Search

Various classification systems for tear trough deformity (TTD) have been published; however, their complexity can pose challenges in clinical use, especially for less experienced surgeons. It is believed that artificial intelligence (AI) technology can address some of these challenges by reducing inadvertent errors and improving the accuracy of medical practice. In this study, we aimed to establish a reliable and precise digital image grading model for TTD using smartphone-based photography enhanced using AI deep learning technology. This model is designed to aid and guide surgeons, particularly those who are less experienced or from younger generations, during clinical examinations and in making decisions regarding further surgical interventions. A total of 504 patients and 983 photos were included in the study. We adopted the Barton's grading system for TTD. All photos were taken using the same smartphone and processed and analyzed using the medical AI assistant (MAIA™) software. The photos were then randomly divided into two groups to establish training and testing models. The confusion matrix for the training model demonstrated a sensitivity of 56%, specificity of 87.3%, F1 score of 0.57, and an area under the curve (AUROC) of 0.85. For the testing group, the sensitivity was 49.3%, specificity was 85%, F1 score was 0.49, and AUROC was 0.83. Representative heatmaps were also generated. Our study is the first to demonstrate that tear trough deformities can be easily categorized using a built-in smartphone camera in conjunction with an AI deep learning program. This approach can reduce errors during clinical patient evaluations, particularly for less experienced practitioners. [ABSTRACT FROM AUTHOR]

• After adjusting for age, sex, ASPECTS, tPA, premorbid mRS, admission NIH stroke scale, prior history of TIA, stroke, atrial fibrillation, diabetes mellitus, hyperlipidemia, heart disease and hypertension, admission LAMS was found to be independently associated with CBV Index (adjusted OR:0.82, p < 0.01). • After adjusting for age, sex, ASPECTS, tPA, premorbid mRS, admission NIH stroke scale, prior history of TIA, stroke, atrial fibrillation, diabetes mellitus, hyperlipidemia, heart disease and hypertension, admission LAMS was found to be independently associated with and HIR (adjusted OR:1.70, p < 0.05). • Admission LAMS is independently associated with CTP CS markers, CBV index and HIR. This finding suggests that LAMS may also provide an indirect estimate of CS, among other factors. The Los Angeles Motor Scale (LAMS) is an objective tool that has been used to rapidly assess and predict the presence of large vessel occlusion (LVO) in the pre-hospital setting successfully in several studies. However, studies assessing the relationship between LAMS score and CT perfusion collateral status (CS) markers such as cerebral blood volume (CBV) index, and hypoperfusion intensity ratio (HIR) are sparse. Our study therefore aims to assess the association of admission LAMS score with established CTP CS markers CBV Index and HIR in AIS-LVO cases. In this prospectively collected, retrospectively reviewed analysis, inclusion criteria were as follows: a) CT angiography (CTA) confirmed anterior circulation LVO from 9/1/2017 to 10/01/2023, and b) diagnostic CT perfusion (CTP). Logistic regression analysis was performed to assess the relationship between admission LAMS with CTP CS markers HIR and CBV Index. p ≤ 0.05 was considered significant. In total, 285 consecutive patients (median age = 69 years; 56 % female) met our inclusion criteria. Multivariable logistic regression analysis adjusting for sex, age, ASPECTS, tPA, premorbid mRS, admission NIH stroke scale, prior history of TIA, stroke, atrial fibrillation, diabetes mellitus, hyperlipidemia, coronary artery disease and hypertension, admission LAMS was found to be independently associated with CBV Index (adjusted OR:0.82, p < 0.01), and HIR (adjusted OR:0.59, p < 0.05). LAMS is independently associated with CTP CS markers, CBV index and HIR. This finding suggests that LAMS may also provide an indirect estimate of CS. [ABSTRACT FROM AUTHOR]

