RESUMEN
Importance: Diagnosing solid lesions in the pancreas via endoscopic ultrasonographic (EUS) images is challenging. Artificial intelligence (AI) has the potential to help with such diagnosis, but existing AI models focus solely on a single modality. Objective: To advance the clinical diagnosis of solid lesions in the pancreas through developing a multimodal AI model integrating both clinical information and EUS images. Design, Setting, and Participants: In this randomized crossover trial conducted from January 1 to June 30, 2023, from 4 centers across China, 12 endoscopists of varying levels of expertise were randomly assigned to diagnose solid lesions in the pancreas with or without AI assistance. Endoscopic ultrasonographic images and clinical information of 439 patients from 1 institution who had solid lesions in the pancreas between January 1, 2014, and December 31, 2022, were collected to train and validate the joint-AI model, while 189 patients from 3 external institutions were used to evaluate the robustness and generalizability of the model. Intervention: Conventional or AI-assisted diagnosis of solid lesions in the pancreas. Main Outcomes and Measures: In the retrospective dataset, the performance of the joint-AI model was evaluated internally and externally. In the prospective dataset, diagnostic performance of the endoscopists with or without the AI assistance was compared. Results: The retrospective dataset included 628 patients (400 men [63.7%]; mean [SD] age, 57.7 [27.4] years) who underwent EUS procedures. A total of 130 patients (81 men [62.3%]; mean [SD] age, 58.4 [11.7] years) were prospectively recruited for the crossover trial. The area under the curve of the joint-AI model ranged from 0.996 (95% CI, 0.993-0.998) in the internal test dataset to 0.955 (95% CI, 0.940-0.968), 0.924 (95% CI, 0.888-0.955), and 0.976 (95% CI, 0.942-0.995) in the 3 external test datasets, respectively. The diagnostic accuracy of novice endoscopists was significantly enhanced with AI assistance (0.69 [95% CI, 0.61-0.76] vs 0.90 [95% CI, 0.83-0.94]; P < .001), and the supplementary interpretability information alleviated the skepticism of the experienced endoscopists. Conclusions and Relevance: In this randomized crossover trial of diagnosing solid lesions in the pancreas with or without AI assistance, the joint-AI model demonstrated positive human-AI interaction, which suggested its potential to facilitate a clinical diagnosis. Nevertheless, future randomized clinical trials are warranted. Trial Registration: ClinicalTrials.gov Identifier: NCT05476978.
Asunto(s)
Inteligencia Artificial , Estudios Cruzados , Humanos , Masculino , Femenino , Persona de Mediana Edad , Anciano , Endosonografía/métodos , Neoplasias Pancreáticas/diagnóstico , Neoplasias Pancreáticas/diagnóstico por imagen , Adulto , Páncreas/diagnóstico por imagen , China , Estudios RetrospectivosRESUMEN
INTRODUCTION: Early gastric cancer with current Helicobacter pylori infection (HpC-EGC) is common, but it is still unclear whether H. pylori eradication therapy (Hp-ET) or endoscopic submucosal dissection (ESD) should be performed first. We evaluated Hp-ETs short-term effects on horizontal boundary delineations of HpC-EGC in ESD. METHODS: Prospectively enrolled HpC-EGC patients were randomly assigned to eradication or control groups. Operation scopes of HpC-EGC lesions were delineated with marking dots at 5 mm out of the endoscopic demarcation line by an independent endoscopist, unaware of eradication status, before formal circumferential incision. As representatives, precise delineation rate, the shortest distance of all marking dots to the pathological demarcation line in all slices of one intact resected specimen (Dmin), and negative marking dot specimen rate were examined. RESULTS: Twenty-three HpC-EGC patients (25 lesions) were allocated to eradication group and 26 patients (27 lesions) were allocated to the control group with similar eradication success rates and all were differentiated type. With improving background mucosa inflammation after Hp-ET and similar gastritis-like epithelium rates, 10 lesions (40.0%) in the eradication group were of precise delineation compared to control group with 2 lesions (7.4%) (relative risk = 5.40, 95% CI 1.31-22.28). Dmin of eradication and control groups were 4.17 ± 2.52 mm and 2.67 ± 2.30 mm (p = 0.029), accompanied by 4 (14.8%) and none (0.0%) specimens that exhibited positive marking dots (p = 0.11), respectively. CONCLUSION: For HpC-EGC patients, administrating eradication medication before ESD is beneficial for the precise delineation of lesions and reducing the risk of positive horizontal resection margins.
