Category: Statistics

Eur Radiol. 2025 Oct 11. doi: 10.1007/s00330-025-12048-4. Online ahead of print.

ABSTRACT

OBJECTIVES: Fractional flow reserve (FFR) and instantaneous wave-Free Ratio (iFR) pressure measurements during invasive coronary angiography (ICA) are the gold standard for assessing vessel-specific ischemia. Artificial intelligence has emerged to compute FFR based on coronary computed tomography angiography (CCTA) images (CT-FFR_AI). We assessed a CT-FFR_AI deep learning model for the prediction of vessel-specific ischemia compared to invasive FFR/iFR measurements.

MATERIALS AND METHODS: We retrospectively selected 322 vessels from 275 patients at two centers who underwent CCTA and invasive FFR and/or iFR measurements during ICA within three months. A junior and senior radiologist at each center supervised vessel centerline-building to generate curvilinear reformats that were processed for CT-FFR_AI binary outcomes (≤ 0.80 or > 0.80) prediction. Reliability for CT-FFR_AI outcomes based on radiologists’ supervision was assessed with Cohen’s κ. Diagnostic values of CT-FFR_AI were calculated using invasive FFR ≤ 0.80 (n = 224) and invasive iFR ≤ 0.89 (n = 238) as the gold standard. A multinomial logistic regression model, including all false-positive and false-negative cases, assessed the impact of patient- and CCTA-related factors on diagnostic values of CT-FFR_AI.

RESULTS: Concordance for CT-FFR_AI binary outcomes was substantial (κ = 0.725, p < 0.001). Sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy of CT-FFR_AI in predicting vessel-specific ischemia on a per-vessel analysis, based on senior radiologists’ evaluations, were 85% (58/68) and 91% (78/86), 82% (128/156) and 78% (119/152), 67% (58/86) and 70% (78/111), 93% (128/138) and 94% (119/127), and 83% (186/224) and 83% (197/238), respectively. Coronary calcifications significantly reduced the diagnostic accuracy of CT-FFR_AI (p < 0.001; OR, 1.002; 95% CI 1.001-1.003).

CONCLUSION: CT-FFR_AI demonstrates high diagnostic performance in predicting vessel-specific coronary ischemia compared to invasive FFR and iFR. Coronary calcifications negatively affect specificity, suggesting that further improvements in spatial resolution could enhance accuracy.

KEY POINTS: Question How accurately can a new deep learning model (CT-FFR_AI) assess vessel-specific ischemia from CCTA non-invasively compared to two validated pressure measurements during invasive coronary angiography? Findings CT-FFR_AI achieved high diagnostic accuracy in predicting vessel-specific ischemia, with high sensitivity and negative predictive value, independent of scanner type and radiologists’ experience. Clinical relevance CT-FFR_AI provides a non-invasive alternative to Fractional Flow Reserve and instantaneous wave-Free Ratio measurements during invasive coronary angiography for detecting vessel-specific ischemia, potentially reducing the need for invasive procedures, lowering healthcare costs, and improving patient safety.

PMID:41076470 | DOI:10.1007/s00330-025-12048-4

J Robot Surg. 2025 Oct 11;19(1):681. doi: 10.1007/s11701-025-02884-3.