A 55-year-old female was referred to an oral pathology/oral medicine specialty clinic by her general dentist for evaluation of multiple intraoral white lesions. Upon examination, the following were detected: a well-defined, flat, white lesion on the left maxillary palatal gingiva; bilateral diffuse vestibular white lesions extending to the adjacent buccal gingiva (Fig. 1A-1C); bilateral white lesions on the hard palate with mild telangiectasia; and a thin curvilinear white lesion extending to mucogingival margin on the anterior maxillary and mandibular labial gingiva with gingival recession (Fig. 1D-1E). There was no edema or tenderness associated with any of the lesions. She denied any history of surgery, infection, or injury to the affected sites. The patient did not wear any removable oral appliance at that time. A longstanding history of dry mouth sensation was endorsed without anesthesia, paresthesia, swelling, ulceration, or bleeding. She had an active history of parafunctional habits including grinding and clenching. Her oral care was regular involving dental visits every 6 months, she brushed with fluoridated toothpaste twice per day, she did not floss, and she did not use mouthwash. The patient denied currently smoking or chewing tobacco and betel quid use. However, she drank one serving of alcohol per week and quit cigarette smoking 23 years ago. The patient's medical history was significant for psoriasis, depression, and longstanding thickened cutaneous plaques at the neck, left shoulder, chest, and right abdomen (Fig. 1F). She reported recent hair loss with three alopecia patches (Fig. 1G). There was also a reported a history of worsening stiffness in her fingers and multiple weight-bearing joints with lower back pain, as well as a history of discoloration of her hands during winter childhood. Her medications included citalopram. A family history of celiac disease was reported. Per the patient, there was no history of dysphagia, muscle pain or shortness of breath. Extraoral examination showed normal head and neck range of motion. No facial asymmetry was appreciated, and no head and neck lymphadenopathy was palpated. Multiple asymptomatic intraoral white lesions with cutaneous lesion are highly suggestive of immune-mediated disorders such as lichen planus, lichenoid drug reaction, or chronic graft-versus-host disease (GVHD). Chronic GVHD was ruled out because the patient did not have a history of hematopoietic stem cell transplantation. Given the history of psoriasis, the authors initially included secondary involvement of psoriasis, but the present case did not show the classic mixed erythroleukoplakic lesions that wax and wane. Other clinical categories that present with a similar clinical appearance include genodermatoses, storage diseases, and connective tissue disorders. Lichen planopilaris (LPP) is a primary lymphocytic cicatricial alopecia with unknown etiology. LPP is commonly seen in Caucasian women between the ages of 25 to 70 years. Clinically, LPP presents as alopecia along the vertex of the scalp with erythematous, pruritic, tender, and hyperkeratotic scaly permanent scarring.1 Graham-Little-Piccardi-Lassueur syndrome (GLS) is a rare subtype of LPP that is characterized by the triad of patchy nodular scarring alopecia of the scalp, non-scarring alopecia of the axilla and groin, and follicular spinous papules on the body, scalp or both.2 Our presented case had smooth, non-nodular asymptomatic, and non-scarring alopecia areata without axillary or groin alopecia. Therefore, LPP and GLS were ruled out. Systemic lupus erythematosus (SLE) is an immune-mediated multiorgan disorder that most commonly affect women between the ages of 15 to 50 years. SLE usually involves the kidneys, heart, and oral mucosa. The oral manifestation of SLE includes desquamative gingivitis with or without asymptomatic erythematous lesions that most commonly present on the hard palate. Sm antibodies are a specific serum marker for SLE and a high titer of RNP antibodies in the absence of antinuclear (ANA) is highly suggestive of the diagnosis mixed connective tissue disease such as SLE.3 Our patient's immunological tests for ANA and Sm-RNP antibodies were negative, so SLE was excluded. Pachyonychia congenita is a rare hereditary disorder characterized by hypertrophic nail dystrophy, painful palmoplantar keratoderma and blistering, oral leukokeratosis, pilosebaceous cysts (including steatocystoma and vellus hair cysts), palmoplantar hyperhidrosis, and follicular keratoses on the trunk and extremities.4 Due to the absence of the classic clinical presentation of pachyonychia congenita it was eliminated from the differential diagnosis list. Lipoid proteinosis (LIP) is a rare multisystem autosomal recessive disorder due to mutations of the ECM1 gene that encodes a glycoprotein known as extracellular matrix protein. Symptoms usually start early in infancy with hoarseness and seizures, followed by yellow thickening of the skin and mucous membranes as well as cicatricial alopecia.5 LIP patients usually present with a row of beaded papules along the eyelid margin, known as moniliform blepharitis, which is thought to be pathognomonic. 6,7. Our patient presented with white plaque lesions and had no history of dysphonia, seizures, or beaded papules on the eyelids, hence lipoid proteinosis was excluded. Systemic sclerosis (SS) is a rare generalized heterogenous connective disorder of unknown etiology, unclear pathogenesis, and a broad spectrum of clinical manifestations.8 SS usually presents as multifocal areas of skin fibrosis, internal organ involvement with vascular abnormalities, and immune disturbances. The main subtypes of SS are limited cutaneous sclerosis (LCS) and the diffuse form. Oral manifestations of both subtypes include the presence of multiple mucosal white lesions, microstomia, limited mouth opening, xerostomia, and telangiectasia. LCS usually presents with calcinosis cutis, Raynaud's phenomenon, esophageal dysmotility, sclerodactyly and telangiectasias (formerly abbreviated as CREST) with absent or delayed minor internal organ involvement unlike the diffuse form. Both subtypes express different autoantibodies. SS will show autoantibodies against topoisomerase I (anti-Scl70) while LCS will show anticentromere antibodies.9,10 We originally considered that the lesions may be an early manifestation of diffuse SS or LCS. However, the patient's tests were negative for anti-Scl70 and anticentromere antibodies, so both SS and LCS were precluded. Histologic examination showed intact hyperparakeratotic stratified squamous mucosa with normal maturation. The underlying lamina propria showed eosinophilic homogenous hyalinization with loss of elastic fibers; mild subepithelial lymphocytic infiltrate was also identified (Fig. 2A-2B). No evidence of cytologic atypia or invasion was present. Immunohistochemical staining for CD34 demonstrated loss of normal dendritic cells with replacement of typical sub-basilar stroma by homogenized collagen. An elastic special stain showed loss of elastic fibers in the lamina propria (Fig. 3A-3B). A PAS-D special stain was negative for fungal organisms. Congo-Red special staining was negative. Thus, the final diagnosis was rendered: features consistent with oral lichen sclerosus. Upon further investigation and appropriate referral to dermatology and gynecology, the patient had biopsy-proven lichen sclerosus from a cutaneous lesion on her back. Interestingly, a biopsy of one of the scalp lesions confirmed the diagnosis of alopecia areata (Fig. 4A-4B). The gynecologic exam revealed lichen sclerosus involving the labia, perineum, and perianal region. Clobetasol ointment was prescribed for extragenital lesions to be used daily for 4-6 weeks before switching to tacrolimus ointment for maintenance. The genital lesions were initially treated with clobetasol ointment twice weekly for 4-6 weeks, then thereafter with tacrolimus every other day. Follow-up with her family physician, annual gynecologic examination for squamous cell carcinoma screening and routine monitoring by an oral pathologist/oral medicine specialist were recommended. Lichen sclerosus (LS) is a rare chronic relapsing inflammatory mucocutaneous disorder. The etiology of LS is unknown but may be explained by local trauma or an autoimmune mechanism associated with underlying systemic conditions such as thyroiditis, diabetes mellitus, pernicious anemia.11-13 An infective etiology such as Borrelia burgdorferi has been proposed in the pathogenesis of LS. However, this hypothesis is controversial because the microorganism has not been identified in patients with LS.14,15 Almost 85% of LS cases have genital lesions, 15-20% of cases will have extragenital LS (EGLS) simultaneously with genital lesions, and only in 6% of the cases have EGLS without genital involvement.16 The majority of genital LS (GLS) lesions are asymptomatic but may present with burning, pruritic symptoms.17 Oral lichen sclerosus (OLS) is rare with less than 40 cases reported in the English literature. Typically, OLS present as well-circumscribed shiny, ivory or porcelain-white plaques or macules of various sizes. Upon palpation, the lesions usually feel soft to firm.18 According to Bevans et al. , their recent review described that OLS occurs most commonly at the labial mucosa (67.6%) followed by the buccal mucosa (32.4%), gingiva (24.3%), tongue (21.6%), buccal sulcus (13.5%), alveolar mucosa (2.7%), and tonsillar region (2.7%).19 Our case involved the gingiva and palate. LS often manifests at the anogenital region with the potential to cause atrophy, destructive scarring, functional impairment, and malignant transformation into squamous cell carcinoma (SCC). However, none of the OLS cases reported in the literature had malignant transformation to SCC. Due to the malignant propensity of the genital lesions, oral lesions should have long-term follow-up by an oral pathologist/oral medicine specialist along with a dermatologist when clinically relevant. Histologic examination of OLS alone is non-specific and has a broad differential diagnosis including amyloidosis. Unlike the homogenized collagen in OLS, amyloid will react with Congo-red histochemical stain. Also, lipoid proteinosis and scleroderma share the same histologic appearance as OLS and should only be diagnosed after correlation with the clinical presentation and laboratory testing. In conclusion, the oral pathologist who receives a biopsy with features of OLS should consider this diagnosis and perform further investigation with the referring clinician. The patient in our case had multiple genital and skin lesions that were only revealed when the patient's dermatologist and gynecologist were involved. We believe that adding this case to the literature will aid future studies and our understanding of the pathogenesis of OLS and its malignant potential. [ABSTRACT FROM AUTHOR]

