Nevin Manimala

BMC Med Inform Decis Mak. 2022 Jun 16;22(Suppl 2):156. doi: 10.1186/s12911-022-01871-0.

ABSTRACT

BACKGROUND: Sepsis is one of the most life-threatening circumstances for critically ill patients in the United States, while diagnosis of sepsis is challenging as a standardized criteria for sepsis identification is still under development. Disparities in social determinants of sepsis patients can interfere with the risk prediction performances using machine learning.

METHODS: We analyzed a cohort of critical care patients from the Medical Information Mart for Intensive Care (MIMIC)-III database. Disparities in social determinants, including race, sex, marital status, insurance types and languages, among patients identified by six available sepsis criteria were revealed by forest plots with 95% confidence intervals. Sepsis patients were then identified by the Sepsis-3 criteria. Sixteen machine learning classifiers were trained to predict in-hospital mortality for sepsis patients on a training set constructed by random selection. The performance was measured by area under the receiver operating characteristic curve (AUC). The performance of the trained model was tested on the entire randomly conducted test set and each sub-population built based on each of the following social determinants: race, sex, marital status, insurance type, and language. The fluctuations in performances were further examined by permutation tests.

RESULTS: We analyzed a total of 11,791 critical care patients from the MIMIC-III database. Within the population identified by each sepsis identification method, significant differences were observed among sub-populations regarding race, marital status, insurance type, and language. On the 5783 sepsis patients identified by the Sepsis-3 criteria statistically significant performance decreases for mortality prediction were observed when applying the trained machine learning model on Asian and Hispanic patients, as well as the Spanish-speaking patients. With pairwise comparison, we detected performance discrepancies in mortality prediction between Asian and White patients, Asians and patients of other races, as well as English-speaking and Spanish-speaking patients.

CONCLUSIONS: Disparities in proportions of patients identified by various sepsis criteria were detected among the different social determinant groups. The performances of mortality prediction for sepsis patients can be compromised when applying a universally trained model for each subpopulation. To achieve accurate diagnosis, a versatile diagnostic system for sepsis is needed to overcome the social determinant disparities of patients.

PMID:35710407 | DOI:10.1186/s12911-022-01871-0

BMC Surg. 2022 Jun 16;22(1):231. doi: 10.1186/s12893-022-01622-y.

ABSTRACT

BACKGROUND: Although there are common postoperative complications, Roux-en-Y cholangiojejunostomy is still broadly used as a standard surgical procedure for patients with biliary stricture. This study aimed to explore long-term risk factors of cholangiojejunostomy in patients with biliary stricture who underwent revisional cholangiojejunostomy.

METHODS: Clinical data of 61 patients with biliary stricture undergoing revisional cholangiojejunostomy were retrospectively analyzed. These patients were classified into two groups (patients with traumatic biliary stricture and non-traumatic biliary stricture). Postoperative complications and survival time were successfully followed up.

RESULTS: Among the patients, 34 underwent revisional cholangiojejunostomy due to traumatic biliary stricture, and 27 underwent revisional cholangiojejunostomy due to non-traumatic biliary surgery. Although there was no statistical difference in most clinical data between two groups, biliary dilation or not during the first surgery, cholelithiasis or not during the first surgery, long-term complications after first surgery, cholelithiasis or not during the second surgery, identifying abnormalities during the second surgery and long-term complications after second surgery were significantly different. All patients were successfully followed up and average follow-up time for patients with traumatic and non-traumatic biliary stricture was (88.44 ± 35.67) months and (69.48 ± 36.61) months respectively. Survival analysis indicated that there was no statistical difference in overall survival between two groups. Additionally, cox proportional hazard analysis demonstrated that first preoperative bilirubin level, short-term complication after first surgery and identifying abnormalities during the second surgery were independent risk factors that may have significant effects on patients’ overall survival and long-term prognosis after cholangiojejunostomy. Among the intraoperative abnormal findings, residual lesions after the first operation had significant effects on the patients overall survival in the earlier stage. Relatively, anastomotic stoma stricture and biliary output loop problems had obvious effects on patients’ overall survival at later stages.

CONCLUSION: First preoperative bilirubin level, short-term complication after first surgery and abnormal findings during the second surgery were independent risk factors of revisional cholangiojejunostomy, which may affect patients’ long-term survival. Therefore, surgeons should minimize incidence of postoperative complications through fully evaluating optimal operative time and standardizing surgical procedures.

