Preventing Adverse Drug Reactions After Hospital Discharge (PADR-AD): Protocol for a randomised-controlled trial in older people

BackgroundAdverse drug reactions (ADRs) and adverse drug events (ADEs) in older people contribute to a significant proportion of hospital admissions and are common following discharge. Effective interventions are therefore required to combat the growing burden of preventable ADRs. The Prediction of Hospitalisation due to Adverse Drug Reactions in Elderly Community Dwelling Patients (PADR-EC) score is a validated risk score developed to assess the risk of ADRs in people aged 65 years and older and has the potential to be utilised as part of an intervention to reduce ADRs.ObjectivesThis trial was designed to investigate the effectiveness of an intervention to reduce ADR incidence in older people and to obtain further information about ADRs and ADEs in the 12–24 months following hospital discharge.MethodsThe study is an open-label randomised-controlled trial to be conducted at the Royal Hobart Hospital, a 500-bed public hospital in Tasmania, Australia. Community-dwelling patients aged 65 years and older with an unplanned overnight admission to a general medical ward will be recruited. Following admission, the PADR-EC ADR score will be calculated by a research pharmacist, with the risk communicated to clinicians and discussed with participants. Following discharge, nominated general practitioners and community pharmacists will receive the risk score and related medication management advice to guide their ongoing care of the patient. Follow-up with participants will occur at 3 and 12 and 18 and 24 months to identify ADRs and ADEs. The primary outcome is moderate-severe ADRs at 12 months post-discharge, and will be analysed using the cumulative incidence proportion, survival analysis and Poisson regression.SummaryIt is hypothesised that the trial will reduce ADRs and ADEs in the intervention population. The study will also provide valuable data on post-discharge ADRs and ADEs up to 24 months post-discharge.     

基于超声影像组学的乳腺癌预测模型的诊断性能研究

目的评估并比较基于不同机器学习算法建立的乳腺癌超声影像组学预测模型的诊断性能。 方法回顾性收集2017年1月至2019年4月就诊皖南医学院第一附属医院、有明确病理结果的乳腺肿块病例828例，以2018年8月31日为节点将其分为训练集（526例）和验证集（302例），提取肿块的超声影像组学特征并进行特征筛选，运用k最近邻（kNN）、逻辑回归（LR）、朴素贝叶斯（NB）、随机森林（RF）和支持向量机（SVM）5种机器学习算法分别建立预测模型，使用重复交叉验证方法做内部验证，计算比较各模型的敏感度、特异度、阳性预测值（PPV）和阴性预测值（NPV），并实施外部验证，绘制ROC曲线并比较ROC曲线下面积（AUC）以评价模型的鉴别诊断性能，绘制校准曲线评价模型校准度。 结果从提取的109个影像组学特征中筛选出19个特征建立了5种机器学习算法模型。在内部验证中，5种模型的敏感度、特异度、PPV、NPV比较，总体差异均有统计学意义（P均＜0.001）。LR模型的特异度、PPV、NPV中位数分别为0.769、0.816、0.778，3项指标均高于其他4种模型；敏感度中位数为0.824，高于kNN、RF和SVM模型。此外，SVM模型的特异度、PPV、NPV中位数分别为0.706、0.774、0.759，虽均低于LR模型，但均高于其他3种模型。在外部验证中，LR、SVM、RF、kNN和NB的AUC依次为0.890、0.832、0.821、0.746和0.703，其中LR与SVM的AUC差异有统计学意义（P=0.012）；此外，各模型在校准性能上表现并不一致，LR和SVM模型的校准曲线显示乳腺癌实际概率与预测概率之间的一致性较好。 结论以超声影像组学特征为基础，运用不同机器学习算法建立的乳腺癌超声预测模型，均表现出较高的诊断性能，其中LR模型表现最为突出；选择合适的机器学习算法有助于进一步提高预测模型的诊断性能，提供更加准确的量化预测结果。     

Isokinetic quadriceps symmetry helps in the decision to return to running after anterior cruciate ligament reconstruction

