Browse
Recent Submissions
Item Types of Anesthetics and IV Adjuvants Used for Cesarean Section(2025-07-24) Mendoza, Francis; Traylor, Beth Ann; Dammann, Erin; Yeap, YarBackground/Objective: For cesarean sections, urgency, maternal medical history, and patient preferences determine the form of anesthesia administration. Spinal anesthesia is the most common form of anesthesia administration due to its speed, patient safety, and overall reliability. Conversely, general anesthesia is a less common option due to its higher risk of difficult intubation, aspiration, and the mother being deprived of witnessing the baby’s birth. Moreover, intravenous (IV) adjuvant medications, such as ketamine and propofol, are occasionally utilized to further provide analgesic and anesthetic effects. The objective of our study was to describe the different types of anesthesia and IV adjuvants used for cesarean sections. Methods: We conducted a retrospective chart review of all patients undergoing cesarean sections at Riley Maternity Tower from 1/01/25 to 3/31/25. Patients with additional procedures performed were excluded. Our exposures were the different types of anesthesia used, including spinal, epidural, combined spinal-epidural (CSE), dural puncture epidural (DPE), and general anesthesia. Additionally, our outcomes of interest were the type of anesthesia selected, the incidence of intravenous adjuvants, and operative times. Results: Of 278 patients undergoing a cesarean section, 160 patients (58%) received spinal anesthesia, 69 patients (25%) received epidural anesthesia, and 19 patients (7%) received general anesthesia. Additional intravenous adjuvants were required in 55 (20%) patients. Moreover, dexmedetomidine was the most frequently administered adjuvant (22%), followed by fentanyl (11%). Patients who required IV adjuvants had a longer mean operative time (63 min) compared to those who did not receive IV adjuvants (51 min). Conclusions: In this single-site observational study, spinal anesthesia was the most common procedure done in patients undergoing a cesarean section (58%). General anesthesia was used in a smaller proportion of patients (7%). Approximately 1 in 5 patients (55/278) required an intravenous adjuvant. Those requiring IV adjuvants tended to have longer operative times.Item Types of Epidural Analgesia for Labor Pain(2025-07-24) Mendoza, Francis; Traylor, Beth Ann; Dammann, Erin; Yeap, YarBackground/Objective: Neuraxial analgesia administered in the epidural space is an effective strategy in controlling labor pain in expecting mothers. Epidural analgesia can also facilitate the transition to surgical anesthesia in patients requiring an emergent cesarean section. However, epidural placement does come with a small risk of accidental dural puncture, or "wet tap,” which can manifest with post-dural puncture headaches. Per recent literature, for patients receiving epidural analgesia for labor pain, the incidence of “wet taps” is 0.5-1.5%. Our study aimed to assess the incidence of “wet taps” in patients undergoing epidural analgesia for labor pain. Methods: We conducted a retrospective chart review of all patients receiving epidural analgesia to control labor pain at Riley Maternity Tower from 1/01/25 to 3/31/25. Our exposures included the different varieties of epidural analgesia used, such as epidural analgesia, combined spinal-epidural (CSE) analgesia, and dural puncture epidural (DPE) analgesia. The primary outcome of interest was the occurrence of a “wet tap.” Results: Of the 338 patients who received epidural analgesia for labor pain, the most common type was epidural (88.46%), followed by DPE (8.88%) and CSE (2.66%). Among patients who received an epidural, 274 patients (81%) underwent vaginal delivery while 64 patients (19%) had a cesarean delivery. Overall, the incidence of a “wet tap” during epidural placement was 1%. Conclusions: In this single-site study, the incidence of “wet tap” in patients receiving epidural analgesia for labor pain was 1%, which is consistent with reports from prior studies. 