Artificial intelligence in histopathological image analysis of central nervous system tumours: A systematic review.

The convergence of digital pathology and artificial intelligence could assist histopathology image analysis by providing tools for rapid, automated morphological analysis. This systematic review explores the use of artificial intelligence for histopathological image analysis of digitised central nervous system (CNS) tumour slides. Comprehensive searches were conducted across EMBASE, Medline and the Cochrane Library up to June 2023 using relevant keywords. Sixty-eight suitable studies were identified and qualitatively analysed. The risk of bias was evaluated using the Prediction model Risk of Bias Assessment Tool (PROBAST) criteria. All the studies were retrospective and preclinical. Gliomas were the most frequently analysed tumour type. The majority of studies used convolutional neural networks or support vector machines, and the most common goal of the model was for tumour classification and/or grading from haematoxylin and eosin-stained slides. The majority of studies were conducted when legacy World Health Organisation (WHO) classifications were in place, which at the time relied predominantly on histological (morphological) features but have since been superseded by molecular advances. Overall, there was a high risk of bias in all studies analysed. Persistent issues included inadequate transparency in reporting the number of patients and/or images within the model development and testing cohorts, absence of external validation, and insufficient recognition of batch effects in multi-institutional datasets. Based on these findings, we outline practical recommendations for future work including a framework for clinical implementation, in particular, better informing the artificial intelligence community of the needs of the neuropathologist.

Use of liothyronine (T3) in hypothyroidism: Joint British Thyroid Association/Society for endocrinology consensus statement.

Persistent symptoms in patients treated for hypothyroidism are common. Despite more than 20 years of debate, the use of liothyronine for this indication remains controversial, as numerous randomised trials have failed to show a benefit of treatment regimens that combine liothyronine (T3) with levothyroxine over levothyroxine monotherapy. This consensus statement attempts to provide practical guidance to clinicians faced with patients who have persistent symptoms during thyroid hormone replacement therapy. It applies to non-pregnant adults and is focussed on care delivered within the UK National Health Service, although it may be relevant in other healthcare environments. The statement emphasises several key clinical practice points for patients dissatisfied with treatment for hypothyroidism. Firstly, it is important to establish a diagnosis of overt hypothyroidism; patients with persistent symptoms during thyroid hormone replacement but with no clear biochemical evidence of overt hypothyroidism should first have a trial without thyroid hormone replacement. In those with established overt hypothyroidism, levothyroxine doses should be optimised aiming for a TSH in the 0.3-2.0 mU/L range for 3 to 6 months before a therapeutic response can be assessed. In some patients, it may be acceptable to have serum TSH below reference range (e.g. 0.1-0.3 mU/L), but not fully suppressed in the long term. We suggest that for some patients with confirmed overt hypothyroidism and persistent symptoms who have had adequate treatment with levothyroxine and in whom other comorbidities have been excluded, a trial of liothyronine/levothyroxine combined therapy may be warranted. The decision to start treatment with liothyronine should be a shared decision between patient and clinician. However, individual clinicians should not feel obliged to start liothyronine or to continue liothyronine medication provided by other health care practitioners or accessed without medical advice, if they judge this not to be in the patient's best interest.

COVID-19-related adrenal haemorrhage: Multicentre UK experience and systematic review of the literature.

OBJECTIVE: Adrenal haemorrhage (AH) is an uncommon, usually incidental imaging finding in acutely unwell patients. AH has been reported during coronavirus disease 2019 (COVID-19) infection and following ChAdOx1 nCoV-19 (Oxford-AstraZeneca) vaccination. The Society for Endocrinology (SfE) established a task force to describe the UK experience of COVID-19-related AH. DESIGN: A systematic literature review was undertaken. A survey was conducted through the SfE clinical membership to identify patients with COVID-19-related AH using a standardized data collection tool. RESULTS: The literature search yielded 25 cases of COVID-19-related AH (19 bilateral; 13 infection-related, and 12 vaccine-related). Eight UK centres responded to the survey with at least one case. A total of 18 cases were included in the descriptive study, including 11 from the survey and 7 UK-based patients from the systematic review. Seven patients (4 males; median age 53 (range 26-70) years), had infection-related AH (four bilateral). Median time from positive COVID-19 test to AH detection was 8 (range 1-30) days. Eleven cases of vaccine-related AH (eight bilateral) were captured (3 males; median age 47 (range 23-78) years). Median time between vaccination (nine Oxford-AstraZeneca and two Pfizer-BioNTech) and AH was 9 (range 2-27) days; 9/11 AH occurred after the first vaccine dose. Acute abdominal pain was the commonest presentation (72%) in AH of any cause. All 12 patients with bilateral AH and one patient with unilateral AH required glucocorticoid replacement. CONCLUSION: Adrenal haemorrhage with consequential adrenal insufficiency can be a complication of COVID-19 infection and vaccination. Adrenal function assessment is mandatory to avoid the potentially fatal consequences of unrecognized adrenal insufficiency.

