Retrospective Validation Cohort Research Articles

Incidence and Predictors of Pericardial Effusion After Chemoradiation Therapy for Locally Advanced Non-Small Cell Lung Cancer

Findings from Radiation Therapy Oncology Group (RTOG) 0617 suggested that collateral radiation to the heart may contribute to early death in patients receiving chemoradiation therapy for non-small cell lung cancer (NSCLC); however, reports of cardiac toxicity after thoracic radiation therapy (RT) remain limited. Because pericardial disease is the most common cardiac complication of thoracic RT, we investigated the incidence of and risk factors for pericardial effusion (PCE) in patients enrolled in a phase 2 prospective randomized study of intensity modulated RT versus proton therapy for locally advanced NSCLC. From July 2009 through April 2014, 201 patients were prospectively treated with proton beam therapy or intensity modulated RT to 60 to 74Gy with concurrent chemotherapy. The primary endpoint (grade ≥2 PCE) was diagnosed on review of follow-up images. Clinical characteristics and cardiac dose-volume parameters associated with PCE were identified via Cox proportional hazards modeling and recursive partitioning analysis of null Martingale residuals. Reproducibility was evaluated in a separate retrospective cohort of 301 patients. The cumulative incidence rates of PCE among patients in the trial were 31.4% at 1year and 45.4% at 2years, with a median time to PCE of 8.9months. Several cardiac dose-volume parameters (eg, V20 [volume receiving ≥20Gy] to V65 [volume receiving ≥65Gy]) predicted PCE, but heart volume receiving ≥35Gy (HV35) was the most strongly associated, with a cutoff volume of 10%.On multivariate analysis, HV35>10% independently predicted PCE (hazard ratio [HR], 2.14; P=.002), a finding that maintained reproducibility in the retrospective validation cohort. Other factors associated with PCE included receipt of adjuvant chemotherapy (HR, 2.82; P<.001) and prior cardiac disease (HR, 1.68; P=.020). PCE was common after RT for NSCLC, occurring in nearly half of patients even after moderate radiation doses to the heart. Adjuvant chemotherapy may increase the risk of PCE, and HV35>10% may identify patients at risk of development of this cardiac toxicity.

A nomogram based on serum bilirubin and albumin levels predicts survival in gastric cancer patients.

Decreases in serum bilirubin and albumin levels are associated with poorer prognoses in some types of cancer. Here, we examined the predictive value of serum bilirubin and albumin levels in 778 gastric cancer patients from a single hospital in China who were divided among prospective training and retrospective validation cohorts. X-tile software was used to identify optimal cutoff values for separating training cohort patients into higher and lower overall survival (OS) groups, based on total bilirubin (TBIL) and albumin levels. In univariate analysis, tumor grade and TNM stage were associated with OS. After adjusting for tumor grade and TNM stage, TBIL and albumin levels were still clearly associated with OS. These results were confirmed in the 299 patients in the validation cohort. A nomogram based on TBIL and albumin levels was more accurate than the TNM staging system for predicting prognosis in both cohorts. These results suggest that serum TBIL and albumin levels are independent predictors of OS in gastric cancer patients, and that an index that combines TBIL and albumin levels with the TNM staging system might have more predictive value than any of these measures alone.

A simple score for early risk stratification in acute heart failure

IntroductionThe use of many acute heart failure (AHF) risk scores is cumbersome. We therefore developed a simple AHF risk score (AHFRS) for early risk stratification. MethodsThe study consisted of a prospective derivation cohort (PDC; N=104; age, 77[21] years; LVEF (%), 35[29]) and a retrospective validation cohort (RVC; N=141; age, 76[15] years; LVEF (%), 35[25]). Clinical, echocardiography and laboratory assessment was performed at admission. The study end-point was death from any cause or HF-rehospitalization at 1year. ResultsIn the PDC 46 (44.2%) patients experienced the end-point. Independent prognostic factors of outcome were hypertension (HTN) history, myocardial infarction (MI) history, and admission red cell distribution width (RDW). Multivariate logistic regression indicated 8-, 4-, and 3-times higher odds ratio for development of study end-point in patients without a HTN history, with MI history, and RDW≥15% (median) respectively. Thus in AHFRS, 2 points were assigned for absence of HTN history, 1 point for presence of MI history, and 1 point for RDW values ≥15% (0 best possible, whereas 4 worst possible score). The AHFRS identified patients who developed the end-point in the PDC with an area under the ROC curve (AUC) of 0.80 [95% C.I.: (0.71, 0.87)] denoting a high discriminative ability. These findings were confirmed in the RVC, in which the endpoint occurred in 52 (36.9%) patients and the AUC for the AHFRS was 0.82 [95% C.I.: (0.73, 0.89)]. ConclusionsAHFRS is easily obtained at admission and accurately risk stratifies AHF patients.

