University Of Pittsburgh Medical Center Research Articles

Outcomes and Treatment Algorithm in Glioblastoma Patients 80 Years and Older

The current standard of care for patients with glioblastoma (GBM) is maximal safe resection followed by adjuvant radiation therapy with concurrent temozolomide chemotherapy. Previous studies that identified this treatment regimen focused on younger patients with GBM. The proportion of patients with GBM over the age of 80 years is increasing. We investigate whether elderly patients benefit from the current standard of care with additional maximal safe resection. Clinical, operative, radiographic, demographic, genetic, and outcomes data were retrospectively collected for patients treated for histologically confirmed World Health Organization grade 4 GBM at University of Pittsburgh Medical Center from 2009 to 2020. Only patients 80 years and older were included (n= 123). Statistically significant values were set at P < 0.05. A univariate Cox proportional hazards analysis of GBM patients aged >80 years identified the use of temozolomide, radiation, Karnofsky Performance Status (KPS) > 70, and methylguanine DNA methyltransferase methylation with increased overall survival (OS). Further multivariate Cox proportional hazards model analysis showed that the variables identified in the univariate analysis passed multicollinearity testing, and that use of temozolomide, KPS >70, and gross total resection were shown to significantly impact survival. Survival analysis showed that patients with biopsy alone had a shorter median OS compared with patients who received resection, temozolomide, and radiation (P < 0.0001, median OS 1.6 vs. 7.5 months). Additionally, patients who underwent biopsy and then received temozolomide and radiation had a shorter median OS when compared with patients who received resection, temozolomide, and radiation (P= 0.0047, median OS 3.6 vs. 7.5 months). For elderly patients with KPS >70, GTR followed by radiation and temozolomide is associated with maximum OS.

A Novel Framework to Address the Complexities of Housing Insecurity and Its Associated Health Outcomes and Inequities: "Give, Partner, Invest".

The association between housing insecurity and reduced access to healthcare, diminished mental and physical health, and increased mortality is well-known. This association, along with structural racism, social inequities, and lack of economic opportunities, continues to widen the gap in health outcomes and other disparities between those in higher and lower socio-economic strata in the United States and throughout the advanced economies of the world. System-wide infrastructure failures at municipal, state, and federal government levels have inadequately addressed the difficulty with housing affordability and stability and its associated impact on health outcomes and inequities. Healthcare systems are uniquely poised to help fill this gap and engage with proposed solutions. Strategies that incorporate multiple investment pathways and emphasize community-based partnerships and innovation have the potential for broad public health impacts. In this manuscript, we describe a novel framework, "Give, Partner, Invest," which was created and utilized by the University of Pittsburgh Medical Center (UPMC) Insurance Services Division (ISD) as part of the Integrated Delivery and Finance System to demonstrate the financial, policy, partnership, and workforce levers that could make substantive investments in affordable housing and community-based interventions to improve the health and well-being of our communities. Further, we address housing policy limitations and infrastructure challenges and offer potential solutions.

Development and Validation of a Machine Learning Model to Identify Patients Before Surgery at High Risk for Postoperative Adverse Events

Identifying patients at high risk of adverse outcomes prior to surgery may allow for interventions associated with improved postoperative outcomes; however, few tools exist for automated prediction. To evaluate the accuracy of an automated machine-learning model in the identification of patients at high risk of adverse outcomes from surgery using only data in the electronic health record. This prognostic study was conducted among 1 477 561 patients undergoing surgery at 20 community and tertiary care hospitals in the University of Pittsburgh Medical Center (UPMC) health network. The study included 3 phases: (1) building and validating a model on a retrospective population, (2) testing model accuracy on a retrospective population, and (3) validating the model prospectively in clinical care. A gradient-boosted decision tree machine learning method was used for developing a preoperative surgical risk prediction tool. The Shapley additive explanations method was used for model interpretability and further validation. Accuracy was compared between the UPMC model and National Surgical Quality Improvement Program (NSQIP) surgical risk calculator for predicting mortality. Data were analyzed from September through December 2021. Undergoing any type of surgical procedure. Postoperative mortality and major adverse cardiac and cerebrovascular events (MACCEs) at 30 days were evaluated. Among 1 477 561 patients included in model development (806 148 females [54.5%; mean [SD] age, 56.8 [17.9] years), 1 016 966 patient encounters were used for training and 254 242 separate encounters were used for testing the model. After deployment in clinical use, another 206 353 patients were prospectively evaluated; an additional 902 patients were selected for comparing the accuracy of the UPMC model and NSQIP tool for predicting mortality. The area under the receiver operating characteristic curve (AUROC) for mortality was 0.972 (95% CI, 0.971-0.973) for the training set and 0.946 (95% CI, 0.943-0.948) for the test set. The AUROC for MACCE and mortality was 0.923 (95% CI, 0.922-0.924) on the training and 0.899 (95% CI, 0.896-0.902) on the test set. In prospective evaluation, the AUROC for mortality was 0.956 (95% CI, 0.953-0.959), sensitivity was 2148 of 2517 patients (85.3%), specificity was 186 286 of 203 836 patients (91.4%), and negative predictive value was 186 286 of 186 655 patients (99.8%). The model outperformed the NSQIP tool as measured by AUROC (0.945 [95% CI, 0.914-0.977] vs 0.897 [95% CI, 0.854-0.941], for a difference of 0.048), specificity (0.87 [95% CI, 0.83-0.89] vs 0.68 [95% CI, 0.65-0.69]), and accuracy (0.85 [95% CI, 0.82-0.87] vs 0.69 [95% CI, 0.66, 0.72]). This study found that an automated machine learning model was accurate in identifying patients undergoing surgery who were at high risk of adverse outcomes using only preoperative variables within the electronic health record, with superior performance compared with the NSQIP calculator. These findings suggest that using this model to identify patients at increased risk of adverse outcomes prior to surgery may allow for individualized perioperative care, which may be associated with improved outcomes.

