Machine Learning to Predict Three Types of Outcomes After Traumatic Brain Injury Using Data at Admission: A Multi-Center Study for Development and Validation.

The difficulty of accurately identifying patients who would benefit from promising treatments makes it challenging to prove the efficacy of novel treatments for traumatic brain injury (TBI). Although machine learning is being increasingly applied to this task, existing binary outcome prediction models are insufficient for the effective stratification of TBI patients. The aim of this study was to develop an accurate 3-class outcome prediction model to enable appropriate patient stratification. To this end, retrospective balanced data of 1200 blunt TBI patients admitted to six Japanese hospitals from January 2018 onwards (200 consecutive cases at each institution) were used for model training and validation. We incorporated 21 predictors obtained in the emergency department, including age, sex, six clinical findings, four laboratory parameters, eight computed tomography findings, and an emergency craniotomy. We developed two machine learning models (XGBoost and dense neural network) and logistic regression models to predict 3-class outcomes based on the Glasgow Outcome Scale-Extended (GOSE) at discharge. The prediction models were developed using a training dataset with n = 1000, and their prediction performances were evaluated over two validation rounds on a validation dataset (n = 80) and a test dataset (n = 120) using the bootstrap method. Of the 1200 patients in aggregate, the median patient age was 71 years, 199 (16.7%) exhibited severe TBI, and emergency craniotomy was performed on 104 patients (8.7%). The median length of stay was 13.0 days. The 3-class outcomes were good recovery/moderate disability for 709 patients (59.1%), severe disability/vegetative state in 416 patients (34.7%), and death in 75 patients (6.2%). XGBoost model performed well with 69.5% sensitivity, 82.5% accuracy, and an area under the receiver operating characteristic curve of 0.901 in the final validation. In terms of the receiver operating characteristic curve analysis, the XGBoost outperformed the neural network-based and logistic regression models slightly. In particular, XGBoost outperformed the logistic regression model significantly in predicting severe disability/vegetative state. Although each model predicted favorable outcomes accurately, they tended to miss the mortality prediction. The proposed machine learning model was demonstrated to be capable of accurate prediction of in-hospital outcomes following TBI, even with the three GOSE-based categories. As a result, it is expected to be more impactful in the development of appropriate patient stratification methods in future TBI studies than conventional binary prognostic models. Further, outcomes were predicted based on only clinical data obtained from the emergency department. However, developing a robust model with consistent performance in diverse scenarios remains challenging, and further efforts are needed to improve generalization performance.

Particle therapy using protons or carbon ions for cancer patients with cardiac implantable electronic devices (CIED): a retrospective multi-institutional study.

PURPOSE: To evaluate the outcomes of particle therapy in cancer patients with cardiac implantable electronic devices (CIEDs). MATERIALS AND METHODS: From April 2001 to March 2013, 19,585 patients were treated with proton beam therapy (PBT) or carbon ion therapy (CIT) at 8 institutions. Of these, 69 patients (0.4%, PBT 46, CIT 22, and PBT + CIT 1) with CIEDs (64 pacemakers, 4 implantable cardioverter defibrillators, and 1 with a cardiac resynchronization therapy defibrillator) were retrospectively reviewed. All the patients with CIEDs in this study were treated with the passive scattering type of particle beam therapy. RESULTS: Six (13%) of the 47 PBT patients, and none of the 23 CIT patients experienced CIED malfunctions (p = 0.105). Electrical resets (7) and over-sensing (3) occurred transiently in 6 patients. The distance between the edge of the irradiation field and the CIED was not associated with the incidence of malfunctions in 20 patients with lung cancer. A larger field size had a higher event rate but the test to evaluate trends as not statistically significant (p = 0.196). CONCLUSION: Differences in the frequency of occurrence of device malfunctions for patients treated with PBT and patients treated with CIT did not reach statistical significance. The present study can be regarded as a benchmark study about the incidence of malfunctioning of CIED in passive scattering particle beam therapy and can be used as a reference for active scanning particle beam therapy.

Liver phantom design and dosimetric verification in participating institutions for a proton beam therapy in patients with resectable hepatocellular carcinoma: Japan Clinical Oncology Group trial (JCOG1315C).

