Statistical primer: performing repeated-measures analysis

Effect of aortic valve phenotype and sex on aorta dilation in patients with aortic stenosis

BACKGROUND

Bicuspid aortic valve (BAV) is often associated with a concomitant aortopathy. However, few studies have evaluated the effect of the aortic valve (AV) phenotype on the rate of dilation of the aorta. This study aimed to compare the progression rate of aorta dimensions according to AV phenotype (BAV vs tricuspid AV (TAV)), fusion type and sex in patients with aortic stenosis (AS).

METHODS

310 patients with AS (224 TAV and 86 BAV) recruited in the Metabolic Determinants of the Progression of Aortic Stenosis study (PROGRESSA, NCT01679431) were included in this analysis. Doppler echocardiography was performed annually to assess AS severity and measure ascending aorta (AA) dimensions. Baseline and last follow-up visit measurements were used to assess the annualised change.

RESULTS

Median AA annualised change was larger in BAV versus TAV (0.33±0.65 mm/year vs 0.21±0.56 mm/year, p=0.04). In the whole cohort, BAV phenotype and higher low-density lipoprotein (LDL) levels were significantly associated with fast progression of AA dilation in univariate analysis (OR 1.80, 95% CI 1.08 to 2.98, p=0.02; 1.37, 95% CI 1.04 to 1.80, p=0.03, respectively). AA dilation rate did not vary according to the BAV subtype (p=0.142). Predictors of AA progression rate were different between valve phenotypes, with higher apolipoprotein B/apolipoprotein A-I ratio, higher baseline peak aortic jet velocity (Vpeak) and smaller baseline AA diameter in the TAV cohort (all p<0.05) versus absence of hypertension, higher LDL levels and smaller baseline AA diameter in the BAV cohort (all p<0.02). In men, higher baseline Vpeak and smaller baseline AA (p<0.001) were independently associated with increased annualised AA dilation, while in women, higher LDL levels (p=0.026) were independently associated with faster AA dilation.

CONCLUSION

This study suggests that BAV is associated with faster dilation of the AA. Predictors of AA dilation are different between valve phenotype and sex, with higher LDL levels being associated with faster AA dilation in BAV.

Optimizing short-term in-class process evaluation: analyzing the effectiveness of teaching interventions in pharmaceutical education using repeated measures analysis

BACKGROUND

The assessment of the effectiveness of teaching interventions in enhancing students' understanding of the Pharmaceutical Care Network Europe (PCNE) Classification System is crucial in pharmaceutical education. This is especially true in regions like China, where the integration of the PCNE system into undergraduate teaching is limited, despite its recognized benefits in addressing drug-related problems in clinical pharmacy practice. Therefore, this study aimed to evaluate the effectiveness of teaching interventions in improving students' understanding of the PCNE Classification System in pharmaceutical education.

METHODS

Undergraduate pharmacy students participated in a series of sessions focused on the PCNE system, including lectures (t1), case analyses (t2), and practical implementation (t3). The levels of understanding were evaluated using time-course questionnaires. Initially, paired samples t-Tests were used to compare understanding levels between different time points. Subsequently, Repeated Measures Analysis (RMA) was employed. Pearson correlation analysis was conducted to examine the relationship between understanding levels and the usability and likelihood of using the PCNE system, as reported in the questionnaires.

RESULTS

The paired samples t-Tests indicated insignificant differences between t2 and t3, suggesting limited improvement following the practical implementation of the PCNE system. However, RMA revealed significant time effects on understanding levels in effective respondents and the focused subgroup without prior experience (random intercept models: all p < 0.001; random slope models: all p < 0.001). These results confirmed the effectiveness of all three teaching interventions. Pearson correlation analysis demonstrated significant positive correlations between understanding levels and the usability and likelihood of using the PCNE system at all examined time points. This finding highlighted the reliability of the understanding levels reported in the questionnaires. The homework scores were used as external calibration standards, providing robust external validation of the questionnaire's validity.

