Audrey Renson

Accessible, curated metagenomic data through ExperimentHub

Affiliations: 1 Centre for Integrative Biology, University of Trento, Trento, Italy 2 Institute for Implementation Science and Population Health, City University of New York School of Public Health, New York, New York, United States of America 3 Roswell Park Cancer Institute, University of Buffalo, Buffalo, New York, United States of America 4 Department of Global Health and Social Medicine, King’s College London 5 Biostatistics Department, Harvard School of Public Health, Boston, Massachusetts, United States of America 6 The Broad Institute, Cambridge, Massachusetts, United States of America

Discriminative Ability of Commonly Used Indexes to Predict Adverse Outcomes After Radical Cystectomy: Comparison of Demographic Data, American Society of Anesthesiologists, Modified Charlson Comorbidity Index, and Modified Frailty Index.

Micro‐Abstract Given the high rate of adverse events after radical cystectomy, we evaluated the discriminative ability of commonly used comorbidity indexes and demographic factors for perioperative complications in patients undergoing radical cystectomy. We found the predictive ability of these factors to be universally poor, highlighting the need for newer models built to identify patients with a greater risk of adverse events after radical cystectomy. Background: The American Society of Anesthesiologists physical status classification system, modified Charlson Comorbidity Index (mCCI), and modified Frailty Index have been associated with complications after urologic surgery. No study has compared the predictive performance of these indexes for postoperative complications after radical cystectomy (RC) for bladder cancer. Materials and Methods: Data from 1516 patients undergoing elective RC for bladder cancer were extracted from the 2005 to 2011 American College of Surgeons National Surgical Quality Improvement Program for a retrospective review. The perioperative outcome variables assessed were occurrence of minor adverse events, severe adverse events, infectious adverse events, any adverse event, extended length of hospital stay, discharge to a higher level of care, and mortality. Patient comorbidity indexes and demographic data were assessed for their discriminative ability in predicting perioperative adverse outcomes using an area under the curve (AUC) analysis from the receiver operating characteristic curves. Results: The most predictive comorbidity index for any adverse event was the mCCI (AUC, 0.511). The demographic factors were the body mass index (BMI; AUC, 0.519) and sex (AUC, 0.519). However, the overall performance for all predictive indexes was poor for any adverse event (AUC < 0.52). Combining the most predictive demographic factor (BMI) and comorbidity index (mCCI) resulted in incremental improvements in discriminative ability compared with that for the individual outcome variables. Conclusion: For RC, easily obtained patient mCCI, BMI, and sex have overall similar discriminative abilities for perioperative adverse outcomes compared with the tabulated indexes, which are more difficult to implement in clinical practice. However, both the demographic factors and the comorbidity indexes had poor discriminative ability for adverse events.

Earlier Goals of Care Discussions in Hospitalized Terminally Ill Patients and the Quality of End-of-Life Care: A Retrospective Study

Background: The association between physician-directed goals of care discussions (GOCDs) and the use of aggressive interventions in terminally ill patients has not been well characterized in the literature. We examined the associations between the timing of physician-directed GOCDs in terminally ill patients and the use of aggressive interventions, probability of dying in the inpatient setting, and intensive care unit (ICU) utilization. Methods: This retrospective cohort study included patients admitted to our urban community hospital in 2015 who had a terminal diagnosis on admission and either died on an inpatient unit or were discharged to hospice. The primary independent variable was the number of days from admission to GOCD, expressed as a proportion of the patient’s length of stay (LOS). We used robust variance Poisson and zero-inflated negative binomial regression, as appropriate, to estimate the associations between goals of care timing and risk of having an intervention, risk of dying in the inpatient setting, odds of ICU admission, and ICU LOS. Results: A total of 197 cases were included. After adjusting for age, language, gender, insurance, dementia, and decision maker (patient versus surrogate decision maker), later GOCD was significantly associated with greater risk of having an aggressive intervention (risk ratio [RR] = 1.04, 95% confidence interval [CI] = 1.02-1.06), greater risk of death as an inpatient (RR = 1.04, 95% CI = 1.02-1.06), and greater odds of ICU admission (odds ratio = 1.19, 95% CI = 1.02-1.39). Conclusion: Later GOCDs were associated with greater risk of aggressive interventions and death as an inpatient and greater odds of ICU admission. Goals of care discussion should be done routinely and early during the hospitalization of terminally ill patients.