BACKGROUND AND PURPOSE: In Japan, the first domestic clinical trial of proton beam therapy for the liver was initiated as the Japan Clinical Oncology Group trial (JCOG1315C: Non-randomized controlled study comparing proton beam therapy and hepatectomy for resectable hepatocellular carcinoma). Purposes of this study were to develop a new dosimetric verification system and to carry out a credentialing for the JCOG1315C clinical trial. MATERIALS AND METHODS: Accuracy and differences in doses in proton treatment planning among participating institutions were surveyed and investigated. We designed and developed a suitable water tank-type liver phantom for a dosimetric verification of proton beam therapy for liver. In a visiting survey of five institutions participating in the clinical trial, we performed the dosimetric verification using the liver phantom and an air-filled ionization chamber. RESULTS: The shape of the dose distributions calculated in proton treatment planning was characteristic and dependent on the manufacturers of the proton beam therapy system, the proton treatment planning system and the setup at the participating institutions. Widths of the lateral penumbra were 5.8-12.7 mm among participating institutions. The accuracy between the calculated and the measured doses in the proton irradiation was within 3% at five measurement points including both points on the isocenter and off the isocenter. CONCLUSIONS: These findings confirmed the accuracy of the delivery doses in the institutions participating in the clinical trial, and the clinical trial with integration of all institutions (five institutions) could be initiated.

Simplified estimation method for dose distributions around field junctions in proton craniospinal irradiation.

In radiotherapy involving craniospinal irradiation (CSI), field junctions of therapeutic beams are necessary, because a CSI target is generally several times larger than the maximum field size of the beams. The purpose of this study was to develop a simplified method for estimating dose uniformity around the field junctions in proton CSI. We estimated the dose profiles around the field junctions of proton beams using a simplified field-junction model, in which partial lateral dose distributions around the field edge were assumed to be approximated using the error function. We measured the lateral dose distributions of the proton beams planned for the CSI treatment using a two-dimensional (2D) ionization chamber array. Although dose hot spots and cold spots tend to be underestimated by a chamber array because of the partial volume effect of the sensitive volume and discrete chamber positions, the model estimation results were fairly consistent with the measurements obtained using a 2D chamber array subjected to CSI-simulated serial irradiation. The simplified junction model enabled us to estimate the dose distributions and dependence of the setup position gap on the dose uniformity around the field junctions on the basis of the field-by-field dose profiles measured using the 2D chamber array.

Lifetime attributable risk of radiation-induced secondary cancer from proton beam therapy compared with that of intensity-modulated X-ray therapy in randomly sampled pediatric cancer patients.

To investigate the amount that radiation-induced secondary cancer would be reduced by using proton beam therapy (PBT) in place of intensity-modulated X-ray therapy (IMXT) in pediatric patients, we analyzed lifetime attributable risk (LAR) as an in silico surrogate marker of the secondary cancer after these treatments. From 242 pediatric patients with cancers who were treated with PBT, 26 patients were selected by random sampling after stratification into four categories: (i) brain, head and neck, (ii) thoracic, (iii) abdominal, and (iv) whole craniospinal (WCNS) irradiation. IMXT was replanned using the same computed tomography and region of interest. Using the dose-volume histograms (DVHs) of PBT and IMXT, the LARs of Schneider et al. were calculated for the same patient. All the published dose-response models were tested for the organs at risk. Calculation of the LARs of PBT and IMXT based on the DVHs was feasible for all patients. The means ± standard deviations of the cumulative LAR difference between PBT and IMXT for the four categories were (i) 1.02 ± 0.52% (n = 7, P = 0.0021), (ii) 23.3 ± 17.2% (n = 8, P = 0.0065), (iii) 16.6 ± 19.9% (n = 8, P = 0.0497) and (iv) 50.0 ± 21.1% (n = 3, P = 0.0274), respectively (one tailed t-test). The numbers needed to treat (NNT) were (i) 98.0, (ii) 4.3, (iii) 6.0 and (iv) 2.0 for WCNS, respectively. In pediatric patients who had undergone PBT, the LAR of PBT was significantly lower than the LAR of IMXT estimated by in silico modeling. Although a validation study is required, it is suggested that the LAR would be useful as an in silico surrogate marker of secondary cancer induced by different radiotherapy techniques.