ABSTRACT

The study design is retrospective study. Robotic spine surgery has evolved rapidly over the past two decades, with growing adoption. However, research on error sources in robotic-assisted pedicle screw placement remains scarce. This study investigates the incidence, risk factors, and intraoperative error types linked to screw malposition. We retrospectively analyzed patients who underwent robotic-assisted thoracic and lumbar spine surgeries at our center since Oct 2023. The incidence of screw breaches, intraoperative errors, and risk factors, such as BMI, pathology, and surgical approach, were assessed. Screw malpositions were identified on postoperative O-arm scans and classified by error cause. Intraoperative revision protocols were documented. A total of 1060 patients (5644 pedicle screws) were included. Screw malposition occurred in 13 patients (1.2%) involving 22 screws (0.39%). The mean BMI of patients with malposition was 27.3 versus 26.5 overall. Among the 13 cases, 8 had lumbar degenerative pathology, 3 scoliosis, 1 high-grade listhesis, and 1 revision for proximal junctional failure. Skiving was the most frequent error (6 cases), followed by arm shift (4), patient movement (2), and registration error (1). Learning curve had a significant role with 12/22 screw malpositioning happening in the initial 50 robotic cases. No significant correlation was found between malposition and pathology type or surgical approach. Robotic systems improve pedicle screw accuracy but are not immune to errors. Skiving is the most common issue and can be minimized by selecting flat bony entry points and avoiding sloping surfaces. Other causes include arm shift, patient movement, and registration errors.

PMID:41076469 | DOI:10.1007/s11701-025-02884-3

Clin Rheumatol. 2025 Oct 11. doi: 10.1007/s10067-025-07722-3. Online ahead of print.

ABSTRACT

AIM: This study aimed to compare talar cartilage thickness in patients with RA and healthy controls and to investigate its association with clinical and demographic variables.

METHOD: Thirty-seven healthy controls and 63 patients with RA diagnosed using the American College of Rheumatology’s (ACR) 2010 criteria were included in this cross-sectional observational study. All participants’ age, gender, and body mass index (BMI) were recorded. The hospital record system’s data was retrieved for the patient group, including medication use, C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), seropositivity, and disease duration. VAS was used to assess pain, and DAS-28 was used to assess disease activity. Talar cartilage thickness measurement was performed by the same doctor as the ultrasound. Descriptive statistics of the data obtained from the study were given by mean, standard deviation for numerical variables, and frequency and percentage analysis for categorical variables. Mann-Whitney U test was used for categorical variables with two groups, and Kruskal Wallis test was used for categorical variables with three or more groups in the comparison of parameters according to categorical variables. Analyses were performed with the help of SPSS 22.0 program. p < 0.05 significance level was selected.

RESULTS: There was no discernible difference between the patient and control groups in terms of age, height, weight, gender, or BMI values (p > 0.05). Talar cartilage thickness did not significantly correlate with age, BMI, ESR, CRP, CCP, DAS-28 RF, number of swollen joints, or sensitive joints (p > 0,05). However, talar cartilage thickness, VAS, HAQ values, and disease duration were significantly correlated negatively (p < 0.05).

CONCLUSION: Our study showed that there is a significant decrease in talar cartilage thickness in RA patients, and this decrease is associated with disease duration, pain severity, and functional impairment. Key Points • Talar cartilage thickness was found to be lower in RA patients compared to the control group • Talar cartilage measurements were negatively correlated with disease duration, VAS and HAQ scores • Talar cartilage thickness may be a potential biomarker for assessing early joint damage and monitoring disease progression in RA.

PMID:41076467 | DOI:10.1007/s10067-025-07722-3

J Robot Surg. 2025 Oct 11;19(1):680. doi: 10.1007/s11701-025-02828-x.