19% of patients who received labor analgesia had cesarean deliveries.Item Clinician Perspectives on Eye-Related Content in Medical Curricula(2025-07-25) Bertram, Michael; Husmann, PollyThe medical school pre-clinical curriculum is vast and seemingly all encompassing. Though the first two years are geared towards learning what is necessary for Step 1 of boards, certain specialties often go overlooked in medical school curricula. One of the most underrepresented specialties is ophthalmology. Six providers across three specialties were interviewed regarding what they thought was currently being taught in medical school, what they think should be taught regarding the eyes, and some suggestions on how medical schools could improve their coverage of ophthalmology. These interviews were semi-structured and recorded using Zoom, they were transcribed, and thematic analysis determined themes among the providers’ answers. This study focused primarily on whether medical schools have allowed their curricula to overlook ophthalmology and how this can be mitigated. Though constant change of medical curricula is a good thing to keep material fresh, we have a responsibility as educators to ensure that all of the important material is being covered to help develop the most well-rounded physicians possible.Item Exploring Alzheimer's Disease Progression Through a Graph-based Retrieval System(2025-07-25) Rai, Aryan; Lee, Eun Hye; Jo, TaehoBackground and Objective: The cognitive decline spectrum includes cognitively normal (CN), mild cognitive impairment (MCI), and dementia stages. Alzheimer’s Disease (AD) is a progressive neurodegenerative disorder that predominantly causes dementia. Thus, identifying patients likely to progress to AD is clinically pertinent. The Alzheimer Disease Neuroimaging Initiative (ADNI) dataset features chronological imaging, biomarker levels, and cognitive scores for over 2,500 individuals across the cognitive decline spectrum. A current challenge in AD research is comprehensively utilizing the volume and variety of information in databases like ADNI. Knowledge graphs (KG) simplify information using points (nodes) and connections (edges). Retrieval systems extract information from a source. In this work, we developed a graph-based retrieval system that utilized ADNI data to classify patients progressing from CN to MCI or AD. Methods: The ADNI KG, containing 2,513 patient nodes, 15,497 visit nodes, and 1,135,912 measurement nodes, was processed using NetworkX. Our retrieval system was coded in python and performed Amyloid-Tau-Neurodegeneration (ATN) scoring for each patient in the KG using conventional criteria. Patient progression was determined by comparing diagnoses from the first 25% of visits with the last 25% of visits. The system’s ability to make predictions using cognitive, biomarker, and imaging data was validated on 200 randomly selected patients from the KG. Results: The system processed a KG with 1,153,922 nodes and 1,164,395 edges. For N=200 validation, the system removed 6 outliers, found 19 progressors (10 CN→MCI, 9 CN→AD), and achieved 76.7% accuracy when provided adequate biomarker data. Conclusion and Potential Impact: We present a graph-based retrieval system that can analyze large quantities of patient data, identify potential AD database errors, output patient-specific insights, and predict cognitive decline progression with reasonable accuracy. Future work will integrate this system with graph neural networks (GNN) and large language models (LLMs) to further improve its accuracy and clinical utility.Item Communication and Learning Improvement Model for Bedside Skills (CLIMBS): An AI-Based Feedback Model for the Objective Standardized Clinical Examination(2025-07-25) Hobson, Liam; Lutz, Chandler; Rodgers, DavidBackground: The Objective Standardized Clinical Examination (OSCE) is a tool designed to assess and provide feedback to healthcare students; however, feedback quality is a common point of frustration for students. Common concerns with post-OSCE feedback include that it is generally not timely, comprehensive, and/or individualized. In this retrospective, quasi-experimental pilot study, we created and implemented an artificial intelligence (AI) feedback model and assessed its ability to provide accurate and reliable OSCE feedback. Methods: Using ChatGPT-4o, we developed the Communication and Learning Improvement Model for Bedside Skills (CLIMBS) to reference a 16-metric OSCE rubric and provide feedback on OSCE transcripts