A predictive model to identify patients with suspected acute coronary syndromes at high risk of cardiac arrest or in-hospital mortality: An IMMEDIATE Trial sub-study,,

BACKGROUND

The IMMEDIATE Trial of emergency medical service use of intravenous glucose-insulin-potassium (GIK) very early in acute coronary syndromes (ACS) showed benefit for the composite outcome of cardiac arrest or in-hospital mortality.

OBJECTIVES

This analysis of IMMEDIATE Trial data sought to develop a predictive model to help clinicians identify patients at highest risk for this outcome and most likely to benefit from GIK.

METHODS

Multivariable logistic regression was used to develop a predictive model for the composite endpoint cardiac arrest or in-hospital mortality using the 460 participants in the placebo arm of the IMMEDIATE Trial.

RESULTS

The final model had four variables: advanced age, low systolic blood pressure, ST elevation in the presenting electrocardiogram, and duration of time since ischemic symptom onset. Predictive performance was good, with a C statistic of 0.75, as was its calibration. Stratifying patients into three risk categories based on the model's predictions, there was an absolute risk reduction of 8.6% with GIK in the high-risk tertile, corresponding to 12 patients needed to treat to prevent one bad outcome. The corresponding values for the low-risk tertile were 0.8% and 125, respectively.

CONCLUSIONS

The multivariable predictive model developed identified patients with very early ACS at high risk of cardiac arrest or death. Using this model could assist treating those with greatest potential benefit from GIK.

Three simple rules to ensure reasonably credible subgroup analyses

The limitations of subgroup analyses are well established—false positives due to multiple comparisons, false negatives due to inadequate power, and limited ability to inform individual treatment decisions because patients have multiple characteristics that vary simultaneously. In this article, we apply Bayes’s rule to determine the probability that a positive subgroup analysis is a true positive. From this framework, we derive simple rules to determine when subgroup analyses can be performed as hypothesis testing analyses and thus inform when subgroup analyses should influence how we practice medicine.

Anticoagulant vs. antiplatelet therapy in patients with cryptogenic stroke and patent foramen ovale: an individual participant data meta-analysis

AIMS

The preferred antithrombotic strategy for secondary prevention in patients with cryptogenic stroke (CS) and patent foramen ovale (PFO) is unknown. We pooled multiple observational studies and used propensity score-based methods to estimate the comparative effectiveness of oral anticoagulation (OAC) compared with antiplatelet therapy (APT).

METHODS AND RESULTS

Individual participant data from 12 databases of medically treated patients with CS and PFO were analysed with Cox regression models, to estimate database-specific hazard ratios (HRs) comparing OAC with APT, for both the primary composite outcome [recurrent stroke, transient ischaemic attack (TIA), or death] and stroke alone. Propensity scores were applied via inverse probability of treatment weighting to control for confounding. We synthesized database-specific HRs using random-effects meta-analysis models. This analysis included 2385 (OAC = 804 and APT = 1581) patients with 227 composite endpoints (stroke/TIA/death). The difference between OAC and APT was not statistically significant for the primary composite outcome [adjusted HR = 0.76, 95% confidence interval (CI) 0.52-1.12] or for the secondary outcome of stroke alone (adjusted HR = 0.75, 95% CI 0.44-1.27). Results were consistent in analyses applying alternative weighting schemes, with the exception that OAC had a statistically significant beneficial effect on the composite outcome in analyses standardized to the patient population who actually received APT (adjusted HR = 0.64, 95% CI 0.42-0.99). Subgroup analyses did not detect statistically significant heterogeneity of treatment effects across clinically important patient groups.

CONCLUSION

We did not find a statistically significant difference comparing OAC with APT; our results justify randomized trials comparing different antithrombotic approaches in these patients.

Effectiveness and Implications of Alternative Placebo Treatments: A Systematic Review and Network Meta-analysis of Osteoarthritis Trials

BACKGROUND

Placebo controls are essential in evaluating the effectiveness of medical treatments. Although it is unclear whether different placebo interventions for osteoarthritis vary in efficacy, systematic differences would substantially affect interpretation of the results of placebo-controlled trials.

OBJECTIVE

To evaluate the effects of alternative placebo types on pain outcomes in knee osteoarthritis.

DATA SOURCES

MEDLINE, EMBASE, Web of Science, Google Scholar, and Cochrane Database from inception through 1 June 2015 and unpublished data.

