Historical and future trends in emergency pituitary referrals: a machine learning analysis.

PURPOSE: Acute pituitary referrals to neurosurgical services frequently necessitate emergency care. Yet, a detailed characterisation of pituitary emergency referral patterns, including how they may change prospectively is lacking. This study aims to evaluate historical and current pituitary referral patterns and utilise state-of-the-art machine learning tools to predict future service use. METHODS: A data-driven analysis was performed using all available electronic neurosurgical referrals (2014-2021) to the busiest U.K. pituitary centre. Pituitary referrals were characterised and volumes were predicted using an auto-regressive moving average model with a preceding seasonal and trend decomposition using Loess step (STL-ARIMA), compared against a Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) algorithm, Prophet and two standard baseline forecasting models. Median absolute, and median percentage error scoring metrics with cross-validation were employed to evaluate algorithm performance. RESULTS: 462 of 36,224 emergency referrals were included (referring centres = 48; mean patient age = 56.7 years, female:male = 0.49:0.51). Emergency medicine and endocrinology accounted for the majority of referrals (67%). The most common presentations were headache (47%) and visual field deficits (32%). Lesions mainly comprised tumours or haemorrhage (85%) and involved the pituitary gland or fossa (70%). The STL-ARIMA pipeline outperformed CNN-LSTM, Prophet and baseline algorithms across scoring metrics, with standard accuracy being achieved for yearly predictions. Referral volumes significantly increased from the start of data collection with future projected increases (p < 0.001) and did not significantly reduce during the COVID-19 pandemic. CONCLUSION: This work is the first to employ large-scale data and machine learning to describe and predict acute pituitary referral volumes, estimate future service demands, explore the impact of system stressors (e.g. COVID pandemic), and highlight areas for service improvement.

IgG4-related hypophysitis: a retrospective cohort study.

PURPOSE: IgG4-related hypophysitis (IgG4-RH) is a rare chronic inflammatory condition of the pituitary gland. This study reports the presentation, management and outcomes for patients with histologically proven IgG4-related hypophysitis. METHODS: A prospectively maintained electronic database was searched over a 14-year period from 1 January 2007 to 31 December 2020 at a single academic centre to identify all patients with a histological diagnosis of IgG4-RH. A retrospective case note review from electronic health records was conducted for each case to extract data on their presentation, management and outcomes. RESULTS: A total of 8 patients (5 male) with a median age of 51 years were identified. The most common presenting symptoms were headache (4/8; 50%), fatigue (3/8; 37.5%) and visual impairment (2/8; 25%). Three patients were initially treated with high-dose steroids aiming for reduction of the pituitary mass. However, ultimately all patients underwent transsphenoidal surgery. Post-operative changes included radiological reduction in pituitary mass in all patients that had imaging (7/7; 100%), improvement in vision (1/2; 50%), residual thick pituitary stalk (5/7; 71.4%), persistent anterior hypopituitarism (4/8; 50%) and panhypopopituitarism including diabetes insipidus (3/8; 37.5%). CONCLUSIONS: IgG4-RH is an increasingly recognised entity presenting with a variety of symptoms and signs. Clinical presentation is similar to other forms of hypophysitis. It is therefore important to consider IgG4-RH as a differential and to have a low threshold for pituitary biopsy, the diagnostic gold standard. The diagnosis of IgG4-RH will guide decisions for additional workup for IgG4-related disease, multi-disciplinary team involvement and follow-up.

The use of computed tomography as a first-line imaging modality in patients with primary hyperparathyroidism.

BACKGROUND: The success of minimally invasive parathyroidectomy (MIP) relies on accurate localization of the abnormal parathyroid glands. Concordant findings on ultrasound (US) and 99mTc-scintigraphy (sestamibi) are currently considered the 'gold standard'. Computed tomography (CT) has also recently been used in preoperative planning. We sought to assess the accuracy of CT for localization of abnormal parathyroid glands in such patients. METHODS: An audit of 75 patients with primary hyperparathyroidism (PHPT) who underwent neck US and CT between 2017 and 2019 at our center as their first-line imaging. RESULTS: All 75 patients underwent US and CT and 54 (72.0%) also had sestamibi. CT alone identified a potential target in all patients, of which the location was correct in 63 (84.0%). The overall combined sensitivity of US and CT was 88% (95% CI 78-94) and was higher than the combined sensitivity of US and sestamibi (65% [95% CI 53-76]; p < 0.001). Twenty-one patients (28.0%) had an ectopic gland, and the sensitivity of US and CT was 86% (95% CI 64-96) versus US and sestamibi (57% [95% CI 34-77]; p = 0.016). For adenomas < 1.0 g (n = 36; 48%), the accuracy of CT was 81% (95% CI 64-91) compared with 62% (95% CI 44-77) for US and sestamibi (p = 0.04). The correct preoperative diagnosis of multiglandular disease (n = 9; 12%) seemed to be the most difficult, with similar accuracy for US and sestamibi (40% [95% CI 14-73]) and US and CT (50% [95% CI 20-80]) (p > 0.99). CONCLUSION: The combination of US and CT was able to correctly identify the location of the abnormal parathyroid in 88% of patients and, in comparison with US and sestamibi, had better diagnostic accuracy, especially for smaller and ectopic adenomas. This finding suggests that US and CT could be considered as a first-line imaging modality in patients with PHPT considered for MIP.