CONCLUSION

The implementation of RMA provided robust evidence of the positive impact of time on understanding levels. This affirmed the effectiveness of all teaching interventions in enhancing students' comprehension of the PCNE Classification System. By utilizing RMA, potential errors inherent in common statistical methods, such as t-Tests, were mitigated. This ensured a more comprehensive and accurate assessment of the effectiveness of the teaching interventions.

Using routinely collected clinical data for circadian medicine: A review of opportunities and challenges

A wealth of data is available from electronic health records (EHR) that are collected as part of routine clinical care in hospitals worldwide. These rich, longitudinal data offer an attractive object of study for the field of circadian medicine, which aims to translate knowledge of circadian rhythms to improve patient health. This narrative review aims to discuss opportunities for EHR in studies of circadian medicine, highlight the methodological challenges, and provide recommendations for using these data to advance the field. In the existing literature, we find that data collected in real-world clinical settings have the potential to shed light on key questions in circadian medicine, including how 24-hour rhythms in clinical features are associated with-or even predictive of-health outcomes, whether the effect of medication or other clinical activities depend on time of day, and how circadian rhythms in physiology may influence clinical reference ranges or sampling protocols. However, optimal use of EHR to advance circadian medicine requires careful consideration of the limitations and sources of bias that are inherent to these data sources. In particular, time of day influences almost every interaction between a patient and the healthcare system, creating operational 24-hour patterns in the data that have little or nothing to do with biology. Addressing these challenges could help to expand the evidence base for the use of EHR in the field of circadian medicine.

Do Different Types of Microphones Affect Listening Effort in Cochlear Implant Recipients? A Pupillometry Study

BACKGROUND

It is known that subjects with a cochlear implant (CI) need to exert more listening effort to achieve adequate speech recognition compared to normal hearing subjects. One tool for assessing listening effort is pupillometry. The aim of this study is to evaluate the effectiveness of adaptive directional microphones in reducing listening effort for CI recipients.

METHODS

We evaluated listening in noise and listening effort degree (by pupillometry) in eight bimodal subjects with three types of CI microphones and in three sound configurations.

RESULTS

We found a correlation only between sound configurations and listening in noise score (p-value 0.0095). The evaluation of the microphone types shows worse scores in listening in noise with Opti Omni (+3.15 dB SNR) microphone than with Split Dir (+1.89 dB SNR) and Speech Omni (+1.43 dB SNR). No correlation was found between microphones and sound configurations and within the pupillometric data.

CONCLUSIONS

Different types of microphones have different effects on the listening of CI patients. The difference in the orientation of the sound source is a factor that has an impact on the listening effort results. However, the pupillometry measurements do not significantly correlate with the different microphone types.

Long-Term Clinical and Echocardiographic Outcomes Following the Ross Procedure: A Post Hoc Analysis of a Randomized Clinical Trial

Importance

The Ross procedure as treatment for adults with aortic valve disease (AVD) has been the subject of renewed interest.

Objective

To evaluate the long-term clinical and echocardiographic outcomes following the Ross procedure for the treatment of adults with AVD.

Design, Setting, and Participants

This post hoc analysis of a randomized clinical trial included adult patients (age <69 years) who underwent a Ross procedure for the treatment of AVD, including those with active endocarditis, rheumatic AVD, decreased ejection fraction, and previous cardiac surgery. The trial, conducted from September 1, 1994, to May 31, 2001, compared homograft root replacement with the Ross procedure at a single center. Data after 2010 were collected retrospectively in November and December 2022.

Exposure

Ross procedure.

Main Outcomes and Measures

The primary end point was long-term survival among patients who underwent the Ross procedure compared with that in the age-, country of origin- and sex-matched general population. Secondary end points were freedom from any reintervention, autograft reintervention, or homograft reintervention and time-related valve function, autograft diameter, and functional status.