STUDY SELECTION

149 randomized trials of adults with knee osteoarthritis that reported pain outcomes and compared widely used pharmaceuticals against oral, intra-articular, topical, and oral plus topical placebos.

DATA EXTRACTION

Study data were independently double-extracted; study quality was assessed by using the Cochrane risk of bias tool.

DATA SYNTHESIS

Placebo effects that were evaluated by using a network meta-analysis with 4 separate placebo nodes (differential model) showed that intra-articular placebo (effect size, 0.29 [95% credible interval, 0.09 to 0.49]) and topical placebo (effect size, 0.20 [credible interval, 0.02 to 0.38]) had significantly greater effect sizes than did oral placebo. This differential model showed marked differences in the relative efficacies and hierarchy of the active treatments compared with a network model that considered all placebos equivalent. In the model accounting for differential effects, intra-articular and topical therapies were superior to oral treatments in reducing pain. When these differential effects were ignored, oral nonsteroidal anti-inflammatory drugs were superior.

LIMITATIONS

Few studies compared different placebos directly. The study could not decisively conclude whether disease severity and co-interventions systematically differed between trials evaluating different placebos.

CONCLUSION

All placebos are not equal, and some can trigger clinically relevant responses. Differential placebo effects can substantially alter estimates of the relative efficacies of active treatments, an important consideration for the design of clinical trials and interpretation of their results.

PRIMARY FUNDING SOURCE

Agency for Healthcare Research and Quality.

Development and validation of a simplified Stroke-Thrombolytic Predictive Instrument

OBJECTIVES

The Stroke-Thrombolytic Predictive Instrument (Stroke-TPI) predicts the probability of good and bad outcomes with and without recombinant tissue plasminogen activator (rtPA). We sought to rebuild and externally validate a simpler Stroke-TPI to support implementation in routine clinical care.

METHODS

Using the original derivation cohort of 1,983 patients from a combined database of randomized clinical trials (NINDS [National Institute of Neurological Disorders and Stroke] 1 and 2; ATLANTIS [Alteplase Thrombolysis for Acute Noninterventional Therapy in Ischemic Stroke] A and B; and ECASS [European Cooperative Acute Stroke Study] II), we simplified the Stroke-TPI by reducing variables and interaction terms and by exploring simpler (3- and 8-item) stroke severity scores. External validation was performed in the ECASS III trial (n = 821).

RESULTS

The following 6 variables were most predictive of good outcomes: age, systolic blood pressure, diabetes, stroke severity, symptom onset to treatment time, and rtPA therapy. Treatment effect modifiers included onset to treatment time and systolic blood pressure. For the models predicting a bad outcome (modified Rankin Scale [mRS] score ≥5), significant variables included age, stroke severity, and serum glucose. rtPA therapy did not change the risk of a poor outcome. Compared with models using the full NIH Stroke Scale, models using the 3-item severity score showed similar discrimination and excellent calibration. External validation on ECASS III showed similar performance (C statistics 0.75 [mRS score ≤1] and 0.80 [mRS score ≤2]).

CONCLUSION

A simpler model using a 3-item stroke severity score, instead of the 15-item NIH Stroke Scale, has similar prognostic value and may be easier to use in routine care. Future studies are needed to test whether it can improve process and clinical outcomes.

The RoPE Score and Right-to-Left Shunt Severity by Transcranial Doppler in the CODICIA Study

BACKGROUND

For patients with cryptogenic stroke (CS) and patent foramen ovale (PFO), it is unknown whether the magnitude of right-to-left shunt (RLSh) measured by contrast transcranial Doppler (c-TCD) is correlated with the likelihood an identified PFO is related to CS as determined by the Risk of Paradoxical Embolism (RoPE) score. Additionally, for patients with CS, it is unknown whether PFO assessment by c-TCD is more sensitive for identifying RLSh compared with transesophageal echocardiography (TEE). Our aim was to determine the significance of RLSh grade by c-TCD in patients with PFO and CS.

METHODS

We evaluated patients with CS who had RLSh quantified by c-TCD in the Multicenter Study into RLSh in Cryptogenic Stroke (CODICIA) to determine whether there is an association between c-TCD shunt grade and the RoPE Score. For patients who underwent c-TCD and TEE, we determined whether there is agreement in identifying and grading RLSh between these two modalities.

