Fine-tuning large language models for rare disease concept normalization.

OBJECTIVE: We aim to develop a novel method for rare disease concept normalization by fine-tuning Llama 2, an open-source large language model (LLM), using a domain-specific corpus sourced from the Human Phenotype Ontology (HPO). METHODS: We developed an in-house template-based script to generate two corpora for fine-tuning. The first (NAME) contains standardized HPO names, sourced from the HPO vocabularies, along with their corresponding identifiers. The second (NAME+SYN) includes HPO names and half of the concept's synonyms as well as identifiers. Subsequently, we fine-tuned Llama 2 (Llama2-7B) for each sentence set and conducted an evaluation using a range of sentence prompts and various phenotype terms. RESULTS: When the phenotype terms for normalization were included in the fine-tuning corpora, both models demonstrated nearly perfect performance, averaging over 99% accuracy. In comparison, ChatGPT-3.5 has only ∼20% accuracy in identifying HPO IDs for phenotype terms. When single-character typos were introduced in the phenotype terms, the accuracy of NAME and NAME+SYN is 10.2% and 36.1%, respectively, but increases to 61.8% (NAME+SYN) with additional typo-specific fine-tuning. For terms sourced from HPO vocabularies as unseen synonyms, the NAME model achieved 11.2% accuracy, while the NAME+SYN model achieved 92.7% accuracy. CONCLUSION: Our fine-tuned models demonstrate ability to normalize phenotype terms unseen in the fine-tuning corpus, including misspellings, synonyms, terms from other ontologies, and laymen's terms. Our approach provides a solution for the use of LLMs to identify named medical entities from clinical narratives, while successfully normalizing them to standard concepts in a controlled vocabulary.

GestaltMML: Enhancing Rare Genetic Disease Diagnosis through Multimodal Machine Learning Combining Facial Images and Clinical Texts.

Individuals with suspected rare genetic disorders often undergo multiple clinical evaluations, imaging studies, laboratory tests and genetic tests, to find a possible answer over a prolonged period of time. Addressing this "diagnostic odyssey" thus has substantial clinical, psychosocial, and economic benefits. Many rare genetic diseases have distinctive facial features, which can be used by artificial intelligence algorithms to facilitate clinical diagnosis, in prioritizing candidate diseases to be further examined by lab tests or genetic assays, or in helping the phenotype-driven reinterpretation of genome/exome sequencing data. Existing methods using frontal facial photos were built on conventional Convolutional Neural Networks (CNNs), rely exclusively on facial images, and cannot capture non-facial phenotypic traits and demographic information essential for guiding accurate diagnoses. Here we introduce GestaltMML, a multimodal machine learning (MML) approach solely based on the Transformer architecture. It integrates facial images, demographic information (age, sex, ethnicity), and clinical notes (optionally, a list of Human Phenotype Ontology terms) to improve prediction accuracy. Furthermore, we also evaluated GestaltMML on a diverse range of datasets, including 528 diseases from the GestaltMatcher Database, several in-house datasets of Beckwith-Wiedemann syndrome (BWS, over-growth syndrome with distinct facial features), Sotos syndrome (overgrowth syndrome with overlapping features with BWS), NAA10-related neurodevelopmental syndrome, Cornelia de Lange syndrome (multiple malformation syndrome), and KBG syndrome (multiple malformation syndrome). Our results suggest that GestaltMML effectively incorporates multiple modalities of data, greatly narrowing candidate genetic diagnoses of rare diseases and may facilitate the reinterpretation of genome/exome sequencing data.

D-TERMINED, a phase 1 trial in newly diagnosed high-grade glioma with temozolomide, radiation, and minocycline followed by adjuvant minocycline/temozolomide.

