Intubation Decision Based on Illness Severity and Mortality in COVID-19: An International Study.

OBJECTIVES: To evaluate the impact of intubation timing, guided by severity criteria, on mortality in critically ill COVID-19 patients, amidst existing uncertainties regarding optimal intubation practices. DESIGN: Prospective, multicenter, observational study conducted from February 1, 2020, to November 1, 2022. SETTING: Ten academic institutions in the United States and Europe. PATIENTS: Adults (≥ 18 yr old) confirmed with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and hospitalized specifically for COVID-19, requiring intubation postadmission. Exclusion criteria included patients hospitalized for non-COVID-19 reasons despite a positive SARS-CoV-2 test. INTERVENTIONS: Early invasive mechanical ventilation (EIMV) was defined as intubation in patients with less severe organ dysfunction (Sequential Organ Failure Assessment [SOFA] < 7 or Pa o2 /F io2 ratio > 250), whereas late invasive mechanical ventilation (LIMV) was defined as intubation in patients with SOFA greater than or equal to 7 and Pa o2 /F io2 ratio less than or equal to 250. MEASUREMENTS AND MAIN RESULTS: The primary outcome was mortality within 30 days of hospital admission. Among 4464 patients, 854 (19.1%) required mechanical ventilation (mean age 60 yr, 61.7% male, 19.3% Black). Of those, 621 (72.7%) were categorized in the EIMV group and 233 (27.3%) in the LIMV group. Death within 30 days after admission occurred in 278 patients (42.2%) in the EIMV and 88 patients (46.6%) in the LIMV group ( p = 0.28). An inverse probability-of-treatment weighting analysis revealed a statistically significant association with mortality, with patients in the EIMV group being 32% less likely to die either within 30 days of admission (adjusted hazard ratio [HR] 0.68; 95% CI, 0.52-0.90; p = 0.008) or within 30 days after intubation irrespective of its timing from admission (adjusted HR 0.70; 95% CI, 0.51-0.90; p = 0.006). CONCLUSIONS: In severe COVID-19 cases, an early intubation strategy, guided by specific severity criteria, is associated with a reduced risk of death. These findings underscore the importance of timely intervention based on objective severity assessments.

Time-varying associations of patient and tumor characteristics with cancer survival: an analysis of SEER data across 14 cancer sites, 2004-2017.

PURPOSE: Surveillance, Epidemiology, and End Results (SEER) cancer registries provides information about survival duration and cause of death for cancer patients. Baseline demographic and tumor characteristics such as age, sex, race, year of diagnosis, and tumor stage can inform the expected survival time of patients, but their associations with survival may not be constant over the post-diagnosis period. METHODS: Using SEER data, we examined if there were time-varying associations of patient and tumor characteristics on survival, and we assessed how these relationships differed across 14 cancer sites. Standard Cox proportional hazards models were extended to allow for time-varying associations and incorporated into a competing-risks framework, separately modeling cancer-specific and other-cause deaths. For each cancer site and for each of the five factors, we estimated the relative hazard ratio and absolute hazard over time in the presence of competing risks. RESULTS: Our comprehensive consideration of patient and tumor characteristics when estimating time-varying hazards showed that the associations of age, tumor stage at diagnosis, and race/ethnicity with risk of death (cancer-specific and other-cause) change over time for many cancers; characteristics of sex and year of diagnosis exhibit some time-varying patterns as well. Stage at diagnosis had the largest associations with survival. CONCLUSION: These findings suggest that proportional hazards assumptions are often violated when examining patient characteristics on cancer survival post-diagnosis. We discuss several interesting results where the relative hazards are time-varying and suggest possible interpretations. Based on the time-varying associations of several important covariates on survival after cancer diagnosis using a pan-cancer approach, the likelihood of the proportional hazards assumption being met or corresponding interpretation should be considered in survival analyses, as flawed inference may have implications for cancer care and policy.

SuPAR, biomarkers of inflammation, and severe outcomes in patients hospitalized for COVID-19: The International Study of Inflammation in COVID-19.

