Electronic Medical Record Search Engine Research Articles

Multiple gastrointestinal immune related adverse events from immune checkpoint inhibitor therapy.

Describe immune related adverse events (irAE) affecting multiple organs of the gastrointestinal system in patients who received immune checkpoint inhibitors (ICI). Within a 2843-patient retrospective cohort consisting of patients with cancer treated with ICIs, EMERSE (Electronic Medical Record Search Engine), an information retrieval system, was used to search free text in the medical record to identify patients with multiple gastrointestinal irAEs. 13 patients developed multiple gastrointestinal irAEs (0.46%). The most common patterns of multisystem gastrointestinal irAE were colitis + pancreatitis and colitis + enteritis. Multisystem gastrointestinal irAEs are rare but warrant further characterization and attention.

A Real-World Analysis of the Diagnostic Evaluation for Hemophagocytic Lymphohistiocytosis in a Pediatric Population

Hemophagocytic lymphohistiocytosis (HLH) is a severe hyperinflammatory syndrome that is often fatal without immunosuppressive treatment to quell the overactive immune response. Accordingly, it is crucial to diagnose HLH promptly and accurately. Nevertheless, diagnosing HLH is challenging due to its phenotypic overlap with other inflammatory conditions. Furthermore, existing HLH diagnostic criteria consist largely of nonspecific clinical and laboratory findings. Together, these factors contribute to significant diagnostic uncertainty, which can delay initiation of treatment in patients who actually have HLH while exposing other patients who do not have HLH to unnecessary immunosuppressive therapy. We sought to better understand how these challenges manifest in a real-world setting and identify opportunities to improve upon existing diagnostic algorithms. To this end, we queried the UCSF Electronic Medical Record Search Engine for clinical notes containing mention of HLH or the related disease, macrophage activation syndrome (MAS), in patients presenting to the UCSF Medical Center between 2008 and 2021. The resulting 2,360 patients were filtered to include only those who were admitted to an inpatient unit and managed by a pediatric care team. Patients with existing malignant, autoimmune, or immunodeficiency conditions were excluded. The resulting 635 patients were divided into an “HLH-diagnosed” group, consisting of 68 patients assigned an ICD-10 code for HLH/MAS, and an “HLH-considered” group, containing the remaining 567 patients without such a code but in whom HLH/MAS was considered on the differential diagnosis. There was an equal distribution of males and females across the cohort. Ages ranged from 0 to 25 years, with no significant difference between groups. Patients in the HLH-diagnosed group were more likely to be deceased at the end of the admission (21% versus 5%, p<0.0001). The most common presenting symptoms were fever, rash, emesis, and fatigue, and there were no differences in the frequency of any presenting features between the two groups. Among the discharge diagnoses for patients in the HLH-considered group, infections were most common (23%), followed by rheumatologic diseases (20%), new malignancies (9%), and fever of unknown origin (8%). We evaluated the performance of the two most commonly used diagnostic algorithms, namely the HLH-2004 criteria and H-score. The HLH-2004 criteria demonstrated a sensitivity of 65% and a specificity of 92%, versus 82% and 83% for the H-score. Patients in the HLH-considered group with newly-diagnosed malignancies had the highest false positivity rate, with 20% testing positive by the HLH-2004 criteria and 34% by the H-score. We also applied these algorithms to patients in the HLH-considered group who were assigned an ICD-10 code for sepsis (n=62). Fifteen and 42% had a false positive result by the HLH-2004 criteria and the H-score, respectively. Given that many HLH-considered patients tested positive by these existing algorithms but were ultimately ruled out based on other clinical data, we sought to create a new model relying only on basic laboratory parameters to help better differentiate HLH-diagnosed and HLH-considered patients. To address this, we performed multivariate logistic regression analysis. Cytopenias and levels of ferritin, triglycerides, and fibrinogen were all significant predictors in univariate analyses and independent predictors in the multivariate analysis. The resulting model produced an area under the curve of 0.86 (0.81-0.90). HLH is commonly considered on the differential diagnosis in hospitalized children, and in our cohort, was ruled out eight times more often than it was diagnosed. Infections, malignancies, and rheumatologic diseases are the most common HLH mimics and cannot be readily distinguished from HLH on the basis of presenting features. Existing diagnostic algorithms showed suboptimal performance in this cohort, particularly in patients with malignancy and sepsis. Our alternative model consisting only of basic laboratory parameters may help streamline the initial diagnostic workup for pediatric patients with concern for HLH and more accurately identify those who require further specialized testing.

Developing a Semantically Based Query Recommendation for an Electronic Medical Record Search Engine: Query Log Analysis and Design Implications.

