Frimley Health NHS Foundation Trust

We evaluated the effects of growth hormone (GH) pretreatment on oocyte retrieval and embryo quality in patients undergoing assisted reproductive technology (ART) for diminished ovarian reserve (DOR). In this prospective cohort study, conducted at Shahida Islam Medical Complex from October 2023 to June 2025, 2,000 women aged ≤40 years with DOR were recruited before starting ART. DOR diagnosis was based on the 2017 criteria and the 2011 Bologna criteria. Participants were nonrandomly allocated to four equal groups: G1 (1 month of GH before ovulation induction+standard ART), G2 (2 months of GH pretreatment+ART), G3 (GH supplementation during the ovulation induction phase+ART), and G4 (control; standard ART). Hormonal and metabolic profiling was performed before, during, and after GH treatment. Accordingly, secondary stratification was applied to enable personalized treatment. Based on early response to GH pretreatment, participants underwent dynamic regrouping into good and poor responders. Key outcomes included oocyte retrieval, embryo quality, and maternal-fetal outcomes. GH pretreatment improved ART results, with the G2 group achieving the best estradiol levels, follicular growth, and oocyte retrieval, albeit with increased insulin resistance (homeostatic model assessment of insulin resistance [HOMA-IR]). The G1 and G3 groups also outperformed controls (G4). GH dose and anti-Müllerian hormone (AMH) level strongly predicted successful oocyte retrieval. Testosterone was negatively associated with retrieval; interleukin 6 and dehydroepiandrosterone sulfate (DHEA-S) had weaker effects. G2 embryos exhibited the best development and blastocyst formation outcomes, with the most advanced development by day 5. A nomogram was developed to predict oocyte retrieval outcomes from GH dose, AMH, HOMA-IR, and testosterone. In patients with DOR, GH pretreatment improves treatment outcomes. Over 1 month of pretreatment can increase the oocytes retrieved but confers higher insulin resistance. Thus, when using GH for more than 4 weeks, clinicians should closely monitor HOMA-IR and perform continuous glucose monitoring; otherwise, 1-month GH pretreatment is preferred when initiating ART.

IntroductionThe sacral erector spinae plane (S-ESP) block is a recently described regional anesthesia technique that targets sacral dermatomes. First reported in 2019, it has been increasingly explored as a potential option for perioperative analgesia in surgeries involving the sacral, perineal, and pelvic regions. This review aims to summarize and critically appraise the current anatomical, technical, and clinical evidence on the S-ESP block, outlining its mechanisms of action, approaches, efficacy, and safety while identifying research gaps and future directions.MethodsWe performed a narrative review integrating anatomical descriptions, sonographic techniques, and available randomized controlled trials (RCTs) evaluating the clinical efficacy of the S-ESP block. A comprehensive literature search was conducted in PubMed, Embase, Scopus, Web of Science, and Ovid (Medline) from inception to July 21, 2025, without language or date restrictions. Full-text RCTs investigating ultrasound-guided S-ESP blocks for perioperative analgesia were reviewed and summarized.ResultsFourteen RCTs (published 2023–2025) were included, conducted predominantly in Turkey, India, and Egypt, spanning pediatric (6 months to 12 years) and adult populations undergoing circumcision, hypospadias repair, hemorrhoidectomy, pilonidal sinus surgery, transurethral resection of prostate, lumbar discectomy, and total hip arthroplasty. Most pediatric trials compared midline S-ESP with caudal or penile blocks, generally demonstrating a longer time to first rescue analgesia and reduced postoperative analgesic consumption, although one study reported a superior duration with caudal block. In adults, the S-ESP block consistently reduced pain scores and opioid requirements.ConclusionsCurrent evidence suggests that both midline and paramedian S-ESP techniques are technically feasible, appear safe in the short term, and provide clinically relevant postoperative analgesia in selected pediatric urogenital, adult anorectal, and orthopedic surgeries. However, heterogeneity in techniques, small single-center samples, and a narrow range of indications limit its generalizability. Large, multicenter RCTs with standardized protocols are needed to clarify the optimal approaches, dosing strategies, and comparative role of S-ESP block versus established neuraxial and peripheral techniques.

Macular phototoxicity is a rare form of retinal injury caused by intense light exposure, most commonly from direct sun-gazing or viewing a solar eclipse without adequate eye protection. We report a case of bilateral macular phototoxicity in a 35-year-old female photographer with no significant past medical history, who developed acute central vision loss after approximately one hour of inadvertent sun exposure. She presented 48 hours after the incident with markedly reduced central vision (only light perception centrally) in both eyes, while peripheral vision remained intact. Dilated fundus examination revealed bilateral circumscribed focal foveal hypopigmentation in the absence of other retinal or optic nerve findings. Optical coherence tomography (OCT) scans confirmed foveal phototoxic damage, showing a normal overall contour but with juxtafoveal microcystic changes and hyperreflective lesions at the fovea in both eyes, consistent with acute macular phototoxicity. Conservative management with lubricating eye drops and oral nutritional supplements was implemented. Over the ensuing weeks, her visual acuity improved from an initial profound central scotoma to 20/40 in both eyes at 10 days, and 20/40 (left eye) and 20/30 (right eye) by one month post-exposure. Follow-up OCT demonstrated resolution of macular swelling and reconstitution of retinal layers. By five months, her visual acuity had fully recovered to baseline (20/20); however, a small persistent central scotoma remained in the left eye. This case illustrates the clinical course of macular phototoxicity and highlights the importance of patient education on ocular sun safety. Early recognition is important, as most cases show significant spontaneous recovery, but preventative measures are paramount since treatment options are limited.