Severe coronavirus disease 2019 (COVID-19) is a hyperinflammatory syndrome. The biomarkers of inflammation best suited to triage patients with COVID-19 are unknown. We conducted a prospective multicenter observational study of adult patients hospitalized specifically for COVID-19 from February 1, 2020 to October 19, 2022. Biomarkers measured included soluble urokinase plasminogen activator receptor (suPAR), C-reactive protein, interleukin-6, procalcitonin, ferritin, and D-dimer. In-hospital outcomes examined include death and the need for mechanical ventilation. Patients admitted in the United States (US, n = 1962) were used to compute area under the curves (AUCs) and identify biomarker cutoffs. The combined European cohorts (n = 1137) were used to validate the biomarker cutoffs. In the US cohort, 356 patients met the composite outcome of death (n = 197) or need for mechanical ventilation (n = 290). SuPAR was the most important predictor of the composite outcome and had the highest AUC (0.712) followed by CRP (0.642), ferritin (0.619), IL-6 (0.614), D-dimer (0.606), and lastly procalcitonin (0.596). Inclusion of other biomarkers did not improve discrimination. A suPAR cutoff of 4.0 ng/mL demonstrated a sensitivity of 95.4% (95% CI: 92.4%-98.0%) and negative predictive value (NPV) of 92.5% (95% CI: 87.5%-96.9%) for the composite outcome. Patients with suPAR < 4.0 ng/mL comprised 10.6% of the cohort and had a 0.8% probability of the composite outcome. Applying this cutoff to the validation cohort yielded a sensitivity of 93.8% (90.4%-96.7%) and NPV of 95.5% (93.1%-97.8%) for the composite outcome. Among commonly measured biomarkers, suPAR offered stronger discriminatory ability and may be useful in triaging low-risk patients with COVID-19.

Contemporary Care and Outcomes of Critically-ill Children With Clinically Diagnosed Myocarditis.

PURPOSE: To describe contemporary management and outcomes in children with myocarditis who are admitted to a cardiac intensive care unit (CICU) and to identify the characteristics associated with mortality. METHODS: All patients in the Pediatric Cardiac Critical Care Consortium (PC4) registry between August 2014 and June 2021 who were diagnosed with myocarditis were included. Univariable analyses and multivariable logistic regression evaluated the factors associated with in-hospital mortality. RESULTS: There were 847 CICU admissions for myocarditis in 51 centers. The median age was 12 years (IQR 2.7-16). In-hospital mortality occurred in 53 patients (6.3%), and 60 (7.1%) had cardiac arrest during admission. Mechanical ventilation was required in 339 patients (40%), and mechanical circulatory support (MCS) in 177 (21%); extracorporeal membrane oxygenation (ECMO)-only in 142 (16.7%), ECMO-to-ventricular assist device (VAD) in 20 (2.4%), extracorporeal cardiac resuscitation in 43 (5%), and VAD-only in 15 (1.8%) patients. MCS was associated with in-hospital mortality; 20.3% receiving MCS died compared to 2.5% without MCS (P < 0.001). Mortality rates were similar in ECMO-only, ECMO-to-VAD and VAD-only groups. The median time from CICU admission to ECMO was 2.0 hours (IQR 0-9.4) and to VAD, it was 9.9 days (IQR 6.3-16.8). Time to MCS was not associated with mortality. In multivariable modeling of patients' characteristics, smaller body surface area (BSA) and low eGFR were independently associated with mortality, and after including critical therapies, mechanical ventilation and ECMO were independent predictors of mortality. CONCLUSION: This contemporary cohort of children admitted to CICUs with myocarditis commonly received high-resource therapies; however, most patients survived to hospital discharge and rarely received VAD. Smaller patient size, acute kidney injury and receipt of mechanical ventilation or ECMO were independently associated with mortality.

Mechanical Ventilation and Outcomes of Children Who Undergo Ventricular Assist Device Placement: 2014-2020 Linked Analysis From the Advanced Cardiac Therapies Improving Outcomes Network and Pediatric Cardiac Critical Care Consortium Registries.

OBJECTIVES: Placement of a ventricular assist device (VAD) improves outcomes in children with advanced heart failure, but adverse events remain important consequences. Preoperative mechanical ventilation (MV) increases mortality, but it is unknown what impact prolonged postoperative MV has. DESIGN: Advanced Cardiac Therapies Improving Outcomes Network (ACTION) and Pediatric Cardiac Critical Care Consortium (PC 4 ) registries were used to identify and link children with initial VAD placement admitted to the cardiac ICU (CICU) from August 2014 to July 2020. Demographics, cardiac diagnosis, preoperative and postoperative CICU courses, and outcomes were compiled. Univariable and multivariable statistics assessed association of patient factors with prolonged postoperative MV. Multivariable logistic regression sought independent associations with outcomes. SETTING: Thirty-five pediatric CICUs across the United States and Canada. PATIENTS: Children on VADs included in both registries. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Two hundred forty-eight ACTION subjects were linked to a matching patient in PC 4 . Median (interquartile) age 7.7 years (1.5-15.5 yr), weight 21.3 kg (9.1-58 kg), and 56% male. Primary diagnosis was congenital heart disease (CHD) in 35%. Pre-VAD explanatory variables independently associated with prolonged postoperative MV included: age (incidence rate ratio [IRR], 0.95; 95% CI, 0.93-0.96; p < 0.01); preoperative MV within 48 hours (IRR, 2.76; 95% CI, 1.59-4.79; p < 0.01), 2-7 days (IRR, 1.82; 95% CI, 1.15-2.89; p = 0.011), and greater than 7 days before VAD implant (IRR, 2.35; 95% CI, 1.62-3.4; p < 0.01); and CHD (IRR, 1.96; 95% CI, 1.48-2.59; p < 0.01). Each additional day of postoperative MV was associated with greater odds of mortality (odds ratio [OR], 1.09 per day; p < 0.01) in the full cohort. We identified an associated greater odds of mortality in the 102 patients with intracorporeal devices (OR, 1.24; 95% CI, 1.04-1.48; p = 0.014), but not paracorporeal devices (77 patients; OR, 1.04; 95% CI, 0.99-1.09; p = 0.115). CONCLUSIONS: Prolonged MV after VAD placement is associated with greater odds of mortality in intracorporeal devices, which may indicate inadequacy of cardiopulmonary support in this group. This linkage provides a platform for future analyses in this population.

