Prevalence and Association of Sarcopenia with Mortality in Patients with Head and Neck Cancer: A Systematic Review and Meta-Analysis.

BACKGROUND: The objective of this meta-analysis was to assess the association of sarcopenia defined on computed tomography (CT) head and neck with survival in head and neck cancer patients. METHODS: Following a PROSPERO-registered protocol, two blinded reviewers extracted data and evaluated the quality of the included studies using the Quality In Prognostic Studies (QUIPS) tool, in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The quality of evidence was assessed using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) framework. A meta-analysis was conducted using maximally adjusted hazard ratios (HRs) with the random-effects model. Heterogeneity was measured using the I2 statistic and was investigated using meta-regression and subgroup analyses where appropriate. RESULTS: From 37 studies (11,181 participants), sarcopenia was associated with poorer overall survival (HR 2.11, 95% confidence interval [CI] 1.81-2.45; p < 0.01), disease-free survival (HR 1.76, 95% CI 1.38-2.24; p < 0.01), disease-specific survival (HR 2.65, 95% CI 1.80-3.90; p < 0.01), progression-free survival (HR 2.24, 95% CI 1.21-4.13; p < 0.01) and increased chemotherapy or radiotherapy toxicity (risk ratio 2.28, 95% CI 1.31-3.95; p < 0.01). The observed association between sarcopenia and overall survival remained significant across different locations of cancer, treatment modality, tumor stages and geographical region, and did not differ between univariate and multivariate HRs. Statistically significant correlations were observed between the C3 and L3 cross-sectional area, skeletal muscle mass, and skeletal muscle index. CONCLUSIONS: Among patients with head and neck cancers, CT-defined sarcopenia was consistently associated with poorer survival and greater toxicity.

Cataract Surgery and Cognitive Benefits in the Older Person: A Systematic Review and Meta-analysis.

TOPIC: This systematic review and meta-analysis aims to clarify the association of cataract surgery with cognitive impairment and dementia. CLINICAL RELEVANCE: The association between vision impairment and cognitive decline is well-established. However, the cognitive benefits of cataract surgery are less clear. Given the lack of cure for dementia, identifying modifiable risk factors is key in caring for patients with cognitive deficits. METHODS: The study was conducted following Preferred Reporting Items for Systematic Review and Meta-analyses guidelines. PubMed, Embase, and Cochrane Library were searched from inception through October 11, 2022, for studies reporting the effect of cataract surgery on cognitive impairment and dementia. We pooled maximally adjusted hazard ratios (HRs) for dichotomous outcomes and ratio of means (RoM) for continuous outcomes using a random-effects model. Heterogeneity was examined using sensitivity and subgroup analyses. The quality of evidence was evaluated using the Newcastle-Ottawa scale, Cochrane risk-of-bias tool for randomized trials, and Grading of Recommendations, Assessment, Development and Evaluations (GRADE) guidelines. RESULTS: This review included 24 articles comprising 558 276 participants, of which 19 articles were analyzed qualitatively. The bias of studies ranged from low to moderate, and GRADE extended from very low to low. Cataract surgery was associated with a 25% reduced risk of long-term cognitive decline compared with those with uncorrected cataracts (HR, 0.75; 95% confidence interval [CI], 0.72-0.78). This cognitive benefit was seen across various cognitive outcomes and remained robust to sensitivity analyses. Participants who underwent cataract surgery showed a similar risk of long-term cognitive decline as healthy controls without cataracts (HR, 0.84; 95% CI, 0.66-1.06). Additionally, cataract surgery was associated with a 4% improvement in short-term cognitive test scores among participants with normal cognition (RoM, 0.96; 95% CI, 0.94-0.99), but no significant association was observed among participants with preexisting cognitive impairment. DISCUSSION: Cataract surgery may be associated with a lower risk of cognitive impairment and dementia, and cataract-associated vision impairment may be a modifiable risk factor for cognitive decline. Physicians should be aware of the cognitive sequelae of cataracts and the possible benefits of surgery. The cognitive benefits of cataract surgery should be investigated further in randomized trials. FINANCIAL DISCLOSURE(S): The author(s) have no proprietary or commercial interest in any materials discussed in this article.

Meta-analysis of multi-modality therapies in sinonasal undifferentiated carcinoma-A timely update.