LA Strain for Categorization of LV Diastolic Dysfunction

ObjectivesThis study sought to observe the relationship between left atrial (LA) strain and left ventricular diastolic function and determine whether LA strain could be used to detect diastolic dysfunction (DD) and classify its degree when present. BackgroundThe assessment of diastolic function is complex and multiparametric because most conventional parameters do not follow the progression of DD. Strain imaging is an emerging index of LA function, with recent data demonstrating that LA strain is diminished in diastolic heart failure. However, LA strain is not part of the standard assessment of diastolic function. We hypothesized that LA strain decreases with worsening DD in a stepwise fashion and could thus be useful in evaluating DD. MethodsWe performed a retrospective derivation and validation cohort study to derive and test LA strain thresholds for DD grades (0 to 3) in patients with preserved left ventricular ejection fraction (N = 229). Two-dimensional speckle tracking was used to measure peak LA strain, which was applied as a single parameter to classify DD. American Society of Echocardiography guidelines were used as the reference standard. ResultsIn the derivation cohort (n = 90), peak LA strain was significantly different between DD groups, with gradual decreases seen with worsening DD. Receiver-operating characteristic analysis resulted in 3 distinct LA strain thresholds for categorization of DD grades, with good to excellent diagnostic utility (area under the curve: 0.86 to 0.91). In an independent validation group (n = 139) with a spectrum of diastolic function, 11 patients (8%) had indeterminate DD grades using standard criteria, whereas LA strain was measured in all patients and its cutoffs resulted in diagnostic accuracy up to 95%. ConclusionsLA strain measurements are feasible and allow accurate categorization of DD, because unlike the traditional parameters, it changes progressively with severity of DD. LA strain may become a useful tool for diastolic assessment in future clinical practice.

A new prognostic histopathologic classification of nasopharyngeal carcinoma.

BackgroundThe current World Health Organization (WHO) classification of nasopharyngeal carcinoma (NPC) conveys little prognostic information. This study aimed to propose an NPC histopathologic classification that can potentially be used to predict prognosis and treatment response.MethodsWe initially developed a histopathologic classification based on the morphologic traits and cell differentiation of tumors of 2716 NPC patients who were identified at Sun Yat-sen University Cancer Center (SYSUCC) (training cohort). Then, the proposed classification was applied to 1702 patients (retrospective validation cohort) from hospitals outside SYSUCC and 1613 patients (prospective validation cohort) from SYSUCC. The efficacy of radiochemotherapy and radiotherapy modalities was compared between the proposed subtypes. We used Cox proportional hazards models to estimate hazard ratios (HRs) with 95% confidence intervals (CI) for overall survival (OS).ResultsThe 5-year OS rates for all NPC patients who were diagnosed with epithelial carcinoma (EC; 3708 patients), mixed sarcomatoid-epithelial carcinoma (MSEC; 1247 patients), sarcomatoid carcinoma (SC; 823 patients), and squamous cell carcinoma (SCC; 253 patients) were 79.4%, 70.5%, 59.6%, and 42.6%, respectively (P < 0.001). In multivariate models, patients with MSEC had a shorter OS than patients with EC (HR = 1.44, 95% CI = 1.27–1.62), SC (HR = 2.00, 95% CI = 1.76–2.28), or SCC (HR = 4.23, 95% CI = 3.34–5.38). Radiochemotherapy significantly improved survival compared with radiotherapy alone for patients with EC (HR = 0.67, 95% CI = 0.56–0.80), MSEC (HR = 0.58, 95% CI = 0.49–0.75), and possibly for those with SCC (HR = 0.63; 95% CI = 0.40–0.98), but not for patients with SC (HR = 0.97, 95% CI = 0.74–1.28).ConclusionsThe proposed classification offers more information for the prediction of NPC prognosis compared with the WHO classification and might be a valuable tool to guide treatment decisions for subtypes that are associated with a poor prognosis.

