Fertility and pregnancy outcomes in primary hyperparathyroidism: Observations from a large insured population.

CONTEXT: Primary hyperparathyroidism (PHPT) has initially been implicated in adverse maternal and neonatal outcomes, while subsequent population studies have failed to show an association. OBJECTIVE: To compare maternal, pregnancy, and neonatal outcomes in patients with and without PHPT. DESIGN: Retrospective matched-cohort study (2005-2020). SETTING: An integrated healthcare delivery system in Southern California. PATIENTS: Women aged 18-44 years were included. Patients with biochemical diagnosis of PHPT were matched 1:3 with eucalcemic controls (non-PHPT). MAIN OUTCOME MEASURES: Achievement of pregnancy, pregnancy outcomes (including rates of abortion, maternal complications), and neonatal outcomes (including hypocalcemia, need for intensive care). RESULTS: The cohort comprised 386 women with PHPT and 1158 age-matched controls. Pregnancy rates between PHPT and control groups were similar (10.6% vs 12.8%). The adjusted rate ratio of pregnancy was 0.89 (95% CI: 0.64-1.24) (PHPT vs non-PHPT). Twenty-nine pregnancies occurred in women with co-existing PHPT and 191 pregnancies occurred in controls, resulting in 23 (79.3%) and 168 (88.0%) live births, respectively (p=0.023). Neonatal outcomes were similar. Live birth rates were similar (86.4%, 80%, 79.2%) for those undergoing parathyroidectomy prior (n=22), during (n=5), or after pregnancy/never (n=24). Among patients who underwent parathyroidectomy during pregnancy, no spontaneous abortions occurred in women entering pregnancy with peak calcium <11.5 mg/dL [2.9 mmol/L]. CONCLUSIONS: We observed no difference in pregnancy rates between women with or without PHPT. Performing parathyroidectomy before pregnancy or during the second trimester appears to be a safe and successful strategy, and adherence to this strategy may be most critical for patients with higher calcium levels (≥11.5 mg/dL [2.9 mmol/L]).

From Bench-to-Bedside: How Artificial Intelligence is Changing Thyroid Nodule Diagnostics, a Systematic Review.

CONTEXT: Use of artificial intelligence (AI) to predict clinical outcomes in thyroid nodule diagnostics has grown exponentially over the past decade. The greatest challenge is in understanding the best model to apply to one's own patient population, and how to operationalize such a model in practice. EVIDENCE ACQUISITION: A literature search of PubMed and IEEE Xplore was conducted for English-language publications between January 1, 2015 and January 1, 2023, studying diagnostic tests on suspected thyroid nodules that used AI. We excluded articles without prospective or external validation, nonprimary literature, duplicates, focused on nonnodular thyroid conditions, not using AI, and those incidentally using AI in support of an experimental diagnostic outside standard clinical practice. Quality was graded by Oxford level of evidence. EVIDENCE SYNTHESIS: A total of 61 studies were identified; all performed external validation, 16 studies were prospective, and 33 compared a model to physician prediction of ground truth. Statistical validation was reported in 50 papers. A diagnostic pipeline was abstracted, yielding 5 high-level outcomes: (1) nodule localization, (2) ultrasound (US) risk score, (3) molecular status, (4) malignancy, and (5) long-term prognosis. Seven prospective studies validated a single commercial AI; strengths included automating nodule feature assessment from US and assisting the physician in predicting malignancy risk, while weaknesses included automated margin prediction and interobserver variability. CONCLUSION: Models predominantly used US images to predict malignancy. Of 4 Food and Drug Administration-approved products, only S-Detect was extensively validated. Implementing an AI model locally requires data sanitization and revalidation to ensure appropriate clinical performance.

A multitask approach for automated detection and segmentation of thyroid nodules in ultrasound images.

An increase in the incidence and diagnosis of thyroid nodules and thyroid cancer underscores the need for a better approach to nodule detection and risk stratification in ultrasound (US) images that can reduce healthcare costs, patient discomfort, and unnecessary invasive procedures. However, variability in ultrasound technique and interpretation makes the diagnostic process partially subjective. Therefore, an automated approach that detects and segments nodules could improve performance on downstream tasks, such as risk stratification. Ultrasound studies were acquired from 280 patients at UCLA Health, totaling 9888 images, and annotated by collaborating radiologists. Current deep learning architectures for segmentation are typically semi-automated because they are evaluated solely on images known to have nodules and do not assess ability to identify suspicious images. However, the proposed multitask approach both detects suspicious images and segments potential nodules; this allows for a clinically translatable model that aptly parallels the workflow for thyroid nodule assessment. The multitask approach is centered on an anomaly detection (AD) module that can be integrated with any UNet architecture variant to improve image-level nodule detection. Of the evaluated multitask models, a UNet with a ImageNet pretrained encoder and AD achieved the highest F1 score of 0.839 and image-wide Dice similarity coefficient of 0.808 on the hold-out test set. Furthermore, models were evaluated on two external validations datasets to demonstrate generalizability and robustness to data variability. Ultimately, the proposed architecture is an automated multitask method that expands on previous methods by successfully both detecting and segmenting nodules in ultrasound.