A patient-reported outcome measure for patients with pituitary adenoma undergoing transsphenoidal surgery.

PURPOSE: Pituitary adenomas affect patients' quality-of-life (QoL) across several domains, with long-term implications even following gross-total resection or disease remission. While clinical outcomes can assess treatment efficacy, they do not capture variations in QoL. We present the development and validation of a patient reported outcome measure (PROM) for patients with pituitary adenomas undergoing transsphenoidal surgery. METHODS: The COSMIN checklist informed the development of the pituitary outcome score (POS). Consecutive patients undergoing surgical treatment for suspected pituitary adenoma at a single centre were included prospectively. An expert focus group and patient interviews informed item generation. Item reduction was conducted through exploratory factor analysis and expert consensus, followed by assessment of the tool's validity, reliability, responsiveness, and interpretability. RESULTS: 96 patients with a median age of 50 years validated the POS. The final questionnaire included 25 questions with four subscales: EQ-5D-5L-QoL, Visual Symptoms, Endocrine Symptoms and Nasal Symptoms. CONCLUSION: The POS is the first validated PROM for patients undergoing transsphenoidal surgery for a pituitary adenoma. This PROM could be integrated into contemporary practice to provide patient-centred outcomes assessment for this patient group, aligning more closely with patient objectives.

The management and outcome of hyponatraemia following transsphenoidal surgery: a retrospective observational study.

PURPOSE: Hyponatraemia is a common complication following transsphenoidal surgery. However, there is sparse data on its optimal management and impact on clinical outcomes. The aim of this study was to evaluate the management and outcome of hyponatraemia following transsphenoidal surgery. METHODS: A prospectively maintained database was searched over a 4-year period between January 2016 and December 2019, to identify all patients undergoing transsphenoidal surgery. A retrospective case-note review was performed to extract data on hyponatraemia management and outcome. RESULTS: Hyponatraemia occurred in 162 patients (162/670; 24.2%) with a median age of 56 years. Female gender and younger age were associated with hyponatraemia, with mean nadir sodium being 128.6 mmol/L on postoperative day 7. Hyponatraemic patients had longer hospital stay than normonatraemic group with nadir sodium being inversely associated with length of stay (p < 0.001). In patients with serum sodium ≤ 132 mmol/L, syndrome of inappropriate antidiuretic hormone secretion (SIADH) was the commonest cause (80/111; 72%). Among 76 patients treated with fluid restriction as a monotherapy, 25 patients (25/76; 32.9%) did not achieve a rise in sodium after 3 days of treatment. Readmission with hyponatraemia occurred in 11 cases (11/162; 6.8%) at a median interval of 9 days after operation. CONCLUSION: Hyponatraemia is a relatively common occurrence following transsphenoidal surgery, is associated with longer hospital stay and risk of readmission and the effectiveness of fluid restriction is limited. These findings highlight the need for further studies to better identify and treat high-risk patients, including the use of arginine vasopressin receptor antagonists.

Temporal trends in craniopharyngioma management and long-term endocrine outcomes: A multicentre cross-sectional study.

BACKGROUND: The optimal management of craniopharyngiomas remains controversial. OBJECTIVES: To examine temporal trends in the management of craniopharyngioma with a focus on endocrine outcomes. METHODS: This was a cross-sectional, multicentre study. Patients treated between 1951 and 2015 were identified and divided into four quartiles. Demographics, presentation, treatment and outcomes were collected. RESULTS: In total, 142 patients with childhood-onset craniopharyngioma (48/142; 34%) and adult-onset disease (94/142; 66%) were included. The median follow-up was 15 years (IQR 5-23 years). Across quartiles, there was a significant trend towards using transsphenoidal surgery (P < .0001). The overall use of radiotherapy was not different among the four quartiles (P = .33). At the latest clinical review, the incidence of GH, ACTH, gonadotrophin deficiencies and anterior panhypopituitarism fell significantly across the duration of the study. Anterior panhypopituitarism was not affected by treatment modality (surgery vs surgery and radiotherapy) (P = .23). There was no difference in the incidence of high BMI (≥25 kg/m2 ) among the four quartiles (P = .14). BMI was higher in patients who treated with surgery and radiotherapy than those treated with surgery only (P = .006). Tumour regrowth occurred in 51 patients (51/142; 36%) with no difference in regrowth among quartiles over the time course of the study (P = .15). CONCLUSION: We demonstrate a significant reduction in panhypopituitarism in craniopharyngioma patients over time, most likely because of a trend towards more transsphenoidal surgery. However, long-term endocrine sequelae remain common and lifelong follow-up is required.