Background: Standard treatment for newly diagnosed high-grade gliomas remains suboptimal. Preclinical data indicate that mesenchymal transition and radiation resistance in glioblastoma are driven by NF-κB and microglia activation, which can be inhibited by minocycline. We assessed the safety and efficacy of minocycline combined with standard radiation and temozolomide in newly diagnosed high-grade gliomas. Methods: Adults with newly diagnosed high-grade glioma were eligible. Minocycline was given with concurrent and adjuvant temozolomide. Minocycline doses were escalated using a 3 + 3 design and expanded to identify the maximum tolerated dose (MTD) and adverse event profile. Individual progression-free survival (PFS) was compared to predicted PFS based on RTOG RPA class using a binomial test. The relationships between mesenchymal and microglial biomarkers were analyzed with immunohistochemistry. Results: The MTD of minocycline was 150 mg twice per day (N = 20); 1 patient (5%) experienced CTCAE grade 3 + nausea and dizziness, and 2 patients (10%) demonstrated thrombocytopenia requiring temozolomide interruptions. Twelve patients exceeded their predicted PFS (60%), which did not meet the predefined efficacy endpoint of 70%. Symptoms increased during post-radiation treatment but remained mild. No significant correlation was seen between biomarkers and PFS. Expression levels of P-p65, a marker of NF-κB activation, were correlated with the microglia marker IBA-1. Conclusions: Minocycline at 150 mg twice per day is well tolerated with standard chemoradiation in patients with newly diagnosed high-grade gliomas. PFS was not significantly increased with the addition of minocycline when compared to historical controls. NF-κB activation correlates with microglia levels in high-grade glioma.

Assessing the Utility of Large Language Models for Phenotype-Driven Gene Prioritization in Rare Genetic Disorder Diagnosis.

Phenotype-driven gene prioritization is a critical process in the diagnosis of rare genetic disorders for identifying and ranking potential disease-causing genes based on observed physical traits or phenotypes. While traditional approaches rely on curated knowledge graphs with phenotype-gene relations, recent advancements in large language models have opened doors to the potential of AI predictions through extensive training on diverse corpora and complex models. This study conducted a comprehensive evaluation of five large language models, including two Generative Pre-trained Transformers series, and three Llama2 series, assessing their performance across three key metrics: task completeness, gene prediction accuracy, and adherence to required output structures. Various experiments explored combinations of models, prompts, input types, and task difficulty levels. Our findings reveal that even the best-performing LLM, GPT-4, achieved an accuracy of 16.0%, which still lags behind traditional bioinformatics tools. Prediction accuracy increased with the parameter/model size. A similar increasing trend was observed for the task completion rate, with complicated prompts more likely to increase task completeness in models smaller than GPT-4. However, complicated prompts are more likely to decrease the structure compliance rate, but no prompt effects on GPT-4. Compared to HPO term-based input, LLM was also able to achieve better than random prediction accuracy by taking free-text input, but slightly lower than with the HPO input. Bias analysis showed that certain genes, such as MECP2, CDKL5, and SCN1A, are more likely to be top-ranked, potentially explaining the variances observed across different datasets. This study provides valuable insights into the integration of LLMs within genomic analysis, contributing to the ongoing discussion on the utilization of advanced LLMs in clinical workflows.

Two Acentric Zn-Based l-Tartrates with Moderate Second Harmonic Generation Responses and Large Birefringence.

The discovery of novel organic-inorganic hybrid nonlinear optical (NLO) crystal materials holds great potential in advancing laser science and technology as they offer a wide range of benefits including easy synthesis, structural versatility, and high hyperpolarizability. Herein, the integration of l-tartaric acid (L-C4H6O6) and ZnSO4 gave rise to two acentric Zn-based organic-inorganic hybrid crystals, namely, Zn2(H2O)2(C4H4O6)2·3H2O (Zn-LT) and Zn2(H2O)(C4H4O6)(C4H6O6)(SO4)·4H2O (Zn-LTS). They both feature layered structures constructed by [ZnO6] octahedron, l-C4H6O6/l-C4H4O62-, water molecule, or [SO4] tetrahedron. Interestingly, they possess moderate second-order NLO effects of 0.28 × KH2PO4 (Zn-LT) and 0.57 × KH2PO4 (Zn-LTS), large birefringence of 0.075 (Zn-LT) and 0.069 (Zn-LTS), suggesting that the introduction of [SO4] groups with intrinsically weak polarizability and weak optical anisotropy induces the enhancement of the NLO effect while without reducing birefringence much. In addition, both of them display UV cutoff edges near 210 nm, indicating their potential as NLO crystals applied in the UV and solar-blind region.

Fine-tuning Large Language Models for Rare Disease Concept Normalization.

