Factors Associated With Loss to Follow-Up in Surgical Trials: A Systematic Review and Meta-Analysis.

INTRODUCTION: Loss to follow-up (LTFU) distorts results of randomized controlled trials (RCTs). Understanding trial characteristics that contribute to LTFU may enable investigators to anticipate the extent of LTFU and plan retention strategies. The objective of this systematic review and meta-analysis was to investigate the extent of LTFU in surgical RCTs and evaluate associations between trial characteristics and LTFU. METHODS: MEDLINE, Embase, and PubMed Central were searched for surgical RCTs published between January 2002 and December 2021 in the 30 highest impact factor surgical journals. Two-hundred eligible RCTs were randomly selected. The pooled LTFU rate was estimated using random intercept Poisson regression. Associations between trial characteristics and LTFU were assessed using metaregression. RESULTS: The 200 RCTs included 37,914 participants and 1307 LTFU events. The pooled LTFU rate was 3.10 participants per 100 patient-years (95% confidence interval [CI] 1.85-5.17). Trial characteristics associated with reduced LTFU were standard-of-care outcome assessments (rate ratio [RR] 0.17; 95% CI 0.06-0.48), surgery for transplantation (RR 0.08; 95% CI 0.01-0.43), and surgery for cancer (RR 0.10; 95% CI 0.02-0.53). Increased LTFU was associated with patient-reported outcomes (RR 14.21; 95% CI 4.82-41.91) and follow-up duration ≥ three months (odds ratio 10.09; 95% CI 4.79-21.28). CONCLUSIONS: LTFU in surgical RCTs is uncommon. Participants may be at increased risk of LTFU in trials with outcomes assessed beyond the standard of care, surgical indications other than cancer or transplant, patient-reported outcomes, and longer follow-up. Investigators should consider the impact of design on LTFU and plan retention strategies accordingly.

Predicting adverse outcomes in pregnant patients positive for SARS-CoV-2: a machine learning approach- a retrospective cohort study.

BACKGROUND: Pregnant people are particularly vulnerable to SARS-CoV-2 infection and to ensuing severe illness. Predicting adverse maternal and perinatal outcomes could aid clinicians in deciding on hospital admission and early initiation of treatment in affected individuals, streamlining the triaging processes. METHODS: An international repository of 1501 SARS-CoV-2-positive cases in pregnancy was created, consisting of demographic variables, patient comorbidities, laboratory markers, respiratory parameters, and COVID-19-related symptoms. Data were filtered, preprocessed, and feature selection methods were used to obtain the optimal feature subset for training a variety of machine learning models to predict maternal or fetal/neonatal death or critical illness. RESULTS: The Random Forest model demonstrated the best performance among the trained models, correctly identifying 83.3% of the high-risk patients and 92.5% of the low-risk patients, with an overall accuracy of 89.0%, an AUC of 0.90 (95% Confidence Interval 0.83 to 0.95), and a recall, precision, and F1 score of 0.85, 0.94, and 0.89, respectively. This was achieved using a feature subset of 25 features containing patient characteristics, symptoms, clinical signs, and laboratory markers. These included maternal BMI, gravidity, parity, existence of pre-existing conditions, nicotine exposure, anti-hypertensive medication administration, fetal malformations, antenatal corticosteroid administration, presence of dyspnea, sore throat, fever, fatigue, duration of symptom phase, existence of COVID-19-related pneumonia, need for maternal oxygen administration, disease-related inpatient treatment, and lab markers including sFLT-1/PlGF ratio, platelet count, and LDH. CONCLUSIONS: We present the first COVID-19 prognostication pipeline specifically for pregnant patients while utilizing a large SARS-CoV-2 in pregnancy data repository. Our model accurately identifies those at risk of severe illness or clinical deterioration, presenting a promising tool for advancing personalized medicine in pregnant patients with COVID-19.

A new formula for estimating gestational age by crown-rump length.