An effective and scalable information retrieval (IR) system plays a crucial role in enabling clinicians and researchers to harness the valuable information present in electronic health records. In a previous study, we developed a prototype medical IR system, which incorporated a semantically based query recommendation (SBQR) feature. The system was evaluated empirically and demonstrated high perceived performance by end users. To delve deeper into the factors contributing to this perceived performance, we conducted a follow-up study using query log analysis. One of the primary challenges faced in IR is that users often have limited knowledge regarding their specific information needs. Consequently, an IR system, particularly its user interface, needs to be thoughtfully designed to assist users through the iterative process of refining their queries as they encounter relevant documents during their search. To address these challenges, we incorporated "query recommendation" into our Electronic Medical Record Search Engine (EMERSE), drawing inspiration from the success of similar features in modern IR systems for general purposes. The query log data analyzed in this study were collected during our previous experimental study, where we developed EMERSE with the SBQR feature. We implemented a logging mechanism to capture user query behaviors and the output of the IR system (retrieved documents). In this analysis, we compared the initial query entered by users with the query formulated with the assistance of the SBQR. By examining the results of this comparison, we could examine whether the use of SBQR helped in constructing improved queries that differed from the original ones. Our findings revealed that the first query entered without SBQR and the final query with SBQR assistance were highly similar (Jaccard similarity coefficient=0.77). This suggests that the perceived positive performance of the system was primarily attributed to the automatic query expansion facilitated by the SBQR rather than users manually manipulating their queries. In addition, through entropy analysis, we observed that search results converged in scenarios of moderate difficulty, and the degree of convergence correlated strongly with the perceived system performance. The study demonstrated the potential contribution of the SBQR in shaping participants' positive perceptions of system performance, contingent upon the difficulty of the search scenario. Medical IR systems should therefore consider incorporating an SBQR as a user-controlled option or a semiautomated feature. Future work entails redesigning the experiment in a more controlled manner and conducting multisite studies to demonstrate the effectiveness of EMERSE with SBQR for patient cohort identification. By further exploring and validating these findings, we can enhance the usability and functionality of medical IR systems in real-world settings.

Social Determinants of Health Documentation in Structured and Unstructured Clinical Data of Patients With Diabetes: Comparative Analysis.

Electronic health records (EHRs) have yet to fully capture social determinants of health (SDOH) due to challenges such as nonexistent or inconsistent data capture tools across clinics, lack of time, and the burden of extra steps for the clinician. However, patient clinical notes (unstructured data) may be a better source of patient-related SDOH information. It is unclear how accurately EHR data reflect patients' lived experience of SDOH. The manual process of retrieving SDOH information from clinical notes is time-consuming and not feasible. We leveraged two high-throughput tools to identify SDOH mappings to structured and unstructured patient data: PatientExploreR and Electronic Medical Record Search Engine (EMERSE). We included adult patients (≥18 years of age) receiving primary care for their diabetes at the University of California, San Francisco (UCSF), from January 1, 2018, to December 31, 2019. We used expert raters to develop a corpus using SDOH in the compendium as a knowledge base as targets for the natural language processing (NLP) text string mapping to find string stems, roots, and syntactic similarities in the clinical notes of patients with diabetes. We applied advanced built-in EMERSE NLP query parsers implemented with JavaCC. We included 4283 adult patients receiving primary care for diabetes at UCSF. Our study revealed that SDOH may be more significant in the lives of patients with diabetes than is evident from structured data recorded on EHRs. With the application of EMERSE NLP rules, we uncovered additional information from patient clinical notes on problems related to social connectionsisolation, employment, financial insecurity, housing insecurity, food insecurity, education, and stress. We discovered more patient information related to SDOH in unstructured data than in structured data. The application of this technique and further investment in similar user-friendly tools and infrastructure to extract SDOH information from unstructured data may help to identify the range of social conditions that influence patients' disease experiences and inform clinical decision-making.

(281) Pelvic Floor Dysfunction in Premenopausal Women with Recurrent Urinary Tract Infections