Diagnostic performance of molecular testing in indeterminate (Bethesda III and IV) thyroid nodules with Hürthle cell cytology.

BACKGROUND: Indeterminate thyroid nodules with Hürthle cell cytology remain a diagnostic challenge. The low benign call rate and positive predictive value of first-generation molecular tests precluded their use to rule out malignancy. We examined the diagnostic performance of current tests. METHOD: This subset analysis of our prospective randomized trial compared the benign call rate and positive predictive value of Afirma Gene Sequencing Classifier and Thyroseq v3 in Bethesda III and IV nodules with Hürthle cell cytology. Molecular test samples were obtained at initial fine-needle aspiration (8/2017-7/2022) and reflexively sent for processing. RESULTS: Molecular testing was performed on 140 Hürthle cell nodules. Of 79 nodules tested with the Afirma Gene Sequencing Classifier, the benign call rate was 84% (66/79). Nine of 66 nodules with benign results were resected, with no malignancies. Twelve of 13 nodules with suspicious results were resected, revealing 3 malignancies-2 papillary thyroid carcinomas and one Hürthle cell carcinoma (positive predictive value 25%). Of 61 nodules tested with Thyroseq v3, the benign call rate was 56% (34/61; (P < .01 versus Afirma Gene Sequencing Classifier). Five of 34 nodules with negative results were resected, with no malignancies. Nineteen of 27 nodules with positive results were resected, revealing 3 malignancies-2 papillary thyroid carcinomas and 1 Hürthle cell carcinoma (positive predictive value 16%). CONCLUSION: The high benign call rate of current molecular tests in Hürthle cell nodules strengthens their value in enabling patients to avoid surgery.

A Multitask Approach for Automated Detection and Segmentation of Thyroid Nodules in Ultrasound Images.

An increase in the incidence and diagnosis of thyroid nodules and thyroid cancer underscores the need for a better approach to nodule detection and risk stratification in ultrasound (US) images that can reduce healthcare costs, patient discomfort, and unnecessary invasive procedures. However, variability in ultrasound technique and interpretation makes the diagnostic process partially subjective. Therefore, an automated approach that detects and segments nodules could improve performance on downstream tasks, such as risk stratification.Current deep learning architectures for segmentation are typically semi-automated because they are evaluated solely on images known to have nodules and do not assess ability to identify suspicious images. However, the proposed multitask approach both detects suspicious images and segments potential nodules; this allows for a clinically translatable model that aptly parallels the workflow for thyroid nodule assessment. The multitask approach is centered on an anomaly detection (AD) module that can be integrated with any U-Net architecture variant to improve image-level nodule detection. Ultrasound studies were acquired from 280 patients at UCLA Health, totaling 9,888 images, and annotated by collaborating radiologists. Of the evaluated models, a multi-scale UNet (MSUNet) with AD achieved the highest F1 score of 0.829 and image-wide Dice similarity coefficient of 0.782 on our hold-out test set. Furthermore, models were evaluated on two external validations datasets to demonstrate generalizability and robustness to data variability. Ultimately, the proposed architecture is an automated multitask method that expands on previous methods by successfully both detecting and segmenting nodules in ultrasound.

Automated extraction of incidental adrenal nodules from electronic health records.

BACKGROUND: Many adrenal incidentalomas do not undergo appropriate biochemical testing and complete imaging characterization to assess for hormone hypersecretion and malignancy. With the growing availability of clinical narratives in the electronic medical record, automated surveillance using advanced data analytic techniques may represent a promising method to improve management. METHODS: A data provisioning process using a series of structured query language scripts was used to abstract all chest and abdominal computed tomography and magnetic resonance imaging reports from an academic health care system in 2018. The narratives and impressions were queried for key text relating to the identification of adrenal incidentalomas. Patients with a history of extra-adrenal malignancy undergoing staging or surveillance imaging were excluded. The prevalence and radiographic characteristics were analyzed. Patients with adrenal incidentalomas newly identified in 2018 were assessed for biochemical testing and nodule stability through August 2021. RESULTS: Of 36,618 patients queried, 8,557 were excluded owing to a history of extra-adrenal malignancy. Data from 447 patients were flagged by the structured query language scripts and electronically abstracted. On internal validation, 307/447 (69%) patients were correctly identified as having adrenal nodules (1.1% overall prevalence). The median patient age was 67 years, and 56% were female. The median nodule size was 1.7 (IQR 1.3-2.5) cm, 9% were bilateral, and 63% were low density (unenhanced Hounsfield units <10). Adrenal carcinoma was identified in 10 patients. In 2018, 121 patients were diagnosed with a new adrenal incidentaloma. Of 32 (27%) patients who had follow-up imaging at a median of 1.9 years, 97% of nodules were stable in size. Biochemical testing was performed in 53 patients (44%), of which 31 (26%) had complete hormonal assessment; 14 (26%) were functional nodules: 7 aldosterone-secreting, 4 cortisol-secreting, and 3 pheochromocytoma. CONCLUSION: Only one-fourth of patients received appropriate biochemical testing after incidental diagnosis of an adrenal nodule, and most nodules with indeterminate imaging characteristics did not undergo follow-up imaging. Advanced data analytic techniques on electronic imaging reports may aid in the clinical identification and improved management of patients with adrenal incidentalomas.

