Introduction: Adrenal insufficiency (AI) is a life-threatening condition that frequently presents with non-specific symptoms, complicating early diagnosis. Morning serum cortisol serves as an effective initial screening tool for assessing the hypothalamic-pituitary-adrenal (HPA) axis, potentially reducing the necessity for dynamic investigations such as the Short Synacthen Test (SST). This study aims to assess the clinical indications for morning cortisol testing and evaluate institutional compliance with guideline-based cortisol cut-offs, the initiation of hydrocortisone therapy, and the provision of “sick day” education. Methodology: A retrospective cross-sectional study was conducted among inpatients at a tertiary hospital in Central Pahang who underwent morning cortisol evaluation between March and August 2025. Data regarding demographics, indications, biochemical results, and clinical management were analyzed using SPSS version 26.0. Results: Among the 111 patients included (mean age: 54.96 ± 15.88 years), the primary indications for testing were hyponatremia (54.1%), hypotension (27.9%), and a history of traditional medication use (26.1%). Based on Endocrine Society guidelines: <150 nmol/L (AI likely): 16.2%, 150–300 nmol/L (Borderline): 26.1%, and >300 nmol/L (AI unlikely): 57.7%. Notably, only 10 of the 18 patients with cortisol levels <150 nmol/L were formally diagnosed and initiated on appropriate therapy. In the borderline group, only two patients underwent an SST. Although 12 patients overall were treated for AI, only eight received documented “sick day” education. Low cortisol levels were significantly associated with hyponatremia, hypotension, and traditional medication use, and demonstrated a negative correlation with serum sodium and albumin levels. Conclusion Morning serum cortisol is a valuable screening tool for AI, particularly in patients presenting with hyponatremia or hypotension. However, significant gaps persist in follow-up management, confirmatory testing, and patient education. Implementing standardized protocols and enhancing clinician education are essential to optimize care.
Hydrocortisone ; Adrenal Insufficiency ; Tertiary Care Centers

Introduction: Diabetic nephropathy is a major complication of type 1 diabetes mellitus (T1DM) and a leading cause of endstage renal disease. This study aims to correlate long-term glycemic control and disease duration with renal decline, measured via estimated glomerular filtration rate (eGFR) and urinalysis proteinuria. Methodology: A retrospective cohort study was conducted among 15 patients with T1DM at Hospital Bentong. Independent variables were 3-year mean glycated hemoglobin A1c (HbA1c) and duration of diabetes. Dependent variables included the change in eGFR derived from baseline and latest serum creatinine (calculated using the CKD-EPI 2021 equation), and the shift in urine protein severity, mapped to an ordinal scale, over a 3-year follow-up. Multivariate linear regression and Spearman’s rank correlation were utilized for analysis. Results: The cohort consisted of 5 male patients and 10 female patients with a mean initial presentation HbA1c of 11.4%, a mean 3-year HbA1c of 9.15%, and a median disease duration of 11 years. Baseline eGFR was preserved (mean 117.15 mL/ min/1.73 m²). Regression analysis revealed that for every 1% increase in mean HbA1c, eGFR declined significantly by 9.6 mL/min/1.73 m² (p = 0.002). Furthermore, each additional year of disease duration independently reduced eGFR by 1.1 units (p = 0.008). Spearman validation confirmed the strong negative correlation between HbA1c and eGFR (Rho = -0.664, p = 0.006). Conversely, the correlation between 3-year mean HbA1c and worsening proteinuria was not statistically significant (Rho = -0.067, p = 0.827), and patient gender did not significantly impact progression (p = 0.129). Conclusion Poor long-term glycemic control and longer duration of diabetes mellitus are strong independent predictors of decline in functional glomerular filtration in T1DM. While worsening proteinuria may not be detected during a short 3-year window using a standard urinalysis (UFEME), a more sensitive test such as Urine Albumin-to-Creatinine Ratio may better detect early structural kidney damage.
Diabetes Mellitus, Type 1 ; Proteinuria