OBJECTIVES: Sinonasal undifferentiated carcinoma (SNUC) is a rare but aggressive tumour with very poor prognosis. There are currently no well-established clinical trials to guide therapy and the impact of various treatment modalities on survival is not well defined. We aim to provide an updated systematic review on current treatment modalities on survival outcomes. DESIGN AND SETTING: Individual patient data were extracted, and survival data pooled in a one-stage meta-analysis. Descriptive statistics were analysed using the Kaplan-Meier method. Patient-level comparisons stratified by treatment modalities, adjusted for demographics, were conducted using shared-frailty Cox regression. PARTICIPANTS AND MAIN OUTCOME MEASURES: Participants include all patients diagnosed with SNUC based on histological evidence. We looked at the overall cumulative survival outcome for different treatment modalities and overall survival by treatment modality in low versus high stage SNUC patients. RESULTS AND CONCLUSION: Seventeen studies were identified, comprising 208 patients from 1993 to 2020. There was no significant difference in cumulative overall survival in low versus high stage patients, and no significant difference in outcomes by treatment modality. The overall cumulative survival of SNUC is 30% at 95 months. Among patients treated with various combinations of treatment modalities, patients with chemoradiotherapy had the highest cumulative survival of 42% at 40 months. Definitive chemoradiotherapy was associated with improved disease survival rate. Regardless of tumour stage, patients should be treated early and aggressively, with no superiority of one treatment regimen over another. Trimodality treatment does not confer survival advantage over bimodality treatment.

Clinical relevance of PD-1 positive CD8 T-cells in gastric cancer.

BACKGROUND: We evaluated the relevance of PD-1+CD8+ T-cells in gastric cancer (GC) including prognostic significance, association with chemotherapy and immunotherapy sensitivity and correlations with the tumor microenvironment (TME). METHODS: Discovery cohort: GC samples were evaluated for AE1/3, CD8, PD-1, Ki-67 and Granzyme-B expression with fluorescence-based multiplex immunohistochemistry (mIHC). Validation cohorts: we analyzed bulk RNAseq GC datasets from TCGA, the "3G" chemotherapy trial and an immunotherapy phase 2 trial. The cox proportional hazards model was used to identify factors that influenced overall survival (OS). To study the TME, we analyzed single-cell RNAseq performed on GCs. RESULTS: In the discovery cohort of 350 GCs, increased PD-1 expression of CD8 T-cells was prognostic for OS (HR 0.822, p = 0.042). PD-1 expression in CD8 T-cells highly correlated with cytolytic [Granzyme-B+] (r = 0.714, p < 0.001) and proliferative [Ki-67+] (r = 0.798, p < 0.001) activity. Analysis of bulk RNAseq datasets showed tumors with high PD-1 and CD8A expression levels had improved OS when treated with immunotherapy (HR 0.117, p = 0.036) and chemotherapy (HR 0.475, p = 0.017). Analysis of an scRNAseq dataset of 152,423 cells from 40 GCs revealed that T-cell and NK-cell proportions were higher (24% vs 18% and 19% vs 15%, p < 0.0001), while macrophage proportions were lower (7% vs 11%, p < 0.0001) in CD8PD-1high compared to CD8PD-1low tumors. CONCLUSION: This is one of the largest GC cohorts of mIHC combined with analysis of multiple datasets providing orthogonal validation of the clinical relevance of PD-1+CD8+ T-cells being associated with improved OS. CD8PD-1high tumors have distinct features of an immunologically active, T-cell inflamed TME.

The efficacy of adjuvant radioactive iodine after reoperation in patients with persistent or recurrent differentiated thyroid cancer: a systematic review.