First-Line Chemotherapy for Metastatic Esophageal Squamous Cell Carcinoma: Clinico-Biological Predictors of Disease Control

Objective: This study aimed to identify predictors of tumor control (TC) in metastatic esophageal squamous cell carcinoma patients receiving first-line chemotherapy. Methods: A development cohort of 68 patients from a prospective multicenter trial (NCT01248299) was used to identify predictors of TC at first radiological tumor assessment and to generate a predictive score for TC. That score was applied in an independent retrospective single-center validation cohort of 60 consecutive patients. Results: Multivariate analysis identified three predictors of TC: body mass index ≥18.5 (OR 4.5, 95% CI 0.91-22.5), absence of bone metastasis (OR 4.6, 95% CI 0.91-23.2) and albumin ≥35 g/l (OR 3.5, 95% CI 1.0-12.1). Based on the presence or absence of these three independent prognosticators, we built a predictive model using a score from 0 to 3. In the development cohort, the TC rates were 14.3 and 78.0% and in the validation cohort 12.5 and 44.2%, for scores of 0-1 and 2-3, respectively. With negative predictive values of 85 and 88% in the development and validation cohorts, respectively, we were able to identify patients with a very low probability of TC. Conclusion: We have developed and validated a score that can be easily determined at the bedside to predict TC in metastatic esophageal squamous cell carcinoma patients.

A scoring system for the prediction of choledocholithiasis: a prospective cohort study.

BACKGROUNDTechniques for diagnosing choledocholithiasis pose significant morbidity and mortality risks.OBJECTIVESWe aimed to develop and validate a clinical scoring system for predicting choledocholithiasis.DESIGNData from a prospectively maintained database of all patients with gallstones.SETTINGPatients were admitted to the general surgery department of a military hospital.PATIENTS AND METHODSWe enrolled consecutive patients with symptomatic gallstones, biliary pancreatitis, obstructive jaundice, or cholangitis, who subsequently underwent biochemical testing and ultrasonography. A predictive model was developed from a scoring system using their imaging and laboratory data. Endoscopic retrograde cholangiopancreatography (ERCP) or intraoperative cholangiography were used for confirmatory diagnoses. The predictive efficacy of the scoring system was validated using a retrospective cohort of 272 patients.MAIN OUTCOME MEASURESPredictive accuracy of the scoring system.RESULTSWe enrolled 155 patients in the development group. The common bile duct diameter, alkaline phosphatase of ≥200 IU, elevated bilirubin levels, alanine transaminase of ≥220 IU, and male age of ≥50 years were significantly associated with choledocholithiasis and were included in the scoring system. Ninety-six patients (35%) had scores of ≥8 (high risk), 86 patients (32%) had scores of 4–7 (intermediate risk), and 27 patients (10%) had scores of 1–3 (low risk). In the validation cohort, the positive predictive value for a score of ≥8 was 91.7%, and the scoring system had an area under the curve of 0.896.CONCLUSIONScores of ≥8 were strongly correlated with choledocholithiasis in the developmental and validation groups, which indicates that our scoring system may be useful for predicting the need for therapeutic ERCP. However, prospective validation in a large multicenter cohort is needed to fully understand the benefits of the system.LIMITATIONSThe retrospective validation cohort might have introduced selection and observational biases. The study may have been underpowered because of the sample size of the developmental cohort. The delay between admission and the time of ERCP theoretically may have increased the number of negative ERCP results, but our false negative rate for ERCP was consistent with the previously reported rates.

Loss of function of Notch1 identifies a poor prognosis group of early stage hepatocellular carcinoma following hepatectomy.