RESULTS

The RoPE score predicted the presence versus the absence of RLSh documented by c-TCD (c-statistic = 0.66). For patients with documented RLSh by c-TCD, shunt severity was correlated with increasing RoPE score (rank correlation (r) = 0.15, p = 0.01). Among 293 patients who had both c-TCD and TEE performed, c-TCD was more sensitive (98.7%) for detecting RLSh. Of the 97 patients with no PFO identified on TEE, 28 (29%) had a large amount of RLSh seen on c-TCD.

CONCLUSIONS

For patients with CS, severity of RLSh by c-TCD is positively correlated with the RoPE score, indicating that this technique for shunt grading identifies patients more likely to have pathogenic rather than incidental PFOs. c-TCD is also more sensitive in detecting RLSh than TEE. These findings suggest an important role for c-TCD in the evaluation of PFO in the setting of CS.

Clinical Prediction Models for Cardiovascular Disease: Tufts Predictive Analytics and Comparative Effectiveness Clinical Prediction Model Database

BACKGROUND

Clinical prediction models (CPMs) estimate the probability of clinical outcomes and hold the potential to improve decision making and individualize care. For patients with cardiovascular disease, there are numerous CPMs available although the extent of this literature is not well described.

METHODS AND RESULTS

We conducted a systematic review for articles containing CPMs for cardiovascular disease published between January 1990 and May 2012. Cardiovascular disease includes coronary heart disease, heart failure, arrhythmias, stroke, venous thromboembolism, and peripheral vascular disease. We created a novel database and characterized CPMs based on the stage of development, population under study, performance, covariates, and predicted outcomes. There are 796 models included in this database. The number of CPMs published each year is increasing steadily over time. Seven hundred seventeen (90%) are de novo CPMs, 21 (3%) are CPM recalibrations, and 58 (7%) are CPM adaptations. This database contains CPMs for 31 index conditions, including 215 CPMs for patients with coronary artery disease, 168 CPMs for population samples, and 79 models for patients with heart failure. There are 77 distinct index/outcome pairings. Of the de novo models in this database, 450 (63%) report a c-statistic and 259 (36%) report some information on calibration.

CONCLUSIONS

There is an abundance of CPMs available for a wide assortment of cardiovascular disease conditions, with substantial redundancy in the literature. The comparative performance of these models, the consistency of effects and risk estimates across models and the actual and potential clinical impact of this body of literature is poorly understood.

Myocardial Infarction, Stroke, and Mortality in cART-Treated HIV Patients on Statins

Despite combination antiretroviral therapy (cART), people living with HIV (PLWH) continue to have more systemic inflammation and metabolic disturbances than the general population. These risk factors for atherosclerosis and organ dysfunction may be ameliorated by statins. We retrospectively analyzed 438 cART treated PLWH from the Nutrition For Healthy Living (NFHL) cohort to determine the association between statins and myocardial infarction (MI), stroke, and all-cause mortality as a composite. We used Cox proportional hazards regression as our main analysis. The average age was 44 years, 32% were women, and 67 of the 438 subjects used statins. There was no association between statins and our composite endpoint in two separate models [1.26 (0.57-2.79) in statin history model and 0.93 (0.65-1.32) per year in statin duration model]. The composite outcome was significantly associated with CD4 count, age, and smoking status in both models. CD4 count remained significant even after exclusion of mortality from the composite (HR=0.88, p=0.02). Confounding control via propensity scoring and multiple imputations did not change the results. Statins did not have an effect on MI, stroke, and mortality. Interestingly, CD4 count appears to be an important predictor of these outcomes, even after exclusion of death from the composite.

Abstract 113: Predicting Short and Medium Term Hospital Readmission Following Heart Valve Surgery

Background: Although models exist for predicting hospital readmission following coronary artery bypass surgery, no similar models exist for predicting hospital readmission following heart valve surgery (HVS). Given Medicare’s financial penalties for 30-day hospital readmission and the ongoing implementation of 90-day bundled payments for cardiac surgical procedures, we developed a clinical predictive model to estimate both short and medium-term readmission for patients with HVS.

Methods: Utilizing a geographically and structurally diverse sample of US hospitals (Premier Inpatient Database), we examined patient characteristics, hospital characteristics, and treatment factors predictive of short and medium term hospital readmission following HVS. The study period was from January 2007 to June 2011. A generalized estimating equation model accounted for clustering within hospitals. We set aside a random 20% of hospitals during model derivation for later use as an internal validation cohort.

