Propensity score trimming mitigates bias due to covariate measurement error in inverse probability of treatment weighted analyses: A plasmode simulation

Long-term efficacy and safety of tailored immunosuppressive therapy in immune-mediated biopsy-proven myocarditis: A propensity-weighted study

AIMS

Standardized immunosuppressive therapy (IS) had been previously investigated in biopsy-proven (BP) lymphocytic myocarditis with heart failure (HF). This study evaluated efficacy and safety of tailored IS in BP immune-mediated myocarditis, irrespective of histology and clinical presentation.

METHODS AND RESULTS

Consecutive BP myocarditis patients treated with long-term tailored IS on top of optimal medical therapy (OMT), were compared with OMT non-IS controls using propensity-score weighting. The primary outcome was a composite of death or heart transplant, the secondary outcome was a composite of biventricular function, New York Heart Association (NYHA) class variation, and relapse. IS was managed by a multidisciplinary Cardioimmunology Team, involved a safety checklist and active patients' education. Ninety-one IS patients were compared with 267 non-IS patients. IS patients more frequently had systemic immune-mediated diseases (35% vs. 9.7%), lower baseline echocardiographic left ventricular ejection fraction (35% vs. 43%), lower right ventricular fractional area change (34% vs. 41%) and higher frequency of active lymphocytic, eosinophilic and giant cell myocarditis (71% vs. 58%, 12% vs. 1.1%, and 6.6% vs. 1.5%, respectively). At 5-year follow up, no difference was observed in the primary outcome (survival rate 93% in IS vs. 87% in non-IS), but IS patients had a higher relapse rate. Thus, IS patients, with a lower biventricular function and a higher risk profile at baseline, presented similar biventricular function and NYHA class to non-IS patients at follow-up. Minor adverse drug reactions occurred in 13% of patients, all resolved with therapy switch.

CONCLUSIONS

Prolonged tailored IS is effective and safe in BP immune-mediated myocarditis irrespective of histology and clinical presentation.

Statistical plasmode simulations-Potentials, challenges and recommendations

Statistical data simulation is essential in the development of statistical models and methods as well as in their performance evaluation. To capture complex data structures, in particular for high-dimensional data, a variety of simulation approaches have been introduced including parametric and the so-called plasmode simulations. While there are concerns about the realism of parametrically simulated data, it is widely claimed that plasmodes come very close to reality with some aspects of the "truth" known. However, there are no explicit guidelines or state-of-the-art on how to perform plasmode data simulations. In the present paper, we first review existing literature and introduce the concept of statistical plasmode simulation. We then discuss advantages and challenges of statistical plasmodes and provide a step-wise procedure for their generation, including key steps to their implementation and reporting. Finally, we illustrate the concept of statistical plasmodes as well as the proposed plasmode generation procedure by means of a public real RNA data set on breast carcinoma patients.

Impella Versus Intra-Aortic Balloon Pump in Patients With Cardiogenic Shock Treated With Venoarterial Extracorporeal Membrane Oxygenation: An Observational Study

BACKGROUND

Venoarterial extracorporeal membrane oxygenation (VA-ECMO) is increasingly used for patients with cardiogenic shock. Although Impella or intra-aortic balloon pump (IABP) is frequently used for left ventricular unloading (LVU) during VA-ECMO treatment, there are limited data on comparative outcomes. We compared outcomes of Impella and IABP for LVU during VA-ECMO.

METHODS AND RESULTS

Using the Nationwide Readmissions Database between 2016 and 2020, we analyzed outcomes in 3 groups of patients with cardiogenic shock requiring VA-ECMO based on LVU strategies: extracorporeal membrane oxygenation (ECMO) only, ECMO with IABP, and ECMO with Impella. Of 15 980 patients on VA-ECMO, IABP and Impella were used in 19.4% and 16.4%, respectively. The proportion of patients receiving Impella significantly increased from 2016 to 2020 (6.5% versus 25.8%; P-trend<0.001). In-hospital mortality was higher with ECMO with Impella (54.8%) compared with ECMO only (50.4%) and ECMO with IABP (48.4%). After adjustment, ECMO with IABP versus ECMO only was associated with lower in-hospital mortality (adjusted odds ratio [aOR], 0.83; P=0.02). ECMO with Impella versus ECMO only had similar in-hospital mortality (aOR, 1.09; P=0.695) but was associated with more bleeding (aOR, 1.21; P=0.007) and more acute kidney injury requiring hemodialysis (aOR, 1.42; P<0.001). ECMO with Impella versus ECMO with IABP was associated with greater risk of acute kidney injury requiring hemodialysis (aOR, 1.49; P=0.002), higher in-hospital mortality (aOR, 1.32; P=0.001), and higher 40-day mortality (hazard ratio, 1.25; P<0.001).