Objective: We aim to develop a novel method for rare disease concept normalization by fine-tuning Llama 2, an open-source large language model (LLM), using a domain-specific corpus sourced from the Human Phenotype Ontology (HPO). Methods: We developed an in-house template-based script to generate two corpora for fine-tuning. The first (NAME) contains standardized HPO names, sourced from the HPO vocabularies, along with their corresponding identifiers. The second (NAME+SYN) includes HPO names and half of the concept's synonyms as well as identifiers. Subsequently, we fine-tuned Llama2 (Llama2-7B) for each sentence set and conducted an evaluation using a range of sentence prompts and various phenotype terms. Results: When the phenotype terms for normalization were included in the fine-tuning corpora, both models demonstrated nearly perfect performance, averaging over 99% accuracy. In comparison, ChatGPT-3.5 has only ∼20% accuracy in identifying HPO IDs for phenotype terms. When single-character typos were introduced in the phenotype terms, the accuracy of NAME and NAME+SYN is 10.2% and 36.1%, respectively, but increases to 61.8% (NAME+SYN) with additional typo-specific fine-tuning. For terms sourced from HPO vocabularies as unseen synonyms, the NAME model achieved 11.2% accuracy, while the NAME+SYN model achieved 92.7% accuracy. Conclusion: Our fine-tuned models demonstrate ability to normalize phenotype terms unseen in the fine-tuning corpus, including misspellings, synonyms, terms from other ontologies, and laymen's terms. Our approach provides a solution for the use of LLM to identify named medical entities from the clinical narratives, while successfully normalizing them to standard concepts in a controlled vocabulary.

Enhancing phenotype recognition in clinical notes using large language models: PhenoBCBERT and PhenoGPT.

To enhance phenotype recognition in clinical notes of genetic diseases, we developed two models-PhenoBCBERT and PhenoGPT-for expanding the vocabularies of Human Phenotype Ontology (HPO) terms. While HPO offers a standardized vocabulary for phenotypes, existing tools often fail to capture the full scope of phenotypes due to limitations from traditional heuristic or rule-based approaches. Our models leverage large language models to automate the detection of phenotype terms, including those not in the current HPO. We compare these models with PhenoTagger, another HPO recognition tool, and found that our models identify a wider range of phenotype concepts, including previously uncharacterized ones. Our models also show strong performance in case studies on biomedical literature. We evaluate the strengths and weaknesses of BERT- and GPT-based models in aspects such as architecture and accuracy. Overall, our models enhance automated phenotype detection from clinical texts, improving downstream analyses on human diseases.

Classification of integers based on residue classes via modern deep learning algorithms.

Judging whether an integer can be divided by prime numbers such as 2 or 3 may appear trivial to human beings, but it can be less straightforward for computers. Here, we tested multiple deep learning architectures and feature engineering approaches to classifying integers based on their residues when divided by small prime numbers. We found that the ability of classification critically depends on the feature space. We also evaluated automated machine learning (AutoML) platforms from Amazon, Google, and Microsoft and found that, without appropriately engineered features, they failed on this task. Furthermore, we introduced a method that utilizes linear regression on Fourier series basis vectors and demonstrated its effectiveness. Finally, we evaluated large language models (LLMs) such as GPT-4, GPT-J, LLaMA, and Falcon, and we demonstrated their failures. In conclusion, feature engineering remains an important task to improve performance and increase interpretability of machine learning models, even in the era of AutoML and LLMs.

Use of Disease-Modifying Treatments in Patients With Sickle Cell Disease.