Rapid tumor growth with glial differentiation of central neurocytoma after stereotactic radiosurgery.

Although stereotactic radiosurgery (SRS) is effective for central neurocytoma (CN), the long-term outcome of SRS remains unclear. We present a case of recurrent CN that was diagnosed 10years after surgical resection and consecutive stereotactic radiotherapy. The patient was treated with SRS for the recurrent tumor, but underwent two-staged surgery once again due to rapid tumor growth. Histological features of the recurrent tumor were consistent with the diagnosis of CN. However, an increased Ki-67 proliferation index (3.4%), aberrant angiogenesis and glial differentiation of the tumor cells were observed, which were not identified in the initial CN. In addition, vascular endothelial growth factor (VEGF) and VEGF receptor were highly expressed in the recurrent tumor cells, as well as in the vascular endothelial cells. Our case suggests that malignant transition with aberrant angiogenesis and glial differentiation may be attributable to SRS.

Dose escalation study of proton beam therapy with concurrent chemotherapy for stage III non-small cell lung cancer.

The purpose of this study is to determine the recommended dose (RD) of proton beam therapy (PBT) for inoperable stage III non-small cell lung cancer (NSCLC). We tested two prescribed doses of PBT: 66 Gy (relative biological effectiveness [RBE]) in 33 fractions and 74 Gy (RBE) in 37 fractions in arms 1 and 2, respectively. The planning target volume (PTV) included the primary tumor and metastatic lymph nodes with adequate margins. Concurrent chemotherapy included intravenous cisplatin (60 mg/m(2) , day 1) and oral S-1 (80, 100 or 120 mg based on body surface area, days 1-14), repeated as four cycles every 4 weeks. Dose-limiting toxicity (DLT) was defined as grade 3 or severe toxicities related to PBT during days 1-90. Each dose level was performed in three patients, and then escalated to the next level if no DLT occurred. When one patient developed a DLT, three additional patients were enrolled. Overall, nine patients (five men, four women; median age, 72 years) were enrolled, including six in arm 1 and three in arm 2. The median follow-up time was 43 months, and the median progression-free survival was 15 months. In arm 1, grade 3 infection occurred in one of six patients, but no other DLT was reported. Similarly, no DLT occurred in arm 2. However, one patient in arm 2 developed grade 3 esophageal fistula at 9 months after the initiation of PBT. Therefore, we determined that 66 Gy (RBE) is the RD from a clinical viewpoints. (Clinical trial registration no. UMIN000005585).

Semi-analytical model for output factor calculations in proton beam therapy with consideration for the collimator aperture edge.

In the development of an external radiotherapy treatment planning system, the output factor (OPF) is an important value for the monitor unit calculations. We developed a proton OPF calculation model with consideration for the collimator aperture edge to account for the dependence of the OPF on the collimator aperture and distance in proton beam therapy. Five parameters in the model were obtained by fitting with OPFs measured by a pinpoint chamber with the circular radiation fields of various field radii and collimator distances. The OPF model calculation using the fitted model parameters could explain the measurement results to within 1.6% error in typical proton treatment beams with 6- and 12 cm SOBP widths through a range shifter and a circular aperture more than 10.6 mm in radius. The calibration depth dependences of the model parameters were approximated by linear or quadratic functions. The semi-analytical OPF model calculation was tested with various MLC aperture shapes that included circles of various sizes as well as a rectangle, parallelogram, and L-shape for an intermediate proton treatment beam condition. The pre-calculated OPFs agreed well with the measured values, to within 2.7% error up to 620 mm in the collimator distance, though the maximum difference was 5.1% in the case of the largest collimator distance of 740 mm. The OPF calculation model would allow more accurate monitor unit calculations for therapeutic proton beams within the expected range of collimator conditions in clinical use.

A model-based analysis of a simplified beam-specific dose output in proton therapy with a single-ring wobbling system.