BackgroundAfter anterior cruciate ligament reconstruction (ACLR), the decision to allow a return to running is empirical, and the post-operative delay is the most-used criterion. The Quadriceps isokinetic-strength Limb Symmetry Index (Quadriceps LSI), with a cutoff of 60%, could be a useful criterion.ObjectiveTo determine the association between a Quadriceps LSI ≥ 60% and return to running after ACLR.MethodsOver a 10-year period, we retrospectively included 470 patients who underwent ACLR. Four months after ACLR, participants performed an isokinetic test; quadriceps concentric peak torque was used to calculate the Quadriceps LSI at 60?/s. With a Quadriceps LSI ≥ 60%, a return to running was suggested. At 6 months after ACLR, participants were clinically evaluated for a return to sport and post-operative middle-term complications. A multivariable predictive model was built to assess the efficiency diagnosis of this cutoff in order to consider cofounding factors. Quadriceps LSI cutoff ≥ 60% was assessed with sensitivity, specificity and the area under the receiver operating characteristic curve (AUC).ResultsAccording to our decision-making process with the 60% Quadriceps LSI cutoff at 60?/s, 285 patients were authorized to return to running at 4 months after ACLR and 185 were not, but 21% (n = 59) and 24% (n = 45), respectively, were not compliant with the recommendation. No iterative autograft rupture or meniscus pathology occurred at 6 months of follow-up. On multivariable logistic regression analysis, a return to running by using the 60% Quadriceps LSI cutoff was associated with undergoing the hamstring strand procedure (odds ratio 2.60, 95% confidence interval [CI] 1.75–3.84; P < 0.0001) and the absence of knee complications (1.18, 1.07–1.29; P = 0.001) at 4 months. The sensitivity and specificity of the 60% Quadriceps LSI cutoff were 83% and 70%, respectively. The AUC was 0.840 (95% CI 0.803–0.877).ConclusionsUsing the 60% cutoff of the isokinetic Quadriceps LSI at 4 months after ACLR could help in the decision to allow a return to running.     

Machine learning in clinical diagnosis,prognostication, and management of acute traumatic spinal cord injury (SCI): A systematic review

BackgroundMachine learning has been applied to improve diagnosis and prognostication of acute traumatic spinal cord injury. We investigate potential for clinical integration of machine learning in this patient population to navigate variability in injury and recovery.Materials and methodsWe performed a systematic review using PRISMA guidelines through PubMed database to identify studies that use machine learning algorithms for clinical application toward improvements in diagnosis, management, and predictive modeling.ResultsOf the 132 records identified, a total of 13 articles met inclusion criteria and were included in final analysis. Of the 13 articles, 5 focused on diagnostic accuracy and 8 were related to prognostication or management of traumatic spinal cord injury. Across studies, 1983 patients with spinal cord injury were evaluated with most classifying as ASIA C or D. Retrospective designs were used in 10 of 13 studies and 3 were prospective. Studies focused on MRI evaluation and segmentation for diagnostic accuracy and prognostication, investigation of mean arterial pressure in acute care and intraoperative settings, prediction of ambulatory and functional ability, chronic complication prevention, and psychological quality of life assessments. Decision tree, random forests (RF), support vector machines (SVM), hierarchical cluster tree analysis (HCTA), artificial neural networks (ANN), convolutional neural networks (CNN) machine learning subtypes were used.ConclusionsMachine learning represents a platform technology with clinical application in traumatic spinal cord injury diagnosis, prognostication, management, rehabilitation, and risk prevention of chronic complications and mental illness. SVM models showed improved accuracy when compared to other ML subtypes surveyed. Inherent variability across patients with SCI offers unique opportunity for ML and personalized medicine to drive desired outcomes and assess risks in this patient population.     

Inhibiting Pathways Predicted From a Steroid Hormone Gene Signature Yields Synergistic Antitumor Effects in NSCLC

IntroductionMounting evidence supports a role for estrogen signaling in NSCLC progression. We previously reported a seven-gene signature that predicts prognosis in estrogen receptor β positive (ERβ+) NSCLC. The signature defines a network comprised of ER and human EGFR-2/3 (HER2/HER3) signaling.MethodsWe tested the efficacy of combining the pan-HER inhibitor, dacomitinib, with the estrogen antagonist, fulvestrant, in ERβ+ NSCLC models with differing genotypes. We assessed the potency of this combination on xenograft growth and survival of host mice, and the ability to reverse the gene signature associated with poor outcome.ResultsSynergy was observed between dacomitinib and fulvestrant in three human ERβ+ NSCLC models: 201T (wild-type EGFR), A549 (KRAS mutant), and HCC827 (EGFR 19 deletion) with combination indices of 0.1-0.6. The combination, but not single agents, completely reversed the gene signature associated with poor prognosis in a mechanism that is largely mediated by activator protein 1 downregulation. In vivo, the combination also induced tumor regression and reversed the gene signature. In HCC827 xenografts treated with the combination, survival of mice was prolonged after therapy discontinuation, tumors that recurred were less aggressive, and two mechanisms of HER inhibitor resistance involving c-Met activation and PTEN loss were blocked.ConclusionsThe combination of an ER blocker and a pan-HER inhibitor provides synergistic efficacy in different models of ERβ+ NSCLC. Our data support the use of this combination clinically, considering its ability to induce potent antitumor effects and produce a gene signature that predicts better clinical outcomes.     