The effect of surgical weight loss on diabetes complications in individuals with class II/III obesity.

AIMS/HYPOTHESIS: The aim of this study was to determine the effect of bariatric surgery on diabetes complications in individuals with class II/III obesity (BMI > 35 kg/m2). METHODS: We performed a prospective cohort study of participants with obesity who underwent bariatric surgery. At baseline and 2 years following surgery, participants underwent metabolic phenotyping and diabetes complication assessments. The primary outcomes for peripheral neuropathy (PN) were a change in intra-epidermal nerve fibre density (IENFD, units = fibres/mm) at the distal leg and proximal thigh, the primary outcome for cardiovascular autonomic neuropathy (CAN) was a change in the expiration/inspiration (E/I) ratio, and the primary outcome for retinopathy was a change in the mean deviation on frequency doubling technology testing. RESULTS: Among 127 baseline participants, 79 completed in-person follow-up (age 46.0 ± 11.3 years [mean ± SD], 73.4% female). Participants lost a mean of 31.0 kg (SD 18.4), and all metabolic risk factors improved except for BP and total cholesterol. Following bariatric surgery, one of the primary PN outcomes improved (IENFD proximal thigh, +3.4 ± 7.8, p<0.01), and CAN (E/I ratio -0.01 ± 0.1, p=0.89) and retinopathy (deviation -0.2 ± 3.0, p=0.52) were stable. Linear regression revealed that a greater reduction in fasting glucose was associated with improvements in retinopathy (mean deviation point estimate -0.7, 95% CI -1.3, -0.1). CONCLUSIONS/INTERPRETATION: Bariatric surgery may be an effective approach to reverse PN in individuals with obesity. The observed stability of CAN and retinopathy may be an improvement compared with the natural progression of these conditions; however, controlled trials are needed.

The Impact of Nurse Practitioner Care and Accountable Care Organization Assignment on Skilled Nursing Services and Hospital Readmissions.

BACKGROUND: Accountable care organizations (ACOs) and the employment of nurse practitioners (NP) in place of physicians are strategies that aim to reduce the cost and improve the quality of routine care delivered in skilled nursing facilities (SNF). The recent expansion of ACOs and nurse practitioners into SNF settings in the United States may be associated with improved health outcomes for patients. OBJECTIVES: To determine the relationship between ACO attribution and NP care delivery during SNF visits and the relationship between NP care delivery during SNF visits and unplanned hospital readmissions. METHODS: We obtained a sample of 527,329 fee-for-service Medicare beneficiaries with 1 or more SNF stays between 2012 and 2017. We used logistic regression to measure the association between patient ACO attribution and evaluation and management care delivered by NPs in addition to the relationship between evaluation and management services delivered by NPs and hospital readmissions. RESULTS: ACO beneficiaries were 1.26% points more likely to receive 1 or more E&M services delivered by an NP during their SNF visits [Marginal Effect (ME): 0.0126; 95% CI: (0.009, 0.0160)]. ACO-attributed beneficiaries receiving most of their E&M services from NPs during their SNF visits were at a lower risk of readmission than ACO-attributed beneficiaries receiving no NP E&M care (5.9% vs. 7.1%; P <0.001). CONCLUSIONS: Greater participation by the NPs in care delivery in SNFs was associated with a reduced risk of patient readmission to hospitals. ACOs attributed beneficiaries were more likely to obtain the benefits of greater nurse practitioner involvement in their care.

Hidden Disparities: How Language Influences Patients' Access to Cancer Care.