Abstract Introduction Urinary tract infections (UTI) are the most common outpatient infections in the United States. Recurrent urinary tract infections (rUTI) are 30 times more common in women than men, suggesting differences in genitourinary anatomy and physiologic factors exist that could explain an increased susceptibility in females. One in three women are diagnosed with a UTI by age 24, with a recurrence rate of approximately 2.6 infections/year. A bidirectional relationship exists between rUTI and pelvic floor dysfunction (PFD) in women, however, little is known about the role of pelvic floor dysfunction in the pathogenesis of UTI in premenopausal women. Objective We aimed to document the occurrence of PFD among a group of premenopausal women with rUTI to further understand the relationship and identify factors associated with pelvic pain and dyspareunia. We hypothesize PFD will have a high prevalence in this population. Methods The Electronic Medical Record Search Engine (EMERSE) database for a single institution was utilize to identify eligible patients. Inclusion criteria included women aged 18-49 with an ICD code for UTI, who still have their ovaries, and who have who have seen a urologist, gynecologist, or urogynecologist within the past 5 years. A total of 610 eligible patients were identified and contacted via email and asked to participate in our survey study. The survey included queries on demographics, medical history, personal experience with UTI symptoms, diagnosis, and treatment. Survey responses were analyzed using chi-squared tests to analyze factors associated with dyspareunia. Pelvic floor dysfunction was defined as baseline pelvic pain outside of the context of UTI or dyspareunia. Internal Review Board approval was obtained for this study (STUDY20211407). Results Of the 610 women contacted, 65 (10.7%) responded. Outside the context of UTI, 52 (80%) women reported some degree of pelvic pain (Figure 1) and 40 (62%) women reported dyspareunia. The presence of pelvic pain was significantly associated with dyspareunia (90% with dyspareunia vs 64% without, p<0.05). On multivariable analysis, patients with pelvic pain were more likely to have dyspareunia (aOR 6.6). There was no association between degree of dyspareunia and duration, number, or severity of UTI. Conclusions PFD is highly prevalent among premenopausal women with rUTI, and screening for pelvic pain and dyspareunia should be performed in this population. Patients who report pelvic pain are more likely to have dyspareunia. Further investigation is needed to evaluate causation between PFD and rUTI in premenopausal women. Disclosure No

Validation of ICD-9 Codes for Identification of Chronic Overlapping Pain Conditions

Chronic overlapping pain conditions (COPCs) are a collection of chronic pain syndromes that often co-occur and are thought to share underlying nociplastic pathophysiology. Since they can manifest as seemingly unrelated syndromes they have historically been studied in isolation. Use of International Classification of Diseases (ICD) codes in medical records has been proposed as a means to identify and study trends in COPCs at the population level, however validated code sets are needed. Recently, a code set comprising ICD-10 codes as proxies for 11 COPCs was validated. The goal of this project was to validate a code set composed of ICD-9 codes for the identification of COPCs in administrative datasets. Data was extracted using the Electronic Medical Record Search Engine at the University of Michigan Health System from January 1st, 2011 to January 1st, 2015. The source population were patients with one of the candidate ICD-9 codes corresponding to various COPCs. Natural language searches were used as a reference standard. If code sets met a pre-specified threshold of agreement between ICD-9 codes and natural language searches (≥ 70%), they were retained and diagnostic accuracy statistics were calculated for each code set. Validated ICD-9 code sets were generated for 10 of the 11 COPCs evaluated. The majority had high levels of diagnostic accuracy, with all but one code set achieving ≥ 80% specificity, sensitivity, and predictive values. This code set may be used by pain researchers to identify COPCs using ICD-9 codes in administrative datasets.

What Do EHRs Tell Us about How We Deploy Health Professionals to Address the Social Determinants of Health

ABSTRACT The objective of this study was to determine how well Electronic Health Record (EHR) documentation identifies which health professionals act to address patient social determinants of health (SDOH) and what interventions are documented. The Electronic Medical Record Search Engine was used to identify food and housing insecurity EHR notes. From the notes, researchers randomly sampled 60 from each SDOH category. Of 120 notes, which contained a reference to food or housing insecurity, 72% also contained information on an intervention taken. Interventions were documented by social workers 63% of the time, followed by dietitians and physicians. Addressing patient SDOH is a crucial part of comprehensive healthcare. Findings contribute to a broader conversation on the documentation and interventions in healthcare settings to address patients’ SDOH. Findings support the critical importance of standardizing SDOH documentation in the EHR across more members of the health workforce to ensure patient needs are met.

Real-World Outcomes with Immunosuppressive Therapy for Aplastic Anemia in Patients Treated at the University of Michigan