Seatbelt use is associated with lower risk of high-grade hepatic injury in motor vehicle crashes in a national sample

Background Seatbelt use, alone and in conjunction with an airbag, is associated with lower risk of mortality, blunt abdominal trauma and kidney injury in motor vehicle crashes (MVCs). However, the effect of these protective devices on risk of severe liver injury is not well characterised. Methods This retrospective cohort study included patient admissions with liver injuries from MVCs from the National Trauma Data Bank (NTDB), collected from 2010 to 2015 in the USA. We examined associations between injury severity and seatbelt use and airbag presence individually and in the presence of additive interaction. Secondary outcomes were mortality, complications and discharge disposition. Results We analysed 55 543 records from the National Trauma Data Bank. In adjusted analysis, seatbelt use alone was protective against severe (AAST VI or above) hepatic injury (risk ratio (RR) 0.79, 95% CI 0.75 to 0.84), while airbag presence alone was not (RR 1.05, 95% CI 0.8 to 1.12). The joint association of seatbelt use and airbag presence with injury severity was greater than seatbelts alone (RR 0.74, 95% CI 0.70 to 0.79), with 13% of the joint lower risk attributable to interaction (95% CI 3% to 24%). The adjusted mortality risk of those without protective devices (10.3%, n=2297) was nearly double that of patients who used a seatbelt in conjunction with a present airbag (5.3%, n=699, p<0.001). Conclusions Seatbelts are associated with lower liver injury severity and are more protective with airbags present, while airbags without seatbelt use were not protective against severe injury among patients with liver injury.

Impact of Trauma Center Designation and Interfacility Transfer on Renal Trauma Outcomes: Evidence for Universal Management

BACKGROUND Renal trauma may be managed differently in tiered trauma systems and among those who requireinterfaculty transfer. OBJECTIVE To evaluate the initial management of renal trauma, assess patterns of management based on hospital trauma level designation and interfacility transfer status, and analyze management trends over time. DESIGN, SETTING, AND PARTICIPANTS Retrospective cohort study of renal trauma from the National Trauma Data Bank 2010-2015. INTERVENTION Nephrectomy, angioembolization, or nonoperative management. OUTCOMES MEASUREMENTS AND STATISTICAL ANALYSIS We used generalized estimating equations to compare odds of each management outcome in patients transferred and directly admitted to a level I center, versus those directly admitted to a non-level I center, adjusting for vital signs, injury, demographic, and facility characteristics. We also used generalized estimating equations to examine linear time trends in management outcome, adjusting for injury characteristics. RESULTS AND LIMITATIONS A total of 51798 renal trauma records were included: 44 838 low-grade (American Association for the Surgery of Trauma I-III) and 6359 high grade (IV-V) injuries. After adjusting for comorbidities, demographics, and hospital characteristics, odds of nephrectomy, angioembolization, and nonoperative management were similar in patients transferred or directly admitted to a level I center compared with those treated at a non-level I center. Changes in management over time demonstrated a decreased rate of nephrectomy (p=0.007) in high-grade injuries, while the rate of angioembolization remained constant (p=0.33). Study limitations include mortality prior to hospital transfer or arrival, and its retrospective nature. CONCLUSIONS In this contemporary trauma analysis, outcomes of both low- and high-grade renal trauma are similar across patients managed in tiered trauma centers and those undergoing transfer, signifying dissemination of collective renal trauma management. The rate of nephrectomy has decreased for high-grade renal injury over our study period, suggesting new adoption of kidney-sparing management. PATIENT SUMMARY Renal trauma is now managed similarly in tiered trauma centers and in patients requiring interfacility transfer. The rate of nephrectomy for high-grade renal injuries has decreased over time.