Nonmetastatic (M0) clear cell renal cell carcinoma (ccRCC) recurs in ∼20% of patients within 5yr after surgery. With no biomarkers available, recurrence detection relies on radiological imaging. Urine glycosaminoglycan profiles (GAGomes) were previously associated with M0 ccRCC recurrence. We conducted an observational prospective multicentre diagnostic test cohort study to evaluate GAGomes for postsurgery recurrence detection in M0 ccRCC. Postsurgical M0 ccRCC patients with a Leibovich score of ≥5 points were included. Follow-up imaging up to 18mo assessed radiological recurrence (reference standard). Urine GAGomes were measured every 3mo to compute a GAGome score (index test). Sensitivity and specificity to radiological recurrence were calculated. The lead time between the first positive GAGome score and radiological recurrence was estimated. Bayesian joint modelling estimated recurrence-free survival hazard ratio (HR). Of the 393 patients screened (January 2020 to November 2021), 134 met the inclusion criteria. The median follow-up was 16mo (interquartile range [IQR]: 12-18) for those without recurrence. At the last follow-up visit, 16% had recurred. The GAGome score had 90% sensitivity (95% confidence interval [CI]: 62-100%) and 51% specificity (95% CI: 30-71%) to radiological recurrence. The positive and negative predictive values were 26% (95%CI: 4-46%) and 97% (95% CI: 87-100%), respectively. The median lead time was 4.2mo (IQR: 1.6-6.4). A 10-point GAGome score increase was associated with an HR of 1.62 (95% high density interval: 1.11-2.30) for recurrence. The main limitation was short follow-up time. GAGome score had very high sensitivity to ccRCC recurrence, resulting in a negative predictive value of 97%. External validation foreseen in the study design aims to confirm its utility to personalise follow-up for M0 ccRCC patients.

BackgroundDiagnostic errors and administrative burdens, including medical coding, remain major challenges in health care. Large language models (LLMs) have the potential to alleviate these problems, but their adoption has been limited by concerns regarding reliability, transparency, and clinical safety.ObjectiveThis study introduces and evaluates 2 LLM-based frameworks, implemented within the Rhazes Clinician platform, designed to address these challenges: generation-assisted retrieval-augmented generation (GARAG) for automated evidence-based treatment planning and generation-assisted vector search (GAVS) for automated medical coding.MethodsGARAG was evaluated on 21 clinical test cases created by medically qualified authors. Each case was executed 3 times independently, and outputs were assessed using 4 criteria: correctness of references, absence of duplication, adherence to formatting, and clinical appropriateness of the generated management plan. GAVS was evaluated on 958 randomly selected admissions from the Medical Information Mart for Intensive Care (MIMIC)–IV database, in which billed International Classification of Diseases, Tenth Revision (ICD-10) codes served as the ground truth. Two approaches were compared: a direct GPT-4.1 baseline prompted to predict ICD-10 codes without constraints and GAVS, in which GPT-4.1 generated diagnostic entities that were each mapped onto the top 10 matching ICD-10 codes through vector search.ResultsAcross the 63 outputs, 62 (98.4%) satisfied all evaluation criteria, with the only exception being a minor ordering inconsistency in one repetition of case 14. For GAVS, the 958 admissions contained 8576 assigned ICD-10 subcategory codes (1610 unique). The vanilla LLM produced 131,329 candidate codes, whereas GAVS produced 136,920. At the subcategory level, the vanilla LLM achieved 17.95% average recall (15.86% weighted), while GAVS achieved 20.63% (18.62% weighted), a statistically significant improvement (P<.001). At the category level, performance converged (32.60% vs 32.58% average weighted recall; P=.99).ConclusionsGARAG demonstrated a workflow that grounds management plans in diagnosis-specific, peer-reviewed guideline evidence, preserving fine-grained clinical detail during retrieval. GAVS significantly improved fine-grained diagnostic coding recall compared with a direct LLM baseline. Together, these frameworks illustrate how LLM-based methods can enhance clinical decision support and medical coding. Both were subsequently integrated into Rhazes Clinician, a clinician-facing web application that orchestrates LLM agents to call specialized tools, providing a single interface for physician use. Further independent validation and large-scale studies are required to confirm generalizability and assess their impact on patient outcomes.

BackgroundMay-Thurner syndrome (MTS) is a rare cause of left lower extremity venous disorders, including deep vein thrombosis (DVT). Classic MTS is caused by compression of the left common iliac vein (LCIV) by the right common iliac artery (RCIA). We report a case of a left lower extremity DVT caused by a rare variant of the classic MTS wherein the LCIV is compressed by the left common iliac artery (LCIA) instead of the RCIA.Case DescriptionA 77-year-old female presented with left lower extremity swelling for two days. Computed tomography (CT) venogram demonstrated extensive ileo-femoral DVT from the LCIV to the peroneal vein, raising suspicion of MTS. She was found to have an unusual variant of MTS wherein the LCIV was compressed by the LCIA instead of the RCIA. She was started on full-dose anticoagulation with heparin and subsequently underwent mechanical thrombectomy followed by endoluminal stenting of the LCIV and external iliac vein (EIV) with restoration of venous flow post-stenting as confirmed by venogram. The patient was discharged home with oral anticoagulation to be continued for 6 months and instructions for follow-up duplex ultrasound to ensure stent patency.ConclusionsMTS should be considered as an underlying etiology of left lower extremity DVT, especially in females and the atypical variants of MTS should be considered while formulating a management plan.