Large language models (LLM) are increasingly applied in the medical field, yet their clinical implementation faces numerous ethical and regulatory challenges. This paper reviews seven major ethical challenges: patient safety and accuracy, bias and fairness, privacy and data protection, transparency and explainability, accountability and legal liability, patient autonomy and informed consent, and the doctor-patient relationship and trust. At the regulatory level, international research indicates that United States currently lacks specific regulations for medical LLM use, while is exploring the regulation of high-risk LLM. The EU’s AI Act classifies medical AI as high-risk and imposes stringent compliance requirements. China has issued generative AI management measures and advocates industry standards, though its legal framework remains incomplete. Solutions include embedding ethical principles during model development, strengthening human-machine collaboration and manual oversight in clinical settings, establishing clear legal standards for accountability, safeguarding data privacy and security, implementing continuous monitoring and improvement, and deepening international cooperation and multidisciplinary governance.

OBJECTIVE Aimed at investigating whether reflux pharyngitis is an independent risk factor for the failure of type Ⅰ tympanoplasty for chronic otitis media.This is achieved by analyzing the relationship between the postoperative tympanic membrane healing in patients who underwent type Ⅰ tympanoplasty and pharyngolaryngeal reflux finding score(RFS).METHODS Patients who underwent type Ⅰ tympanoplasty in the Department of Otolaryngology Head and Neck Surgery,Nanshan People's Hospital,Shenzhen,China,from January 2023 to July 2024 were retrospectively included.All the patients received preoperative perfect nasal endoscopy,laryngoscopy,evaluation by the RFS questionnaire,preoperative otoscopy for tympanoplasty,pure tone hearing threshold,and temporal bone thin-layer CT examination.Postoperative otoscopic examination was performed to observe tympanic membrane healing and followed up for 3 months.The patients were divided into surgery success group and failure group based on the criterion of whether a complete tympanic membrane was formed by endoscopic examination within 3 months.The RFS scores of the two groups were statistically analyzed.RESULTS A total of 135 patients with an average age of 44.78 years(±12.22 years)took part in this study,with 60 males and 75 females included,and 68 left ears and 67 right ears involved.There were 120 patients in the surgery success group,and 15 patients in the failure group.Statistical analysis revealed that the RFS score of the patients in the tympanoplasty failure group was remarkably higher than that of the patients in the tympanoplasty success group.Moreover,there were significantly more cases with suspected reflux pharyngitis in the surgery failure group(P=0.007).Reflux-induced tympanic membrane lesion and reperforation mostly occurred in the central part of the tympanic membrane graft.CONCLUSION Reflux pharyngitis has been implicated with tympanoplasty failure,and thus may be a causative factor.Additionally,the RFS can be used to screen patients with chronic suppurative otitis media for suspected reflux pharyngitis.Findings from this work indicate that perioperative anti-reflux therapy,combined with dietary and lifestyle counselling for the patients who suffer from reflux pharyngitis and are about to undergo the tympanoplasty surgery may improve surgical success rate.