While bariatric surgery is an effective method for achieving long-term weight loss, postoperative readmissions are associated with negative clinical outcomes and significant costs. We aimed to use machine learning (ML) algorithms to predict readmissions and compare results to logistic regression. Hospitals participating in the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program, United States. Patients who underwent sleeve gastrectomy (SG), Roux-en-Y gastric bypass (RYGB), and biliopancreatic diversion with duodenal switch between 2016 and 2020 were selected from the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP) database. Patient variables reported by the MBSAQIP database were analyzed by ML algorithms random forest (RF), gradient boosting (XGB), and deep neural networks (NN), and the results of the predictive models were compared to logistic regression using area under the receiver operating characteristic curve (AUROC). Our study included 863,348 patients, of which 39,068 (4.52%) were readmitted. AUROC scores were XGB.785 (95% CI.784–.786), RF.785 (95% CI.784–.785), and NN.754 (95% CI.753–.754), compared with.62 (95% CI.62–.621) for logistic regression (LR) (P <.001). The sensitivity and specificity for XGB, the best performing model, were 73.81% and 70%, compared with 52.94% and 70% for logistic regression. The most important variables were intervention or reoperation prior to discharge, unplanned ICU admission, initial procedure, and the intraoperative transfusion. ML demonstrates significant advantages over logistic regression when predicting 30-day readmission following bariatric surgery. With external validation, models could identify the best candidates for early discharge or targeted postdischarge resources. • Machine learning (ML) predicted readmissions after bariatric surgery better than logistic regression when using all predischarge information. • All model types showed low predictive ability for readmissions based on pre and intraoperative factors alone. [ABSTRACT FROM AUTHOR]