OBJECTIVE: The effectiveness of adjuvant radioiodine (RAI) after reoperation in patients with persistent or recurrent differentiated thyroid cancer (DTC) is controversial. Although various organizations recognize that strong evidence for the use of RAI is lacking, they continue to recommend the use of adjuvant RAI therapy for select groups of patients. This is concerning as RAI therapy has potential side effects such as gastrointestinal symptoms, bone marrow suppression, and gonadal damage. METHODS: Four electronic databases were systematically searched for randomized trials or observational studies that examined the outcomes of adjuvant RAI after reoperation for recurrent DTC, among patients of any age. The baseline characteristics, treatment response, disease progression, and overall survival of these studies were synthesized and reported. A meta-analysis of the use of RAI on progression-free survival was also performed. RESULTS: Six observational studies, comprising a combined cohort of 437 patients who underwent reoperation, were included from 1212 records. Adjuvant RAI after reoperation in recurrent DTC was not associated with longer progression-free or overall survival. There was also no association of RAI with excellent structural or biochemical treatment response, lower thyroglobulin levels, nor a lower rate of second recurrence or distant metastases. CONCLUSIONS: Adjuvant RAI after reoperation in recurrent DTC was not associated with improved cancer or treatment-related outcomes. However, as the included studies were of inadequate quality, there is an urgent need for randomized trials and well-analyzed cohort studies. Physicians should exercise clinical judgment to prescribe adjuvant RAI for only selected, high-risk patients.

KMSubtraction: reconstruction of unreported subgroup survival data utilizing published Kaplan-Meier survival curves.

BACKGROUND: Data from certain subgroups of clinical interest may not be presented in primary manuscripts or conference abstract presentations. In an effort to enable secondary data analyses, we propose a workflow to retrieve unreported subgroup survival data from published Kaplan-Meier (KM) plots. METHODS: We developed KMSubtraction, an R-package that retrieves patients from unreported subgroups by matching participants on KM plots of the overall cohort to participants on KM plots of a known subgroup with follow-up time. By excluding matched patients, the opposing unreported subgroup may be retrieved. Reproducibility and limits of error of the KMSubtraction workflow were assessed by comparing unmatched patients against the original survival data of subgroups from published datasets and simulations. Monte Carlo simulations were utilized to evaluate the limits of error of KMSubtraction. RESULTS: The validation exercise found no material systematic error and demonstrates the robustness of KMSubtraction in deriving unreported subgroup survival data. Limits of error were small and negligible on marginal Cox proportional hazard models comparing reconstructed and original survival data of unreported subgroups. Extensive Monte Carlo simulations demonstrate that datasets with high reported subgroup proportion (r = 0.467, p < 0.001), small dataset size (r = - 0.374, p < 0.001) and high proportion of missing data in the unreported subgroup (r = 0.553, p < 0.001) were associated with uncertainty are likely to yield high limits of error with KMSubtraction. CONCLUSION: KMSubtraction demonstrates robustness in deriving survival data from unreported subgroups. The limits of error of KMSubtraction derived from converged Monte Carlo simulations may guide the interpretation of reconstructed survival data of unreported subgroups.

Choice of PD-L1 immunohistochemistry assay influences clinical eligibility for gastric cancer immunotherapy.

BACKGROUND: Immune checkpoint inhibitors (ICI) are now standard-of-care treatment for patients with metastatic gastric cancer (GC). To guide patient selection for ICI therapy, programmed death ligand-1 (PD-L1) biomarker expression is routinely assessed via immunohistochemistry (IHC). However, with an increasing number of approved ICIs, each paired with a different PD-L1 antibody IHC assay used in their respective landmark trials, there is an unmet clinical and logistical need for harmonization. We investigated the interchangeability between the Dako 22C3, Dako 28-8 and Ventana SP-142 assays in GC PD-L1 IHC. METHODS: In this cross-sectional study, we scored 362 GC samples for PD-L1 combined positive score (CPS), tumor proportion score (TPS) and immune cells (IC) using a multiplex immunohistochemistry/immunofluorescence technique. Samples were obtained via biopsy or resection of gastric cancer. RESULTS: The percentage of PD-L1-positive samples at clinically relevant CPS ≥ 1, ≥ 5 and ≥ 10 cut-offs for the 28-8 assay were approximately two-fold higher than that of the 22C3 (CPS ≥ 1: 70.3 vs 49.4%, p < 0.001; CPS ≥ 5: 29.1 vs 13.4%, p < 0.001; CPS ≥ 10: 13.7 vs 7.0%, p = 0.004). The mean CPS score on 28-8 assay was nearly double that of the 22C3 (6.39 ± 14.5 vs 3.46 ± 8.98, p < 0.001). At the clinically important CPS ≥ 5 cut-off, there was only moderate concordance between the 22C3 and 28-8 assays. CONCLUSION: Our findings suggest that scoring PD-L1 CPS with the 28-8 assay may result in higher PD-L1 scores and higher proportion of PD-L1 positivity compared to 22C3 and other assays. Until stronger evidence of inter-assay concordance is found, we urge caution in treating the assays as equivalent.