Results: At 219 hospitals, we identified 38,532 patients with HVS (73 years old, 56% male, 71% non-Hispanic White, with 61% aortic valve, 29% mitral valve, and 10% combination valve surgery.) A total of 3084 (8%) and 4943 (12.9%) patients were readmitted within 1 month and 3 months, respectively, with the most common reasons for readmission being heart failure, atrial fibrillation, and pleural/pericardial effusions. Among the >60 factors examined, 18 independent factors were retained in the final model for predicting 3-month readmission (see table.) The strongest predictors of 3-month readmission were type of valve surgery, urgency of hospital admission, hospital length of stay, and discharge disposition. The final model had fair discrimination (C-statistic 0.667, p < 0.001) and good calibration (predicted vs. observed differences of <1-2% across all deciles of predicted readmission risk.)

Conclusions: We described key factors that predict 3-month hospital readmission following HVS. This model should enable clinicians to identify individuals with HVS who are at increased risk of hospital readmission and who are most likely to benefit from improved coordination of post-discharge care and follow-up.

Abstract 164: A Field Synopsis of Gender Effects in Clinical Prediction Models for Cardiovascular Disease

Introduction: Gender differences in incidence, prognosis and treatment response have been observed across the spectrum of cardiovascular diseases (CVD). However, despite several decades of investigation, consistent findings regarding the magnitude and directionality of gender differences in CVD are elusive. We therefore conducted the first field synopsis of the role of gender on CVD conditions using a registry of clinical prediction models (CPMs).

Methods: The Tufts PACE Center (CPM) Registry is based on a systematic review of cardiovascular CPMs published in English-language articles from 1/1990-5/2012. All included CPMs permit calculation of outcome probabilities from information provided in an equation, point score or nomogram. For the 15 most common unique index condition-outcome pair models, we calculated the proportion of models that included coefficients for the effect of gender on CVD incidence or prognosis, or presented gender-stratified models. The sample size, age distribution and proportion of females in the model development cohorts were summarized.

Results: Out of 579 CPMs with CVD as either an index condition or outcome, 169 (29%) contained a coefficient for gender and 33 (6%) presented gender-stratified models. Gender was more frequently included as a covariate or stratification variable in models predicting incident CVD versus prognosis for patients with known CVD. Gender was included in 60/74 (81%) models predicting morbidity and/or mortality among a population sample, yet in only 9/53 (17%) of models predicting morbidity and/or mortality among patients with stroke, and 9/53 (17%) of models predicting mortality among patients with congestive heart failure. Gender was more likely to be included in CPMs developed from cohorts with larger sample sizes (150/299 cohorts with n≥2000 versus 54/277 cohorts with n<2000, p<0.001). For each 10% increase in the proportion of women in the model development cohort, there was a 22% increased odds of including gender in the CPM (OR = 1.22, 95% CI 1.08-1.39, p=0.002).

Conclusions: Gender is an important prognostic factor in CVD, but is only included in about one third of published CPMs. Gender is much more frequently included as a predictor of incident CVD among the disease-free than of prognosis in those with established CVD.

Abstract 174: Clinical Prediction Models for Cardiovascular Disease: The Tufts PACE CPM Database

Background: Clinical prediction models (CPMs) estimate the probability of clinical outcomes and hold the potential to improve decision making and individualize care. For patients with cardiovascular disease (CVD) there are numerous CPMs available though the extent of this literature is not well described.

Methods and Results: We conducted a systematic review for articles containing CPMs for CVD published between January 1990 through May 2012. CVD includes coronary artery disease (CAD), congestive heart failure (CHF), arrhythmias, stroke, venous thromboembolism (VTE) and peripheral vascular disease (PVD). We created a novel database and characterized CPMs based on the stage of development, population under study, performance, covariates, and predicted outcomes. We included articles that describe newly developed CPMs that predict the risk of developing an outcome (prognostic models) or the probability of a specific diagnosis (diagnostic models). There are 796 models included in this database representing 31 distinct index conditions. 717 (90%) are de novo CPMs, 21 (3%) are CPM recalibrations, and 58 (7%) are CPM adaptations. There are 215 CPMs for patients with CAD, 168 CPMs for population samples at risk for incident CVD, and 79 models for patients with CHF (Figure). De novo CPMs predicting mortality were most commonly published for patients with known CAD (98 models) followed by HF (63 models) and stroke (24 models). There are 77 distinct index/ outcome (I/O) pairings and models are roughly evenly split between those predicting short term outcomes (< 3 months) and those predicting long term outcomes (< 6 months). There are 41 diagnostic CPMs included in this database, most commonly predicting diagnoses of CAD (11 models), VTE (10 models), and acute coronary syndrome (5 models). Of the de novo models in this database 450 (63%) report a c-statistic and 259 (36%) report either the Hosmer-Lemeshow statistic or show a calibration plot.