Social determinants of health (SDOH) are linked to poor health care outcomes across the different medical specialties. We conducted a scoping review to understand the existing literature and identify further areas of research to address disparities within hand surgery. A systematic search of PubMed, Scopus, and Cochrane was conducted. Inclusion criteria were English studies examining health disparities in hand surgery. The following were assessed: the main SDOH, study design/phase/theme, and main disease/injury/procedure. A previously described health disparities research framework was used to determine study phase: detecting (identifying risk factors), understanding (analyzing risk factors), and reducing (assessing interventions). Studies were categorized according to themes outlined at the National Institute of Health and American College of Surgeons: Summit on Surgical Disparities. The initial search yielded 446 articles, with 49 articles included in final analysis. The majority were detecting-type (31/49, 63%) or understanding-type (12/49, 24%) studies, with few reducing-type studies (6/49, 12%). Patient factors (31/49, 63%) and systemic/access factors (16/49, 33%) were the most frequently studied themes, with few investigating clinical care/quality factors (4/49, 8%), clinician factors (3/49, 6%), and postoperative/rehabilitation factors (1/49, 2%). The most commonly studied SDOH include insurance status (13/49, 27%), health literacy (10/49, 20%), and social deprivation (6/49, 12%). Carpal tunnel syndrome (9/49, 18%), upper extremity trauma (9/49, 18%), and amputations (5/49, 10%) were frequently assessed. Most investigations involved retrospective or database designs (29/49, 59%), while few were prospective, cross-sectional, or mixed-methods. Despite an encouraging upward trend in health disparities research, existing studies are in the early phases of investigation. Most of the literature focuses on patient factors and systemic/access factors in regard to insurance status. Further work with prospective, cross-sectional, and mixed-method studies is needed to better understand health disparities in hand surgery, which will inform future interventions. [ABSTRACT FROM AUTHOR]

Background: Optimal treatment of anal squamous cell carcinoma (ASCC) is definitive chemoradiation. Patients with persistent or recurrent disease require abdominoperineal resection (APR). Current models for predicting need for APR and overall survival are limited by low accuracy or small datasets. This study sought to use machine learning (ML) to develop more accurate models for locoregional failure and overall survival for ASCC. Methods: This study used the National Cancer Database from 2004–2018, divided into training, validation, and test sets. We included patients with stage I-III ASCC who underwent chemoradiation. Our primary outcomes were need for APR and 3-year overall survival. Random forest (RF), gradient boosting (XGB), and neural network (NN) ML-based models were developed and compared with logistic regression (LR). Accuracy was assessed using area under the receiver operating characteristic curve (AUROC). Results: APR was required in 5.3% (1,015/18,978) of patients. XGB performed best with AUROC of 0.813, compared with 0.691 for LR. Tumor size, lymphovascular invasion, and tumor grade showed the strongest influence on model predictions. Mortality was 23.6% (7,988/33,834). AUROC for XGB and LR were similar at 0.766 and 0.748, respectively. For this model, age, radiation dose, sex, and insurance status were the most influential variables. Conclusions: We developed and internally validated machine learning-based models for predicting outcomes in ASCC and showed higher accuracy versus LR for locoregional failure, but not overall survival. After external validation, these models may assist clinicians with identifying patients with ASCC at high risk of treatment failure. [ABSTRACT FROM AUTHOR]

Grading of oral epithelial dysplasia (OED) can be challenging with considerable intra- and inter-observer variability. Abnormal immunohistochemical staining patterns of the tumor suppressor protein, p53, have been recently shown to be an independent factor associated with progression. We retrospectively identified 232 oral biopsies from 1998 to 2008 with diagnosis of reactive lesions (n=30), low-grade lesions (LGLs, mild dysplasia (D1, 51), moderate dysplasia (D2, 98)) and high-grade lesions (HGLs, severe dysplasia/carcinoma in situ, 53). Tissue microarray (TMA) blocks were constructed with 1-mm-diameter core from the most representative area of the pathology. Three consecutive sections were sectioned and stained for H&E, p53 (DO7, DAKOTM) and p16 (CINTec®) per manufacturer's protocols. The staining results were reviewed by a pathologist (YCKK) blinded to clinical outcomes and categorized into three groups using a pattern-based approach: 1) conventional OED (OEDc) if p16-/p53w; 2) HPV OED (OEDHPV) if p16+/p53basal-sparing; and 3) p53 abnormal OED (OEDp53) if the pattern was p53null/cytoplasmic/overexpression. All analysis was performed using R statistical software (v4.2.2). Differences with p<0.05 was considered significant. There were 160 cases of OEDc, including 14 OEDHPV, and 42 OEDp53. While there was no age differences among the three groups, the OEDHPV group had a higher proportion of males (p=0.04). All cases of OEDHPV were found in HGLs, whereas OEDp53 were observed in both LGL (17, 11.4%) and HGLs (25, 47%); none detected in the 30 reactive lesions. Remarkably, of the 149 LGLs, 24 progressed with a median time of 4.4 years (Q1-Q3, 2.1-8.7 years), and 9 (38%) of these progressed within 3 years. Survival analysis showed that OEDp53 had significantly poorer progress-free probability (p<0.001) with hazard ratio of 14.9 (95%CI, 3.7-59.7), compared to OEDc LGL. We demonstrated that OED with abnormal p53 pattern is at high-risk of progression irrespective of dysplasia grade. [ABSTRACT FROM AUTHOR]

