Purpose: We investigated the correlation between the number of examined lymph nodes (ELNs) and correct staging and long-term survival in non–small-cell lung cancer (NSCLC) by using large databases and determined the minimal threshold for the ELN count.Methods: Data from a Chinese multi-institutional registry and the US SEER database on stage I to IIIA resected NSCLC (2001 to 2008) were analyzed for the relationship between the ELN count and stage migration and overall survival (OS) by using multivariable models. The series of the mean positive LNs, odds ratios (ORs), and hazard ratios (HRs) were fitted with a LOWESS smoother, and the structural break points were determined by Chow test. The selected cut point was validated with the SEER 2009 cohort.Results: Although the distribution of ELN count differed between the Chinese registry (n = 5,706) and the SEER database (n = 38,806; median, 15 versus seven, respectively), both cohorts exhibited significantly proportional increases from N0 to N1 and N2 disease (SEER OR, 1.038; China OR, 1.012; both P < .001) and serial improvements in OS (N0 disease: SEER HR, 0.986; China HR, 0.981; both P < .001; N1 and N2 disease: SEER HR, 0.989; China HR, 0.984; both P < .001) as the ELN count increased after controlling for confounders. Cut point analysis showed a threshold ELN count of 16 in patients with declared node-negative disease, which were examined in the derivation cohorts (SEER 2001 to 2008 HR, 0.830; China HR, 0.738) and validated in the SEER 2009 cohort (HR, 0.837).Conclusion: A greater number of ELNs is associated with more-accurate node staging and better long-term survival of resected NSCLC. We recommend 16 ELNs as the cut point for evaluating the quality of LN examination or prognostic stratification postoperatively for patients with declared node-negative disease.

To construct a prediction model based on peritumoral radiomics signatures from CT images and investigate its efficiency in predicting early recurrence (ER) of hepatocellular carcinoma (HCC) after curative treatment. In total, 156 patients with primary HCC were randomly divided into the training cohort (109 patients) and the validation cohort (47 patients). From the pretreatment CT images, we extracted 3-phase two-dimensional images from the largest cross-sectional area of the tumor. A region of interest (ROI) was manually delineated around the lesion for tumoral radiomics (T-RO) feature extraction, and another ROI was outlined with an additional 2 cm peritumoral area for peritumoral radiomics (PT-RO) feature extraction. The least absolute shrinkage and selection operator (LASSO) logistic regression model was applied for feature selection and model construction. The T-RO and PT-RO models were constructed. In the validation cohort, the prediction efficiencies of the two models and peritumoral enhancement (PT-E) were evaluated qualitatively by receiver operating characteristic (ROC) curves, calibration curves and decision curves and quantitatively by area under the curve (AUC), the category-free net reclassification index (cfNRI) and integrated discrimination improvement values (IDI). By comparing AUC values, the prediction accuracy in the validation cohort was good for the PT-RO model (0.80 vs. 0.79, P = 0.47) but poor for the T-RO model (0.82 vs. 0.62, P < 0.01), which was significantly overfitted. In the validation cohort, the ROC curves, calibration curves and decision curves indicated that the PT-RO model had better calibration efficiency and provided greater clinical benefits. CfNRI indicated that the PT-RO model correctly reclassified 47% of ER patients and 32% of non-ER patients compared to the T-RO model (P < 0.01); additionally, the PT-RO model correctly reclassified 24% of ER patients and 41% of non-ER patients compared to PT-E (P = 0.02). IDI indicated that the PT-RO model could improve prediction accuracy by 0.22 (P < 0.01) compared to the T-RO model and by 0.20 (P = 0.01) compared to PT-E. The CT-based PT-RO model can effectively predict the ER of HCC and is more efficient than the T-RO model and the conventional imaging feature PT-E.

The evidence of the impact of traditional statistical (TS) and artificial intelligence (AI) tool interventions in clinical practice was limited. This study aimed to investigate the clinical impact and quality of randomized controlled trials (RCTs) involving interventions evaluating TS, machine learning (ML), and deep learning (DL) prediction tools. A systematic review on PubMed was conducted to identify RCTs involving TS/ML/DL tool interventions in the past decade. A total of 65 RCTs from 26,082 records were included. A majority of them had model development studies and generally good performance was achieved. The function of TS and ML tools in the RCTs mainly included assistive treatment decisions, assistive diagnosis, and risk stratification, but DL trials were only conducted for assistive diagnosis. Nearly two-fifths of the trial interventions showed no clinical benefit compared to standard care. Though DL and ML interventions achieved higher rates of positive results than TS in the RCTs, in trials with low risk of bias (17/65) the advantage of DL to TS was reduced while the advantage of ML to TS disappeared. The current applications of DL were not yet fully spread performed in medicine. It is predictable that DL will integrate more complex clinical problems than ML and TS tools in the future. Therefore, rigorous studies are required before the clinical application of these tools.

Results concerning the association between peritoneal dialysis-related peritonitis and mortality in peritoneal dialysis patients are inconclusive, with one potential reason being that the time-dependent effect of peritonitis has rarely been considered in previous studies. This study aimed to evaluate whether peritonitis has a negative impact on mortality in a large cohort of peritoneal dialysis patients. We also assessed the changing impact of peritonitis on patient mortality with respect to duration of follow-up. This retrospective cohort study included incident patients who started peritoneal dialysis from 1 January 2006 to 31 December 2011. Episodes of peritonitis were recorded at the time of onset, and peritonitis was parameterized as a time-dependent variable for analysis. We used the Cox regression model to assess whether peritonitis has a negative impact on mortality. A total of 1321 patients were included. The mean age was 48.1 ± 15.3 years, 41.3% were female, and 23.5% with diabetes mellitus. The median (interquartile) follow-up time was 34 (21–48) months. After adjusting for confounders, peritonitis was independently associated with 95% increased risk of all-cause mortality (hazard ratio, 1.95; 95% confidence interval: 1.46–2.60), 90% increased risk of cardiovascular mortality (hazard ratio, 1.90; 95% confidence interval: 1.28–2.81) and near 4-fold increased risk of infection-related mortality (hazard ratio, 4.94; 95% confidence interval: 2.47–9.86). Further analyses showed that peritonitis was not significantly associated with mortality within 2 years of peritoneal dialysis initiation, but strongly influenced mortality in patients dialysed longer than 2 years. Peritonitis was independently associated with higher risk of all-cause, cardiovascular and infection-related mortality in peritoneal dialysis patients, and its impact on mortality was more significant in patients with longer peritoneal dialysis duration.