BACKGROUND: The accurate estimation of gestational age by ultrasound is crucial in prenatal care for the monitoring of fetal growth and development. As changes in maternal childbearing age, body habitus, and ultrasound technology occur, previously published formulas may not be accurate for today's population. OBJECTIVE: This study aimed to develop new formulas for calculating the gestational age based on a first-trimester ultrasound scan and to compare the new formulas with preexisting formulas. STUDY DESIGN: This study was a single-center, retrospective observational study that included pregnancies conceived using in vitro fertilization. The pregnancies had known dates of embryo transfer and multiple standard ultrasound examinations in the first trimester of pregnancy. Pregnancies ending with a miscarriage or termination in the first trimester of pregnancy were excluded. A polynomial regression analysis was performed to determine the optimal model that represented the relationship between gestational age and crown-rump length. The models were evaluated using systematic error, random error, absolute difference of the calculated gestational age and actual gestational age, and proportion of estimation within 0 and 2 days of the known gestational age. The optimal model was chosen and compared with preexisting formulas. RESULTS: Overall, 1436 ultrasound results were included in the analysis. The analysis produced 3 models: linear, cubic, and quadratic models with correlation coefficients of 0.968, 0.989, and 0.991, respectively. The cubic formula was superior to the linear and quadratic formulas concerning systematic error, random error, absolute difference, and proportion of estimation within 0 and 2 days. The new formula had a lower systematic error, random error, and mean absolute difference (0.06%, 2.43%, and 0.97 days, respectively) and the highest proportion of estimation within 0 and 2 days (37.4% and 93.5%, respectively) than previously published formulas. CONCLUSION: The formula proposed in this study followed a cubic model and seemed to be able to more accurately estimate gestational age in the first trimester of pregnancy based on crown-rump length compared with previously published formulas.

Predictors of 1-year mortality after adult lung transplantation: Systematic review and meta-analyses.

BACKGROUND: Prognostic factors in lung transplantation are those variables that are associated with transplant outcomes. Knowledge of donor and recipient prognostic variables can aid in the optimal allocation of donor lungs to transplant recipients and can also inform post-operative discussions with patients about prognosis. Current research findings related to prognostic factors in lung transplantation are inconsistent and the relative importance of various factors is unclear. This review aims to provide the best possible estimates of the association between putative prognostic variables and 1-year all-cause mortality in adult lung transplant recipients. METHODS: We searched 5 bibliographic databases for studies assessing the associations between putative predictors (related to lung donors, recipients, or the transplant procedure) and 1-year recipient mortality. We pooled data across studies when justified and utilized GRADE methodology to assess the certainty in the evidence. RESULTS: From 72 eligible studies (2002-2020), there were 34 recipient variables, 4 donor variables, 10 procedural variables, and 7 post-transplant complication variables that were amenable to a meta-analysis. With a high degree of certainty in the evidence only post-transplant need for extra-corporeal membrane oxygenation (ECMO) (HR 1.91, 95% CI 1.79-2.04) predicted 1-year mortality. No donor variables appeared to predict transplant outcome with high or even moderate certainty. CONCLUSION: Across the range of contemporary donors and recipients that clinicians accept for lung transplantation, this review, with high certainty, found 1 prognostic factor that predicted 1-year mortality, and 37 additional factors with a moderate degree of certainty. The lack of prognostic significance for some widely accepted factors (e.g., donor smoking, age) likely relates to existing limits in the range of these variables at the time of donor and recipient selection.

Universal masking to prevent SARS-CoV-2 transmission from Taiwan's practices.

In the fight against the COVID-19 pandemic, Taiwan, with its universal masking policy, slowed down the spread of cases and flattened its epidemic curve without enforcing lockdown or mass quarantine in 2020. This study identifies the distinguishing features of Taiwan's universal masking policy practice, such as priority, continuous improvement, multi-stakeholder partnership, transparency and accountability, and altruism and social solidarity. By confronting uncertainty through the COVID-19 crisis, this study suggests that face masking, rather than being just a physical barrier of non-pharmacological intervention, can be adopted as an interactive policy platform to empower the public for stimulating cross-sector collaboration towards social innovation and creating spillover effects, such as acts of public trust, altruism, and solidarity.