Human papillomavirus (HPV)-associated oral epithelial dysplasia (OED) and squamous cell carcinoma of the oral cavity (OSCC) are uncommon. They are known to present with distinct histomorphology and diffuse block-like p16 immunohistochemistry (IHC). We recently described a novel OED classification system using p53 and p16 IHC and encountered several cases of OED with both diffuse p16 expression and aberrant p53 expression ("double-positive"). We hereby report five cases of "double-positive" p53 and p16 OED and OSCC with their unique morphology and results of high-risk (HR) HPV in situ hybridization (ISH) and TP53 mutation analysis. Five cases of "double-positive" OED and/or OSCC with basaloid morphology were identified from the British Columbia Oral Biopsy Service, Vancouver General Hospital and Pathology and Laboratory Medicine, Foothills Medical Centre, University of Calgary. HR HPV ISH was performed on all cases. TP53mutation analysis by targeted Next-Generation Sequencing (NGS) was performed on four of the five cases. All five cases showed abundant karyorrhectic cells, apoptotic keratinocytes, and basaloid or "Bowenoid" morphology. All five cases demonstrated p16 diffuse block-like positivity and p53 abnormal staining patterns (3 overexpression, 2 null). HR HPV RNA ISH was negative in all five cases. Three of the four cases sent for targeted NGS demonstrated pathogenic TP53 mutations. Our results suggest these "double-positive" cases should be classified as p53 abnormal-OED and p53 abnormal-OSCC, instead of HPV-associated OED and HPV-associated OSCC, to accurately reflect their underlying pathophysiology. Interpretation of p16 IHC in OED and OSCC with basaloid histomorphology should be done with extreme caution. Confirmation with p53 IHC, with consideration for HR HPV ISH is highly recommended to prevent misdiagnosis and misclassification of these oral lesions. [ABSTRACT FROM AUTHOR]

Recent breakthroughs in circulating tumor DNA (ctDNA) technologies present a compelling opportunity to combine this emerging liquid biopsy approach with the field of radiogenomics, the study of how tumor genomics correlate with radiotherapy response and radiotoxicity. Canonically, ctDNA levels reflect metastatic tumor burden, although newer ultrasensitive technologies can be used after curative-intent radiotherapy of localized disease to assess ctDNA for minimal residual disease (MRD) detection or for post-treatment surveillance. Furthermore, several studies have demonstrated the potential utility of ctDNA analysis across various cancer types managed with radiotherapy or chemoradiotherapy, including sarcoma and cancers of the head and neck, lung, colon, rectum, bladder, and prostate. Additionally, because peripheral blood mononuclear cells are routinely collected alongside ctDNA to filter out mutations associated with clonal hematopoiesis, these cells are also available for single nucleotide polymorphism analysis and could potentially be used to detect patients at high risk for radiotoxicity. Lastly, future ctDNA assays will be utilized to better assess locoregional MRD in order to more precisely guide adjuvant radiotherapy after surgery in cases of localized disease, and guide ablative radiotherapy in cases of oligometastatic disease. [ABSTRACT FROM AUTHOR]

Indications for reverse total shoulder arthroplasty (RTSA) have been expanding. In addition to degenerative joint disease (DJD), RTSA is now being used to treat proximal humerus fractures (PHF). The purpose of this study was to compare postoperative complications in RTSA performed for DJD versus PHF. A retrospective analysis of the PearlDiver National Database was performed. International Classification of Diseases 10 codes were used to identify RTSA patients from 2015-2018 and separate them into DJD and PHF cohorts. Demographics, comorbidities, and hospital data were identified and compared using a two-sample t -test and chi-squared test. Systemic complications at 90 days and surgical complications at 90 days, 1 year, and 2 years were compared using multivariable logistic regression. Fifteen thousand six hundred seventy eight patients (92.6% DJD, 7.4% PHF) were identified. PHF patients were more likely to be older (70.3 vs. 69.7 years, P =.026), female (83.5% vs. 62.2%, P <.001), and have more medical comorbidities (Charlson Comorbidity Index 3.42 vs. 3.17, P =.006) than DJD patients. After controlling for patient factors, PHF patients were more likely than DJD patients to develop urinary tract infection (odds ratio [OR] 1.65, P <.001), deep vein thrombosis (OR 1.76, P =.024), and hematoma (OR 3.83, P <.001) within 90 days of RTSA. At 90 days, 1 year, and 2 years postoperatively, RTSA for PHF patients were also more likely than RTSA for DJD patients to sustain a periprosthetic fracture (OR 2.57, P <.001) and instability (OR 2.02, P <.001). Patients with DJD and PHF undergoing RTSA represent different patient populations with distinct postoperative clinical outcomes. RTSA for PHF has inferior outcomes, which is significant in an era of bundled payments. [ABSTRACT FROM AUTHOR]

