Purpose: High glucose concentrations and swings are associated with endothelial dysfunction. We examined the effects of variability in fasting plasma glucose on peripheral artery disease (PAD) in patients with diabetes mellitus (DM). Patients and Methods: In this screening study for the risk factors of PAD, we retrospectively collected data on the ankle-brachial index (ABI) and the percentage of mean arterial pressure (%MAP) at the ankle between August 01, 2016 and July 31, 2017. We defined low ABI ≤ 0.90, high %MAP ≥ 45%, or both as high-risk PAD and others as low-risk PAD. We compared the standard deviation (SD) of the first fasting plasma glucose data available each year after January 01, 2007. Results: In 2577 patients, a higher SD of annual fasting glucose was observed in those with an ABI ≤ 0.90 than in patients with an ABI > 0.90 (2.6 ± 2.1 vs 2.2 ± 2.3, P = 0.009), and in patients with %MAP ≥ 45% than in those with %MAP < 45% (2.4 ± 2.1 vs 2.2 ± 2.3, P = 0.034). A high-risk PAD was significantly associated with the SD (P = 0.032) but not with the mean (P = 0.338) of annual fasting glucose. The former was an independent risk factor for high-risk PAD (odds ratio = 1.424; 95% CI = 1.118‒1.814; P = 0.004). Conclusion: Variability but not mean of annual fasting plasma glucose was significantly associated with a high risk of PAD in patients with DM.

Background: No systematic review has assessed genicular nerve blocks (GNB) efficacy in total knee arthroplasty (TKA) widely. This study evaluated its effectiveness in patients undergoing TKA. Methods: We conducted a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)-guided systematic review and meta-analysis (PROSPERO CRD42024593742) of randomized controlled trials on GNB in TKA. Searches of PubMed, Embase, and Cochrane Library (July 2014–July 2024) included English-only studies. Risk of bias was assessed using Cochrane 2.0, and random-effects models were applied. No funding was received. Results: This meta-analysis included 6 randomized controlled trials (335 patients; 169 GNB). Primary outcomes were pain scores (Visual Analog Scale), morphine use, and knee flexion. GNB significantly lowered pain at all time points: 4 hours (rest mean difference [MD]: –3.38, 95% confidence interval [CI]: –6.62 to –0.15, P = .04; activity MD: –3.91, 95% CI: –4.16 to –3.66, P < .00001), 8 hours (rest MD: –2.88, P = .01; activity MD: –3.58, P = .01), 12 hours (rest MD: –2.27, P = .004; activity MD: –2.33, P = .02), and 24 hours (rest MD: –1.40, P < .00001; activity MD: –2.42, P < .00001). However, GNB did not significantly affect knee flexion (MD: 19.86, 95% CI: –7.26 to 46.98, P = .15) or morphine use (SMD: 0.36, 95% CI: –2.22 to 2.93, P = .79). Subgroup analysis confirmed pain reduction with GNB alone or combined with other blocks at 12 and 24 hours: rest (MD: –1.10, P = .0001; MD: –2.85, P < .0001) and activity (MD: –1.30, P < .0001; MD: –4.00, P < .00001). Conclusion: GNB reduces early postoperative pain after TKA without improving function or opioid use. Limitations include few trials, English-only data, short follow-up, and no complication assessment. Further research should confirm long-term efficacy and safety.

Background and Objectives: Knee osteoarthritis (KOA) is a major cause of global disability. The efficacy of a non-invasive treatment, pulsed electromagnetic field (PEMF) therapy, remains debated. This systematic review and meta-analysis evaluate PEMF’s effectiveness on KOA, exploring the influence of device parameters. Materials and Methods: We systematically searched PubMed, Embase, and the Cochrane Library for randomized controlled trials (RCTs) from 2015 to 2025. Nine RCTs with a total of 457 patients were included. Primary outcomes were pain (Visual Analog Scale—VAS) and function (Western Ontario and McMaster Universities Osteoarthritis Index—WOMAC). Data were pooled using a random-effects model with subgroup analyses based on PEMF amplitude and frequency. Results: No significant improvement in VAS pain or total WOMAC scores was found at one month. However, time-dependent effects were observed. WOMAC-pain improved significantly at 18–21 days (MD = −1.63, 95% CI: −2.43 to −0.82, I² = 28%) but not at one month. Conversely, WOMAC-stiffness (MD = −1.11, 95% CI: −1.386 to −0.85, I² = 0%) and daily activity (MD = −3.39, 95% CI: −4.81 to −1.97, I² = 0%) improved significantly only at the one-month. Objective functional measures did not improve, and the overall risk of bias across studies was high. The efficacy of PEMF is also influenced by the amplitude and frequency. Conclusions: PEMF efficacy for KOA is nuanced, with benefits dependent on timing and device parameters. High frequency gives fast pain relief; high amplitude builds function. Though statistically significant, these improvements may not reach thresholds for clinical meaningfulness. Significant heterogeneity in treatment protocols is a major barrier to clear conclusions. Standardized, large-scale RCTs are needed to determine optimal parameters and confirm PEMF’s clinical role.

BACKGROUND Nearly half of children with ADHD experience comorbid language difficulties. While SNAP-IV behavioral assessment is routinely administered to all children with ADHD, comprehensive language evaluation is not universally conducted due to time and resource constraints, resulting in many children with language difficulties going undetected. OBJECTIVE To develop a clinical screening tool that leverages routinely collected SNAP-IV data to identify ADHD children at risk for language difficulties, thereby reducing missed cases. We used interpretable machine learning to identify key predictors and designed a two-tier screening workflow. METHODS We analyzed data from 188 children with ADHD (88.3% male, mean age 10.47 years). Language difficulty was defined as GCC < 86 (48.4% prevalence). Random Forest models were developed using SNAP-IV items, executive function measures, and demographics. SHAP (SHapley Additive exPlanations) analysis quantified feature contributions. A two-tier screening workflow was designed and evaluated. RESULTS The model achieved AUC 0.853. SHAP analysis revealed that inattention symptoms were substantially more predictive than hyperactivity symptoms. The top predictors were: SNAP-3 ("mind elsewhere during conversation"), planning ability, and SNAP-5 ("difficulty organizing"). A two-tier workflow combining quick rule-out (4-item score) with ML-based risk stratification achieved 90.1% sensitivity while reducing unnecessary referrals by 24.5%. CONCLUSIONS Inattention symptoms and planning deficits are key predictors of language difficulties in ADHD. A clinical decision support system integrating ML-based screening can systematically identify at-risk children while optimizing clinical resources. A web-based application is under development for prospective validation.