PMID:35710403 | DOI:10.1186/s12893-022-01622-y

BMC Med Inform Decis Mak. 2022 Jun 16;22(Suppl 2):155. doi: 10.1186/s12911-022-01869-8.

ABSTRACT

BACKGROUND: Opioid use disorder (OUD) has become an urgent health problem. People with OUD often experience comorbid medical conditions. Systematical approaches to identifying co-occurring conditions of OUD can facilitate a deeper understanding of OUD mechanisms and drug discovery. This study presents an integrated approach combining data mining, network construction and ranking, and hypothesis-driven case-control studies using patient electronic health records (EHRs).

METHODS: First, we mined comorbidities from the US Food and Drug Administration Adverse Event Reporting System (FAERS) of 12 million unique case reports using frequent pattern-growth algorithm. The performance of OUD comorbidity mining was measured by precision and recall using manually curated known OUD comorbidities. We then constructed a disease comorbidity network using mined association rules and further prioritized OUD comorbidities. Last, novel OUD comorbidities were independently tested using EHRs of 75 million unique patients.

RESULTS: The OUD comorbidities from association rules mining achieves a precision of 38.7% and a recall of 78.2 Based on the mined rules, the global DCN was constructed with 1916 nodes and 32,175 edges. The network-based OUD ranking result shows that 43 of 55 known OUD comorbidities were in the first decile with a precision of 78.2%. Hypothyroidism and type 2 diabetes were two top-ranked novel OUD comorbidities identified by data mining and network ranking algorithms. Based on EHR-based case-control studies, we showed that patients with OUD had significantly increased risk for hyperthyroidism (AOR = 1.46, 95% CI 1.43-1.49, p value < 0.001), hypothyroidism (AOR = 1.45, 95% CI 1.42-1.48, p value < 0.001), type 2-diabetes (AOR = 1.28, 95% CI 1.26-1.29, p value < 0.001), compared with individuals without OUD.

CONCLUSION: Our study developed an integrated approach for identifying and validating novel OUD comorbidities from health records of 87 million unique patients (12 million for discovery and 75 million for validation), which can offer new opportunities for OUD mechanism understanding, drug discovery, and multi-component service delivery for co-occurring medical conditions among patients with OUD.

PMID:35710401 | DOI:10.1186/s12911-022-01869-8

Trials. 2022 Jun 16;23(1):500. doi: 10.1186/s13063-022-06386-7.

ABSTRACT

BACKGROUND: Anorexia nervosa (AN) is a serious and disabling mental disorder with a high disease burden. In a proportion of cases, intensive hospital-based treatments, i.e. inpatient or day patient treatment, are required, with day patient treatment often being used as a ‘step-down’ treatment after a period of inpatient treatment. Demand for such treatment approaches has seen a sharp rise. Despite this, the relative merits of these approaches for patients, their families, and the NHS and wider society are relatively unknown. This paper describes the rationale for, and protocol of, a two-arm multi-centre open-label parallel group non-inferiority randomised controlled trial, evaluating the effectiveness and cost-effectiveness of these two intensive treatments for adults with severe AN: inpatient treatment as usual and a stepped care day patient approach (the combination of day patient treatment with the option of initial inpatient treatment for medical stabilisation). The main aim of this trial is to establish whether, in adults with severe AN, a stepped care day patient approach is non-inferior to inpatient treatment as usual in relation to improving body mass index (BMI) at 12 months post-randomisation.

METHODS: 386 patients with a Diagnostic and Statistical Manual 5th edition diagnosis of severe AN or related disorder, with a BMI of ≤16 kg/m² and in need of intensive treatment will be randomly allocated to either inpatient treatment as usual or a stepped care day patient approach. Patients in both groups will receive treatment until they reach a healthy weight or get as close to this point as possible. Assessments will be conducted at baseline (prior to randomisation), and at 6 and 12 months post-randomisation, with additional monthly symptom monitoring. The primary outcome will be BMI at the 12-month post-randomisation assessment. Other outcomes will include psychosocial adjustment; treatment motivation, expectations and experiences; cost-effectiveness; and carer burden.

DISCUSSION: The results of this study will provide a rigorous evaluation of two intensive treatment approaches which will inform future national and international treatment guidelines and service provision.

TRIAL REGISTRATION: ISRCTN ISRCTN10166784 . Registered 28 February 2020. ISRCTN is a primary registry of the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) network and includes all items from the WHO Trial Registration Data Set.

