Image Reconstruction for Accelerated MR Scan With Faster Fourier Convolutional Neural Networks.

High quality image reconstruction from undersampled k -space data is key to accelerating MR scanning. Current deep learning methods are limited by the small receptive fields in reconstruction networks, which restrict the exploitation of long-range information, and impede the mitigation of full-image artifacts, particularly in 3D reconstruction tasks. Additionally, the substantial computational demands of 3D reconstruction considerably hinder advancements in related fields. To tackle these challenges, we propose the following: 1) A novel convolution operator named Faster Fourier Convolution (FasterFC), aims at providing an adaptable broad receptive field for spatial domain reconstruction networks with fast computational speed. 2) A split-slice strategy that substantially reduces the computational load of 3D reconstruction, enabling high-resolution, multi-coil, 3D MR image reconstruction while fully utilizing inter-layer and intra-layer information. 3) A single-to-group algorithm that efficiently utilizes scan-specific and data-driven priors to enhance k -space interpolation effects. 4) A multi-stage, multi-coil, 3D fast MRI method, called the faster Fourier convolution based single-to-group network (FAS-Net), comprising a single-to-group k -space interpolation algorithm and a FasterFC-based image domain reconstruction module, significantly minimizes the computational demands of 3D reconstruction through split-slice strategy. Experimental evaluations conducted on the NYU fastMRI and Stanford MRI Data datasets reveal that the FasterFC significantly enhances the quality of both 2D and 3D reconstruction results. Moreover, FAS-Net, characterized as a method that can achieve high-resolution (320, 320, 256), multi-coil, (8 coils), 3D fast MRI, exhibits superior reconstruction performance compared to other state-of-the-art 2D and 3D methods.

Association between Coronary Artery Disease Reporting and Data System-recommended Post-Coronary CT Angiography Management and Clinical Outcomes in Patients with Stable Chest Pain from a Chinese Registry.

Background Coronary Artery Disease Reporting and Data System (CAD-RADS) was developed to standardize and optimize disease management in patients after coronary CT angiography (CCTA), but the impact of CAD-RADS management recommendations on clinical outcomes remains unclear. Purpose To retrospectively assess the association between the appropriateness of post-CCTA management according to CAD-RADS version 2.0 and clinical outcomes. Materials and Methods From January 2016 to January 2018, consecutive participants with stable chest pain referred for CCTA were prospectively included in a Chinese registry and followed for 4 years. Retrospectively, CAD-RADS 2.0 classification and the appropriateness of post-CCTA management were determined. Propensity score matching (PSM) was used to adjust for confounding variables. Hazard ratios (HRs) for a major adverse cardiovascular event (MACE), relative risks for invasive coronary angiography (ICA), and the corresponding number needed to treat were estimated. Results Of the 14 232 included participants (mean age, 61 years ± 13 [SD]; 8852 male), 2330, 2756, and 2614 were retrospectively categorized in CAD-RADS 1, 2, and 3, respectively. Only 26% of participants with CAD-RADS 1-2 disease and 20% with CAD-RADS 3 received appropriate post-CCTA management. After PSM, appropriate post-CCTA management was associated with lower risk of MACEs (HR, 0.34; 95% CI: 0.22, 0.51; P < .001), corresponding to a number needed to treat of 21 in CAD-RADS 1-2 but not CAD-RADS 3 (HR, 0.86; 95% CI: 0.49, 1.85; P = .42). Appropriate post-CCTA management was associated with decreased use of ICA in CAD-RADS 1-2 (relative risk, 0.40; 95% CI: 0.29, 0.55; P < .001) and 3 (relative risk, 0.33; 95% CI: 0.28, 0.39; P < .001), resulting in a number needed to treat of 14 and 2, respectively. Conclusion In this retrospective secondary analysis, appropriate disease management after CCTA according to CAD-RADS 2.0 was associated with lower risk of MACEs and more prudent use of ICA. ClinicalTrials.gov registration no. NCT04691037 © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Leipsic and Tzimas in this issue.

Local Axial Scale-Attention for Universal Lesion Detection.