IntroductionAplastic anemia is a challenging disease to treat due to its rarity and severity. The current standard of care for severe and very severe aplastic anemia includes consideration of frontline allogeneic stem cell transplant (ASCT) in patients who are under 40 years of age and have an HLA-matched sibling. For patients who do not meet this criteria, immunosuppressive therapy (IST) consisting typically of horse antithymocyte-globulin (ATG) and cyclosporine (CsA) is considered, with recent data supporting the addition of eltrombopag. The main risks of IST are infusion reactions, serum sickness, kidney injury, and infection. While it is generally well tolerated in younger patients, older patients have historically been treated with transfusion support alone due to concerns for higher complications. Unfortunately, data to support clinical decisions regarding these concerns is limited, with small studies showing that reduced dosing or attenuated doses of ATG with or without CsA is safe and can be effective. At the University of Michigan, we have been treating older patients with attenuated doses of ATG +/- CsA since the late-2000s and have recently added eltrombopag to our standard of care. Herein, we present the results of our experience using IST to treat aplastic anemia, specifically presenting our outcomes in patients over 60 years old at the time of initiation of IST.MethodsWe used the Electronic Medical Record Search Engine (EMERSE) to query patients with aplastic anemia who were treated at the University of Michigan between 1995 and 2021. Patients included in our analysis were diagnosed with idiopathic aplastic anemia or hepatitis-associated aplastic anemia, were 18 years or older at the date of initiating IST, were treated with IST first-line, had adequate physician follow up at the University of Michigan, and were reported as being complaint with IST. Patients diagnosed or with high concern for hypoplastic myelodysplastic syndrome (MDS) were excluded in addition to those who received treatment for paroxysmal nocturnal hemoglobinuria (PNH). Statistical analysis was performed using SPSS Statistics v.27 (IBM, 2020).ResultsWe identified 62 patients aged 18 - 60 and 54 patients over 60 years old who met inclusion criteria. Descriptive statistics of the two cohorts are available in Table 1. All patients in our study received horse ATG in addition to cyclosporine. 63% of patients over 60 received full dose horse ATG (20mg/kg for 5 days) and 37% received attenuated horse ATG (20mg/kg for 5 days), while 46% received eltrombopag as part of their IST. OS was superior in patients 18-60 receiving IST with median OS NR vs. 2760 days (Figure 1; p=0.009). In patients over 60, no difference in OS was observed in those who received attenuated (3646 days) vs. full dose IST (2760 days) (Figure 2, p=0.83). Lastly, no difference in median OS was seen in patients > 60 years old who received eltrombopag compared to those who did not (2821 vs 2760 days, Figure 3, p=0.795).Using a Chi-squared test, no statistically significant differences in rates of PR at 12 months, CR at 12 months, primary refractory disease, febrile neutropenia within 6 months of IST, bacterial infections within 6 months IST, and viral infections within 6 months of IST were noted between those who received dose attenuated ATG and those who received full dose ATG in patients > 60. PR in older patients was 48% (12) and 36% (9) at 3 and 12 months with eltrombopag, compared to 40% (10) and 36% (9) without. CR in older patients was 12% (3) and 16% (4) with eltrombopag at 3 and 12 months, compared to 12% (3) and 12% (3) without. In patients > age 60 who received dose attenuated IST, the addition of eltrombopag did not affect OS (Figure 4). A trend toward better OS was seen in patients 18-60 who received eltrombopag, but this was not statistically significant (Figure 5).DiscussionIn conclusion, we present our experience using IST to treat aplastic anemia at the University of Michigan. Dose attenuated IST has been proposed as a potential option for elderly patients, but we found no significant difference in OS, rates of response, or rates of infection between patients who received full dose and those who did not. There was also a trend toward improved survival with the addition of eltrombopag in younger patients, but additional follow up time is needed. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Using Symptoms and Healthcare Encounters to Capture a Rare Disease: A Study of Clinical Notes of the Alpha‐Gal Meat Allergy

AbstractThis paper examines clinical notes to identify reported symptoms and investigate patient‐provider communication processes in alpha‐gal syndrome (AGS). Clinical notes appear to be a credible and stable source of research where the researcher can find information regarding both symptoms and environmental factors of AGS. Compilation of notes could be used for a checklist to aid in diagnosis. This study analyzed clinicians' notes in patient records retrieved from the Electronic Medical Record Search Engine (EMERSE). The most reported symptoms fell into four general categories: skin (42%), inflammation (17%), gastrointestinal (20%), and anaphylaxis (21%). Environmental triggers were also commonly reported. This in‐depth analysis of clinical notes of AGS can serve as a basis for future automation of rare disease analysis; moreover, it provides a basic understanding of the granularity of information that an electronic health record (EHR) may provide for rare disease identification.

Relationships between hallucinations, delusions, depression, suicide ideation, and plan among adults presenting with psychosis in psychiatric emergency care

ABSTRACT Background Suicide is a leading cause of preventable death among adults diagnosed with schizophrenia spectrum disorders. Fewer studies have examined suicide thoughts and behaviors among community populations of adults with psychosis symptoms. The current study examined the relationships between psychosis symptomatology, depression, and suicide ideation and plan among a sample of adults in psychiatric emergency care. Methods An Electronic Medical Record Search Engine was used to obtain data from Psychiatric Emergency Services (PES) of an academic healthcare system in the United States. Patients included 712 adults who presented to PES from 2013 to 2020 with one or more symptom of psychosis and either suicide ideation or a recent suicide attempt. Data were examined in SPSS27 using logistic regressions. Results Patients were twice as likely to experience suicidal ideation when hallucinations were present and eight times more likely when depression was present. Also, patients were twice as likely to have a suicide plan when hallucinations were present and three-times more likely when depression was present. Delusions did not significantly relate to suicide ideation or plan. Discussion Study findings suggest the importance of incorporating positive symptom evaluations into suicide risk assessments and the formulation of risk among patients with psychosis symptoms.