The direct and indirect impact of the COVID-19 pandemic on the care of patients with pituitary disease: a cross sectional study.

PURPOSE: The coronavirus disease 2019 (COVID-19) pandemic is widely believed to have had a major impact on the care of patients with pituitary disease. The virus itself may directly result in death, and patients with adrenal insufficiency, often a part of hypopituitarism, are thought to represent a particularly susceptible subgroup. Moreover, even in patients that do not contract the virus, the diversion of resources by healthcare institutions to manage the virus may indirectly result in delays in their management. To this end, the aim of this study was to determine the direct and indirect impact of the COVID-19 pandemic on patients with pituitary disease. METHODS: A cross-sectional study design was adopted, with all adult patients seen by our pituitary service in the year prior to the nationwide lockdown on March 23rd 2020 invited to participate in a telephone survey. RESULTS: In all, 412 patients (412/586; 70.3%) participated in the survey. 66 patients (66/412; 16.0%) reported having suspected COVID-19 infection. Of the 10 patients in this group tested for COVID-19 infection, three received a positive test result. No deaths due to COVID-19 were identified. 267 patients (267/412; 64.8%) experienced a delay or change in the planned care for their pituitary disease, with 100 patients (100/412; 24.3%) perceiving an impact to their care. CONCLUSIONS: Whilst only a small percentage of patients had confirmed or suspected COVID-19 infection, over half were still indirectly impacted by the pandemic through a delay or change to their planned care.

OFF-LABEL USE AND MISUSE OF TESTOSTERONE, GROWTH HORMONE, THYROID HORMONE, AND ADRENAL SUPPLEMENTS: RISKS AND COSTS OF A GROWING PROBLEM.

Over the past few decades, there has been an unprecedented rise in off-label use and misuse of testosterone, growth hormone, thyroid hormone, and adrenal supplements. Testosterone therapy is often promoted to men for the treatment of low energy, lower libido, erectile dysfunction, and other symptoms. Growth hormone is used in attempts to improve athletic performance in athletes and to attenuate aging in older adults. Thyroid hormone and/or thyroid supplements or boosters are taken to treat fatigue, obesity, depression, cognitive impairment, impaired physical performance, and infertility. Adrenal supplements are used to treat common nonspecific symptoms due to "adrenal fatigue," an entity that has not been recognized as a legitimate medical diagnosis. Several factors have contributed to the surge in off-label use and misuse of these hormones and supplements: direct-to-consumer advertising, websites claiming to provide legitimate medical information, and for-profit facilities promoting therapies for men's health and anti-aging. The off-label use and misuse of hormones and supplements in individuals without an established endocrine diagnosis carries known and unknown risks. For example, the risks of growth hormone abuse in athletes and older adults are unknown due to a paucity of studies and because those who abuse this hormone often take supraphysiologic doses in sporadic intervals. In addition to the health risks, off-label use of these hormones and supplements generates billions of dollars of unnecessary costs to patients and to the overall health-care system. It is important that patients honestly disclose to their providers off-label hormone use, as it may affect their health and treatment plan. General medical practitioners and adult endocrinologists should be able to begin a discussion with their patients regarding the unfavorable balance between the risks and benefits associated with off-label use of testosterone, growth hormone, thyroid hormone, and adrenal supplements. Abbreviations: DHEA = dehydroepiandrosterone; FDA = U.S. Food and Drug Administration; GH = growth hormone; IGF-1 = insulin-like growth factor 1; LT3 = L-triiodothyronine; LT4 = levothyroxine; T3 = total triiodothyronine; T4 = thyroxine; TSH = thyroid-stimulating hormone.