ABSTRACT

Robotic-assisted total knee arthroplasty (RA-TKA) aims to improve outcomes and reduce costs, but its effect on in-hospital direct costs remains unclear. This study compares in-hospital direct costs between RA-TKA and conventional TKA (C-TKA). A retrospective review of elective TKAs from 1/2018-5/2023 at a single center was performed. RA-TKA and C-TKA patients were matched by demographics and surgery date. Cost metrics included total, fixed, variable direct costs and indirect costs. Operative times, length of stay (LOS), and in-hospital costs were compared using independent t-tests, Mann-Whitney U, and Chi-square tests. Pearson correlations assessed relationships between costs and clinical metrics. A hierarchical multilinear model was used to control for demographics, operative time, and LOS to isolate the cost of robotic assistance. Of 2590 primary TKAs, 440 used robotic assistance. After matching, 788 patients remained (397 RA-TKA, 391 C-TKA). Preoperative Knee Injury and Osteoarthritis Outcome Score for Joint Replacement (KOOS JR) and Veterans Rand 12 Item Health Survey (VR-12) scores were similar. RA-TKAs had longer operative times (“cut-to-close”: 128 vs. 110 min; “room-in to room-out”: 202 vs. 182 min; p < 0.001) and longer LOS (2.39 vs. 1.56 days; p < 0.001). RA-TKAs had 17.8% higher total direct costs and 18.5% higher variable direct costs (p < 0.001). After adjusting for operative time and LOS, robotic assistance independently increased total direct cost by 4.2% of median C-TKA cost (p < 0.001). Robotics remained an independent predictor of total and variable direct cost (p < 0.001). RA-TKA incurs higher in-hospital direct costs than C-TKA, largely due to increased variable costs. Even after adjusting for operative time and LOS, robotic assistance independently contributes to higher total direct cost.

PMID:41076447 | DOI:10.1007/s11701-025-02828-x

J Prosthet Dent. 2025 Oct 10:S0022-3913(25)00727-9. doi: 10.1016/j.prosdent.2025.09.005. Online ahead of print.

ABSTRACT

STATEMENT OF PROBLEM: Immediate loading of dental implants in the esthetic zone combines implant placement and the interim restoration in a single stage, addressing the demand for reduced treatment duration and enhanced outcomes. The impact of prosthetic workflows, traditional versus digital, on marginal bone loss, esthetic success, and patient satisfaction remains unclear.

PURPOSE: The purpose of this systematic review and meta-analysis was to compare the clinical and patient-centered outcomes of traditional versus digital prosthetic workflows in the immediate loading of implants in the esthetic zone, focusing on marginal bone loss, esthetic outcomes, and patient satisfaction.

MATERIAL AND METHODS: Following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) 2020 guidelines, randomized controlled trials and non-randomized comparative studies published between January 2015 and January 2025 were identified through searches in the PubMed, Cochrane Library, and Scopus databases. Studies reporting on marginal bone loss, Pink and White Esthetic Scores, and patient satisfaction for immediate implant loading were included. Two reviewers independently extracted data and assessed the risk of bias. A qualitative and quantitative synthesis of findings was conducted. Review Manager (RevMan) was used to conduct the meta-analysis. The results were presented with a 95% confidence interval (CI), and the mean difference (MD) was calculated as the summary statistic for continuous outcomes.

RESULTS: Four studies, comprising 2 randomized controlled trials and 2 non-randomized prospective studies, satisfied the inclusion criteria. The meta-analysis showed a non-significant trend favoring digital workflows for patient satisfaction (SMD: 0.26, 95% CI: -0.01 to 0.53; P=.06, I²=0%). Digital workflows demonstrated significant improvements in esthetic outcomes, with higher Pink Esthetic Scores (MD: 0.14, 95% CI: 0.05 to 0.24; P=.003, I²=17%) and White Esthetic Scores (MD: 0.09, 95% CI: 0.01 to 0.17; P=.02, I²=0%). Marginal bone loss was significantly reduced with digital workflows (MD:-0.08, 95% CI: -0.15 to -0.01; P=.02, I²=0%), indicating superior precision compared to conventional workflows.

CONCLUSIONS: Digital workflows for implant placement showed significant advantages in esthetic outcomes and less marginal bone loss, with a trend toward higher patient satisfaction. These findings support the growing adoption of digital workflows in clinical dental practice.

PMID:41076437 | DOI:10.1016/j.prosdent.2025.09.005

J Prosthet Dent. 2025 Oct 10:S0022-3913(25)00764-4. doi: 10.1016/j.prosdent.2025.09.034. Online ahead of print.