derived from a custom workflow. Using a single, full-marks OSCE recording, we assessed CLIMBS by calculating transcription and feedback accuracy and examining feedback reliability against an artificially-generated ‘Poor Performance’ transcript and a manually corrected transcript. Results: The CLIMBS workflow exhibited 92.39% transcript accuracy and good inter-rater reliability (cosine similarity = 0.972 +/- 0.015; ICC2 = 0.744 (95% CI: [0.59, 0.88]). Overall OSCE scores assigned by CLIMBS to the unedited transcript (93.84; 95% CI [91.53, 96.17]) and the manually corrected transcript (87.69, 95% CI [81.37, 94.02]) were statistically different from the ‘Poor Performance’ transcript (78.08; 95% CI [71.59, 84.57]). The metric-based score pattern distribution of the manual transcript was similar to both the unedited and the ‘Poor Performance’ transcripts (cosine similarity = 0.890 and 0.819, respectively), while the unedited and ‘Poor Performance’ transcripts exhibited low overlap with each other (cosine similarity = 0.566). Conclusion: This pilot study demonstrated that an AI model can analyze and provide accurate summative feedback to OSCE recordings – differentiating good from poor performance. The accuracy and reliability of formative feedback require further study. With future testing to improve inter-rater reliability and examine formative feedback, CLIMBS has potential to improve the timeliness, comprehensiveness, and personalization of OSCE feedback.Item Lifetime Prevalence of Traumatic Brain Injury in the Previously Unhoused Population(2025-07) Sasic, Suki; Messmore, NikiIt is well described in the literature that the unhoused or previously unhoused population has a higher prevalence of traumatic brain injuries (TBI) as compared to the general population. Some of the various long-term consequences of TBIs are cognitive/functional impairments, attention deficits, and impulsive/risky behavior. Therefore, because a large proportion of the unhoused population may be experiencing the persistent challenges that stem from unidentified TBIs, identification and education of those involved in supportive services to the unhoused could result in more effective support for long-term housing success. This project compared rates of TBIs in two permanent supportive housing models that are run by the nonprofit Beacon Inc. Interviews were conducted according to the Ohio State University TBI ID tool. This screening test included several subcategories of TBI: occurrence of moderate or severe TBI (“worst”), occurrence of first severe TBI before age 20 (“first”), and occurrence of multiple TBIs within a short time period (“multiple”). Residents of the clustered site had a higher incidence of TBI (78%, n=32) than residents of the scattered sites (41%, n=17). Both groups had a higher prevalence of TBI than the nationwide average of 18.2% with p values of 0.02 and <0.001, respectively. Failure to diagnose TBIs, especially in the unhoused population, is common due to lack of healthcare accessibility and medical distrust. This gap in care may lead to underdiagnosis and therefore untreated, unacknowledged long-term consequences. Furthermore, those cognitive consequences have been shown to contribute to the cycle of homelessness by predisposing patients to future TBIs via assault, substance abuse, and falls, further hampering their ability to exit homelessness. Having an understanding of clients’ cognitive history can better inform case workers of their clients’ ability to live independently and perform daily tasks of living. Therefore, with this information, case managers may be able to better understand what supports are necessary for clients’ long-term housing success.Item Association Between Anxiety, Depression, and Opioid Use Disorder in Adult Surgical Patients: A Real-World Data Study(2025-07-25) Bowman, Zachary; Cook, Nolan; Quach, Tommy; Jordan, Kevin; Roscoe, MichaelBackground/Objective: The opioid crisis remains a central issue in public health. Prior research has explored how postoperative opioid prescribing contributes to the development of opioid use disorder (OUD). Pre-existing psychiatric conditions may influence postoperative misuse of opioid medications, but existing research on this association is limited in scope and generalizability. This