Notch1 has previously been implicated in the carcinogenesis of hepatocellular carcinoma (HCC). The present study aimed to investigate the prognostic value of Notch1 in early stage HCC patients after hepatectomy. The differential expression of Notch1 in paired tumor and non-tumorous tissue was evaluated by RT-PCR, western blotting and immunohistochemistry. The correlation between Notch1 expression and the surgical outcome of patients at BCLC stage 0/A and its ≤5 cm subgroup was retrospectively investigated in 206 patients from the Eastern Hepatobiliary Surgery Hospital (training cohort), and prospectively validated in 185 patients from the same center and retrospectively verified in 129 patients from the Fujian Medical University (validation cohort 1 and 2, respectively). Compared with paired non-tumorous tissues, loss of Notch1 was observed in tumor tissue. Patients with normal Notch1 had better prognosis than those with loss of Notch1 in the training cohort and ≤5 cm subgroup (time to recurrence: 38.5±6.1 vs. 16.0±3.2 months, P<0.001 and 53.0±6.1 vs. 21.7±3.5 months, P=0.004; 1-, 3-, 5-year survival rates: 91, 64 and 49% vs. 73, 31 and 22%, P<0.001 and 93, 71, 57% vs. 76, 39, 24%, P<0.001). Notch1 expression was an independent factor for recurrence and survival (hazard ratio: 1.901, 2.154; 2.038 and 2.337). Moreover, Notch1 status affected early tumor recurrence, as the 2-year recurrence rate was 61.2 vs. 26.9% (P<0.001) and 51.2 vs. 21.3% (P=0.002) in tumors with reduced or increased Notch1 expression in this cohort and subgroup. These results were fully confirmed by the study in our prospective and retrospective validation cohorts. The status of Notch1 is useful for predicting the prognosis of patients with early stage HCC undergoing hepatectomy.

Derivation and preliminary validation of a risk score to predict 30-day ED revisits for sickle cell pain

BackgroundEmergency department (ED) revisits and 30-day readmissions have been proposed as markers for quality of ED care for sickle cell disease (SCD). ObjectiveTo create a scoring system that quantifies the risk of 30-day revisit after ED discharge for SCD vaso-occlusive pain MethodsThis was a dual-center retrospective derivation and validation cohort study. The derivation was performed at an academic, tertiary care center and the validation at an urban community hospital. The primary outcome was revisit to the ED within 30 days after an ED discharge for SCD pain. Recursive partitioning was used to derive a scoring system to predict 30-day revisits. ResultsOf a total of 1456 ED visits for SCD pain, there were 680 ED discharges (admission rate of 53%) in 193 unique individuals included in the derivation cohort. There were 240 (35.3%) 30-day revisits. Of a total of 126 ED visits for SCD, there were 79 ED discharges in 41 unique individuals in the validation cohort. The final risk score included 4 variables: (1) age, (2) insurance status, (3) triage pain score, and (4) amount of opioids administered during the ED visit. Possible scores range from 0 to 6. The areas under the receiver operating characteristic curves were 0.746 (95% confidence interval, 0.71-0.78—derivation cohort) and 0.753 (95% confidence interval, 0.65-0.86—validation cohort). A cutoff of 4 or greater identified 60% of 30-day ED revisits in the derivation cohort and 80% of revisits in the validation cohort. ConclusionsA risk score can identify ED visits for SCD pain with high risk of 30-day revisit.

Predicting Early Death Among Elderly Dialysis Patients: Development and Validation of a Risk Score to Assist Shared Decision Making for Dialysis Initiation

A shared decision-making tool could help elderly patients with advanced chronic kidney disease decide about initiating dialysis therapy. Because mortality may be high in the first few months after initiating dialysis therapy, incorporating early mortality predictors in such a tool would be important for an informed decision. Our objective is to derive and validate a predictive risk score for early mortality after initiating dialysis therapy. Retrospective observational cohort, with development and validation cohorts. US Renal Data System and claims data from the Centers for Medicare & Medicaid Services for 69,441 (aged≥67 years) patients with end-stage renal disease with a previous 2-year Medicare history who initiated dialysis therapy from January 1, 2009, to December 31,2010. Demographics, predialysis care, laboratory data, functional limitations, and medical history. All-cause mortality in the first 3 and 6 months. Predicted mortality by logistic regression. The simple risk score (total score, 0-9) included age (0-3 points), low albumin level, assistance with daily living, nursing home residence, cancer, heart failure, and hospitalization (1 point each), and showed area under the receiver operating characteristic curve (AUROC)=0.69 in the validation sample. A comprehensive risk score with additional predictors was also developed (with AUROC=0.72, high concordance between predicted vs observed risk). Mortality probabilities were estimated from these models, with the median score of 3 indicating 12% risk in 3 months and 20% in 6 months, and the highest scores (≥8) indicating 39% risk in 3 months and 55% in 6 months. Patients who did not choose dialysis therapy and did not have a 2-year Medicare history were excluded. Routinely available information can be used by patients with chronic kidney disease, families, and their nephrologists to estimate the risk of early mortality after dialysis therapy initiation, which may facilitate informed decision making regarding treatment options.