Importance: Despite hydroxyurea being an established treatment for sickle cell disease (SCD), it remains underused. The recent approval of the disease-modifying treatments (DMTs) l-glutamine, crizanlizumab, and voxelotor underscores the need to understand the uptake of DMTs in the current treatment landscape. Objective: To explore characteristics that may be associated with DMT use and to describe observed patterns of yearly DMT use from 2014 to 2021. Design, Setting, and Participants: This cross-sectional study used administrative claims data from Optum's deidentified Clinformatics Data Mart Database from January 1, 2014, to September 30, 2021, to identify adults and children with SCD. Data were analyzed from August 1, 2022, to August 28, 2023. Exposure: Use of DMTs. Main Outcomes and Measures: Patient characteristics across groups with varying patterns of DMT use and yearly patterns of prescription fills for hydroxyurea, crizanlizumab, voxelotor, and l-glutamine. Results: A total of 5022 beneficiaries with SCD (2081 [41.4%] aged 18-45 years; 2929 [58.3%] female) were included in sample A (144 [2.9%] inconsistent users, 274 [5.5%] incident users, 892 [17.8%] consistent users, and 3712 [73.9%] non-DMT users). Inconsistent users had a higher prevalence of vaso-occlusive crises (mean [SD], 3.7 [4.7]), splenic complications (6 of 144 [4.2%]), pulmonary complications (36 of 144 [25.0%]), kidney disease (21 of 144 [14.6%]), acute chest syndrome (18 of 144 [12.5%]), and health care visits (eg, mean [SD] inpatient visits, 7.0 [10.7]) compared with the other use groups. Non-DMT users had the lowest prevalence of vaso-occlusive crises (mean [SD], 0.8 [2.4]), acute chest syndrome (109 of 3712 [2.9%]), and inpatient (mean [SD], 2.0 [6.6]) and emergency department (mean [SD], 0.7 [3.1]) visits and the highest proportion of adults 65 years and older (593 of 3712 [16.0%]). In sample B (6387 beneficiaries with SCD), hydroxyurea use modestly increased from 428 of 2188 participants (19.6%) in 2014 to 701 of 2880 (24.3%) in 2021. Use of l-glutamine increased briefly but gradually decreased throughout the study period. In 2021, out of 2880 participants, 102 (3.5%) had at least 1 fill for crizanlizumab and 131 (4.6%) had at least 1 fill for voxelotor. Overall, total DMT use increased from 428 of 2188 participants (19.6%) in 2014 to 815 of 2880 patients (28.3%) in 2021. Conclusions and Relevance: In this cross-sectional analysis of adults and children with SCD, uptake of DMTs remained low from 2014 to 2021, despite the approval of newer therapies. Notable differences in patient characteristics across varied DMT exposure types necessitate further exploration into factors that facilitate DMT use and the creation of strategies to enhance DMT uptake.

Enhancing Phenotype Recognition in Clinical Notes Using Large Language Models: PhenoBCBERT and PhenoGPT.

To enhance phenotype recognition in clinical notes of genetic diseases, we developed two models - PhenoBCBERT and PhenoGPT - for expanding the vocabularies of Human Phenotype Ontology (HPO) terms. While HPO offers a standardized vocabulary for phenotypes, existing tools often fail to capture the full scope of phenotypes, due to limitations from traditional heuristic or rule-based approaches. Our models leverage large language models (LLMs) to automate the detection of phenotype terms, including those not in the current HPO. We compared these models to PhenoTagger, another HPO recognition tool, and found that our models identify a wider range of phenotype concepts, including previously uncharacterized ones. Our models also showed strong performance in case studies on biomedical literature. We evaluated the strengths and weaknesses of BERT-based and GPT-based models in aspects such as architecture and accuracy. Overall, our models enhance automated phenotype detection from clinical texts, improving downstream analyses on human diseases.

Effects of carbohydrate drinks ingestion on executive function in athletes: a systematic review and meta-analysis.

Background: Carbohydrates are often used as boosters for endurance and high-intensity exercise. However, it is unclear whether carbohydrate drinks intake before or during exercise can affect specific domains of cognitive function, such as Executive Function (EF). Methods: Following the guidance of PRISMA 2020, we searched six major databases including PubMed, WOS, SPORTDiscus, Cochrane, Embase, and Scopus. Outcomes were presented in the form of Reaction Time (RT), Accuracy (ACC), and Scores (Score) for performing EF tests. Effect sizes were calculated from the test results of EF and expressed as standardized mean differences (SMDs). After analyzing the overall results, we performed subgroup analyses based on the athletes' program characteristics. Results: After retrieving a total of 5,355 articles, ten randomized controlled trials (RCTs) were identified and included in this review. The overall results showed that the intake of carbohydrate drinks before or during exercise did not have a significant effect on the reduction of EF after exercise (ACC (-0.05 [-0.27, 0.18]); RT (-0.18 [-0.45, 0.09]); Score (0.24 [-0.20, 0.68])). The subgroup analyses based on open skill sports and close skill sports also showed invalid results, but the results of RT ended up with different preference (ACC of open skill sports athletes (-0.10 [-0.34, 0.14]); RT of open skill athletes (-0.27 [-0.60, 0.07]); RT of close skill athletes (0.29 [-0.24, 0.82])). Conclusion: The intake of 6-12% of single or mixed carbohydrates before or during exercise was not significantly effective in reducing the decline in EF after exercise. Our findings may have been influenced by the type of intervention, dose, mode of administration, or individual variability of the included subjects.