Conclusions: There is an abundance of CPMs available for many CVD conditions, with substantial redundancy in the literature. The comparative performance of these models, the consistency of effects and risk estimates across models and the actual and potential clinical impact of this body of literature is poorly understood.

Prevention of recurrent stroke in patients with patent foramen ovale

Patent foramen ovale (PFO) is common and only rarely related to stroke. The high PFO prevalence in healthy individuals makes for difficult decision making when a PFO is found in the setting of a cryptogenic stroke, because the PFO may be an incidental finding. Recent clinical trials of device-based PFO closure have had negative overall summary results; these trials have been limited by low recurrence rates. The optimal antithrombotic strategy for these patients is also unknown. Recent work has identified a risk score that estimates PFO-attributable fractions based on individual patient characteristics, although whether this score can help direct therapy is unclear.

Improving diabetes prevention with benefit based tailored treatment: risk based reanalysis of Diabetes Prevention Program

OBJECTIVE

To determine whether some participants in the Diabetes Prevention Program were more or less likely to benefit from metformin or a structured lifestyle modification program.

DESIGN

Post hoc analysis of the Diabetes Prevention Program, a randomized controlled trial.

SETTING

Ambulatory care patients.

PARTICIPANTS

3060 people without diabetes but with evidence of impaired glucose metabolism.

INTERVENTION

Intervention groups received metformin or a lifestyle modification program with the goals of weight loss and physical activity.

MAIN OUTCOME MEASURE

Development of diabetes, stratified by the risk of developing diabetes according to a diabetes risk prediction model.

RESULTS

Of the 3081 participants with impaired glucose metabolism at baseline, 655 (21%) progressed to diabetes over a median 2.8 years' follow-up. The diabetes risk model had good discrimination (C statistic=0.73) and calibration. Although the lifestyle intervention provided a sixfold greater absolute risk reduction in the highest risk quarter than in the lowest risk quarter, patients in the lowest risk quarter still received substantial benefit (three year absolute risk reduction 4.9% v 28.3% in highest risk quarter; numbers needed to treat of 20.4 and 3.5, respectively). The benefit of metformin, however, was seen almost entirely in patients in the top quarter of risk of diabetes. No benefit was seen in the lowest risk quarter. Participants in the highest risk quarter averaged a 21.4% three year absolute risk reduction (number needed to treat 4.6).

CONCLUSIONS

Patients at high risk of diabetes have substantial variation in their likelihood of receiving benefit from diabetes prevention treatments. Using this knowledge could decrease overtreatment and make prevention of diabetes far more efficient, effective, and patient centered, provided that decision making is based on an accurate risk prediction tool.

Abstract W MP74: Development And Validation Of A Modified Stroke Thrombolytic Predictive Instrument To Improve Implementation And Ease Of Use

Introduction: The Stroke Thombolytic Predictive Instrument (Stroke TPI), derived on data from the first 5 randomized clinical trials of standard dose rtPA, predicts the probability of good and bad outcomes with and without rtPA. We sought to rebuild and externally validate a simpler Stroke TPI to support implementation in routine clinical care. Methods: Using a derivation cohort of 1,983 patients from a combined database of randomized clinical trials (NINDS 1 and 2; ATLANTIS A and B; ECASS 2), we simplified the Stroke TPI by: 1) removing lower credibility interaction terms; and 2) exploring reduced stroke severity scores, including previously developed 8-item and 3-item measures. Additionally, we included alternative thresholds of a good 90-day functional outcome (i.e. prediction of both modified Rankin Score [mRS] ≤ 1 and mRS ≤ 2). Bootstrapping methods were used for internal validation. External validation was performed on the ECASS 3 trial (n=821). Results: The following 6 variables were included to predict good outcomes: age, systolic blood pressure (SBP), diabetes, stroke severity, symptom onset to treatment time (OTT) and rtPA therapy. Treatment effect modifiers (interaction terms) included OTT and SBP. For the models predicting a bad outcome (mRS ≥ 5), significant variables included: age, stroke severity, and serum glucose. rtPA therapy did not change the risk of a poor outcome. As compared to models using the full NIHSS, models employing the 3-item severity score showed similar discrimination, with c-statistics on bootstrap (internal) validation of: 0.76 (mRS ≤ 1); 0.78 (mRS ≤ 2); and 0.76 (mRS ≥ 5), and with excellent calibration. External validation on ECASS 3 showed similar performance (c-statistics 0.75 [mRS ≤ 1] and 0.80 [mRS ≤ 2]), with good calibration. Conclusion: A simpler prediction model using a 3-item, instead of the 15-item NIHSS stroke severity score, has similar prognostic value and may be easier to use in routine care. Future studies are needed to test whether the reduced model can improve treatment rates, time to treatment, and outcomes.