Prediction of COVID-19 Patients' Emergency Room Revisit using Multi-Source Transfer Learning.

The coronavirus disease 2019 (COVID-19) has led to a global pandemic of significant severity. In addition to its high level of contagiousness, COVID-19 can have a heterogeneous clinical course, ranging from asymptomatic carriers to severe and potentially life-threatening health complications. Many patients have to revisit the emergency room (ER) within a short time after discharge, which significantly increases the workload for medical staff. Early identification of such patients is crucial for helping physicians focus on treating life-threatening cases. In this study, we obtained Electronic Health Records (EHRs) of 3,210 encounters from 13 affiliated ERs within the University of Pittsburgh Medical Center between March 2020 and January 2021. We leveraged a Natural Language Processing technique, ScispaCy, to extract clinical concepts and used the 1001 most frequent concepts to develop 7-day revisit models for COVID-19 patients in ERs. The research data we collected were obtained from 13 ERs, which may have distributional differences that could affect the model development. To address this issue, we employed a classic deep transfer learning method called the Domain Adversarial Neural Network (DANN) and evaluated different modeling strategies, including the Multi-DANN algorithm (which considers the source differences), the Single-DANN algorithm (which doesn't consider the source differences), and three baseline methods: using only source data, using only target data, and using a mixture of source and target data. Results showed that the Multi-DANN models outperformed the Single-DANN models and baseline models in predicting revisits of COVID-19 patients to the ER within 7 days after discharge (median AUROC = 0.8 vs. 0.5). Notably, the Multi-DANN strategy effectively addressed the heterogeneity among multiple source domains and improved the adaptation of source data to the target domain. Moreover, the high performance of Multi-DANN models indicates that EHRs are informative for developing a prediction model to identify COVID-19 patients who are very likely to revisit an ER within 7 days after discharge.

Pretreatment pain predicts perineural invasion in patients with head and neck squamous cell carcinoma.

Perineural invasion (PNI) in head and neck cancer (HNC) is a distinct pathological feature used to indicate aggressive tumor behavior and drive treatment strategies. Our study examined the prevalence and predictors of PNI in HNC patients stratified by tumor site. A retrospective analysis of head and neck squamous cell carcinoma (HNSCC) patients who underwent surgical resection at the University of Pittsburgh Medical Center between 2015 and 2018 was performed. Pretreatment pain was assessed at least 1 week before surgery using the Functional Assessment of Cancer Therapy-Head and Neck (FACT-H&N). Demographics, clinical characteristics, and concomitant medications were obtained from medical records. Patients with cancers at the oropharynx and non-oropharynx (i.e., cancer at oral cavity, mandible, larynx) sites were separately analyzed. Tumor blocks were obtained from 10 patients for histological evaluation of intertumoral nerve presence. A total of 292 patients (202 males, median age = 60.94 ± 11.06) were assessed. Pain and PNI were significantly associated with higher T stage (p < 0.001) and tumor site (p < 0.001); patients with non-oropharynx tumors reported more pain and had a higher incidence of PNI compared to oropharynx tumors. However, multivariable analysis identified pain as a significant variable uniquely associated with PNI for both tumor sites. Evaluation of nerve presence in tumor tissue showed 5-fold higher nerve density in T2 oral cavity tumors compared to oropharyngeal tumors. Our study finds that PNI is associated with pretreatment pain and tumor stage. These data support the need for additional research into the impact of tumor location when investigating targeted therapies of tumor regression.

Longitudinal Analysis of Caregiver Burden in Head and Neck Cancer

Despite the critical role of caregivers in head and neck cancer (HNC), there is limited literature on caregiver burden (CGB) and its evolution over treatment. Research is needed to address evidence gaps that exist in understanding the causal pathways between caregiving and treatment outcomes. To evaluate the prevalence of and identify risk factors for CGB in HNC survivorship. This longitudinal prospective cohort study took place at the University of Pittsburgh Medical Center. Dyads of treatment-naive patients with HNC and their caregivers were recruited between October 2019 and December 2020. Eligible patient-caregiver dyads were 18 years or older and fluent in English. Patients undergoing definitive treatment identified a caregiver as the primary, nonprofessional, nonpaid person who provided the most assistance to them. Among 100 eligible dyadic participants, 2 caregivers declined participation, resulting in 96 enrolled participants. Data were analyzed from September 2021 through October 2022. Participants were surveyed at diagnosis, 3 months postdiagnosis, and 6 months postdiagnosis. Caregiver burden was evaluated with the 19-item Social Support Survey (scored 0-100, with higher scores indicating more support), Caregiver Reaction Assessment (CRA; scored 0-5, with higher scores on 4 subscales [disrupted schedule, financial problems, lack of family support, and health problems] indicating negative reactions, and higher scores on the fifth subscale [self-esteem] indicating favorable influence); and 3-item Loneliness Scale (scored 3-9, with higher scores indicating greater loneliness). Patient health-related quality of life was assessed using the University of Washington Quality of Life scale (UW-QOL; scored 0-100, with higher scores indicating better QOL). Of the 96 enrolled participants, half were women (48 [50%]), and a majority were White (92 [96%]), married or living with a partner (81 [84%]), and working (51 [53%]). Of these participants, 60 (63%) completed surveys at diagnosis and at least 1 follow-up. Of the 30 caregivers, most were women (24 [80%]), White (29 [97%]), married or living with a partner (28 [93%]), and working (22 [73%]). Caregivers of nonworking patients reported higher scores on the CRA subscale for health problems than caregivers of working patients (mean difference, 0.41; 95% CI, 0.18-0.64). Caregivers of patients with UW-QOL social/emotional (S/E) subscale scores of 62 or lower at diagnosis reported increased scores on the CRA subscale for health problems (UW-QOL-S/E score of 22: CRA score mean difference, 1.12; 95% CI, 0.48-1.77; UW-QOL-S/E score of 42: CRA score mean difference, 0.74; 95% CI, 0.34-1.15; and UW-QOL-S/E score of 62: CRA score mean difference, 0.36; 95% CI, 0.14-0.59). Woman caregivers had statistically significant worsening scores on the Social Support Survey (mean difference, -9.18; 95% CI, -17.14 to -1.22). The proportion of lonely caregivers increased over treatment. This cohort study highlights patient- and caregiver-specific factors that are associated with increased CGB. Results further demonstrate the potential implications for negative health outcomes for caregivers of patients who are not working and have lower health-related quality of life.