Unequal crossing over between CYP11B2 and CYP11B1 causes 11 ß -hydroxylase deficiency in a consanguineous family.

Cytochrome P450 (CYP) family CYP11B2/CYP11B1 chimeric genes have been shown to arise from unequal crossing over of the genes encoding aldosterone synthase (CYP11B2) and 11ß-hydroxylase (CYP11B1) during meiosis. The activity deficiency or impaired activity of aldosterone synthase and 11ß-hydroxylase resulting from these chimeric genes are important reasons for 11ß-hydroxylase deficiency (11ß-OHD). Here，two patients with pseudoprecocious puberty and hypokalemia hypertension and three carriers in a consanguineous marriage family were studied. A single CYP11B2/CYP11B1 chimera consisting of the promoter and exons 1 through 5 of CYP11B2, exons 8 and 9 of CYP11B1, and a breakpoint consisting of part of exon 6 of CYP11B2 and part of exon 6, intron 6, and exon 7 of CYP11B1 were detected in the patients and carriers. At the breakpoint of the chimera, a c 0.1086 G > C ( p.Leu.362 =) synonymous mutation in exon 6 of CYP11B2, a c 0.1157 C＞G(p. A386V) missense mutation in exon 7 of CYP11B1, and an intronic mutation in intron 6 were detected. The allele model of the CYP11B2/CYP11B1 chimera demonstrated homozygosity and heterozygosity in the patients and the carriers, respectively. Molecular docking and enzymatic activity analyses indicated that the CYP11B2/CYP11B1 chimeric protein interacted with the catalytic substrate of aldosterone synthase and had similar enzymatic activity to aldosterone synthase. Our study indicated that deletion of CYP11B1 and CYP11B2 abolished the enzymatic activity of 11 ß-hydroxylase and aldosterone synthase; however, the compensation of the enzymatic activity of aldosterone synthase by the CYP11B2/CYP11B1 chimeric protein maintained normal aldosterone levels in vitro. All of the above findings explained the 11ß-OHD phenotypes of the proband and patients in the family.

Trajectories of Pelvic Floor Symptoms and Support After Vaginal Delivery in Primiparous Women Between Third Trimester and 1 Year Postpartum.

OBJECTIVES: The objectives of this study were to describe trajectories of pelvic floor symptoms and support from the third trimester to 1 year postpartum in primiparous women after vaginal delivery and to explore factors associated with their resolution between 8 weeks postpartum and 1 year postpartum. METHODS: Five hundred ninety-seven nulliparous women 18 years or older who gave birth vaginally at term completed the Epidemiology of Prolapse and Incontinence Questionnaire and the Pelvic Organ Prolapse Quantification examination at the third trimester, 8 weeks postpartum, and 1 year postpartum. RESULTS: At 1 year postpartum, 41%, 32%, and 23% of participants reported stress urinary incontinence, nocturia, and flatus incontinence, respectively, and 9% demonstrated maximal vaginal descent (MVD) ≥ 0 cm. For more common symptoms, incidence rates between the third trimester and 8 weeks postpartum ranged from 6% for urinary frequency to 22% for difficult bowel movements, and resolution rates between 8 weeks postpartum and 1 year postpartum ranged from 23% for stress urinary incontinence to 73% for pain. Between the third trimester and 8 weeks postpartum, 13% demonstrated de novo MVD ≥ 0 cm. For most symptoms, the presence of the same symptom before delivery decreased the probability of resolution between 8 weeks postpartum and 1 year. However, the sensitivities of predelivery vaginal bulge and MVD of 0 cm or greater for those outcomes at 1 year postpartum was overall low (10-12%). CONCLUSIONS: One year postpartum, urinary and bowel symptoms are common in primiparous women who gave birth vaginally. A substantial portion of this burden is represented by symptoms present before delivery, while most of the prevalence of worse anatomic support is accounted for by de novo changes after delivery.