Background: Readmission after colorectal surgery is common and often implies complications for patients and costs for hospitals. Previous works have created predictive models using logistic regression for this outcome but have shown limited accuracy. Machine learning has shown promise in improving predictions by identifying non-linear patterns in data. We sought to create a more accurate predictive model for readmission after colorectal surgery using machine learning.Methods: Patients who underwent colorectal surgery were identified in the National Quality Improvement Program (NSQIP) database including years 2012-2019 and split into training, validation, and test sets. The primary outcome was readmission within 30 days of surgery. Three types of machine learning models were created, including random forest (RF), gradient boosting (XGB), and neural network (NN). A logistic regression (LR) model was also created for comparison. Model performance was evaluated using area under the receiver operating characteristic curve (AUROC).Results: The dataset included 213,827 patients after application of exclusion criteria. A total of 23,083 (10.8%) of patients experienced readmission. NN obtained an AUROC of 0.751 (95% CI 0.743-0.759), compared with 0.684 (95% CI 0.676-0.693) for LR. RF and XGB performed similarly with AUROCs of 0.749 (95% CI 0.741-0.757) and 0.745 (95% CI 0.737-0.753) respectively. Ileus, index admission length of stay, organ-space surgical site infection present at time of surgery, and ostomy placement were identified as the most contributory variables.Conclusions: Machine learning approaches outperformed traditional statistical methods in the prediction of readmission after colorectal surgery. After external validation, this improved prediction model could be used to target interventions to reduce readmission rate. [ABSTRACT FROM AUTHOR]

Evaluation of p53 immunohistochemistry (IHC) in oral epithelial dysplasia (OED) has demonstrated a characteristic staining pattern in cases associated with high-risk human papillomavirus (HPV), identical to the pattern of expression reported in HPV associated dysplasia of the vulva. The accuracy of p53, in combination with p16, as a surrogate marker for HR HPV infection in OED, has not been extensively studied. Expression of p53 was evaluated in 105 cases of HR HPV positive oral cavity OED, in which 104 cases were p16 positive and 1 case was p16 negative. HPV status was confirmed by RT-qPCR for E6 mRNA or RNA in situ hybridization (ISH). 7 cases of p16 positive oral cavity OED with abnormal p53 expression and/or TP53 mutation had negative HPV RNA ISH and were excluded. Our cohort included 96 males and 9 females, with a median age of 57 years (range 32-89). 93% of cases demonstrated classic HPV-associated basaloid morphology, and 7% of cases were keratinizing. The floor of mouth/ventral tongue were the most affected sites (61%), followed by the lateral tongue (18%) and gingiva (13%). 76% of cases demonstrated null-like/basal sparing pattern with p53 staining and 24% demonstrated a mid-epithelial/basal sparing pattern. 11 cases had an associated squamous cell carcinoma at the time of initial biopsy. A single case progressed to squamous cell carcinoma in an immunosuppressed patient, and 16 cases recurred. Of the cases with recurrence, 38% recurred more than once. This study unequivocally demonstrates that p53 IHC, in combination with p16, can accurately identify HPV OED. Further, p53 can correctly identify p16 positive HR HPV negative p53 mutant OED with basaloid morphology. HPV OED can present with invasive SCC, but malignant progression in HPV OED is extremely rare. [ABSTRACT FROM AUTHOR]

Sixty to eighty percent of oral cavity invasive squamous cell carcinoma (OSCC) demonstrate molecular alterations in TP53. The presence of TP53 mutations in other organ systems have been associated with a more aggressive clinical course. The purpose of this study was to classify OSCC into three subtypes—conventional, Human Papillomavirus (HPV)-associated, and p53 abnormal OSCC—using immunohistochemistry to determine if subtype correlates with higher risk of lymph node metastasis at the time of surgery. A total of 104 consecutive patients with OSCC resection and associated lymph node dissection were identified from the Vancouver General Hospital database. p53 immunohistochemistry was performed for all cases and scored into p53 wild-type (conventional: scattered basal, patchy basal/parabasal; HPV-associated: mid-epithelial/basal sparing, markedly reduced/basal sparing) and p53 abnormal (overexpression basal/parabasal only, overexpression basal/parabasal to diffuse, null, cytoplasmic) patterns. p16 immunohistochemistry and high-risk HPV RNA in situ hybridization were used to confirm HPV status. Logistic regression analysis was performed to investigate the association of p53 status, tumor size, depth of invasion, and pT stage against lymph node status. We identified 24 cases with p53 wild-type patterns (18 conventional, 6 HPV-associated) and 80 (77%) cases with p53 abnormal patterns. Three of 24 p53 wild-type cases had positive lymph nodes (2 conventional, 1 HPV-associated; 13%), while 41 of 80 p53 abnormal cases had positive lymph nodes (51%) (p=0.0008). Multivariate analysis showed that p53 abnormal pattern was an independent risk factor associated with positive node(s) with an odds ratio of 5.12 (95% CI, 1.47-24.1; p=0.02). This study demonstrates that p53 abnormal OSCC were significantly more likely to be associated with positive lymph node status, compared to p53 wild-type cases at time of surgery. Further investigation with long-term follow-up is required to determine its clinical application prior to surgery planning. [ABSTRACT FROM AUTHOR]

