Nevin Manimala

Trials. 2021 Dec 27;22(1):959. doi: 10.1186/s13063-021-05900-7.

ABSTRACT

BACKGROUND: Clustered or correlated outcome data is common in medical research studies, such as the analysis of national or international disease registries, or cluster-randomized trials, where groups of trial participants, instead of each trial participant, are randomized to interventions. Within-group correlation in studies with clustered data requires the use of specific statistical methods, such as generalized estimating equations and mixed-effects models, to account for this correlation and support unbiased statistical inference.

METHODS: We compare different approaches to estimating generalized estimating equations and mixed effects models for a continuous outcome in R through a simulation study and a data example. The methods are implemented through four popular functions of the statistical software R, “geese”, “gls”, “lme”, and “lmer”. In the simulation study, we compare the mean squared error of estimating all the model parameters and compare the coverage proportion of the 95% confidence intervals. In the data analysis, we compare estimation of the intervention effect and the intra-class correlation.

RESULTS: In the simulation study, the function “lme” takes the least computation time. There is no difference in the mean squared error of the four functions. The “lmer” function provides better coverage of the fixed effects when the number of clusters is small as 10. The function “gls” produces close to nominal scale confidence intervals of the intra-class correlation. In the data analysis and the “gls” function yields a positive estimate of the intra-class correlation while the “geese” function gives a negative estimate. Neither of the confidence intervals contains the value zero.

CONCLUSIONS: The “gls” function efficiently produces an estimate of the intra-class correlation with a confidence interval. When the within-group correlation is as high as 0.5, the confidence interval is not always obtainable.

PMID:34961539 | DOI:10.1186/s13063-021-05900-7

Biomed Eng Online. 2021 Dec 28;20(1):130. doi: 10.1186/s12938-021-00970-9.

ABSTRACT

BACKGROUND: Ankle sprain affects the structure and function of ankle cartilage. However, it is not clear whether the daily training and competition affect the ankle cartilage without acute injury. Changes in ankle cartilage without injury may influence future strategies to protect ankle function in athletes. This study aimed to evaluate whether the composition of ankle cartilage significantly altered in asymptomatic adolescent football players after a whole season of training and competition using T2-mapping magnetic resonance imaging (MRI).

MATERIALS AND METHODS: 12 local club’s U17 asymptomatic adolescent football players without abnormalities in routine MRI were included. Routine and T2-mapping MRI were performed to measure the cartilage thickness of tibiotalar joint (TT) and posterior subtalar joint (pST) and T2 values in pre- and post-seasons. All of them took the right side as dominant foot.

RESULTS: In the pre- and post-seasons, cartilage T2 values in TT (talus side) and pST (calcaneus side) were higher than that of TT (tibial side) and pST (talus side) (all p < 0.05), which was caused by magic angle effect and gravity load. No statistically significant differences in thickness after season in the other cartilages of ankle were found compared with that before the season (all p > 0.05). However, T2 values of TT (tibial side and talus side) cartilage in the dominant foot were significantly reduced after season (p = 0.008; p = 0.034). These results indicate that the microstructure of articular cartilage changes in the joints with greater mobility, although no trauma occurred and the gross morphology of cartilage did not change.

CONCLUSION: Changes in the T2 values of tibiotalar joint cartilage in the dominant foot of healthy young athletes before and after the season suggest that the microstructure of cartilage had changed during sports even without injury. This finding suggests that the dominant ankle joint should be protected during football to delay degeneration of the articular cartilage.

PMID:34961538 | DOI:10.1186/s12938-021-00970-9

J Cardiothorac Surg. 2021 Dec 27;16(1):353. doi: 10.1186/s13019-021-01732-3.

ABSTRACT

BACKGROUND: Pre-operative ejection fraction (EF) and comorbidities affect post-op outcomes. We aimed to compare the mortality and adverse events of patients with different baseline EF and also to evaluate the distribution of comorbidities in each EF group.

METHODS: A total of 20,937 patients who underwent isolated coronary artery bypass graft (CABG) surgery from January 2006 to December 2016 was included. Patients were divided into three groups based on their pre-operative left ventricular EF as follows; (1) Normal: EF ≥ 50%; (2) Mild to moderately reduced: 50% < EF ≤ 35%; and (3) Severely reduced: EF < 35%. The backward elimination method was considered for multivariate Cox-regression analysis to locate predictors of mortality and non-fatal cerebro-cardiovascular events (CCVEs). The median follow-up time was 5.61 [3.12-8.0] years.