Objective:To summarize the clinicopathologic characteristics of malignant hypertension (MHT) patients with acute kidney injury (AKI) and application of renin-angiotensin-aldosterone system inhibitor (RAASi).Methods:It was a retrospective cohort study. The adult patients with MHT and AKI admitted to Peking University First Hospital from January 1, 2012 to July 14, 2022. The patients were categorized into RAASi group and non-RAASi group based on RAASi administration from AKI onset to discharge. The clinicopathological data between the two groups were compared, and application of RAASi was analyzed.Results:A total of 179 patients were enrolled with age of 31 (26, 37) years and 148 males (82.7%). Ninety-five patients (53.1%) received dialysis treatment. The common causes of MHT were essential hypertension (125 patients, 69.8%), renal hypertension (39 patients, 21.8%) and endocrine hypertension (7 patients, 3.9%). AKI severity distribution showed 41 patients (22.9%) in stage 1, 1 patient (0.5%) in stage 2 and 137 patients (76.5%) in stage 3. Among MHT patients, 94 patients (52.5%) had been treated with RAASi before AKI, and 13 patients (7.3%) discontinued RAASi after AKI. Among 85 patients (47.5%) without receiving RAASi treatment before AKI, 68 new patients (38.0%) received RAASi treatment after AKI, and 40 patients (22.3%) were treated with the support of dialysis. Compared with non-RAASI group ( n=30), proportions of chronic kidney disease ( χ2=6.324, P=0.012) and post-AKI hyperkalemia ( χ2=4.048, P=0.044) in RAASi group ( n=149) were lower, and the proportion of dialysis treatment ( χ2=5.638, P=0.018), admission diastolic blood pressure ( Z=-3.609, P<0.001) and maximum diastolic blood pressure during hospitalization ( Z=-1.978, P=0.048) were higher. There were no statistically significant differences in the rates of target blood pressure control and renal function recovery between the two groups during hospitalization (all P>0.05). During hospitalization, 64 patients received renal biopsies, of which 50 patients (78.1%) had typical MHT vascular lesions such as "onion skin" in renal arterioles. Twenty-seven patients (42.2%) were complicated with glomerular diseases, and IgA nephropathy was the most common type (85.2%, 23/27). The proportions of glomerular ischemia and sclerosis, endothelial cell proliferation and acute renal tubular injury in RAASi group ( n=54) were lower than those in non-RAASi group ( n=10), and proportions of thrombosis and "onion skin" change were higher than those in RAASi group ( n=10), but the differences were not statistically significant (all P>0.05). Renal function recovery occurred in 47 patients (26.3%) by discharge. Among 95 dialysis patients, 26 patients (27.4%) achieved dialysis independence at discharge. Conclusions:MHT patients with AKI exhibit severe renal pathology and short-term poor prognosis. RAASi is primarily prescribed to those with relatively better kidney function or those receiving dialysis support.

Objective:To investigate the correlation between statins and contrast-induced acute kidney injury (CI-AKI) and provide a reference basis for clinical practice.Methods:It was a retrospective cohort study. The adult patients were admitted to Peking University First Hospital from January 1, 2018, to December 31, 2020, and received at least one intravascular iodinated contrast administration during hospitalization. The clinical data of the patients were collected. The enrolled patients were divided into statin group and non-statin group according to statin exposure. The exposure of statins was defined as use of any type of statins within 48 hours before iodinated contrast administration. The primary outcome was in-hospital AKI defined as AKI developed after contrast administration and before discharge, with 30 days as the endpoint observation time, and the secondary outcome was post-contrast AKI (PC-AKI) defined as AKI onset within 72 hours after contrast administration. Cox regression model was applied to investigate the correlation between statin prescription prior to contrast administration and clinical outcomes. Pre-specified interaction analysis was conducted to examine modification effect of age, gender, baseline estimated glomerular filtration rate (eGFR), diabetes and the injection method of contrast.Results:Among 10 321 enrolled patients, the age was 63 (54, 71) years old, and 6 274 (60.8%) patients were males. There were 2 372 (23.0%) patients taking statins before the use of iodinated contrast agents, and the person-time incidence rate of in-hospital AKI was 2.5 per 1 000 person-days. The person-time incidence rate of statin users and statin non-users was 3.2 and 2.4 per 1 000 person-days, respectively. Compared with the non-statin group, age, serum creatinine and the proportions of males, admitted to the intensive care unit, lipid metabolism disorder, hypertension, diabetes, cerebrovascular diseases, cardiovascular diseases, using renin-angiotensin- aldosterone inhibitors, using diuretics, using non-steroidal anti-inflammatory drugs, using proton pump inhibitors, iodinated contrast administration via artery, eGFR<60 ml·min -1·(1.73 m 2) -1 were higher, while the proportions of general anesthesia surgery, severe liver diseases and tumors, and eGFR were lower in the statin group (all P<0.05). Among 10 321 patients, 5 867 patients had serum creatinine measurement within 72 hours after iodinated contrast administration, among which 70 patients (4.0 per 1 000 person-days) developed PC-AKI. Multivariate Cox regression analysis showed that statin use was an independent protective factor for in-hospital AKI ( HR=0.65, 95% CI 0.45?0.93, P=0.017) and PC-AKI ( HR=0.44, 95% CI 0.22?0.88, P=0.020). Subgroup analysis showed the significant interaction between diabetes and statin use ( P for interaction=0.039), and the protective effect of statins against in-hospital AKI was only observed in non-diabetic group ( HR=0.45, 95% CI 0.26?0.77). There were no significant differences in subgroups stratified by age, sex, baseline eGFR and the injection method of contrast (all P for interaction>0.05). Conclusions:Statin use prior to iodinated contrast administration is correlated with reduced risks of in-hospital AKI and PC-AKI in hospitalized patients, and the correlation between statin use and in-hospital AKI is more significant in non-diabetic patients. It is suggested that statin use before the application of iodinated contrast agents in hospitalized patients may prevent the occurrence of AKI.