Evaluation of a hospital admission prediction model adding coded chief complaint data using neural network methodology

Our objective was to apply neural network methodology to determine whether adding coded chief complaint (CCC) data to triage information would result in an improved hospital admission prediction model than one without CCC data. We carried out a retrospective derivation and validation cohort study of all adult emergency department visits to a single center. We downloaded triage, chief complaint, and admission/discharge data on each included visit. Using a CCC algorithm and the Levenberg-Marquardt back-propagation learning method, we derived hospital admission prediction models without and with CCC data and applied these to the validation cohort, reporting the prediction models' characteristics. A total of 74 056 emergency department visits were included in the derivation cohort, 85 144 in the validation cohort with 213 CCC categories. The sensitivity/specificity of the derivation cohort models without and with CCC data were 64.0% [95% confidence interval (CI): 63.7-64.3], 87.7% (95% CI: 87.4-88.0), 59.8% (95% CI: 59.5-60.3%), and 91.7% (95% CI: 91.4-92.0) respectively. The sensitivity/specificity of the derived models without and with CCC data applied to the validation cohort were 60.7% (95% CI: 60.4-61.0), 87.7% (95% CI: 87.4-88.0), 59.8% (95% CI: 59.5-60.3), and 90.6% (95% CI: 90.3-90.9) respectively. The area under the curve in the validation cohort for the derived models without and with CCC data were 0.840 (95% CI: 0.838-0.842) and 0.860 (95% CI: 0.858-0.862). Net reclassification index (0.156; 95% CI: 0.148-0.163) and integrated discrimination improvement (0.060; 95% CI: 0.058-0.061) in the CCC model were significant. Neural net methodology application resulted in the derivation and validation of a modestly stronger hospital admission prediction model after the addition of CCC data.

An international longitudinal registry of patients with atrial fibrillation at risk of stroke (GARFIELD): the UK protocol

BackgroundAtrial fibrillation (AF) is an independent risk factor for stroke and a significant predictor of mortality. Evidence-based guidelines for stroke prevention in AF recommend antithrombotic therapy corresponding to the risk of stroke. In practice, many patients with AF do not receive the appropriate antithrombotic therapy and are left either unprotected or inadequately protected against stroke. The purpose of the Global Anticoagulant Registry in the FIELD (GARFIELD) is to determine the real-life management and outcomes of patients newly diagnosed with non-valvular AF.Methods/designGARFIELD is an observational, international registry of newly diagnosed AF patients with at least one additional investigator-defined risk factor for stroke. The aim is to enrol 55,000 patients at more than 1000 centres in 50 countries worldwide. Enrolment will take place in five independent, sequential, prospective cohorts; the first cohort includes a retrospective validation cohort. Each cohort will be followed up for 2 years. The UK stands to be a significant contributor to GARFIELD, aiming to enrol 4,582 patients, and reflecting the care environment in which patients with AF are managed. The UK protocol will also focus on better understanding the validity of the two main stroke risk scores (CHADS2 and CHA2DS2VASC) and the HAS-BLED bleeding risk score, in the context of a diverse patient population.DiscussionThe GARFIELD registry will describe how therapeutic strategies, patient care, and clinical outcomes evolve over time. This study will provide UK-specific comprehensive data that will allow a range of evaluations both at a national level and in relation to global data and contribute to a better understanding of AF management in the UK.Trial registrationClinicalTrial.gov: NCT01090362

Risk score for identifying adults with CSF pleocytosis and negative CSF Gram stain at low risk for an urgent treatable cause

We aimed to derive and validate a risk score that identifies adults with cerebrospinal fluid (CSF) pleocytosis and a negative CSF Gram stain at low risk for an urgent treatable cause. Patients with CSF pleocytosis and a negative CSF Gram stain were stratified into a prospective derivation (n = 193) and a retrospective validation (n = 567) cohort. Clinically related baseline characteristics were grouped into three composite variables, each independently associated with a set of predefined urgent treatable causes. We subsequently derived a risk score classifying patients into low (0 composite variables present) or high (≥ 1 composite variables present) risk for an urgent treatable cause. The sensitivity of the risk score was determined in the validation cohort and in a prospective case series of 214 adults with CSF-culture proven bacterial meningitis, CSF pleocytosis and a negative Gram stain. A total of 41 of 193 patients (21%) in the derivation cohort and 71 of 567 (13%) in the validation cohort had an urgent treatable cause. Sensitivity of the dichotomized risk score to detect an urgent treatable cause was 100.0% (95% CI 93.9-100.0%) in the validation cohort and 100.0% (95% CI 97.8-100.0%) in bacterial meningitis patients. The risk score can be used to identify adults with CSF pleocytosis and a negative CSF Gram stain at low risk for an urgent treatable cause.