Clinical Documentation of Peanut Allergy Prevention in an Academic Learning Health System

Consensus guidelines recommend early peanut introduction (EPI) as an effective strategy for preventing peanut allergy. Real-world practice patterns of EPI guidance are poorly understood. We aimed to described EPI practice patterns within an academic learning health system. We queried the health system’s central data repository to identify individuals eligible to receive EPI guidance from 1/6/2017 - 03/30/2020: persons less than 12 months of age at the time of a clinical encounter for either a health supervision visit or a problem-based visit related to eczema. We used the Electronic Medical Record Search Engine (EMERSE) to search clinical notes for the word “peanut.” These were manually reviewed for EPI documentation and other practice pattern characteristics. Of a total of 18,149 individuals, 152 (0.8%, mean age 6.4 months) had clinical documentation of EPI guidance (72% clinical note, 20% patient instructions, 5% both, 3% telephone note). EPI guidance was performed by 61 providers among 6 specialties (43% pediatrics, 40% family medicine, 15% allergy/immunology, 2% other) in 24 unique practices. Of those who received EPI guidance, the following comorbid diagnoses and/or characteristics were documented: 35% eczema, 11% egg allergy, 21% family history of food allergy, 31% topical corticosteroid use. Twenty percent were referred to an allergist for peanut allergy risk assessment and 12% had a peanut IgE obtained. Clinical documentation of EPI counseling is low. Further dissemination and implementation efforts are needed to encourage providers to prioritize EPI and increase standard utilization of consensus guidelines.

INCIDENTAL RIGHT VENTRICULAR ABNORMALITIES ON SINGLE PHOTON EMISSION CT SCAN SIGNIFICANTLY PREDICTS DEVELOPMENT OF PULMONARY HYPERTENSION IN BLACK FEMALES

SESSION TITLE: Pulmonary Vascular Disease Posters SESSION TYPE: Original Investigation Posters PRESENTED ON: October 18-21, 2020 PURPOSE: Single Photon Emission Computed Tomography (SPECT) imaging is a diagnostic tool used to risk-stratify patients presenting with chest pain for myocardial ischemia. While SPECT abnormalities of the right ventricle (RV) are known to correlate with invasive hemodynamic measures of disease severity and prognosis in patients with pulmonary hypertension (PH), the prognostic significance of isolated RV abnormalities on SPECT studies in black females who do not carry a PH or cardiac diagnosis is unknown. METHODS: This was a retrospective study. Using the electronic medical record search engine (EMERSE) tool, we identified 12,000 SPECT studies completed at our center between 1/1/07 and 1/1/20. After excluding studies with abnormalities of left ventricular perfusion, elevated cardiac risk, and previous diagnoses of PH, and focusing on black females, we identified 943 unique studies for analysis, and categorized them by the presence or absence of RV abnormalities, including hypertrophy, dilation, or increased uptake. Patient characteristics, including age and body mass index (BMI) at the time of index SPECT study (time zero), were obtained from the medical record, and patients were followed until either an event occurred or the study period ended (2/1/20). The primary outcome was mortality, secondary outcomes included hospital admission and a diagnosis of PH, defined as an estimated right ventricular systolic pressure (RVSP) > 35 mmHg on echocardiogram. Patients lost to follow-up were censored at the time of their last clinical encounter. Multivariable Cox Proportional Hazards models and logistic regression models, adjusted for age and BMI, were used in the analysis. This study was reviewed and approved by the UC Institutional Review Board (IRB 2019-1282). RESULTS: A total of 943 black females were included in this analysis, with a median age of 52 years and a median BMI of 33. The median duration of follow-up was 10.5 years, and 10.3% of patients died during the follow-up period. In the full cohort, RV abnormalities were present in 4% of patients, and PH was subsequently diagnosed in 18.3% of patients during follow-up (median RVSP 45 mmHg). Age was the only significant predictor of readmission (HR 1.01, p=0.007) or mortality (HR 1.07, p=1E-14). RV abnormalities (OR 3.2, p=0.001), age (OR 1.04, p=4E-8), and BMI (OR 1.02, p=0.004) all significantly and independently predicted development of PH. CONCLUSIONS: RV abnormalities incidentally found on otherwise normal SPECT studies were significantly predictive of the development of PH in black females, even after adjustment for age and BMI. The combination of these factors may be useful in risk-stratification of black female patients for PH, and identify high-risk patients who would benefit from a more comprehensive workup for PH and serial echocardiographic evaluation. CLINICAL IMPLICATIONS: RV abnormalities on SPECT imaging may help risk-stratify black females for PH DISCLOSURES: no disclosure on file for Rachel Baker; Chair of a Committee relationship with PHA Please note: PHA Medically Led Sessions, PHA Preceptorship Added 05/22/2020 by Jean Elwing, source=Web Response Removed 05/22/2020 by Jean Elwing, source=Web Response Educator relationship with PH CME Programs - PHA, Simply Speaking, Impact PH Please note: Intermittent, Ongoing Added 05/22/2020 by Jean Elwing, source=Web Response, value=Honoraria Principal Investigator relationship with Actelion, Arena, Reata, United Therapeutics Please note: Ongoing - University is Paid Directly Added 05/22/2020 by Jean Elwing, source=Web Response, value=Grant/Research Support Advisory Committee Member relationship with United Therapeutics Please note: $5001 - $20000 Added 06/05/2020 by Jean Elwing, source=Web Response, value=Consulting fee Principal Investigator relationship with Lung LLC, Liquidia, Phase Bio, Complexa, Gossamer Bio Please note: Ongoing - University is Paid Directly Added 05/22/2020 by Jean Elwing, source=Web Response, value=Grant/Research Support Member of a Committee relationship with PHA Please note: PHA Education Committee Added 05/22/2020 by Jean Elwing, source=Web Response Removed 05/22/2020 by Jean Elwing, source=Web Response Advisory Committee Member relationship with Liquidia, Acceleron Please note: $1001 - $5000 Added 06/05/2020 by Jean Elwing, source=Web Response, value=Consulting fee Advisory Committee Member relationship with Acceleron Please note: $5001 - $20000 Added 06/05/2020 by Jean Elwing, source=Web Response, value=Consulting fee Removed 06/05/2020 by Jean Elwing, source=Web Response Grant - Entelligence Award relationship with Actelion Please note: >$100000 Added 05/25/2020 by Arun Jose, source=Web Response, value=Grant/Research Support No relevant relationships by Francis McCormack, source=Web Response No relevant relationships by Humna Abid Memon, source=Web Response