Rathke's cleft cysts following transsphenoidal surgery: long-term outcomes and development of an optimal follow-up strategy.

BACKGROUND: In patients with symptomatic Rathke's cleft cyst, transsphenoidal surgery is highly effective at preventing further visual loss and usually allows for some recovery of vision. However, cyst recurrence and the need for re-operation are well recognized. To this end, the aim of this study was to investigate patterns of recurrence and long-term outcomes and to use this information to develop an optimal follow-up strategy. METHOD: A prospectively maintained database was searched over a 10-year period between 1 January 2008 and the 1 January 2018 to identify all adults that underwent transsphenoidal surgery with a new diagnosis of Rathke's cleft cyst. A retrospective case note review was performed for each patient to extract data on their presentation, investigation, treatment, and outcome. RESULTS: In all, 61 eligible patients were identified. The median follow-up was 34 months (range 2-112 months). In the 22 patients with pre-operative visual loss, the outcomes at 6 months were as follows: normal vision (2/22; 9.1%), improved but not normal (7/22; 31.8%), stable (12/22; 54.5%), worse but not blind (1/22; 4.5%), and blind (0/22; 0%). The overall rate of regrowth and re-operation in our study was 19.7 and 11.5%, respectively. The only factor that was significantly associated with recurrence was the presence of residual cystic disease on the post-operative MRI (p < 0.001). CONCLUSIONS: We propose a follow-up strategy that stratifies patients at "low risk" if there is no residual cyst, with increasing interval scans, or "high risk" if there is residual cyst, with annual visual assessment and scans.

Increased Population Risk of AIP-Related Acromegaly and Gigantism in Ireland.

The aryl hydrocarbon receptor interacting protein (AIP) founder mutation R304* (or p.R304* ; NM_003977.3:c.910C>T, p.Arg304Ter) identified in Northern Ireland (NI) predisposes to acromegaly/gigantism; its population health impact remains unexplored. We measured R304* carrier frequency in 936 Mid Ulster, 1,000 Greater Belfast (both in NI) and 2,094 Republic of Ireland (ROI) volunteers and in 116 NI or ROI acromegaly/gigantism patients. Carrier frequencies were 0.0064 in Mid Ulster (95%CI = 0.0027-0.013; P = 0.0005 vs. ROI), 0.001 in Greater Belfast (0.00011-0.0047) and zero in ROI (0-0.0014). R304* prevalence was elevated in acromegaly/gigantism patients in NI (11/87, 12.6%, P < 0.05), but not in ROI (2/29, 6.8%) versus non-Irish patients (0-2.41%). Haploblock conservation supported a common ancestor for all the 18 identified Irish pedigrees (81 carriers, 30 affected). Time to most recent common ancestor (tMRCA) was 2550 (1,275-5,000) years. tMRCA-based simulations predicted 432 (90-5,175) current carriers, including 86 affected (18-1,035) for 20% penetrance. In conclusion, R304* is frequent in Mid Ulster, resulting in numerous acromegaly/gigantism cases. tMRCA is consistent with historical/folklore accounts of Irish giants. Forward simulations predict many undetected carriers; geographically targeted population screening improves asymptomatic carrier identification, complementing clinical testing of patients/relatives. We generated disease awareness locally, necessary for early diagnosis and improved outcomes of AIP-related disease.

The screening and management of pituitary dysfunction following traumatic brain injury in adults: British Neurotrauma Group guidance.

Pituitary dysfunction is a recognised, but potentially underdiagnosed complication of traumatic brain injury (TBI). Post-traumatic hypopituitarism (PTHP) can have major consequences for patients physically, psychologically, emotionally and socially, leading to reduced quality of life, depression and poor rehabilitation outcome. However, studies on the incidence of PTHP have yielded highly variable findings. The risk factors and pathophysiology of this condition are also not yet fully understood. There is currently no national consensus for the screening and detection of PTHP in patients with TBI, with practice likely varying significantly between centres. In view of this, a guidance development group consisting of expert clinicians involved in the care of patients with TBI, including neurosurgeons, neurologists, neurointensivists and endocrinologists, was convened to formulate national guidance with the aim of facilitating consistency and uniformity in the care of patients with TBI, and ensuring timely detection or exclusion of PTHP where appropriate. This article summarises the current literature on PTHP, and sets out guidance for the screening and management of pituitary dysfunction in adult patients with TBI. It is hoped that future research will lead to more definitive recommendations in the form of guidelines.