Association of obstructive sleep apnea with thyroid cancer incidence: a systematic review and meta-analysis.

PURPOSE: Obstructive sleep apnea (OSA) is a postulated carcinogen based on epidemiological associations with all-cancer incidence and non-thyroid biological models. However, associations with thyroid carcinoma are unclear. METHODS: We included observational/randomized studies of associations of OSA with thyroid carcinoma incidence/mortality in adults, from four databases. Random-effects meta-analyses and the population attributable fraction (PAF; from published global OSA prevalence estimates) were computed. RESULTS: We included four observational studies (N = 2,839,325), all with moderate/low risk of bias. OSA diagnosis was associated with twofold incidence of thyroid carcinoma (pooled HR 2.32, 95% CI 1.35-3.98, I2 = 95%), after multi-adjustment for demographics, BMI, smoking, alcohol, and comorbidities. Subgroup analysis of studies with at least 5 years of follow-up showed a stronger association of OSA with thyroid cancer incidence (pooled HR 3.27, 95% CI 2.80-3.82, I2 = 0%). Up to 14.5% (95% CI 4.29-27.6%) of incident thyroid carcinomas globally may be associated with OSA. Thyroid carcinoma mortality data was unavailable. CONCLUSIONS: OSA is associated with higher thyroid carcinoma incidence, though this does not prove causation. Biological/clinical studies should investigate OSA severity in relation to thyroid carcinoma progression and mortality, stratified by tumor histology.

Association Between Hearing Loss and Cardiovascular Disease: A Meta-analysis.

OBJECTIVE: Hearing loss (HL) has been postulated to be linked to cardiovascular diseases (CVDs) via vascular mechanisms, but epidemiological associations remain unclear. The study aims to clarify the association between HL and stroke, coronary artery disease (CAD), and any CVD. DATA SOURCES: PubMed, Embase, and SCOPUS from inception until April 27, 2022. REVIEW METHODS: Three blinded reviewers selected observational studies reporting stroke, CAD, and any CVD in patients with HL, compared to individuals without HL. We extracted data, evaluated study bias using the Newcastle-Ottawa scale, following Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines and a PROSPERO-registered protocol (CRD42022348648). We used random-effects inverse variance meta-analyses to pool the odds ratios (ORs) for the association of HL with stroke, CAD, and any CVD. RESULTS: We included 4 cohort studies (N = 940,771) and 6 cross-sectional studies (N = 680,349). Stroke, CAD, and any CVD were all strongly associated with HL. The overall pooled OR of the association between HL and stroke was 1.26 (95% confidence interval [CI] = 1.16-1.37, I2 = 78%), and was 1.33 (95% CI = 1.12-1.58) and 1.29 (95% CI = 1.14-1.45) for low- and high-frequency HL, respectively. Minimal publication bias was observed, with minimal change to pooled effect size following trim and fill. Similarly, the pooled OR of the association between HL and CAD was 1.36 (95% CI = 1.13-1.64, I2 = 96%), while that between HL and any CVD was 1.38 (95% CI = 1.07-1.77, I2 = 99%). CONCLUSION: Our findings suggest that HL and CVD are closely related. Physicians treating patients with HL should be cognizant of this association and view HL in the broader context of general health and aging.

Risk of Suicide and Psychiatric Disorders Among Isotretinoin Users: A Meta-Analysis.