In radiation therapy, it is necessary to preset a monitor unit in an irradiation control system to deliver a prescribed absolute dose to a reference point in the planning target volume. The purpose of this study was to develop a model-based monitor unit calculation method for proton-beam therapy with a single-ring wobbling system. The absorbed dose at a calibration point per monitor unit had been measured for each beam-specific measurement condition without a patient-specific collimator or range compensator before proton therapeutic irradiation at Shizuoka Cancer Center. In this paper, we propose a simplified dose output model to obtain the output ratio between a beam-specific dose and a reference field dose, from which a monitor unit for the proton treatment could be derived without beam-specific measurements. The model parameters were determined to fit some typical data measured in a proton treatment room, called a Gantry 1 course. Then, the model calculation was compared with 5456 dose output ratios that had been measured for 150-, 190- and 220 MeV therapeutic proton beams in two treatment rooms over the past decade. The mean value and standard deviation of the difference between the measurement and the model calculation were respectively 0.00% and 0.27% for the Gantry 1 course, and -0.25% and 0.35% for the Gantry 2 course. The model calculation was in good agreement with the measured beam-specific doses, within 1%, except for conditions less frequently used for treatment. The small variation for the various beam conditions shows the high long-term reproducibility of the measurement and high degree of compatibility of the two treatment rooms. Therefore, the model was expected to assure the setting value of the dose monitor for treatment, to save the effort required for beam-specific measurement, and to predict the dose output for new beam conditions in the future.

Assessment of organ dose reduction and secondary cancer risk associated with the use of proton beam therapy and intensity modulated radiation therapy in treatment of neuroblastomas.

BACKGROUND: To compare proton beam therapy (PBT) and intensity-modulated radiation therapy (IMRT) with conformal radiation therapy (CRT) in terms of their organ doses and ability to cause secondary cancer in normal organs. METHODS: Five patients (median age, 4 years; range, 2-11 years) who underwent PBT for retroperitoneal neuroblastoma were selected for treatment planning simulation. Four patients had stage 4 tumors and one had stage 2A tumor, according to the International Neuroblastoma Staging System. Two patients received 36 Gy, two received 21.6 Gy, and one received 41.4 Gy of radiation. The volume structures of these patients were used for simulations of CRT and IMRT treatment. Dose-volume analyses of liver, stomach, colon, small intestine, pancreas, and bone were performed for the simulations. Secondary cancer risks in these organs were calculated using the organ equivalent dose (OED) model, which took into account the rates of cell killing, repopulation, and the neutron dose from the treatment machine. RESULTS: In all evaluated organs, the mean dose in PBT was 20-80% of that in CRT. IMRT also showed lower mean doses than CRT for two organs (20% and 65%), but higher mean doses for the other four organs (110-120%). The risk of secondary cancer in PBT was 24-83% of that in CRT for five organs, but 121% of that in CRT for pancreas. The risk of secondary cancer in IMRT was equal to or higher than CRT for four organs (range 100-124%). CONCLUSION: Low radiation doses in normal organs are more frequently observed in PBT than in IMRT. Assessments of secondary cancer risk showed that PBT reduces the risk of secondary cancer in most organs, whereas IMRT is associated with a higher risk than CRT.

A revision of proton machine quality assurance for wobbled-proton-beam therapy.

Periodic checks for proton machine quality assurance (QA) are significant for machine users safely and accurately to provide proton-beam treatment for cancer. Our aim in this study was to describe a revision to proton machine QA procedures for wobbled-proton-beam therapy at the Shizuoka Cancer Center (SCC) in Japan. The previous daily, monthly, and annual QA procedures were determined by reference to our past operational experience and to QA papers for medical accelerators. The revised QA procedures were initiated in May 2011 after preliminary measurements to decide baselines for the QA procedures. This paper presents the proton machine QA procedures and the results of representative QA measurements. Three action levels were decided on by reference to the American Association of Physicists in Medicine Task Group 142 report. Tolerances of inspection action were decided on based on the provisional operational results and actual fluctuations of the QA measurement for a year, and those of scheduled action and stop-treatment action were determined by reference to the machine QA papers and those of the inspection action. No deviation from the tolerance of the scheduled action has been observed so far. Although a few QA procedures exceeded the tolerance of the inspection action, these excesses were resolved by inspection and improvement of the respective measuring procedure within the designated QA time. Hereafter, the proton machine QA procedures proposed in this study will be performed continuously at the SCC to assure patient safety and accurate operation of proton therapy.

