Neurological diseases have high morbidity and disability rates， posing a severe threat to human health. Cli nical manifestations include motor， sensory， cognitive and conscious disorders. Chaihu jia longgu muli decoction is derived from Treatise on Febrile Diseases ， with the effects of harmonizing Shaoyang， activating Yang and clearing heat， and tranquilizing the mind. This paper systematically reviews the research progress in clinical application and mechanism of Chaihu jia longgu muli decoction in the field of neurological diseases. It has been found that the decoction shows favorable efficacy in various neurological diseases such as insomnia， depression， epilepsy， vertigo， migraine and vascular dementia. The specific mechanisms are related to regulating neurotransmitter levels， repairing neuronal function， alleviating neuroinflammation， improving mitochondrial dysfunction and regulating intestinal flora. In the future， standardized prospective follow-up cohorts should be established， and core outcome indicators should be clearly defined to strengthen the evidence base. Furthermore， multidisciplinary research should be leveraged to expand the therapeutic value of Chaihu jia longgu muli decoction in the management of neurological diseases.

Objective: To examine the prevalence of e-cigarette use and associated factors among primary and secondary school students in Pudong New Area, Shanghai, and to explore the pathways linking harm perception of e-cigarettes, interpersonal social influence, and attitudes toward e-cigarette use with experimentation behavior, in order to provide scientific basis for further optimizing the prevention and control strategies of e-cigarette among adolescents. Methods: From September to October 2025, a multi stage cluster random sampling method was used to select 5 144 primary and secondary school students aged 8-19 years from 47 primary and secondary schools in Pudong New Area, Shanghai, to conduct an anonymous questionnaire survey. The questionnaire collected information on e-cigarette use, harm perception of e-cigarette, interpersonal social influence, and attitudes toward e-cigarette use. The Chi-square test was applied to analyze the differences in cigarette and e-cigarette use behavior among primary and secondary school students with different demographic characteristics. Structural equation modeling (SEM) was constructed using Mplus 8.3 software, and the bootstrap method was utilized to test the mediating effects. Results: The reported rates of cigarette experimentation, e-cigarette experimentation, and current e-cigarette use among primary and secondary school students were 1.85%, 2.10%, and 0.70%, respectively. Higher reporting rates of e-cigarette experimentation were observed among boys (2.82%), secondary school students aged 16-19 (3.39%), senior high school students (3.05%), and those with weekly pocket money >100 yuan ( 6.11% ) ( χ 2=11.67, 8.61, 8.00, 54.18, all P <0.05). Structural pathway analysis results demonstrated that e-cigarette harm perception positively predicted interpersonal social influence ( β =0.61) and e-cigarette use attitude ( β = 0.53 ), and interpersonal social influence ( β =0.65) and e-cigarette use attitude ( β =0.25) further promoted e-cigarette experimentation behavior among primary and secondary school students (all P <0.01), with interpersonal social influence playing a major mediating role (indirect effect=0.40). Conclusions E-cigarette experimentation behavior among primary and secondary school students in Pudong New Area is jointly influenced by multiple psychosocial factors. Intervention strategies should strengthen interpersonal social factors such as family and peers on the basis of health education, thereby constructing a comprehensive prevention and control strategy for e-cigarette use among primary and secondary school students.

