Evaluation of propensity score used in cardiovascular research: a cross-sectional survey and guidance document

Additive Prognostic Value of Left Ventricular Dispersion and Deformation in Patients With Severe Aortic Stenosis

BACKGROUND

Speckle tracking strain echocardiography allows one to visualize the timing of maximum regional strain and quantifies left ventricular-mechanical dispersion (LV-MD). Whether LV-MD and LV-global longitudinal strain (LV-GLS) provide similar or complementary information in mortality risk stratification in patients with severe aortic stenosis (SAS) remains unknown.

OBJECTIVES

The authors hypothesized that LV mechanical dyssynchrony assessed by LV-MD is associated with an increased risk of mortality and provides additional prognostic information on top of LV-GLS in patients with SAS.

METHODS

A total of 364 patients with SAS (aortic valve area indexed ≤0.6 cm2/m2 and/or aortic valve area ≤1 cm2), LV ejection fraction ≥50% and no or mild symptoms were enrolled. The endpoint was overall mortality.

RESULTS

During a median follow-up period of 41 months, 149 patients died. After adjustment, LV-MD ≥68 ms was significantly associated with an increased risk of mortality (adjusted HR: 1.41; 95% CI: 1.01-1.96; P = 0.044). Adding LV-MD ≥68 ms to a multivariable Cox regression model including LV-GLS ≥-15% improved predictive performance in terms of overall mortality, with improved global model fit, reclassification, and better discrimination. Patients with both criteria had an important increase in mortality compared to patients with none or one criterion (adjusted HR: 2.02; 95% CI: 1.34-3.03; P = 0.001). Interobserver reproducibility of LV-MD was good with an intraclass correlation coefficient of 0.90 (95% CI: 0.72-0.97).

CONCLUSIONS

LV-MD is a reproducible parameter independently associated with an increased risk of mortality in SAS. Increased LV-MD associated with depressed LV-GLS identifies a subgroup of patients with an increased mortality risk. Whether early aortic valve replacement improves the outcome of these patients deserves further studies.

Visualizing the target estimand in comparative effectiveness studies with multiple treatments

Aim: Comparative effectiveness research using real-world data often involves pairwise propensity score matching to adjust for confounding bias. We show that corresponding treatment effect estimates may have limited external validity, and propose two visualization tools to clarify the target estimand. Materials & methods: We conduct a simulation study to demonstrate, with bivariate ellipses and joy plots, that differences in covariate distributions across treatment groups may affect the external validity of treatment effect estimates. We showcase how these visualization tools can facilitate the interpretation of target estimands in a case study comparing the effectiveness of teriflunomide (TERI), dimethyl fumarate (DMF) and natalizumab (NAT) on manual dexterity in patients with multiple sclerosis. Results: In the simulation study, estimates of the treatment effect greatly differed depending on the target population. For example, when comparing treatment B with C, the estimated treatment effect (and respective standard error) varied from -0.27 (0.03) to -0.37 (0.04) in the type of patients initially receiving treatment B and C, respectively. Visualization of the matched samples revealed that covariate distributions vary for each comparison and cannot be used to target one common treatment effect for the three treatment comparisons. In the case study, the bivariate distribution of age and disease duration varied across the population of patients receiving TERI, DMF or NAT. Although results suggest that DMF and NAT improve manual dexterity at 1 year compared with TERI, the effectiveness of DMF versus NAT differs depending on which target estimand is used. Conclusion: Visualization tools may help to clarify the target population in comparative effectiveness studies and resolve ambiguity about the interpretation of estimated treatment effects.

The risk of revision is higher following shoulder hemiarthroplasty compared with total shoulder arthroplasty for osteoarthritis: a matched cohort study of 11,556 patients from the National Joint Registry, UK

BACKGROUND AND PURPOSE

Total shoulder arthroplasty (TSA) and hemiarthroplasty (HA) are used in the management of osteoarthritis of the glenohumeral joint. We aimed to determine whether TSA or HA resulted in a lower risk of adverse outcomes in patients of all ages with osteoarthritis and an intact rotator cuff and in a subgroup of patients aged 60 years or younger.

PATIENTS AND METHODS

Shoulder arthroplasties recorded in the National Joint Registry, UK, between April 1, 2012 and June 30, 2021, were linked to Hospital Episode Statistics in England. Elective TSAs and HAs were matched on propensity scores based on 11 variables. The primary outcome was all-cause revision. Secondary outcomes were combined revision/non-revision reoperations, 30-day inpatient complications, 1-year mortality, and length of stay. 95% confidence intervals (CI) were reported.

RESULTS

11,556 shoulder arthroplasties were included: 7,641 TSAs, 3,915 HAs. At 8 years 95% (CI 94-96) of TSAs and 91% (CI 90-92) of HAs remained unrevised. The hazard ratio (HR) varied across follow-up: 4-year HR 2.7 (CI 1.9-3.5), 8-year HR 2.0 (CI 0.5-3.5). Rotator cuff insufficiency was the most common revision indication. In patients aged 60 years or younger prosthesis survival at 8 years was 92% (CI 89-94) following TSA and 84% (CI 80-87) following HA.