Universal lesion detection (ULD) in computed tomography (CT) images is an important and challenging prerequisite for computer-aided diagnosis (CAD) to find abnormal tissue, such as tumors of lymph nodes, liver tumors, and lymphadenopathy. The key challenge is that lesions have a tiny size and high similarity with non-lesions, which can easily lead to high false positives. Specifically , non-lesions are nearby normal anatomy that include the bowel, vasculature, and mesentery, which decrease the conspicuity of small lesions since they are often hard to differentiate. In this study, we present a novel scale-attention module that enhances feature discrimination between lesion and non-lesion regions by utilizing the domain knowledge of radiologists to reduce false positives effectively. Inspired by the domain knowledge that radiologists tend to divide each CT image into multiple areas, then detect lesions in these smaller areas separately, a local axial scale-attention (LASA) module is proposed to re-weight each pixel in a feature map by aggregating local features from multiple scales adaptively. In addition, to keep the same weight, a combination of axial pixels in the height- and width-axes is designed, attached with position embedding. The model can be used in CNNs easily and flexibly. We test our method on the DeepLesion dataset. The sensitivities at 0.5, 1, 2, 4, 8, and 16 false positives (FPs) per image and average sensitivity at [0.5, 1, 2, 4] are calculated to evaluate the accuracy. The sensitivities are 78.30%, 84.96%, 89.86%, 93.14%, 95.36%, and 95.54% at 0.5, 1, 2, 4, 8, and 16 FPs per image; the average sensitivity is 86.56%, outperforming the former methods. The proposed method enhances feature discrimination between lesion and non-lesion regions by adding LASA modules. These encouraging results illustrate the potential advantage of exploiting the domain knowledge for lesion detection.

Optimized Machine Learning Models to Predict In-Hospital Mortality for Patients with ST-Segment Elevation Myocardial Infarction.

PURPOSE: This study aimed to optimize machine learning (ML) models for predicting in-hospital mortality in patients with ST-segment elevation acute myocardial infarction (STEMI). PATIENTS AND METHODS: A total of 5708 STEMI patients were enrolled and divided into two groups according to patients' hospital outcomes. Both groups were randomly split into a training set (75%) and a testing set (25%). Four ML models were trained with data, which applied random under-sampling (RUS). The performance of optimized ML models was evaluated with respect to accuracy, sensitivity, specificity, G-mean and AUC. Two sets of features in chronological order were considered: a full set that included all variables during hospitalization and a simplified set that only included variables prior to reperfusion therapy, and the performance of the prediction models trained with these two sets of features was compared. RESULTS: For the comprehensive metric - G-mean, the models trained with RUS outperformed those without, 80.54% vs 23.31% on average in the full set and 75.72% vs 35.76% on average in the simplified set. For models trained with the full set, the SVM achieved the best performance with 85.62% accuracy, 84.21% sensitivity, 85.66% specificity, 84.93% G-mean and 0.919 AUC. For models trained with the simplified set, the SVM achieved 83.48% G-mean, which was comparable to the models trained using the full set. For the most critical metric - sensitivity, the SVM trained using the simplified set achieved 89.47%, which even exceed the SVM (84.21%), DT (81.58%) and RF (81.58%) trained using the full set. CONCLUSION: Applying RUS can improve the performance of prediction models, and the models trained with simplified set, which only included variables prior to reperfusion therapy can accurately predict high-risk patients.

Impact of subthalamic nucleus stimulation did not differ on young-onset and older-onset Parkinson's disease: A three-year follow up.

OBJECTIVE: To assess the role of onset age in the results of bilateral subthalamic nucleus deep brain stimulation (STN-DBS), we carried out a retrospective study of two groups of patients regarding age at disease onset. METHODS: We compared, up to 3 years after surgery, the clinical effects, quality of life and the levodopa equivalent daily dose (LEDD) in patients with young-onset Parkinson's disease (onset age <50 years, YOPD) vs patients with older-onset Parkinson's disease (onset age ≥50 years, OOPD). RESULTS: A dramatic improvement in motor symptoms was equally observed in both groups of patients after DBS. The improvements of the Unified Parkinson's Disease Rating Scale part III motor scale (UPDRS-III) score, axial sub-score and non-axial sub-score from baseline gradually decreased over time. The benefit of STN-DBS for the axial symptoms decreased most rapidly, which directly resulted in a progressive decline in stimulation efficacy in both groups. Nevertheless, the improvement in non-axial symptoms after DBS was remarkable and long-lasting. The quality of life in both groups were also improved after DBS but were slightly decreased in the following years. The reduced LEDD were equivalent in both groups. CONCLUSIONS: STN-DBS alleviates motor symptoms and improves quality of life equally in both YOPD and OOPD patients with similar LEDD. The initial therapeutic benefit of STN-DBS for PD gradually decreases over time, mainly due to the progression of PD and the rapid withdrawal of the benefit for axial symptoms.

