3D Breast Cancer Segmentation in DCE-MRI Using Deep Learning With Weak Annotation

BACKGROUND

Deep learning models require large-scale training to perform confidently, but obtaining annotated datasets in medical imaging is challenging. Weak annotation has emerged as a way to save time and effort.

PURPOSE

To develop a deep learning model for 3D breast cancer segmentation in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) using weak annotation with reliable performance.

STUDY TYPE

Retrospective.

POPULATION

Seven hundred and thirty-six women with breast cancer from a single institution, divided into the development (N = 544) and test dataset (N = 192).

FIELD STRENGTH/SEQUENCE

3.0-T, 3D fat-saturated gradient-echo axial T1-weighted flash 3D volumetric interpolated brain examination (VIBE) sequences.

ASSESSMENT

Two radiologists performed a weak annotation of the ground truth using bounding boxes. Based on this, the ground truth annotation was completed through autonomic and manual correction. The deep learning model using 3D U-Net transformer (UNETR) was trained with this annotated dataset. The segmentation results of the test set were analyzed by quantitative and qualitative methods, and the regions were divided into whole breast and region of interest (ROI) within the bounding box.

STATISTICAL TESTS

As a quantitative method, we used the Dice similarity coefficient to evaluate the segmentation result. The volume correlation with the ground truth was evaluated with the Spearman correlation coefficient. Qualitatively, three readers independently evaluated the visual score in four scales. A P-value <0.05 was considered statistically significant.

RESULTS

The deep learning model we developed achieved a median Dice similarity score of 0.75 and 0.89 for the whole breast and ROI, respectively. The volume correlation coefficient with respect to the ground truth volume was 0.82 and 0.86 for the whole breast and ROI, respectively. The mean visual score, as evaluated by three readers, was 3.4.

DATA CONCLUSION

The proposed deep learning model with weak annotation may show good performance for 3D segmentations of breast cancer using DCE-MRI.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: Stage 2.

[MRI-Guided Breast Intervention: Biopsy and Needle Localization]

In Korea, the number of institutions providing breast MRI, as well as the number of breast MRIs, has recently increased. However, MRI-guided procedures, including biopsy and needle localization, are rarely performed compared to ultrasound-guided or stereotactic biopsy. As breast MRI has high sensitivity but limited specificity, lesions detected only on MRI require pathologic confirmation through MRI-guided biopsy or surgical excision with MRI-guided needle localization. Thus, we aimed to review MRI-guided procedures, including their indications, techniques, procedural considerations, and limitations.

Diagnostic Usefulness of Diffusion-Weighted MRI for Axillary Lymph Node Evaluation in Patients with Breast Cancer

This study aimed to determine whether apparent diffusion coefficient (ADC) and morphological features on diffusion-weighted MRI (DW-MRI) can discriminate metastatic axillary lymph nodes (ALNs) from benign in patients with breast cancer. Two radiologists measured ADC, long and short diameters, long-to-short diameter ratio, and cortical thickness and assessed eccentric cortical thickening, loss of fatty hilum, irregular margin, asymmetry in shape or number, and rim sign of ALNs on DW-MRI and categorized them into benign or suspicious ALNs. Pathologic reports were used as a reference standard. Statistical analysis was performed using the Mann-Whitney U test and chi-square test. Overall sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and diagnostic accuracy of DW-MRI were calculated. The ADC of metastatic ALNs was 0.905 × 10^-3 mm²/s, and that of benign ALNs was 0.991 × 10^-3 mm²/s (p = 0.243). All morphologic features showed significant difference between the two groups. The sensitivity, specificity, PPV, NPV, and diagnostic accuracy of the final categorization on DW-MRI were 77.1%, 93.3%, 79.4%, 92.5%, and 86.2%, respectively. Our results suggest that morphologic evaluation of ALNs on DWI can discriminate metastatic ALNs from benign. The ADC value of metastatic ALNs was lower than that of benign nodes, but the difference was not statistically significant.

The Frequency and Causes of Not-Detected Breast Malignancy in Dynamic Contrast-Enhanced MRI

Breast MR is the most sensitive imaging modality, but there are cases of malignant tumors that are not detected in MR. This study evaluated the frequency and main causes of malignant breast lesions not detected in dynamic contrast-enhanced (DCE) MR. A total of 1707 cases of preoperative breast MR performed between 2020 and 2021 were included. Three radiologists individually reviewed the DCE MRs and found not-detected malignancy cases in the MRs. The final cases were decided through consensus. For the selected cases, images other than DCE MRIs, such as mammography, ultrasounds, diffusion-weighted MRs, and, if possible, contrast-enhanced chest CTs, were analyzed. In the final sample, 12 cases were not detected in DCE MR, and the frequency was 0.7% (12/1707). Six cases were not detected due to known non-enhancing histologic features. In four cases, tumors were located in the breast periphery and showed no enhancement in MR. In the remaining two cases, malignant lesions were not identified due to underlying marked levels of BPE. The frequency of not-detected malignancy in DCE MR is rare. Knowing the causes of each case and correlating it with other imaging modalities could be helpful in the diagnosis of breast malignancy in DCE MR.

Radiomics Nomogram: Prediction of 2-Year Disease-Free Survival in Young Age Breast Cancer

This study aimed to predict early breast cancer recurrence in women under 40 years of age using radiomics signature and clinicopathological information. We retrospectively investigated 155 patients under 40 years of age with invasive breast cancer who underwent MRI and surgery. Through stratified random sampling, 111 patients were assigned as the training set, and 44 were assigned as the validation set. Recurrence-associated factors were investigated based on recurrence within 5 years during the total follow-up period. A Rad-score was generated through texture analysis (3D slicer, ver. 4.8.0) of breast MRI using the least absolute shrinkage and selection operator Cox regression model. The Rad-score showed a significant association with disease-free survival (DFS) in the training set (p = 0.003) and validation set (p = 0.020) in the Kaplan–Meier analysis. The nomogram was generated through Cox proportional hazards models, and its predictive ability was validated. The nomogram included the Rad-score and estrogen receptor negativity as predictive factors and showed fair DFS predictive ability in both the training and validation sets (C-index 0.63, 95% CI 0.45–0.79). In conclusion, the Rad-score can predict the disease recurrence of invasive breast cancer in women under 40 years of age, and the Rad-score-based nomogram showed reasonably high DFS predictive ability, especially within 2 years of surgery.

