Artificial Intelligence Evaluation of Fascial Plane Block Quality and Postoperative Pain in Cardiac Surgery

Postoperative pain is a common and significant problem following open heart surgery. Fascial plane blocks (FPBs) such as the serratus anterior plane block (SAPB) and the pecto-intercostal fascial block (PIFB) are increasingly used as part of multimodal analgesia in cardiac surgery. However, objective assessment of block quality and its relationship with clinical outcomes remains limited in the literature. This prospective observational study evaluated the anatomical success of ultrasound-guided fascial plane blocks applied in elective open heart surgery (median sternotomy). Block images were recorded in DICOM format and scored independently and blindly on a 3-point scale by five pre-trained, publicly available large vision-language models used as independent observers via zero-shot prompting (Gemini 3.5 Flash, Claude Sonnet 4.6, ChatGPT 5.5, Grok 4.3, Kimi AI 2.6), by three experienced anaesthesiologists, and by an adjudicator. The primary analysis was the correlation between anatomical block success scores and postoperative pain (Numerical Rating Scale at rest and on coughing). Agreement between AI and human assessments, total analgesic consumption and clinical outcomes were analysed as secondary aims. Note: this registration was originally entered from a superseded version of the protocol. The current approved protocol (version dated 29 April 2026) specifies pre-trained large vision-language models rather than a bespoke convolutional neural network. The record has been corrected accordingly; see Detailed Description.

NCT Number	NCT07581678
Lead Sponsor	Bursa Yuksek Ihtisas Training and Research Hospital
Conditions	Pain, Postoperative, Cardiac Surgical Procedures, Nerve Block
Enrollment	120 participants
Start Date	2025-09-01
Primary Completion	2026-05-01 (estimated)
Study Completion	2026-05-01 (estimated)
Updated on ClinicalTrials.gov	2026-07-17