Objective:To assess the viability of reducing radiation dose, contrast media volume, injection flow rate and contrast medium concentration (quadruple low-dose protocol) by utilizing virtual monoenergetic images (VMI) in photon-counting detector CT (PCD-CT) for aortic CT angiography (CTA), while maintaining image quality in comparison to images obtained from energy-integrating detector CT (EID-CT).Methods:From April 2024 to June 2024, a total of 40 participants who underwent aortic CTA on PCD-CT were prospectively enrolled in the experimental group (PCD-CT group), while 40 patients with similar baseline characteristics who had previously undergone aortic CTA using EID-CT were retrospectively selected for the conventional group (EID-CT group). The EID-CT group used a tube voltage of 90 kVp, a contrast media volume of 60 ml of contrast, an injection flow rate of 3 ml/s, and a contrast concentration of 350 mgI/ml; the PCD-CT group used the QuantumPlus mode, with a tube voltage of 140 kVp, a total amount of iodine in the contrast media of 140 mgI/kg, and an injection flow rate=contrast media volume/(delay time+scan time), and a contrast media concentration of 320 mgI/ml. VMIs in PCD-CT group were reconstructed in 5-keV intervals ranging from 45 to 65 keV. The effective radiation dose and contrast injection protocols were recorded and compared between two groups. Objective image quality assessment was performed for each group. CT attenuation, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were measured at five anatomical locations (ascending aorta, aortic arch, descending aorta, abdominal aorta, and right common iliac artery), and image noise was recorded. Subjective image quality was independently evaluated by two readers using a 5-point Likert scale in a blinded manner. Based on data normality, the one-way ANOVA or Kruskal-Wallis test was used for image quality assessment, with Bonferroni-corrected post-hoc analysis for multiple comparisons.Results:There were no significant differences in the baseline characteristics between two groups (all P0.05). The PCD-CT group demonstrated significantly lower effective radiation dose [(3.88±0.65) mSv vs. (5.97±1.15)mSv], contrast media volume [(29.25±4.56) ml vs. 60 ml], and injection rate [(2.65±0.42) ml/s vs. 3 ml/s] than the EID-CT group, with reductions of 35%, 51%, and 12%, respectively (all P0.001). For objective image quality, except for the ascending aortic CT attenuation, the CT attenuation, SNR, and CNR of other vessels in the 55 keV PCD-CT group were comparable to those in the EID-CT group. Additionally, the difference in image noise between these two groups was not statistically significant ( P0.05). Concerning subjective image quality, at 55 keV, the PCD-CT group had similar image noise scores and vessel attenuation scores (both P0.05) and better visualization of renal artery branching ( P=0.001) compared to the EID-CT group. Conclusion:In comparison to EID-CT, the use of a 55 keV image in PCD-CT for aortic CTA has demonstrated reductions in radiation dose, contrast media volume, injection flow rate and contrast medium concentration, while maintaining image quality.

Objective:To evaluate the feasibility of head and neck vascular imaging using photon-counting detector computed tomography (PCD-CT) combined with a “quadruple lows” protocol—characterized by low contrast media volume, low iodine concentration, low injection rate, and low radiation dose—and to compare the image quality with that obtained by energy-integrating detector CT (EID-CT).Methods:A total of 105 patients with suspected cerebrovascular disease were prospectively enrolled at the First Affiliated Hospital of Zhengzhou University between April and June 2024. Patients were randomly assigned to three groups ( n=35). Group A underwent conventional head and neck CTA using EID-CT. Group B underwent PCD-CT with a protocol involving ultra-low contrast media volume, low iodine concentration, and low injection rate. Group C underwent PCD-CT with the full “quadruple low” protocol. Objective image quality parameters—including CT attenuation, image noise (standard deviation, SD), signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR)—were measured at the ascending aorta, common carotid artery, internal carotid artery, vertebral artery, basilar artery, posterior cerebral artery, and middle cerebral artery. Two radiologists independently rated subjective image quality using a 5-point Likert scale. Differences among groups were analyzed using one-way ANOVA and the Kruskal-Wallis test. Results:Compared to Group A [contrast volume: (42.78±6.64)ml], contrast agent volume was significantly reduced in Groups B and C[ (26.26±4.45) ml and (26.54±3.83)ml, respectively], demonstrating reductions of 39% and 38% (both P<0.01). The iodine concentration was 320 mg/ml in Groups B and C, lower than 350 mg/ml in Group A (8.5%). The injection rate was also reduced in Groups B and C [(3.39±0.61) and (3.55±0.51)ml/s, respectively] compared to Group A [(4.28±0.66) ml/s], with reductions of 21% and 17% (both P<0.01). The effective dose (ED) was similar between Groups A and B [(1.40±0.15) vs. (1.40±0.19)mSv, P>0.05], while Group C demonstrated a significantly lower ED [(0.99±0.09) mSv], with a reduction of 30% compared to Group A and 29% compared to Group B (both P<0.01).In terms of objective image quality, significant differences in image noise (SD) were observed among the three groups at the vertebral artery, internal carotid artery, posterior cerebral artery, and middle cerebral artery (all P<0.05). Groups B and C showed significantly lower SD compared to Group A ( P<0.05), with no significant difference between B and C ( P>0.05). SNR was significantly higher in Groups B and C than in Group A at multiple vascular segments (all P<0.05). CNR differed only at the internal carotid artery, where Groups B and C demonstrated superior performance compared to Group A ( P<0.05).Subjective image quality scores showed no significant difference between Groups A and C ( P>0.05), while Group B had significantly higher scores than both A and C ( P<0.05). All images were deemed diagnostically acceptable. Conclusion:Compared with conventional EID-CT, PCD-CT combined with a “quadruple lows” protocol enables substantial reductions in contrast media and radiation dose while further improving image quality in head and neck CTA.