BACKGROUND: Patients with limited English proficiency, a vulnerable patient population, remain understudied in the literature addressing cancer disparities. Although it is well documented that language discordance between patients and physicians negatively impacts the quality of patient care, little is known about how patients' preferred spoken language impacts their access to cancer care. PATIENTS AND METHODS: Between November 2021 and June 2022, we conducted an audit study of 144 hospitals located across 12 demographically diverse states. Using a standardized script, trained investigators assigned to the roles of English-speaking, Spanish-speaking, and Mandarin-speaking patients called the hospital general information telephone line seeking to access care for 3 cancer types that disproportionately impact Hispanic and Asian populations (colon, lung, and thyroid cancer). Primary outcome was whether the simulated patient caller was provided with the next steps to access cancer care, defined as clinic number or clinic transfer. We used chi-square tests and logistic regression analysis to test for associations between the primary outcome and language type, region type, hospital teaching status, and cancer care requested. We used multivariable logistic regression analysis to determine factors associated with simulated patient callers being provided the next steps. RESULTS: Of the 1,296 calls, 52.9% (n=686) resulted in simulated patient callers being provided next steps to access cancer care. Simulated non-English-speaking (vs English-speaking) patient callers were less likely to be provided with the next steps (Mandarin, 27.5%; Spanish, 37.7%; English, 93.5%; P<.001). Multivariable logistic regression found significant associations of the primary outcome with language spoken (Mandarin: odds ratio [OR], 0.02 [95% CI, 0.01-0.04] and Spanish: OR, 0.04 [95% CI, 0.02-0.06] vs English) and hospital teaching status (nonteaching: OR, 0.43 [95% CI, 0.32-0.56] vs teaching). CONCLUSIONS: Linguistic disparities exist in access to cancer care for non-English-speaking patients, emphasizing the need for focused interventions to mitigate systems-level communication barriers.

Real-World Use of Systemic Therapies for the Treatment of Advanced Thyroid Cancers.

OBJECTIVE: In the last decade, new systemic treatment options have been made available for patients with advanced thyroid cancer. However, little is known about the real-world utilization of these systemic therapies. METHODS: We used Optum's de-identified Clinformatics® Data Mart Database to characterize trends in the use of 15 systemic therapies that are available for the treatment of advanced thyroid cancer between 2013 and 2021. Joinpoint regression was used to calculate annual percentage changes in the use of systemic therapy by patients' race/ethnicity. The sequence of therapies was determined by the date of prescription claims. RESULTS: Between 2013 and 2021, the annual number of patients treated for advanced thyroid cancer with systemic therapy increased from 45 patients in 2013 to 114 patients in 2021 (N of total cohort = 885). Most patients were female (54.7%) and non-Hispanic White (62.1%). Between 2013 and 2021, there was a significant decrease in the proportion of non-Hispanic White patients treated for advanced thyroid cancer with systemic therapy (annual percentage change -3.9%, 95% confidence intervals, -6.0% to -1.8%). Since its approval by the US Food and Drug Administration (FDA) in 2015, lenvatinib remains the most frequently prescribed first-line therapy for the treatment of radioiodine-refractory thyroid cancer (48.8% of patients between 2017 and 2021). Between 2017 and 2021, most patients (79.7%) were initiated on 1 of the 10 FDA-approved agents and 81.7% received only a first-line therapy. CONCLUSIONS: Between 2013 and 2021, the use of systemic treatment options for advanced thyroid cancer increased significantly, largely driven by the prescription of lenvatinib following its approval by the FDA in 2015, with an increasing trend for use in non-White patients.

Unplanned Extubation During Pediatric Cardiac Intensive Care: U.S. Multicenter Registry Study of Prevalence and Outcomes.

OBJECTIVES: The epidemiology of unplanned extubations (UEs) and associated adverse outcomes in pediatric cardiac ICUs (CICU). DESIGN: Registry data (August 2014 to October 2020). SETTING: Forty-five Pediatric Cardiac Critical Care Consortium hospitals. PATIENTS: Patients receiving mechanical ventilation (MV) via endotracheal tube (ETT). INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Fifty-six thousand five hundred eight MV courses occurred in 36,696 patients, with a crude UE rate of 2.8%. In cardiac surgical patients, UE was associated with longer duration of MV, but we failed to find such association in medical patients. In both cohorts, UE was associated with younger age, being underweight, and airway anomaly. In multivariable logistic regression, airway anomaly was associated with UE in all patients. Younger age, higher Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery score category, longer duration of MV, and initial oral rather than nasal ETT are associated with UE in the surgical group, but we failed to find such associations in the medical group. UE was associated with a higher reintubation rate compared with elective extubation (26.8 vs 4.8%; odds ratio [OR], 7.35; 95% CI, 6.44-8.39; p < 0.0001) within 1 day of event. After excluding patients having redirection of care, UE was associated with at least three-fold greater odds for each of ventilator-associated pneumonia (VAP), cardiac arrest, and use of mechanical circulatory support (MCS). However, we failed to identify an association between UE and greater odds of mortality (1.2 vs 0.8%; OR, 1.48; 95% CI, 0.86-2.54; p = 0.15), but uncertainty remains. CONCLUSIONS: UE in CICU patients is associated with greater odds of cardiac arrest, VAP, and MCS. Cardiac medical and surgical patients in the CICU appear to have different explanatory factors associated with UE, and perhaps these may be modifiable and tested in future collaborative population research.

