Informed consent was obtained from all patients and were counseled for the use of stimulating catheters and radio-contrast media for the spread of local anesthetic (LA). In 12 patients, a CEA was administered for the isolated proximal humerus, shoulder girdle stabilization, and shoulder disarticulation. Continuous CEA was performed in a sitting position on the operating table after securing an intravenous (IV) line and a continuous monitoring of heart rate with two lead ECGs and oxygen saturation. Precautions were taken in the event of a vasovagal shock (atropine, intubation cart).

After sterile preparation, with the patients awake and sitting, a 17-G Tuohy-stimulating needle (Arrow®, Inc.) was inserted at the level of C7-T1 with a midline approach after skin infiltration with 4 ml of 1% lidocaine at the point of needle insertion. The cervical epidural space was identified with a loss of resistance technique, following which the catheter (Arrow®, Inc.) was inserted. The epidural needle hub was rotated 90° on the side to be operated before catheter insertion. A nerve stimulator (B-Braun, USA) was connected to the stimulating epidural catheter with a stylet which was primed with 0.9% saline prior to threading. Electrical leads were positioned on the abdomen, and an electrical current (1–3 mA) was applied during advancement. Contractions of triceps, biceps, and deltoid muscles were observed and noted as the catheter was threaded. The side of the evoked motor response (EMR) and current was recorded for each patient if the twitch was unilateral. The end-point was the EMR of either the biceps or the deltoid muscles (C5-7). The catheter was secured with a sterile dressing (Tegaderm; 3 M Medical, USA). Patients were made to lie supine, and the tip of the radio-opaque catheter was located on the image intensifier in the antero-posterior and lateral views. With an ongoing continuous neurostimulation, the unilateral EMR we re-confirmed and an initial dose of 1 ml of iohexol (300 mg iodine/ml) diluted in 5 ml 0.5% bupivacaine was injected. The position of the catheter was confirmed on fluoroscopy after the initial dose. Analgesia was confirmed with a light pinprick after 20 min using a 26-G hypodermic needle. Top-ups of 0.5% bupivacaine were injected in aliquots of 1 ml when the surgeon complained of inadequate relaxation, patient bleated pain on the surgical incision and at the completion of the surgical procedure when pain first appeared. The total dose of LA injected (bolus and top-ups) was noted in all patients. All patients received IV midazolam 1–2 mg prior to surgical incision. The assessment was made during the surgical incision for pain, discomfort, and change in the heart rate. Three patients for shoulder disarticulation were sedated with a propofol infusion at 100 μg/kg/min, after surgical incision confirmed block efficacy. Intraoperatively, patients were monitored with non-invasive blood pressure, electrocardiogram, and oxygen saturation. Intraoperative bradycardia or hypotension if any was recorded. An infusion of 0.1% ropivacaine at 5 ml/h was initiated immediately after the surgical procedure. Pain scores (visual analog scale) were monitored for the next 48 h at various time points (0, 6, 12, 24, and 48 h). Patients were assessed over 24 h for pain [visual analog scale pain scores 0–10 (VAS)] and side effects. IV diclofenac 75 mg was injected 12th hourly as a part of multimodal analgesia and intravenous tramadol 50 mg in 100 ml normal saline was prescribed as a rescue analgesic for a VAS more than 4. The amount of tramadol required was calculated in the first 24 h.

Block efficacy was considered adequate if there was no pain on incision, no supplementation of intravenous analgesics in the intraoperative period, and a VAS of not more than 4 in the postoperative period for the first 48 h. At the end of the 48th hour and completion of infusion, epidural catheters were removed. Prior to discharge patients were assessed for local neck pain and neurological complications, if any.

CEA was attempted in 12 patients, and surgeries were successfully conducted in 11 patients (4 males, 7 females). One patient with a dural puncture received general anesthesia. The mean age was 46.45 years, and the mean current at which the desired EMRs were achieved was 2.4 mA. With the needle hub rotation to the right, EMRs were successfully obtained in the right C5-6 myotomes in 7 patients and to the left in 4 patients. Images obtained with fluoroscope depicted the catheter tip (Fig. 1A) at vertebral levels of C5-6 in 8, at C6 in 2, and at C6-7 in 2 patients. The mean LA volume used initially and as a top-up dose was 8.08 and 4.72 ml, respectively. Contrast spread (Figs. 1B, 2A, B, 3A, B) was visualized from the lower end of C2 to the upper end of T3 with 5 to 10 ml of radiocontrast diluted in LA. The mean surgical duration was 78.66 min. The mean VAS score was 1.85 at the end of the 48th hour. Six patients required rescue analgesics (IV tramadol, mean of 75 mg). In patients who received propofol infusion for shoulder disarticulation, all awoke with VAS of 0, 1, and 2 in the postoperative recovery room. Intraoperative bradycardia in one patient and hypotension in three patients were noted after top-ups with LA. Hand grip was weak in 6 patients who received top-ups of LA and 4 among them had contralateral weakness in the immediate postoperative period. Catheters removed at the 48th hour were uneventful, and no local pain or neurological adverse effects were reported. The patient who suffered from a dural puncture did not complain of headaches or visual disturbances. Table 1 depicts the current used in milliamperes, the muscle in which evoked motor response was observed, the catheter tip location, the contrast spread visualized, and the local anesthetic volume used.

A Catheter tip located in the lateral epidural space at the level of C6 vertebral level. B Catheter tip is lateral to the mid-line at the C6 vertebral level. Contrast injections delineate the cervical nerve roots (Cnr). T1, first costotransverse junction

A Catheter tip in the posterior epidural space (PES). The upper limit of contrast is at the C2 level. B Catheter tip is lateralized in the cervical epidural space with maximum contrast spread on the surgical side and minimal spread on the non-operative side

A Lateral view: contrast spread in the posterior epidural space (PES) and anterior epidural space (AES) till the C5 upper vertebral level. B Anterior–posterior view: contrast injection depicting the spread from C5 till T2 and delineates the cervical nerve roots