CONCLUSIONS

In patients with cardiogenic shock on VA-ECMO, LVU with Impella, particularly with 2.5/CP, was not associated with improved survival at 40 days but was associated with increased adverse events compared with IABP. More data are needed to assess Impella platform-specific comparative outcomes of LVU.

Characterizing Imbalance in the Tails of the Propensity Score Distribution

Understanding characteristics of patients with propensity scores in the tails of the propensity score (PS) distribution has relevance for inverse-probability-of-treatment-weighted and PS-based estimation in observational studies. Here we outline a method for identifying variables most responsible for extreme propensity scores. The approach is illustrated in 3 scenarios: 1) a plasmode simulation of adult patients in the National Ambulatory Medical Care Survey (2011-2015) and 2) timing of dexamethasone initiation and 3) timing of remdesivir initiation in patients hospitalized for coronavirus disease 2019 from February 2020 through January 2021. PS models were fitted using relevant baseline covariates, and tails of the PS distribution were defined using asymmetric first and 99th percentiles. After fitting of the PS model in each original data set, values of each key covariate were permuted and model-agnostic variable importance measures were examined. Visualization and variable importance techniques were helpful in identifying variables most responsible for extreme propensity scores and may help identify individual characteristics that might make patients inappropriate for inclusion in a study (e.g., off-label use). Subsetting or restricting the study sample based on variables identified using this approach may help investigators avoid the need for trimming or overlap weights in studies.

The effect of serum calcium on the association of depression with infertility among U.S. women

This study aimed to explored the association between depressive symptoms and infertility among U.S. women, and the effect of serum calcium on this association. We used data from the National Health and Nutrition Examination Survey (2013-2018), relating to women aged 20-45 years. Depressive symptoms were determined using the nine-item Patient Health Questionnaire (PHQ-9 scores ≥10), and interview data were used to identify self-reported infertility. Of 2708 women (mean age: 32.7 ± 7.5 years), 274 were depressed and 12.0 % self-reported being "ever-infertile." Depressive symptoms were associated with infertility in multivariable logistic regression (OR, 1.62; 95 % CI, 1.11-2.38). Depressive symptoms were associated with infertility among participants who were obese (OR, 1.68; 95 % CI, 1.03-2.74), had not received psychological counseling (OR, 1.60; 95 % CI, 1.03-2.50), were antidepressant users (OR 3.22; 95 % CI, 1.15-9.00), and had high serum calcium levels (OR, 2.05; 95 % CI, 1.25-3.35). A significant interaction between serum calcium and depression was observed for infertility (P = .038, interaction likelihood ratio test). In sensitivity analyses, the association between depressive symptoms and infertility remained after excluding women aged ≥35 years (OR, 1.87; 95 % CI, 1.08-3.23), lowering the cut-off for PHQ-9 scores (≥5) (OR, 1.48; 95 % CI, 1.12-1.96), excluding women with some gynecological diseases (OR, 1.63; 95 % CI, 1.07-2.49), and using inverse probability of treatment weighting (OR, 1.64; 95 % CI, 1.17-2.31).

Conclusion

Our findings indicate that depression is associated with infertility among U.S. women and serum calcium may have an effect on the association. Interventions such as serum calcium reduction, weight management and psychosocial counseling for infertility treatment in individuals with depression may be integrated into routine clinical practice. Additionally, more caution could be exercised when using antidepressants.

Longitudinal plasmode algorithms to evaluate statistical methods in realistic scenarios: an illustration applied to occupational epidemiology

INTRODUCTION

Plasmode simulations are a type of simulations that use real data to determine the synthetic data-generating equations. Such simulations thus allow evaluating statistical methods under realistic conditions. As far as we know, no plasmode algorithm has been proposed for simulating longitudinal data. In this paper, we propose a longitudinal plasmode framework to generate realistic data with both a time-varying exposure and time-varying covariates. This work was motivated by the objective of comparing different methods for estimating the causal effect of a cumulative exposure to psychosocial stressors at work over time.