ASSESSMENT OF QUALITY AND ADEQUACY OF PANRETINAL PHOTOCOAGULATION IN PROLIFERATIVE DIABETIC RETINOPATHY USING ULTRA-WIDEFIELD IMAGING.

PURPOSE: To assess the characteristics of completed panretinal photocoagulation (PRP), using ultra-widefield imaging in proliferative diabetic retinopathy. METHODS: Quantitative assessment of ultra-widefield imaging images of 133 patients with proliferative diabetic retinopathy with completed PRP was made using ImageJ software. The parameters assessed included distance of laser spots from the optic disk, foveal center, superior and inferior arcades, and extent of the maximum width of laser. Areas assessed were total area of the image, area of the inner limit within which laser spots are restricted, minimum areas of unlasered patches, total area lasered, and ideal area to be covered by PRP. RESULTS: Two hundred one images were assessed for the final analysis. The mean distance of laser spots was 4.2 ± 2.4 mm from the optic disk (nasal) and 6.6 ± 2.5 mm from the foveal center (temporal). The mean distance of laser spots from the superior arcade vessel was 3.2 ± 1.9 mm and 6.2 ± 4.4 mm from the inferior arcade. The mean area of the retina that should have been ideally lasered was found to be 900 ± 267 mm 2 , and the actual area lasered was found to be 681 ± 254.4 mm 2 . CONCLUSION: Approximately one-quarter area of the retina continues to remain ischemic because of the lack of inadequate coverage of PRP. Further longitudinal studies are recommended, using ultra-widefield imaging to objectively assess the adequacy of PRP and its role in modulating the course of progression of the retinopathy.

Critical Care Unit Organizational and Personnel Factors Impact Cardiac Arrest Prevention and Rescue in the Pediatric Cardiac Population.

OBJECTIVES: Patient-level factors related to cardiac arrest in the pediatric cardiac population are well understood but may be unmodifiable. The impact of cardiac ICU organizational and personnel factors on cardiac arrest rates and outcomes remains unknown. We sought to better understand the association between these potentially modifiable organizational and personnel factors on cardiac arrest prevention and rescue. DESIGN: Retrospective analysis of the Pediatric Cardiac Critical Care Consortium registry. SETTING: Pediatric cardiac ICUs. PATIENTS: All cardiac ICU admissions were evaluated for cardiac arrest and survival outcomes. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Successful prevention was defined as the proportion of admissions with no cardiac arrest (inverse of cardiac arrest incidence). Rescue was the proportion of patients surviving to cardiac ICU discharge after cardiac arrest. Cardiac ICU organizational and personnel factors were captured via site questionnaires. The associations between organizational and personnel factors and prevention/rescue were analyzed using Fine-Gray and multinomial regression, respectively, accounting for clustering within hospitals. We analyzed 54,521 cardiac ICU admissions (29 hospitals) with 1,398 cardiac arrest events (2.5%) between August 1, 2014, and March 5, 2019. For both surgical and medical admissions, lower average daily cardiac ICU occupancy was associated with better cardiac arrest prevention. Better rescue for medical admissions was observed for higher registered nursing hours per patient day and lower proportions of "part time" cardiac ICU physician staff (< 6 service weeks/yr). Increased registered nurse experience was associated with better rescue for surgical admissions. Increased proportion of critical care certified nurses, full-time intensivists with critical care fellowship training, dedicated respiratory therapists, quality/safety resources, and annual cardiac ICU admission volume were not associated with improved prevention or rescue. CONCLUSIONS: Our multi-institutional analysis identified cardiac ICU bed occupancy, registered nurse experience, and physician staffing as potentially important factors associated with cardiac arrest prevention and rescue. Recognizing the limitations of measuring these variables cross-sectionally, additional studies are needed to further investigate these organizational and personnel factors, their interrelationships, and how hospitals can modify structure to improve cardiac arrest outcomes.

Mortality from Pulmonary Hypertension in the Pediatric Cardiac ICU.

Rationale: Patients with pulmonary hypertension (PH) admitted to pediatric cardiac ICUs are at high risk of mortality. Objectives: To identify factors associated with mortality in cardiac critical care admissions with PH. Methods: We evaluated medical admissions with PH to Pediatric Cardiac Critical Care Consortium institutions over 5 years. PH was standardly defined in the clinical registry by diagnosis and/or receipt of intensive care-level pulmonary vasodilator therapy. Multivariable logistic regression identified independent associations with mortality. Measurements and Main Results: We analyzed 2,602 admissions; mortality was 10% versus 3.9% for all other medical admissions. Covariates most strongly associated with mortality included invasive ventilation (adjusted odds ratio, 44.8; 95% confidence interval, 6.2-323), noninvasive ventilation (19.7; 2.8-140), cardiopulmonary resuscitation (8.9; 5.6-14.1), and vasoactive infusions (4.8; 2.6-8.8). Patients receiving both invasive ventilation and vasoactive infusions on admission Days 1 and 2 had an observed mortality rate of 29.2% and 28.6%, respectively, compared with <5% for those not receiving either. Vasoactive infusions emerged as the dominant early risk factor for mortality, increasing the absolute risk of mortality on average by 6.4% when present on admission Day 2. Conclusions: Patients with PH admitted to pediatric cardiac critical care units have high mortality rates. Those receiving invasive ventilation and vasoactive infusions on Day 1 or Day 2 had an observed mortality rate that was more than fivefold greater than that of those who did not. These data highlight the illness severity of patients with PH in this setting and could help inform conversations with families regarding the prognosis.