Objective:To investigate the serum sodium level and its fluctuation in patients with hospitalized acquired acute kidney injury (AKI) and explore their impacts on in-hospital mortality.Methods:It was a single-center retrospective study. The adult patients developing hospital-acquired AKI and receiving at least twice serum sodium tests admitted to Peking University First Hospital from January 1, 2018, to December 31, 2020 were included. Dysnatremia included hyponatremia (< 135 mmol/L) and hypernatremia (>145 mmol/L). The patients were divided into hyponatremia group, normal serum sodium group and hypernatremia group, and the differences of clinical data among the three groups were compared. The fluctuation of serum sodium level was evaluated by coefficient of variation. A restricted cubic spline was applied to investigate the association between serum sodium level at AKI onset and mortality. Poisson regression analysis was used to explore the mortality risk of dysnatremia at AKI onset, dysnatremia at admission, and coefficient of variation of serum sodium, respectively.Results:Among the enrolled 1 475 AKI patients, the age was 66.0 (55.0, 78.0) years, and 850 patients (57.6%) were males. The estimated glomerular filtration rate was 77.3 (50.4, 97.6) ml·min -1·(1.73 m 2) -1. The time from admission to AKI onset was 8 (4, 15) days. The incidence of hyponatremia and hypernatremia at admission were 19.6% (289/1 475) and 2.6% (39/1 475), respectively, while the incidence at AKI onset was 24.0% (354/1 475) and 12.7% (188/1 475), respectively. There was statistically significant difference in terms of age, the initial classification distribution of AKI, serum sodium at admission, serum sodium at the occurrence of AKI, the lowest serum sodium at hospitalization, the highest serum sodium at hospitalization, the coefficient of variation of serum sodium, and the proportions of heart failure, stroke, disseminated intravascular coagulation, sepsis, acute respiratory distress syndrome, shock, prerenal causes, circle diuretics and aldosterone antagonists among hyponatremia group, normal serum sodium group and hypernatremia group (all P<0.05). The restricted cubic spline analysis showed a "U"-shaped correlation between serum sodium level at AKI onset and in-hospital mortality. Poisson regression analysis showed that after adjusting for age, gender, number of chronic comorbidities, initial classification of AKI, basal estimated glomerular filtration rate and number of acute disease state, with normal serum sodium as the reference, hyponatremia ( RR=1.56, 95% CI 1.14-2.13) and hypernatremia ( RR=1.71, 95% CI 1.23-2.39) at AKI onset were correlated with an increased risk of in-hospital mortality. Hyponatremia at admission was correlated with an increased risk of in-hospital mortality ( RR=2.13, 95% CI 1.62-2.79), while there was no statistically significant association between hypernatremia and in-hospital mortality ( RR=1.22, 95% CI 0.62-2.44). After further adjusting serum sodium levels at admission and at the occurrence of AKI, the coefficient of variation of serum sodium level was still correlated with an increased risk of in-hospital mortality ( RR=1.23, 95% CI 1.14-1.33). Conclusions:Dysnatremia is common in patients with hospital-acquired AKI. The serum sodium level at AKI onset is correlated with in-hospital death in a "U" shape. Dysnatremia and serum sodium fluctuation are associated with an increased risk of in-hospital mortality.