Abstract W P177: The Frequency and Directionality of the Effect of Sex in Prediction Models for Stroke

Introduction: There are apparent sex differences in stroke, with women having a higher lifetime risk and worse outcomes. However, there remain critical gaps in our understanding of sex differences in risk, treatment response and outcomes following stroke. We conducted the first systematic summary of the role of sex on stroke-related conditions using a registry of clinical prediction models (CPMs).

Methods: The Tufts PACE CPM Registry is based on a systematic review of cerebrovascular and cardiovascular CPMs published in English-language articles from 1/1990-5/2012, and includes 585 unique CPMs extracted from 506 articles. All included CPMs permit calculation of outcome probabilities from information provided in the form of an equation, point score or nomogram. We calculated the proportion of models with coefficients for the effect of sex on stroke incidence or prognosis, and summarized the directionality (harmful vs. protective) of the coefficients for sex.

Results: Out of 75 CPMs with stroke as either an index condition or outcome, 23 (31%) contained a coefficient for sex or presented sex-stratified models. Only 8/48 (17%) models of stroke prognosis included sex or presented sex-specific models, as compared to 14/24 (58%) of models predicting stroke incidence. In models categorized by unique index-outcome pairs, sex was most commonly included in models predicting stroke among a general population (67%). Female sex was associated with reduced risk of mortality after ischemic stroke and a higher risk of stroke from arrhythmias or CABG/PCI. In a general population, women typically had a lower risk of stroke, although stratified models suggest this depends on the presence or absence of other risk factors.

Conclusions: Sex is an important prognostic factor in CVD but is inconsistently included in stroke CPMs. Sex is more frequently included in models of stroke incidence than models of prognosis. Being female seems protective for some outcomes, but harmful for others.

Comparative effectiveness of pharmacologic interventions for knee osteoarthritis: a systematic review and network meta-analysis

BACKGROUND

The relative efficacy of available treatments of knee osteoarthritis (OA) must be determined for rational treatment algorithms to be formulated.

PURPOSE

To examine the efficacy of treatments of primary knee OA using a network meta-analysis design, which estimates relative effects of all treatments against each other.

DATA SOURCES

MEDLINE, EMBASE, Web of Science, Google Scholar, Cochrane Central Register of Controlled Trials from inception through 15 August 2014, and unpublished data.

STUDY SELECTION

Randomized trials of adults with knee OA comparing 2 or more of the following: acetaminophen, diclofenac, ibuprofen, naproxen, celecoxib, intra-articular (IA) corticosteroids, IA hyaluronic acid, oral placebo, and IA placebo.

DATA EXTRACTION

Two reviewers independently abstracted study data and assessed study quality. Standardized mean differences were calculated for pain, function, and stiffness at 3-month follow-up.

DATA SYNTHESIS

Network meta-analysis was performed using a Bayesian random-effects model; 137 studies comprising 33,243 participants were identified. For pain, all interventions significantly outperformed oral placebo, with effect sizes from 0.63 (95% credible interval [CrI], 0.39 to 0.88) for the most efficacious treatment (hyaluronic acid) to 0.18 (CrI, 0.04 to 0.33) for the least efficacious treatment (acetaminophen). For function, all interventions except IA corticosteroids were significantly superior to oral placebo. For stiffness, most of the treatments did not significantly differ from one another.

LIMITATION

Lack of long-term data, inadequate reporting of safety data, possible publication bias, and few head-to-head comparisons.

CONCLUSION

This method allowed comparison of common treatments of knee OA according to their relative efficacy. Intra-articular treatments were superior to nonsteroidal anti-inflammatory drugs, possibly because of the integrated IA placebo effect. Small but robust differences were observed between active treatments. All treatments except acetaminophen showed clinically significant improvement from baseline pain. This information, along with the safety profiles and relative costs of included treatments, will be helpful for individualized patient care decisions.