Results

This study included 108 adults (92 [85%] male) with a median age of 38 years (range, 19-66 years). Median duration of clinical follow-up was 24.1 years (IQR, 22.6-26.1 years; 2488 patient-years), with 98% follow-up completeness. Of these patients, 9 (8%) had active endocarditis and 45 (42%) underwent reoperations. The main hemodynamic lesion was stenosis in 30 (28%) and regurgitation in 49 (45%). There was 1 perioperative death (0.9%). Twenty-five year survival was 83.0% (95% CI, 75.5%-91.2%), representing a relative survival of 99.1% (95% CI, 91.8%-100%) compared with the general population (83.7%). At 25 years, freedom from any reintervention was 71.1% (95% CI, 61.6%-82.0%); from autograft reintervention, 80.3% (95% CI, 71.9%-89.6%); and from homograft reintervention, 86.3% (95% CI, 79.0%-94.3%). Thirty-day mortality after the first Ross-related reintervention was 0% and after all Ross-related reinterventions was 3.8% (n = 1); 10-year survival after reoperation was 96.2% (95% CI, 89.0%-100%).

Conclusions and Relevance

This study found that the Ross procedure provided excellent survival into the third decade postoperatively that was comparable to that in the general population. Long-term freedom from reintervention demonstrated that the Ross procedure may be a durable substitute into late adulthood, showing a delayed but progressive functional decline.

Trial Registration

isrctn.org Identifier: ISRCTN03530985.

Mid-term clinical and echocardiographic results of the INSPIRIS RESILIA aortic valve: a retrospective comparison to the Magna Ease

OBJECTIVES

The INSPIRIS aortic valve combines the RESILIA proprietary tissue preservation process and an expandable stent frame to benefit future transcatheter valve-in-valve procedures. As the INSPIRIS valve became commercially available in 2017, mid-term outcome reports are scarce. We aimed to evaluate mid-term safety and echocardiographic performance of the INSPIRIS valve in comparison to its predecessor, the Carpentier Edwards Perimount Magna Ease (ME).

METHODS

This study was a retrospective single-centre study. Clinical results included early postoperative outcomes, mid-term mortality and readmission for cardiovascular cause or stroke. Echocardiographic follow-up (FU) was performed at discharge and 1-3, 6, 12 and 24 months. Clinical end point analyses were accomplished with a propensity score matching analysis and FU echocardiographic data comparisons using pairwise analyses and linear mixed-effect models.

RESULTS

We included 953 patients who received an INSPIRIS (n = 488) or ME (n = 463) bioprosthesis between January 2018 and July 2021. In the matched population (n = 217 per group), no significant difference in short-term outcomes was observed, survival was similar at 30 months (INSPIRIS: 94% vs ME: 91%, P = 0.89), but freedom from readmission was higher in the INSPIRIS group (94% vs 86%, P = 0.014). INSPIRIS valves had a lower gradient at discharge (∼10 vs 14 mmHg, P < 0.001), 1-3 months (∼10 vs 12 mmHg, P < 0.001) and 24 months (∼11 vs 17 mmHg, P < 0.001) in paired analyses and significantly lower evolution of mean transvalvular gradients compared to ME.

CONCLUSIONS

This study represents the largest comparative evaluation of the INSPIRIS to the ME valves, which demonstrated safe clinical outcomes and favourable haemodynamic performance at 2 years. Long-term FU is underway.

TREM2 modulates neuroinflammation with elevated IRAK3 expression and plays a neuroprotective role after experimental SAH in rats

BACKGROUND

The modulation of neuroinflammation is a new direction that may alleviate the early brain injury after subarachnoid hemorrhage (SAH). Brain resident microglia/macrophages (Mi/MΦ) are the key drivers of neuroinflammation. Triggering receptor expressed on myeloid cells 2 (TREM2) has been reported to play a neuroprotective role by activating phagocytosis and suspending inflammatory response in experimental ischemic stroke and intracerebral hemorrhage. This study was designed to investigate the role of TREM2 on neuroinflammation and neuroprotective effects in a rat SAH model.