CONCLUSION

The risk of revision was higher following HA in patients with osteoarthritis and an intact rotator cuff. Patients aged 60 years and younger had a higher risk of revision following HA.

Development and validation of multicentre study on novel Artificial Intelligence-based Cardiovascular Risk Score (AICVD)

OBJECTIVE

Cardiovascular diseases (CVD) are one of the most prevalent diseases in India amounting for nearly 30% of total deaths. A dearth of research on CVD risk scores in Indian population, limited performance of conventional risk scores and inability to reproduce the initial accuracies in randomised clinical trials has led to this study on large-scale patient data. The objective is to develop an Artificial Intelligence-based Risk Score (AICVD) to predict CVD event (eg, acute myocardial infarction/acute coronary syndrome) in the next 10 years and compare the model with the Framingham Heart Risk Score (FHRS) and QRisk3.

METHODS

Our study included 31 599 participants aged 18-91 years from 2009 to 2018 in six Apollo Hospitals in India. A multistep risk factors selection process using Spearman correlation coefficient and propensity score matching yielded 21 risk factors. A deep learning hazards model was built on risk factors to predict event occurrence (classification) and time to event (hazards model) using multilayered neural network. Further, the model was validated with independent retrospective cohorts of participants from India and the Netherlands and compared with FHRS and QRisk3.

RESULTS

The deep learning hazards model had a good performance (area under the curve (AUC) 0.853). Validation and comparative results showed AUCs between 0.84 and 0.92 with better positive likelihood ratio (AICVD -6.16 to FHRS -2.24 and QRisk3 -1.16) and accuracy (AICVD -80.15% to FHRS 59.71% and QRisk3 51.57%). In the Netherlands cohort, AICVD also outperformed the Framingham Heart Risk Model (AUC -0.737 vs 0.707).

CONCLUSIONS

This study concludes that the novel AI-based CVD Risk Score has a higher predictive performance for cardiac events than conventional risk scores in Indian population.

TRIAL REGISTRATION NUMBER

CTRI/2019/07/020471.

Cardiometabolic sex differences in adults born small for gestational age

Aim

This study aimed to assess the cardiometabolic sex similarities and differences in adults born small for gestational age.

Methods

This study was an ambispective cohort study from a birth registry in Barcelona, Spain, including 523 adult participants (20-40 years-old) subdivided as born small for gestational age (SGA, if birth weight <10th centile) or adequate fetal growth for gestational age (AGA). Cardiometabolic health was assessed by echocardiography, electrocardiogram, blood pressure measurement, vascular ultrasound, anthropometric measurements, and serum glycemia and lipid profile. Stratified analyses by sex were performed by estimation of adjusted absolute difference (AAD) using inverse probability weighting.

Results

Compared with AGA, the stratified analyses by sex showed a more pronounced reduction in ejection fraction [AAD: female -1.73 (95% CI -3.2 to -0.28) vs. male -1.33 (-3.19 to 0.52)] and increment in heart rate [female 3.04 (0.29-5.8) vs. male 2.25 (-0.82 to 5.31)] in SGA females compared with SGA males. In contrast, a more pronounced reduction in PR interval [female -1.36 (-6.15 to 3.42) vs. male -6.61 (-11.67 to -1.54)] and an increase in systolic blood pressure [female 0.06 (-2.7 to 2.81) vs. male 2.71 (-0.48 to 5.9)] and central-to-peripheral fat ratio [female 0.05 (-0.03 to 0.12) vs. male 0.40 (0.17-0.62)] were mainly observed in SGA male compared with SGA female.

Conclusions

Sex differences were observed in the effect of SGA on cardiometabolic endpoints with female being more prone to cardiac dysfunction and male to electrocardiographic, vascular, and metabolic changes. Future research including sex-stratification data is warranted.

Reevaluating the protective effect of smoking on preeclampsia risk through the lens of bias

Preeclampsia is a hypertensive disorder that is usually diagnosed after 20 weeks' gestation. Despite the deleterious effect of smoking on cardiovascular disease, it has been frequently reported that smoking has a protective effect on preeclampsia risk and biological explanations have been proposed. However, in this manuscript, we present multiple sources of bias that could explain this association. First, key concepts in epidemiology are reviewed: confounder, collider, and mediator. Then, we describe how eligibility criteria, losses of women potentially at risk, misclassification, or performing incorrect adjustments can create bias. We provide examples to show that strategies to control for confounders may fail when they are applied to variables that are not confounders. Finally, we outline potential approaches to manage this controversial effect. We conclude that there is probably no single epidemiological explanation for this counterintuitive association.

Recommendations for the use of propensity score methods in multiple sclerosis research

BACKGROUND

With many disease-modifying therapies currently approved for the management of multiple sclerosis, there is a growing need to evaluate the comparative effectiveness and safety of those therapies from real-world data sources. Propensity score methods have recently gained popularity in multiple sclerosis research to generate real-world evidence. Recent evidence suggests, however, that the conduct and reporting of propensity score analyses are often suboptimal in multiple sclerosis studies.