ABSTRACT

STATEMENT OF PROBLEM: Existing evidence suggests that conventional and bulk-fill resin composites (RCs) do not fully restore the physiological fracture resistance of teeth with extensive MOD cavities, leaving posterior restorations susceptible to bulk fracture, particularly in structurally compromised or endodontically treated teeth. Whether short fiber-reinforced resin-based composites (SFRCs) can address these limitations and improve fracture resistance remains unclear.

PURPOSE: The purpose of this systematic review and network meta-analysis was to evaluate the fracture resistance of extensive direct restorations restored with SFRCs, either as a single component or combined with other materials, compared with bulk-fill RCs, conventional RC, and intact teeth.

MATERIAL AND METHODS: A systematic search was conducted in the PubMed, Scopus, Web of Science, and EMBASE databases for studies published through July 2025. In vitro studies evaluating fracture resistance in MOD cavities restored with SFRC compared with other restorative techniques were included. A random-effects network meta-analysis was performed using standardized mean differences (SMDs) with 95% confidence intervals (CIs). Heterogeneity (I² and τ²) and inconsistency were assessed using node-splitting and loop inconsistency models. The confidence in the results was evaluated using the confidence in network meta-analysis (CINeMA) framework.

RESULTS: Thirty studies met the inclusion criteria. SFRC with overlay materials demonstrated significantly higher fracture resistance than conventional RC (SMD=1.30; 95% CI: 0.50 to 2.09) but not significantly higher than bulk-fill RC (SMD=0.81; 95% CI:-0.26 to 1.88). SFRC alone also showed greater resistance than conventional RC (SMD=2.12; 95% CI:0.22 to 4.03). Notably, no statistically significant difference was found between SFRC alone and intact teeth (SMD=-0.74; 95% CI:-2.64 to 1.16). Confidence in comparisons ranged from low to moderate, primarily associated with concerns regarding imprecision and heterogeneity. Intact teeth exhibited the highest fracture resistance across all conditions (SUCRA: 95.5%).

CONCLUSIONS: SFRCs exhibited better fracture resistance compared with conventional RCs, while no significant difference was observed between SFRC-based restorations and bulk-fill RCs. SFRCs may serve as a clinically suitable alternative for extensive restorations. Future research should focus on optimizing material combinations and addressing methodological variability in in vitro studies.

PMID:41076436 | DOI:10.1016/j.prosdent.2025.09.034

Br J Oral Maxillofac Surg. 2025 Sep 23:S0266-4356(25)00243-8. doi: 10.1016/j.bjoms.2025.08.015. Online ahead of print.

ABSTRACT

This study evaluates the performance of six large language models (LLMs) in generating content relevant to oral and maxillofacial surgery (OMFS), focusing on their ability to provide accurate, comprehensive, and relevant information across five specific tasks. Each LLM was assessed based on its responses to five prompts: (1) postoperative instructions for third molar surgery; (2) a list of best-selling books on orthognathic surgery; (3) the most cited articles in OMFS; (4) novel ideas for systematic reviews; and (5) emerging trends in OMFS. Responses were scored for relevance, comprehensiveness, and accuracy using predefined criteria. Statistical analysis was performed using the Kruskal-Wallis test to compare tool performance. The LLMs performed similarly overall, with varying strengths and weaknesses. For postoperative instructions, they all provided comparable recommendations, though Perplexity underperformed. In identifying best-selling books, Gemini and Perplexity excelled, while ChatGPT and Copilot struggled with retrieving highly cited articles. Copilot and Claude were more effective in suggesting novel systematic review topics, while ChatGPT, Claude, Copilot, and DeepSeek identified emerging trends most accurately. LLMs demonstrate significant potential in supporting OMFS-related tasks, but their performance varies depending on the specific application. While they excel at synthesising existing information and identifying trends, limitations in accuracy and occasional hallucinations highlight the need for human oversight. These findings underscore the importance of integrating artificial intelligence (AI) as a supplementary tool in clinical, academic, and research settings, ensuring its use complements, rather than replaces, human expertise.