PRIMARY FUNDING SOURCE

Agency for Healthcare Research and Quality.

Controversies in cardioembolic stroke

OPINION STATEMENT

Cardioembolic (CE) stroke mechanisms account for a significant number of ischemic strokes; however, the true burden is likely underestimated. It is critically important to identify patients with CE strokes because these individuals have high recurrence rates and represent a subgroup of patients who may benefit from targeted therapy in the form of anticoagulation or device based treatments. Current guidelines offer recommendations for diagnosis and treatment of these patients; however, important questions remain. First, appropriate cardiac testing in the setting of CE must be individualized and the optimal duration of electrocardiographic monitoring to rule out atrial fibrillation (AF) is unclear. Second, risk stratification tools for AF remain understudied, and there is controversy about which anticoagulant agents are most appropriate. Lastly, important potential CE sources of stroke such as patent foramen ovale have garnered significant attention recently, and debate regarding how to manage these patients persists. In this review, we discuss some of the important controversies in diagnosing and treating patients with possible CE stroke, pointing to areas where future research might be particularly valuable.

Multiple chronic conditions in type 2 diabetes mellitus: prevalence and consequences

OBJECTIVES

Multiple chronic comorbidities (MCCs) are an issue of growing significance in diabetes because they are highly prevalent and can increase disease burden and costs. We examined MCC patterns among patients with type 2 diabetes mellitus and identified specific comorbidity clusters associated with poor patient outcomes.

STUDY DESIGN AND METHODS

We conducted a cross-sectional analysis of 161,174 patients with diabetes using electronic health record data supplied by US providers in the 2008 to 2012 Humedica data sets. We examined prevalence of MCC clusters in younger and older patients. For each of the 15 most common MCC clusters, we reported predicted probabilities for diabetes face-to-face visits, reaching glycated hemoglobin < 8%; emergency department (ED) visits; and 30-day hospital readmissions, based on logistic regression results.

RESULTS

The leading MCC combination was the presence of hypertension-hyperlipidemia-obesity and no other diagnosed comorbidities (19% of the sample). The most notable difference, by age, was a higher prevalence of obesity in the younger cohort. MCC clusters were more diverse among the older population: the top 10 MCC clusters accounted for 66% of older patients, compared with 78% of younger patients. Patients with certain comorbidity profiles, such as those with obesity only, were less likely to have diabetes-related face-to-face visits and to meet A1C treatment goals, and more likely to have ED visits and 30-day readmissions.

CONCLUSIONS

Patients with diabetes have substantial comorbidities, but the patterns vary considerably across patients and by age. Diabetes care remained suboptimal among many types of MCC patients, and patient outcomes varied by MCC profile. Specific management strategies should be developed for common MCC clusters, such as hypertension-hyperlipidemia-obesity.

Determinants of antithrombotic choice for patent foramen ovale in cryptogenic stroke

OBJECTIVE

We examined the influence of clinical, radiologic, and echocardiographic characteristics on antithrombotic choice in patients with cryptogenic stroke (CS) and patent foramen ovale (PFO), hypothesizing that features suggestive of paradoxical embolism might lead to greater use of anticoagulation.

METHODS

The Risk of Paradoxical Embolism Study combined 12 databases to create the largest dataset of patients with CS and known PFO status. We used generalized linear mixed models with a random effect of component study to explore whether anticoagulation was preferentially selected based on the following: (1) younger age and absence of vascular risk factors, (2) "high-risk" echocardiographic features, and (3) neuroradiologic findings.

RESULTS

A total of 1,132 patients with CS and PFO treated with anticoagulation or antiplatelets were included. Overall, 438 participants (39%) were treated with anticoagulation with a range (by database) of 22% to 54%. Treatment choice was not influenced by age or vascular risk factors. However, neuroradiologic findings (superficial or multiple infarcts) and high-risk echocardiographic features (large shunts, shunt at rest, and septal hypermobility) were predictors of anticoagulation use.

CONCLUSION

Both antithrombotic regimens are widely used for secondary stroke prevention in patients with CS and PFO. Radiologic and echocardiographic features were strongly associated with treatment choice, whereas conventional vascular risk factors were not. Prior observational studies are likely to be biased by confounding by indication.