Can a machine-learning model improve the prediction of nodal stage after a positive sentinel lymph node biopsy in breast cancer?

Background: The current standard for evaluating axillary nodal burden in clinically node negative breast cancer is sentinel lymph node biopsy (SLNB). However, the accuracy of SLNB to detect nodal stage N2-3 remains debatable. Nomograms can help the decision-making process between axillary treatment options. The aim of this study was to create a new model to predict the nodal stage N2-3 after a positive SLNB using machine learning methods that are rarely seen in nomogram development.Material and methods: Primary breast cancer patients who underwent SLNB and axillary lymph node dissection (ALND) between 2012 and 2017 formed cohorts for nomogram development (training cohort, N = 460) and for nomogram validation (validation cohort, N = 70). A machine learning method known as the gradient boosted trees model (XGBoost) was used to determine the variables associated with nodal stage N2-3 and to create a predictive model. Multivariate logistic regression analysis was used for comparison.Results: The best combination of variables associated with nodal stage N2-3 in XGBoost modeling included tumor size, histological type, multifocality, lymphovascular invasion, percentage of ER positive cells, number of positive sentinel lymph nodes (SLN) and number of positive SLNs multiplied by tumor size. Indicating discrimination, AUC values for the training cohort and the validation cohort were 0.80 (95%CI 0.71-0.89) and 0.80 (95%CI 0.65-0.92) in the XGBoost model and 0.85 (95%CI 0.77-0.93) and 0.75 (95%CI 0.58-0.89) in the logistic regression model, respectively.Conclusions: This machine learning model was able to maintain its discrimination in the validation cohort better than the logistic regression model. This indicates advantages in employing modern artificial intelligence techniques into nomogram development. The nomogram could be used to help identify nodal stage N2-3 in early breast cancer and to select appropriate treatments for patients.

A pilot study of the implementation of WHO surgical checklist in Finland: improvements in activities and communication.

BACKGROUND: World Health Organisation (WHO) has introduced a surgical safety checklist that has reduced post-operative morbidity and mortality. Prior to national checklist implementation, we assessed its possible impact on the operating room (OR) process, safety-related issues and communication among surgical staff in a high-income country. METHODS: In four university and teaching hospitals, a structured questionnaire was delivered to OR personnel involved in consecutive operations over 4-6 weeks before and after the checklist implementation. The questionnaire resembled the WHO checklist and comprised multiple-choice questions relating to performance of safety checks and communication. Anaesthesiologists (A), surgeons (S) and circulating nurses (CN) answered the questions independently. The WHO checklist was modified for national needs. RESULTS: Questionnaires were returned from 1748 operations, 901 before and 847 after the checklist. Patient's identity was more often confirmed (A: 62.7% vs. 84.0%, S: 71.6% vs. 85.5%, CN: 81.6% vs. 94.2%, P < 0.001) and knowledge of names and roles among team members (A: 65.7% vs. 81.8%, S: 71.1% vs. 83.6%, CN: 87.7% vs. 93.2%, P < 0.01) improved with the checklist. Anaesthesiologists and surgeons discussed critical events pre-operatively (A: 22.0% vs. 42.6%, S: 34.7% vs. 46.2%, P < 0.001) more frequently after the checklist. In addition, fewer communication failures (43 vs. 17, P < 0.05) were reported with checklist. CONCLUSIONS: The checklist increased OR teams' awareness of patient-related issues, the procedure and expected risks. It also enhanced team communication and prevented communication failures. Our findings support use of the WHO checklist in various surgical fields.

Characteristics of clinically node negative breast cancer patients needing preoperative MRI.