Microdosimetric calculation of relative biological effectiveness for design of therapeutic proton beams.

The authors attempt to establish the relative biological effectiveness (RBE) calculation for designing therapeutic proton beams on the basis of microdosimetry. The tissue-equivalent proportional counter (TEPC) was used to measure microdosimetric lineal energy spectra for proton beams at various depths in a water phantom. An RBE-weighted absorbed dose is defined as an absorbed dose multiplied by an RBE for cell death of human salivary gland (HSG) tumor cells in this study. The RBE values were calculated by a modified microdosimetric kinetic model using the biological parameters for HSG tumor cells. The calculated RBE distributions showed a gradual increase to about 1cm short of a beam range and a steep increase around the beam range for both the mono-energetic and spread-out Bragg peak (SOBP) proton beams. The calculated RBE values were partially compared with a biological experiment in which the HSG tumor cells were irradiated by the SOBP beam except around the distal end. The RBE-weighted absorbed dose distribution for the SOBP beam was derived from the measured spectra for the mono-energetic beam by a mixing calculation, and it was confirmed that it agreed well with that directly derived from the microdosimetric spectra measured in the SOBP beam. The absorbed dose distributions to planarize the RBE-weighted absorbed dose were calculated in consideration of the RBE dependence on the prescribed absorbed dose and cellular radio-sensitivity. The results show that the microdosimetric measurement for the mono-energetic proton beam is also useful for designing RBE-weighted absorbed dose distributions for range-modulated proton beams.

A treatment planning comparison of passive-scattering and intensity-modulated proton therapy for typical tumor sites.

Intensity-modulated proton therapy (IMPT) is expected to improve treatment results with fewer side effects than other proton therapies. The purpose of this study was to evaluate the tumor sites for which IMPT was effective under the same beam calculation conditions by planning IMPT for typical cases treated with passive scattering proton therapy (PSPT). We selected 16 cases of nasal cavity, lung, liver or prostate cancers as typical tumor sites receiving PSPT. The dose distributions and dose volume histograms optimized by the IMPT were compared with those optimized by the PSPT. We took particular note of the doses to the skin and organs at risk (OAR) when PSPT was replaced by IMPT. Furthermore, an improvement of the beam angles was also performed to obtain better dose distributions in the IMPT. The IMPT with the same beam angles resulted in near-maximum doses to the skin of average 78%, 64%, 84% and 99% of the PSPT doses for nasal cavity, lung, liver, and prostate cancers, respectively. However, it was difficult to improve the dose homogeneity of the target volume. The change of the IMPT beam angles could reduce the doses to OARs and skin in the case of the nasal cavity, while it had limited effect in the other cases. We concluded that IMPT was effective for reducing the doses to some OARs when treating nasal cavity, lung, liver and prostate cancers. The selection of beam angles was important in the IMPT optimization, especially for nasal cavity cancers.

Utility of diffusion tensor imaging in the acute stage of mild to moderate traumatic brain injury for detecting white matter lesions and predicting long-term cognitive function in adults.