METHODS

Adult male Sprague-Dawley rats were induced SAH through endovascular perforation. Lentivirus vectors were administered by i.c.v. to induce TREM2 overexpression or knockdown 7 days before SAH induction. Short- and long-term neurobehavioral tests, western blotting, immunofluorescence, enzyme-linked immunosorbent assay, terminal deoxynucleotidyl transferase dUTP nick end labeling and Nissl staining were performed to explore the neuroprotective role of TREM2 after SAH.

RESULTS

The expression of TREM2 elevated in a rat SAH model with a peak at 48 h after SAH and mainly expressed in Mi/MΦ in brain. TREM2 overexpression improved short- and long-term neurological deficits induced by SAH in rats, while TREM2 knockdown worsened neurological dysfunction. The rats with TREM2 overexpressed presented less neuronal apoptosis and more neuronal survival at 48 h after SAH, while the rats with TREM2 knockdown presented on the contrary. TREM2 overexpression manifested activated phagocytosis and suppressed inflammatory response, with the increase of CD206⁺/CD11b⁺ cells and IL-10 expression as well as the decrease of the infiltration of MPO⁺ cells and the expression of TNF-α, IL-1β. While TREM2 knockdown abolished these effects. The protein level of IRAK3, a negative regulatory factor of inflammation, was significantly elevated after TREM2 overexpression and declined after TREM2 knockdown.

CONCLUSIONS

Our research suggested TREM2 played a neuroprotective role and improved the short- and long-term neurological deficits by modulating neuroinflammation after SAH. The modulation on neuroinflammation of TREM2 after SAH was related with the elevated protein level of IRAK3.

Repeated-Measures Analysis in the Context of Heteroscedastic Error Terms with Factors Having Both Fixed and Random Levels

The design and analysis of experiments which involve factors each consisting of both fixed and random levels fit into linear mixed models. The assumed linear mixed-model design matrix takes either a full-rank or less-than-full-rank form. The complexity of the data structures of such experiments falls in the model-selection and parameter-estimation process. The fundamental consideration in the estimation process of linear models is the special case in which elements of the error vector are assumed equal and uncorrelated. However, different assumptions on the structure of the variance–covariance matrix of error vector in the estimation of parameters of a linear mixed model may be considered. We conceptualise a repeated-measures design with multiple between-subjects factors, in which each of these factors has both fixed and random levels. We focus on the construction of linear mixed-effects models, the estimation of variance components, and hypothesis testing in which the default covariance structure of homoscedastic error terms is not appropriate. We illustrate the proposed approach using longitudinal data fitted to a three-factor linear mixed-effects model. The novelty of this approach lies in the exploration of the fixed and random levels of the same factor and in the subsequent interaction effects of the fixed levels. In addition, we assess the differences between levels of the same factor and determine the proportion of the total variation accounted for by the random levels of the same factor.

Reporting sample size calculations for randomized controlled trials published in nursing journals: A cross-sectional study

BACKGROUND

Prior sample size calculation is essential to ensure that a randomized controlled trial (RCT) has enough power to detect any statistical differences between two groups while not over-recruiting participants. However, the compliance among RCTs published in nursing field is unknown.

OBJECTIVE

To describe the reporting of sample size methodology and parameters used in RCTs published in nursing journals.

DESIGN

A cross-sectional database search of all RCTs published in nursing journals was conducted.

SETTING

116 journals in the category of nursing.

PARTICIPANTS

Not applicable.

METHODS

A database search was conducted to identify all RCTs published in nursing journals from January 2016 to December 2016. Two-arm RCTs were reviewed to see if sample size estimation was mentioned in the text and if the parameters were sufficient for the estimation. For RCTs with effect size, sample sizes were recomputed and compared with the reported size.

RESULTS

Two hundred and twenty-three RCTs were included in this study, and 143 (64.1%) studies mentioned how they obtained their sample sizes. Among these 143 studies, 132 (92.3%), 133 (93.0%), 33 (23.1%), and 84 (58.7%) specified their levels of significance (α), powers, tails of the test, and effect size related information, respectively. Only 22 (15.4%) provided all the necessary parameters to compute their sample sizes. Out of the 84 RCTs with effect size information, 49 (58.3%) had a relative difference of less than 25%.