XRT-04. THE ROLE OF STEREOTACTIC RADIOSURGERY FOR MENINGIOMAS IN PEDIATRIC PATIENTS

Abstract Despite the generally benign and slow-growing nature, meningiomas in pediatric populations often exhibit more aggressive behavior. Stereotactic radiosurgery (SRS) is a non-invasive modality that is widely used as an upfront or adjuvant treatment for patients with meningiomas. Due to the rarity of meningiomas in pediatric patients, the outcomes of SRS for childhood meningiomas have never been described. From 1987 to 2022, a total of 2,030 patients with meningiomas underwent SRS at the University of Pittsburgh Medical Center. Nine patients (2 female; incidence rate = 0.44%) with a median age of 14.5 years (range, 9—17 years) at SRS were identified. Three patients (33.3%) had a history of total body irradiation. Seven patients (77.8%) were asymptomatic at the time of SRS presentation. Seven patients (77.8%) had a pre-SRS craniotomy; two patients (22.2%) had pre-SRS fractionated radiotherapy. The median cumulative tumor volume was 3.65 cc (range, 0.01—11.84 cc). The median margin dose prescribed was 14 Gy (range, 11—25 Gy). The median follow-up time was 107 months (range, 24—194 months). Five patients (55.6%) experienced tumor progression at a median of 25 months (range, 13—161 months). Local tumor control at 1-, 5-, and 10 years was 100%, 55.56%, and 55.56%, respectively. After local tumor progression, 2 patients received repeat SRS, 2 patients had surgical resection and 1 was placed on hydroxyurea for an unresectable tumor. At the date of the last clinical follow-up, 5 patients (55.6%) were still alive while 4 patients (44.4%) expired due to the progression of intracranial disease. 1-, 5-, and 10-year survival was 100%, 88.89%, and 64.81%, respectively. Childhood meningioma is a rare tumor that exhibits aggressive behaviors. SRS should be considered as an upfront or adjuvant treatment option in the multi-modal management of childhood meningiomas for optimal tumor control and maximal quality of life preservation.

Predicting Severe Complications from Cytoreductive Surgery with Hyperthermic Intraperitoneal Chemotherapy: A Data-Driven, Machine Learning Approach to Augment Clinical Judgment.

CRS-HIPEC provides oncologic benefit in well-selected patients with peritoneal carcinomatosis; however, it is a morbid procedure. Decision tools for preoperative patient selection are limited. We developed a risk score to predict severity of 90day complications for cytoreductive surgery with hyperthermic intraperitoneal chemotherapy (CRS-HIPEC). Adults who underwent CRS-HIPEC at the University of Pittsburgh Medical Center (March 2001-April 2020) were analyzed as part of this study. Primary endpoint was severe complications within 90days following CRS-HIPEC, defined using Comprehensive Complication Index (CCI) scores as a dichotomous (determined using restricted cubic splines) and continuous variable. Data were divided into training and test sets. Several machine learning and traditional algorithms were considered. For the 1959 CRS-HIPEC procedures included, CCI ranged from 0 to 100 (median 32.0). Adjusted restricted cubic splines model defined severe complications as CCI > 61. A minimum of 20 variables achieved optimal performance of any of the models. Linear regression achieved the highest area under the receiving operator characteristic curve (AUC, 0.74) and outperformed the NSQIP Surgical Risk calculator (AUC 0.80 vs. 0.66). Factors most positively associated with severe complications included peritoneal carcinomatosis index score, symptomatic status, and undergoing pancreatectomy, while American Society of Anesthesiologists 2 class, appendiceal diagnosis, and preoperative albumin were most negatively associated with severe complications. This study refines our ability to predict severe complications within 90days of discharge from a hospitalization in which CRS-HIPEC was performed. This advancement is timely and relevant given the growing interest in this procedure and may have implications for patient selection, patient and referring provider comfort, and survival.