CBCT影像组学联合构建Nomogram模型预测食管癌放疗患者放射性肺炎

目的 通过对放疗疗程中不同时段CBCT图像的影像组学分析,寻找早期定量预测食管癌放疗放射性肺炎(RP)的参数,结合临床特征和肺剂量体积参数建立联合Nomogram模型并探讨这一模型对食管癌RP的预测价值。方法 回顾分析2017—2019年间临床资料、剂量学参数、CBCT图像资料完整的 96例胸中段食管鳞癌调强放疗患者资料,每例患者均分别获取放疗期间3个不同时段的肺CBCT图像。全组病例随机分成训练集(67例)和验证集(29例),以CBCT上双肺实质作为感兴趣区,运用3D-Slicer软件进行图像分割和特征提取,经LASSO-Logistics回归分析方法进行特征参数筛选并构建影像组学标签(Rad-score)。从3个不同时段建立的RP预测模型中选择最优模型联合经回归分析获得的最佳临床及剂量学参数,建立联合Nomogram模型,并进行受试者工作特征曲线分析,基于曲线下的面积(AUC)验证其诊断效能。结果 第一时段的影像组学预测模型优于其他两个时段,在训练集中的AUC值为0.700(95%CI为 0.568~0.832),敏感性和特异性分别为61.5%、75.0%;在验证集中的AUC值为0.765(95%CI为 0.588~0.941),敏感性和特异性分别为84.6%、64.7%。影像组学联合临床及剂量学构建的Nomogram模型在训练集中的AUC值为0.836(95%CI为 0.700~0.918),敏感性和特异性分别为96.0%、54.8%;在验证集中的AUC值为0.905(95%CI为 0.799~1.000),敏感性和特异性分别为92.9%、73.3%。联合Nomogram模型诊断效能最佳。结论 基于放疗早期肺CBCT影像组学特征构建的模型,对于食管癌RP具有一定的预测效能,Rad-score联合 肺V5Gy、肺 Dmean、肿瘤分期建立的Nomogram模型具有更好的预测准确性,可作为一种定量预测模型用于RP的预测。     

Development and validation of a novel scoring system to predict severe intraventricular hemorrhage in very low birth weight infants

ObjectiveWe sought to develop and validate a novel scoring system for the prediction of severe intraventricular hemorrhage (SIVH) in very low birth weight infants (VLBWI).MethodsThis retrospective cohort multicenter study included 615 VLBWI born between 24 and 32 weeks gestational age (GA). Multivariable logistic regression analyses were used to determine which factors evaluated within the first 5 days of life were associated with SIVH and the weights of these variables. The accuracy of the predictive scoring system was prospectively tested in the same units.ResultsThe final SIVH scoring system included the following variables: antenatal steroid therapy, GA, birth weight (BW), 1-min Apgar score, mechanical ventilation and hypotension. The SIVH scores used to divide the subjects into three tiers (low-risk (<5), moderate-risk (5–8) and high-risk (>8)) were developed based on these variables. Of infants with a score below 5, 1.2% (4/339) developed SIVH. Of those with a score above 8, 42.5% (17/40) developed SIVH. The scores were successfully verified in 99 VLBWI.ConclusionsThese findings suggest that among infants with a score >8, the incidence of SIVH is significantly higher. This scoring system can be used to predict the incidence of SIVH during the first 5 days after birth and may contribute to the early intervention in and prevention of SIVH.     