Objective To analyze the respective abnormal brain regions and commonly altered brain regions of spontaneous brain ac-tivity in post-stroke aphasia(PSA)and post-stroke depression(PSD),and to explore the potential pathological mechanisms underlying single disease and comorbidity.Methods Literatures were retrieved from PubMed,Embase,Web of Science,Cochrane Library,CNKI,Wanfang data,VIP and SinoMed,from establishment to April 19th,2025.Resting-state functional magnetic resonance imaging(fMRI)studies were included if they focused on patients with PSA or PSD,with healthy controls(HC)serving as control group,and amplitude of low-frequency fluctuation(ALFF),fractional amplitude of low-frequency fluctu-ation(fALFF)and regional homogeneity(ReHo)were adopted as outcome indicator.Data regarding differential brain regions were extracted from the original studies,and a neuroimaging meta-analysis was conducted using SDM-PSI V6.23 beta.Results A total of 17 articles on PSA(339 patients and 351 HC)and five articles on PSD(102 patients and 149 HC)were included.The spontaneous brain activity in the left cerebellar area 9,right middle temporal gyrus and right insula was significantly higher in PSA patients than in HC(P<0.05),while activity in the right cerebellar area 6,left medial superior frontal gyrus(SFGmed),left middle frontal gyrus and right anterior cingulate/paracingulate gyrus was lower(P<0.05).For PSD,spontaneous brain activity was elevated in the right cuneus gyrus and right superior occipital gyrus compared to HC(P<0.05),while it was reduced in the left SFGmed and left inferior pa-rietal lobe(P<0.05).Additionally,spontaneous brain activity in the left medial superior frontal gyrus was lower in both diseases compared to HC(P<0.05).Conclusion The pathological mechanism of PSA may involve a pattern of right-sided compensatory hyperfunction and left-sided inhibitory damage within the language-related network,accompanied by cross-hemispheric synergistic activity of the cerebellum.The occurrence of PSD may be attributed to hyperactivity of the right occipital net-work,and functional inhibition involving the left inferior parietal lobe.The inhibition observed in the left SFG-med(MNI coordinates x=0,y=26,z=44)likely represents the language-emotion integration hub,mediating the co-occurrence of PSA and PSD.

Objective To analyze the respective abnormal brain regions and commonly altered brain regions of spontaneous brain ac-tivity in post-stroke aphasia(PSA)and post-stroke depression(PSD),and to explore the potential pathological mechanisms underlying single disease and comorbidity.Methods Literatures were retrieved from PubMed,Embase,Web of Science,Cochrane Library,CNKI,Wanfang data,VIP and SinoMed,from establishment to April 19th,2025.Resting-state functional magnetic resonance imaging(fMRI)studies were included if they focused on patients with PSA or PSD,with healthy controls(HC)serving as control group,and amplitude of low-frequency fluctuation(ALFF),fractional amplitude of low-frequency fluctu-ation(fALFF)and regional homogeneity(ReHo)were adopted as outcome indicator.Data regarding differential brain regions were extracted from the original studies,and a neuroimaging meta-analysis was conducted using SDM-PSI V6.23 beta.Results A total of 17 articles on PSA(339 patients and 351 HC)and five articles on PSD(102 patients and 149 HC)were included.The spontaneous brain activity in the left cerebellar area 9,right middle temporal gyrus and right insula was significantly higher in PSA patients than in HC(P<0.05),while activity in the right cerebellar area 6,left medial superior frontal gyrus(SFGmed),left middle frontal gyrus and right anterior cingulate/paracingulate gyrus was lower(P<0.05).For PSD,spontaneous brain activity was elevated in the right cuneus gyrus and right superior occipital gyrus compared to HC(P<0.05),while it was reduced in the left SFGmed and left inferior pa-rietal lobe(P<0.05).Additionally,spontaneous brain activity in the left medial superior frontal gyrus was lower in both diseases compared to HC(P<0.05).Conclusion The pathological mechanism of PSA may involve a pattern of right-sided compensatory hyperfunction and left-sided inhibitory damage within the language-related network,accompanied by cross-hemispheric synergistic activity of the cerebellum.The occurrence of PSD may be attributed to hyperactivity of the right occipital net-work,and functional inhibition involving the left inferior parietal lobe.The inhibition observed in the left SFG-med(MNI coordinates x=0,y=26,z=44)likely represents the language-emotion integration hub,mediating the co-occurrence of PSA and PSD.