WHO grading of oral epithelial dysplasia (OED) is employed by pathologists to aid in prognostication and treatment planning. However, considerable inter-observer variability exists. Recently, p53 abnormal OED has been proposed as a unique entity with an increased risk of malignant transformation. Subsequently, a pattern-based approach to p53/p16 immunohistochemistry (IHC) interpretation has been developed to classify OED into Human Papillomavirus (HPV)-associated, p53 wild-type (conventional), and p53 abnormal OED. In this study, the grading (three-tiered and two-tiered) and classification of OED were compared using hematoxylin and eosin-stained sections (H&E) with and without p53/p16 IHC. Formalin-fixed, paraffin-embedded oral biopsy specimens with targeted Next-Generation Sequencing results and/or high-risk HPV in situ hybridization results were collected from 51 patients. Inter-observer variability among 18 pathologists was determined using Fleiss' kappa (κ). 93% of HPV-associated OED and 77% of p53 abnormal OED were successfully identified using p53/p16 IHC. In contrast, 55% of HPV-associated OED and 37% of p53 abnormal OED were identified using H&E. 19% of p53 abnormal OED were initially interpreted as "reactive, no dysplasia" using H&E; this decreased to 10% following reviewing of p53/p16 IHC. There is excellent agreement beyond chance and fair to good agreement beyond chance for the classification of OED using p53/p16 IHC (overall, κ = 0.61, HPV-associated OED, κ = 0.92; p53 abnormal OED, κ = 0.56). Both the three-tiered and two-tiered dysplasia grading systems demonstrate poor agreement beyond chance in H&E (κ = 0.35 and 0.39; p < 0.0001) and p53/p16 IHC (κ = 0.33 and 0.37; p < 0.0001). A panel of p53/p16 IHC can improve both diagnostic accuracy and inter-observer agreement in the diagnosis of OED. p53 abnormal OED cannot be reliably identified on H&E alone, and the histologic spectrum of HPV-associated dysplasia is broader than currently appreciated. [ABSTRACT FROM AUTHOR]

Purpose: The COVID-19 pandemic largely suspended in-person scientific meetings because of risk of disease spread. In the era of vaccination and social distancing practices, meetings have begun returning to in-person formats. We surveyed attendees and potential attendees of 2 oncology meetings in the United States to identify rates of mixing behavior and the subsequent rate of self-reported COVID-19 infection.Methods and Materials: We collected via survey reported social mixing behavior and COVID-19 positivity (within 21 days of meeting conclusion) of actual and potential in-person attendees of the American Society of Clinical Oncology (ASCO) Quality Care Symposium held September 24 to 25, 2021, and the American Society for Radiation Oncology (ASTRO) Annual Meeting held October 24 to 27, 2021. Conference speakers and other participants were identified through publicly available meeting materials and targeted via e-mail when possible. Recruitment of additional attendees and potential attendees was also conducted through a sharable link promoted via oncology newsletters and social media. Descriptive statistics alone were performed owing to low COVID-19 event rates.Results: Response rates from targeted conference participants with publicly available e-mails were 27.4% for the ASCO and 14.3% for the ASTRO meetings. The ASCO survey produced 94 responses (48 in-person attendees). The ASTRO survey produced 370 responses (267 in-person attendees). Across both meetings, 3 of 308 (1.0%) in-person attendees versus 2 of 141 (1.4%) nonattendees tested positive for COVID-19. Low COVID-19 positivity rates were reported among in-person attendees spending more (>20) versus fewer (≤20) hours attending live sessions (2.2% vs 0%) and among indoor social event participants versus nonparticipants (0.8% vs 1.9%). Attendees largely felt comfortable attending additional in-person meetings after experiencing ASCO (87.5%) or ASTRO (91.9%) and felt mask compliance was good or excellent at ASCO (100%) and ASTRO (94.6%) meetings.Conclusions: In-person meetings do not seem to be contributing to high rates of new COVID-19 infections in the setting of vaccine and social distancing mandates, supporting paths forward for at least partially in-person conferences as COVID-19 becomes endemic. [ABSTRACT FROM AUTHOR]