study aimed to evaluate whether a diagnosis of depression and/or anxiety is associated with increased odds of OUD among adult patients undergoing common opioid-prescribing surgical procedures. Methods: In this retrospective case-control study, we used de-identified Electronic Health Record data from the IU School of Medicine-Evansville RWEdataLab (CRC/Sidus Insights) Psychiatric database. Adult patients aged 18-70 who underwent a common opioid-prescribing surgical procedure were identified using CPT codes. Psychiatric and OUD diagnoses were identified using ICD-10 codes. Patients were grouped based on the presence or absence of an OUD diagnosis. Odds ratios were calculated to assess the association between OUD and prior diagnosis of anxiety, depression, or both. Results: Among 18,440 patients who underwent a qualifying surgery, 653 were diagnosed with OUD. Of these, 288 had depression, 280 had anxiety, and 223 had both diagnoses, indicating substantial overlap between conditions. Patients with depression (OR: 1.91; 95% CI: 1.63-2.23), anxiety (OR: 1.26; 95% CI: 1.08–1.48), and both conditions (OR: 3.84; 95% CI: 3.24–4.54) had significantly higher odds of OUD compared to surgical patients with other psychiatric diagnoses. Conclusion and Clinical Implications: These findings suggest that adult surgical patients with a history of anxiety and/or depression have increased odds of developing OUD. This underscores the clinical importance of individualized pain management and enhanced perioperative support for patients with psychiatric comorbidities. Future research should explore these associations in broader and more diverse populations and evaluate interventions that integrate mental health screening into preoperative planning.Item Generation of CRISPR-Cas9 Engineered OPTNE478G Human Embryo Stem Cell Line for Investigation into Mitophagy Defects(2025-07-25) Tebbe, Maverick; Kathri, Saajid; Surma, Michelle; Das, ArupratanBackground: Optineurin (OPTN) is a mitophagy adaptor protein linking damaged, ubiquitinated mitochondria to autophagosomes for lysosomal degradation. The E50K and E478G OPTN mutations are associated with normal tension glaucoma and ALS, respectively. Currently, it is unknown whether the E50K mutation selectively affects retinal ganglion cells and the E478G mutation selectively affects motor neurons. Given OPTN’s role in mitophagy, investigating the mitophagy defects in human stem cell-differentiated (hRGCs) and induced motor neurons (iMNs) harboring these mutations provides an avenue to explore the cellular mechanisms of these diseases. Methods: This project generated a OPTN-E478G human embryonic stem cell (hESC) line through CRISPR-Cas9 gene editing. gRNA oligomers were annealed and cloned via transformation of DH5ɑ cells. Following sequencing, the gRNA-Cas9-GFP plasmid and donor plasmid with E478G insert were transfected into H7-WT-hESCs. The CRISPR-Cas9 system cut and repaired the double stranded break by homology-directed repair introducing the E478G mutation. GFP-positive colonies were isolated, expanded, and screened by PCR and restriction enzyme digestion. Western blot analysis of E50K and E478G mutants for hRGCs and iMNs were performed following introduction of the mitochondrial uncoupler CCCP, inducing mitophagy. Results: Insertion of the gRNA into the Cas9-GFP plasmid was confirmed by sequencing. The hESCs were successfully transfected with the E478G plasmid as confirmed by restriction enzyme digestion of SapI and BspHI. Isolation of this stem cell population is ongoing and will be screened with restriction enzymes prior to sequencing. Western blots of NBR1 showed increased levels in WT-hRGCs and iMNs (E50K and E50K corrected), but not in E50K-hRGCs, indicating impaired mitophagy. Conclusions: Our study reveals distinct mitophagy defects seen in the OPTN mutations E50K and E478G for the RGCs and iMNs. This project lays the groundwork for further studies, including live cell imaging, OPTN activation, and LC3b lipidation, to further define the implications of these mutations on mitophagyItem Is the Social Vulnerability Index Related to Outcomes in Pregnancies Complicated by Type 2 Diabetes?