Management and treatment of patients with stage III unresected NSCLC at UPMC: A quality initiative by Integra Connect Precision Q.

e20588 Background: The PACIFIC trial led to improved OS in patients (pts) with unresectable stage 3 NSCLC who received durvalumab (durva) post chemoradiation (CRT).1 In 2019, the University of Pittsburgh Medical Center (UPMC) and IC initiated a QI to improve immunotherapy (IO) use post CRT. A baseline analysis was done to assess scanning and treatment patterns post-CRT, and duration of treatment (DOT) with durva. Drawing on this analysis, IC and UPMC conducted a focus group to address gaps in care and initiated improvements. This real-world study aims to assess the change in IO use post-CRT, the time to scan (TTS), time to treatment (TTT), and duration of treatment (DOT) compared to baseline. The DOT of durva was compared between pts initiated &lt;14 days of CRT and those who started ≥14 days. Methods: From the UPMC and IC Precision Q real-world databases, we identified 555 pts with stage 3 unresectable NSCLC for manual chart abstraction. Pts received CRT between 2/16/2018 and 12/31/2021. TTS was calculated from the last day of CRT to the 1st scan and the TTT obtained from the last day of CRT to the 1st durva dose. The baseline period was from 2/16/2018 to 8/31/2020 and the post-intervention stage from 9/1/2020 to 12/31/2021 based on CRT start date. Pts were excluded if they had no scan data, a 1st scan ≥90 days post CRT (TTS analyses), a duration of zero days (DOT analyses) or a TTT &gt;120 days (TTT analyses). At least 1 yr of follow up was required for DOT comparisons. Descriptive analyses were used. A two-tailed 2 sample z-test was used for proportions and the nonparametric Wilcoxon rank sum test was used for TTS, TTT and DOT comparisons. Results: The baseline pts were 46% female with an average age of 67.7 years and the post-intervention were 47% female with an average age of 69.4 years. The rate of durva use post CRT in pts who did not progress increased from 66% (n =128/194) at baseline to 83% (n =96/116) post intervention (p &lt; 0.01). The baseline median TTS decreased from 22.5 days (n=174) to 20 days (n=121) post intervention (p &lt;0.05), and median TTT decreased from 35 days (n=121) to 28 days (n=90) (p &lt; 0.05). The median DOT increased from 6 (n=127) to 8.2 months (n=76) (p &gt;0.05). During the study period, 8% (n =17/223) of pts started durva &lt; 14 days post CRT and 92% started ≥14 days (n =206/223). The median DOT in the &lt; 14 days group was 10.8 months versus 6.2 months in ≥14 days (p &gt;0.05). The median overall survival (OS) was 36.5 months in the &lt; 14 days vs. 27.8 months in the ≥14 days group (p &gt; 0.05). Conclusions: QIs in NSCLC are effective tools that can improve appropriate durva utilization, TTT and DOT in eligible pts. Practices that assess their clinical performance can identify gaps in care and act. In this study, a superior OS, though lacking statistical significance, was seen in the group that started IO &lt;14 days post CRT compared to ≥14 days. Further studies with larger sample size, are needed to validate these findings.

Understanding testing for and treatment of homologous recombination repair (HRR)-mutated metastatic castrate-resistant prostate cancer (mCRPC): A quality initiative by Integra Connect (IC) PrecisionQ.

e17052 Background: In May of 2020, the approval of two Poly (ADP-ribose) polymerase inhibitors (PARPi) changed the landscape for treatment among patients (pts) with HRR mutations (HRRm) in mCRPC. The University of Pittsburgh Medical Center (UPMC) conducted a quality initiative (QI) with IC PrecisionQ to assess the rates of HRR testing, HRR positivity, and study the rates of PARPi use in HRRm patients with mCRPC. This QI included clinical interventions to optimize testing and treatment in these patients. We aim to show the improvement in HRR testing and rates of treatment with PARPi after implementation of the QI, in patients with HRRm mCRPC at UPMC. Methods: Using the de-identified IC PrecisionQ database, 223 UPMC pts with mCRPC were selected by medical chart curation as baseline group. Patients in the baseline group had received treatment between 5/1/2020 to 9/30/2021 that were confirmed as mCRPC. The baseline data was reviewed with UPMC clinical leadership, discussed among a physician roundtable, and HRR testing rates were shared among providers. Post-baseline, we evaluated 152 mCRPC pts who started a new line of therapy between 1/1/2022 to 12/31/2022 to assess HRR testing rates and use of PARPi in eligible mCRPC pts. The presence of one or more HRR genes was used to determine testing rates. Descriptive analyses were performed and a two-proportion z-test was used for comparisons of HRR testing in patients with mCRPC and treatment with a PARPi in HRRm mCRPC pts. Results: The average age at diagnosis was 67 years, 47% of patients had a Gleason score of &gt; = 8, and 62% were stage IV. The rate of HRR testing at baseline was 52% (116 of 223). The results post baseline and intervention showed an increase in HRR testing to 69% (105 of 152) of mCRPC patients who started a new line of therapy in 2022 (p &lt; 0.01). At baseline, 60% (21 of 35) of HRRm pts were treated with a PARPi after prior NHA treatment compared to 68% (19 of 28) post baseline and intervention (p &gt; 0.05). Among all HRR mCRPC tested pts, there was a 30.9% (84/272) positivity rate, of which 11.4% were BRCA1/2 positive (30/262) and 10.5% were ATM positive (26/247). Not all patients tested for HRR were tested for all genes. Conclusions: This quality initiative to improve testing and identification of HRR mutations in patients with mCRPC led to a 33% relative increase in HRR testing. Quality initiative programs are critical to transforming cancer care and moving the field closer toward individualized, precision medicine.