Dorsal Midbrain Syndrome: Clinical and Imaging Features in 75 Cases.

Dorsal midbrain syndrome (DMS) consists of a constellation of clinical features, including reduced upgaze, pupillary light-near dissociation, lid retraction, convergence retraction, and eye misalignment. This syndrome results mostly from intrinsic or extrinsic mesodiencephalic tumors or strokes, obstructive hydrocephalus, failure of cerebrospinal fluid shunting to correct obstructive hydrocephalus, and head trauma. Published reports that include imaging corroboration are based on relatively small cohorts and have not included comprehensive patient self-reports on the impact of these abnormalities on quality of life. We conducted a retrospective review of cases of DMS identified between 1998 and 2019 at the University of Michigan using the Electronic Medical Record Search Engine. Patients were included only if they had been evaluated by a neuro-ophthalmologist and had a corroborative imaging abnormality. We collected data on symptoms and on neuro-ophthalmic and neurologic signs. We reviewed brain imaging reports on all 75 patients, and the study neuroradiologist analyzed the imaging in 57 patients. Using a uniform list of questions, we conducted telephone interviews of 26 patients to assess lingering symptoms and their impact on quality of life. There were 75 patients, only 5 of whom were younger than 10 years. Neoplasms accounted for 47%, strokes (mostly thalamic) for 25%, nonneoplastic masses for 12%, nonneoplastic hydrocephalus for 7%, traumatic brain injury for 5%, and demyelination for 4%. Reduced upgaze occurred in 93% of patients, being completely absent or reduced to less than 50% amplitude in 67%. Convergence retraction on attempted upgaze occurred in 52%, horizontal misalignment in 49%, vertical misalignment in 47%, and pupillary light-near dissociation in 37%. Optic neuropathy attributed to chronic papilledema occurred in only 3%. Three or more neuro-ophthalmic signs were present in 84%, and only 4% had a single sign-reduced upgaze. Imaging features did not correlate with the frequency or severity of clinical signs. There was some improvement in the clinical signs among the patients with stroke but no change among the patients with neoplasms. In the 26 telephone interviews, patients with neoplasms reported that imbalance had a greater impact on quality of life than did diplopia. Patients with strokes reported that imbalance had the greatest impact initially but that its effect dissipated. Neither group reported lingering effects of impaired upgaze. This large series expands on the clinical profile of DMS. Neoplasms and strokes were the most common causes. Obstructive hydrocephalus alone, identified as a major cause in the largest previously published series, was uncommon. At least 3 neuro-ophthalmic signs were present in nearly all patients, with upgaze deficit as predominant. Unlike an earlier report, this study found no correlation between brain imaging and clinical signs. Neuro-ophthalmic signs persisted even after neoplasms were successfully treated and improved only slightly after stroke. Telephone interviews with patients revealed that diplopia and upgaze deficit had less lasting impact on quality of life than did ataxia and concurrent nonneurologic problems.