Osteonecrosis of the jaw risk factors in bisphosphonate-treated patients with metastatic cancer.

BACKGROUND: A case-control study was performed to define clinical and genetic risk factors associated with osteonecrosis of the jaw in patients with metastatic cancer treated with bisphosphonates. METHODS: Clinical data and tissues were collected from patients treated with bisphosphonates for metastatic bone disease who were diagnosed with osteonecrosis of the jaw (cases) and matched controls. Clinical data included patient, behavioral, disease, and treatment information. Genetic polymorphisms in CYP2C8 (rs1934951) and other candidate genes were genotyped. Odds ratios from conditional logistic regression models were examined to identify clinical and genetic characteristics associated with case or control status. RESULTS: The study population consisted of 76 cases and 126 controls. In the final multivariable clinical model, patients with osteonecrosis of the jaw were less likely to have received pamidronate than zoledronic acid (odds ratio = 0.18, 95% Confidence interval: 0.03-0.97, p = .047) and more likely to have been exposed to bevacizumab (OR = 5.15, 95% CI: 1.67-15.95, p = .005). The exploratory genetic analyses suggested a protective effect for VEGFC rs2333496 and risk effects for VEGFC rs7664413 and PPARG rs1152003. CONCLUSIONS: We observed patients with ONJ were more likely to have been exposed to bevacizumab and zoledronic and identified potential genetic predictors that require validation prior to clinical translation.

Prediction of extubation failure in the paediatric cardiac ICU using machine learning and high-frequency physiologic data.

BACKGROUND: Cardiac intensivists frequently assess patient readiness to wean off mechanical ventilation with an extubation readiness trial despite it being no more effective than clinician judgement alone. We evaluated the utility of high-frequency physiologic data and machine learning for improving the prediction of extubation failure in children with cardiovascular disease. METHODS: This was a retrospective analysis of clinical registry data and streamed physiologic extubation readiness trial data from one paediatric cardiac ICU (12/2016-3/2018). We analysed patients' final extubation readiness trial. Machine learning methods (classification and regression tree, Boosting, Random Forest) were performed using clinical/demographic data, physiologic data, and both datasets. Extubation failure was defined as reintubation within 48 hrs. Classifier performance was assessed on prediction accuracy and area under the receiver operating characteristic curve. RESULTS: Of 178 episodes, 11.2% (N = 20) failed extubation. Using clinical/demographic data, our machine learning methods identified variables such as age, weight, height, and ventilation duration as being important in predicting extubation failure. Best classifier performance with this data was Boosting (prediction accuracy: 0.88; area under the receiver operating characteristic curve: 0.74). Using physiologic data, our machine learning methods found oxygen saturation extremes and descriptors of dynamic compliance, central venous pressure, and heart/respiratory rate to be of importance. The best classifier in this setting was Random Forest (prediction accuracy: 0.89; area under the receiver operating characteristic curve: 0.75). Combining both datasets produced classifiers highlighting the importance of physiologic variables in determining extubation failure, though predictive performance was not improved. CONCLUSION: Physiologic variables not routinely scrutinised during extubation readiness trials were identified as potential extubation failure predictors. Larger analyses are necessary to investigate whether these markers can improve clinical decision-making.

Epidemiology of Acute Kidney Injury After Neonatal Cardiac Surgery: A Report From the Multicenter Neonatal and Pediatric Heart and Renal Outcomes Network.