Association of patient and tumor characteristics with outcomes in young patients with head and neck squamous cell carcinoma (HNSCC).

e18054 Background: HNSCC is increasingly being diagnosed in a younger population, with a variable prognosis. While studies have evaluated outcomes in young compared to older patients, evaluation of prognostic factors specifically in younger patients, that may help guide treatment decisions, is needed. Therefore, we conducted a retrospective study to investigate the association of patient and tumor characteristics with disease specific survival (DSS) in HNSCC patients ≤45 years old (yo). Methods: Patients treated at the University of Pittsburgh Medical Center from January 1, 2007 to June 30, 2020, that were ≤45 yo at time of diagnosis with HNSCC without distant metastatic disease, were included. Those who received treatments at other institutions and who had no follow-up after index treatments were excluded. Patient demographics, social history, tumor pathology and treatment course were collected. The primary endpoint of DSS was defined as the interval between the time of pathologic diagnosis and the time of cancer-related death. Proportional hazards regression was applied separately for surgically and non-surgically treated patients. A multivariate analysis was conducted for surgically treated patients only. Results: 230 patients were included. Median age was 42, 73% male, 45% oral cavity primary, 35% oropharynx (89% Human papillomavirus positive), 14% larynx, 69% were smokers, and 73% received combined modality therapy. 35% of patients had a recurrence with the majority locoregional only (61%). Median DSS was not reached. The 5 year DSS was 73% (95% CI 67% to 80%). There was no difference in DSS between surgically treated (n = 169) and non-surgically treated (n = 61, p = 0.94) patients. For surgically treated patients, univariate analysis showed significantly worse DSS in those with higher pathologic T and N stages, positive margins, perineural invasion (PNI), extranodal extension (ENE), and negative HPV status. For non-surgically treated patients, only negative HPV status was associated with significantly worse DSS (p = 0.034). In the multivariate analysis for surgical treated patients, which included all significant characteristics from univariate analysis, pathologic T and N stage, positive margins, (HR 4.92, 95% CI 1.98-12.25, p = 0.006) and PNI (HR 2.66, 95% CI 1.17-6.08, p = 0.02) were significantly associated with worse DSS. Conclusions: Pathologic T and N stage, positive margins, and PNI are independently associated with significantly worse DSS in surgically treated HNSCC patients aged ≤45. Compared with classic prognostic factors seen in the overall population with HNSCC, PNI may be a uniquely strong prognostic factor in younger-aged HNSCC patients.

Association of Elevated Glycated Hemoglobin (HbA1c) in COVID-19 Patients Admitted to the Intensive Care Unit and Their Clinical Outcomes.

Aim The study aimed to collect retrospective data to investigate the association between elevated glycated hemoglobin (HbA1c) levels and clinical outcomes in COVID-19 patients admitted to the ICU, including in-hospital mortality and 90-day mortality. Methods This is an observational retrospective study using electronic health records of patients with diabetes admitted to the ICU with COVID-19 across the University of Pittsburgh Medical Center (UPMC) in Central PA Hospitals. Our retrospective analysis was performed on patients admitted to the ICU between May 1st, 2021, to May 1st, 2022. The HbA1clevel obtained within three months before their admission was evaluated and stratified to show their association with clinical outcomes, including in-hospital mortality and 90-day mortality. Additionally, the need for insulin drip and ICU and hospital length of stay were compared among these patients. Results We analyzed 384 patients, which were distributedin three groups. The majority of the patients (183 patients or 47.66%) had an HbA1c below 7%, 113 patients (29.43%) had an HbA1c between 7-9%, and 88 patients (22.92%) had an HbA1c above 9%. The group with an HbA1c<7% had a mortality rate of 54.1% during the hospital stay, with a median stay of 13 days.The patients with an HbA1c between 7-9% had a higher mortality rate of 65.49% with a median stay of 12 days. The patients with HbA1c>9% had a mortality rate of 43.18% with a median stay of 11.5 days. Conclusion This retrospective studyfound that there was no linear association between higher HbA1c levels and a higher risk of mortality during hospitalization. The 90-day mortalityrate was not statistically different among the three HbA1c groups.The need for insulin drip was higher in patients with higher HbA1c levels. The majority of patients in all three groups were classified as low-risk based on their BMI, and there were no significant differences in the distribution of patients across BMI categories in the HbA1c groups.

Herpes zoster ophthalmicus: frequency and risk factors for developing uncommon ocular manifestations

ObjectifDéterminer la fréquence de l'herpès zoster ophtalmique (HZO), communément appelé « zona ophtalmique », et évaluer les facteurs de risque d'apparition de manifestations oculaires peu fréquentes de l'HZO confirmé par analyse de laboratoire. NatureÉtude de cohorte rétrospective. MéthodesLa fréquence de l'HZO, comparativement à l'ensemble des cas d'herpès zoster, a été calculée d'après les codes de la Classification internationale des maladies attribués aux patients examinés au centre médical de l'Université de Pittsburgh entre le 1er janvier 2004 et le 31 octobre 2021. Nous avons en outre recueilli les données démographiques et cliniques des patients atteints de zona ophtalmique chez lesquels la présence du virus varicelle-zona a été objectivée par la réaction en chaîne de la polymérase (PCR) entre le 1er janvier 2011 et le 31 décembre 2020. RésultatsLa fréquence des cas d'HZO diagnostiqués entre 2004 et 2021 chez les patients de tous âges se chiffrait à 4,2 %, le taux annuel étant de 2,7 % à 6,7 %. Les taux ont systématiquement augmenté de 2,9 % par année entre 2012 et 2021. Après la mise en marché du vaccin à virus vivant contre le zona en 2008, la fréquence des diagnostics d'HZO a diminué de 5,1 % entre 2008 et 2012 chez les patients de 60 ans et plus. Dans un échantillon de 50 cas d'HZO diagnostiqué par PCR, 62 % des sujets avaient des manifestations oculaires fréquentes sur le plan clinique, dont la plupart étaient des cas de kératite (n = 13) et d'uvéite antérieure (n = 10). Enfin, la majorité des manifestations peu fréquentes du zona ophtalmique étaient des cas de nécrose rétinienne aiguë (n = 15; 38 %), lesquels qui étaient significativement plus susceptibles de se produire en cas d'immunodépression (rapport de cotes non ajusté : 4,55; intervalle de confiance à 95 % : 1,29–13,83). ConclusionsEntre 2004 et 2021, la fréquence globale d'HZO était de 4,2 %, pourcentage qui a augmenté tous les ans depuis 2012. Les manifestations oculaires peu fréquentes du zona ophtalmique diagnostiqué par PCR consistent surtout en la nécrose rétinienne aiguë et sont plus susceptibles de se produire chez des patients immunodéprimés.