Electronic Medical Record Search Engine (EMERSE): An Information Retrieval Tool for Supporting Cancer Research.

PURPOSEThe Electronic Medical Record Search Engine (EMERSE) is a software tool built to aid research spanning cohort discovery, population health, and data abstraction for clinical trials. EMERSE is now live at three academic medical centers, with additional sites currently working on implementation. In this report, we describe how EMERSE has been used to support cancer research based on a variety of metrics.METHODSWe identified peer-reviewed publications that used EMERSE through online searches as well as through direct e-mails to users based on audit logs. These logs were also used to summarize use at each of the three sites. Search terms for two of the sites were characterized using the natural language processing tool MetaMap to determine to which semantic types the terms could be mapped.RESULTSWe identified a total of 326 peer-reviewed publications that used EMERSE through August 2019, although this is likely an underestimation of the true total based on the use log analysis. Oncology-related research comprised nearly one third (n = 105; 32.2%) of all research output. The use logs showed that EMERSE had been used by multiple people at each site (nearly 3,500 across all three) who had collectively logged into the system > 100,000 times. Many user-entered search queries could not be mapped to a semantic type, but the most common semantic type for terms that did match was “disease or syndrome,” followed by “pharmacologic substance.”CONCLUSIONEMERSE has been shown to be a valuable tool for supporting cancer research. It has been successfully deployed at other sites, despite some implementation challenges unique to each deployment environment.

SAT-172 FDG PET/CT in Benign vs. Malignant Adrenocortical Tumors

18F-fluorodeoxyglucose positron emission tomography computed tomography (FDG PET/CT) is frequently used for evaluation of indeterminate adrenal nodules.The objective of this study was to evaluate the diagnostic performance of FDG PET/CT in distinguishing benign adrenocortical adenomas (ACA) from adrenocortical carcinoma (ACC).We identified 139 patients in our institution who completed both FDG PET/CT and adrenal biopsy or adrenalectomy from 2001 to 2019 using the institutional Electronic Medical Record Search Engine (EMERSE). Patients with adrenal pathology revealing non-adrenocortical tumors were excluded. The imaging characteristics of ACAs were compared with ACCs.Sixteen patients with ACAs and 50 patients with ACC were identified; of the latter, 30 were excluded, because FDG PET/CT imaging was completed after adrenalectomy. The average age of the ACC vs. ACA group was 48±14 and 52±17 years, respectively, with a gender composition of 35% and 25% men, respectively. All 20 patients with ACC had FDG avid lesions, defined by imaging report and increased ratio of adrenal tumor maximum standardized uptake value (SUVmax) to hepatic parenchymal average SUV (SUVmean) of greater than 2.5. Among those with ACAs, 11 had positive, 1 had mildly positive, 3 had indeterminate, and 1 had negative FDG PET/CT findings according to the report. Ratio of adrenal SUVmax to hepatic SUVmean was greater in ACC compared to ACA (13.5 versus 1.6, respectively, p = 0.13), but did not reach statistical significance, likely due to the small sample size and unavailability of SUV ratios for all lesions.Our study identified a large number of ACAs that were deemed FDG avid but without adrenal SUVmax to hepatic SUVmean ratio of greater than 2.5. Although the sensitivity, and therefore the negative predictive value, of FDG PET/CT scan is excellent for the diagnosis of ACC, our data does not allow for the calculation of specificity. This is due to the selection bias inherent in our patient population referred for evaluation of unusual adrenal masses, and the fact that ACAs with negative FDG PET/CT often do not undergo surgery or biopsy, and therefore pathology results are not available. Despite this limitation, our findings show that calculation of the ratio of adrenal SUVmax to hepatic SUVmean, rather than lesion SUVmax alone, helps for characterization of adrenal lesions as malignant vs. benign, and that a negative FDG PET/CT is valuable in excluding ACC.