Importance: Isotretinoin is hypothesized to contribute to the development of psychiatric disorders, but the epidemiological association and risk factors associated with psychiatric disorders among isotretinoin users remain unclear. Objective: To clarify the absolute and relative risk and risk factors associated with suicide and psychiatric disorders among isotretinoin users. Data Sources: PubMed, Embase, Web of Science, and Scopus were searched from inception until January 24, 2023. Study Selection: Randomized trials and observational studies were selected if they reported the absolute risk, relative risk, and risk factors for suicide and psychiatric disorders among isotretinoin users. Data Extraction and Synthesis: Relevant data were extracted and risk of bias was evaluated at the study level using the Newcastle-Ottawa Scale. Data were pooled using inverse variance-weighted meta-analyses. Heterogeneity was measured using the I2 statistic, and meta-regression analyses were performed. Main Outcomes and Measures: Absolute risk (percentage), relative risks (risk ratios [RR]), and risk factors (RR) of suicide and psychiatric disorders among isotretinoin users. Results: A total of 25 studies including 1 625 891 participants were included in the review and 24 in the meta-analysis. Among the included studies, participants' average age ranged from 16 to 38 years, and distribution by sex ranged from 0% to 100% male. The 1-year pooled absolute risk from between 2 and 8 studies of completed suicide, suicide attempt, suicide ideation, and self-harm were each less than 0.5%, while that of depression was 3.83% (95% CI, 2.45-5.93; I2 = 77%) in 11 studies. Isotretinoin users were less likely than nonusers to attempt suicide at 2 years (RR, 0.92; 95% CI, 0.84-1.00; I2 = 0%), 3 years (RR, 0.86; 95% CI, 0.77-0.95; I2 = 0%), and 4 years (RR, 0.85; 95% CI, 0.72-1.00; I2 = 23%) following treatment. Isotretinoin was not associated with the risk of all psychiatric disorders (RR, 1.08; 95% CI, 0.99-1.19; I2 = 0%). Study-level meta-regression found that studies with participants of older age reported lower 1-year absolute risk of depression, while those with a higher percentage of male participants reported a higher 1-year absolute risk of completed suicide. Conclusions and Relevance: The findings suggest that at a population level, isotretinoin users do not have increased risk of suicide or psychiatric conditions but may instead have a lower risk of suicide attempts at 2 to 4 years following treatment. While these findings are reassuring, clinicians should continue to practice holistic psychodermatologic care and monitor patients for signs of mental distress during isotretinoin treatment.

Association Between Smoking and Chronic Rhinosinusitis: A Systematic Review and Meta-Analysis.

OBJECTIVE: Chronic rhinosinusitis (CRS) is a prevalent inflammatory disease of the upper airway. The impact of smoking on CRS has not been clearly established. We aim to clarify the association between first-hand cigarette smoking and the prevalence and prognoses of CRS. REVIEW METHODS: PubMed, Embase, SCOPUS, and Cochrane Library were searched from inception until May 15, 2022. Three blinded reviewers selected relevant studies, extracted data, and evaluated study bias following a PROSPERO-registered protocol (CRD42022345585). We used random-effects meta-analyses to pool the prevalence of smoking in CRS, association between smoking status and CRS, and association of smoking with quality of life (QOL) before and after functional endoscopic sinus surgery (FESS). We also performed descriptive analyses of olfactory function, CT scores, and endoscopy scores before and after FESS. RESULTS: We included 23 cross-sectional studies, 19 cohort studies, two case-control studies, and one prospective clinical trial. The pooled prevalence of ever-smokers was 40% (95% CI = 0.30-0.51) and 33% (95% CI = 0.25-0.43) in patients with and without CRS. Compared to never-smokers, active smokers and past smokers had 1.35 (95% CI = 1.18-1.55) and 1.23 (95% CI = 1.17-1.29) higher odds of having CRS. Among patients with CRS, non-smokers reported higher initial QOL than smokers (standardized mean difference [SMD] = 0.23, 95% CI = 0.11-0.35), although post-FESS QOL was similar (SMD = 0.10, 95% CI = -0.30-0.51). Descriptive analysis found no significant correlations between smoking and post-FESS olfactory function and endoscopy scores. CONCLUSIONS: Cigarette smoking is associated with higher prevalence and odds of CRS. Clinicians should be aware that smoking predisposes to CRS, but does not negatively impact the rhinologic outcomes of FESS. Laryngoscope, 134:2513-2524, 2024.

Olfactory Impairment and Frailty: A Systematic Review and Meta-Analysis.