RESULTS: The mean age in the total population was 60.94 ± 9.51 years and 73.6% of the total population was male. Diabetes mellitus was the common risk factor of mortality and CCVE in all EF groups. Impaired renal function (GFR < 60 ml/min) was associated with a higher risk of mortality after CABG regardless of EF level. The median 5-year mortality rate in patients with normal EF, mild-moderately reduced EF and severely reduced EF were 9.5%, 12.8%, and 22.7% respectively (P < 0.001). Although the trend of CCVEs was higher in severe left ventricle (LV) dysfunction, it was not statistically significant (p = 0.071).

CONCLUSION: Patients with severely reduced EF are at higher risk of mortality after CABG compared to those with higher EF levels; however, the rate of CCVEs may not be necessarily higher after adjustment for multiple pre-operative comorbidities.

PMID:34961534 | DOI:10.1186/s13019-021-01732-3

J Neuroeng Rehabil. 2021 Dec 28;18(1):184. doi: 10.1186/s12984-021-00977-2.

ABSTRACT

BACKGROUND: A client-centred task-oriented approach has advantages towards motivation and adherence to therapy in neurorehabilitation, but it is costly to integrate in practice. An intelligent Activity-based Client-centred Training (i-ACT), a low-cost Kinect-based system, was developed which integrates a client-centred and task-oriented approach. The objectives were (1) to investigate the effect of additional i-ACT training on functioning. And (2) to assess whether training with i-ACT resulted in more goal oriented training.

METHODS: A single-blind randomised controlled trial was performed in 4 Belgian rehabilitation centres with persons with central nervous system deficits. Participants were randomly allocated through an independent website-based code generator using blocked randomisation (n = 4) to an intervention or control group. The intervention group received conventional care and additional training with i-ACT for 3 × 45 min/week during 6 weeks. The control group received solely conventional care. Functional ability and performance, quality of life (QoL), fatigue, trunk movement, and shoulder active range of motion (AROM) were assessed at baseline, after 3 weeks and 6 weeks of training, and 6 weeks after cessation of training. Data were analysed using non-parametric within and between group analysis.

RESULTS: 47 persons were randomised and 45 analysed. Both intervention (n = 25) and control (n = 22) group improved over time on functional ability and performance as measured by the Wolf Motor Function Test, Manual Ability Measure-36, and Canadian Occupational Performance Measure, but no major differences were found between the groups on these primary outcome measures. Regarding QoL, fatigue, trunk movement, and shoulder AROM, no significant between group differences were found. High adherence for i-ACT training was found (i.e. 97.92%) and no adverse events, linked to i-ACT, were reported. In the intervention group the amount of trained personal goals (88%) was much higher than in the control group (46%).

CONCLUSIONS: Although additional use of i-ACT did not have a statistically significant added value regarding functional outcome over conventional therapy, additional i-ACT training provides more individualised client-centred therapy, and adherence towards i-ACT training is high. A higher intensity of i-ACT training may increase therapy effects, and should be investigated in future research.

TRIAL REGISTRATION: ClinicalTrials.gov Identifier NCT02982811. Registered 29 November 2016.

PMID:34961531 | DOI:10.1186/s12984-021-00977-2

Respir Res. 2021 Dec 27;22(1):321. doi: 10.1186/s12931-021-01917-3.

ABSTRACT

BACKGROUND: Pulmonary hypertension (PH) is one of the common complications in chronic obstructive pulmonary disease (COPD). The study aimed to evaluate the predicting ability of N-terminal pro brain natriuretic peptide (NT-pro BNP) in patients with AECOPD-PH and its relationship with the severity of PH.

METHODS: A large retrospective case-controlled study (n = 1072) was performed in the First Affiliated Hospital of Xinjiang Medical University from January 2018 to December 2020, and patients were divided into stable COPD (n = 178), AECOPD (n = 688) and AECOPD-PH group (n = 206). Different statistical models were used to screen for reliable and stable biomarkers.

RESULTS: In unadjusted analysis and PSM (model 1, 2, 3), red cell distribution width (RDW), total bilirubin (TBIL), and NT-pro BNP were higher in patients with AECOPD-PH than those in AECOPD group. Logistic regression analysis showed, when the range of NT-proBNP was 271-1165 pg/mL (OR: 0.293; 95%CI: 0.184-0.467; P < 0.001) and NT-proBNP > 1165 pg/mL (OR: 0.559; 95%CI: 0.338-0.926; P = 0.024), the morbidity risk of PH in AECOPD patients was increased, so did TBIL. In receiver operating characteristic (ROC) curves, at the cut-off value of NT-proBNP was 175.14 pg/mL, AUC was 0.651 (P < 0.001), which was better than TBIL (AUC: 0.590, P < 0.001). As for the results of rank correlation analysis, NT-proBNP had a weak correlation with severity of PH with AECOPD (r_s = 0.299, P = 0.001) and its relative relevance with other biomarkers (RDW was 0.359 and TBIL was 0.238, P < 0.001).