The Usefulness of Magnetic Resonance Imaging (MRI) for the Detection of Local Recurrence after Mastectomy with Reconstructive Surgery in Breast Cancer Patients

The purpose of this study is to investigate the usefulness of magnetic resonance imaging (MRI) for the detection of local recurrence after nipple-sparing mastectomy (NSM) or skin-sparing mastectomy (SSM) with immediate reconstructive surgery for breast cancer. Two hundred and eighty-six NSM or SSM procedures and immediate reconstruction cases between August 2015 and February 2020 were reviewed. The detectability rates of for local recurrence using MRI and ultrasound were assessed, and the characteristics of recurrent and primary cancers were evaluated. The patients with multifocal or multicentric primary cancer and a dense parenchymal pattern showed a higher recurrence rate (p < 0.001). A total of 22 cases showed recurrence, and due to multifocal recurrence, a total of 27 recurrent lesions were identified in the reconstructed breast, of which 12 were symptomatic and 15 were asymptomatic (p < 0.001). With the exception of skin recurrence (n = 6), MRI showed a significantly higher detectability rate (95.2%, 20 of 21) than ultrasound (38.1%, 8 of 21) for the recurrence of cancer in the reconstructed breast (p < 0.001), especially for small-sized (<1 cm) asymptomatic lesions. In addition, the mean recurrence interval of MRI-detected asymptomatic lesions was 21.7 months (SD ± 17.7), which was significantly longer than that of symptomatic recurrence. In conclusion, postoperative MRI can be useful for identifying small-sized (<1 cm) asymptomatic recurrence lesions in reconstructed breast tissue after NSM or SSM, which can be implemented within two years of surgery.

Ipsilateral Recurrence of DCIS in Relation to Radiomics Features on Contrast Enhanced Breast MRI

The purpose of this retrospective study was to investigate the association between ipsilateral recurrence of ductal carcinoma in situ (DCIS) and radiomics features from DCIS and contralateral normal breast on contrast enhanced breast MR imaging. A total of 163 patients with DCIS who underwent preoperative MR imaging between January 2010 and December 2014 were included (training cohort; n = 117, validation cohort; n = 46). Radiomics features were extracted from whole tumor volume of DCIS on early dynamic T1-subtraction images and from the contralateral normal breast on precontrast T1 and early dynamic T1-subtraction images. After feature selection, a Rad-score was established by LASSO Cox regression model. Performance of Rad-score was evaluated by the receiver operating characteristic (ROC) curve and Kaplan Meier curve with log rank test. The Rad-score was significantly associated with ipsilateral recurrence free survival (RFS). The low-risk group with a low Rad-score showed higher ipsilateral RFS than the high-risk group with a high Rad-score in both training and validation cohorts (p < 0.01). The Rad-score based on radiomics features from DCIS and contralateral normal breast on breast MR imaging showed the potential for prediction of ipsilateral RFS of DCIS.

Evaluation of Computer-Aided Detection (CAD) in Screening Automated Breast Ultrasound Based on Characteristics of CAD Marks and False-Positive Marks

The present study evaluated the effectiveness of computer-aided detection (CAD) system in screening automated breast ultrasound (ABUS) and analyzed the characteristics of CAD marks and the causes of false-positive marks. A total of 846 women who underwent ABUS for screening from January 2017 to December 2017 were included. Commercial CAD was used in all ABUS examinations, and its diagnostic performance and efficacy in shortening the reading time (RT) were evaluated. In addition, we analyzed the characteristics of CAD marks and the causes of false-positive marks. A total of 1032 CAD marks were displayed based on the patient and 534 CAD marks on the lesion. Five cases of breast cancer were diagnosed. The sensitivity, specificity, PPV, and NPV of CAD were 60.0%, 59.0%, 0.9%, and 99.6% for 846 patients. In the case of a negative study, it was less time-consuming and easier to make a decision. Among 530 false-positive marks, 459 were identified clearly for pseudo-lesions; the most common cause was marginal shadowing, followed by Cooper’s ligament shadowing, peri-areolar shadowing, rib, and skin lesions. Even though CAD does not improve the performance of ABUS and a large number of false-positive marks were detected, the addition of CAD reduces RT, especially in the case of negative screening ultrasound.

Retrospective Review of Missed Cancer Detection and Its Mammography Findings with Artificial-Intelligence-Based, Computer-Aided Diagnosis

To investigate whether artificial-intelligence-based, computer-aided diagnosis (AI-CAD) could facilitate the detection of missed cancer on digital mammography, a total of 204 women diagnosed with breast cancer with diagnostic (present) and prior mammograms between 2018 and 2020 were included in this study. Two breast radiologists reviewed the mammographic features and classified them into true negative, minimal sign or missed cancer. They analyzed the AI-CAD results with an abnormality score and assessed whether the AI-CAD correctly localized the known cancer sites. Of the 204 cases, 137 were classified as true negative, 33 as minimal signs, and 34 as missed cancer. The sensitivity, specificity and diagnostic accuracy of AI-CAD were 84.7%, 91.5% and 86.3% on diagnostic mammogram and 67.2%, 91.2% and 83.38% on prior mammogram, respectively. The AI-CAD correctly localized 27 cases from 34 missed cancers on prior mammograms. The findings in the preceding mammography of AI-CAD-detected missed cancer were common in the order of calcifications, focal asymmetry and asymmetry. Asymmetry was the most common finding among the seven cases, which could not be detected by AI-CAD in the missed cases (5/7). The assistance of AI-CAD can be helpful in the early detection of breast cancer in mammography screenings.

Correlation between antibiotic consumption and the incidence of healthcare facility-onset Clostridioides difficile infection: a retrospective chart review and analysis

Background

Healthcare facility-onset Clostridioides difficile infection is the leading cause of antibiotic-associated diarrhea, and is associated with morbidity and mortality. The use of antibiotics is an important risk factor for healthcare facility-onset C. difficile infection. We evaluated the correlation between the incidence of healthcare facility-onset C. difficile infection and antibiotic consumption, according to antibiotic class.

