytic methods and adjustment for traditional cardiovascular risk factors varied considerably among studies. To provide robust evidence to address these discrepancies, we conducted a systematic review and an individual-participant meta-analysis of population-based cohort studies, adjusting for traditional risk factors, to determine the associations between retinal vessel caliber and CHD risk. We particularly examined whether the associations differed between women and men. Methods Data Extraction We reviewed the literature to identify all studies that measured retinal vessel caliber, documented CHD events, and were conducted in general populations. We conducted a search of MEDLINE and EMBASE of all studies published between 1950 and 4 June 2009. We used the MEDLINE search terms (exp Retinal Diseases/, (retina or retinal).tw., retinopathy.tw., Arteriolar narrowing.tw., Arterio-venous nicking.tw., Arteriovenous nicking.tw., venular dilatation.tw., venular dilation.tw., arterio-venular ratio.tw.) and (Cardiovascular Diseases/, exp Heart Diseases/, exp Vascular Diseases/, cardiovascular.tw., coronary.tw., heart.tw., mortality.tw.) and (incidence/, exp Mortality/, exp epidemiologic studies/, prognos

Prediction of Incident Stroke Events Based on Retinal Vessel Caliber: A Systematic Review and Individual-Participant Meta-Analysis

.tw., Prognosis/, predict

Assessing New Biomarkers and Predictive Models for Use in Clinical Practice: A Clinician's Guide

.mp., course.tw., (score or scoring or scored).tw., observ

Use of a decision aid including information on overdetection to support informed choice about breast cancer screening: a randomised controlled trial

.mp., risk:.mp., between group:.tw.) and (Photography/, Photomicrography/, photo

Three questions that patients can ask to improve the quality of information physicians give about treatment options: A cross-over trial

.tw., image

Helping women make choices about mammography screening : An online randomized trial of a decision aid for 40-year-old women