Objective:To summarize the clinicopathologic characteristics of malignant hypertension (MHT) patients with acute kidney injury (AKI) and application of renin-angiotensin-aldosterone system inhibitor (RAASi).Methods:It was a retrospective cohort study. The adult patients with MHT and AKI admitted to Peking University First Hospital from January 1, 2012 to July 14, 2022. The patients were categorized into RAASi group and non-RAASi group based on RAASi administration from AKI onset to discharge. The clinicopathological data between the two groups were compared, and application of RAASi was analyzed.Results:A total of 179 patients were enrolled with age of 31 (26, 37) years and 148 males (82.7%). Ninety-five patients (53.1%) received dialysis treatment. The common causes of MHT were essential hypertension (125 patients, 69.8%), renal hypertension (39 patients, 21.8%) and endocrine hypertension (7 patients, 3.9%). AKI severity distribution showed 41 patients (22.9%) in stage 1, 1 patient (0.5%) in stage 2 and 137 patients (76.5%) in stage 3. Among MHT patients, 94 patients (52.5%) had been treated with RAASi before AKI, and 13 patients (7.3%) discontinued RAASi after AKI. Among 85 patients (47.5%) without receiving RAASi treatment before AKI, 68 new patients (38.0%) received RAASi treatment after AKI, and 40 patients (22.3%) were treated with the support of dialysis. Compared with non-RAASI group ( n=30), proportions of chronic kidney disease ( χ2=6.324, P=0.012) and post-AKI hyperkalemia ( χ2=4.048, P=0.044) in RAASi group ( n=149) were lower, and the proportion of dialysis treatment ( χ2=5.638, P=0.018), admission diastolic blood pressure ( Z=-3.609, P<0.001) and maximum diastolic blood pressure during hospitalization ( Z=-1.978, P=0.048) were higher. There were no statistically significant differences in the rates of target blood pressure control and renal function recovery between the two groups during hospitalization (all P>0.05). During hospitalization, 64 patients received renal biopsies, of which 50 patients (78.1%) had typical MHT vascular lesions such as "onion skin" in renal arterioles. Twenty-seven patients (42.2%) were complicated with glomerular diseases, and IgA nephropathy was the most common type (85.2%, 23/27). The proportions of glomerular ischemia and sclerosis, endothelial cell proliferation and acute renal tubular injury in RAASi group ( n=54) were lower than those in non-RAASi group ( n=10), and proportions of thrombosis and "onion skin" change were higher than those in RAASi group ( n=10), but the differences were not statistically significant (all P>0.05). Renal function recovery occurred in 47 patients (26.3%) by discharge. Among 95 dialysis patients, 26 patients (27.4%) achieved dialysis independence at discharge. Conclusions:MHT patients with AKI exhibit severe renal pathology and short-term poor prognosis. RAASi is primarily prescribed to those with relatively better kidney function or those receiving dialysis support.

Objective:To investigate the correlation between statins and contrast-induced acute kidney injury (CI-AKI) and provide a reference basis for clinical practice.Methods:It was a retrospective cohort study. The adult patients were admitted to Peking University First Hospital from January 1, 2018, to December 31, 2020, and received at least one intravascular iodinated contrast administration during hospitalization. The clinical data of the patients were collected. The enrolled patients were divided into statin group and non-statin group according to statin exposure. The exposure of statins was defined as use of any type of statins within 48 hours before iodinated contrast administration. The primary outcome was in-hospital AKI defined as AKI developed after contrast administration and before discharge, with 30 days as the endpoint observation time, and the secondary outcome was post-contrast AKI (PC-AKI) defined as AKI onset within 72 hours after contrast administration. Cox regression model was applied to investigate the correlation between statin prescription prior to contrast administration and clinical outcomes. Pre-specified interaction analysis was conducted to examine modification effect of age, gender, baseline estimated glomerular filtration rate (eGFR), diabetes and the injection method of contrast.Results:Among 10 321 enrolled patients, the age was 63 (54, 71) years old, and 6 274 (60.8%) patients were males. There were 2 372 (23.0%) patients taking statins before the use of iodinated contrast agents, and the person-time incidence rate of in-hospital AKI was 2.5 per 1 000 person-days. The person-time incidence rate of statin users and statin non-users was 3.2 and 2.4 per 1 000 person-days, respectively. Compared with the non-statin group, age, serum creatinine and the proportions of males, admitted to the intensive care unit, lipid metabolism disorder, hypertension, diabetes, cerebrovascular diseases, cardiovascular diseases, using renin-angiotensin- aldosterone inhibitors, using