The development and validation of a multivariable prognostic model to predict foot ulceration in diabetes using a systematic review and individual patient data meta-analyses.

AIMS: Diabetes guidelines recommend screening for the risk of foot ulceration but vary substantially in the underlying evidence base. Our purpose was to derive and validate a prognostic model of independent risk factors for foot ulceration in diabetes using all available individual patient data from cohort studies conducted worldwide. METHODS: We conducted a systematic review and meta-analysis of individual patient data from 10 cohort studies of risk factors in the prediction of foot ulceration in diabetes. Predictors were selected for plausibility, availability and low heterogeneity. Logistic regression produced adjusted odds ratios (ORs) for foot ulceration by ulceration history, monofilament insensitivity, any absent pedal pulse, age, sex and diabetes duration. RESULTS: The 10 studies contained data from 16 385 participants. A history of foot ulceration produced the largest OR [6.59 (95% CI 2.49 to 17.45)], insensitivity to a 10 g monofilament [3.18 (95% CI 2.65 to 3.82)] and any absent pedal pulse [1.97 (95% CI 1.62 to 2.39)] were consistently, independently predictive. Combining three predictors produced sensitivities between 90.0% (95% CI 69.9% to 97.2%) and 95.3% (95% CI 84.5% to 98.7%); the corresponding specificities were between 12.1% (95% CI 8.2% to 17.3%) and 63.9% (95% CI 61.1% to 66.6%). CONCLUSIONS: This prognostic model of only three risk factors, a history of foot ulceration, an inability to feel a 10 g monofilament and the absence of any pedal pulse, compares favourably with more complex approaches to foot risk assessment recommended in clinical diabetes guidelines.

Brain choline concentration. Early quantitative marker of ischemia and infarct expansion?

OBJECTIVE: Better prediction of tissue prognosis in acute stroke might improve treatment decisions. We hypothesized that there are metabolic ischemic disturbances measurable noninvasively by proton magnetic resonance spectroscopy ((1)H MRS) that occur earlier than any structural changes visible on diffusion-tensor imaging (DTI), which may therefore serve for territorial identification of tissue at risk. METHODS: We performed multivoxel (1)H MRS plus DTI within a maximum of 26 hours, and DTI at 3-7 days, after ischemic stroke. We compared choline, lactate, N-acetylaspartate, and creatine concentrations in normal-appearing voxels that became infarcted (infarct expansion) with normal-appearing voxels around the infarct that remained "healthy" (nonexpansion) on follow-up DTI. Each infarct expansion voxel was additionally classified as either complete infarct expansion (infarcted tissue on follow-up DTI covered > or =50% of the voxel) or partial infarct expansion (<50% of voxel). RESULTS: In 31 patients (NIH Stroke Scale score 0-28), there were 108 infarct nonexpansion voxels and 113 infarct expansion voxels (of which 80 were complete expansion and 33 partial expansion voxels). Brain choline concentration increased for each change in expansion category from nonexpansion, via partial expansion to complete expansion (2,423, 3,843, 4,158 IU; p < 0.05). Changes in lactate, N-acetylaspartate, and creatine concentrations in expansion category were insignificant although for lactate there was a tendency to such association. CONCLUSIONS: Choline concentration measurable with (1)H MRS was elevated in peri-ischemic normal-appearing brain that became infarcted by 3-7 days. The degree of elevation was associated with the amount of infarct expansion. (1)H MRS might identify DTI-normal-appearing tissue at risk of conversion to infarction in early stroke.

When are summary ROC curves appropriate for diagnostic meta-analyses?

Diagnostic tests are increasingly evaluated with systematic reviews and this has lead to the recent developments of statistical methods to analyse such data. The most commonly used method is the summary receiver operating characteristic (SROC) curve, which can be fitted with a non-linear bivariate random-effects model. This paper focuses on the practical problems of interpreting and presenting data from such analyses. First, many meta-analyses may be underpowered to obtain reliable estimates of the SROC parameters. Second, the SROC model may be inappropriate. In these situations, a summary with two univariate meta-analyses of the true and false positive rates (TPRs and FPRs) may be more appropriate. We characterize the type of problems that can occur in fitting these models and present an algorithm to guide the analyst of such studies, with illustrations from analyses of published data. A set of R functions, freely available to perform these analyses, can be downloaded from (www.diagmeta.info).

Changes in NAA and lactate following ischemic stroke: a serial MR spectroscopic imaging study.

OBJECTIVE: Although much tissue damage may occur within the first few hours of ischemic stroke, the duration of tissue injury is not well defined. We assessed the temporal pattern of neuronal loss and ischemia after ischemic stroke using magnetic resonance spectroscopic imaging (MRSI) and diffusion-weighted imaging (DWI). METHODS: We measured N-acetylaspartate (NAA) and lactate in 51 patients with acute ischemic stroke at five time points, from admission to 3 months, in voxels classified as normal, possibly or definitely abnormal (ischemic) according to the appearance of the stroke lesion on the admission DWI. We compared changes in NAA and lactate in different voxel classes using linear mixed models. RESULTS: NAA was significantly reduced from admission in definitely and possibly abnormal (p < 0.01) compared to contralateral normal voxels, reaching a nadir by 2 weeks and remaining reduced at 3 months. Lactate was significantly increased in definitely and possibly abnormal voxels (p < 0.01) during the first 5 days, falling to normal at 2 weeks, rising again later in these voxels. CONCLUSION: The progressive fall in N-acetylaspartate suggests that some additional neuronal death may continue beyond the first few hours for up to 2 weeks or longer. The mechanism is unclear but, if correct, then it is possible that interventions to limit this ongoing subacute tissue damage might add to the benefit of hyperacute treatment, making further improvements in outcome possible.