Introduction: Artificial intelligence (AI) and digital health technologies are increasingly integrated into diabetes care. However, data on awareness and adoption among endocrinologists in Malaysia remain limited. Methodology: A cross-sectional survey was conducted among endocrinologists in Malaysia using a structured online questionnaire. Items assessed awareness of AI-based diabetes technologies, attitudes toward AI in clinical care, perceived barriers, and willingness to adopt AI tools. Multiple responses were permitted for selected items. Results: The cohort (N = 65) comprised 60% junior endocrinologists and 40% senior endocrinologists (72% governmentbased). Recognition was higher for monitoring and screening technologies, particularly continuous glucose monitoring (CGM) data analytics (66.2%) and automated retinopathy screening (33.8%) versus therapeutic AI tools, such as lifestyle coaching applications (43.1%) and insulin titration algorithms (30.8%). While approximately 60% endocrinologists reported familiarity with AI in diabetes care, 15.4% of them had no prior recognition of AI tools. Attitudes toward AI were largely positive; 76% agreed that AI improves diabetes outcomes, 60% believed it reduces clinician workload, and 90% agreed that AI would complement diabetes care systems. However, approximately 50% remained neutral regarding its reliability and expressed concerns about ethics and data security. In practice, only 40% endocrinologists reported using AI tools, most commonly CGM analytics with automated insights (46.4%), with generally low frequency of use. The main barriers to adoption were lack of formal training (78.5%) and high implementation costs (60%). Despite this, 90% expressed willingness to adopt AI in future practice. Conclusion Awareness of AI technologies in diabetes care among Malaysian endocrinologists is variable, with stronger recognition of tools already embedded in routine practice. While attitudes toward AI are favorable, significant educational and structural barriers must be addressed to facilitate broader implementation.
Artificial Intelligence ; Endocrinologists ; Malaysia ; Diabetes Mellitus

Presentation of Cushing’s syndrome during pregnancy is extremely rare. We report a 21-year-old female with Cushing’s syndrome diagnosed at 23 weeks of gestation and had recurrent acute pulmonary oedema during the antepartum and postpartum period. She delivered prematurely via emergency caesarean section at 28 weeks of gestation. This case highlights the rare occurrence of recurrent acute pulmonary oedema during pregnancy and consequential premature birth in a patient with adrenal Cushing’s. She was diagnosed with adrenal Cushing’s during the postpartum period based on unsuppressed serum cortisol after overnight and low-dose dexamethasone suppression test with a suppressed ACTH. CT scan of the adrenal glands revealed a right adrenal cortical adenoma. The risk of complications in infants and mothers who suffer from Cushing’s syndrome needs to be handled carefully. The diagnosis of Cushing’s syndrome in pregnant women often overlaps and is difficult to establish in early pregnancy.
Pregnancy

INTRODUCTIONThe use of non-fluoroscopic systems (NFS) to guide radiofrequency catheter ablation (RFCA) for the treatment of supraventricular tachycardia (SVT) is associated with lower radiation exposure. This study aimed to determine if NFS reduces fluoroscopy time, radiation dose and procedure time.

METHODSWe prospectively enrolled patients undergoing RFCA for SVT. NFS included EnSiteTM NavXTM or CARTO® mapping. We compared procedure and fluoroscopy times, and radiation exposure between NFS and conventional fluoroscopy (CF) cohorts. Procedural success, complications and one-year success rates were reported.

RESULTSA total of 200 patients over 27 months were included and RFCA was guided by NFS for 79 patients; those with atrioventricular nodal reentrant tachycardia (AVNRT), left-sided atrioventricular reentrant tachycardia (AVRT) and right-sided AVRT were included (n = 101, 63 and 36, respectively). Fluoroscopy times were significantly lower with NFS than with CF (10.8 ± 11.1 minutes vs. 32.0 ± 27.5 minutes; p < 0.001). The mean fluoroscopic dose area product was also significantly reduced with NFS (NSF: 5,382 ± 5,768 mGy*cm2 vs. CF: 21,070 ± 23,311 mGy*cm2; p < 0.001); for all SVT subtypes. There was no significant reduction in procedure time, except for left-sided AVRT ablation (NFS: 79.2 minutes vs. CF: 116.4 minutes; p = 0.001). Procedural success rates were comparable (NFS: 97.5% vs. CF: 98.3%) and at one-year follow-up, there was no significant difference in the recurrence rates (NFS: 5.2% vs. CF: 4.2%). No clinically significant complications were observed in both groups.

CONCLUSIONThe use of NFS for RFCA for SVT is safe, with significantly reduced radiation dose and fluoroscopy time.

Adolescent ; Adult ; Aged ; Aged, 80 and over ; Catheter Ablation ; methods ; Child ; Female ; Fluoroscopy ; Humans ; Male ; Middle Aged ; Prospective Studies ; Radiation Dosage ; Radiation, Ionizing ; Tachycardia, Atrioventricular Nodal Reentry ; therapy ; Tachycardia, Supraventricular ; therapy ; Treatment Outcome ; Young Adult