Predicting Benign Prostate Pathology on Magnetic Resonance Imaging/Ultrasound Fusion Biopsy in Men with a Prior Negative 12-core Systematic Biopsy: External Validation of a Prognostic Nomogram

BACKGROUND Magnetic resonance imaging (MRI) of the prostate after a prior negative biopsy may reduce the need for unnecessary repeat biopsies. OBJECTIVE To externally validate a previously developed nomogram predicting benign prostate pathology on MRI/ultrasound (US) fusion-targeted biopsy in men with a Prostate Imaging Reporting and Data System (PI-RADS) 3-5 region of interest and a prior negative 12-core systematic biopsy, and update this nomogram to improve its performance. DESIGN, SETTING, AND PARTICIPANTS A total of 2063 men underwent MRI/US fusion-targeted biopsy from April 2012 to September 2017; 104 men with a negative systematic biopsy followed by MRI-US fusion-targeted biopsy of a PI-RADS 3-5 region of interest (58%) met the study inclusion criteria. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS An MRI-based nomogram that had previously been developed in a multi-institutional clinical setting was externally validated. Predictive characteristics were age, prostate volume, MRI PI-RADS score, and prostate-specific antigen (PSA). Bayesian logistic regression was used to update the previous model. RESULTS AND LIMITATIONS Median age of the external validation cohort was 68 yr, PSA was 7.2ng/ml, and biopsy confirmed benign pathology in 30% (n=31), suggesting a lower baseline risk compared with the nomogram development cohort. Receiver operating characteristic curve analysis showed areas under curve (AUCs) from 0.77 to 0.80 for nomogram validation. An updated model was constructed with improved calibration and similar discrimination (AUC 0.79). CONCLUSIONS Age, prostate volume, PI-RADS, and PSA predict benign pathology on MRI/US fusion-targeted biopsy in men with a prior negative 12-core systematic biopsy. The validated and updated nomogram demonstrated high diagnostic accuracy and may further aid in the decision to avoid a biopsy in men with a prior negative biopsy. PATIENT SUMMARY We externally validated a clinically useful tool that predicts benign prostate pathology on magnetic resonance imaging/ultrasound fusion-targeted biopsy in men with a prior negative 12-core systematic biopsy and updated this predictive tool to improve its performance in patient counseling regarding the need for a repeat biopsy.

Development of a Novel Prognostic Risk Score for Predicting Complications of Penectomy in the Surgical Management of Penile Cancer

Introduction Penectomy for PC is useful in staging, disease prognosis, and treatment. Limited studies have evaluated its surgical complications. We sought to assess these complications and determine predictive models to create a novel risk score for penectomy complications. Patients and Methods A retrospective review of patients undergoing PC surgical management from the 2005‐2016 American College of Surgeons National Surgical Quality Improvement Program was performed. Data were queried for partial and total penectomy among those with PC. To develop predictive models of complications, we fit LASSO logistic, random forest, and stepwise logistic models to training data using cross‐validation, demographic, comorbidity, laboratory, and wound characteristics as candidate predictors. Each model was evaluated on the test data using receiver operating characteristic curves. A novel risk score was created by rounding coefficients from the LASSO logistic model. Results A total of 304 cases met the inclusion criteria. Overall incidence of penectomy complications was 19.7%, where urinary tract infection (3.0%), superficial surgical site infection (3.0%), and bleeding requiring transfusion (3.9%) were most common. LASSO logistic, random forest, and stepwise logistic models for predicting complications had area under the curve (AUC) [95% confidence interval] values of 0.66 [0.52‐0.81], 0.73 [0.63‐0.83], and 0.59 [0.45‐0.74], respectively. Eleven variables were included in the risk score. The LASSO model–derived risk score had moderately good performance (area under the curve [95% confidence interval] 0.74 [0.66‐0.82]). Using a cutoff point of 6, the score attains sensitivity 0.58, specificity 0.74, and kappa 0.26. Conclusion PC management through penectomy is associated with appreciable complications rates. Predictive models of penectomy complications performed moderately well. Our novel prognostic risk score may allow for improved preoperative counseling and risk stratification of men undergoing surgical management of PC. Micro‐Abstract Limited studies exist on complications of penile cancer surgical treatment. We sought to create a novel risk score for penectomy. Using the American College of Surgeons National Surgery Quality Improvement Program database, we fit least absolute shrinkage and selection operator, random forest, and stepwise logistic models to training data and evaluated each model using receiver operating characteristic curves. Predictive models performed moderately well. Our novel prognostic risk score may allow for improved preoperative counseling.