PMID:35710394 | DOI:10.1186/s13063-022-06386-7

Ital J Pediatr. 2022 Jun 16;48(1):98. doi: 10.1186/s13052-022-01292-9.

ABSTRACT

BACKGROUND: Breastfeeding might prevent childhood cancer by stimulating the immune system.

METHODS: The following databases, including PubMed, Embase, and Cochrane Library, were searched from inception to January 10, 2021.

RESULTS: In dose-dependent manner, there was a statistically significant inverse association between any breastfeeding and the incidence of childhood cancer. There was no evidence that breastfeeding was inversely related to childhood cancer of the skeletal, reproductive, or sensory systems. However, breastfeeding was inversely associated with the incidence of hematological malignancies and cancers of the nervous and urinary systems. Among hematological malignancies, the relationship was significant for acute lymphocytic leukemia (ALL) and acute myeloid leukemia (AML), but not for acute non-lymphocytic leukemia (ANLL), Hodgkin’s lymphoma (HL), or non-HL.

CONCLUSIONS: The evidences demonstrated that breastfeeding have a potential protective role in preventing selective childhood cancer growth, especially for ALL, AML, cancer of nervous and urinary systems. This study recommended that breastfeeding be extended for as long as possible or maintained for at least 6 months to prevent selective childhood cancer growth.

PMID:35710389 | DOI:10.1186/s13052-022-01292-9

JCO Oncol Pract. 2022 Jun 16:OP2200056. doi: 10.1200/OP.22.00056. Online ahead of print.

ABSTRACT

PURPOSE: Patients with hematologic malignancies are extremely vulnerable to financial toxicity (FT) because of the high costs of treatment and health care utilization. This pilot study identified patients at high risk because of FT and attempted to improve clinical outcomes with comprehensive intervention.

METHODS: All patients who presented to the Levine Cancer Institute’s Leukemia Clinic between May 26, 2019, and March 10, 2020, were screened for inclusion by standardized two question previsit survey. Patients screening positive were enrolled in the comprehensive intervention that used nurse navigators, clinical pharmacists, and community pro bono financial planners. Primary outcomes were defined as improvement in mental and physical quality of life in all patients and improvement in overall survival in the high-risk disease group.

RESULTS: One hundred seven patients completed comprehensive intervention. Patients experiencing FT had increased rates of noncompliance including to prescription (16.8%) and over-the-counter medications (15.9%). The intervention resulted in statistically significantly higher quality of life when measured by using Patient-Reported Outcomes Measurement Information System physical (12.5 ± 2.2 v 13.7 ± 1.8) and mental health scores (11.4 ± 2.2 v 12.4 ± 2.2; all P < .001). In patients with high-risk disease (as determined by using disease-specific scoring systems), risk of death in those receiving the intervention was 0.44 times the risk of death in those without the intervention after adjusting for race, and treatment with stem-cell transplant, oral chemotherapy, or immunotherapy (95% CI, 0.21 to 0.94; P = .034).

CONCLUSION: Screening and intervention on FT for patients with hematologic malignancies is associated with increased quality of life and survival.

PMID:35709421 | DOI:10.1200/OP.22.00056

Otol Neurotol. 2022 Jun 17. doi: 10.1097/MAO.0000000000003547. Online ahead of print.

ABSTRACT

OBJECTIVE: To analyze the relationship of electrode array (EA) type and position on hearing preservation longevity following cochlear implantation.

STUDY DESIGN: Retrospective chart review.

SETTING: Tertiary referral center.

PATIENTS: Adult cochlear implant recipients between 2013 and 2019 with hearing preserved postoperatively and postoperative CT scans.

INTERVENTIONS: CT scan analysis of EA position. Stepwise regression to determine influence of EA position, EA type, and patient demographics on postoperative low frequency hearing.

MAIN OUTCOME MEASURES: Low frequency pure tone average (LFPTA), LFPTA shift, angular insertion depth, base insertion depth, scalar position, mean perimodiolar distance.