Asunto(s)
Resección Endoscópica de la Mucosa , Infecciones por Helicobacter , Helicobacter pylori , Neoplasias Gástricas , Humanos , Infecciones por Helicobacter/complicaciones , Infecciones por Helicobacter/tratamiento farmacológico , Neoplasias Gástricas/cirugía , Neoplasias Gástricas/patología , Mucosa Gástrica/cirugía , Mucosa Gástrica/patologíaRESUMEN
Evidence comparing ultrasound endoscopy-guided fine-needle biopsy (EUS-FNB) with EUS-guided fine-needle aspiration (EUS-FNA) in deep-seated lymphoma tissue sampling is insufficient. This study aims to evaluate the diagnostic efficacy of immunohistochemistry (IHC) or flow cytometry (FCM) on specimens obtained from EUS-FNB and EUS-FNA in the diagnosis and staging of deep-seated lymphomas. This real-world, dual-center study prospectively evaluated all eligible specimens from patients who underwent EUS-FNB/FNA over an 8-year period. 53 patients were enrolled, with 23 patients in the EUS-FNB group and 30 patients in the EUS-FNA group. FNB yielded specimens with longer core tissues (0.80 mm [0.55, 1.00] vs. 0.45 mm [0.30, 0.50], p = 0.009) and higher scores of specimen adequacy [4 (3.75, 4.00) vs. 3 (1.00, 4.00), p = 0.025]. Overall analysis revealed that the diagnostic accuracy of IHC based on specimens acquired from EUS-FNB was significantly higher than that of EUS-FNA (91.30% vs. 60.00%, p = 0.013). After controlling confounding factors including lesion size and endoscopists, EUS-FNB with IHC maintained a higher-level diagnostic accuracy compared to EUS-FNA (OR = 1.292 [1.037-1.609], p = 0.023). When FCM was additionally used to analyze the specimen acquired from EUS-FNA, the diagnostic yield was significantly improved (ROC AUC: 0.733 vs. 0.550, p = 0.015), and the AUC of FNB alone or combined with FCM was 0.739 and 0.761. Conclusions: FNB needles generate higher histopathological diagnostic accuracy and specimen quality than FNA for the deep-seated lymphoma. Though the application of FCM significantly improves the diagnostic efficacy of EUS-FNA, FNB was still the preferred diagnostic modality with a shorter procedure time, comparable diagnostic accuracy, and better cost-effectiveness.
RESUMEN
BACKGROUND: Endoscopic ultrasound (EUS)-guided tissue acquisition (TA) by EUS-guided fine needle aspiration (FNA) or fine needle biopsy (FNB) is a standard diagnostic procedure for solid pancreatic lesions. Whether rapid on-site evaluation (ROSE) should be used to support EUS-TA remains controversial. Here we assessed the diagnostic performance of EUS-TA with or without self-ROSE for solid pancreatic masses. METHODS: Three hundred and seventy EUS-TA cases with self-ROSE and 244 cases without ROSE were retrospectively enrolled between August 2018 and June 2022. All procedures including ROSE were performed by the attending endoscopist. Clinical data, EUS characteristics, and diagnostic performance for distinguishing benign from malignant solid pancreatic masses including accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were compared between groups. RESULTS: Self-ROSE improved the diagnostic accuracy of solid pancreatic lesions by 16.7% in the EUS-TA group (p < 0.001) and by 18.9% in the EUS-FNA alone group (p < 0.001). Self-ROSE also improved the diagnostic sensitivity by 18.6% in the EUS-TA group (p < 0.001) and by 21.2% in the EUS-FNA alone group (p < 0.001). Improvements in the diagnostic accuracy by self-ROSE in the EUS-FNB group were not significant. 2.2 ± 0.7, 2.4 ± 0.9, 2.3 ± 0.7, 2.5 ± 0.9, 2.1 ± 0.6, and 2.1 ± 0.7 needle passes were required in the EUS-TA, EUS-FNA, and EUS-FNB with or without self-ROSE groups, respectively. CONCLUSIONS: Self-ROSE significantly improved the accuracy and sensitivity of EUS-FNA alone and EUS-TA diagnosis of solid pancreatic lesions and helped to reduce needle passes during the procedure. Whether self-ROSE benefits EUS-FNB and whether EUS-FNB alone is comparable to EUS-FNA with self-ROSE require further clarification.