Purpose: Upper airway obstruction seen in Robin Sequence (RS) is commonly treated with mandibular distraction osteogenesis (MDO). The purpose of this study is to evaluate the impact of distraction distance on sleep study outcomes in patients with obstructive sleep apnea (OSA) secondary to RS.Methods: A retrospective cohort study was conducted for patients with isolated RS who underwent MDO at Children's Hospital Los Angeles between January 2006-September 2021. The predictor variable was distraction distance (maximal distraction using a 30 mm device vs sub-maximal distraction), and the primary outcome variable was OSA scores. Relationships between covariates, including demographic characteristics, preoperative sleep variables, and postoperative OSA outcomes using polysomnography, were also analyzed. Descriptive statistics and tests of statistical significance were performed using the Statistical Package for Social Sciences (SPSS) (version 28.0), including Student's t-test, proportions testing, multiple linear regression, and correlation analysis.Results: Seventy-one patients met inclusion criteria (39.4% female, 60.6% male). Average age at MDO was 3.0 ± 10.2 months. Fifty-six patients were distracted maximally with a 30 mm distractor, while the remaining 15 patients experienced shorter distraction due to distractor limitations (25 mm distractor), persistent infection or family request. Looking at absolute values of postoperative sleep study variables, there were no significant differences between patients who were maximally and sub-maximally distracted across apnea-hypopnea index (AHI), highest carbon dioxide, lowest oxygen saturation, and oxygen requirement. However, both cohorts demonstrated significant improvements in lowest oxygen saturation, AHI, highest carbon dioxide level, and highest oxygen requirement compared to their pre-distraction levels. Compared with patients distracted <30 mm, maximal distraction had a significantly greater improvement in AHI when controlling for preoperative sleep study variables (P = .047).Conclusion: Patients with isolated RS who have more severe OSA experienced greater improvements in AHI, oxygen requirement, and oxygen saturation after MDO. Two-thirds of patients no longer had oxygen requirements after MDO. Our results suggest that MDO is helpful in treating patients with RS regardless of distraction distance. However, our study provides evidence that increasing the distraction distance may further improve AHI, which is particularly beneficial to patients with a significant preoperative AHI. [ABSTRACT FROM AUTHOR]

Background: Procedure-specific complications can have devastating consequences. Machine learning–based tools have the potential to outperform traditional statistical modeling in predicting their risk and guiding decision-making. We sought to develop and compare deep neural network (NN) models, a type of machine learning, to logistic regression (LR) for predicting anastomotic leak after colectomy, bile leak after hepatectomy, and pancreatic fistula after pancreaticoduodenectomy (PD). Methods: The colectomy, hepatectomy, and PD National Surgical Quality Improvement Program (NSQIP) databases were analyzed. Each dataset was split into training, validation, and testing sets in a 60/20/20 ratio, with fivefold cross-validation. Models were created using NN and LR for each outcome. Models were evaluated primarily with area under the receiver operating characteristic curve (AUROC). Results: A total of 197,488 patients were included for colectomy, 25,403 for hepatectomy, and 23,333 for PD. For anastomotic leak, AUROC for NN was 0.676 (95% 0.666–0.687), compared with 0.633 (95% CI 0.620–0.647) for LR. For bile leak, AUROC for NN was 0.750 (95% CI 0.739–0.761), compared with 0.722 (95% CI 0.698–0.746) for LR. For pancreatic fistula, AUROC for NN was 0.746 (95% CI 0.733–0.760), compared with 0.713 (95% CI 0.703–0.723) for LR. Variables related to intra-operative information, such as surgical approach, biliary reconstruction, and pancreatic gland texture were highly important for model predictions. Discussion: Machine learning showed a marginal advantage over traditional statistical techniques in predicting procedure-specific outcomes. However, models that included intra-operative information performed better than those that did not, suggesting that NSQIP procedure-targeted datasets may be strengthened by including relevant intra-operative information. [ABSTRACT FROM AUTHOR]

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.ctim.2009.05.001 Byline: Myeong Soo Lee (a), Kevin W. Chen (b), Tae-Young Choi (a), Edzard Ernst (b)(c) Keywords: Qigong; Type 2 diabetes; Complementary therapies; Systematic review Abstract: To summarize and critically evaluate the evidence available from clinical trials (CTs) of qigong for patients with type 2 diabetes. Author Affiliation: (a) Division of Standard Research, Korea Institute of Oriental Medicine, Daejeon 305-811, South Korea (b) Center for Integrative Medicine, University of Maryland School of Medicine, Baltimore, USA (c) Complementary Medicine, Peninsula Medical School, Universities of Exeter & Plymouth, Exeter, UK

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.addbeh.2006.12.022 Byline: William A. Vega, Kevin W. Chen, Jill Williams Keywords: Smoking; Behavioral problems; Psychiatric disorders; Hispanic; African American Abstract: Data from the NHSDA (2000) which contained screening measures for assessing risk for DSM-IV psychiatric disorders, were used to estimate smoking prevalence and its association with these disorders, among European American, Hispanic, and African American adolescents. Prevalence estimates, odds ratios, and hazard models were used to compare ethnic subgroups. European American and Hispanic adolescents born in the U.S. had a higher prevalence of smoking and DSM-IV tobacco dependence, and girls were higher than boys. Lifetime smokers had statistically significant odds ratios for anxiety, affective, substance use, and any behavior disorder, while Current smokers had a similar risk pattern except anxiety disorder. The increased risk for substance use disorder among smokers was notably higher. A hazard analysis showed that early onset of smoking (before 12 years) was related to earlier illicit drug use initiation as contrasted with later onset (12 years +), and non-smoking sharply reduced risk and delayed initiation into drug use. African American and Hispanic immigrants had lower risk of smoking initiation and tobacco dependence, however, ethnic group adolescents who initiated smoking shared increased risk of addictive and non-addictive disorders. Author Affiliation: Department of Psychiatry, Robert Wood Johnson Medical School, United States