Methods

Patients with healthcare facility-onset C. difficile infection from January 2017 to December 2018 at Konkuk University Medical Center (a tertiary medical center) were included. We evaluated changes in the incidence of healthcare facility-onset C. difficile infection and antibiotic consumption. The correlation between the incidence of healthcare facility-onset C. difficile infection and antibiotic consumption was evaluated two ways: without a time interval and with 1-month interval matching.

Results

A total of 446 episodes of healthcare facility-onset C. difficile infection occurred during the study period. The incidence of healthcare facility-onset C. difficile infection was 9.3 episodes per 10,000 patient-days, and increased significantly. We observed an increase in the consumption of β-lactam/β-lactamase inhibitors, and a decrease in the consumption of other classes of antibiotics, with a significant decrease in the consumption of fluoroquinolones, glycopeptides, and clindamycin (P = 0.01, P < 0.001, and P = 0.001, respectively). The consumption of β-lactam/β-lactamase inhibitors was independently correlated with the incidence of healthcare facility-onset C. difficile infection in the analysis without a time interval. When the analysis was conducted with 1-month interval matching, glycopeptide consumption was independently associated with the incidence of healthcare facility-onset C. difficile infection.

Conclusions

Despite the reduction in fluoroquinolone and clindamycin consumption, the incidence of healthcare facility-onset C. difficile infection increased during the study period, and was correlated with increased consumption of β-lactam/β-lactamase inhibitors. Reduced consumption of specific antibiotics may be insufficient to reduce the incidence of healthcare facility-onset C. difficile infection.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13756-021-00986-9.

Risk Stratification of Ductal Carcinoma In Situ and Texture Analysis of Contrast-Enhanced Breast Magnetic Resonance Imaging

OBJECTIVE

The aim of this study was to investigate whether texture analysis of contrast-enhanced T1 weighted images could predict risk of ductal carcinoma in situ (DCIS).

METHODS

The study included 185 DCIS lesions that were classified as either low risk or non-low risk using surgical pathology records. All magnetic resonance imaging texture analyses were performed using postprocessing software, and texture-derived parameters were extracted.

RESULTS

The sphericity, compactness, and spherical disproportion were significantly different in the low-risk and non-low risk groups using the Van Nuys Prognostic Index (mean ± SD, 0.479 ± 0.189 vs 0.414 ± 0.176, 0.161 ± 0.159 vs 0.112 ± 0.134, and 2.569 ± 1.434 vs 2.934 ± 1.374, respectively; P < 0.05). In the univariate analyses, sphericity (odds ratio, 7.091; 95% confidence interval, 1.236-40.666; P = 0.028) and compactness (odds ratio, 9.267; 95% confidence interval, 1.125-76.360; P = 0.039) were significantly associated with a high probability of being low risk according to the Van Nuys Prognostic Index.

CONCLUSIONS

Whole-lesion texture analysis may be helpful in identifying patients classified as having low-risk DCIS before surgery.

Impact of antiviral treatment on long-term prognosis in non-immunocompromised patients with CMV reactivation

Background

Reactivation of human cytomegalovirus (CMV) occurs in non-immunocompromised patients with or without specific organ involvement, but it is still unknown whether it has a clinical implication on long-term prognosis or not.

Methods

A retrospective cohort study evaluating non-immunocompromised adult patients with CMV reactivation was conducted during the period between January 2010 and February 2018. Patients were divided into ganciclovir-treated and non-treated groups. Patients who died within 30 days from CMV reactivation were excluded as they died from complex causes of conditions. Survivors were followed for 30-months to evaluate long-term prognosis.

Results

A total of 136 patients with CMV reactivation was included, consisting of 66 ganciclovir-treated (48.5%) and 70 non-treated (51.5%) patients. Overall, patients were old-aged (median 70 years old) and most were treated with pneumonia of any cause (91.2%). More patients in ganciclovir-treated group were treated at intensive care unit (43.9% vs 24.3%, respectively) and had higher viral load over 5000 copies/ml (48.5% vs 22.9%) than non-treated group (all P < 0.05). Primary and secondary endpoints including 30-months survival (28.0 vs 38.9%, respectively) and 12-months survival (40.3% vs 49.2%) were not statistically different between the ganciclovir-treated and non-treated groups. In the multivariate analyses, ganciclovir treatment was not associated with 30-months survival (HR 1.307, 95% CI 0.759–2.251) and 12-months survival (HR 1.533, 95% CI 0.895–2.624).

Conclusion

In a retrospective cohort study evaluating non-immunocompromised patients with CMV reactivation, ganciclovir treatment was not associated with long-term prognosis. Antiviral treatment in this condition would not be necessary unless organ involvement is suspected.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-021-06098-4.

Fully Automatic Assessment of Background Parenchymal Enhancement on Breast MRI Using Machine-Learning Models

BACKGROUND

Automated measurement and classification models with objectivity and reproducibility are required for accurate evaluation of the breast cancer risk of fibroglandular tissue (FGT) and background parenchymal enhancement (BPE).

PURPOSE

To develop and evaluate a machine-learning algorithm for breast FGT segmentation and BPE classification.

STUDY TYPE

Retrospective.

POPULATION

A total of 794 patients with breast cancer, 594 patients assigned to the development set, and 200 patients to the test set.

FIELD STRENGTH/SEQUENCE

3T and 1.5T; T₂ -weighted, fat-saturated T₁ -weighted (T₁ W) with dynamic contrast enhancement (DCE).

ASSESSMENT

Manual segmentation was performed for the whole breast and FGT regions in the contralateral breast. The BPE region was determined by thresholding using the subtraction of the pre- and postcontrast T₁ W images and the segmented FGT mask. Two radiologists independently assessed the categories of FGT and BPE. A deep-learning-based algorithm was designed to segment and measure the volume of whole breast and FGT and classify the grade of BPE.

STATISTICAL TESTS

Dice similarity coefficients (DSC) and Spearman correlation analysis were used to compare the volumes from the manual and deep-learning-based segmentations. Kappa statistics were used for agreement analysis. Comparison of area under the receiver operating characteristic (ROC) curves (AUC) and F1 scores were calculated to evaluate the performance of BPE classification.

