Prediction of Local Anaesthetic Dosing During Labour Epidural Analgesia

Condition(s) studied

Investigational drug(s) / intervention(s)

Machine-learning prediction model: A machine-learning prediction model will be developed to anticipate the parturient's requirement of LA at admission in the Labour Suite, according to demographic, obstetric and anaesthetic features ongoing before administration of the first epidural bolus.

Study summary

Epidural analgesia is the gold standard for controlling labour pain. However, labour pain happens during neuraxial analgesia, due to anaesthetic, obstetric, maternal factors.

The investigators hypothesized that relevant variables, able to predict the local anaesthetic (LA) requirement during labour, can be identified at admission and each parturient may therefore be accordingly classified in "low-requirement" and "high-requirement". In this way, a predictive score may be developed, and the analgesic regimen may be matched to the individual patient, thus ensuring a timely and appropriate treatment of patients likely to require higher doses of LA, while minimizing potentially side effects of excessive treatment in the low-dose group.

Eligibility

Primary outcome measure(s)

• Machine-learning algorithm able to predict the LA consumption — From the epidural catheter placement to delivery.
  To develop a machine-learning algorithm able to predict the mean hourly cumulative LA consumption administered via the epidural catheter from the catheter placement up to delivery, expressed as time-weighted LA consumption per hour (mg/h) in patients receiving top-up analgesia or the need for adjunctive rescue LA boluses in patients receiving PIEB (Programmed Intermittent Epidural Bolus) analgesia.

Trial sites (1)

More Fondazione Policlinico Universitario Agostino Gemelli IRCCS trials in Italy