DISCUSSION

Around 36% of the RCTs published in nursing journals did not report how they reached their sample sizes. For those that had reported, only 15% provided all the necessary parameters to re-compute their sizes. It is concluded that the methodology and parameters used for sample size determination are inadequately reported in RCTs published in nursing journals.

Statistical primer: checking model assumptions with regression diagnostics

Regression modelling is an important statistical tool frequently utilized by cardiothoracic surgeons. However, these models-including linear, logistic and Cox proportional hazards regression-rely on certain assumptions. If these assumptions are violated, then a very cautious interpretation of the fitted model should be taken. Here, we discuss several assumptions and report diagnostics that can be used to detect departures from these assumptions. Most of the diagnostics discussed are based on residuals: a measure of the difference between the observed and model fitted values. Reliable and generalizable results depend on correctly developed statistical models, and proper diagnostics should play an integral part in the model development.

D-optimal designs for multiarm trials with dropouts

Multiarm trials with follow-up on participants are commonly implemented to assess treatment effects on a population over the course of the studies. Dropout is an unavoidable issue especially when the duration of the multiarm study is long. Its impact is often ignored at the design stage, which may lead to less accurate statistical conclusions. We develop an optimal design framework for trials with repeated measurements, which takes potential dropouts into account, and we provide designs for linear mixed models where the presence of dropouts is noninformative and dependent on design variables. Our framework is illustrated through redesigning a clinical trial on Alzheimer's disease, whereby the benefits of our designs compared with standard designs are demonstrated through simulations.

Sonographic follow-up after endoscopic carpal tunnel release for severe carpal tunnel syndrome: a one-year neuroanatomical prospective observational study

BACKGROUND

Endoscopic carpal tunnel release (ECTR) has been gradually adopted for the treatment of severe carpal tunnel syndrome (CTS). However, perioperative assessment of neuroanatomical parameters of median nerve, which are important determinant of median nerve recovery, has rarely been reported. This one-year prospective study aimed to investigate the natural history of the neuroanatomical morphology of the median nerve after ECTR in severe CTS patients by high-frequency ultrasonography and assess the ability of neuroanatomical measures to quantify morphological recovery of the median nerve after ECTR.

METHODS

This study recruited 31 patients (44 wrists) with a definitive diagnosis of severe CTS and underwent ECTR operation. The edema length (EL) of median nerve from the inlet of the carpal tunnel to the proximal wrist was detected on long axis imaging plane and the anteroposterior diameter (D) and cross-sectional area (CSA) at the inlet of the carpal tunnel on short axis imaging plane were detected by high frequency ultrasound. All these metrics were detected at 3 days before surgery and at the 2nd week, 4th week, 3rd month, 6th month and 12th month after surgery separately.

RESULTS

There was no significant difference of each parameter between the 2-week postoperative (1.914 ± 0.598 cm in EL, 0.258 ± 0.039 cm in D and 0.138 ± 0.015 cm2 in CSA) and 3-days preoperative time points (P-EL =0.250; P-D = 0.125; P-CSA =0.712). From the fourth week to the third month after surgery, the parameters quickly improved. The EL (0.715 ± 0.209 cm), D (0.225 ± 0.017 cm) and CSA (0.117 ± 0.012 cm2) at the 3- month postoperative time points were more reduced than at the fourth week after surgery (P-EL < 0.001; P-D = 0.038; P-CSA =0.014). Thereafter, the neurological anatomy parameters recovered slowly. By the 12-month postoperative time points, the three parameters were neuroanatomically close to normal. Compared to the control group in D (0.213 ± 0.005 cm), there was no difference at the 12-month time point (0.214 ± 0.009 cm, P = 0.939). However, the difference in EL (0.098 ± 0.030 cm vs. 0.016 ± 0.011 cm) and CSA (0.103 ± 0.008 cm2 vs. 0.073 ± 0.005 cm2) between patients and healthy volunteers at the 12-month time point still existed (P-EL < 0.001; P-CSA < 0.001).