Comparing Administrative and Clinical Data for Central Line Associated Blood Stream Infections in Pediatric Intensive Care Unit and Pediatric Cardiothoracic Intensive Care Unit

Central line associated bloodstream infections (CLABSIs) are a frequent source of health complication for patients of all ages, including for patients in the pediatric intensive care unit (PICU) and Pediatric Cardiothoracic Intensive Care Unit (PCTU). Many hospitals, including the University of Michigan Health System, currently use the International Classification of Disease (ICD) coding system when coding for CLABSI.The purpose of this study was to determine the accuracy of coding for CLABSI infections with ICD-9CM codes in PICU and PCTU patients. A retrospective chart review was conducted for 75 PICU and PCTU patients with 90 events of hospital acquired central line infections at the University of Michigan Health System (from 2007-2011). The different variables examined in the chart review included the type of central line the patient had, the duration of the stay of the line, the type of organism infecting the patient, and the treatment the patient received. A review was conducted to assess if patients had received the proper ICD-9CM code for their hospital acquired infection. In addition, each patient chart was searched using Electronic Medical Record Search Engine to determine if any phrases that commonly referred to hospital acquired CLABSIs were present in their charts. Our review found that in most CLABSI cases the hospital’s administrative data diagnosis using ICD-9CM coding systems did not code for the CLABSI. Our results indicate a low sensitivity of 32% in the PICU and an even lower sensitivity of 12% in the PCTU. Using these results, we can conclude that the ICD-9CM coding system cannot be used for accurately defining hospital acquired CLABSIs in administrative data. With the new use of the ICD-10CM coding system, further research is needed to assess the effects of the ICD-10CM coding system on the accuracy of administrative data.

Supporting information retrieval from electronic health records: A report of University of Michigan’s nine-year experience in developing and using the Electronic Medical Record Search Engine (EMERSE)

ObjectiveThis paper describes the University of Michigan’s nine-year experience in developing and using a full-text search engine designed to facilitate information retrieval (IR) from narrative documents stored in electronic health records (EHRs). The system, called the Electronic Medical Record Search Engine (EMERSE), functions similar to Google but is equipped with special functionalities for handling challenges unique to retrieving information from medical text. Materials and methodsKey features that distinguish EMERSE from general-purpose search engines are discussed, with an emphasis on functions crucial to (1) improving medical IR performance and (2) assuring search quality and results consistency regardless of users’ medical background, stage of training, or level of technical expertise. ResultsSince its initial deployment, EMERSE has been enthusiastically embraced by clinicians, administrators, and clinical and translational researchers. To date, the system has been used in supporting more than 750 research projects yielding 80 peer-reviewed publications. In several evaluation studies, EMERSE demonstrated very high levels of sensitivity and specificity in addition to greatly improved chart review efficiency. DiscussionIncreased availability of electronic data in healthcare does not automatically warrant increased availability of information. The success of EMERSE at our institution illustrates that free-text EHR search engines can be a valuable tool to help practitioners and researchers retrieve information from EHRs more effectively and efficiently, enabling critical tasks such as patient case synthesis and research data abstraction. ConclusionEMERSE, available free of charge for academic use, represents a state-of-the-art medical IR tool with proven effectiveness and user acceptance.

IntheNEWS: A ROUNDUP OF NEWS AND INFORMATION FROM OUR COMMUNITY

In<i>the</i>NEWS: A ROUNDUP OF NEWS AND INFORMATION FROM OUR COMMUNITY

An Electronic Medical Record Search Engine for Clinical Outcome Research

An Electronic Medical Record Search Engine for Clinical Outcome Research

OP180: Mucosal malignancy with solid organ transplantation

Purpose 1. Measure the prevalence of mucosal head and neck squamous cell carcinoma (HNSCC) using a large database of solid organ transplant recipients. 2. Assess the risk of malignancies considering organ type, age at transplant and use of immunosuppressive medications. 3. Investigate the impact of on overall survival of each patient subset. Methods A database containing all solid organ transplant recipients at UMHS was used, and medical records from 4125 patients transplanted from 1996 to 2006 were queried using EMERSE (Electronic Medical Record Search Engine), with a bundle of search terms designed to screen for target malignancies. Once cases were identified, analysis of differences in incidence based on transplant site and age was performed using a generalized linear model with transplant organ and age at transplant as predictors. For all patients transplanted in the year 2002, medication usage history and survival data were gathered for subgroup analysis of these endpoints. Results The transplant recipient cohort consisted of 59% kidney, 22% liver, 8% lung, 10% heart and 5% receiving dual organ transplant. Head and neck cancer (HNSCC) was diagnosed in 0.6% of patients (n = 26). The mean age (std., range) at transplant was 44 years (17, Conclusion Transplant recipients carry an elevated risk of mucosal malignancies. Older age is correlated with increased risk of malignancy and shorter post-transplant interval. There is a trend towards worse prognosis in HNSCC patients with solid organ transplant.