Predictive model for the inadequate labor epidural analgesia: an outcome of the prospective observational study at university women's hospital.

BACKGROUND Continuous epidural analgesic infusions provide superior analgesia as compared to other forms of labor analgesia. However, inadequate analgesia after labor epidurals is not uncommon and has been found to be as high as 24% in some studies. The mechanism of these failures include inappropriate epidural catheter location, tissue compartmentalization within epidural space, delayed migration, kinking, occlusion or disconnection of correctly placed epidural catheter. AIMS The aim of our study was to examine the effect of various factors on the incidence of inadequate pain relief with labor epidurals. METHODS Eighteen independent potential risk factors for failed epidurals were collected from each parturient: patient characteristics (body mass index, history of failed epidural, opioid tolerance, illicit drug use and back abnormalities), labor details (parity, singleton versus multiples pregnancies, induced versus spontaneous labor, augmentation with oxytocin, malpresentation and cervical dilatation greater than 7 cm), epidural technique (experience of the operator-resident/specialist, method of loss of resistance-air/saline, paresthesia during epidural insertion, difficult insertion, ultrasound used, and number of attempts) and other factors (time of epidural insertion). RESULTS Data collected from 502 parturients showed that difficulty in placement of epidural catheter was reported in 43 (8.6%) patients. Inadequate pain reliefwas seen in 104 (21%) parturients. Cervical dilatation > 7 cm, previous failed epidural analgesia, paresthesia during epidural insertion, and loss of resistance using air were found to be the best predictors of inadequate epidural analgesia A constructed classification table showed that the predictive model correctly classified 96.7% of successful epidurals of producing adequate pain relief. However, the predictive model correctly classified only eighteen failed/inadequate epidurals (16.8%) as failures. Overall, 79.7% of the epidurals placed were successfully classified by the predictive model. CONCLUSION In parturients identified as being at high risk for failed epidural, ultrasound guidance, saline-based loss of resistance technique, and appropriate intra-epidural-space length of catheter are the methods that should be utilized to lower the incidence of failure.