Determinación del ancho de distribución eritrocitaria. Utilidad en la insuficiencia cardiaca

The use of many acute heart failure (AHF) risk scores is cumbersome. We therefore developed a simple AHF risk score (AHFRS) for early risk stratification.The study consisted of a prospective derivation cohort (PDC; N = 104; age, 77[21] years; LVEF (%), 35[29]) and a retrospective validation cohort (RVC; N = 141; age, 76[15] years; LVEF (%), 35[25]). Clinical, echocardiography and laboratory assessment was performed at admission. The study end-point was death from any cause or HF-rehospitalization at 1 year.In the PDC 46 (44.2%) patients experienced the end-point. Independent prognostic factors of outcome were hypertension (HTN) history, myocardial infarction (MI) history, and admission red cell distribution width (RDW). Multivariate logistic regression indicated 8-, 4-, and 3-times higher odds ratio for development of study end-point in patients without a HTN history, with MI history, and RDW ≥ 15% (median) respectively. Thus in AHFRS, 2 points were assigned for absence of HTN history, 1 point for presence of MI history, and 1 point for RDW values ≥ 15% (0 best possible, whereas 4 worst possible score). The AHFRS identified patients who developed the end-point in the PDC with an area under the ROC curve (AUC) of 0.80 [95% C.I.: (0.71, 0.87)] denoting a high discriminative ability. These findings were confirmed in the RVC, in which the endpoint occurred in 52 (36.9%) patients and the AUC for the AHFRS was 0.82 [95% C.I.: (0.73, 0.89)].AHFRS is easily obtained at admission and accurately risk stratifies AHF patients.

Pharmacogenetic warfarin dose refinements remain significantly influenced by genetic factors after one week of therapy.

By guiding initial warfarin dose, pharmacogenetic (PGx) algorithms may improve the safety of warfarin initiation. However, once international normalised ratio (INR) response is known, the contribution of PGx to dose refinements is uncertain. This study sought to develop and validate clinical and PGx dosing algorithms for warfarin dose refinement on days 6-11 after therapy initiation. An international sample of 2,022 patients at 13 medical centres on three continents provided clinical, INR, and genetic data at treatment days 6-11 to predict therapeutic warfarin dose. Independent derivation and retrospective validation samples were composed by randomly dividing the population (80%/20%). Prior warfarin doses were weighted by their expected effect on S-warfarin concentrations using an exponential-decay pharmacokinetic model. The INR divided by that "effective" dose constituted a treatment response index . Treatment response index, age, amiodarone, body surface area, warfarin indication, and target INR were associated with dose in the derivation sample. A clinical algorithm based on these factors was remarkably accurate: in the retrospective validation cohort its R(2) was 61.2% and median absolute error (MAE) was 5.0 mg/week. Accuracy and safety was confirmed in a prospective cohort (N=43). CYP2C9 variants and VKORC1-1639 G→A were significant dose predictors in both the derivation and validation samples. In the retrospective validation cohort, the PGx algorithm had: R(2)= 69.1% (p<0.05 vs. clinical algorithm), MAE= 4.7 mg/week. In conclusion, a pharmacogenetic warfarin dose-refinement algorithm based on clinical, INR, and genetic factors can explain at least 69.1% of therapeutic warfarin dose variability after about one week of therapy.