Asunto(s)
Neoplasias Pancreáticas , Evaluación in Situ Rápida , Humanos , Estudios Retrospectivos , Biopsia por Aspiración con Aguja Fina Guiada por Ultrasonido Endoscópico/métodos , Páncreas/diagnóstico por imagen , Páncreas/patología , Neoplasias Pancreáticas/diagnóstico por imagen , Neoplasias Pancreáticas/patologíaRESUMEN
Objectives: The superiority of EUS-guided fine-needle biopsy (EUS-FNB) over fine-needle aspiration (FNA) remains controversial. This study aimed to compare the efficacy of FNB and FNA in immunohistochemistry (IHC)-required lesions, including, type 1 autoimmune pancreatitis (AIP), neuroendocrine tumor (NET), mesenchymal tumor, and lymphoma. Methods: In this multicenter study, specimens from all eligible patients who underwent EUS-FNB/FNA with these specific lesions were prospectively evaluated. Demographics, adequacy of specimens for IHC, diagnostic accuracy, and integrity of tissue were analyzed. Subgroup analysis and multivariate logistic regression were also performed to control confounders. Results: A total of 439 patients were included for analysis. Most lesion types were type 1 AIP (41.69%), followed by NET, mesenchymal tumor, and lymphoma. FNB yielded specimens with better adequacy for IHC (82.41% vs. 66.67%, P < 0.001) and higher diagnostic accuracy (74.37% vs. 55.42%, P < 0.001). The superiority of FNB over FNA in adequacy for IHC (odds ratio, 2.786 [1.515-5.291]) and diagnostic accuracy (odds ratio, 2.793 [1.645-4.808]) remained significant after control of confounders including needle size, lesion site, lesion size, and endoscopists. In subgroup analysis, FNB showed higher diagnostic accuracy in AIP and mesenchymal tumor, whereas no statistically significant difference was observed in NET and lymphoma. Conclusions: FNB was superior to FNA needles in obtaining tissues with better adequacy and integrity. These results suggest that FNB should be considered a first-line modality in the diagnosis of IHC-required lesions, especially AIP and mesenchymal tumor. However, a randomized controlled trial with larger sample size is needed to further confirm our findings.
RESUMEN
AIM: To present the prevalence of tonsillitis in synovitis, acne, pustulosis, hyperostosis and osteitis (SAPHO) patients, to compare the clinical characteristics and disease activities between SAPHO patients with and without tonsillitis and to preliminarily explore the efficacy of tonsillectomy in SAPHO syndrome. METHOD: A total of 58 SAPHO patients were included. Clinical data were collected, including demographic characteristics and acute phase reactants (erythrocyte sedimentation rate, high-sensitivity C-reactive protein). The visual analog scale (VAS), Palmoplantar Pustule Psoriasis Area Severity Index (PPPASI) and Nail Psoriasis Severity Index (NAPSI) were used to measure the severity of bone pain, skin lesions and nail lesions, respectively. Patients were referred to the otolaryngology department for tonsil examinations, including tonsil hypertrophy (grade ≥ 2), chronic congestion, inflammatory secretion and tonsil stones. The patients who underwent tonsillectomy were followed up after the surgery. RESULTS: A total of 67.2% of patients had tonsillitis. Patients with tonsillitis had markedly higher PPPASI (1.2 [0, 7.4] vs. 7.6 [1.75, 15.5], P = .018) and NAPSI (0 [0, 21] vs. 8 [3, 28], P = .032) scores. After tonsillectomy, the patients experienced significantly improved bone pain (VAS, 5 [4, 7] vs. 3 [1, 4], P = .034) and skin lesions (PPPASI, 16.2 [7.05, 18.35] vs 1.8 [0.7, 3.7], P = .028). CONCLUSION: Approximately 2/3 of SAPHO patients had tonsillitis. Patients with tonsillitis had more severe skin and nail lesions. Tonsillectomy might be associated with improved bone and skin symptoms in SAPHO patients. Future prospective controlled studies are warranted.