ObjectiveTo investigate whether the effect of Taohong Siwutang on cerebral ischemia-reperfusion （CIRI） injury in rats is related to the regulation of astrocyte polarization and explore the related mechanism. MethodsEighty-four male SD rats were randomly assigned to the following groups： A sham operation group， a model group， Taohong Siwutang treatment groups （low dose， medium dose， and high dose）， ligustrazine phosphate tablet （LPT） group， and AG490 group. All groups， except for the sham operation group， underwent middle cerebral artery occlusion/reperfusion （MCAO/R） modeling and were treated for seven days. The neurological impairment was evaluated using the Longa score. The volume of cerebral infarction was assessed through 2，3，5-triphenyltetrazolium chloride （TTC） staining. Real-time fluorescent quantitative polymerase chain reaction （Real-time PCR） and Western blot analyses were performed to analyze the mRNA and protein expression levels of cortical complement 3 （C3）， S100 calcium-binding protein A10 （S100A10）， Janus kinase 2 （JAK2）， and signal transducer and activator of transcription 3 （STAT3）. Additionally， protein expression levels of vascular endothelial growth factor-A （VEGF-A） were assessed， and the mRNA expression levels of inflammatory factors， including interleukin-6 （IL-6）， interleukin-1β （IL-1β）， and tumor necrosis factor-α （TNF-α）， were evaluated. Glial fibrillary acidic protein （GFAP） and C3， S100A10 and Co-localization was detected via immunofluorescence double staining. Lastly， VEGF expression levels were measured using enzyme-linked immunosorbent assay （ELISA）. ResultsCompared with the sham operation group， the model group showed a significant increase in cerebral infarction volume and neurological impairment （P<0.01）. C3 protein levels were elevated， while S100A10 levels were decreased. Pathway-related markers were significantly upregulated （P<0.05， P<0.01）， and VEGF-A protein levels were significantly reduced （P<0.01）. The mRNA expression of inflammatory factors was significantly upregulated （P<0.01）. Co-localization analysis showed significantly increased GFAP and C3 fluorescence intensity （P<0.01） and greatly decreased GFAP and S100A10 fluorescence intensity （P<0.01）. Additionally， VEGF content was significantly elevated （P<0.01）. Compared with the model group， medium- and high-dose Taohong Siwutang and LPT groups exhibited a significant reduction in cerebral infarction volume and neurological impairment （P<0.01）. Groups treated with low， medium， and high doses of Taohong Siwutang and LPT group exhibited a decrease in C3 protein expression levels and an increase in S100A10 expression levels （P<0.01）. In the high-dose Taohong Siwutang and AG490 groups， both protein and mRNA expression of C3 and pathway-related markers were significantly downregulated （P<0.05， P<0.01）， while S100A10 expression and VEGF-A protein levels were significantly increased （P<0.01）. Additionally， the mRNA expression levels of inflammatory factors were significantly reduced （P<0.01）. The co-localization fluorescence intensity of GFAP and C3 significantly decreased （P<0.01）， while that of GFAP and S100A10 greatly increased （P<0.01）. Furthermore， VEGF content exhibited a marked elevation （P<0.01）. ConclusionTaohong Siwutang exerts a protective effect in rats with cerebral CIRI injury. The underlying mechanism is associated with the downregulation of the JAK2/STAT3 signaling pathway， promotion of A2-type astrocyte polarization， reduction of inflammatory factor release， and enhancement of VEGF production.