RESULTS: Of 792 cochlear implant recipients, 121 had preoperative LFPTA ≤80 dB HL with 60 of the 121 (49.6%) implanted with straight, 32 (26.4%) with precurved, styletted, and 29 (24.0%) implanted precurved, nonstyletted EA. Mean follow up was 28.6 months (range 1-103). There was no statistically significant difference in activation, 6- and 12-month, and last follow-up LFPTA (125, 250, and 500 Hz) shift based on EA type (straight p = 0.302, precurved, styletted p = 0.52, precurved, nonstyletted p = 0.77). Preoperative LFPTA and age of implantation were significant predictors of LFPTA shift at activation, accounting for 30.8% of variance (F[2, 113] = 26.603, p < 0.0001). LFPTA shift at activation, scalar position, and base insertion depth were significant predictors of variability and accounted for 39.1% of variance in LFPTA shift at 6 months (F[3, 87] = 20.269, p < 0.0001). Only LFPTA shift at 12 months was found to be a significant predictor of LFPTA shift at last follow up, accounting for 41.0% of variance (F[1, 48] = 32.653, p < 0.0001).

CONCLUSIONS: Patients had excellent long-term residual hearing regardless of EA type. Age, preoperative acoustic hearing, and base insertion depth may predict short term preservation, while 12-month outcomes significantly predicted long-term hearing preservation.

PMID:35709407 | DOI:10.1097/MAO.0000000000003547

Spec Care Dentist. 2022 Jun 16. doi: 10.1111/scd.12747. Online ahead of print.

ABSTRACT

AIMS: To assess the presence of alterations suggestive of reduced bone mineral density (BMD) by using mandibular cortical index (MCI) in panoramic radiographs of cirrhotic individuals and to evaluate their relationship with other characteristics of hepatic cirrhosis (HC).

METHODS AND RESULTS: This is an observational case-control study assessing the medical records of 165 cirrhotic patients matched by sex and age with healthy individuals. MELD (model of end stage liver disease) score, etiology, complications, comorbidities, and serum levels of vitamin D were collected. MCI was used to obtain BMD. Binary logistic regression was used to test associations and the risk estimates were expressed in odds ratio. Most of the sample consisted of men (73.93%) with median age of 56 years old. In the study group, the mean value of MELD was 16.5 and hepatitis C was the main etiology of HC (33.9%). Cirrhotic individuals are 3.99 times more likely to present alterations suggestive of reduced BMD (p = .00). There was no statistical significance in the association of MCI with levels of vitamin D, comorbidities, etiology or cirrhosis complications.

CONCLUSIONS: MCI suggestive of reduced BMD is more likely to be identified in panoramic radiographs of cirrhotic individuals than of healthy ones.

PMID:35709388 | DOI:10.1111/scd.12747

J Appl Stat. 2021 Sep 1;48(16):3251-3252. doi: 10.1080/02664763.2021.1973388. eCollection 2021.

ABSTRACT

[This corrects the article DOI: 10.1080/02664763.2016.1142945.].

PMID:35709375 | PMC:PMC9041910 | DOI:10.1080/02664763.2021.1973388

Aesthet Surg J. 2022 Jun 16:sjac158. doi: 10.1093/asj/sjac158. Online ahead of print.

ABSTRACT

BACKGROUND: Since the association between anaplastic large cell lymphoma (ALCL) and textured breast implants has led to the recall of Allergan Biocell (Irvine, CA, USA) devices, plastic surgeons have been faced with the challenge of caring for patients with these implants in situ. Cosmetic and reconstructive surgeons have been contacting patients with these implants in order to encourage them to follow up and discuss the most appropriate risk reduction strategies.

OBJECTIVES: To evaluate patient concerns about the risk of breast implant associated ALCL and compare management differences between cosmetic and reconstruction patients.

METHODS: A retrospective review was performed of 432 patients with macro textured implants that presented to clinic after being contacted (121 reconstructive and 311 cosmetic). These records were analyzed for their presenting concerns, surgery wait times and management plans. Statistical analysis was performed to compare the cohorts and odds ratios were computed to determine the association between patient concerns and their choice of management.

RESULTS: After consultation, 59.5% of the reconstruction cohort and 49.5% of the cosmetic cohort scheduled implant removal or exchange. The reconstructive population had a higher rate of ALCL concern (62.7%), however both cohorts had a significant odds ratio demonstrating an expressed fear of ALCL likely contributed to their subsequent clinical management (1.66 OR cosmetic, 2.17 OR reconstructive).

CONCLUSIONS: Although the risk of ALCL appears to be more concerning to the reconstructive population, both cohorts were equally motivated to have their implants removed. Informing patients about their ALCL risk is crucial to ensure a patient supported risk reduction plan.

PMID:35709374 | DOI:10.1093/asj/sjac158