Asunto(s)
Síndrome de Hiperostosis Adquirido/epidemiología , Tonsilitis/epidemiología , Síndrome de Hiperostosis Adquirido/diagnóstico , Adulto , Beijing/epidemiología , Femenino , Humanos , Masculino , Persona de Mediana Edad , Prevalencia , Estudios Retrospectivos , Medición de Riesgo , Factores de Riesgo , Índice de Severidad de la Enfermedad , Tonsilectomía , Tonsilitis/diagnóstico , Tonsilitis/cirugía , Resultado del TratamientoRESUMEN
Importance: Nail involvement is common in synovitis, acne, pustulosis, hyperostosis, and osteitis (SAPHO) syndrome, which has a strong association with quality of life in patients with SAPHO. Tofacitinib is an oral Janus kinase inhibitor that has been previously shown to be effective for nail psoriasis. Objective: To assess the efficacy and safety of tofacitinib for the treatment of nail involvement in SAPHO syndrome. Interventions: Participants received tofacitinib, 5 mg, twice daily, for 12 weeks. Design, Setting, and Participants: This open-label, single-arm, prospective pilot study included 13 patients with SAPHO syndrome accompanied by nail lesions and active palmoplantar pustulosis who were recruited from Peking Union Medical College Hospital from September 2019 to December 2019. Follow-up was completed in March 2020. Analysis began March 2020. Main Outcomes and Measures: The primary end point was the percentage of the change from baseline in Nail Psoriasis Severity Index scores at week 12. Secondary end points included the percentage of the change from baseline in Palmoplantar Psoriasis Area and Severity Index scores, change from baseline in Visual Analogue Scale scores in global osteoarticular pain, Dermatology Life Quality Index scores, and inflammatory markers. Adverse events were recorded throughout the study. Results: Thirteen female Asian patients (means [SD] age, 39.7 [12.3] years) were included, all of whom completed the study. At week 12, significant improvements were observed in Nail Psoriasis Severity Index scores (median, -67% [interquartile range (IQR), -56% to -77%]; P < .001) and Palmoplantar Psoriasis Area and Severity Index scores (median, -71% [IQR, -58% to -78%]; P < .001). Significant improvement was also noted in Dermatology Life Quality Index scores (median, -12 [IQR, -8.5 to -15]; P < .001) at week 12. A significant decrease in Visual Analogue Scale scores in global osteoarticular pain was observed at week 8 (median, -4 [IQR, 0 to -5]; P = .02) but was not significant at week 12. Inflammatory marker levels were decreased, as indicated by erythrocyte sedimentation rate (median, -8 mm/h [IQR, -4 mm/h to -11 mm/h]; P < .001) and high-sensitivity C-reactive protein levels (median, -1.6 [IQR, -0.3 to -4.1]; P = .01). No severe adverse events were observed. Conclusions and Relevance: In this pilot study, tofacitinib yielded significant remission of nail lesions and palmoplantar psoriasis accompanied by an improvement in quality of life in patients with SAPHO syndrome. Additional follow-up studies to evaluate the long-term efficacy and safety of tofacitinib for nail involvement in SAPHO syndrome are warranted. Trial Registration: Chinese Clinical Trial Registry number: ChiCTR1900025941.