SOCIETY FOR ENDOCRINOLOGY CLINICAL GUIDANCE: Inpatient management of cranial diabetes insipidus.

Cranial diabetes insipidus (CDI) is a treatable chronic condition that can potentially develop into a life-threatening medical emergency. CDI is due to the relative or absolute lack of the posterior pituitary hormone vasopressin (AVP), also known as anti-diuretic hormone. AVP deficiency results in uncontrolled diuresis. Complete deficiency can lead to polyuria exceeding 10 L/24 h. Given a functioning thirst mechanism and free access to water, patients with CDI can normally maintain adequate fluid balance through increased drinking. Desmopressin (DDAVP, a synthetic AVP analogue) reduces uncontrolled water excretion in CDI and is commonly used in treatment. Critically, loss of thirst perception (through primary pathology or reduced consciousness) or limited access to water (through non-availability, disability or inter-current illness) in a patient with CDI can lead to life-threatening dehydration. This position can be further exacerbated through the omission of DDAVP. Recent data have highlighted serious adverse events (including deaths) in patients with CDI. These adverse outcomes and deaths have occurred through a combination of lack of knowledge and treatment failures by health professionals. Here, with our guideline, we recommend treatment pathways for patients with known CDI admitted to hospital. Following these guidelines is essential for the safe management of patients with CDI.

Implications of the United Kingdom prospective diabetes study.

The United Kingdom Prospective Diabetes Study (UKPDS) is the largest ever study to clarify the role of glycemia and blood pressure control in improving morbidity and mortality in patients with type 2 diabetes. The study examined the impact of glycemia and blood pressure control on microvascular and macrovascular risk, and if any particular therapy was advantageous for these patients. This articles discusses the background, methodology and results of the UKPDS and develops recommendations for treatment and management of patients with type 2 diabetes in the clinical setting.

A spectrum of behaviour in silent corticotroph pituitary adenomas.

Silent corticotroph adenomas (SCA) are pituitary tumours positive on immunohistochemical staining for ACTH but without clinical evidence of Cushing's disease in the patient. Previous reports suggest that these tumours may behave in a more aggressive way then other pituitary adenomas. We have followed the natural history of SCA and assessed whether histopathological indices predict tumour behaviour. We identified 22 patients in whom trans-sphenoidal surgery was performed for a non-functioning adenoma (NFA) with positive immunostaining for ACTH between 1990 and 2000 and examined the history of their disease. Patients were followed up for a mean of 4.8 years. A total of 86.7% of patients had documented visual deficits at presentation. In four cases hypercortisolaemia was observed later in the course of the disease. Two patients died as a result of their SCA and 33.3% of tumours recurred. Recurrence was more frequent in patients treated with adjuvant radiotherapy. Pathological indices (increased mitotic range and Ki-67) did not predict recurrence or malignant transformation. We suggest that certain 'silent' corticotroph tumours may have the potential for ACTH secretion leading to hypercortisolaemia at a later stage in the disease. The possibility of transformation to a more aggressive tumour needs to be considered in all SCA.

Insulin resistance, lipid and fatty acid concentrations in 867 healthy Europeans. European Group for the Study of Insulin Resistance (EGIR).

BACKGROUND: Insulin resistance, dyslipidaemia and abnormal nonesterified fatty acid (NEFA) metabolism are features of the 'metabolic syndrome', but the mechanisms of these relationships are uncertain. MATERIALS AND METHODS: We studied associations between insulin resistance and lipoprotein concentrations by retrospective analysis of euglycaemic hyperinsulinaemic clamp data from 867 normoglycaemic subjects in 21 European centres. Data on NEFA concentrations were available in a subgroup of 541 subjects from 9 clinical centres. These subjects' characteristics do not vary significantly from those of the whole cohort. RESULTS: After adjustment for the effects of age, sex, obesity and intercentre variability, regression analysis showed relationships between triglycerides and markers of insulin sensitivity. There were significant correlations between triglycerides and fasting plasma glucose (P < 0.0001), fasting plasma insulin (P < 0.0001) and mean glucose infusion rate at steady state (M-value, P < 0.0001). Indices of insulin resistance were related to NEFA concentrations. Fasting NEFA were negatively correlated with the M-value (P < 0.0001). Non-esterified fatty acids at steady state were positively correlated with fasting markers of insulin resistance: fasting plasma glucose (P < 0.05), fasting plasma insulin (P < 0.005) and negatively correlated with the M-value (P < 0.0005). There were relationships between fasting concentrations of plasma lipids and of NEFAs. Non-esterified fatty acids at steady state correlated with fasting triglycerides (P < 0.0001), but not with any of the other plasma lipoprotein concentrations. The associations of fasting triglycerides with the M-value and with NEFAs at steady state were independent of each other. All these associations were independent of obesity and geographical location CONCLUSION: The results in this large cohort of healthy European subjects suggest that triglyceride concentrations depend upon both insulin's gluco-regulation (estimated by glucose uptake) and antilipolytic insulin action (measured by NEFA levels) during an euglycaemic clamp.