Patient-level thyroid cancer classification using attention multiple instance learning on fused multi-scale ultrasound image features.

For patients with thyroid nodules, the ability to detect and diagnose a malignant nodule is the key to creating an appropriate treatment plan. However, assessments of ultrasound images do not accurately represent malignancy, and often require a biopsy to confirm the diagnosis. Deep learning techniques can classify thyroid nodules from ultrasound images, but current methods depend on manually annotated nodule segmentations. Furthermore, the heterogeneity in the level of magnification across ultrasound images presents a significant obstacle to existing methods. We developed a multi-scale, attention-based multiple-instance learning model which fuses both global and local features of different ultrasound frames to achieve patient-level malignancy classification. Our model demonstrates improved performance with an AUROC of 0.785 (p<0.05) and AUPRC of 0.539, significantly surpassing the baseline model trained on clinical features with an AUROC of 0.667 and AUPRC of 0.444. Improved classification performance better triages the need for biopsy.

Nasal Feeding Tubes Are Associated with Fewer Adverse Events than Feeding via Ostomy in Hospitalized Patients Receiving Enteral Nutrition.

BACKGROUND: Surgical feeding ostomies (eg, gastrostomy) have become required by many nursing facilities for all patients receiving enteral nutrition, whether for short- or long-term use. These policies lack supportive evidence. Comparisons of adverse event rates between surgical and natural orifice tubes are few and lacking in the inpatient setting. Additionally, we hypothesize that adverse events related to feeding tubes are underreported. We sought to quantify adverse events to test the relative safety of surgical feeding ostomies and natural orifice (eg. nasogastric or orogastric) feeding tubes in hospitalized patients. METHODS: This was a prospective observational cohort study of enterally fed inpatients using semiweekly focused physical examination, scripted survey, and chart review. RESULTS: All tube-fed patients admitted to a large, urban, academic hospital received semiweekly bedside evaluation and chart review over a 9-week period (n = 226 unique patients, mean 6.25 visits each, total 1118 observations). Demographics were comparable between 148 subjects with natural orifice and 113 subjects with surgical feeding tubes. A higher incidence of adverse events was observed with surgical tubes (3.34 vs 1.25 events per 100 subject days, P < .001). Only 50% of all adverse events were documented in the medical record. More patients with surgical tubes were discharged to skilled nursing facilities (58% vs 24%). CONCLUSIONS: Surgical feeding tubes are associated with significantly higher in-hospital adverse event rates when compared with natural orifice (nasal or oral) feeding tubes. Policies requiring surgical feeding ostomies should be reevaluated.

Attenuated familial adenomatous polyposis (AFAP) in a patient associated with a novel mutation in APC.

Familial adenomatous polyposis (FAP) is an autosomal dominant syndrome associated with mutation in the adenomatous polyposis coli (APC) gene, a tumour suppressor located on chromosome 5q21. Attenuated familial adenomatous polyposis (AFAP) is a variant associated with fewer and later onset of colon polyps. AFAP-associated APC mutations have largely been found before codon 157, in exon 9 or after codon 1595. We present the case of a 44-year-old man incidentally found to have numerous gastric polyps during bariatric surgery, with innumerable polyps in the remaining part of the stomach and the entire colon, with rectal sparing, consistent with AFAP phenotype. Genetic testing demonstrated the c.7682dup (p.Ser2562Lysfs*21) variant in exon 15 of APC. This represents a previously undescribed APC mutation. This mutation likely yields end-binding protein 1 and human disc large binding protein inactivation, causing cell cycle microtubule dysregulation and tumour suppressor inactivation. Through loss of these regulatory mechanisms, this mutation is associated with AFAP phenotype. The patient was treated surgically and is doing well.

Zinc Deficiency With Dermatitis in a Parenteral Nutrition-Dependent Patient Due to National Shortage of Trace Minerals.

The shortages of intravenous drugs remains critical, with sterile injectables accounting for 80% of the approximately 300 shortages. The impact is being felt in patients dependent on parenteral nutrition (PN), and severe deficiencies are becoming more commonplace. We report here a man who developed severe zinc deficiency, manifesting as a painful desquamative rash, due to an inability to obtain multi-trace element additives for his PN.