RESULTS

The mean (±SD) DSC for manual and deep-learning segmentations was 0.85 ± 0.11. The correlation coefficient for FGT volume from manual- and deep-learning-based segmentations was 0.93. Overall accuracy of manual segmentation and deep-learning segmentation in BPE classification task was 66% and 67%, respectively. For binary categorization of BPE grade (minimal/mild vs. moderate/marked), overall accuracy increased to 91.5% in manual segmentation and 90.5% in deep-learning segmentation; the AUC was 0.93 in both methods.

DATA CONCLUSION

This deep-learning-based algorithm can provide reliable segmentation and classification results for BPE.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY STAGE: 2.

Clinical Benefits of Piperacillin/Tazobactam versus a Combination of Ceftriaxone and Clindamycin in the Treatment of Early, Non-Ventilator, Hospital-Acquired Pneumonia in a Community-Based Hospital

Purpose

There is an increasing prevalence of multidrug-resistant (MDR) organisms worldwide. Therefore, broad-spectrum antibiotics are recommended in the treatment of hospital-acquired pneumonia (HAP). However, it remains controversial whether patients with early onset, non-ventilator HAP (NV-HAP) should also be empirically treated with broad-spectrum antibiotics. We compared the clinical benefit of ceftriaxone plus clindamycin vs piperacillin/tazobactam as the initial empirical treatment of adults with early NV-HAP.

Patients and Methods

Retrospective cohort study was conducted in adult patients who were diagnosed with early, NV-HAP between January 2013 and June 2017 at a community-based tertiary care hospital. Patients were eligible for inclusion if they had received empiric treatment with either ceftriaxone and clindamycin or piperacillin/tazobactam for at least 3 days. Patients with increased risk of MDR pathogens were excluded.

Results

A total of 89 patients were treated with ceftriaxone and clindamycin, while 124 received piperacillin/tazobactam. There were no significant differences between the two antibiotic groups with regard to median age, sex, or risk of pneumonia. The 30-day all-cause mortality did not differ significantly between the ceftriaxone plus clindamycin and piperacillin/tazobactam groups (4.5% vs 1.6%, P=0.202, respectively). However, in multivariate analysis, clinical failure was more frequent in the ceftriaxone plus clindamycin group than in the piperacillin/tazobactam group (HR 3.316; 95% CI, 1.589–6918, P=0.001).

Conclusion

Treatment with piperacillin/tazobactam was more effective than that with ceftriaxone plus clindamycin in patients with early NV-HAP. This study supports the recent treatment recommendations that patients with early NV-HAP should be treated empirically with broad-spectrum antibiotics.

Clinical factors influencing the performance of bacterial multiplex polymerase chain reaction in patients with community-onset pneumonia

The etiologic diagnostic yield of community-onset pneumonia (COP) using conventional methods is low. Bacterial multiplex polymerase chain reaction (mPCR) has been shown to be more sensitive than conventional methods. This study assessed the clinical factors influencing bacterial mPCR results in patients with COP. Patients with COP admitted to a tertiary care hospital between November 2015 and April 2016 were retrospectively assessed. Conventional methods included culture-based methods and serology for Mycoplasma pneumoniae. Bacterial mPCR that could identify Streptococcus pneumoniae, Haemophilus influenzae, Mycoplasma pneumoniae, and Legionella pneumophilia was performed. Bacterial mPCR was performed in a total of 342 patients with COP in the study. Bacterial mPCR alone provided etiology in 99 patients. The total etiologic diagnosis rates improved from 22.2 to 51.1% when bacterial mPCR was added to conventional methods. Additional diagnostic benefits of bacterial mPCR were more prominent in the prior antibiotic non-exposure group (77.8% vs 63.5%, P = 0.015) and in the low-risk group with low CURB 65 score (62.6% vs 44.9%, P = 0.005). Patients who required ICU care, those with healthcare-associated pneumonia (HCAP), and patients with any underlying diseases were not associated with the additional pathogen detection rates using bacterial mPCR. By supplementing conventional diagnostic methods with bacterial mPCR-based methods, the overall pathogen detection rates improved in patients with COP. Moreover, the additional diagnostic usefulness of bacterial mPCR was significantly higher in patients without prior antibiotic exposure and in the mild-to-moderate-risk group with lower CURB 65 score.

Evaluation of the effect of computer aided diagnosis system on breast ultrasound for inexperienced radiologists in describing and determining breast lesions

AIM

To investigate the effect of a computer-aided diagnosis (CAD) system on breast ultrasound (US) for inexperienced radiologists in describing and determining breast lesions.

MATERIALS AND METHODS

Between October 2015 to January 2017, 500 suspicious or probable benign lesions in 413 patients were reviewed. Five experienced readers retrospectively reviewed for each of 100 lesions according to the Breast Imaging Reporting and Data System (BI-RADS) lexicon and category, with CAD system (S-detectTM). The readers then made final decisions by combining CAD results to their US results. Using the nested experiment design, five inexperienced readers were asked to select the appropriate BI-RADS lexicons, categories, CAD results, and combination results for each of the 100 lesions, retrospectively. Diagnostic performance of experienced and inexperienced radiologists and CAD were assessed. For each case, agreements in the lexicons and categories were analyzed among the experienced reader, inexperienced reader and CAD.

RESULTS

Indicators of the diagnostic performance for breast malignancy of the experienced group (AUC=0.83, 95%CI [0.80, 0.86]) were similar or higher than those of CAD (AUC = 0.79, 95%CI[0.74, 0.83], p=0.101), except for specificity. Conversely, indicators of diagnostic performance of inexperienced group (AUC=0.65, 95%CI[0.58, 0.71]) did not differ from or were lower than those of CAD(AUC=0.73, 95%CI[0.67, 0.78], p=0.013). Also, the diagnostic performance of the inexperienced group after combination with the CAD result was significantly improved (0.71, 95% CI [0.65, 0.77], p=0.001), whereas that of the experienced group did not change after combination with the CAD result, except for specificity and positive predictive value (PPV). Kappa values for the agreement of the categorization between CAD and each radiologist group were increased after applying the CAD result to their result of general US. Especially, the increase of the Kappa value was higher in the inexperienced group than in the experienced group. Also, for all the lexicons, the Kappa values between the experienced group and CAD were higher than those between the inexperienced group and CAD.