diuretics, using non-steroidal anti-inflammatory drugs, using proton pump inhibitors, iodinated contrast administration via artery, eGFR<60 ml·min -1·(1.73 m 2) -1 were higher, while the proportions of general anesthesia surgery, severe liver diseases and tumors, and eGFR were lower in the statin group (all P<0.05). Among 10 321 patients, 5 867 patients had serum creatinine measurement within 72 hours after iodinated contrast administration, among which 70 patients (4.0 per 1 000 person-days) developed PC-AKI. Multivariate Cox regression analysis showed that statin use was an independent protective factor for in-hospital AKI ( HR=0.65, 95% CI 0.45?0.93, P=0.017) and PC-AKI ( HR=0.44, 95% CI 0.22?0.88, P=0.020). Subgroup analysis showed the significant interaction between diabetes and statin use ( P for interaction=0.039), and the protective effect of statins against in-hospital AKI was only observed in non-diabetic group ( HR=0.45, 95% CI 0.26?0.77). There were no significant differences in subgroups stratified by age, sex, baseline eGFR and the injection method of contrast (all P for interaction>0.05). Conclusions:Statin use prior to iodinated contrast administration is correlated with reduced risks of in-hospital AKI and PC-AKI in hospitalized patients, and the correlation between statin use and in-hospital AKI is more significant in non-diabetic patients. It is suggested that statin use before the application of iodinated contrast agents in hospitalized patients may prevent the occurrence of AKI.

Objective:To investigate the serum sodium level and its fluctuation in patients with hospitalized acquired acute kidney injury (AKI) and explore their impacts on in-hospital mortality.Methods:It was a single-center retrospective study. The adult patients developing hospital-acquired AKI and receiving at least twice serum sodium tests admitted to Peking University First Hospital from January 1, 2018, to December 31, 2020 were included. Dysnatremia included hyponatremia (< 135 mmol/L) and hypernatremia (>145 mmol/L). The patients were divided into hyponatremia group, normal serum sodium group and hypernatremia group, and the differences of clinical data among the three groups were compared. The fluctuation of serum sodium level was evaluated by coefficient of variation. A restricted cubic spline was applied to investigate the association between serum sodium level at AKI onset and mortality. Poisson regression analysis was used to explore the mortality risk of dysnatremia at AKI onset, dysnatremia at admission, and coefficient of variation of serum sodium, respectively.Results:Among the enrolled 1 475 AKI patients, the age was 66.0 (55.0, 78.0) years, and 850 patients (57.6%) were males. The estimated glomerular filtration rate was 77.3 (50.4, 97.6) ml·min -1·(1.73 m 2) -1. The time from admission to AKI onset was 8 (4, 15) days. The incidence of hyponatremia and hypernatremia at admission were 19.6% (289/1 475) and 2.6% (39/1 475), respectively, while the incidence at AKI onset was 24.0% (354/1 475) and 12.7% (188/1 475), respectively. There was statistically significant difference in terms of age, the initial classification distribution of AKI, serum sodium at admission, serum sodium at the occurrence of AKI, the lowest serum sodium at hospitalization, the highest serum sodium at hospitalization, the coefficient of variation of serum sodium, and the proportions of heart failure, stroke, disseminated intravascular coagulation, sepsis, acute respiratory distress syndrome, shock, prerenal causes, circle diuretics and aldosterone antagonists among hyponatremia group, normal serum sodium group and hypernatremia group (all P<0.05). The restricted cubic spline analysis showed a "U"-shaped correlation between serum sodium level at AKI onset and in-hospital mortality. Poisson regression analysis showed that after adjusting for age, gender, number of chronic comorbidities, initial classification of AKI, basal estimated glomerular filtration rate and number of acute disease state, with normal serum sodium as the reference, hyponatremia ( RR=1.56, 95% CI 1.14-2.13) and hypernatremia ( RR=1.71, 95% CI 1.23-2.39) at AKI onset were correlated with an increased risk of in-hospital mortality. Hyponatremia at admission was correlated with an increased risk of in-hospital mortality ( RR=2.13, 95% CI 1.62-2.79), while there was no statistically significant association between hypernatremia and in-hospital mortality ( RR=1.22, 95% CI 0.62-2.44). After further adjusting serum sodium levels at admission and at the occurrence of AKI, the coefficient of variation of serum sodium level was still correlated with an increased risk of in-hospital mortality ( RR=1.23, 95% CI 1.14-1.33). Conclusions:Dysnatremia is common in patients with hospital-acquired AKI. The serum sodium level at AKI onset is correlated with in-hospital death in a "U" shape. Dysnatremia and serum sodium fluctuation are associated with an increased risk of in-hospital mortality.