Importance: Olfactory impairment (OI) and frailty are prevalent conditions associated with aging, but studies investigating their association with each other have been discordant. Objective: To summarize current evidence surrounding the association between OI and frailty. Data Sources: PubMed, Embase, Cochrane Library, SCOPUS, and CINAHL from inception to November 28, 2023. Study Selection: This study included observational studies investigating the association between objectively or subjectively assessed OI and objectively evaluated frailty among adults. Data Extraction and Synthesis: Two independent authors extracted data into a structured template. Maximally adjusted estimates were pooled using a random-effects model, and statistical heterogeneity was evaluated using I2 values. Additional prespecified subgroup and sensitivity analyses were performed. This study used the Newcastle-Ottawa Scale for bias assessment and the Grading of Recommendations Assessment, Development and Evaluation framework for overall evidence quality evaluation. Main Outcomes and Measures: The primary outcome was the cross-sectional association between OI and frailty, for which the odds of frailty were compared between participants with and without OI. The secondary outcome was the cross-sectional association between frailty and OI, for which the odds of OI were compared between participants with and without frailty. Results: This study included 10 studies with 10 624 patients (52.9% female; mean [SD] age, 62.9 [9.6] years). The Newcastle-Ottawa Scale score of studies ranged from low to moderate. Grading of Recommendations Assessment, Development and Evaluation scores ranged from low to moderate. OI was associated with a 2.32-fold (odds ratio [OR], 2.32; 95% CI, 1.63-3.31; I2 = 0%) greater odds of frailty compared with individuals with healthy olfactory function. The odds of OI was progressively greater with categorical frailty status, with a 1.55-fold (OR, 1.55; 95% CI, 1.32-1.82; I2 = 0%), 2.28-fold (OR, 2.28; 95% CI, 1.96-2.65; I2 = 0%), and 4.67-fold (OR, 4.67; 95% CI, 2.77-7.86; I2 = 0%) increase in odds for individuals with prefrailty, frailty, and the most frailty, respectively, compared with robust individuals. The results demonstrated stability in subgroup analyses (geographical continent of study, objective vs subjective olfactory assessment) and sensitivity tests. Conclusions and Relevance: The results of this systematic review and meta-analysis suggest that there is an association between OI and frailty, with an increase in the odds of OI with worsening categorical frailty status among individuals with prefrailty, frailty, and the most frailty. OI may be a potential biomarker for frailty. Future studies could delve into whether OI may be a modifiable risk factor for frailty.

Serum Biomarkers after Adenotonsillectomy for Pediatric OSA: A Systematic Review and Meta-Analysis.

OBJECTIVE: To assess whether adenotonsillectomy improves levels of inflammatory and cardiometabolic markers in children with polysomnographically diagnosed obstructive sleep apnea (OSA). DATA SOURCES: Two authors independently searched PubMed, Embase, and Cochrane databases up to August 16, 2022, for studies relating to pre- and post-operative levels of serum markers in pediatric patients undergoing adenotonsillectomy. REVIEW METHODS: Data were extracted from included articles into a structured proforma. Meta-analyses of the standardized mean difference (SMD) were conducted in random-effects models. We calculated the probability of benefit (POB) and number needed to treat (NNT) for outcomes that demonstrated a statistically significant effect after adenotonsillectomy. The primary outcomes were changes in serum markers including C-reactive protein (CRP), high-sensitivity CRP (hs-CRP), Insulin-like growth factor 1 (IGF-1), interleukin-10 (IL-10), interleukin-6 (IL-6), tumor necrosis factor-alpha (TNF-α), Brain natriuretic peptide (BNP), insulin, glucose, total cholesterol, triglyceride, low-density lipoprotein (LDL), high-density lipoprotein (HDL). RESULTS: We screened 1616 studies and included 26 studies with 1331 participants. Meta-analysis was performed on 20 of the included studies. Adenotonsillectomy was associated with a significant decrease in insulin levels (SMD = -0.322, 95% Confidence Interval (CI) = -0.583 to -0.061), CRP (SMD = -0.946, 95% CI = -1.578 to -0.314), and BNP (SMD = -1.416, 95% CI = -2.355 to -0.477) and significant increase in levels of IGF-1 (SMD = 0.691, 95% CI = 0.207 to 1.176). There were no significant changes in levels of triglyceride, total cholesterol, TNF-α, LDL, HDL, glucose, IL-10, and IL-6. CONCLUSION: In children with polysomnographically diagnosed OSA, adenotonsillectomy was associated with improvements in serum biomarkers, comprising lower CRP, insulin, and BNP, and higher IGF-1. Laryngoscope, 134:3030-3037, 2024.