CONCLUSION

By using the CAD system for classification of breast lesions, diagnostic performance of the inexperienced radiologists for malignancy was significantly improved, and better agreement was observed in lexicons between the experienced group and CAD than between the inexperienced group and CAD. CAD may be beneficial and educational for the inexperienced group.

Treatment of community-onset pneumonia in neutropenic cancer patients: β-lactam monotherapy versus combination antibiotic regimens

Background

Although β-lactam monotherapy may be sufficient in non-critically ill patients with community-acquired pneumonia, the value of combination antibiotic regimens in community-onset neutropenic pneumonia remains unclear.

Methods

A retrospective cohort study was conducted to compare the effects of combination antibiotic regimens to those of β-lactam monotherapy in cancer patients with community-onset neutropenic pneumonia. Electronic medical records of patients diagnosed with community-onset neutropenic pneumonia between March 1995 and February 2015 at a tertiary care center were reviewed.

Results

During the study period, 165 cancer patients with community-onset neutropenic pneumonia were identified. Seventy-two patients received β-lactam monotherapy and 93 received combination therapy (β-lactam plus either a macrolide or fluoroquinolone). Causative pathogens were identified in 27.9% of the patients, and only two were positive for atypical pathogens. Although 30-day mortality was higher in the β-lactam group (15.3% versus 4.3%; P = 0.015), combination therapy was not associated with a statistically significant survival benefit in the multivariate analysis (hazard ratio 0.85, 95% confidence interval 0.20–3.67; P = 0.827). Duration of neutropenia, C-reactive protein level, and Multinational Association for Supportive Care in Cancer risk index were significant factors for 30-day mortality. In a subgroup analysis of patients treated with cefepime, the most frequently used β-lactam (63.0%), combination therapy also showed no significant survival benefit.

Conclusions

Combination antibiotic regimens were not associated with a survival benefit over β-lactam monotherapy in the treatment of community-onset neutropenic pneumonia. Unnecessary combination therapy should be reconsidered in cancer patients who are at high risk for adverse drug reactions and colonization with multi-drug resistant organisms.

Electronic supplementary material

The online version of this article (10.1186/s41479-019-0061-1) contains supplementary material, which is available to authorized users.

Infectious Complications in Patients Who Received High-Volume Plasma Exchange Prior to Liver Transplant: A Case Report

Acute liver failure is a rare but life-threatening medical emergency. Despite advancements in medical management, mortality rates of acute liver failure remain high. Currently, liver transplant is the only definitive therapeutic option available. High-volume plasma exchange has been shown to increase transplant-free survival in patients with acute liver failure before liver transplant. However, the occurrence of infectious complications in patients who receive this treatment has not been well studied. We report 2 cases of severe opportunistic infections occurring within 30 days of transplant in patients who underwent high-volume plasma exchange before liver transplant.

Differentiation of multiple myeloma and metastases: Use of axial diffusion-weighted MR imaging in addition to standard MR imaging at 3T

Background

Metastasis and multiple myeloma are common malignant bone marrow lesions which may be difficult to distinguish because of similar imaging findings. The purpose of this study was to determine the value of adding diffusion-weighted imaging (DWI) to standard MR imaging to differentiate multiple myeloma from metastasis.

Methods

25 patients with metastasis and 18 patients with multiple myeloma underwent 3T MR imaging with DWI (b = 0, 800 s/mm²) were enrolled. They all had pathologically confirmed bone lesions and were in a treatment naïve state. Two readers who were blind of final diagnosis measured the average ADC (ADCav) and minimum ADC (ADCmin) on the DWI. They then estimated the diagnosis, based on the standard MR imaging and measured ADC values. Another reader performed histogram analysis on the whole tumor volume and obtained mean ADC (ADCvol), standard deviation (SDvol), skewness, and kurtosis. Comparison of the obtained values from DWI was performed by the t-test or Mann-Whitney U test. The receiver operating characteristic (ROC) curve with areas under the curve (AUC) was used to obtain the cut off values and to evaluate the diagnostic performance of the two readers.

Results

ADCav, ADCmin, and ADCvol of multiple myeloma were significantly lower than those of metastasis: ADCav, 752 μm²/sec versus 1081 μm²/sec; ADCmin, 704 μm²/sec vs 835 μm²/sec; ADCvol 761 μm²/sec vs 1184 μm²/sec (p < .001). In histogram analysis, ADC values of multiple myeloma showed narrow distribution than metastasis: SDvol, 144 vs 257 (p < .001). Areas under the receiver operating characteristic curve was significantly higher with additive DWI than standard MR alone: 0.762 vs 0.953; 0.706 vs 0.950 (p < .05) for two readers.

Conclusions

This study suggested that the addition of axial DWI to standard MR imaging can be helpful to diagnose multiple myeloma from metastasis at 3T.

Reproducibility of mRECIST in Measurement and Response Assessment for Hepatocellular Carcinoma Treated by Transarterial Chemoembolization

RATIONALE AND OBJECTIVES

To evaluate the reproducibility of Modified Response Evaluation Criteria in Solid Tumors (mRECIST) for hepatocellular carcinoma (HCC) after transarterial chemoembolization (TACE) using multiphasic computed tomography.

MATERIALS AND METHODS

The institutional review board approved this retrospective study. We evaluated 97 patients who underwent TACE (60 conventional TACE [cTACE] and 37 drug-eluting bead TACE [DEB-TACE]) for HCC from 2010 to 2014. Four radiologists evaluated pairs of dynamic liver CTs scanned within 2 months before and after TACE based on mRECIST. Assessment of intra- or interobserver reproducibility for response categorization and sum of long diameter were evaluated using weighted kappa statistics (κ) and intraclass correlation coefficients, respectively. The relationship between concordance of target lesion selection and agreement of target lesion response was evaluated using Fisher exact test.

RESULTS

Intraobserver reproducibility for overall response was moderate to excellent (κ = 0.525-0.865). Interobserver reproducibility was improved on the second review compared to the first review and it was good in both treatment groups (κ = 0.627 for cTACE and 0.602 for DEB-TACE). Between the two treatment methods, intra- or interobserver reproducibility was better after cTACE than DEB-TACE. Intraclass correlation coefficients for sum of long diameter measurement showed excellent intra- or interobserver reproducibility. The concordance rate of target lesion selection was significantly higher for patients with radiologists' agreement for target lesion response than patients with disagreed response (P = .003).