Development of the Depist’HTA® score to predict a rise in blood pressure during a self-testing session: PREDIC-HTA bakery survey

The use of an automatic blood pressure monitor with the help of a protocol given by an e-health application allows self-testing of blood pressure without the help of a health professional.ObjectiveConduct a screening campaign for high blood pressure in artisanal bakeries in France and develop a score to predict the observation of an increase in blood pressure during self-testing.MethodCross-sectional survey conducted in mainland France in 2018 using 1000 screening kits including a tablet, Predic’HTA® application and a connected electronic blood pressure monitor (BP Track, IHealth). Available for one week in 14,000 artisanal bakeries, in turn. Used in the workplace, on a voluntary basis, by bakery employees. Completion of a health questionnaire with an estimate of excessive salt consumption. Self-screening of 3 blood pressure measurements with centralization of data on a secure database for epidemiological analysis. The average of the last 2 measurements characterizes each subject.ResultsFull use of Predic’HTA® was obtained in 62.4% of users or 7502 subjects (44% under 35 years of age, 44% 35 to 54 years of age, 12% 55 years of age and over). Antihypertensive treatment was followed by 27% of those aged 55 and over and 1% of those under 35. Among untreated subjects, the prevalence of BP > 140/90 was 21.1%, increasing with age (13%, 27%, 33%) and higher among men (27.2%) than among women (15.4%). With the parameters age, sex, weight, IMC, family history of high blood pressure, excessive salt consumption, previous intake of an antihypertensive drug, a score is calculated to associate a probability of having an BP >120/80 with self-screening. The ROC curve shows an AUC at 0.723 with a PPV at 91% and a NPV at 41%. The depist’HTA® score is available at www.depisthta.netConclusionSelf-testing of blood pressure is possible with an electronic blood pressure monitor if a dedicated appliance is used at the same time. The development of the prediction score depist’HTA® for a blood pressure level > 120/80 would be useful to target subjects who need to self-test their blood pressure in the short term.     

Contribution of components of the quadriceps femoris muscle for producing external torque in patients with patellofemoral pain syndrome

IntroductionDifferent muscular activities of the quadriceps components for producing necessary torque may change in patients with patellofemoral pain syndrome (PFPS). The aim of the current study, therefore, was to assess the contribution of each component of the quadriceps femoris muscle for producing external torque in patients with PFPS.MethodTwelve females with PFPS (24.7 ± 2.3 years) and twelve healthy matched females (25.4 ± 2.4 years) performed three consecutive knee flexion and extension movements with maximum effort at 45°/s and 300°/s using a Biodex system 3 dynamometer. Simultaneously, electromyographic (EMG) activities of the vastus medialis oblique (VMO), RF (rectus femoris) and vastus lateralis (VL) muscles were recorded using a DataLog instrument. Standard multiple regressions were used to assess the ability of EMG activities of the VMO, RF and VL muscles to predict normalized quadriceps femoris isokinetic concentric and eccentric torques at 45°/s and 300°/s in the normal and patient groups.ResultsIn the normal group, the VL and the VMO were the good predictors of quadriceps concentric torque at 45°/s and 300°/s, respectively. The VL and the RF were the good predictors of quadriceps eccentric torque at 300°/s in the patient group. No other conditions showed a considerable prediction for quadriceps torque in the normal or patient group.ConclusionFemales with PFPS differ with normal females in terms of the contribution of each component of the quadriceps femoris for producing external torque. Training the VMO for concentric contraction at both high and low velocities should be included in the management of the patients with PFPS.     

Expectation affects neural repetition suppression in infancy

Recent work provides evidence that the infant brain is able to make top-down predictions, but this has been explored only in limited contexts and domains. We build upon this evidence of predictive processing in infants using a new paradigm to examine auditory repetition suppression (RS). RS is a well-documented neural phenomenon in which repeated presentations of the same stimulus result in reduced neural activation compared to non-repeating stimuli. Many theories explain RS using bottom-up mechanisms, but recent work has posited that top-down expectation and predictive coding may bias, or even explain, RS. Here, we investigate whether RS in the infant brain is similarly sensitive to top-down mechanisms. We use fNIRS to measure infants’ neural response in two experimental conditions, one in which variability in stimulus presentation is expected (occurs 75% of the time) and a control condition where variability and repetition are equally likely (50% of the time). We show that 6-month-old infants exhibit attenuated frontal lobe response to blocks of variable auditory stimuli during contexts when variability is expected as compared to the control condition. These findings suggest that young infants’ neural responses are modulated by predictions gained from experience and not simply by bottom-up mechanisms.