OBJECTIVES: Cardiac surgery-associated acute kidney injury occurs commonly following congenital heart surgery and is associated with adverse outcomes. This study represents the first multicenter study of neonatal cardiac surgery-associated acute kidney injury. We aimed to describe the epidemiology, including perioperative predictors and associated outcomes of this important complication. DESIGN: This Neonatal and Pediatric Heart and Renal Outcomes Network study is a multicenter, retrospective cohort study of consecutive neonates less than 30 days. Neonatal modification of The Kidney Disease Improving Global Outcomes criteria was used. Associations between cardiac surgery-associated acute kidney injury stage and outcomes (mortality, length of stay, and duration of mechanical ventilation) were assessed through multivariable regression. SETTING: Twenty-two hospitals participating in Pediatric Cardiac Critical Care Consortium. PATIENTS: Twenty-two-thousand forty neonates who underwent major cardiac surgery from September 2015 to January 2018. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Cardiac surgery-associated acute kidney injury occurred in 1,207 patients (53.8%); 983 of 1,657 in cardiopulmonary bypass patients (59.3%) and 224 of 583 in noncardiopulmonary bypass patients (38.4%). Seven-hundred two (31.3%) had maximum stage 1, 302 (13.5%) stage 2, 203 (9.1%) stage 3; prevalence of cardiac surgery-associated acute kidney injury peaked on postoperative day 1. Cardiac surgery-associated acute kidney injury rates varied greatly (27-86%) across institutions. Preoperative enteral feeding (odds ratio = 0.68; 0.52-0.9) and open sternum (odds ratio = 0.76; 0.61-0.96) were associated with less cardiac surgery-associated acute kidney injury; cardiopulmonary bypass was associated with increased cardiac surgery-associated acute kidney injury (odds ratio = 1.53; 1.01-2.32). Duration of cardiopulmonary bypass was not associated with cardiac surgery-associated acute kidney injury in the cardiopulmonary bypass cohort. Stage 3 cardiac surgery-associated acute kidney injury was independently associated with hospital mortality (odds ratio = 2.44; 1.3-4.61). No cardiac surgery-associated acute kidney injury stage was associated with duration of mechanical ventilation or length of stay. CONCLUSIONS: Cardiac surgery-associated acute kidney injury occurs frequently after neonatal cardiac surgery in both cardiopulmonary bypass and noncardiopulmonary bypass patients. Rates vary significantly across hospitals. Only stage 3 cardiac surgery-associated acute kidney injury is associated with mortality. Cardiac surgery-associated acute kidney injury was not associated with any other outcomes. Kidney Disease Improving Global Outcomes criteria may not precisely define a clinically meaningful renal injury phenotype in this population.

Time-varying survival effects for squamous cell carcinomas at oropharyngeal and nonoropharyngeal head and neck sites in the United States, 1973-2015.

BACKGROUND: Anatomical site is strongly associated with head and neck cancer etiology, and etiology and patient sociodemographic characteristics are prognostic factors for survival. It is not known whether the effects of these predictors persist over the postdiagnosis period or are strongest proximal to the time of diagnosis. METHODS: Using survival times and causes of death for 180,434 patients with head and neck cancer in the Surveillance, Epidemiology, and End Results cancer registry (1973-2015), the empirical cumulative incidences of cancer-specific death and other-cause death were calculated with a competing risks framework, and the time-dependent effects (hazard ratios) of anatomical tumor site (oropharynx, oral cavity, or hypopharynx/larynx), age, sex, race, and year of diagnosis on cancer-specific death and other-cause death, stratified by tumor stage, were estimated. RESULTS: All effects were significantly time-varying (P < .001). Patients with nonoropharyngeal cancer had a higher hazard of cancer-specific death but a similar cumulative fraction of deaths because of a higher rate of death from other causes. Cancer-specific survival has not changed for patients with nonoropharyngeal cancer over the past decades but has improved since 2000 for patients with oropharyngeal cancer. The effects of age and sex on cancer survival were strongest proximal to the diagnosis, whereas the effect of race persisted over time. CONCLUSIONS: Recent improvements in survival for patients with oropharyngeal cancer may be due more to an increasing fraction of cancers attributable to human papillomavirus than to increasing treatment effectiveness. The prognostic strength of anatomical site and other predictors changes over the postdiagnosis period. LAY SUMMARY: It is generally assumed that the effects of tumor and personal characteristics on the survival of patients with head and neck cancer are fixed over time, but this study shows that many factors are most important only in the first few years after diagnosis. Also, recent improvements in the survival of patients with head and neck cancer appear to benefit only patients with cancers of the oropharynx. The improvements may be due more to an increasing fraction of cancers caused by human papillomavirus (which generally have better outcomes) than to advances in head and neck cancer treatment overall.

The Relationship Between Imaging and Thyroid Cancer Diagnosis and Survival.