.tw.). We then searched the selected papers to identify studies that met our inclusion criteria: carried out in general populations, measured retinal vessel caliber from either photographic film or digital photographs by using computer-assisted methods, and recorded incident CHD events. We contacted the principal or lead investigators of the chosen studies and obtained individual-participant data from each study to investigate heterogeneity in the published results and, if appropriate, to calculate pooled estimates of the associations between retinal vessel caliber and CHD risk. If the investigators agreed to participate, we then requested original recorded data on individual retinal vessel caliber measurements, fatal and nonfatal CHD events and time to these events, baseline measurements of variables included in the Framingham Risk Score (age, sex, systolic blood pressure, serum total cholesterol and high-density lipoprotein [HDL] cholesterol levels, current smoking status, use of blood pressurelowering medications, and presence of diabetes), body mass index, diastolic blood pressure, leukocyte count, and previous CHD. Statistical Analysis We analyzed the data for women and men separately because our primary hypothesis was that retinal vessel caliber predicts incident CHD more strongly in women than in men (8). In addition, separate Framingham Risk Scores, with different coefficients for the variables in the score, are used for men and women (12). The standard deviation for the means of arteriolar and venular caliber was approximately 20 m. We estimated the hazard ratio associated with each 20-m decrease in arteriolar caliber and each 20-m increase in venular caliber, which were adjusted for the other retinal vessel caliber, the variables that make up the Framingham Risk Score, and other risk factors associated with CHD and retinal caliber (13, 14). We estimated these separately for each study by using a proportional hazards model. We then combined data from all studies and used a stratified proportional hazards model to test for interaction between the study stratification variable and retinal vessel caliber variables, as well as sex and the CHD risk factors. This tests heterogeneity across studies in associations with retinal vessel caliber. Where no heterogeneity was present, we obtained a pooled hazard ratio adjusted for the CHD risk factors. The stratified proportional hazards model allows the baseline hazard function to differ among the studies but assumes that the effect of the retinal vessel caliber and the other variables are fixed. We defined nonfatal CHD events as myocardial infarction, coronary artery bypass graft, or coronary angioplasty. For those events that were coded by using International Classification of Diseases, Tenth Edition, codes, we classified fatal events as CHD deaths if the main or underlying cause of death received a code from I21 to I25 or if the study-adjudicated cause of death was CHD. Within each study, we assessed the appropriate functional form of each of the continuous variables in the models by using fractional polynomials, and we tested the proportional hazards assumption by using plots of the Schoenfeld residuals and by testing for the effect of adding time-dependent covariates (15). We repeated the main analysis to examine the robustness of our results, this time standardizing the retinal vessel caliber measurements by dividing them by the study-specific standard deviations to allow for different means and standard deviations of the retinal vessel caliber measurements in the different studies (16). We also pooled the study-specific hazard ratios by using a random-effects model (17). Results Characteristics of the Studies Identified We found 3946 studies with our initial search strategy. We then identified 25 studies that had performed retinal assessments or vessel caliber measurements and had followed participants over time to monitor CHD events. Of these, 18 studies (1835) recorded only the presence of retinopathy and not retinal vessel caliber, were conducted exclusively in persons with diabetes, or recorded only fatal events (Figure 1). This left 7 studies that met our inclusion criteria. One study, MESA (Multi-Ethnic Study of Atherosclerosis) (36), had insufficient outcome data available at the time of the analysis. Investigators from the other 6 studies (ARIC, CHS, AusDiab [Australian Diabetes, Obesity and Lifestyle] study, BMES [Blue Mountains Eye Study], BDES [Beaver Dam Eye Study], and RS [Rotterdam Study]) agreed to provide data for the individual-participant meta-analysis. Figure 1. Literature search and selection. CHD = coronary heart disease. Table 1 shows the characteristics for 22159 participants from each of the 6 studies we included. The measurement of retinal vessel caliber was similar in each study with some slight variations (7, 3741). Briefly, participants in each study had retinal photographs (film or digital) taken for either a single eye or both eyes, centered on the optic disc and macula. The BDES and BMES both used the Zeiss FF3 camera (Carl Zeiss Meditec, Jena, Germany) and 30 fields (10), ARIC and CHS used the Canon CR6-45NM (Canon, Tokyo, Japan) with 45 fields (37, 41), AusDiab used the Canon CR45UAF (Canon) with 45 fields (38), and RS used the Topcon TRC-SS2 (Topcon, Tokyo, Japan) with 20 fields (11). Trained graders, who were masked to participant characteristics, then viewed the optic disc photographs. The graders measured the diameters of all arterioles and venules that coursed through a zone that surrounded the optic disc, one-half to one-whole disc diameter away from the optic disc margin, by using a computer-assisted software program specifically developed for this purpose (37). The measurement module was custom programmed in Khoros (public domain image-processing software from the University of New MexicoAlbuquerque, Albuquerque, New Mexico) and used the green channel of the digital image to enhance contrast of the retinal vessels against the retinal pigment epithelium. Table 1. Participant Characteristics The ARIC, BDES, and BMES used an earlier version of this software to measure the retinal vessel caliber, whereas AusDiab, CHS, and RS used a later version of the same software. Both versions are available on request from the reviewers or the Wisconsin Fundus Photograph Reading Center, University of WisconsinMadison, Madison, Wisconsin. For this meta-analysis, we summarized the individual mean retinal vessel calibers from each study by using the ParrHubbard formula (37). Reproducibility statistics were high for thes

Retinal vascular caliber and the development of hypertension: a meta-analysis of individual participant data.

The caliber of the retinal vessels has been shown to be associated with stroke events. However, the consistency and magnitude of association, and the changes in predicted risk independent of traditional risk factors, are unclear. To determine the association between retinal vessel caliber and the risk of stroke events, the investigators combined individual data from 20,798 people, who were free of stroke at baseline, in 6 cohort studies identified from a search of the Medline (National Library of Medicine, Bethesda, Maryland) and EMBASE (Elsevier B.V., Amsterdam, the Netherlands) databases. During follow-up of 5-12 years, 945 (4.5%) incident stroke events were recorded. Wider retinal venular caliber predicted stroke (pooled hazard ratio = 1.15, 95% confidence interval: 1.05, 1.25 per 20-micron increase in caliber), but the caliber of retinal arterioles was not associated with stroke (pooled hazard ratio = 1.00, 95% confidence interval: 0.92, 1.08). There was weak evidence of heterogeneity in the hazard ratio for retinal venular caliber, which may be attributable to differences in follow-up strategies across studies. Inclusion of retinal venular caliber in prediction models containing traditional stroke risk factors reassigned 10.1% of people at intermediate risk into different, mostly lower, risk categories.