CONCLUSIONS

The intra- and interobserver reproducibility of mRECIST in patients with HCC after TACE was moderate to excellent, and the reproducibility was slightly better after cTACE than DEB-TACE.

Host susceptibility to MERS-CoV infection, a retrospective cohort study of the 2015 Korean MERS outbreak

To evaluate host susceptibility factors to Middle East respiratory syndrome coronavirus (MERS-CoV) infection, we conducted a retrospective cohort study from the single largest exposure event of the 2015 Korean MERS outbreak. A total of 175 patients were closely exposed to a super-spreader, 26 of which were infected (14.9%). In a multivariate analysis, history of autologous stem cell transplantation (HR, 31.151; 95% CI, 5.447-178.145; P < 0.001) and tachypnea at ED (HR, 4.392; 95% CI, 1.402-13.761; P = 0.011) were significantly associated with MERS-CoV infection.

Histogram analysis of volume-based apparent diffusion coefficient in breast cancer

Background Breast cancer is a heterogeneous disease. Recent studies showed that apparent diffusion coefficient (ADC) values have various association with tumor aggressiveness and prognosis. Purpose To evaluate the value of histogram analysis of ADC values obtained from the whole tumor volume in invasive ductal cancer (IDC) and ductal carcinoma in situ (DCIS). Material and Methods This retrospective study included 201 patients with confirmed DCIS (n = 37) and IDC (n = 164). The IDC group was divided into two groups based on the presence of a DCIS component: IDC-DCIS (n = 76) and pure IDC (n = 88). All patients underwent preoperative breast magnetic resonance imaging (MRI) with diffusion-weighted images at 3.0 T. Histogram parameters of cumulative ADC values, skewness, and kurtosis were calculated and statistically analyzed. Results The differences between DCIS, IDC-DCIS, and pure IDC were significant in all percentiles of ADC values, in descending order of DCIS, IDC-DCIS, and pure IDC. IDC showed significantly lower ADC values than DCIS, and ADC50 was the best indicator for discriminating IDC from DCIS, with a threshold of 1.185 × 10^-3 mm²/s (sensitivity of 82.9%, specificity of 75.7%). However, multivariate analysis of obtained ADC values showed no significant differences between DCIS, IDC-DCIS, and pure IDC ( P > 0.05). Conclusion Volume-based ADC values showed association with heterogeneity of breast cancer. However, there was no additional diagnostic performance in histogram analysis for differentiating between DCIS, IDC-DCIS, and pure IDC.

Serologic responses of 42 MERS-coronavirus-infected patients according to the disease severity

We evaluated serologic response of 42 Middle East respiratory syndrome coronavirus (MERS-CoV)-infected patients according to 4 severity groups: asymptomatic infection (Group 0), symptomatic infection without pneumonia (Group 1), pneumonia without respiratory failure (Group 2), and pneumonia progressing to respiratory failure (Group 3). None of the Group 0 patients showed seroconversion, while the seroconversion rate gradually increased with increasing disease severity (0.0%, 60.0%, 93.8%, and 100% in Group 0, 1, 2, 3, respectively; P = 0.001). Group 3 patients showed delayed increment of antibody titers during the fourth week, while Group 2 patients showed robust increment of antibody titer during the third week. Among patients having pneumonia, 75% of deceased patients did not show seroconversion by the third week, while 100% of the survived patients were seroconverted (P = 0.003).

Suggested new breakpoints of anti-MERS-CoV antibody ELISA titers: performance analysis of serologic tests

To provide optimal cut-off values of anti-Middle East respiratory syndrome coronavirus (MERS-CoV) serologic tests, we evaluated performance of ELISA IgG, ELISA IgA, IFA IgM, and IFA IgG using 138 serum samples of 49 MERS-CoV-infected patients and 219 serum samples of 219 rRT-PCR-negative MERS-CoV-exposed healthcare personnel and patients. The performance analysis was conducted for two different purposes: (1) prediction of neutralization activity in MERS-CoV-infected patients, and (2) epidemiologic surveillance of MERS-CoV infections among MERS-CoV-exposed individuals. To evaluate performance according to serum collection time, we used ‘days post onset of illness (dpoi)’ and ‘days post exposure (dpex)’ assessing neutralization activity and infection diagnosis, respectively. Performance of serologic tests improved with delayed sampling time, being maximized after a seroconversion period. In predicting neutralization activity, ELISA IgG tests showed optimal performance using sera collected after 21 dpoi at cut-off values of OD ratio 0.4 (sensitivity 100% and specificity 100%), and ELISA IgA showed optimal performance using sera collected after 14 dpoi at cut-off value of OD ratio 0.2 (sensitivity 85.2% and specificity 100%). In diagnosis of MERS-CoV infection, ELISA IgG exhibited optimal performance using sera collected after 28 dpex, at a cut-off value of OD ratio 0.2 (sensitivity 97.3% and specificity 92.9%). These new breakpoints are markedly lower than previously suggested values (ELISA IgG OD ratio 1.1, sensitivity 34.8% and specificity 100% in the present data set), and the performance data help serologic tests to be practically used in the field of MERS management.

Atypical presentations of MERS-CoV infection in immunocompromised hosts

During the 2015 Korean MERS outbreak, we experienced atypical presentations of MERS-CoV infections in three immunocompromised hosts that warranted exceptional management. Case 1 showed delayed symptom development after a four-day asymptomatic period, Case 2 experienced a 20-day incubation period, and Case 3 exhibited persistent viral shedding without clinical deterioration. Recognizing these exceptions is extremely important in the management of MERS-CoV-exposed or -infected patients and for control of potential MERS outbreaks.

