A methodological appraisal of research on prognosis after transient ischemic attacks

Identifying transient ischemic attack (TIA) patients at high-risk of adverse outcomes: development and validation of an approach using electronic health record data

Background

Risk-stratification tools that have been developed to identify transient ischemic attack (TIA) patients at risk of recurrent vascular events typically include factors which are not readily available in electronic health record systems. Our objective was to evaluate two TIA risk stratification approaches using electronic health record data.

Methods

Patients with TIA who were cared for in Department of Veterans Affairs hospitals (October 2015—September 2018) were included. The six outcomes were mortality, recurrent ischemic stroke, and the combined endpoint of stroke or death at 90-days and 1-year post-index TIA event. The cohort was split into development and validation samples. We examined the risk stratification of two scores constructed using electronic health record data. The Clinical Assessment Needs (CAN) score is a validated measure of risk of hospitalization or death. The PREVENT score was developed specifically for TIA risk stratification.

Results

A total of N = 5250 TIA patients were included in the derivation sample and N = 4248 in the validation sample. The PREVENT score had higher c-statistics than the CAN score across all outcomes in both samples. Within the validation sample the c-statistics for the PREVENT score were: 0.847 for 90-day mortality, 0.814 for 1-year mortality, 0.665 for 90-day stroke, and 0.653 for 1-year stroke, 0.699 for 90-day stroke or death, and 0.744 for 1-year stroke or death. The PREVENT score classified patients into categories with extreme nadir and zenith outcome rates. The observed 1-year mortality rate among validation patients was 7.1%; the PREVENT score lowest decile of patients had 0% mortality and the highest decile group had 30.4% mortality.

Conclusions

The PREVENT score had strong c-statistics for the mortality outcomes and classified patients into distinct risk categories. Learning healthcare systems could implement TIA risk stratification tools within electronic health records to support ongoing quality improvement.

Registration

ClinicalTrials.gov Identifier: NCT02769338.

Prediction and prevention of stroke after transient ischemic attack in the short and long term

Over the last 5 years, a number of studies have shown the early risk of stroke following transient ischemic attack (TIA) to be of the order of 5-10% at 1 week and 10-20% at 3 months, considerably higher than previously estimated. Because these studies have been carried out in a variety of different clinical settings, their findings are likely to be generalizable. Various independent prognostic factors for this early risk of stroke have been identified and models, based on clinical features at presentation, have been derived and validated to predict risk of stroke within 7 and 90 days after TIA. At the same time, diffusion-weighted magnetic resonance imaging and carotid imaging provide prognostic information and are likely to refine risk prediction further, although no unified model combining clinical and imaging data currently exists. Uncertainty continues surrounding the most effective secondary prevention in the hyperacute phase after TIA, especially in the choice of antiplatelet agents, although clinical trials to address this question are ongoing. However, the need for carotid endarterectomy in patients with symptomatic carotid stenosis is well established. The risk of vascular disease in the medium term (1-5 years) following TIA has been more widely studied, and predictive models for this are available. Recent data on the long-term (10 years and beyond) vascular risk after TIA demonstrate ongoing mortality from both cerebrovascular and cardiovascular causes, highlighting the need for continued secondary prevention.

Prognostic social factors in the subacute phase after a stroke for the discharge destination from the hospital stroke-unit. A systematic review of the literature

PURPOSE

The objective of our study was to identify prognostic social factors in the subacute phase after stroke for the discharge destination from the hospital stroke-unit.

METHODS

A systematic literature search was performed, designed in accordance with the Cochrane Collaboration criteria. Internal, statistical and external validity of the studies were assessed using a checklist with 11 methodological criteria.

RESULTS

Characteristics of the social situation that proved to be important for prediction of the discharge destination are marital status and social support. Quantity and methodological quality of the research studies were insufficient, and the number of possible social prognostic factors investigated was limited by the absence of a conceptual framework of social subdomains in the studies, including an unambiguous definition of the prognostic social factors within these subdomains.

CONCLUSIONS

A great need exists for research into the prognostic qualities of the following social factors: the ability to provide support, presence, and readiness of the homefront; the availability of professional care, personal financial means, membership of societies and clubs, frequency of contacts with close relatives and friends; the quality of the patient's residence with regard to the adaptation to the needs and abilities of the patient. A commitment about the aforementioned conceptual framework is mandatory.

Microvessel density as a prognostic factor in non-small-cell lung carcinoma: a meta-analysis of individual patient data

BACKGROUND

Angiogenesis is a potential prognostic factor that has been investigated in patients with non-small-cell lung carcinoma. However, published studies of the role of angiogenesis as a prognostic factor are inconclusive. We aimed to collect individual patient data to assess microvessel-density counts (ie, a measure of angiogenesis) as a prognostic factor in non-small-cell lung carcinoma.