Effects of Oxytocin for Induction and Augmentation of Labor on Pelvic Floor Symptoms and Support in the Postpartum Period.

OBJECTIVE: The aim of the study was to determine whether oxytocin for induction or augmentation of labor impacts the incidence or persistence of pelvic floor symptoms and support 5 to 10 weeks after first vaginal delivery. METHODS: Participants in this prospective cohort study were nulliparous women 18 years or older that delivered vaginally at 37 weeks gestation or more and completed the Epidemiology of Prolapse and Incontinence Questionnaire (EPIQ) and the Pelvic Organ Prolapse Quantification examination in third trimester and 5 to 10 weeks postpartum. We compared the incidence and persistence of symptomatic EPIQ domains and worse vaginal support (maximal vaginal descent ≥0 cm) between women who received oxytocin with those that did not (with or without prostaglandin or mechanical methods in both groups). We performed modified binomial regression to calculate adjusted relative risks of each outcome with 95% confidence intervals. RESULTS: The mean (SD) age of the 722 participants was 28.3 (5.2) years; 20% were Hispanic. There were no significant differences according to oxytocin exposure in either incidence or persistence of symptomatic EPIQ domains or worse vaginal support. We found similar results in sensitivity analyses comparing women who received oxytocin as the sole pharmacologic agent to women who received no pharmacologic agent. After adjusting for demographic and obstetric factors associated with incidence and persistence of symptoms and support, oxytocin exposure continued to have no effect. CONCLUSIONS: Oxytocin during labor does not significantly increase the risks for the incidence or persistence of pelvic floor symptoms or worse vaginal support in the early postpartum period, although power for less frequent outcomes was limited.

Treatment of open and comminuted mid-distal tibial fractures by bilateral external fixation combined with limited-internal fixation.

Open and comminuted mid-distal fractures often result from high-energy trauma, and a concomitant poor blood supply often leads to skin necrosis, infection, and bone union. To circumvent such complications, we used limited-reduction and bilateral-external fixators to treat open and comminuted mid-distal tibial fractures with compromised soft tissue. A retrospective series of 34 patients who had open and comminuted mid-distal tibial fractures and treated by bilateral-external fixators with limited-internal fixation were analyzed. Patients were followed for 10-25 months (mean: 12 months) post-treatment and osseous union was achieved in each case. The average union time was 16.3 weeks. Based on the Johner- Wruhs criteria, the retrospective series consisted of 21 'excellent' cases, 8 'good' cases, 4 'fair' cases, and a 'poor' case. The total percentage of 'excellent' and 'good' cases of fracture recovery was 85.29%. Bilateral-external and limited-internal fixators pro- vided high bone union rate and excellent ankle-joint motion. Hence, it is an appropriate surgical approach for treating open and comminuted mid-distal tibial fractures with compromised soft tissue.

Procalcitonin and N-Terminal Pro-B-Type Natriuretic Peptide for Prognosis in Septic Acute Kidney Injury Patients Receiving Renal Replacement Therapy.