PMID:41076417 | DOI:10.1016/j.bjoms.2025.08.015

Br J Anaesth. 2025 Oct 10:S0007-0912(25)00602-6. doi: 10.1016/j.bja.2025.08.036. Online ahead of print.

ABSTRACT

BACKGROUND: This meta-analysis is the first to compare tracheal intubation conditions and haemodynamic responses produced by various types and doses of neuromuscular blocking agents (NMBAs) in paediatric anaesthesia while also exploring factors associated with variability in outcomes.

METHODS: Randomised controlled and controlled clinical trials involving healthy paediatric participants (0-12 yr) were included. Trials compared intubation conditions using various NMBA interventions or NMBA-free settings under direct laryngoscopy. Outcomes included odds ratios (ORs) for excellent and acceptable intubation conditions, and mean differences for MAP and HR. Bayesian network, pairwise, and cumulative meta-analyses, along with meta-regression, assessed NMBA effectiveness and covariate effects.

RESULTS: Data from 105 trials (8008 participants) were analysed. Suxamethonium ≥1.50 mg kg^-1 and rocuronium ≥0.90 mg kg^-1 provided similar intubation conditions, though not consistently within 60 s. Other NMBAs were, on average, slower and less effective. Opioids decreased MAP and HR but did not improve intubation conditions when combined with suxamethonium ≥1.00 mg kg^-1 or rocuronium ≥0.90 mg kg^-1. Non depolarising NMBAs enhanced excellent (OR: 2.97 [1.82-5.10]) and acceptable intubation conditions (OR: 2.29 [1.14-4.39]) more in younger children (1.64 [1.08-2.20] yr) than in older ones (5.53 [4.04-7.01] yr). Intubation without NMBAs was most difficult in neonates and infants, with conditions improving until about age 4 yr. Beyond this, the difference in intubation quality between groups with and without NMBAs increased with age, indicating a greater benefit of using NMBAs in older children. Values are mean (95% credible interval).

CONCLUSIONS: We present a meta-analytical approach to synthesise and consolidate evidence from previous research and demonstrate how neuromuscular blocking agent type and dose, intubation timing, age, and induction drugs affect the safety and efficacy of paediatric airway management. Low-to-moderate confidence can be assigned to the recommendations from this meta-analysis.

SYSTEMATIC REVIEW PROTOCOL: PROSPERO (CRD42018097146).

PMID:41076411 | DOI:10.1016/j.bja.2025.08.036

Clin Genitourin Cancer. 2025 Sep 11:102432. doi: 10.1016/j.clgc.2025.102432. Online ahead of print.

ABSTRACT

OBJECTIVE: To evaluate the risk of contralateral lymph node involvement (LNI) and the feasibility of safely performing unilateral pelvic lymph node dissection (PLND) in unfavorable-intermediate and high-risk prostate cancer (PCa) patients with cN0 status on preoperative ⁶⁸Ga-PSMA PET/CT, tumor involvement in single lobe, or dominant lobe involvement with worse tumor characteristics on biopsy.

MATERIAL AND METHODS: In this retrospective multicenter study were analyzed 768 patients who underwent RP and bilateral extended-PLND. Patients with cN0 status on PSMA PET/CT and PI-RADS ≥ 3 lesions on multiparametric magnetic resonance imaging(mpMRI) were included. Tumor lobe (single/dominant-prostatic lobe) involvement and LNI status were recorded for all patients. The dominant lobe was determined based on higher ISUP grade group(GG), number and percentage of more positive cores, and more advanced features on MRI, respectively. LNI status was analyzed by tumor side and location. Statistical analysis included univariate and multivariate models to evaluate predictors of contralateral LNI.