ObjectiveTo investigate whether the effect of Taohong Siwutang on cerebral ischemia-reperfusion （CIRI） injury in rats is related to the regulation of astrocyte polarization and explore the related mechanism. MethodsEighty-four male SD rats were randomly assigned to the following groups： A sham operation group， a model group， Taohong Siwutang treatment groups （low dose， medium dose， and high dose）， ligustrazine phosphate tablet （LPT） group， and AG490 group. All groups， except for the sham operation group， underwent middle cerebral artery occlusion/reperfusion （MCAO/R） modeling and were treated for seven days. The neurological impairment was evaluated using the Longa score. The volume of cerebral infarction was assessed through 2，3，5-triphenyltetrazolium chloride （TTC） staining. Real-time fluorescent quantitative polymerase chain reaction （Real-time PCR） and Western blot analyses were performed to analyze the mRNA and protein expression levels of cortical complement 3 （C3）， S100 calcium-binding protein A10 （S100A10）， Janus kinase 2 （JAK2）， and signal transducer and activator of transcription 3 （STAT3）. Additionally， protein expression levels of vascular endothelial growth factor-A （VEGF-A） were assessed， and the mRNA expression levels of inflammatory factors， including interleukin-6 （IL-6）， interleukin-1β （IL-1β）， and tumor necrosis factor-α （TNF-α）， were evaluated. Glial fibrillary acidic protein （GFAP） and C3， S100A10 and Co-localization was detected via immunofluorescence double staining. Lastly， VEGF expression levels were measured using enzyme-linked immunosorbent assay （ELISA）. ResultsCompared with the sham operation group， the model group showed a significant increase in cerebral infarction volume and neurological impairment （P<0.01）. C3 protein levels were elevated， while S100A10 levels were decreased. Pathway-related markers were significantly upregulated （P<0.05， P<0.01）， and VEGF-A protein levels were significantly reduced （P<0.01）. The mRNA expression of inflammatory factors was significantly upregulated （P<0.01）. Co-localization analysis showed significantly increased GFAP and C3 fluorescence intensity （P<0.01） and greatly decreased GFAP and S100A10 fluorescence intensity （P<0.01）. Additionally， VEGF content was significantly elevated （P<0.01）. Compared with the model group， medium- and high-dose Taohong Siwutang and LPT groups exhibited a significant reduction in cerebral infarction volume and neurological impairment （P<0.01）. Groups treated with low， medium， and high doses of Taohong Siwutang and LPT group exhibited a decrease in C3 protein expression levels and an increase in S100A10 expression levels （P<0.01）. In the high-dose Taohong Siwutang and AG490 groups， both protein and mRNA expression of C3 and pathway-related markers were significantly downregulated （P<0.05， P<0.01）， while S100A10 expression and VEGF-A protein levels were significantly increased （P<0.01）. Additionally， the mRNA expression levels of inflammatory factors were significantly reduced （P<0.01）. The co-localization fluorescence intensity of GFAP and C3 significantly decreased （P<0.01）， while that of GFAP and S100A10 greatly increased （P<0.01）. Furthermore， VEGF content exhibited a marked elevation （P<0.01）. ConclusionTaohong Siwutang exerts a protective effect in rats with cerebral CIRI injury. The underlying mechanism is associated with the downregulation of the JAK2/STAT3 signaling pathway， promotion of A2-type astrocyte polarization， reduction of inflammatory factor release， and enhancement of VEGF production.

Objective:To investigate the cognitive characteristics and related influencing factors in Parkinson′s disease (PD) patients with or without olfactory anosognosia (OA).Methods:A total of 113 PD patients who were treated at the Affiliated Hospital of Yangzhou University between March 2023 and April 2024 were selected. The PD Olfactory Dysfunction Auxiliary Diagnostic Card was used to assess olfactory function. Based on the olfactory identification scores and subjective awareness of olfactory dysfunction, patients were divided into the normosmic group, olfactory dysfunction group, and the later was further divided into olfactory dysfunction without OA (OA-) group, and olfactory dysfunction with OA (OA+) group. The results of the Unified Parkinson′s Disease Rating Scale-Ⅲ (UPDRS-Ⅲ) and Hoehn-Yahr (H-Y) staging assessments of the patients were collected. Non-motor symptoms such as cognitive function, anxiety, depression, sleep disturbances, and constipation were evaluated using relevant scales. Logistic regression analysis was used to explore the related factors affecting OA in PD patients with olfactory decline.Results:The Montreal Cognitive Assessment (MoCA) scores of the olfactory dysfunction group were lower than those of the normosmic group (20.30±4.47 vs 22.64±2.50, t=2.907, P=0.007). The Self-Rating Anxiety Scale scores (39.00±8.60 vs 43.86±10.63, t=2.444, P=0.016), visuospatial and executive function scores (2.35±1.32 vs 2.98±1.42, t=2.263, P=0.026), and orientation scores (4.88±1.14 vs 5.34±1.07, t=2.046, P=0.043) of the OA+ group were lower than those of the OA- group. Logistic regression analysis revealed that lower MoCA scores were an independent risk factor for PD combined with OA ( OR=0.853, 95% CI 0.743-0.980, P=0.024). Conclusions:PD patients with olfactory dysfunction exhibit more severe cognitive impairment. Among them, patients with OA show more significant impairments in visuospatial, executive function and orientation. Cognitive impairment may be an independent risk factor for PD combined with OA.