A Real-World Study of the Effectiveness and Safety of Semaglutide for Weight Loss.

Introduction Recent randomized controlled trials (RCTs) have shown the great efficacy of semaglutide in achieving significant weight loss in overweight and obese adults. However, real-world data about its effectiveness are still limited. This study evaluated the effectiveness and adverse events of semaglutide for weight management in a real-life setting, excluding patients with diabetes mellitus (DM). Methods This is a retrospective chart review of 40 overweight or obese individuals with a median age of 47 years, weight of 111.7 kg, and body mass index (BMI) of 39.7 kg/m2 who were prescribed semaglutide for weight management. Results After three months of semaglutide administration, the median weight reduction was 7.4 kg (6.6% of the baseline weight), with 28 (70%) and eight patients (20%) achieving greater than 5% (5.6 kg) and 10% (11.2 kg) weight loss, respectively. Among 25 patients with six-month data, 22 (88%), 17 (68%), and eight (32%) patients exceeded 5% (5.6 kg), 10% (11.2 kg), and 15% (16.8 kg) weight loss, respectively. The maintenance semaglutide dose was 1 mg in 16 cases and 2 mg in nine cases, leading to a similar weight loss of 13.6% (14.9 kg) and 12.8% (14 kg), respectively. Relatively low response rates were observed in males, with seven responders out of 12 (58.4%) compared to 24 out of 28 (85.8%) in females (P value = 0.057), and in five out of nine (55.6%) among those with a history of psychiatric disease. The rate of adverse events was 26 out of 40 patients (65%), mostly mild to moderate and of short duration, leading to discontinuation in only a single case (2.5%). Conclusion This retrospective study demonstrated the significant effectiveness of semaglutide for weight loss, even at lower than approved maintenance doses, combined with a good safety profile. Therefore, semaglutide may dramatically change the landscape of obesity treatment.

Machine Learning Analysis of Post-Operative Tumour Progression in Non-Functioning Pituitary Neuroendocrine Tumours: A Pilot Study.

Post-operative tumour progression in patients with non-functioning pituitary neuroendocrine tumours is variable. The aim of this study was to use machine learning (ML) models to improve the prediction of post-operative outcomes in patients with NF PitNET. We studied data from 383 patients who underwent surgery with or without radiotherapy, with a follow-up period between 6 months and 15 years. ML models, including k-nearest neighbour (KNN), support vector machine (SVM), and decision tree, showed superior performance in predicting tumour progression when compared with parametric statistical modelling using logistic regression, with SVM achieving the highest performance. The strongest predictor of tumour progression was the extent of surgical resection, with patient age, tumour volume, and the use of radiotherapy also showing influence. No features showed an association with tumour recurrence following a complete resection. In conclusion, this study demonstrates the potential of ML models in predicting post-operative outcomes for patients with NF PitNET. Future work should look to include additional, more granular, multicentre data, including incorporating imaging and operative video data.

Video-based performance analysis in pituitary surgery - Part 1: Surgical outcomes.

BACKGROUND: Endoscopic pituitary adenoma surgery has a steep learning curve, with varying surgical techniques and outcomes across centers. In other surgeries, superior performance is linked with superior surgical outcomes. This study aimed to explore the prediction of patient-specific outcomes using surgical video analysis in pituitary surgery. METHODS: Endoscopic pituitary adenoma surgery videos from a single center were annotated by experts for surgical workflow (3 phases, 15 steps) and surgical skill (using modified Objective Structured Assessment of Technical Skills; mOSATS). Quantitative workflow metrics were calculated, including phase duration and step transitions. Poisson or logistic regression was used to assess the association of workflow metrics and mOSATS with common inpatient surgical outcomes. RESULTS: 100 videos from 100 patients were included. Nasal phase mean duration was 24mins and mean mOSATS was 21.2/30. Mean duration was 34mins and mean mOSATS was 20.9/30 for the sellar phase, and 11mins and 21.7/30 respectively for the closure phase. The most common adverse outcomes were new anterior pituitary hormone deficiency (n=26), dysnatremia (n=24) and cerebrospinal fluid (CSF) leak (n=5). Higher mOSATS for all three phases and shorter operation duration was associated with decreased length of stay (p=0.003 & p<0.001). Superior closure phase mOSATS were associated with reduced post-operative CSF leak (p<0.001), and superior sellar phase mOSATS were associated with reduced post-operative visual deterioration (p=0.041). CONCLUSION: Superior surgical skill and shorter surgical time were associated with superior surgical outcomes, at a generic and phase-specific level. Such video-based analysis has promise for integration into data-driven training and service improvement initiatives.