Depth Super-resolution from RGB-D Pairs with Transform and Spatial Domain Regularization.

This paper proposes a depth super-resolution method with both transform and spatial domain regularization. In the transform domain regularization, nonlocal correlations are exploited via an auto-regressive model, where each patch is further sparsified with a locally-trained transform to consider intra-patch correlations. In the spatial domain regularization, we propose a multi-directional total variation (MTV) prior to characterize the geometrical structures spatially orientated at arbitrary directions in depth maps. To achieve adaptive regularization, the MTV is weighted for each directional finite difference considering local characteristics of RGB-D data. We develop an accelerated proximal gradient algorithm to solve the proposed model. Quantitative and qualitative evaluations compared with state-of-the-art methods demonstrate that the proposed method achieves superior depth super-resolution performance for various configurations of magnification factors and datasets.

Successful surgery for refractory seizures associated with bilateral schizencephaly: two case reports and literature review.

Schizencephaly is a rare malformation of cortical development resulting from cell migration defects that occur unilaterally or bilaterally. The type of the schizencephalic cleft can be open lip or closed lip. Patients suffering from refractory seizures secondary to schizencephaly should be considered for surgical treatment. In this paper, we retrospectively analyzed two patients with confirmed schizencephaly and intractable seizures. The evaluation methods included a medical history assessment, a neurological examination and magnetic resonance imaging (MRI). Continuous intracranial video-electroencephalogram (vEEG) monitoring with surface electrodes and deep electrodes was evaluated to confirm the epileptogenic zones associated with the schizencephalic lesions. Cortical electrical stimulation was performed to evaluate the neurophysiology of the relevant brain regions. Epileptic focus resection was performed close to the schizencephalic cleft according to the results of intracranial EEG and stimulation while preserving neurological functions. MRI revealed bilateral open lip schizencephaly in one patient and closed lip schizencephaly in the other patient. The epileptogenic zones were localized close to the schizencephalic clefts. The seizure outcome was Engel's class Ia in both patients at 1-year follow-up. No significant neurological deficits were found, and their activities of daily life were significantly improved. We conclude that abnormal cortex near the schizencephalic clefts may display an extrinsic epileptogenicity. Accurate localization of the epileptogenic zones using intracranial EEG and electrical stimulation can lead to a seizure-free outcome in patients with refractory epilepsy associated with schizencephaly.

[Therapeutic effect of gamma knife on intracranial cavernous angioma].

OBJECTIVE: To evaluate the therapeutic effect of gamma knife on patients with intracranial cavernous angioma (CA). METHODS: The medical records of 122 patients (134 lesions) who underwent radiosurgery were reviewed retrospectively. RESULTS: Th e average follow-up period was 43 months. No patient died. One patient underwent CA resection. In patients with epilepsy, 83% patients showed alleviation of seizures. About 44% of the lesions shrank in size after treatment with gamma knife radiosurgery (59/134). Seven cases had hemorrhage again after radiosurgery, and the overall annual hemorrhage rate was 1.6%. Edema was found in 11.5% patients (14/122) and all patients showed improvement aft er treatment. CONCLUSION: Gamma knife is a safe treatment for CA, which could obviously improve the symptoms of epilepsy. Gamma knife radiosurgery is the fi rst option for the treatment of cavernous sinus angiomas.

[Efficacy and prognosis of trigeminal neuralgia treated with surgical excision or gamma knife surgery].

OBJECTIVE: To evaluate the curative effects and complications when using microvascular decompression (MVD) or gamma knife surgery (GKS) to treat trigeminal neuralgia (TN). And to investigate the prognosis for TN after these treatments. METHODS: Sixty-one TN patients treated using MVD and eighty-six TN patients treated using GKS were enrolled by means of telephone, letter or out-patient recheck; the patients had originally presented from December 1996 to June 2010. A chi-square test was applied to follow-up data on pain relief after 1 month, 6 months and 12 months, and at the final follow-up. RESULTS: One hundred and two patients were followed for 3-151 months. Cumulative pain relief rates at 1, 6, 12 months and at end times were 90.48%, 95.24%, 92.86%, and 95.24%, respectively, for forty-two MVD-treated patients, and 23.33%, 83.33%, 86.67% and 90% for sixty GKS-treated patients, respectively. The effects of two methods for treatment of TN was not statistically different (χ(2) = 2.053, P=0.152). Pain relief rates in the short-term (first month) demonstrated statistically significant differences (P<0.01), but pain relief rates in the long-term showed no significant differences (P>0.05). CONCLUSION: Immediate pain relief with MVD treatment is higher than with GKS, but in the long term both treatments were comparable.

