The patient was sent back to the operating room and in the prone position, skeletonization of D8 down to the middle of the sacrum was performed, including exposure of the lateral transverse processii.

Using Inomed ISIS intraoperative monitoring, the running roots of Th 9-10-11and 12 and L4-5 and S1 were exposed and isolated by umbilical tape. It was clear that both Th 10 are not functioning and only tiny twigs can be found, which are useless for grafting and pinpoint tear of the right pleura was accidentally noticed during exploration for abnormal variation of these neural structures. Due to this fact the Th 10 branches were omitted as candidates for bridging. The pinpoint tear was covered by muscle and the anesthesia team was warned upon.

The Acromed construct was removed and it seems that the D12 down to L2 were bony fused from posterior.

Drilling of the sacrum at the trajectory of S1 was performed to expose the S1 roots at their postganglionic part was achieved in both sides.

Inspection of the running roots at the injury level revealed, that the L1,2 and 3 were damaged anatomically and there was no possibility to find and isolate them.

Using the saphenous graft, anastamosis between Th9 and S1 was performed, connecting the proximal part  of postganglionic Th9 and the distal end of the postganglionic S1 from the left.

The other saphenous graft was used to anastamose the right S1 and right L5, taking into consideration the wide diameter of one the branches of the saphenous graft, which was sufficient to fill the diameter of the right S1 root.

Using the sural nerves, anastamosis between Th11 and both L5 was undertaken and Th12 to L4 both sides using the same technique as for the fist mentioned bridge.

The grafts were lax and they were allocated, so as to be away from the bone, near the muscle in the hope to have acceptable blood supply.

Routine closure of the wound with ready-vac drain with negative pressure under the skin.

Smooth postoperative recovery. The patient sent to the ICU. The second operation took around 20 hours.

Serial check chest X-rays were performed 3 times in the ICU to follow the progression of pneumo or heamothorax, which was negative.

This kind of surgery, still in infancy and needs more perfection and more proper decision-making policies. S2 and below are less important to direct anastomoses to them.

The harvested posterior femoral nerves were not used in this case, which means, that the operation must be staged in other way.

In the next case, it is better to explore the spine first, to see exactly what nerves are useful and what nerves remained with the patient in acceptable anatomical shape. Using intraoperative monitoring is a must and this surgery must be guided with it, otherwise, putting anastamosis between two functionally invalid structures is only a waist of time and effort.

In this case the operation must be divided into 3 stages: stage 1- exploring the above and below the injury level and studying all the involved roots functionally and morphologically. The second stage, must be the harvest of the needed grafts and their required length and diameter and their number. The third stage is the bridging of the acceptable functionally and morphologically dorsal donors with the acceptable anatomically lumbo-sacral recipients.

The timing of surgery is important. This patient was reluctant for long time - 10 years after insult. This kind of surgery must be performed 8-12 months after the insult.

By planning this surgery in successive three days, the surgeon will be able to have the best choices and the more precise actions.

This case brought another task, which needs to be resolved. The damaged L1-2-3 roots, must be approached by another means: The extraperitoneal approach to the major nerves arising from them, to be identified and through tunneling to bring the grafts to the back to follow the other anastomotic connections with the dorsal donors. This must be resolved in the next case.

The patient underwent debridement of the right infragluteal area  28-September-2008 at the incision of the harvested right sensory femoral nerve. The conclusion for future operations is that not to use the sensory femoral nerves, since they are short in length and the area is susceptible to bed sore formation.