Accurate, practical and cost-effective assessment of carotid stenosis in the UK.

OBJECTIVES: To determine whether less invasive imaging tests [ultrasound (US), magnetic resonance angiography (MRA), computed tomographic angiography (CTA) and contrast-enhanced MRA (CEMRA)], alone or combined, could replace intra-arterial angiography (IAA), what effect this would have on strokes and deaths, endarterectomies performed and costs, and whether less invasive tests were cost-effective. DATA SOURCES: Electronic databases covering the years 1980-2003 inclusive, updated to April 2004. Key journals from 1990 to the end of 2002. REVIEW METHODS: The authors constituted a panel of experts in stroke, imaging, vascular surgery, statistics and health economic modelling. The accuracy of less invasive carotid imaging was systematically reviewed using Standards for Reporting of Diagnostic Accuracy (STARD) methodology, supplemented by individual patient data from UK primary research and audit studies. A systematic review of the costs of less invasive tests, outpatient clinics, endarterectomy and stroke was performed, along with a microcosting exercise. A model of the process of care following a transient ischaemic attack (TIA)/minor stroke was developed, populated with data from stroke epidemiology studies in the UK, effects of medical and surgical interventions, outcomes, quality of life and costs. A survey of UK stroke prevention clinics provided typical timings. Twenty-two different carotid imaging strategies were evaluated for short- and long-term outcomes, quality-adjusted life-years (QALYs) and net benefit. RESULTS: In 41 included studies (2404 patients, median age 60-65 years), most data were available on 70-99% stenosis. CEMRA was the most accurate [sensitivity 0.94, 95% confidence interval (CI) 0.88 to 0.97; specificity 0.93, 95% CI 0.89 to 0.96], compared with US, MRA and CTA, which were all similar (e.g. for US: sensitivity 0.89, 95% CI 0.85 to 0.92; specificity 0.84, 95% CI 0.77 to 0.89). Data for 50-69% stenoses and on combinations of tests were too sparse to be reliable. There was heterogeneity between studies for all imaging modalities except for CTA. The individual patient data (2416 patients) showed that the literature overestimated test accuracy in routine practice and that, in general, tests perform with higher sensitivity and specificity in asymptomatic than in symptomatic arteries. In the cost-effectiveness model, on current UK timings, strategies allowed more patients to reach endarterectomy very quickly, and where those with 50-69% stenosis would be offered surgery in addition to those with 70-99%, prevented most strokes and produced greatest net benefit. This included most strategies with US as first or repeat test, and not those with IAA. However, the model was sensitive to less invasive test accuracy, cost and timing of endarterectomy. In patients investigated late after TIA, test accuracy is crucial and CEMRA should be used before surgery. CONCLUSIONS: In the UK, less invasive tests can be used in place of IAA if radiologists trained in carotid imaging are available. Imaging should be carefully audited. Stroke prevention clinics should reduce waiting times at all stages to improve speed of access to endarterectomy. In patients presenting late after TIA, test accuracy is very important and US results should be confirmed by CEMRA, as patients with 50-69% stenosis are less likely to benefit. More data are required to define the accuracy of the less invasive tests, with improvements made in the data collection methods used and how data are presented. Consideration should also be given to the use of new technologies and randomised trials.

Non-invasive imaging compared with intra-arterial angiography in the diagnosis of symptomatic carotid stenosis: a meta-analysis.

BACKGROUND: Accurate carotid imaging is important for effective secondary stroke prevention. Non-invasive imaging, now widely available, is replacing intra-arterial angiography for carotid stenosis, but the accuracy remains uncertain despite an extensive literature. We systematically reviewed the accuracy of non-invasive imaging compared with intra-arterial angiography for diagnosing carotid stenosis in patients with carotid territory ischaemic symptoms. METHODS: We searched for articles published between 1980 and April 2004; included studies comparing non-invasive imaging with intra-arterial angiography that met Standards for Reporting of Diagnostic Accuracy (STARD) criteria; extracted data to calculate sensitivity and specificity of non-invasive imaging, to test for heterogeneity and to perform sensitivity analyses; and categorised percent stenosis by the North American Symptomatic Carotid Endarterectomy Trial (NASCET) method. RESULTS: In 41 included studies (2541 patients, 4876 arteries), contrast-enhanced MR angiography was more sensitive (0.94, 95% CI 0.88-0.97) and specific (0.93, 95% CI 0.89-0.96) for 70-99% stenosis than Doppler ultrasound, MR angiography, and CT angiography (sensitivities 0.89, 0.88, 0.76; specificities 0.84, 0.84, 0.94, respectively). Data for 50-69% stenoses and combinations of non-invasive tests were sparse and unreliable. There was heterogeneity between studies and evidence of publication bias. INTERPRETATION: Non-invasive tests, used cautiously, could replace intra-arterial carotid angiography for 70-99% stenosis. However, more data are required to determine their accuracy, especially at 50-69% stenoses where the balance of risk and benefit for carotid endarterectomy is particularly narrow, and to explore and overcome heterogeneity. Methodology for evaluating imaging tests should be improved; blinded, prospective studies in clinically relevant patients are essential basic characteristics.