Empirical combination of a β-lactam to vancomycin may not improve outcomes of methicillin-susceptible Staphylococcus aureus bacteremia, compared to vancomycin monotherapy

To evaluate effect of empirical combination of a β-lactam to vancomycin and vancomycin monotherapy in Staphylococcus aureus bacteremia (MSSA-B), we conducted a retrospective cohort study. Electronic medical records of individuals who were diagnosed with MSSA-B between January 2005 and February 2015 at a tertiary care center were reviewed. Patients were classified into three groups according to empirical antibiotic regimen (BL group, β-lactam; VAN group, vancomycin; BV group, combination of β-lactam and vancomycin), and 30-day all-cause mortality of each group was compared. During the study period, 561 patients with MSSA-B were identified. After exclusion of 198 patients (36 with poly-microbial infection, 114 expired within 2 days, and 48 already received parenteral antibiotics) and a matching process, 46 patients for each group were included. Baseline characteristics were similar except for severity and comorbidity scores. The 30-day mortality for all three groups were not significantly different (BL 4.3%, VAN 6.5%, BV 8.7%; P = 0.909). In a multivariate analysis, type of empirical antibiotic regimen was not statistically associated with 30-day all-cause mortality. In comparison with the VAN group, the BV group yielded a HR of 0.579 (95% CI = 0.086-3.890, P = 0.574). Pitt bacteremia score was the only significant factor for mortality. The empirical combination of a β-lactam to vancomycin was not associated with lower mortality in treating MSSA-B, compared to vancomycin monotherapy.

Differential Cell Count and CRP Level in Blood as Predictors for Middle East Respiratory Syndrome Coronavirus Infection in Acute Febrile Patients during Nosocomial Outbreak

A case-control study was performed to identify clinical predictors for Middle East respiratory syndrome coronavirus (MERS-CoV) infection among patients with acute febrile illness during the nosocomial outbreak. Patients with MERS-CoV were more likely to have monocytosis with normal white blood cell (WBC) count and lower C-reactive protein (CRP) level. Simple laboratory data such as complete blood counts (CBC) with differential count could be a useful marker for the prediction of MERS and triage at the initial presentation of acute febrile patients in outbreak setting.

MERS-CoV outbreak following a single patient exposure in an emergency room in South Korea: an epidemiological outbreak study

BACKGROUND

In 2015, a large outbreak of Middle East respiratory syndrome coronavirus (MERS-CoV) infection occurred following a single patient exposure in an emergency room at the Samsung Medical Center, a tertiary-care hospital in Seoul, South Korea. We aimed to investigate the epidemiology of MERS-CoV outbreak in our hospital.

METHODS

We identified all patients and health-care workers who had been in the emergency room with the index case between May 27 and May 29, 2015. Patients were categorised on the basis of their exposure in the emergency room: in the same zone as the index case (group A), in different zones except for overlap at the registration area or the radiology suite (group B), and in different zones (group C). We documented cases of MERS-CoV infection, confirmed by real-time PCR testing of sputum samples. We analysed attack rates, incubation periods of the virus, and risk factors for transmission.

FINDINGS

675 patients and 218 health-care workers were identified as contacts. MERS-CoV infection was confirmed in 82 individuals (33 patients, eight health-care workers, and 41 visitors). The attack rate was highest in group A (20% [23/117] vs 5% [3/58] in group B vs 1% [4/500] in group C; p<0·0001), and was 2% (5/218) in health-care workers. After excluding nine cases (because of inability to determine the date of symptom onset in six cases and lack of data from three visitors), the median incubation period was 7 days (range 2-17, IQR 5-10). The median incubation period was significantly shorter in group A than in group C (5 days [IQR 4-8] vs 11 days [6-12]; p<0·0001). There were no confirmed cases in patients and visitors who visited the emergency room on May 29 and who were exposed only to potentially contaminated environment without direct contact with the index case. The main risk factor for transmission of MERS-CoV was the location of exposure.

INTERPRETATION

Our results showed increased transmission potential of MERS-CoV from a single patient in an overcrowded emergency room and provide compelling evidence that health-care facilities worldwide need to be prepared for emerging infectious diseases.

FUNDING

None.

Predictive factors for pneumonia development and progression to respiratory failure in MERS-CoV infected patients

BACKGROUND

After the 2015 Middle East respiratory syndrome (MERS) outbreak in Korea, prediction of pneumonia development and progression to respiratory failure was emphasized in control of MERS outbreak.

METHODS

MERS-CoV infected patients who were managed in a tertiary care center during the 2015 Korean MERS outbreak were reviewed. To analyze predictive factors for pneumonia development and progression to respiratory failure, we evaluated clinical variables measured within three days from symptom onset.

RESULTS

A total of 45 patients were included in the study: 13 patients (28.9%) did not develop pneumonia, 19 developed pneumonia without respiratory failure (42.2%), and 13 progressed to respiratory failures (28.9%). The identified predictive factors for pneumonia development included age ≥45 years, fever ≥37.5 °C, thrombocytopenia, lymphopenia, CRP ≥ 2 mg/dL, and a threshold cycle value of PCR less than 28.5. For respiratory failure, the indicators included male, hypertension, low albumin concentration, thrombocytopenia, lymphopenia, and CRP ≥ 4 mg/dL (all P < 0.05). With ≥ two predictive factors for pneumonia development, 100% of patients developed pneumonia. Patients lacking the predictive factors did not progress to respiratory failure.

CONCLUSION

For successful control of MERS outbreak, MERS-CoV infected patients with ≥ two predictive factors should be intensively managed from the initial presentation.

Control of an Outbreak of Middle East Respiratory Syndrome in a Tertiary Hospital in Korea

BACKGROUND

In 2015, a large outbreak of Middle East respiratory syndrome (MERS) occurred in the Republic of Korea. Half of the cases were associated with a tertiary care university hospital.

OBJECTIVE

To document the outbreak and successful control measures.

DESIGN

Descriptive study.

SETTING

A 1950-bed tertiary care university hospital.

PATIENTS

92 patients with laboratory-confirmed MERS and 9793 exposed persons.

MEASUREMENTS

Description of the outbreak, including a timeline, and evaluation of the effectiveness of the control measures.

RESULTS

During the outbreak, 92 laboratory-confirmed MERS cases were associated with a large tertiary care hospital, 82 of which originated from unprotected exposure to 1 secondary patient. Contact tracing and monitoring exposed patients and assigned health care workers were at the core of the control measures in the outbreak. Nontargeted screening measures, including body temperature screening among employees and visitors at hospital gates, monitoring patients for MERS-related symptoms, chest radiographic screening, and employee symptom monitoring, did not detect additional patients with MERS without existing transmission links. All in-hospital transmissions originated from 3 patients with MERS who also had pneumonia and productive cough.