CONCLUSIONS: Our findings suggest that NT-proBNP has a diagnostic efficacy in AECOPD-PH and NT-proBNP has a weak correlation with severity of PH with AECOPD.

PMID:34961527 | DOI:10.1186/s12931-021-01917-3

J Neuroimaging. 2021 Dec 27. doi: 10.1111/jon.12961. Online ahead of print.

ABSTRACT

BACKGROUND AND PURPOSE: Stent retriever (SR) thrombectomy is commonly used for the treatment of emergent large vessel occlusion (ELVO) in acute ischemic stroke. Clot imaging parameters such as clot length, diameter, distance to the internal carotid artery terminus, and vessel angle where the SR is deployed may predict the likelihood of achieving first pass effect (FPE). Most of the proposed factors that seem to affect recanalization success have been studied individually, and conflicting data derived from clinical versus in vitro studies using 3-dimensional printed models of intracranial circulation currently exist.

METHODS: Using patient-specific 3-dimensional phantoms of the cervical and intracranial circulation, we simulated middle cerebral arteries (MCA) M1 and M2 occlusions treated with SR thrombectomy using Solitaire (Medtronic) or Trevo (Styker). Our primary outcome was FPE, defined as Thrombolysis in Cerebral Infarction score of 2c-3 achieved after a single thrombectomy attempt. We also performed retrospective analysis of same clot imaging characteristics of consecutive cases of MCA occlusion and its association with FPE matching the 3-dimensional in vitro experiments. Analysis was conducted using IBM SPSS Statistics Version 25 (IBM Corp., Armonk, NY). Chi-square tests and bivariate logistic regressions were the main statistical tests used in analysis. A p-value of less than .05 was considered to indicate statistical significance. Ninety-five confidence intervals (95% CI) were generated.

RESULTS: We compared 41 thrombectomy experiments performed using patient-specific 3-dimensional in vitro models with a retrospective cohort of 41 patients treated with SR thrombectomy. We found that in the in vitro cohort, higher MCA angulation was associated with a lower likelihood of FPE (odds ratio [OR] = 0.967, 95% CI = 0.944-0.991, p = .008). Meanwhile in the in vivo cohort, higher MCA angulation was associated with a higher likelihood of FPE (OR = 1.039, 95% CI = 1.003-1.077, p = .033). Neither clot length nor location of clot (M1 vs. M2) was associated with a difference in FPE rates in either cohort.

DISCUSSION: Comparison of SR thrombectomy performed during actual MCA occlusion cases versus patient-specific 3-dimensional replicas revealed MCA angulation as an independent predictor of procedure success or failure. However, the opposite direction of effect was observed between the two studied environments, indicating potential limitations of studying SR thrombectomy using 3-dimensional models of LVO.

PMID:34958701 | DOI:10.1111/jon.12961

Magn Reson Med. 2021 Dec 27. doi: 10.1002/mrm.29125. Online ahead of print.

ABSTRACT

PURPOSE: Multicomponent analysis of MRI T₂ relaxation time (mcT₂ ) is commonly used for estimating myelin content by separating the signal at each voxel into its underlying distribution of T₂ values. This voxel-based approach is challenging due to the large ambiguity in the multi-T₂ space and the low SNR of MRI signals. Herein, we present a data-driven mcT₂ analysis, which utilizes the statistical strength of identifying spatially global mcT₂ motifs in white matter segments before deconvolving the local signal at each voxel.

METHODS: Deconvolution is done using a tailored optimization scheme, which incorporates the global mcT₂ motifs without additional prior assumptions regarding the number of microscopic components. The end results of this process are voxel-wise myelin water fraction maps.

RESULTS: Validations are shown for computer-generated signals, uniquely designed subvoxel mcT₂ phantoms, and in vivo human brain. Results demonstrated excellent fitting accuracy, both for the numerical and the physical mcT₂ phantoms, exhibiting excellent agreement between calculated myelin water fraction and ground truth. Proof-of-concept in vivo validation is done by calculating myelin water fraction maps for white matter segments of the human brain. Interscan stability of myelin water fraction values was also estimated, showing good correlation between scans.

CONCLUSION: We conclude that studying global tissue motifs prior to performing voxel-wise mcT₂ analysis stabilizes the optimization scheme and efficiently overcomes the ambiguity in the T₂ space. This new approach can improve myelin water imaging and the investigation of microstructural compartmentation in general.

PMID:34958690 | DOI:10.1002/mrm.29125

Brief Bioinform. 2021 Dec 28:bbab528. doi: 10.1093/bib/bbab528. Online ahead of print.