A Reflection on Nursing in Neurosurgery

Around the world in many countries, May is “Nurses Month” and May 12, International Nurses Day (Florence Nightingale's birthday). Thinking about this a few months ago led me to think how little we spend in the pages of this journal and others, on the value of nursing in neurosurgery. The cover this month highlights the partnership of nurses and neurosurgeons from years past and the present. My dear friend Dan Leksell shared this image of his father Lars Leksell performing air ventriculography, while his nurse assists with the imaging. Can we even think about the performance of neurosurgery without nursing support, whether in the office, the research environment, the operating room, the hospital wards, or the rehabilitation centers that often help our patients in their recoveries? Do we just take for granted that complex and challenging surgeries will be followed by excellent nursing care that lasts all night, the next day, and the day after that? Nurses talk to our patients when we are not there, reiterate our messages, and clarify and answer patient's questions. They become as familiar with new surgical techniques as we do, to understand their indications and potential outcomes. Then they work to make those outcomes possible. In the surgical suite, they must become familiar with new devices and do so expeditiously to meet our high standards. It is truly a wonder how over the course of their own career they evolve as we evolve, as our field evolves. Neurosurgical nursing is its own calling. As for any of us, the path is different from person to person. To become involved in the care of patients with challenging and debilitating neurological disorders, in one or more different clinical settings, is a commitment toward excellence. I personally think of our work as more a vocation than a career. We are so fortunate to have nurses in the practice setting who help us bring patients and families to an understanding of their own medical challenges and to help them navigate what is assuredly an anxious time. Their calm, unhurried demeanor can help them through a potentially frustrating interaction with the health care system, lessening worry and sleepless nights, at least to some degree. Nurses on the hospital wards provide the surgical aftercare of recovery, hopefully to a prompt return home, but sometimes toward an extended care stay that may develop into more personal contacts with patients, friends, and family. The hands-on, personal interactions involved are truly special and go far beyond the expectations of a “job.” In the surgical suite, our nurses have the same commitment to professionalism, communication, care, and error reduction as we do. This is displayed time and again, each time a new patient enters the room. And last, a personal note. Many years ago now, I married a neurosurgical nurse who had worked on the neurosurgery floor at the Toronto General Hospital and then in the operating room there. She later worked in the OR at the University of Pittsburgh Medical Center. She understood the field, the disorders we treat, and the personalities involved. Her nursing skills were also crucial to our family, including structure, planning, fearlessness, and tenderness. As my dear friend John Golfinos would say, “you can't spell neurosurgery without n-u-r-s-e.” I encourage you to think about the nurses you work with, in any of the different settings, and know that the wonderful science in these journal pages could not have happened without them. Douglas Kondziolka, MD, MSc Editor-in-Chief, Neurosurgery Publications New York, New York, USA

PO-02-021 DIRECTION OF REPOLARIZATION IN BASELINE ECG IS PREDICTIVE OF RESPONSE TO CARDIAC RESYNCHRONIZATION THERAPY

Cardiac resynchronization therapy (CRT) is a cornerstone of treatment for advanced heart failure, but not all patients derive equal benefit from this intervention. Current approaches of using electrocardiographic and echocardiographic parameters to identify likely CRT responders demonstrate poor predictive ability. We seek to identify electrocardiographic parameters that can improve prediction of benefit with CRT This was a retrospective analysis of patients who underwent CRT-D placement at University of Pittsburgh Medical Center from 2014-2020. Patients were excluded if pre-implantation EKG or pre- and post-implantation echocardiograms were not available. Each patient’s immediate pre-implantation ECG was processed using the CineECG software (EKG Excellence) to obtain the mean temporospatial isochrone of ventricular activation. 192 features quantifying the activation and recovery pathways were computed for each patient. Response to CRT was defined as an improvement in ejection fraction (EF) of 5% on follow-up transthoracic echocardiogram 3-12 months after CRT implantation. Forward sequential feature selection was used to identify the features of highest importance based on negative logistic loss. 276 patients were included in this analysis (36% female, mean age 69 years, mean pre-procedural EF 27%). The 10 most important CineECG features were identified. When these features were used in a multivariate logistic regression, three features were independently predictive of CRT response: left bundle branch block (LBBB); leftward direction of late repolarization potentials (i.e., Tpeak - Tend rotated toward LV in the horizontal plane); and location of center-of-mass of terminal cardiac activation. Compared to CRT responders (n=169, 62%), non-responders were more likely to have abnormal direction of late repolarization potentials in both LBBB group (-0.34±0.30 vs. -0.45±0.30 vs, p=0.027) and RBBB group (0.25±0.51 vs. 0.61±0.27, p=0.031). In the latter group, direction of later repolarization potentials correlated with magnitude of change in post-procedural EF (r=0.23, p=0.016). Our findings confirm that the presence of LBBB is predictive of response to CRT. In addition, our data suggests that in patients with RBBB, a more leftward direction of late repolarization potentials is independently predictive of CRT response. Further studies into repolarization dynamics as a predictor of CRT response in RBBB are warranted.