METHODS

We obtained published and unpublished datasets and extracted appropriate data, taking particular care to ensure data quality. Detailed information was obtained for the laboratory methods used by every research centre that generated the data. The outcome of interest was overall survival. We did a meta-analysis to estimate the prognostic role of microvessel density by combining separately estimated hazard ratios (HR) from every study, which were adjusted for tumour stage and age. Analyses were done separately for studies that used the Chalkley method or for those that counted all microvessels.

FINDINGS

17 centres provided data for 3200 patients, 2719 of which were included in the analysis. All but three centres (datasets 9, 10, and 13-367 cases) had already published their findings, and six had updated follow-up information (datasets 1, 2, 3, 6, 7, and 8-1273 cases). For all but three centres (datasets 4, 11, and 13) some data corrections were necessary. For microvessel density counts obtained by the Chalkley method, the HR for death per extra microvessel was 1.05 (95% CI 1.01-1.09, p=0.03) when analysed as a continuous variable. For microvessel density counts obtained by the all vessels method, the HR for death per ten extra microvessels was 1.03 (0.97-1.09, p=0.3) when analysed as a continuous variable.

INTERPRETATION

Microvessel density does not seem to be a prognostic factor in patients with non-metastatic surgically treated non-small-cell lung carcinoma. This conclusion contradicts the results of a meta-analysis of published data only. Therefore, the methodology used to assess prognostic factors should be assessed carefully.

Long-term survival and vascular event risk after transient ischaemic attack or minor ischaemic stroke: a cohort study

BACKGROUND

Determinants of survival and of risk of vascular events after transient ischaemic attack (TIA) or minor ischaemic stroke are not well defined in the long term. We aimed to restudy these risks in a prospective cohort of patients after TIA or minor ischaemic stroke (Rankin grade< or =3), after 10 years or more.

METHODS

We assessed the survival status and occurrence of vascular events in 2473 participants of the Dutch TIA Trial (recruitment in 1986-89; arterial cause of cerebral ischaemia). We included 24 hospitals in the Netherlands that recruited at least 50 patients. Primary outcomes were all-cause mortality and the composite event of death from all vascular causes, non-fatal stroke, and non-fatal myocardial infarction. We assessed cumulative risks by Kaplan-Meier analysis and prognostic factors with Cox univariate and multivariate analysis.

FINDINGS

Follow-up was complete in 2447 (99%) patients. After a mean follow-up of 10.1 years, 1489 (60%) patients had died and 1336 (54%) had had at least one vascular event. 10-year risk of death was 42.7% (95% CI 40.8-44.7). Age and sex-adjusted hazard ratios were 3.33 (2.97-3.73) for age over 65 years, 2.10 (1.79-2.48) for diabetes, 1.77 (1.45-2.15) for claudication, 1.94 (1.42-2.65) for previous peripheral vascular surgery, and 1.50 (1.31-1.71) for pathological Q waves on baseline electrocardiogram. 10-year risk of a vascular event was 44.1% (42.0-46.1). After falling in the first 3 years, yearly risk of a vascular event increased over time. Predictive factors for risk of vascular events were similar to those for risk of death.

INTERPRETATION

Long-term secondary prevention in patients with cerebral ischaemia still has room for further improvement.

Prognosis of Neurologic Diseases

This chapter covers the topic of prognostic studies and the application of research principles to questions of prognosis. Studies of prognosis are the core of clinical neurology and clinical epidemiology, and are central to the work of clinicians and clinician investigators. Prognosis can be regarded as a set of outcomes and their associated probabilities following the occurrence of some defining event or diagnosis that can be a symptom, sign, test result, or disease. The chapter presents information on study design principles for prognostic studies, common study biases, and statistical analysis considerations.

The use of a modified Delphi procedure for the determination of 26 prognostic factors in the sub-acute stage of stroke

The aim of this study was to reach consensus about the prognostic factors when deciding the discharge destination from a hospital stroke unit, and to construct a prognostic conceptual framework. To realise an optimal integration of knowledge from research findings and from clinical experience by expert opinions we used a 'modified Delphi Technique', which is the most commonly used method for the production of clinical guidelines. The process yielded 26 prognostic factors, which were arranged in clinical and social sub-domains. The sub-domains and the factors within each sub-domain were prioritised according to their assumed predictive value for the decision process. The order of importance of the prognostic factors in the clinical domain was: (1) disabilities, (2) pre-morbid disabilities, (3) impairments and (4) disease/biology; and the order of importance of the factors in the social domain was: (1) home front, (2) social situation and (3) residence. The Delphi procedure is an excellent instrument to determine and prioritise prognostic factors. With this procedure research-based and consensus-based knowledge can be combined. For a valid procedure it is mandatory to state explicitly in advance how the scores will be judged, and to explain the scientific level of the evidence during the whole procedure.