ABSTRACT

Batch effects explain a large part of the noise when merging gene expression data. Removing irrelevant variations introduced by batch effects plays an important role in gene expression studies. To obtain reliable differential analysis results, it is necessary to remove the variation caused by technical conditions between different batches while preserving biological variation. Usually, merging data directly with batch effects leads to a sharp rise in false positives. Although some methods of batch correction have been developed, they have some drawbacks. In this study, we develop a new algorithm, adjustment mean distribution-based normalization (AMDBNorm), which is based on a probability distribution to correct batch effects while preserving biological variation. AMDBNorm solves the defects of the existing batch correction methods. We compared several popular methods of batch correction with AMDBNorm using two real gene expression datasets with batch effects and analyzed the results of batch correction from the visual and quantitative perspectives. To ensure the biological variation was well protected, the effects of the batch correction methods were verified by hierarchical cluster analysis. The results showed that the AMDBNorm algorithm could remove batch effects of gene expression data effectively and retain more biological variation than other methods. Our approach provides the researchers with reliable data support in the study of differential gene expression analysis and prognostic biomarker selection.

PMID:34958674 | DOI:10.1093/bib/bbab528

J Biopharm Stat. 2021 Dec 27:1-17. doi: 10.1080/10543406.2021.2011902. Online ahead of print.

ABSTRACT

In this paper, we develop a methodology for leveraging real-world data into single-arm clinical trial studies. In recent years, the idea of augmenting randomized clinical trials data with real-world data has emerged as a particularly attractive technique for health organizations and drug developers to accelerate the drug development process. Major regulatory authorities such as the Food and Drug Administration and European Medicines Agency have recognized the potential of utilizing real-world data and are advancing toward making regulatory decisions based on real-world evidence. Several statistical methods have been developed in recent years for borrowing data from real-world sources such as electronic health records, product and disease registries, as well as claims and billing data. We propose a novel approach to augment single-arm clinical trials with the real-world data derived from single or multiple data sources. Furthermore, we illustrate the proposed method in the presence of missing data and conduct simulation studies to evaluate its performance in diverse settings.

PMID:34958629 | DOI:10.1080/10543406.2021.2011902

Vascular. 2021 Dec 27:17085381211062747. doi: 10.1177/17085381211062747. Online ahead of print.

ABSTRACT

OBJECTIVES: Thoracic outlet syndrome (TOS) is a group of disorders caused by impingement of the neurovascular structures at the thoracic outlet. Neurogenic TOS (nTOS), which is thought to be caused by a compression of the brachial plexus, accounts for more than 90% of the cases. Although treatment for nTOS is successful through physiotherapy and/or surgical decompression, little is known about the impact of psychosocial factors, namely, major depressive disorder (MDD), on postoperative outcomes such as non-routine discharge (NRD). Here, we assess whether MDD predicts the type of discharge following nTOS surgical intervention.

METHODS: A retrospective analysis of the National Inpatient Sample database from the years 2005-2018 was performed. Using the International Classification of Diseases Clinical Modification, Ninth and Tenth revisions, patients who underwent a surgical intervention for nTOS were identified. Our primary outcome was to investigate the effects of MDD on nTOS patient disposition status after surgical management; secondary outcomes included analysis of total hospital charges and length of stay. NRD was defined as anything beyond discharge home without healthcare services. Univariate and multivariable logistic regression analyses were conducted to assess MDD and other potential independent predictors of NRD and prolonged hospital stay (> 2 days) following surgical intervention.

RESULTS: A total of 6099 patients were identified: 596 (9.77%) patients with MDD and 5503 (90.23%) without MDD. On average, patients with MDD were older (39.6 ± 12.0 years vs. 36.0 ± 13.0 years; p < 0.001), female (80.7% vs. 63.5%; p < 0.001), white (89.6% vs. 85.6%; p = 0.030), and on Medicare (9.6% vs 5.2%; p < 0.001). Univariate and multivariable logistic regression models identified MDD as an independent risk factor associated with a higher risk of NRD (adjusted odds ratio [aOR], 1.50; 95% confidence interval [CI], 1.0-2.2). Additionally, chronic kidney disease (aOR, 2.60; 95% CI, 1.2-5.4), postoperative complications (aOR, 1.87; 95% CI, 1.2-2.9), and Medicare (aOR, 2.95; 95% CI, 1.9-4.7) were statistically significant predictors for higher risk of NRD. However, MDD was not associated with prolonged hospital stay (aOR, 1.00; 95% CI, 0.8-1.2) or higher median of total charges (MDD group: $27,867 vs. non-MDD group: $28,123; p = 0.799).

CONCLUSION: Comorbid MDD was strongly associated with higher NRD rates following nTOS surgical intervention. MDD had no significant impact on length of hospital stay or total hospital charges. Additional prospective research is necessary in order to better evaluate the impact of MDD in patients with nTOS.

PMID:34958613 | DOI:10.1177/17085381211062747