Video-Based Performance Analysis in Pituitary Surgery - Part 2: Artificial Intelligence Assisted Surgical Coaching.

BACKGROUND: Superior surgical skill improves surgical outcomes in endoscopic pituitary adenoma surgery. Video-based coaching programs, pioneered in professional sports, have shown promise in surgical training. In this study, we developed and assessed a video-based coaching program, using artificial intelligence (AI). METHODS: AI-assisted video-based surgical coaching was implemented over 6 months with the pituitary surgery team. The program consisted of 1) monthly random video analysis and review; and 2) quarterly 2-hour educational meetings discussing these videos and learning points. Each video was annotated for surgical phases and steps using AI, which improved video interactivity and allowed calculation of quantitative metrics. Primary outcomes were program feasibility, acceptability and appropriateness. Surgical performance (via modified Objective Structured Assessment of Technical Skills; mOSATS) and early surgical outcomes were recorded for every case during the 6-month coaching period, and a preceding 6-month control period. Beta and logistic regression were used to assess the change in mOSATS and surgical outcomes after the coaching program implementation. RESULTS: All participants highly rated the program's feasibility, acceptability and appropriateness. During the coaching program, 63 endoscopic pituitary adenoma cases were included, with 41 in the control group. Surgical performance across all operative phases improved during the coaching period (p<0.001), with a reduction in new post-operative anterior pituitary hormone deficit (p=0.01). CONCLUSION: We have developed a novel AI-assisted video surgical coaching program for endoscopic pituitary adenoma surgery - demonstrating its viability and impact on surgical performance. Early results also suggest improvement in patient outcomes. Future studies should be multicenter and longer term.

Approach to the Patient With Adrenal Hemorrhage.

Adrenal hemorrhage is an uncommon, underrecognized condition that can be encountered in several clinical contexts. Diagnosing adrenal hemorrhage is challenging due to its nonspecific clinical features. Therefore, it remains a diagnosis that is made serendipitously on imaging of acutely unwell patients rather than with prospective clinical suspicion. Adrenal hemorrhage can follow abdominal trauma or appear on a background of predisposing conditions such as adrenal tumors, sepsis, or coagulopathy. Adrenal hemorrhage is also increasingly reported in patients with COVID-19 infection and in the context of vaccine-induced immune thrombocytopenia and thrombosis. Unexplained abdominal pain with hemodynamic instability in a patient with a predisposing condition should alert the physician to the possibility of adrenal hemorrhage. Bilateral adrenal hemorrhage can lead to adrenal insufficiency and potentially fatal adrenal crisis without timely recognition and treatment. In this article, we highlight the clinical circumstances that are associated with higher risk of adrenal hemorrhage, encouraging clinicians to prospectively consider the diagnosis, and we share a diagnostic and management strategy.

Post-Operative Medium- and Long-Term Endocrine Outcomes in Patients with Non-Functioning Pituitary Adenomas-Machine Learning Analysis.

Post-operative endocrine outcomes in patients with non-functioning pituitary adenoma (NFPA) are variable. The aim of this study was to use machine learning (ML) models to better predict medium- and long-term post-operative hypopituitarism in patients with NFPAs. We included data from 383 patients who underwent surgery with or without radiotherapy for NFPAs, with a follow-up period between 6 months and 15 years. ML models, including k-nearest neighbour (KNN), support vector machine (SVM), and decision tree models, showed a superior ability to predict panhypopituitarism compared with non-parametric statistical modelling (mean accuracy: 0.89; mean AUC-ROC: 0.79), with SVM achieving the highest performance (mean accuracy: 0.94; mean AUC-ROC: 0.88). Pre-operative endocrine function was the strongest feature for predicting panhypopituitarism within 1 year post-operatively, while endocrine outcomes at 1 year post-operatively supported strong predictions of panhypopituitarism at 5 and 10 years post-operatively. Other features found to contribute to panhypopituitarism prediction were age, volume of tumour, and the use of radiotherapy. In conclusion, our study demonstrates that ML models show potential in predicting post-operative panhypopituitarism in the medium and long term in patients with NFPM. Future work will include incorporating additional, more granular data, including imaging and operative video data, across multiple centres.