METHODS

We developed two longitudinal plasmode algorithms: a parametric and a nonparametric algorithms. Data from the PROspective Québec (PROQ) Study on Work and Health were used as an input to generate data with the proposed plasmode algorithms. We evaluated the performance of multiple estimators of the parameters of marginal structural models (MSMs): inverse probability of treatment weighting, g-computation and targeted maximum likelihood estimation. These estimators were also compared to standard regression approaches with either adjustment for baseline covariates only or with adjustment for both baseline and time-varying covariates.

RESULTS

Standard regression methods were susceptible to yield biased estimates with confidence intervals having coverage probability lower than their nominal level. The bias was much lower and coverage of confidence intervals was much closer to the nominal level when considering MSMs. Among MSM estimators, g-computation overall produced the best results relative to bias, root mean squared error and coverage of confidence intervals. No method produced unbiased estimates with adequate coverage for all parameters in the more realistic nonparametric plasmode simulation.

CONCLUSION

The proposed longitudinal plasmode algorithms can be important methodological tools for evaluating and comparing analytical methods in realistic simulation scenarios. To facilitate the use of these algorithms, we provide R functions on GitHub. We also recommend using MSMs when estimating the effect of cumulative exposure to psychosocial stressors at work.

Which video technology brings the higher cognitive burden and motion sickness in laparoscopic colorectal surgery: 3D, 2D-4 K or 3D-4 K? a propensity score study

BACKGROUND

Technological development has offered laparoscopic colorectal surgeons new video systems to improve depth perception and perform difficult task in limited space. The aim of this study was to assess the cognitive burden and motion sickness for surgeons during 3D, 2D-4 K or 3D-4 K laparoscopic colorectal procedures and to report post-operative data with the different video systems employed.

METHODS

Patients were assigned to either 3D, 2D-4 K or 3D-4 K video and two questionnaires (Simulator Sickness Questionnaire-SSQ- and NASA Task Load Index -TLX) were used during elective laparoscopic colorectal resections (October 2020-August 2022) from two operating surgeons. Short-term results of the operations performed with the three different video systems were also analyzed.

RESULTS

A total of 113 consecutive patients were included: 41 (36%) in the 3D Group (A), 46 (41%) in the 3D-4 K Group and 26 (23%) in the 2D-4 K Group (C). Weighted and adjusted regression models showed no significant difference in cognitive load amongst the surgeons in the three groups of video systems when using the NASA-TLX. An increased risk for slight/moderate general discomfort and eyestrain in the 3D-4 K group compared with 2D-4 K group (OR = 3.5; p = 0.0057 and OR = 2.8; p = 0.0096, respectively) was observed. Further, slight/moderate difficulty focusing was lower in both 3D and 3D-4 K groups compared with 2D-4 K group (OR = 0.4; p = 0.0124 and OR = 0.5; p = 0.0341, respectively), and higher in the 3D-4 K group compared with 3D group (OR = 2.6; p = 0.0124). Patient population characteristics, operative time, post-operative staging, complication rate and length of stay were similar in the three groups of patients.

CONCLUSIONS

3D and 3D-4 K systems, when compared with 2D-4 K video technology, have a higher risk for slight/moderate general discomfort and eyestrain, but show lower difficulty focusing. Short post-operative outcomes do not differ, whichever imaging system is used.

Low Utilization of Lead Extraction Among Patients With Infective Endocarditis and Implanted Cardiac Electronic Devices

BACKGROUND

Cardiac implantable electronic device (CIED)-associated infections are associated with substantial morbidity, mortality, and costs. Guidelines have cited endocarditis as a Class I indication for transvenous lead removal/extraction (TLE) among patients with CIEDs.

OBJECTIVES

The authors sought to study utilization of TLE among hospital admissions with infective endocarditis using a nationally representative database.

METHODS

Using the Nationwide Readmissions Database (NRD), 25,303 admissions for patients with CIEDs and endocarditis between 2016 and 2019 were evaluated on the basis of International Classification of Diseases-10th Revision, Clinical-Modification (ICD-10-CM) codes.

RESULTS

Among admissions for patients with CIEDs and endocarditis, 11.5% were managed with TLE. The proportion undergoing TLE increased significantly from 2016 to 2019 (7.6% vs 14.9%; P trend < 0.001). Procedural complications were identified in 2.7%. Index mortality was significantly lower among patients managed with TLE (6.0% vs 9.5%; P < 0.001). Presence of Staphylococcus aureus infection, implantable cardioverter-defibrillator, and large hospital size were independently associated with TLE management. TLE management was less likely with older age, female sex, dementia, and kidney disease. After adjustment for comorbidities, TLE was independently associated with significantly lower odds of mortality (adjusted OR: 0.47; 95% CI: 0.37-0.60 by multivariable logistic regression, and adjusted OR: 0.51; 95% CI: 0.40-0.66 by propensity score matching).