LIMITATIONS

This was a retrospective single-center study. Statistical analysis could not be done. Because this MERS outbreak originated from a superspreader, effective control measures could differ in endemic areas or in other settings.

CONCLUSION

Control strategies for MERS outbreaks should focus on tracing contacts of persons with epidemiologic links. Adjusting levels of quarantine and personal protective equipment according to the assumed infectivity of each patient with MERS may be appropriate.

PRIMARY FUNDING SOURCE

Samsung Biomedical Research Institute.

Ultrasonographic findings of low-grade endometrial stromal sarcoma of the uterus with a focus on cystic degeneration

PURPOSE

The goal of this study was to perform a retrospective analysis of the ultrasonographic findings associated with low-grade endometrial stromal sarcoma.

METHODS

Ten pathologically confirmed cases of low-grade endometrial stromal sarcoma at our institution from January 2007 to April 2014 were retrospectively reviewed. All patients underwent a preoperative transvaginal ultrasound. Two radiologists came to a consensus regarding the location, size, margin, and echogenicity of the tumor, as well as the presence of intratumoral cystic degeneration and its extent and configuration.

RESULTS

Low-grade endometrial stromal sarcoma manifested as an intramural mass protruding into the endometrial cavity (n=6) or as a purely intramural mass (n=4). The maximal diameter of the lesion ranged from 4 to 9.1 cm (mean, 6.2 cm). The imaging features of low-grade endometrial stromal sarcoma were variable: six cases involved predominantly solid masses containing cystic degeneration, one was a predominantly unilocular cystic mass, two were ill-defined infiltrative solid masses, and one was a well-defined solid mass. Among the seven cases with internal cystic degeneration, five patients showed a multiseptated cystic area or a cystic area with multiple small clusters, while a unilocular cystic area within the tumor was found in two patients.

CONCLUSION

Low-grade endometrial stromal sarcoma is associated with variable ultrasonographic findings with regard to the location, margin, and configuration of the lesion. Multiseptated cystic areas and multiple small areas of cystic degeneration are common.

Delayed Perforation Occurring after Endoscopic Submucosal Dissection for Early Gastric Cancer

Delayed perforation is a very rare complication of endoscopic submucosal dissection (ESD), with a reported incidence of 0.1% to 0.45%. Few reports exist on the clinical features and outcomes of delayed perforation after ESD, and it is unclear whether the optimal management strategy is emergency surgery or endoscopic closure with conservative treatment. Here, we report two cases of delayed perforation occurring after ESD for early gastric cancer. In both cases, lesions were located in the antrum, and tumor depths were confined to the mucosal layer. Total procedure times for ESD were 25 and 45 minutes, respectively. Because delayed perforation may be associated with excessive thermal damage and necrosis of the muscle layer, treatment with emergency surgery should be used instead of conservative management in cases of delayed perforation after ESD.

Histogram analysis of apparent diffusion coefficient at 3.0t: Correlation with prognostic factors and subtypes of invasive ductal carcinoma

PURPOSE

To evaluate apparent diffusion coefficient (ADC) histogram parameters that show correlations with prognostic factors and subtypes of breast cancer.

MATERIALS AND METHODS

At 3.0T, various ADC histogram parameters were calculated including the entire tumor volume in 173 invasive ductal carcinomas: the minimum, 10th percentile, mean, median, 90th percentile, and maximum. ADC parameters were correlated with prognostic factors and subtype.

RESULTS

The mean ADCmedian value was significantly higher in the group with lymph node metastasis, HER2 positivity, and a Ki-67 value <14% than in the group with negativity for lymph node metastasis, HER2 negativity, and a Ki-67 value ≥14% (0.907, 0.978, and 0.941 vs. 0.735, 0.778, and 0.761 × 10(-3) mm(2) /s, respectively) (P < 0.01). There was no significant correlation between ADCmedian and tumor size, histologic grade, estrogen receptor expression, and progesterone receptor expression (P = 0.272, 0.113, 0.261, and 0.181, respectively). For most ADC parameters except for ADCmin , the mean of variable ADC parameters of HER2-positive, luminal A, luminal B-HER2(+), triple-negative, and luminal B-HER2(-) diseases were arranged in descending order (1.175, 0.936, 0.863, 0.811, and 0.665 × 10(-3) mm(2) /s in ADCmedian , respectively) with statistical significant difference (P < 0.001). In multivariate analysis, histologic grade, the Ki-67 index, and HER2 expression were statistically significant explanatory prognostic factors for ADCmedian and the Ki-67 index had the most robust effects on ADC parameters (standardized coefficient = -0.317).

CONCLUSION

Various ADC parameters were correlated with prognostic factors and subtype, except for ADCmin . HER2 positivity showed high ADC values and high Ki-67 index revealed low ADC values.

The effects of varying volumes of crystalloid administration before cesarean delivery on maternal hemodynamics and colloid osmotic pressure

The value of intravenous crystalloid administration in preventing spinal-induced hypotension in the parturient has recently been questioned. Also, the association between increasing crystalloid volume and decreasing postpartum colloid osmotic pressure (COP) raises concern regarding the risk of maternal and fetal pulmonary edema. To study the dose-response effect of varying amounts of crystalloid volume prior to spinal anesthesia, we measured maternal hemodynamic variables and maternal and fetal COP in three groups of healthy parturients receiving spinal anesthesia for elective cesarean delivery. Fifty-five parturients were randomized in a double-blind fashion to receive one of 10, 20, or 30 mL/kg of crystalloid volumes prior to induction of spinal anesthesia. Measurements included mean arterial blood pressure (MAP), cardiac index (CI), and systemic vascular resistance index (SVRI) recorded using noninvasive thoracic impedance monitoring until delivery. Maternal and neonatal COP were measured. All groups showed declines in MAP and SVRI from baseline at 5 min after spinal anesthesia, but the amount of decline did not differ among groups. Total ephedrine and additional intravenous (i.v.) fluid administered did not differ among groups. The 20- and 30- mL/kg groups showed a larger decline in maternal COP than the 10-mL/kg group; no differences in neonatal COP were seen with varying preload. We conclude that increasing the amount of i.v. crystalloid administered to 30 mL/kg in the healthy parturient does not significantly alter maternal hemodynamics or ephedrine requirements after spinal anesthesia and has no apparent benefit.