Objective Todeterminetherelationshipbetweenthetumorvolumeoftheperipherallungadenocarcinomawith maximum diameter≤3cmandlymphnodemetastasis(LNM).Methods TheMSCTmanifestationsof235subjectswhowerediagnosedasperipheral lungadenocarcinomawithmaximumdiameter≤3cm wereretrospectivelyanalyzed.Thesepatientsweregroupedaccordingtodifferent parametersincludingsmokinghistory,differentiation,tumorconsistencyandavailabilityoftumornecrosis.Tumorvolumeandratesof LNMamongthesegroupswerecompared.ROCanalysiswasusedtocalculatethecut-offvalueanddiagnosticaccuracy.Results (1) ThetumorvolumeofLNMgroupwaslargerthanthatofnoLNMgroup,cut-offvaluewas5.5cm3,andAUCwas0.76;(2)Therates ofLNMofthewell,moderate-well,moderate,moderate-poorandpoordifferentiationgroupswere0%,8.7%,17.7%,45.6%and46.7%respectively.Theratesofpuregroundglassopacity(p-GGO),mixedandsolidtumorwere0%,8.3%and29.3%respectively.The ratesofthetumorpresentandabsentofnecrosiswere47.8%,22.0%respectively(P＜0.05).Conclusion Usingthevolumeoftumor on MSCTtopredictLNMisanewnon-invasivewayofassessingLNM,withhighsensitivityandspecificity,whichcouldsupplymore imaginginformationforsurgeonstochoosethewayoflymphnodedissection.

To determine the multidetector computed tomography (MDCT) features as well as the anatomic-pathological basis in thyroid diseases involving the upper mediastinum, we performed a retrospective analysis of 49 patients who had thyroid diseases involving the upper mediastinum. In the study, 22 cases were nodular goiter, 13 cases were thyroid adenoma, and 14 cases were thyroid cancer. The relevance between MDCT appearances and their diffusing route of common thyroid diseases as well as the anatomic-pathological features in this region were evaluated. It was found that the lesions located in the upper anterior mediastinum, the upper posterior mediastinum, and both sides were 67.3% (33/49), 14.3% (7/49), 18.4% (9/49), respectively. Different diseases had their distinct MDCT features nodular goiter mainly showed localized and multiple nodules or tumor bulk (77.3%), thyroid adenoma mainly showed solitary tumor bulk (92.3%), and thyroid cancer mainly demonstrated solitary tumor bulk (57.1%), respectively. Among the 49 cases, 9 cases had cervical and/or mediastinal metastases in lymph nodes. The thyroid diseases involving the upper mediastinum most commonly occurred in the upper anterior mediastinum. The MDCT features and distribution of diffusing thyroid lesions in cervico-thoracic junctional region closely correlated with the anatomic-pathological characteristics in this region.
Goiter, Nodular ; diagnostic imaging ; pathology ; Humans ; Mediastinal Neoplasms ; diagnostic imaging ; pathology ; secondary ; Mediastinum ; diagnostic imaging ; pathology ; Multidetector Computed Tomography ; Thyroid Neoplasms ; diagnostic imaging ; pathology