(2025-07-25) Kabele, Claudia Linczer; Scifres, ChristinaBackground: Type 2 diabetes (T2D) is associated with increased risk for adverse maternal and fetal outcomes. Using the Social Vulnerability Index (SVI), a measure developed by the CDC, studies show that women in areas of higher social vulnerability face challenges achieving recommended glycemic targets. We investigated whether pregnancies complicated by T2D and higher social vulnerability have increased adverse perinatal outcomes. Methods: We performed a retrospective cohort study looking at pregnancies complicated by T2D who delivered at Eskenazi and IU Health from January 2018 to December 2020. Demographic, medical history, and pregnancy outcome data were abstracted. We used the Census Geocoder Tool to match each patient's address with the respective census tract and SVI value. SVIs were divided into tertiles (< 0.492, 0.492 - 0.7792, > 0.7792). We compared demographic and perinatal outcomes using chi-square test and ANOVA as indicated across SVI tertiles. Results: The SVI was calculated for 349/371 (94%) participants. Those in the highest tertile of SVI were less likely to initiate prenatal care in the first trimester compared to those in the lowest tertiles (79 vs 70 vs 59%, p=0.005). HbA1c values were similar across SVI groups. The proportion of those with T2D who received prenatal care at Eskenazi Health increased across SVI tertiles (21 vs 34 vs 58%, p<0.001). Unexpectedly, birth weight, preterm birth, hypertensive disorders, and neonatal morbidity did not differ among groups. Conclusions: The SVI identifies individuals at increased risk for initiation of prenatal care after the first trimester. Rates of adverse outcomes were high among all individuals with T2D and did not vary significantly by SVI tertile. To better understand the impact of public health support provided by Eskenazi on perinatal outcomes among those with T2D and a high SVI, data abstraction from 2021 to 2024 is ongoing.Item A Regional Analysis of Air Quality and Respiratory Health Outcomes in the Northwest Indiana and Chicagoland Area(2025-07-24) Sullivan, Elizabeth; Muvuka, BarakaBackground: Northwest Indiana (NWI) houses three of seven coal-based steel factories in the US and other major industries that contribute to air pollution in NWI and Chicagoland. Pollutants such as ozone, PM2.5, and PM10 are associated with respiratory disease onset and exacerbation. The American Lung Association has assigned failing air pollution grades to most NWI and Chicagoland counties. While respiratory diseases are prevalent locally, there is limited research on air pollution’s role. This study investigated air pollution exposures and associations with respiratory outcomes in NWI and Chicagoland. Methods: This retrospective study analyzed data from the Chicago Multiethnic Prevention and Surveillance Study (COMPASS), with social determinants of health, demographic, and environmental data for adults from 2013-2023. Additional air quality data were obtained from the Environmental Protection Agency and Washington University in St. Louis. Descriptive, bivariate (Chi-Square; Mann Whitney U, Kruskal Wallis, Spearman Rank), and multivariate (Binary Logistic Regression; Linear Regression, p<0.05) analyses were conducted using SPSS 31.0 (p<0.05). This study was IRB-exempt (Indiana University IRB #27779). Results: This study included 2,971 participants with respiratory conditions including asthma (94.4%), COPD (6.5%), chronic bronchitis (7.4%), and lung cancer (1.0%), with 8.5% residing near a pollution source. Among participants with respiratory conditions, exposure to one pollution source was associated with fewer hospitalizations (B=-.823), while clinic visits predicted more hospitalizations (B=.749). For asthma, less than high school education (OR=2.4), some college education (OR=3.3), exposure to one (OR=2.1) and multiple pollution sources (OR=2.7) increased the odds of ED visits. Contrastingly, exposure to one pollution source (B=-.830, p=.003) was associated with fewer hospitalizations. Former smoking was associated with older age (B=7.620) of asthma diagnosis. Conclusion: Air pollution and SDOH were associated with increased ED utilization for asthma, reflecting their combined impact. NWI is heavily industrialized but underrepresented in environmental research, warranting further environmental justice research and interventions.