Sociodemographic patterning in the oral microbiome of a diverse sample of New Yorkers

1.1 Purpose Variations in the oral microbiome are potentially implicated in social inequalities in oral disease, cancers, and metabolic disease. We describe sociodemographic variation of oral microbiomes in a diverse sample. 1.2 Methods We performed 16S rRNA sequencing on mouthwash specimens in a subsample (n=282) of the 2013-14 population-based New York City Health and Nutrition Examination Study (NYC-HANES). We examined differential abundance of 216 operational taxonomic units (OTUs), and alpha and beta diversity by age, sex, income, education, nativity, and race/ethnicity. For comparison, we also examined differential abundance by diet, smoking status, and oral health behaviors. 1.3 Results 69 OTUs were differentially abundant by any sociodemographic variable (false discovery rate < 0.01), including 27 by race/ethnicity, 21 by family income, 19 by education, three by sex. We also found 49 differentially abundant by smoking status, 23 by diet, 12 by oral health behaviors. Genera differing for multiple sociodemographic characteristics included Lactobacillus, Prevotella, Porphyromonas, Fusobacterium. 1.4 Conclusions We identified oral microbiome variation consistent with health inequalities, with more taxa differing by race/ethnicity than diet, and more by SES variables than oral health behaviors. Investigation is warranted into possible mediating effects of the oral microbiome in social disparities in oral, metabolic and cancers. Highlights Most microbiome studies to date have had minimal sociodemographic variability, limiting what is known about associations of social factors and the microbiome. We examined the oral microbiome in a population-based sample of New Yorkers with wide sociodemographic variation. Numerous taxa were differentially abundant by race/ethnicity, income, education, marital status, and nativity. Frequently differentially abundant taxa include Porphyromonas, Fusobacterium, Streptococcus, and Prevotella, which are associated with oral and systemic disease. Mediation of health disparities by microbial factors may represent an important intervention site to reduce health disparities, and should be explored in prospective studies.Background: Variations in the human oral microbiome are potentially implicated in health inequalities, but existing studies of the oral microbiome have minimal sociodemographic diversity. This study describes sociodemographic variation of the oral microbiome in a diverse sample of New York City residents. Methods: Data come from 296 participants, a subsample of the 2013-14 population-based New York City Health and Nutrition Examination Study (NYC-HANES). Mouthwash samples were processed using using 16S v4 rRNA amplicon sequencing. We examined differential abundance of 216 operational taxonomic units (OTUs), in addition to alpha and beta diversity amongst sociodemographic variables including age, gender, income, education, nativity, and race/ethnicity. Results: A total of 75 OTUs were differentially abundant by any sociodemographic variable (false discovery rate < 0.01), including 27 by race/ethnicity, 23 by family income, 20 by education, and five by gender. Genera differing for more than one sociodemographic characteristic included Lactobacillus, Prevotella, Porphyromonas, and Fusobacterium. Education (p=0.03) and age (p=0.02) were associated with weighted UniFrac distances. Discussion: In a diverse sample, we identified variations in the oral microbiome consistent with health inequalities. Further investigation is warranted into possible mediating effects of the oral microbiome in social disparities in diabetes, inflammation, oral health, and preterm birth.