BACKGROUND: Sepsis is a complex clinical syndrome leading to severe sepsis and septic shock. It is very common in the intensive care unit with high mortality. Thus, judging its prognosis is extremely important. Procalcitonin (PCT) and -N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels are commonly elevated in sepsis patients, but only a few are discussed in the septic acute kidney injury patients (AKI) who received renal replacement therapy (RRT). Our study is aimed at investigating the prognostic value of PCT and NT-proBNP in septic AKI patients who received RRT. METHODS: This was a retrospective study of septic AKI patients who underwent RRT in a Chinese university hospital. All enrolled patients tested PCT and NT-proBNP at RRT initiation. PCT and NT-proBNP levels were compared between the survivors and non-survivors. Receiver operating characteristic (ROC) curves of the 2 biomarkers were performed for predicting in-hospital mortality. According to the median value of PCT (16.2 ng/mL) and NT-proBNP (10,271 pg/mL), patients were divided into 4 groups (low PCT and low NT-proBNP; high PCT and low NT-proBNP; low PCT and high NT-proBNP; high PCT and high NT-proBNP). The Kaplan-Meier survival curves were used to analyze the 28-day survival rate in the 4 groups. RESULTS: A total of 81 patients were enrolled in the study. Of which, 48 (59.3%) patients died during hospitalization. The median of NT-proBNP in non-survivors was significantly higher than in survivors (p = 0.001), while PCT had no significant difference (p = 0.412). The area under the ROC curve of PCT and NT-proBNP for predicting in-hospital mortality was 0.561 (95% CI 0.426-0.695) and 0.729 (95% CI 0.604-0.854). Kaplan-Meier survival curve analysis showed that increased NT-proBNP level was associated with 28-day mortality while combined with PCT there was no statistical difference in 4 different level groups. CONCLUSION: NT-proBNP has a certain predictive value for the prognosis in septic AKI patients who received RRT. It seems that the initial PCT value for prognosis is limited. The combination of PCT and -NT-proBNP to evaluate the prognosis in these critically ill patients is currently unclear.

Trajectories of lower back, upper back, and pelvic girdle pain during pregnancy and early postpartum in primiparous women.

OBJECTIVE: Most studies about truncal pain during and after pregnancy focus on low back pain, few prospectively define change in pain, and even fewer evaluate pain in all three major truncal areas: upper back, lower back, and pelvic girdle. Thus, the objective of this prospective cohort study was to describe, in primiparous women delivered vaginally, prevalence rates and severities of upper back, lower back, and pelvic girdle pain during pregnancy and 6-10 weeks postpartum and to describe the trajectory of pain constellations between time points. STUDY DESIGN: Participants completed questionnaires at each time point. Pain intensity was rated on a visual analogue scale ranging from 0 to 100. RESULTS: Of the 288 participants, 94% reported truncal pain during pregnancy, while 75% did so postpartum. Prevalence rates of upper back, lower back, and pelvic girdle pain with or without other types of pain during pregnancy were 42%, 77%, and 74% and postpartum were 43%, 52%, and 41%, respectively. Pain severity was highest for women endorsing pain in three locations (median 55-60). Of women with the most common pain constellation during pregnancy, lower back and pelvic girdle (32%), 18% had persistent low back and pelvic girdle pain postpartum, 20% had no pain, and the remainder had pain in a different location. Of women with pain in all three locations during pregnancy (27%), 34% had persistent pain in three locations postpartum, 13% had no pain, and the remainder had pain in at least one location. CONCLUSION: More women experience pain in a constellation of locations than in a single location. Severity increases as number of pain sites increase. Women with pain in three sites during pregnancy are least likely to have pain resolve. Interventions should focus on the entire trunk and not simply one site of pain.

MondoA-Mlx transcriptional activity is limited by mTOR-MondoA interaction.

Mammalian target of rapamycin (mTOR) integrates multiple signals, including nutrient status, growth factor availability, and stress, to regulate cellular and organismal growth. How mTOR regulates transcriptional programs in response to these diverse stimuli is poorly understood. MondoA and its obligate transcription partner Mlx are basic helix-loop-helix leucine zipper (bHLHZip) transcription factors that sense and execute a glucose-responsive transcriptional program. MondoA-Mlx complexes activate expression of thioredoxin-interacting protein (TXNIP), which is a potent inhibitor of cellular glucose uptake and aerobic glycolysis. Both mTOR and MondoA are central regulators of glucose metabolism, yet whether they interact physically or functionally is unknown. We show that inhibition of mTOR induces MondoA-dependent expression of TXNIP, coinciding with reduced glucose uptake. Mechanistically, mTOR binds to MondoA in the cytoplasm and prevents MondoA-Mlx complex formation, restricting MondoA's nuclear entry and reducing TXNIP expression. Further, we show that mTOR inhibitors and reactive oxygen species (ROS) regulate interaction between MondoA and mTOR in an opposing manner. Like mTOR's suppression of the MondoA-TXNIP axis, MondoA can also suppress mTOR complex 1 (mTORC1) activity via its direct transcriptional regulation of TXNIP. Collectively, these studies reveal a regulatory relationship between mTOR and the MondoA-TXNIP axis that we propose contributes to glucose homeostasis.