RESULTS: LNI was observed in 96(12.5%) of 768 patients, with 61(7.9%) having ipsilateral LNI and 35(4.6%) contralateral LNI with or without ipsilateral LNI. Patients with contralateral LNI had higher preoperative PSA, more frequent EAU high-risk classification, larger-index lesion diameter, higher ISUP GG on both the dominant and nondominant side, and a higher rate of positive percentages in the nondominant side (all P values < .05). Multivariate analysis identified preoperative PSA (HR 1.028, 95% Cl 1.001-1.057, P = .044), ISUP GG 2 (HR 4.325,95% Cl-1.620-14.374,P = .007) and ≥ ISUP-GG 3 (HR 14.004, 95% Cl 3.025-54.773, P < .001) on the nondominant side as independent predictors for contralateral LNI. The ROC-derived AUC for predicting contralateral LNI was 0.873, indicating good predictive accuracy.

CONCLUSION: In cases where preoperative ⁶⁸Ga-PSMA PET/CT indicates a negative LN status, contralateral PLND may not be necessary in intermediate-risk patients with negative biopsy or ISUP GG 1 tumor on the nondominant side.

PMID:41076397 | DOI:10.1016/j.clgc.2025.102432

Chin J Traumatol. 2025 Sep 17:S1008-1275(25)00114-2. doi: 10.1016/j.cjtee.2025.03.002. Online ahead of print.

ABSTRACT

PURPOSE: Hidden blood loss (HBL) during the perioperative period significantly impacts postoperative recovery and complications, yet it is frequently disregarded. This study aimed to investigate the effects of tip-apex distance (TAD) and calcar-referenced tip-apex distance (calTAD) on HBL in the treatment of intertrochanteric fractures utilizing proximal femoral nail antirotation (PFNA). The study also seeks to evaluate the possible decrease in HBL subsequent to PFNA treatment by optimizing nail positioning.

METHOD: A historical cohort study was conducted from January 2020 to December 2022. Patients diagnosed with unilateral acute closed femoral intertrochanteric fracture and who underwent PFNA internal fixation surgery met the inclusion criteria, and were grouped according to the value of calTAD and TAD. The participants were divided into low TAD group (TAD<20 mm) and high TAD group (TAD≥20 mm); low calTAD group (calTAD<7.625 mm) and high calTAD group (calTAD≥7.625 mm), respectively. The primary outcome measures were intraoperative blood loss (including HBL, overt blood loss, and total blood loss). Continuous data were analyzed using an independent sample t-test or Mann-Whitney U test, and categorical data were analyzed using the Pearson Chi-square test. Univariate analysis was used to evaluate the association between various indicators and perioperative HBL. A stepwise multiple linear regression analysis model was used to determine the independent factors affecting perioperative HBL. A p value less than 0.05 was considered statistically significant.

RESULTS: A total of 131 patients were initially included, of which 80 were assigned to the calTAD group (with 61 in the high calTAD group and 19 in the low calTAD group), and 80 were assigned to the TAD group (with 34 in the high TAD group and 46 in the low TAD group). The average HBL for the low TAD group was 772.85 mL, whereas for the high TAD group it was 919.68 mL (p>0.05). The average HBL for the low calTAD group was 611.42 mL, whereas for the high calTAD group it was 904.97 mL (p<0.05). Subsequent analysis revealed that the patient’s height, preoperative hemoglobin levels, changes in hemoglobin and hematocrit levels from pre- to post-surgery, and calTAD are independent risk factors influencing HBL.

CONCLUSION: In summary, our investigation revealed a significant correlation between the positioning of nails in PFNA and HBL during the perioperative period. By optimizing the placement of the cephalic nail, specifically by ensuring a calTAD of less than 7.625 mm, a significant decrease in HBL can be attained. Additionally, we identified that height, preoperative hemoglobin, differences in preoperative and postoperative hemoglobin and hematocrit, and the positioning of the cephalic nail were independent risk factors for HBL.

PMID:41076394 | DOI:10.1016/j.cjtee.2025.03.002