OBJECT: Traumatic brain injury (TBI) often impairs cognitive function. Diffusion tensor (DT) imaging, a novel modality, permits evaluation of the effects of head trauma on white matter nerve fibers. The objectives of the current study were to investigate where the white matter injury following mild to moderate TBI is specifically located on DT imaging in the acute disease stage and to examine the relationship between the severity of the white matter lesion on DT imaging in the acute stage of TBI and future cognitive function in the chronic disease stage. METHODS: Twenty adult patients with mild to moderate TBI (Glasgow Coma Scale score between 9 and 15) underwent conventional MR and DT imaging a median of 3.5 days after injury, and 27 matched healthy controls also underwent both imaging modalities. The patients with TBI were further subdivided into 2 groups, that is, mild and more severe TBI groups, based on clinical (mild or moderate TBI), CT (diffuse brain injury [DBI] I or II), or MR imaging (normal or pathological appearance) classification. Fractional anisotropies (FAs) were compared between patients and controls using the region of interest method. Regions of interest were located in 8 different areas including the genu, stem, and splenium of the corpus callosum and the corona radiata (CR), anterior limb of the internal capsule (ALIC), posterior limb of the internal capsule (PLIC), frontal white matter (FWM), and occipital white matter (OWM) of the periventricular white matter. Eleven patients with TBI also underwent neuropsychological testing, which included the Trail Making Test, Wisconsin Card Sorting Test, Wechsler Adult Intelligence Scale-Revised, and P300 testing in the chronic disease stage (median 364 days). RESULTS: Region of interest analysis demonstrated significantly lower FA values in the genu, stem, and splenium of the corpus callosum in more severe TBI groups (moderate TBI on clinical classification, DBI II on CT classification, and pathological appearance on MR imaging classification) than in controls. A significant difference was also observed in the FA of the splenium between controls and the mild TBI group of the clinical classification. No significant difference was observed in the FA of the CR, ALIC, PLIC, FWM, and OWM between controls and any of the TBI groups of clinical or imaging classifications. No significant difference was observed in the FA of any regions between mild and more severe TBI groups of the clinical or imaging classifications. Multiple regression analysis showed a statistically significant positive linear relationship between FA in the splenium and total IQ (r = 0.79, p = 0.004). A significant negative linear relationship between FA in the FWM and P300 latency was also observed (r = 0.62, p = 0.04). CONCLUSIONS: Fractional anisotropy reductions in the splenium and FWM in the acute stage of mild to moderate TBI may be a useful prognostic factor for long-term cognitive dysfunction.

Brain imaging modality before systemic thrombolysis for ischemic stroke within three hours.

OBJECTIVE: In Japan, MRI-based thrombolysis after CT screening is the most common imaging strategy prior to intravenous thrombolysis (IVT) with tissue plasminogen activator (tPA) within 3 h after ischemic stroke. A choice of MRI with MR angiography (MRA) provides a higher diagnostic accuracy, but may delay an initiation of thrombolysis. METHODS: In our neuro-unit, brain CT is the first screening image for suspected stroke. We retrospectively examined a delay to thrombolysis, imaging modality, diagnostic accuracy, and clinical outcomes at 3 months by the modified Rankin Scale in patients receiving IVT within 3 h. RESULTS: Among 67 patients receiving IVT with tPA, brain imaging prior to IVT was solely CT in 10 (15%) patients and CT + MRI/MRA in 57 (85%) patients. Final diagnosis of brain ischemia was 100%. Patients receiving CT + MRI had significantly shorter pre-hospital delay (mean 54 vs. 83 min; p = 0.012), but longer door-to-needle time (mean 90 vs. 57 min; p = 0.019) than those receiving CT only. Finally, time from onset to thrombolysis was not different between the two groups and clinical outcomes were also comparable. The earlier patients arrived, the longer door-to-needle times were (p < 0.001). CONCLUSIONS: The imaging strategy of initial CT screening with optional MRI/MRA scans prior to IVT was feasible. However, it resulted in an additional 30 min in-hospital delay of tPA administration, which may affect clinical outcomes.

Residual motion and duty time in respiratory gating radiotherapy using individualized or population-based windows.

PURPOSE: The efficiency and precision of respiratory gated radiation therapy for tumors is affected by variations in respiration-induced tumor motion. We evaluated the use of individualized and population-based parameters for such treatment. METHODS AND MATERIALS: External respiratory signal records and images of respiration-induced tumor motion were obtained from 42 patients undergoing respiratory gated radiation therapy for liver tumors. Gating window widths were calculated for each patient, with 2, 4, and 10 mm of residual motion, and the mean was defined as the population-based window width. Residual motions based on population-based and predefined window widths were compared. Duty times based on whole treatment sessions, at various window levels, were calculated. The window level giving the longest duty time was defined as the individualized most efficient level (MEL). MELs were also calculated based on the first 10 breathing cycles. The duty times for population-based MELs (defined as mean MELs) and individualized MELs were compared. RESULTS: Tracks of respiration-induced tumor motion ranged from 3 to 50 mm. Half of the patients had larger actual residual motions than the assigned residual motions. Duty times were greater when based on individualized, rather than population-based, window widths. The MELs established during whole treatment sessions for 2 mm and 4 mm of residual motion gave significantly increased duty times, whereas those calculated using the first 10 breathing cycles showed only marginal increases. CONCLUSIONS: Using individualized window widths and levels provided more precise and efficient respiratory gated radiation therapy. However, methods for predicting individualized window levels before treatment remain to be explored.