BACKGROUND: Controversy exists over whether there has been a true increase in the occurrence of thyroid cancer or overdiagnosis secondary to imaging practices. Because cancer overdiagnosis is associated with detection of indolent disease, overdiagnosis can be associated with perceived improvement in survival. MATERIALS AND METHODS: Surveillance, Epidemiology, and End Results-Medicare linked database was used to determine the relationship between type of imaging leading to thyroid cancer diagnosis and survival. Disease-specific and overall survival were evaluated in 11,945 patients aged ≥66 years with differentiated thyroid cancer diagnosed between January 1, 2001, and September 30, 2015, who prior to their cancer diagnosis initially underwent thyroid ultrasound versus other imaging capturing the neck. Analyses were performed using the Kaplan-Meier method and Cox proportional hazards model with propensity score. RESULTS: Patients who underwent thyroid ultrasound as compared with other imaging had improved disease-specific and overall survival (p < .001, p < .001). However, those who underwent thyroid ultrasound were less likely to have comorbidities (p < .001) and more likely to be younger (p < .001), be female (p < .001), have localized cancer (p < .001), and have tumor size ≤1 cm (p < .001). After using propensity score analysis and adjusting for tumor characteristics, type of initial imaging still correlated with better overall survival but no longer correlated with better disease-specific survival. CONCLUSION: There is improved disease-specific survival in patients diagnosed with thyroid cancer after thyroid ultrasound as compared with after other imaging. However, better disease-specific survival is related to these patients being younger and healthier and having lower-risk cancer, suggesting that thyroid ultrasound screening contributes to cancer overdiagnosis. IMPLICATIONS FOR PRACTICE: The findings from this study have implications for patients, physicians, and policy makers. Patients who have thyroid ultrasound as their initial imaging are fundamentally different from those who are diagnosed after other imaging. Because patients undergoing ultrasound are younger and healthier and are diagnosed with lower-risk thyroid cancer, they are less likely to die of their thyroid cancer. However, being diagnosed with thyroid cancer can lead to cancer-related worry and create risks for harm from treatments. Thus, efforts are needed to reduce inappropriate use of ultrasound, abide by the U.S. Preventive Services Task Force recommendations, and apply nodule risk stratification tools when appropriate.

Predicting and Surviving Prolonged Critical Illness After Congenital Heart Surgery.

OBJECTIVES: Prolonged critical illness after congenital heart surgery disproportionately harms patients and the healthcare system, yet much remains unknown. We aimed to define prolonged critical illness, delineate between nonmodifiable and potentially preventable predictors of prolonged critical illness and prolonged critical illness mortality, and understand the interhospital variation in prolonged critical illness. DESIGN: Observational analysis. SETTING: Pediatric Cardiac Critical Care Consortium clinical registry. PATIENTS: All patients, stratified into neonates (≤28 d) and nonneonates (29 d to 18 yr), admitted to the pediatric cardiac ICU after congenital heart surgery at Pediatric Cardiac Critical Care Consortium hospitals. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: There were 2,419 neonates and 10,687 nonneonates from 22 hospitals. The prolonged critical illness cutoff (90th percentile length of stay) was greater than or equal to 35 and greater than or equal to 10 days for neonates and nonneonates, respectively. Cardiac ICU prolonged critical illness mortality was 24% in neonates and 8% in nonneonates (vs 5% and 0.4%, respectively, in nonprolonged critical illness patients). Multivariable logistic regression identified 10 neonatal and 19 nonneonatal prolonged critical illness predictors within strata and eight predictors of mortality. Only mechanical ventilation days and acute renal failure requiring renal replacement therapy predicted prolonged critical illness and prolonged critical illness mortality in both strata. Approximately 40% of the prolonged critical illness predictors were nonmodifiable (preoperative/patient and operative factors), whereas only one of eight prolonged critical illness mortality predictors was nonmodifiable. The remainders were potentially preventable (postoperative critical care delivery variables and complications). Case-mix-adjusted prolonged critical illness rates were compared across hospitals; six hospitals each had lower- and higher-than-expected prolonged critical illness frequency. CONCLUSIONS: Although many prolonged critical illness predictors are nonmodifiable, we identified several predictors to target for improvement. Furthermore, we observed that complications and prolonged critical care therapy drive prolonged critical illness mortality. Wide variation of prolonged critical illness frequency suggests that identifying practices at hospitals with lower-than-expected prolonged critical illness could lead to broader quality improvement initiatives.

Randomized controlled trial of a clinical decision support system for painful polyneuropathy.

INTRODUCTION: Despite the existence of guidelines, painful neuropathy is often inappropriately treated. We sought to determine the effectiveness of a clinical decision support system on guideline-recommended medication use. METHODS: We randomized neurology providers, stratified by subspecialty, to a best practice alert (BPA) linked to a Smartset or a BPA alone when seeing patients with neuropathy. The primary outcome was the proportion of patients with uncontrolled nerve pain prescribed a guideline-recommended medication. Generalized estimating equations were used to assess effectiveness. RESULTS: Seventy-five neurology providers (intervention 38, control 37) treated 2697 patients with neuropathy (intervention 1026, control 671). Providers did not acknowledge the BPA in 1928 (71.5%) visits. Only four of eight intervention arm neurologists who treated patients with uncontrolled nerve pain opened the Smartset. The intervention was not associated with guideline-recommended medication use (odds ratio 0.52, 0.18-1.48; intervention 52%, control 54.8%). DISCUSSION: Our intervention did not improve prescribing practices for painful neuropathy. Physicians typically ignored the BPAs/Smartset; therefore, future studies should mandate their use or employ alternate strategies.