Large language models approach expert-level clinical knowledge and reasoning in ophthalmology: A head-to-head cross-sectional study.

Large language models (LLMs) underlie remarkable recent advanced in natural language processing, and they are beginning to be applied in clinical contexts. We aimed to evaluate the clinical potential of state-of-the-art LLMs in ophthalmology using a more robust benchmark than raw examination scores. We trialled GPT-3.5 and GPT-4 on 347 ophthalmology questions before GPT-3.5, GPT-4, PaLM 2, LLaMA, expert ophthalmologists, and doctors in training were trialled on a mock examination of 87 questions. Performance was analysed with respect to question subject and type (first order recall and higher order reasoning). Masked ophthalmologists graded the accuracy, relevance, and overall preference of GPT-3.5 and GPT-4 responses to the same questions. The performance of GPT-4 (69%) was superior to GPT-3.5 (48%), LLaMA (32%), and PaLM 2 (56%). GPT-4 compared favourably with expert ophthalmologists (median 76%, range 64-90%), ophthalmology trainees (median 59%, range 57-63%), and unspecialised junior doctors (median 43%, range 41-44%). Low agreement between LLMs and doctors reflected idiosyncratic differences in knowledge and reasoning with overall consistency across subjects and types (p>0.05). All ophthalmologists preferred GPT-4 responses over GPT-3.5 and rated the accuracy and relevance of GPT-4 as higher (p<0.05). LLMs are approaching expert-level knowledge and reasoning skills in ophthalmology. In view of the comparable or superior performance to trainee-grade ophthalmologists and unspecialised junior doctors, state-of-the-art LLMs such as GPT-4 may provide useful medical advice and assistance where access to expert ophthalmologists is limited. Clinical benchmarks provide useful assays of LLM capabilities in healthcare before clinical trials can be designed and conducted.

Relationship between vision impairment and employment.

AIMS: To examine the relationship between vision impairment (VI) and employment outcomes in a multiethnic Asian population. METHODS: We included 7608 Asian individuals aged ≥40 years (mean (SD) age: 58.4 (10.3) years; 64.8% male) from the Singapore Epidemiology Eye Disease Study (response rate: 78.8%), a population-based cohort study (mean follow-up period: 6.2 years). Presenting visual acuity (VA) was assessed using a logarithm of the minimum angle of resolution (logMAR) chart, with VI defined as mild (VA >0.3 to <0.6 logMAR) and moderate to severe (VA ≥0.6 logMAR). Self-reported employment statuses at both baseline and follow-up were used as outcomes. Underemployment was defined as a decline in occupational skill level, categorised by International Standard Classification of Occupations, at follow-up compared with baseline. Multinomial logistic regression models were used to determine independent associations between VI and various employment outcomes, adjusted for variables that were found to significantly differ across employment statuses. RESULTS: Presenting VI was prevalent in 20.2% (N=1536) of participants. Compared with those without VI, participants with mild and moderate to severe VI were more likely to be unemployed at baseline (OR 1.47, 95% CI 1.15 to 1.87, p=0.002 and 2.74, 95% CI 1.94 to 3.89, p<0.001, respectively). At follow-up, participants with any VI at baseline were more likely to be underemployed (OR 1.46, 95% CI 1.03 to 2.05, p=0.033). CONCLUSION: VI, even when mild, is associated with unemployment and underemployment. Future studies should investigate whether visual interventions could be used as part of a multipronged strategy to improve employment outcomes for the population.

Association of Hearing Aids and Cochlear Implants With Cognitive Decline and Dementia: A Systematic Review and Meta-analysis.

