An Integrated Artificial Intelligence Approach for Predicting Analgesic Time Based on Nalbuphine Versus Morphine as Adjuvants to Bupivacaine in Ultrasound-Guided Supraclavicular Block

An Integrated Artificial Intelligence Approach for Predicting Analgesic Time Based on Nalbuphine Versus Morphine as Adjuvants to Bupivacaine in Ultrasound-Guided Supraclavicular Block

This study investigated the effect of adding nalbuphine or morphine to bupivacaine for supraclavicular brachial plexus block in upper limb surgeries. Sixty adult patients were randomized into three groups: control (bupivacaine + saline), nalbuphine, and morphine. The primary objective was to compare the duration of analgesia between the groups. A secondary goal was to assess whether artificial intelligence (AI), specifically the k-nearest neighbor (KNN) algorithm, could predict analgesic duration based on patient clinical and demographic data. The study concluded that both nalbuphine and morphine significantly prolonged analgesic duration and that the AI model showed high predictive accuracy.

This prospective, randomized, double-blind clinical trial was conducted at Al-Zahraa and Damietta University Hospitals to evaluate the effectiveness of nalbuphine and morphine as adjuvants to bupivacaine in ultrasound-guided supraclavicular brachial plexus block. Sixty ASA I-II adult patients scheduled for upper limb surgeries were enrolled and divided equally into three groups. Group C received 0.5% bupivacaine with saline; Group N received bupivacaine with nalbuphine (50 μg/kg); Group M received bupivacaine with morphine (50 μg/kg). The primary outcome was analgesic duration, measured from block performance until the first request for postoperative analgesia. Secondary outcomes included onset and duration of sensory and motor block, total postoperative analgesic consumption, pain scores, and complications.

In parallel, a machine learning model using the K-Nearest Neighbor (KNN) algorithm was developed to predict analgesic duration from demographic and hemodynamic parameters. Exploratory data analysis and clustering methods confirmed the complex relationship between variables. The KNN model demonstrated high predictive accuracy (correlation coefficient ~0.95). The study concluded that both adjuvants extended analgesic duration and that AI models can assist in personalizing analgesic strategies based on patient profiles.

Inclusion Criteria:

Adult patients aged 21-60 years

ASA physical status I or II

Scheduled for elective upper limb surgery below the elbow

Provided written