[Comparison of prognosis of skull base chordoma treated by surgical resection and gamma knife surgery].

OBJECTIVE: To analyze the complication, survival and life quality, and to explore the efficacy and prognosis of skull base chordoma treated by open surgery and gamma knife surgery (GKS). METHODS: Thirty-eight patients with skull base chordomas who underwent an open surgery and another 19 patients who underwent GKS between January 2002 and January 2010 were followed up. The Karnofsky performance scale (KPS) at admission,discharge and follow-up, and recurrent rates were calculated through SPSS life-table, and survival rate curve was used to evaluate the efficacy and prognosis. RESULTS: Forty patients were followed up, and the survival rate of 1, 3 and 5 years was 92.6%, 74.3% and 60% in 27 patients with open surgery, and 100%, 83.3% and 62.5% in 13 patients with GKS respectively. There was no significant difference between the 2 groups (χ(2)=0.867, P=0.353). There was no significant difference in the KPS score between the 2 groups (P=0.138). CONCLUSION: Despite the poor prognosis of skull base chordoma, GKS can effectively improve the survival rate and present life quality of patients with skull base chordoma.

[Surgical outcomes of pediatric symptomatic epilepsy and their influencing factors].

OBJECTIVE: To investigate the surgical outcomes of pediatric symptomatic epilepsy and the influencing factors for postoperative outcomes. METHODS: A cohort of 48 children with symptomatic epilepsy received surgical treatment from October 2004 to September 2008. The surgical outcomes were followed up. RESULTS: A 27.3 months (range 12-51 months) follow-up was performed in 43 cases. Engel classification for evaluating postoperative epileptic outcomes showed that class I in 32 cases (74%), class II in 4 cases (9%), class III in 4 cases (9%) and class IV in 3 cases (7%). Preoperative seizure frequency is an independent predictor of postoperative epileptic outcomes (P<0.05). CONCLUSIONS: Operative treatment can lead to a favorable result in children with symptomatic epilepsy. Preoperative seizure frequency is an independent influencing factor for postoperative outcomes.

[Diagnosis and treatment of cystic meningioma].

OBJECTIVE: To explore the pathogenetic mechanism of cystic meningioma and the key to its diagnosis and operation. METHODS: Clinical data of 8 patients with cystic meningioma were analyzed retrospectively. The occuring position and cure rate of cystic meningioma were compared with those of solid meningioma in the corresponding period. RESULTS: All patients were subjected to total resection of tumor parenchyma and cystic wall. No operative death and severe complications occurred. The incidence of uppertentorial region of cystic meningioma was higher than that of solid meningioma, and the difference was significant (χ(2)=2.618, P<0.05). The cure rate was not significantly different between cystic meningioma and solid meningioma (χ(2)=0.010, P>0.05). CONCLUSION: Removing tumor totally is the key to preventing its recurrence.

[Clinical characteristics and operative effect of hippocampus lesions].

OBJECTIVE: To study the clinical characteristics and operative effect of hippocampus lesions. METHODS: We retrospectively analyzed the clinical characteristics and operative outcome of 44 patients with hippocampus lesions between August 2005 and April 2010. RESULTS: Seizure attack was the initial symptom among 40 of the 44 patients. Pathological examinations revealed 18 gliomas, 9 cavernous malformations, 12 hippocampus sclerosis, 2 focal cortical dysplasia, 1 atypical hyperplasia, 1 injury glial scar, and 1 encephalomalacia. Thirteen patients received anterior medial temporal lobectomy and the other 31 received lesionectomy or selective amygdalohippocampectomy via transsylvian approach. An average of 15.7 month follow-up was accomplished in 37 patients. Postoperative epileptic outcomes were evaluated according to Engel classification: Grade I 73.0%(27/37), Grade II 13.5%(5/37), Grade III 10.8%(4/37) and Grade IV 2.7%(1/37). No perioperative death occurred. One patient experienced hemiplegia but recovered 8 months after the operation. Noticeable postoperative visual field deficit was left in 2 patients. Two patients with glioma died of remote tumor recurrence during follow-up. CONCLUSION: Seizure attack is a major complaint of hippocampus lesions. Satisfactory seizure and tumor control may be achieved through anterior medial temporal lobectomy or selective amygdalohippocampectomy with lesionectomy.