CE-452776-1 EARLY INITIATION OF ARNI AND SGLT2I IS ASSOCIATED WITH DELAYED ATRIAL FIBRILLATION IN HEART FAILURE PATIENTS

Early use of mineralocorticoid receptor antagonists (MRA), sodium-glucose co-transporter inhibitors (SGLT2i) and angiotensin receptor-neprilysin inhibitors (ARNI) have shown benefit in major trials for heart failure patients and in other studies for recurrence of atrial fibrillation (AF) after ablation. Whether these medications and their resulting improved remodeling are associated with delayed onset of AF in a real-world heart failure cohort is unknown. We aimed to define the clinical phenotype of heart failure patients receiving early use (within 6 months) of MRA, SGLT2i, and ARNIs. We also investigated the association between use of these medications and time to first hospitalization with AF. We analyzed the electronic health records of the University of Pittsburgh Medical Center, a large multi-hospital system, between 2010 and 2022. Differences between baseline demographics, cardiovascular risk factors, and comorbidity burden were evaluated with t tests. Adjusted Cox proportional hazards models were performed to evaluate the association between early use of the medications and onset of AF for all heart failure patients and subgroups with reduced (HFrEF) and preserved (HFpEF) ejection fractions. Among 57689 patients, 2834 (5%) developed AF in the follow up period (median 3.3 years, IQR 4.7). A total 12015 (21%) had early initiation of MRA, 2259 (4%) SGLT2i, and 2909 (5%) ARNI. Patients receiving these medications were significantly more male with more commercial insurance, initial evaluation by cardiologists, and lower CHA2DS2VASC and Elixhauser comorbidity scores (table 1). In multivariate Cox models, delayed onset of AF hospitalization was associated with both use of early SGLT2i (HR 0.56, 95% CI 0.39-0.80, p<0.01) and ARNI (HR 0.52, 95% CI 0.39-0.70, p<0.001) but not MRA (figure 1). This trend was echoed in HFrEF patients (SGLT2i HR 0.46, 95% CI 0.24-0.89, p<0.05; ARNI HR 0.59, 95% CI 0.42-0.84, p<0.01) but associations were not significant in HFpEF. In a large, real-world cohort of heart failure patients, early use of ARNI and SGLT2i is significantly associated with delayed onset of AF. This association was driven largely by the HFrEF cohort. While only 5% of patients developed AF, early use of ARNI and SGLT2i in all indicated heart failure patients may decrease AF onset and morbidity in an already high-risk population.Tabled 1Table 1: Selected Baseline Characteristics of All Heart Failure Patients Receiving Early MRA, SGLT2i, and ARNI TherapyEarly Use of MRA-YESEarly Use of MRA-NOEarly Use of SGLT2i-YESEarly Use of SGLT2i-NOEarly Use of ARNI-YESEarly Use of ARNI-NOAge (mean±SD)67±14.172±13.865±12.671±14.065±13.571±14.0Male Sex (%)545061506550White Race (%)858987878488Commercial Insurance (%)242226222922Initial Evaluation by a Cardiologist (%)483947435540CHA2DS2VASC score (mean±SD)3.7±1.64.1±1.54.0±1.64.0±1.53.5±1.64.0±1.6Mean Elixhauser comorbidity score10.4119.210.91010.9Prior coronary revascularization (%)423346354735Confirmed HFrEF (%)301433165215Confirmed HFpEF (%)17231522423Early use of MRA (%)NANA48205619Early use of SGLT2i (%)93NANA273Early use of ARNI (%)14-NS3-NS35-NS4-NSNANANote: All pairs noted above are significant with p<0.01 except for those labeled "-NS", which are not significant. Open table in a new tab

PC7. Competency-based, Time-variable Resident Education: A Programmatic Update

PURPOSE: In 2018, the University of Pittsburgh Medical Center (UPMC) adopted a Competency-Based, Time-Variable (CBTV) model for plastic surgery training, the first of its kind in the United States. This educational model emphasizes performance and skill mastery, rather than time-based benchmarks, as the measure by which residents are deemed fit for graduation. METHODS: Starting in the PGY1 year, each resident is evaluated regularly via oral exam to assess knowledge of plastic surgery principles. Resident self-assessment of technical competency is directly compared to faculty assessment utilizing the Zwisch Rating Scale, a validated scoring system in rating resident autonomy. Our department has reorganized its rotation schedule to shift Chief-level rotations from the PGY6 year to the PGY5 year. RESULTS: UPMC matched its first CBTV resident class in 2018. Of this three-resident cohort, two residents (66%) were deemed competent for graduation after five clinical years of training. One resident elected to defer graduation until after six years of training due to fellowship application timing constraints. The other resident elected to pursue a year of research and subsequently graduate after five clinical years of training. Thus, all three residents will graduate after six years at UPMC, but with customized experiences based on their specific interests and educational needs. CONCLUSION: The CBTV model has increased the flexibility of our training program to accommodate individualized educational needs of each resident, allowing them variable rotation schedules to address surgical deficiencies as well as to further develop their clinical interests. This is a paradigm shift in surgical education.

Single Nucleotide Polymorphism rs9277336 Controls the Nuclear Alpha Actinin 4-Human Leukocyte Antigen-DPA1 Axis and Pulmonary Endothelial Pathophenotypes in Pulmonary Arterial Hypertension.