CONCLUSIONS

Neuroanatomical parameters were gradually improved after ECTR surgery. The best time for US follow up is at 3-month postoperative time point for patients who do not show clinical improvement, since at this time the change is the greatest for most CTS patients. This study has been registered in Chinese Clinical Trial Registry: ChiCTR-ROC-17014068 (retrospectively registered 20-12-2017).

A Web-Based Gamification Program to Improve Nutrition Literacy in Families of 3- to 5-Year-Old Children: The Nutriscience Project

OBJECTIVE

Assess the impact of a web-based gamification program on nutrition literacy of families and explore differences in impact by socioeconomic status.

DESIGN

Quasi-experimental.

SETTING

Thirty-seven kindergartens from Portugal.

PARTICIPANTS

Eight hundred seventy-seven families.

INTERVENTION

Web-based social network of participants' interactions, educational materials, apps and nutritional challenges, focused on fruit, vegetables, sugar, and salt.

MAIN OUTCOME MEASURES

Parental nutrition literacy (self-reported survey - 4 dimensions: Nutrients, Food portions, Portuguese food wheel groups, Food labeling).

ANALYSIS

General linear model - Repeated measures was used to analyze the effect on the nutrition literacy score.

RESULTS

Families uploaded 1267 items (recipes, photographs of challenges) and educators uploaded 327 items (photographs, videos) onto the interactive platform. For the intervention group (n = 106), the final mean (SD) score of nutrition literacy was significantly higher than the baseline: 78.8% (15.6) vs 72.7% (16.2); P < .001, regardless of parental education and perceived income status. No significant differences in the scores of the control group (n = 83) were observed (final 67.8% [16.1] vs initial 66.4% [15.6]; P = .364).

CONCLUSIONS AND IMPLICATIONS

Gamified digital interactive platform seems to be a useful, easily adapted educational tool for the healthy eating learning process. Future implementations of the program will benefit from longer time intervention and assessment of the eating habits of families before and after intervention.

CEMP-1 Levels in Periodontal Wound Fluid during the Early Phase of Healing: Prospective Clinical Trial

Objectives

Cementogenesis seems to be significantly compromised during tissue inflammation. In dental practice, surgical procedures are performed with the aim to regenerate periodontium including cementum. However, inflammation that occurs during the initial healing phases after surgery may impair regeneration of this tissues. The aim of the present study was to assess if surgical procedures designed to regenerate periodontium might affect levels of cementum protein-1 (CEMP-1) in periodontal wound fluid during early phase of healing.

Materials and Methods

In 36 patients, 18 intrabony periodontal defects were treated with regenerative therapy (REG group) and 18 suprabony periodontal defects were treated with open flap debridement (OFD group). In the experimental sites, gingival crevicular fluid was collected immediately before surgery, and periodontal wound fluid was collected 4, 7, 14, and 21 days after surgery. CEMP-1 levels were detected by indirect enzyme-linked immunosorbent assay technique.

Results

At the analysis, it resulted that there was a significant average difference in CEMP-1 values between the REG and OFD groups at baseline (p = 0.041), the CEMP-1-modeled average in the OFD group was lower by 0.45 ng/ml. There was a significant trend in CEMP-1 over time, and this trend was different among the 2 groups: the REG group showed a statistically significant rising CEMP-1 trend (0.18 ng/ml a week p = 0.012), while the OFD had a trend that was significantly lower (-0.22 ng/ml a week compared to the REG group trend p = 0.023), the OFD group lost on average 0.05 ng/ml a week. In REG sites, GCF protein levels resulted also related to clinical parameters.

Conclusions

During the initial inflammatory phase of periodontal healing, CEMP-1 levels decrease regardless of the surgical protocol applied. The surgical procedures used to regenerate periodontal tissue are able to reverse this trend and to induce significant increase of CEMP-1 in periodontal wound fluid after the first week postop.