[Expression of peroxisome proliferator-activated receptor-gamma in human pituitary adenomas and its correlation to tumor invasiveness].

BACKGROUND & OBJECTIVE: Peroxisome proliferator-activated receptor-gamma (PPARgamma) is a nuclear transcription factor. Its ligant can inhibit proliferation and enhance differentiation of tumor cells, which plays crucial roles in metastasis and invasion of tumors. This study was to investigate the expression of PPARgamma in human pituitary adenomas and its clinical significance. METHODS: Immunohistochemistry (IHC) was used to investigate the expression of PPARgamma protein in 78 human pituitary adenomas, including 41 invasive and 37 non-invasive cases. The expression levels of PPARgamma mRNA in 33 human pituitary adenomas, including 16 invasive and 17 non-invasive cases, and three normal pituitary tissues obtained from autopsy were confirm by reverse transcription polymerase chain reaction (RT-PCR). The expression of PPARgamma in invasive and non-invasive pituitary adenomas was analyzed by Chi2 test and t test of the fourfold table. RESULTS: The positive rate of PPARgamma protein was significantly higher in invasive pituitary adenomas than in non-invasive ones (68.09% vs. 38.71%,P<0.05). The PPARgamma mRNA level was significantly higher in pituitary adenomas than in normal pituitary tissues (2.99+/-0.18 vs. 1.55+/-0.25, P<0.05), and higher in invasive pituitary adenomas than in non-invasive cases (3.95+/-0.43 vs. 2.40+/-0.24, P<0.01). CONCLUSIONS: PPARgamma is highly expressed in human pituitary adenomas, especially in the invasive ones. PPARgamma may be used as a new target for the treatment of pituitary adenomas.

[Electrocorticography monitoring in microsurgical treatment of solitary cavernous angiomas].

OBJECTIVE: To investigate the curative effect of electrocorticography (ECoG) monitoring in the microsurgical treatment of cavernous angiomas. METHODS: Clinical data of 71 patients with epileptogenic cavernous angiomas,who had been performed ECoG monitoring during the operation,were analyzed retrospectively. RESULTS: The foci of cavernous angiomas and epilepsy of the 71 patients were resected during the operation. In the 58 patients who were followed up,42 had not epileptic seizure,and 16 still had epileptic seizure,while the frequencies of 13 patients reduced to below 10%,and 3 patients over 10%. CONCLUSION: The drug treatment of epileptogenic cavernous angiomas can not control epileptic seizure,and the patients should receive the microsurgical treatment early. Electrocorticography monitoring can direct the surgical procedure,and control the postoperative epileptic seizure.

Expression of galectin-3 correlates with apoptosis in pituitary adenoma cells.

OBJECTIVE: To investigate the expressions of Galectin-3 (Gal-3), Bcl-2 and Bax in human pituitary adenomas, and to explore the interrelation among them. METHODS: RT-PCR and immunohistochemistry were applied to detect the mRNA and protein expressions of Gal-3, Bcl-2 and Bax in surgically excised human pituitary adenoma tissues, including invasive and non-invasive pituitary adenomas, and the correlation analysis was performed. RESULTS: The Gal-3 and Bcl-2 expressions in the invasive pituitary group were significantly higher than those in the non-invasive group, and the expression of Bax had no significant difference between the two groups. Pearsonos correlation analyses showed that the Gal-3 expression was positively correlated with Bcl-2, but was not correlated with Bax, which was inversely correlated with expression of Bcl-2. CONCLUSION: Gal-3 may function through a cell death inhibition pathway involving Bcl-2 to enhance cell proliferation, which result in the invasive growth of pituitary adenoma. These results indicate that Gal-3 has an important role in pituitary tumor cell proliferation and may serves as a possible therapeutic target in treatment of pituitary tumors.

[Effect of different tidal volume ventilations on atelectasis in patients during general anesthesia by CT scan].