CONCLUSIONS

Utilization of lead extraction among patients with CIEDs and endocarditis is low, even in the presence of low rates of procedural complications. Lead extraction management is associated with significantly lower mortality, and its use has trended upward between 2016 and 2019. Barriers to TLE for patients with CIEDs and endocarditis require investigation.

Prenatal exposure to pregabalin, birth outcomes and neurodevelopment - a population-based cohort study in four Nordic countries

INTRODUCTION

Pregabalin is an antiepileptic drug frequently prescribed to pregnant women. Risks of adverse birth and postnatal neurodevelopmental outcomes following prenatal exposure to pregabalin are uncertain.

OBJECTIVE

To investigate the association between prenatal exposure to pregabalin and the risks of adverse birth and postnatal neurodevelopmental outcomes.

METHODS

This study was conducted using population-based registries in Denmark, Finland, Norway, and Sweden (2005-2016). We compared pregabalin exposure against no exposure to antiepileptics and against active comparators lamotrigine and duloxetine. We obtained pooled propensity score-adjusted estimates of association using fixed-effect and Mantel-Haenszel (MH) meta-analyses.

RESULTS

The total number of pregabalin-exposed births was 325/666,139 (0.05%) in Denmark, 965/643,088 (0.15%) in Finland, 307/657,451 (0.05%) in Norway, and 1275/1,152,002 (0.11%) in Sweden. The adjusted prevalence ratios (aPRs) with 95% confidence interval (CI) following pregabalin exposure versus no exposure were 1.14 (0.98-1.34) for major congenital malformations and 1.72 (1.02-2.91) for stillbirth, which attenuated to 1.25 (0.74-2.11) in MH meta-analysis. For the remaining birth outcomes, the aPRs were close to or attenuated toward unity in analyses using active comparators. Adjusted hazard ratios (95% CI) contrasting prenatal pregabalin exposure versus no exposure were 1.29 (1.03-1.63) for ADHD and attenuated when using active comparators, 0.98 (0.67-1.42) for autism spectrum disorders, and 1.00 (0.78-1.29) for intellectual disability.

CONCLUSIONS

Prenatal exposure to pregabalin was not associated with low birth weight, preterm birth, small for gestational age, low Apgar score, microcephaly, autism spectrum disorders, or intellectual disability. On the basis of the upper value of the 95% confidence interval, increased risks greater than 1.8 were unlikely for any major congenital malformation and ADHD. For stillbirth and most groups of specific major congenital malformations, the estimates attenuated in MH meta-analysis.

Does 3D laparoscopic video technology affect long-term survival in right hemicolectomy for cancer compared to standard 2D? A propensity score study

BACKGROUND

There are few studies focused on the short-term results of laparoscopic right hemicolectomy performed with 2D (two-dimension) or 3D (three-dimension) video technology and none on the oncologic effects. The aim of the study was to assess the long-term results of laparoscopic right hemicolectomy (LRH) with intracorporeal anastomosis using 3D or 2D video in patients with right colon cancer with at least three years of oncologic follow-up.

METHODS

Data from patients undergoing laparoscopic right hemicolectomy (LRH) with intracorporeal anastomosis for cancer in an 11-year period (June 2008-June 2019) and ≥ 3 years of follow-up were prospectively collected. Surgical procedures were performed by two expert laparoscopic surgeons.

RESULTS

111 patients were included in the study: 56 (50.5%) in the 3D group and 55 (49.5%) in the 2D group. Tumor stage and number of lymph nodes harvested were similar. Overall and disease-free survival were not different in the two groups. Local recurrence occurred in none of the patients, and distant metachronous metastases were similar in the two groups. A propensity score weighting approach was used to account for potential confounding related to patients' nonrandom allocation to the 2 groups. The effects of the intervention on postoperative outcomes were assessed with a weighted regression approach.

CONCLUSIONS

Laparoscopic 3D technology allows similar oncological results as 2D vision in LRH with intracorporeal anastomosis. Larger prospective randomized studies might confirm these results in the long-term follow-up.