OBJECTIVES

To provide practical guidance to clinicians and researchers on the use of propensity score methods within the context of multiple sclerosis research.

METHODS

We summarize recommendations on the use of propensity score matching and weighting based on the current methodological literature, and provide examples of good practice.

RESULTS

Step-by-step recommendations are presented, starting with covariate selection and propensity score estimation, followed by guidance on the assessment of covariate balance and implementation of propensity score matching and weighting. Finally, we focus on treatment effect estimation and sensitivity analyses.

CONCLUSION

This comprehensive set of recommendations highlights key elements that require careful attention when using propensity score methods.

Tranexamic Acid Was Not Associated with Increased Complications in High-Risk Patients with Intertrochanteric Fracture

BACKGROUND

For elective total joint arthroplasty, tranexamic acid (TXA) is considered safe and efficacious. However, evidence of TXA's safety in high-risk patients undergoing nonelective surgery for hip fracture is sparse. This study aimed to assess whether TXA administration to high-risk patients with an intertrochanteric (IT) hip fracture increased the risk of thromboembolic complications or mortality.

METHODS

All patients treated surgically for IT hip fracture between 2015 and 2019 across 4 hospitals of a single hospital system were considered. High- versus low-risk patients and those receiving TXA versus no TXA treatment were identified. Propensity scores adjusted for risk differences between patient groups with TXA and no TXA administration were calculated for (1) high-risk patients (n = 141) and (2) the entire population (n = 316). Postoperative mortality, deep venous thrombosis (DVT), pulmonary embolism (PE), myocardial infarction (MI), and stroke within 90 days of surgery were evaluated.

RESULTS

No association between TXA administration and increased risk of mortality or complications in either group was identified. Specifically, out of 282 matched high-risk patients, no differences in mortality (odds ratio [OR], 0.97 [95% confidence interval (CI), 0.90, 1.05]), DVT (OR, 0.97 [95% CI, 0.93, 1.00]), PE (OR 1.00 [95% CI, 0.95, 1.05]), MI (OR, 1.04 [95% CI, 0.98, 1.10]), or stroke (OR, 1.00 [95% CI, 0.95, 1.05]) were identified.

CONCLUSIONS

In our review of propensity-matched high-risk patients undergoing surgical repair for IT fracture, we found that TXA administration compared with no TXA administration was not associated with an increased risk of mortality, DVT, PE, MI, or stroke within 90 days of surgery.

LEVEL OF EVIDENCE

Therapeutic Level IV . See Instructions for Authors for a complete description of levels of evidence.

Development and Validation of a Nomogram for Predicting Albumin Transfusion After Spinal Tuberculosis Surgery: Based on Propensity Score Matching Analysis

BACKGROUND

There have been few literature reports on the use of perioperative parameters to predict the risk of albumin transfusion after spinal tuberculosis surgery based on the application of nomogram and propensity score matching (PSM) analysis.

OBJECTIVE

The purpose was to predict the risk of albumin transfusion after spinal tuberculosis surgery based on a combination of PSM and nomogram.

METHODS

The clinical data of the patients were collected in our hospital, including preoperative clinical data, preoperative laboratory tests, and postoperative clinical data. All data were divided into 2 groups, including the albumin transfusion group and the non-albumin transfusion group. The PSM analysis was used to adjust the baseline data of the 2 groups. The nomogram was further constructed. The practicability and predictive ability of the model were evaluated.

RESULTS

A total of 494 cases were collected in this article; 102 pairs by PSM analysis were used to construct the nomogram. There were statistical differences in surgical approach, aspartate aminotransferase/alanine aminotransferase levels, drainage, and kyphosis by logistic analysis, and these parameters were included in the construction of the nomogram. The C-index of the prediction model was 0.734. The area under the curve was 0.73 and the net benefit was between 0.13 and 0.99. The calculated C-index was 0.71 by the internal verification method.

CONCLUSIONS

The PSM analysis had a good matching effect and the nomogram had a good predictive ability. Surgical approach, aspartate aminotransferase/alanine aminotransferase levels, drainage, and kyphosis might be predictors of albumin transfusion after spinal tuberculosis surgery.

Emulating a Randomised Controlled Trial With Observational Data: An Introduction to the Target Trial Framework

Randomised controlled trials (RCTs) are often considered to be the highest quality of evidence owing to the absence of baseline confounding, the simplicity of analyses, and direct estimation of causal effects. However, observational studies can be designed to mimic RCTs and estimate causal treatment effects. In this review, we describe the target trial framework to illustrate how observational studies can successfully emulate RCTs. We focus on key design elements of RCTs and how to emulate them with observational data. These elements include 1) eligibility criteria, 2) treatment assignment and randomisation, 3) specification of "time zero", 4) outcomes, 5) follow-up, 6) causal contrasts (intention-to-treat vs per-protocol), and 7) statistical analyses. In addition, we describe the design of an example target trial created to emulate the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition With Prasugrel Thrombolysis in Myocardial Infarction (TRITON-TIMI) 38 trial and compare effect estimates. Overall, careful design of a target trial using observational data can produce causal effect estimates that are often comparable to RCTs.