OBJECTIVE: To investigate the effect of different tidal volume ventilations on the amount of atelectasis during general anesthesia. METHODS: Twenty adults, ASA physical status I and status II patients, who were scheduled for elective excision of intracranial lesion were randomly divided into 2 groups: Group TV (traditional tidal volume ventilation, 10 mL/kg) and Group LV (low tidal volume ventilation, 6 mL/kg). Atelectasis, as determined by CT and artery blood gas (ABG) analysis, was measured before the anesthesia, after the tracheal intubation, and at the end of the operation, respectively. Respiratory mechanical parameters were measured at 30, 120, and 240 min after the intubation. RESULTS: After the tracheal intubation, CT scan showed obvious atelectasis in both groups. The atelectasis area was(4.35+/-2.15)cm2 (3.12%+/-1.94%) in the TV group and (4.80+/-2.45)cm2 (3.89%+/-2.11%) in the LV group, with a nonsignificant difference between the 2 groups. At the end of the operation, there was no significant increase in the amount of atelectasis between and within the 2 groups. Artery blood gas analysis showed no difference after the tracheal intubation and at the end of the operation in either group. Ppeak, Pplat, Pmean and lung compliance(Cst)were significantly higher in the TV group than those in the LV group. CONCLUSION: Low tidal volume(6 mL/kg) ventilation is more feasible during general anesthesia in patients with healthy lungs, and it does not increase the atelectasis and impairment of gas exchange.

Expression and structure of interleukin 4 receptor (IL-4R) complex in human invasive pituitary adenomas.

Pituitary adenomas are frequently invasive of surrounding tissues, which adversely affects the surgical outcome and the disease-free survival of patients. In the present study, Interleukin 4 receptor (IL-4R) complex has been investigated to figure out whether the three subunits are overexpressed in human invasive pituitary adenomas. Reverse transcription-polymerase chain reaction (RT-PCR) analysis for interleukin 4 receptor alpha (IL-4Ralpha), interleukin 13 receptor alpha1 (IL-13Ralpha1), interleukin 2 receptor gammac (IL-2Rgammac) were performed on total RNA extracted from 10 non-invasive pituitary adenomas, 30 invasive pituitary adenomas, one glioblastoma multiforme, one normal human pituitary tissue sample and one normal human brain tissue sample. Quantitative real-time PCR and in situ immunofluorescence assay were performed in five invasive functioning pituitary adenoma samples and five invasive nonfunctioning pituitary adenoma samples. RT-PCR analysis for IL-4Ralpha, IL-13Ralpha1 and IL-2Rgammac chains were overexpressed in invasive pituitary adenomas. The transcripts for three subunits were not/weakly expressed in normal pituitary tissue and normal brain tissue. The quantitative real-time PCR and in situ immunofluorescence assay confirmed the results of the RT-PCR analysis. Our results indicate that human invasive pituitary adenomas express type III IL-4R complex. These receptors may serve as a novel target for immunotoxin therapy in patients with invasive pituitary adenomas who are not amenable to total surgical resection or for recurrent cases.

[Gamma knife treatment for pituitary prolactinomas].

OBJECTIVE: To evaluate the outcome of gamma knife for prolactinomas. METHODS: Eighty-nine patients were treated by gamma knife and 51 were followed up. The dose to the tumor margin ranged from 18 Gy to 35 Gy (mean 26.1 Gy). The maximum radiation dose varied from 36 Gy to 60 Gy (mean 50.41 Gy). The mean tumor diameter was 15.5 mm (5 - 26 mm). RESULTS: The follow-up data were available for 51 patients ranging from 6 to 108 months (mean 37 months). The tumor growth control rate was 100%. The endocrinological remission rate was 40%. The rate of hypopituitarism was 17.6%. CONCLUSION: Gamma knife radiosurgery can be used as a primary treatment for selected prolactinomas,especially for pituitary microadenomas.

[Detection and clinical significance of urinary epidermal growth factor in brain tumor patients].

OBJECTIVE: To investigate the clinical significance of urinary epidermal growth factor (EGF) in patients with brain tumors. METHODS: The levels of EGF in urine samples collected from 20 patients (9 low grade astrocytomas, 6 anaplastic astrocytomas, and 5 meningiomas) and 5 healthy individuals were determined. EGF levels were measured by radioimmunoassay technique. A preoperative and one postoperative determination were performed. RESULTS: Preoperative urinary EGF levels of astrocytoma patients were statistically higher than those of meningioma patients and the controls (P < 0.01). Preoperative urinary EGF levels showed a positive correlation with the degree of malignance in the astrocytoma patients (P < 0.05). A significant decrease of the postoperative levels of EGF was observed in the astrocytoma patients who underwent gross total resection (P < 0.01). The pre/postoperative urinary EGF levels of the meningioma patients showed no significant fluctuations and showed no significant difference with those of healthy individuals (P > 0.05). CONCLUSION: The urinary EGF levels of astrocytoma patients correlate with the WHO grade of malignance and significantly decrease after gross total removal. Urinary EGF may be of practical value in diagnosing and evaluating the surgical efficacy of astrocytomas.