International longitudinal registry of patients with atrial fibrillation at risk of stroke: Global Anticoagulant Registry in the FIELD (GARFIELD)

Atrial fibrillation (AF) is associated with high rates of morbidity and mortality. Patients with AF carry a fivefold increased risk of stroke and the risk of death from AF-related stroke is doubled. Current management is often inadequate, leaving patients at risk for a potentially fatal or disabling event. The purpose of the GARFIELD registry is to evaluate the management and outcomes of patients with newly diagnosed non-valvular AF at risk for stroke. The GARFIELD registry is an observational, multicenter, prospective study of patients with newly diagnosed AF and one or more additional risk factors for stroke. The aim is to enroll 55,000 patients at >1,000 centers in 50 countries. Enrollment will take place in five independent, sequential, prospective cohorts. An additional retrospective validation cohort of 5,000 patients with established AF and at least one additional risk factor for stroke will be conducted in parallel with cohort one. The study started in December 2009, with a planned recruitment period of 4 years and a minimum of 2-year follow-up for each patient. The GARFIELD registry will provide valuable insights into the clinical management and related outcomes of AF patients throughout many regions of the world and across the spectrum of healthcare systems. By capturing data from unselected patients treated in everyday practice, the registry has the potential to identify best practices as well as deficiencies in available treatment options for specific patient populations and to describe how therapeutic strategies, patient care, and outcomes will evolve over time.

Pancreatic resection in the elderly.

Elderly patients undergoing pancreatic resection present unique challenges in postoperative care. Although mortality rates among elderly patients after pancreatectomy at high-volume centers is known to be low, the anticipated decline in functional status and nutritional parameters has received little attention. Functional decline is an unrecognized but critically important consequence of pancreatic resection in older patients. This study is a retrospective review, validation cohort, of older and younger patients undergoing major pancreatic resection. The setting is the state of California (database of all hospitals in the state) and The University of California, San Francisco (UCSF; a tertiary care referral center). The study population is a consecutive sample of older (greater than or equal to 75 years) and younger (16 to 74 years) patients from California (January 1990 to December 1996; n = 3,113) and UCSF (January 1993 to November 2000; n = 218), who underwent radical pancreaticoduodenectomy, distal pancreatectomy, or total pancreatectomy for neoplasia. The main outcomes measures were length of stay, complications, mortality, discharge disposition, supplemental nutrition requirement, and readmissions. Elderly patients had higher mortality rates than the young statewide (10% versus 7%, p = 0.006). Although the 3% mortality at UCSF was the same for both groups, older patients were more often admitted to the ICU (47% versus 20%, p = 0.003), treated for major cardiac events (13% versus 0.5%, p < 0.001), discharged with enteral tube feedings (48% versus 16%, p < 0.001), or malnourished on readmission (17% versus 2%, p < 0.005). Older patients were more frequently discharged to skilled nursing facilities (17% versus 1% at UCSF; 24% versus 7% in California; p < 0.001, both groups). Older patients are more likely than younger patients to require an ICU stay, suffer a cardiac complication, and experience compromised nutritional and functional status after major pancreatic resection.

Langerhans cell histiocytosis presenting in the neonatal period: a retrospective case series.

To describe the morphologic characteristics of skin lesions, extent of extracutaneous disease, and outcomes in patients with neonatal presentation of Langerhans cell histiocytosis (LCH), and to examine clinical predictors of disease prognosis. Retrospective validation cohort study. Maximum duration of follow-up was 10 years. A tertiary care children's hospital in Chicago, Ill. Nineteen children with cutaneous findings in the first 4 weeks of life and subsequently diagnosed with LCH based on compatible tissue histologic analysis, confirmed by electron microscopy and/or immunohistochemical analysis. Cutaneous lesion morphologic characteristics, extracutaneous manifestations, treatments, and outcomes were tabulated and compared. The most common initial skin lesion was erythematous, often crusted, vesiculopustules. Skin lesion morphologic traits did not correlate with extent of extracutaneous disease. One third of patients had disease limited to the skin and/or mucous membranes. All of these patients are alive and well, and 1 has developed diabetes insipidus. Twelve of the 19 patients had multisystem disease, and 2 died of disease. The results of a multiorgan workup performed at the time of diagnosis were predictive of which patients in this cohort manifested multisystem disease. The overall incidence of diabetes insipidus was 21%. Vesiculopustular lesions are common in congenital/neonatal LCH, but the morphologic characteristics of lesions are not helpful in predicting the extent of disease. A multiorgan evaluation at the time of diagnosis may be predictive of the probability of multisystem involvement with LCH.