Prognostic factors in the subacute phase after stroke for the future residence after six months to one year. A systematic review of the literature

OBJECTIVE

To identify evidence-based prognostic factors in the subacute phase after a stroke for future residence at six months to one year post stroke.

DESIGN

Systematic literature search designed in accordance with the Cochrane Collaboration criteria with the following data sources: (1) MEDLINE, EMBASE, CINAHL, Current Contents, Cochrane Database of Systematic Reviews, PsycLIT and Sociological Abstracts. (2) Reference lists, personal archives and consultation of experts in the field. (3) Guidelines.

METHODS

Inclusion criteria were: (1) cohort studies of patients with an ischaemic or haemorrhagic stroke; (2) inception cohort with assessment of prognostic factors within the first two weeks after stroke; (3) outcome measures for future residence; and (4) a follow-up of six months to one year. Internal, statistical and external validity of the studies were assessed using a checklist with 11 methodological criteria in accordance with the recommendations of the Cochrane Collaboration.

RESULTS

From 1027 potentially relevant studies 10 studies involving a total of 3564 patients met the inclusion criteria. No prognostic factor was identified in at least two level A (i.e., a good level of scientific evidence according to the methodological score) studies, our standard for scientific proof. The following factors were found in at least one level A study: low initial ADL functioning, high age, cognitive disturbance, paresis of arm and leg, not alert as initial level of consciousness, old hemiplegia, homonymous hemianopia, visual extinction, constructional apraxia, no transfer to the stroke unit, nonlacunar stroke type, visuospatial construction problems, urinary incontinence and female gender.

CONCLUSIONS

At present there is insufficient evidence concerning possible predictors in the subacute stage of stroke to make an evidence-based prediction of the future residence. In the scientific research until now social factors and their contribution to the possibility of living independently have not been investigated, or at least less well. None of the studies in this review described a conceptual framework as basis for the choice of the examined prognostic factors.

Prognostic factors for ambulation and activities of daily living in the subacute phase after stroke. A systematic review of the literature

OBJECTIVE

To identify evidence-based prognostic factors in the subacute phase after stroke for activities of daily living (ADL) and ambulation at six months to one year after stroke.

DESIGN

Systematic literature search designed in accordance with the Cochrane Collaboration criteria with the following data sources: (1) MEDLINE, EMBASE, CINAHL, Current Contents, Cochrane Database of Systematic Reviews, Psyclit, and Sociological Abstracts. (2) Reference lists, personal archives, and consultation of experts. (3) Guidelines.

METHODS

Inclusion criteria were: (1) cohort studies of patients with an ischaemic or haemorrhagic stroke; (2) inception cohort with assessment of prognostic factors within the first two weeks after stroke; (3) outcome measures for ADL and ambulation; and (4) a follow-up of six months to one year. Internal, statistical and external validity of the studies were assessed using a checklist with 11 methodological criteria in accordance with the recommendations of the Cochrane Collaboration.

RESULTS

From 1,027 potentially relevant studies 26 studies involving a total of 7,850 patients met the inclusion criteria. Incontinence for urine is the only prognostic factor identified in three studies with a level A (i.e., a good level of scientific evidence according to the methodological score). The following factors were found in one level A study: initial ADL disability and ambulation, high age, severe paresis or paralysis, impaired swallowing, ideomotor apraxia, ideational apraxia, and visuospatial construction problems; as well as factors relating to complications of an ischaemic stroke, such as extraparenchymal bleeding, cerebral oedema and size of intraparenchymal haemorrhage.

CONCLUSIONS

The present evidence concerning possible predictors in the subacute stage of stroke has insufficient quality to make an evidence-based prediction of ADL and ambulation after stroke because only one prognostic factor was demonstrated in at least two level A studies, our cut-off for sufficient scientific evidence.

Prediction in idiopathic membranous nephropathy

BACKGROUND

Results of the prognosis of idiopathic membranous nephropathy are conflictive and prevent an effective risk stratification. These conflicts are explained in part by insufficient consideration of methodological principles for prognostic research. This cohort study is aimed at identifying clinical predictors for risk stratification while paying particular attention to methodology.

METHODS

We studied 120 patients with idiopathic membranous nephropathy. Baseline data were extracted at the time of diagnostic renal biopsy, and patients were followed prospectively. Predictors were identified for the end points end-stage renal failure (ESRF) and ESRF or death.