BACKGROUND: International guidelines do not recommend magnetic resonance imaging (MRI) for all breast cancer patients at primary diagnostics. This study aimed to understand which patient or tumor characteristics are associated with the use of MRI. The role of MRI among other preoperative imaging methods in clinically node negative breast cancer was studied. MATERIAL AND METHODS: Patient and tumor characteristics were analyzed in association with the use of MRI by multivariable logistic regression analysis in 461 patients. Primary tumor size was compared between MRI, mammography (MGR), ultrasound (US) and histopathology by Spearman correlation. The delays in surgery and diagnosis were analyzed among patients with or without MRI, and axillary reoperations were evaluated. RESULTS: Age (p < 0.0001), primary operation method (p < 0.0001), tumor histology (p < 0.0001) and HER2 status (p = 0.0064) were associated with the use of MRI. Spearman correlations between tumor size in histopathology and the difference in tumor size between histopathology and imaging methods were 0.52 in MGR, 0.66 in US and 0.36 in MRI (p < 0.0001 for all). A seven-day delay in surgical treatment was observed among patients with MRI compared to patients without MRI (p < 0.0001). Axillary reoperation rates were similar in patients with or without MRI (p = 0.57). CONCLUSION: Patient selection through prearranged characterization is important in deciding on optimal candidates for preoperative MRI among breast cancer patients. MRI causes moderate delays in primary breast cancer surgery. Preoperative MRI is useful in the evaluation of tumor size but might be insufficient in detecting lymph node metastases.

Preoperative Magnetic Resonance Imaging in Patients With Stage I Invasive Ductal Breast Cancer: A Prospective Randomized Study.

BACKGROUND AND AIMS: Preoperative magnetic resonance imaging has become an important complementary imaging technique in patients with breast cancer, providing additional information for preoperative local staging. Magnetic resonance imaging is recommended selectively in lobular breast cancer and in patients with dense breast tissue in the case when mammography and ultrasound fail to fully evaluate the lesion, but the routine use of magnetic resonance imaging in all patients with invasive ductal carcinoma is controversial. The purpose of this randomized study was to investigate the diagnostic value of preoperative magnetic resonance imaging and its impact on short-term surgical outcome in newly diagnosed unifocal stage I invasive ductal carcinoma. MATERIAL AND METHODS: A total of 100 patients were randomized to either receive preoperative breast magnetic resonance imaging or to be scheduled directly to operation without magnetic resonance imaging on a 1:1 basis. There were 50 patients in both study arms. RESULTS: In 14 patients (28%), breast magnetic resonance imaging detected an additional finding and seven of them were found to be malignant. Six additional cancer foci were found in the ipsilateral breast and one in the contralateral breast. Magnetic resonance imaging findings caused a change in planned surgical management in 10 patients (20%). Mastectomy was performed in six patients (12%) in the magnetic resonance imaging group and in two patients (4%) in the control group ( p = 0.140). The breast reoperation rate was 14% in the magnetic resonance imaging group and 24% in the control group ( p = 0.202). The mean interval between referral and first surgical procedure was 34 days in the magnetic resonance imaging group and 21 days in the control group ( p < 0.001). CONCLUSION: Preoperative magnetic resonance imaging may be beneficial for some patients with early-stage invasive ductal carcinoma, but its routine use is not recommended without specific indications.

Local inflammatory response in choriodecidua induced by Ureaplasma urealyticum.

Ureaplasma urealyticum is the bacterial species most often connected with preterm birth, although it often colonises the amniotic fluid without any adverse effects. The induction of preterm labour seems to depend on whether the bacteria produce an inflammatory reaction. In vitro stimulation of choriodecidual tissue with high amounts of U. urealyticum or with lipopolysaccharide induced a qualitatively similar inflammatory response detected by the production of tumour necrosis factor alpha, followed by secretion of anti-inflammatory cytokine interleukin-10 and of prostaglandin E2. Lower quantities of bacteria failed to induce any response.

Comparative bioavailability of two commercial preparations of carbamazepine tablets.

A comparative bioavailability study was performed using two commercially available types of carbamazepine tablets, by statistical analysis of serum levels and other bioavailability parameters. After single oral dose the extent of absorption from the two preparations was similar, although a statistically significant difference in absorption rates was observed. In a multiple-dose study no significant difference in serum carbamazepine levels was found, so the tablets could be considered as bioequivalent drug products.