Risk Prediction of Coronary Heart Disease based on Retinal Vascular Caliber (From The Atherosclerosis Risk in Communities [ARIC] Study)

New biomarkers and predictive models that aim to improve the identification of people at risk of cardiovascular disease are constantly proposed. Clinicians need to be aware of the various methods used to assess these biomarkers and models and how these should be interpreted. New biomarkers and models are assessed in terms of their contribution to global fit, discrimination, calibration, and reclassification. These measures, when used in isolation, do not address the clinically important questions of whether the new model predicts risk more accurately than existing models and whether the risks predicted for individuals are sufficiently different to warrant a change in treatment decisions. We recommend that these measures be supplemented with graphical displays such as a calibration plot for the Hosmer-Lemeshow test and a scatterplot of the risks predicted by the models being compared. We encourage researchers to report such analyses from studies on the clinical utility of new biomarkers because this information is pertinent for the clinician who must decide whether to test for a new biomarker in their clinical practice.

Citation bias in reported smoking prevalence in people with schizophrenia

BACKGROUND Mammography screening can reduce breast cancer mortality. However, most women are unaware that inconsequential disease can also be detected by screening, leading to overdiagnosis and overtreatment. We aimed to investigate whether including information about overdetection of breast cancer in a decision aid would help women aged around 50 years to make an informed choice about breast screening. METHODS We did a community-based, parallel-group, randomised controlled trial in New South Wales, Australia, using a random cohort of women aged 48-50 years. Recruitment to the study was done by telephone; women were eligible if they had not had mammography in the past 2 years and did not have a personal or strong family history of breast cancer. With a computer program, we randomly assigned 879 participants to either the intervention decision aid (comprising evidence-based explanatory and quantitative information on overdetection, breast cancer mortality reduction, and false positives) or a control decision aid (including information on breast cancer mortality reduction and false positives). Participants and interviewers were masked to group assignment. The primary outcome was informed choice (defined as adequate knowledge and consistency between attitudes and screening intentions), which we assessed by telephone interview about 3 weeks after random allocation. The primary outcome was analysed in all women who completed the relevant follow-up interview questions fully. This trial is registered with the Australian New Zealand Clinical Trials Registry, number ACTRN12613001035718. FINDINGS Between January, 2014, and July, 2014, 440 women were allocated to the intervention group and 439 were assigned to the control group. 21 women in the intervention group and 20 controls were lost to follow-up; a further ten women assigned to the intervention and 11 controls did not answer all questions on attitudes. Therefore, 409 women in the intervention group and 408 controls were analysed for the primary outcome. 99 (24%) of 409 women in the intervention group made an informed choice compared with 63 (15%) of 408 in the control group (difference 9%, 95% CI 3-14; p=0·0017). Compared with controls, more women in the intervention group met the threshold for adequate overall knowledge (122/419 [29%] vs 71/419 [17%]; difference 12%, 95% CI 6-18; p<0·0001), fewer women expressed positive attitudes towards screening (282/409 [69%] vs 340/408 [83%]; 14%, 9-20; p<0·0001), and fewer women intended to be screened (308/419 [74%] vs 363/419 [87%]; 13%, 8-19; p<0·0001). When conceptual knowledge alone was considered, 203 (50%) of 409 women in the intervention group made an informed choice compared with 79 (19%) of 408 in the control group (p<0·0001). INTERPRETATION Information on overdetection of breast cancer provided within a decision aid increased the number of women making an informed choice about breast screening. Becoming better informed might mean women are less likely to choose screening. FUNDING Australian National Health and Medical Research Council.

Who Uses Smoking Cessation Apps? A Feasibility Study Across Three Countries via Smartphones

OBJECTIVE To test the effect of three questions (what are my options? what are the benefits and harms? and how likely are these?), on information provided by physicians about treatment options. METHODS We used a cross-over trial using two unannounced standardized patients (SPs) simulating a presentation of mild-moderate depression. One SP was assigned the intervention role (asking the questions), the other the control role. An intervention and control SP visited each physician, order allocated randomly. The study was conducted in family practices in Sydney, Australia, during 2008-09. Data were obtained from consultation audio-recordings. Information about treatment options and patient involvement were analyzed using the Assessing Communication about Evidence and Patient Preferences (ACEPP) tool and the OPTION tool. RESULTS Thirty-six SP visits were completed (18 intervention, 18 control). Scores were higher in intervention consultations than controls: ACEPP scores 21.4 vs. 16.6, p<0.001, difference 4.7 (95% CI 2.3-7.0) and OPTION scores 36 vs. 25, p=0.001, difference 11.5 (95% CI 5.1-17.8), indicating greater information provision and behavior supporting patient involvement. CONCLUSION Asking these three questions improved information given by family physicians and increased physician facilitation of patient involvement. Practice implications. These questions can drive evidence-based practice, strengthen patient-physician communication, and improve safety and quality.