Logistic regression model to predict outcome after in-hospital cardiac arrest: validation, accuracy, sensitivity and specificity

Objective: To develop and validate a logistic regression model to identify predictors of death before hospital discharge after in-hospital cardiac arrest. Design: Retrospective derivation and validation cohorts over two 1 year periods. Data from all in-hospital cardiac arrests in 1986–87 were used to derive a logistic regression model in which the estimated probability of death before hospital discharge was a function of patient and arrest descriptors, major underlying diagnosis, initial cardiac rhythm, and time of year. This model was validated in a separate data set from 1989–90 in the same hospital. Calculated for each case was 95% confidence limits (C.L.) about the estimated probability of death. In addition, accuracy, sensitivity, and specificity of estimated probability of death and lower 95% C.L. of the estimated probability of death in the derivation and validation data sets were calculated. Setting: 560-bed university teaching hospital. Patients: The derivation data set described 270 cardiac arrests in 197 inpatients. The validation data set described 158 cardiac arrests in 120 inpatients. Interventions: none. Measurements and results: Death before hospital discharge was the main outcome measure. Age, female gender, number of previous cardiac arrests, and electrical–mechanical dissociation were significant variables associated with a higher probability of death. Underlying coronary artery disease or valvular heart disease, ventricular tachycardia, and cardiac arrest during the period July–September were significant variables associated with a lower probability of death. Optimal sensitivity and specificity in the validation set were achieved at a cut-off probability of 0.85. Conclusions: Performance of this logistic regression model depends on the cut-off probability chosen to discriminate between predicted survival and predicted death and on whether the estimated probability or the lower 95% C.L. of the estimated probability is used. This model may inform the development of clinical practice guidelines for patients who are at risk of or who experience in-hospital cardiac arrest.

Study of the decomposition of TiCWC solid solutions: S Zhenhui, (Zhuzhou Cemented Carbide Industry Co, Hunan, China), Int J. Refractory Metals & Hard Materials, Vol 9, No 3, 1990, 154–156

Accurate prediction of tumour response to neoadjuvant chemoradiotherapy enables personalised perioperative therapy for locally advanced rectal cancer. We aimed to develop and validate an artificial intelligence radiopathomics integrated model to predict pathological complete response in patients with locally advanced rectal cancer using pretreatment MRI and haematoxylin and eosin (H&E)-stained biopsy slides.In this multicentre observational study, eligible participants who had undergone neoadjuvant chemoradiotherapy followed by radical surgery were recruited, with their pretreatment pelvic MRI (T2-weighted imaging, contrast-enhanced T1-weighted imaging, and diffusion-weighted imaging) and whole slide images of H&E-stained biopsy sections collected for annotation and feature extraction. The RAdioPathomics Integrated preDiction System (RAPIDS) was constructed by machine learning on the basis of three feature sets associated with pathological complete response: radiomics MRI features, pathomics nucleus features, and pathomics microenvironment features from a retrospective training cohort. The accuracy of RAPIDS for the prediction of pathological complete response in locally advanced rectal cancer was verified in two retrospective external validation cohorts and further validated in a multicentre, prospective observational study (ClinicalTrials.gov, NCT04271657). Model performances were evaluated using area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).Between Sept 25, 2009, and Nov 3, 2017, 303 patients were retrospectively recruited in the training cohort, 480 in validation cohort 1, and 150 in validation cohort 2; 100 eligible patients were enrolled in the prospective study between Jan 10 and June 10, 2020. RAPIDS had favourable accuracy for the prediction of pathological complete response in the training cohort (AUC 0·868 [95% CI 0·825–0·912]), and in validation cohort 1 (0·860 [0·828–0·892]) and validation cohort 2 (0·872 [0·810–0·934]). In the prospective validation study, RAPIDS had an AUC of 0·812 (95% CI 0·717–0·907), sensitivity of 0·888 (0·728–0·999), specificity of 0·740 (0·593–0·886), NPV of 0·929 (0·862–0·995), and PPV of 0·512 (0·313–0·710). RAPIDS also significantly outperformed single-modality prediction models (AUC 0·630 [0·507–0·754] for the pathomics microenvironment model, 0·716 [0·580–0·852] for the radiomics MRI model, and 0·733 [0·620–0·845] for the pathomics nucleus model; all p<0·0001).RAPIDS was able to predict pathological complete response to neoadjuvant chemoradiotherapy based on pretreatment radiopathomics images with high accuracy and robustness and could therefore provide a novel tool to assist in individualised management of locally advanced rectal cancer.National Natural Science Foundation of China; Youth Innovation Promotion Association of the Chinese Academy of Sciences.