Emergent surgical repair for penetrating injury of the cervical carotid artery associated with shock.

A 41-year-old male presented with penetrating carotid artery injury after being stabbed in the left neck with a knife. He had profuse bleeding from the wound and was in hemorrhagic shock. No neuroimaging evaluation was possible, so emergent surgery was begun with continuous manual compression of the left neck. Graft reconstruction of the injured carotid artery was performed. He recovered well without neurological deficits. Urgent hemostasis is required for patients with penetrating carotid artery injury presenting with shock. Surgical reconstruction of the carotid artery with a vascular graft is a safe option.

Changes in rectal volume and prostate localization due to placement of a rectum-emptying tube.

PURPOSE: Rectal volume variation has a crucial effect on prostate localization during external beam radiotherapy for prostate cancer. This study investigated the effect of rectal volume reduction by a rectum-emptying tube (RET) on prostate immobilization. MATERIALS AND METHODS: The study group comprised 21 patients who underwent proton beam treatment for prostate cancer. Sigmoid-shaped flexible plastic RETs were used to drain gases from the rectum. Computed tomography (CT) was performed before and after RET placement at the treatment planning stage and at the beginning of treatment. Prostate displacement and changes in rectal volume were measured on the CT images. The feasibility of RET placement was evaluated during and after the procedure. RESULTS: All 21 patients tolerated the procedure. The rectal volume was significantly lower with a RET than without a RET. The differences in rectal volume between the treatment planning stage and the beginning of treatment were significantly lower with a RET than without. RET placement significantly decreased prostate displacement in the anteroposterior and superoinferior directions but not in the left-right direction. CONCLUSION: RET placement reduced both rectal volume and variation in rectal volume. The procedure reduced displacement of the prostate. RET placement thus appears to be an effective technique for immobilizing the prostate.

Unruptured Cerebral Aneurysm Detected after Intravenous Tissue Plasminogen Activator for Stroke.

Therapeutic guidelines of intravenous thrombolysis with tissue plasminogen activator (tPA) for hyperacute ischemic stroke are very strict. Because of potential higher risk of bleeding complications, the presence of unruptured cerebral aneurysm is a contraindication for systemic thrombolysis with tPA. According to the standard CT criteria, a 66-year-old woman who suddenly developed aphasia and hemiparesis received intravenous tPA within 3 h after ischemic stroke. Magnetic resonance angiography during tPA infusion was performed and the presence of a small unruptured cerebral aneurysm was suspected at the anterior communicating artery. Delayed cerebral angiography confirmed an aneurysm with a size of 7 mm. The patient did not experience any adverse complications associated with the aneurysm. Clinical experiences of this kind of accidental off-label thrombolysis may contribute to modify the current rigid tPA guidelines for stroke.

Research on radiation protection in the application of new technologies for proton and heavy ion radiotherapy.

Particle radiotherapy using proton and heavy ion beams has shown improved clinical results and is a promising cancer therapy which is expected to gradually spread in Japan. There are, however, no special regulations for radiotherapy treatment facilities. They have been operated under the same safety regulations as for a research facility using a research accelerator. Significantly high-energy radiation is necessary for particle radiotherapy compared with conventional radiation therapy. The treatment facility, therefore, should have a large accelerator, which is installed in a room with a thick shield wall. Data on radiation protection for such high energy medical facilities is fragmentary and insufficient. In this study, we examined the necessity of other regulations for the safe operation of medical facilities for particle radiotherapy. First, we measured activation levels of the therapeutic devices and of patients. Next the safety level of the medical facility was evaluated from the viewpoint of radiation protection. We have confirmed the facilities can be safely operated by present regulations given in the Law Concerning Prevention from Radiation Hazards due to Radiation Isotopes, etc. or the Law for Health Protection and Medical Care.