Objective:To investigate the long-term efficacy, safety and prognostic factors of intraoperative radiotherapy (IORT) for narrow-margin (resection margin < 1 cm) hepatectomy in patients with hepatocellular carcinoma (HCC) during radical surgery.Methods:A retrospective cohort study was conducted. The data of primary HCC patients undergoing radical surgery and narrow-margin hepatectomy IORT in the Cancer Hospital of the Chinese Academy of Medical Sciences from November 2009 to February 2019 were collected. IORT applied 6 MeV or 9 MeV electron beams and a single irradiation was given to the margin. Kaplan-Meier method was used for the overall survival (OS) and disease-free survival (DFS) analysis; log-rank test was used for survival comparison among subgroups. The recurrence patterns and adverse reactions were recorded. Univariate and multivariate Cox proportional hazards models were used to analyze the factors influencing the OS and DFS.Results:A total of 64 patients were enrolled, with the median age [ M ( Q1, Q3)] of 57 years (49, 63) years. All patients included 55 males (85.9%) and 9 females (14.1%). The median dose of IORT was 15 Gy (range: 12-17 Gy). The median follow-up time was 83.3 (64.4, 91.9) months. The 1-year, 3-year, 5-year, 7-year, 10-year OS rates were 90.4%, 80.6%, 75.5%, 71.4% and 47.6%, respectively; the 1-year, 3-year, 5-year, 7-year,10-year DFS rates were 77.8%, 68.1%, 59.6%, 57.6% and 38.4%, respectively. Univariate Cox regression analysis indicated that preoperative serum alpha-fetoprotein (AFP) > 400 ng/ml was an independent risk factor for poor OS (> 400 ng/ml vs. ≤ 400 ng/ml: HR = 6.57, 95% CI: 2.16-19.96, P < 0.001), while not the independent influencing factor of poor DFS ( HR = 1.71, 95% CI: 0.65-4.52, P = 0.277). The age ≤ 60 years or not, gender, viral hepatitis or not, American Joint Committee on Cancer stage, tumor diameter (> 5 cm or not), tumor number, degree of tumor differentiation, microvascular invasion or not, microsatellite nodules or not, anatomical liver resection or not, and the dose of IORT ≤15 Gy or not were not the independent influencing factors of poor OS and DFS (all P > 0.05). Kaplan-Meier method analysis showed that patients with preoperative serum AFP ≤ 400 ng/ml (48 cases) had better OS compared with those with preoperative serum AFP>400 ng/ml (16 cases) (5-year OS rate: 84.8% vs. 44.9%; 7-year OS rate: 79.9% vs.37.4%), and the difference was statistically significant ( P = 0.002). There was no statistically significant difference in the DFS between the 2 groups ( P = 0.134). During the follow-up, 28 patients (43.8%) relapsed, including 17 cases (26.6%) of early recurrence and 11 cases (17.2%) of late recurrence. No marginal recurrence was observed. There were 22 cases (34.4%) of intrahepatic recurrence alone, 2 cases (3.1%) of extrahepatic recurrence and 4 cases (6.3%) of stimutaneous recurrence inside and outside the liver. The 1-, 3-, 5- and 7-year cumulative recurrence rates inside the liver were 19.0%, 27.2%, 37.4% and 39.3% respectively, and the cumulative recurrence rates outside the liver were 6.4%, 8.0%, 9.6% and 9.6% respectively. There were no adverse reactions above grade 3 in the entire group. There were no surgery-related deaths within 30 d after the operation, and no radiation-induced liver disease occurred. Conclusions:Narrow-margin IORT helps HCC patients receiving hepatectomy to achieve favorable long-term survival and adverse reactions are tolerable. It can be used as a safe and effective adjuvant therapy alternative.