Importance: Hearing loss is associated with cognitive decline. However, it is unclear if hearing restorative devices may have a beneficial effect on cognition. Objective: To evaluate the associations of hearing aids and cochlear implants with cognitive decline and dementia. Data Sources: PubMed, Embase, and Cochrane databases for studies published from inception to July 23, 2021. Study Selection: Randomized clinical trials or observational studies published as full-length articles in peer-reviewed journals relating to the effect of hearing interventions on cognitive function, cognitive decline, cognitive impairment, and dementia in patients with hearing loss. Data Extraction and Synthesis: The review was conducted in accordance with Preferred Reporting Items for Systematic Review and Meta-analyses (PRISMA) reporting guidelines. Two authors independently searched the PubMed, Embase, and Cochrane databases for studies relating to the effect of hearing interventions on cognitive decline and dementia in patients with hearing loss. Main Outcomes and Measures: Maximally adjusted hazard ratios (HRs) were used for dichotomous outcomes and ratio of means for continuous outcomes. Sources of heterogeneity were investigated using sensitivity and subgroup analyses, and publication bias was assessed using visual inspection, the Egger test, and trim and fill. Results: A total of 3243 studies were screened; 31 studies (25 observational studies, 6 trials) with 137 484 participants were included, of which 19 (15 observational studies, 4 trials) were included in quantitative analyses. Meta-analysis of 8 studies, which had 126 903 participants, had a follow-up duration ranging from 2 to 25 years, and studied long-term associations between hearing aid use and cognitive decline, showed significantly lower hazards of any cognitive decline among hearing aid users compared with participants with uncorrected hearing loss (HR, 0.81; 95% CI, 0.76-0.87; I2 = 0%). Additionally, meta-analysis of 11 studies with 568 participants studying the association between hearing restoration and short-term cognitive test score changes revealed a 3% improvement in short-term cognitive test scores after the use of hearing aids (ratio of means, 1.03; 95% CI, 1.02-1.04, I2 = 0%). Conclusions and Relevance: In this meta-analysis, the usage of hearing restorative devices by participants with hearing loss was associated with a 19% decrease in hazards of long-term cognitive decline. Furthermore, usage of these devices was significantly associated with a 3% improvement in cognitive test scores that assessed general cognition in the short term. A cognitive benefit of hearing restorative devices should be further investigated in randomized trials.

Clinical efficacy and long-term immunogenicity of an early triple dose regimen of SARS-CoV-2 mRNA vaccination in cancer patients.

INTRODUCTION: Three doses of SARS-CoV-2 mRNA vaccines have been recommended for cancer patients to reduce the risk of severe disease. Anti-neoplastic treatment, such as chemotherapy, may affect long-term vaccine immunogenicity. METHOD: Patients with solid or haematological cancer were recruited from 2 hospitals between July 2021 and March 2022. Humoral response was evaluated using GenScript cPASS surrogate virus neutralisation assays. Clinical outcomes were obtained from medical records and national mandatory-reporting databases. RESULTS: A total of 273 patients were recruited, with 40 having haematological malignancies and the rest solid tumours. Among the participants, 204 (74.7%) were receiving active cancer therapy, including 98 (35.9%) undergoing systemic chemotherapy and the rest targeted therapy or immunotherapy. All patients were seronegative at baseline. Seroconversion rates after receiving 1, 2 and 3 doses of SARS-CoV-2 mRNA vaccination were 35.2%, 79.4% and 92.4%, respectively. After 3 doses, patients on active treatment for haematological malignancies had lower antibodies (57.3%±46.2) when compared to patients on immunotherapy (94.1%±9.56, P<0.05) and chemotherapy (92.8%±18.1, P<0.05). SARS-CoV-2 infection was reported in 77 (28.2%) patients, of which 18 were severe. No patient receiving a third dose within 90 days of the second dose experienced severe infection. CONCLUSION: This study demonstrates the benefit of early administration of the third dose among cancer patients.

Trainee participation does not adversely affect the safety of thyroid surgery: Systematic review and meta-analysis.

Surgical traineeship is essential but must be safe for patients. In thyroid surgery, surgeon volume correlates with improved clinical/economic outcomes. However, it is presently unclear how far does trainee participation affect post-thyroidectomy complication rates in real-world and randomized data. We systematically searched four databases for associations of trainee participation with any post-thyroidectomy outcome. We conducted univariate meta-analyses, sensitivity analyses, and assessed publication bias qualitatively and quantitatively. We included 1 randomized and 15 observational studies from 3755 records, comprising 34 774 thyroid surgical patients. Trainee participation was associated with 12 min longer operative time, but not higher complication rates (hypoparathyroidism, recurrent laryngeal nerve palsy, hematoma, blood loss, return to operating room, hospitalization duration, readmission, and mortality). Sensitivity, publication bias, and multivariate analyses did not change our findings. Real-world and limited randomized data suggest that trainee participation in thyroid surgery is safe, given adequate consultant supervision and appropriate case selection.