Background Pulmonary arterial hypertension (PAH) is a complex, fatal disease where disease severity has been associated with the single nucleotide polymorphism (SNP) rs2856830, located near the human leukocyte antigen DPA1 (HLA-DPA1) gene. We aimed to define the genetic architecture of functional variants associated with PAH disease severity by identifying allele-specific binding transcription factors and downstream targets that control endothelial pathophenotypes and PAH. Methods and Results Electrophoretic mobility shift assays of oligonucleotides containing SNP rs2856830 and 8 SNPs in linkage disequilibrium revealed functional SNPs via allele-imbalanced binding to human pulmonary arterial endothelial cell nuclear proteins. DNA pulldown proteomics identified SNP-binding proteins. SNP genotyping and clinical correlation analysis were performed in 84 patients with PAH at University of Pittsburgh Medical Center and in 679 patients with PAH in the All of Us database. SNP rs9277336 was identified as a functional SNP in linkage disequilibrium (r2>0.8) defined by rs2856830, and the minor allele was associated with decreased hospitalizations and improved cardiac output in patients with PAH, an index of disease severity. SNP pulldown proteomics showed allele-specific binding of nuclear ACTN4 (alpha actinin 4) protein to rs9277336 minor allele. Both ACTN4 and HLA-DPA1 were downregulated in pulmonary endothelium in human patients and rodent models of PAH. Via transcriptomic and phenotypic analyses, knockdown of HLA-DPA1 phenocopied knockdown of ACTN4, both similarly controlling cell structure pathways, immune pathways, and endothelial dysfunction. Conclusions We defined the pathogenic activity of functional SNP rs9277336, entailing the allele-specific binding of ACTN4 and controlling expression of the neighboring HLA-DPA1 gene. Through inflammatory or genetic means, downregulation of this ACTN4-HLA-DPA1 regulatory axis promotes endothelial pathophenotypes, providing a mechanistic explanation for the association between this SNP and PAH outcomes.

Intracranial Hemorrhage as a Marker for Mycotic Aneurysm in Patients With Infective Endocarditis.

Mycotic aneurysms are a significant complication of IE that can lead to intracranial hemorrhage (ICH) and occasionally require intervention before valve surgery. Digital subtraction angiography (DSA) remains the gold standard for the detection of mycotic aneurysms but is an invasive procedure with contrasted dye exposure. Our objective was to determine predictive factors for the presence of mycotic aneurysms in patients with known infective endocarditis (IE). IRB exemption was obtained from the University of Pittsburgh Medical Center. Patients with left-sided or right-sided endocarditis with patent foramen ovale (PFO) were identified. Records were retrospectively reviewed by an independent investigator. Patients were stratified by demographic information, diagnostic imaging, and clinical outcomes. Standard descriptive statistics were used for characterization. The χ2 analysis and Fisher exact test were used for categorical comparisons. The Student t test was used for group comparisons. 36 cases of left-sided or right-sided endocarditis with PFO undergoing evaluation for valve surgery were identified. Twenty-seven patients underwent CNS imaging, and 24 underwent DSA. Six patients were found to have ICH on CNS imaging. Mycotic aneurysms were identified by DSA in 5 patients (13.9%), and of these patients, 4 (80%) presented with ICH and 1 with ischemic stroke. Notably, the patient who presented with ischemic stroke then developed ICH on follow-up CT imaging. The negative predictive value of the absence of ICH on initial CNS imaging for the presence of mycotic aneurysm was 97.4%. Patients with left-sided and right-sided endocarditis with PFO and ICH on CNS imaging were more likely to have mycotic aneurysms detected through DSA than similar patients without ICH. Routine DSA screening for mycotic aneurysms may not be required in patients with endocarditis undergoing evaluation for valve surgery who present without ICH on CNS imaging.

Changes in body weight and glycemic control in association with COVID-19 Shutdown among 23,000 adults with type 2 diabetes.

To examine the association between COVID-19 Shutdown and within-subjects changes in body weight, body mass index (BMI), and glycemic parameters using electronic health record (EHR) data from 23,000 adults with type 2 diabetes (T2DM). Patients with T2DM with outpatient visit data on body weight, BMI, hemoglobin A1c (HbA1c), and blood glucose (≥ 2 measures before and after 3/16/2020) recorded in the EHR at the University of Pittsburgh Medical Center were included. A within-subjects analysis compared average and clinically significant changes in weight, BMI, HbA1c, and blood glucose during the year POST-Shutdown (Time 2-3) compared to the same interval during the PRE-Shutdown year (Time 0-1) using paired samples t-tests and the McNemar-Bowker test. We studied 23,697 adults with T2DM (51% female; 89% White; mean age = 66 ± 13years; mean BMI = 34 ± 7kg/m2; mean HbA1c = 7 ± 2% [53 ± 21.9mmol/mol]). Weight and BMI decreased during both the PRE- and POST-Shutdown intervals, but the changes were statistically smaller during the year POST-Shutdown relative to PRE (0.32kg and 0.11 units, p < 0.0001). HbA1c showed statistically greater improvements during the POST-Shutdown interval compared to PRE (- 0.18% [-2mmol/mol], p < 0.0001), but changes in glucose did not differ for the two intervals. Despite widespread discussion of weight gain in association with the COVID-19 Shutdown, study data showed no evidence of adverse effects of Shutdown on body weight, BMI, HbA1C, or blood glucose in a large sample of adults with T2DM. This information may help to inform future public health decision-making.