RESULTS

From the 120 patients followed for a median of five years (1 to 24 years), 19% developed end-stage renal failure or deterioration of renal function. Proteinuria of more than 3.5 g/day persisted in 34%, and 47% were in complete or partial remission. The Kaplan-Meier estimated probability of renal survival was 91 +/- 3% at five years and 75 +/- 6% at ten years. The predictors for the primary outcome, ESRF, identified in a Cox proportional hazards model, were histological stage (Ehrenreich-Churg) III-IV (hazard ratio 5.3, CI 1.9 to 15.0, P = 0.002) and nephrotic syndrome (hazard ratio 7.9, CI 1.1 to 61.5, P = 0.04); the predictors for the secondary outcome, ESRF or patient death, were histological stage III-IV (hazard ratio 2.8, CI 1.3 to 6.0, P = 0.008), nephrotic syndrome (hazard ratio 3.0, CI 1.1 to 8.0, P = 0.003) and comorbidity (hazard ratio 2.8, CI 1.3 to 5.9, P = 0.007). Nephrotic syndrome and histological stage III-IV allowed the demarcation of the high-risk group from the remaining patients (P < 0.0001).

CONCLUSION

Histological stage, nephrotic syndrome, and comorbidity predict end-stage renal failure or death in idiopathic membranous nephropathy. Identification of the high-risk group at the time of diagnostic renal biopsy will permit appropriate treatment to be targeted to the patients who might benefit the most from the therapy in future clinical trials.

Reducing Bias in Observational Research on Alcohol Withdrawal Syndrome

Observational methods are often used to study alcohol withdrawal syndrome. Observational cohorts and case-control studies are useful in research on prognosis and in situations where ethical concerns might prevent the use of clinical trial designs. In addition, case-control studies are appropriate for studying the rare but serious complications of alcohol withdrawal syndrome. Despite their appeal and suitability, these methods are subject to a number of biases that can render their results suspect. Susceptibility, diagnostic suspicion, and exposure suspicion bias are among the important biases to consider in observational studies on alcohol withdrawal syndrome. Strategies for reducing these biases include the creation of inception cohorts, the use of operational definitions, and blinding. Increased attention to these strategies should help clinicians and researchers improve the practice of caring for patients with alcohol withdrawal syndrome.

Prognosis of idiopathic membranous nephropathy: a methodologic meta-analysis

Results in studies on prognosis and treatment of membranous nephropathy are conflicting. The aim of this investigation was to analyze the methodology of the existing research and to identify sources of these conflicting results. Studies published on prognosis of membranous nephropathy from 1970 to 1995 were identified using a Medline database literature search. The criteria for inclusion in the methodologic analysis were: (1) original article; (2) cohort study or clinical trial with > or = 50 adults; (3) zero time near the diagnostic renal biopsy; and (4) follow-up > or = six months. Ten well-accepted methodologic standards for prognostic research were applied to each study and the compliance was evaluated. Among the 26 studies that met the inclusion criteria, the median number of standards fulfilled was 4 and the highest was 7. The proportion of studies adhering to the individual standards was: (1) adequate diagnostic criteria, 35%; (2) definite end point, 46%; (3) adequate analysis of a surrogate end point, 52%; (4) analysis of baseline severity, 0%; (5) indication of baseline frequency for candidate predictors, 35%; (6) reproducible classification of predictors, 85%; (7) multivariable analysis, 50%; (8) identification of the variables' importance in multivariable analysis, 38%; (9) evaluation of the effect of treatment on predictors, 19%; and (10) adequate analysis of censored patients, 58%. We conclude that basic methodologic principles have frequently been disregarded. The consideration of these standards in future research can improve the interpretability and applicability of results and help reconcile conflict when results are compared among different studies.

Understanding prognosis to improve rehabilitation: the example of lateral elbow pain

OBJECTIVE

A systematic overview of evidence aimed at determining the clinical course of lateral elbow pain and prognostic factors that affect elbow pain duration and outcomes.

DATA SOURCES

Online bibliographic database searches from 1983 to 1994; information requests from selected authors and bibliography screenings.

STUDY SELECTION

One author reviewed 424 articles; 40 met the following eligibility criteria: any study with primary data on soft tissue injuries specific to the elbow which referred to prognosis or reported use of any outcome measure.

DATA ABSTRACTION

Strength of evidence grade based on clinical epidemiological validity assessment. Criteria included in the validity assessment included case definition, patient selection, follow-up, outcome, prognostic factors, and analysis. All eligible studies were independently assessed by two investigators.

DATA SYNTHESIS

Four studies (10%) were judged to provide moderate strength of evidence; no studies were graded as providing strong evidence on prognosis. All four moderate-quality studies were clinical trials of short duration. One study indicated that site of lesion and prior occurrence may be predictive of poorer outcome in patients with lateral epicondylitis.

CONCLUSION

The majority of studies on lateral elbow pain were limited by methodological weaknesses in selection and definition of the study population, length of follow-up, and analysis of prognostic factors. Estimates of duration were only available from weaker studies with longer follow-up times; significant subject heterogeneity in the weaker studies prevented a determination of usual clinical course. More methodologically rigorous research on prognosis could assist clinicians in patient care and evaluation of interventions.