Total En Bloc Spondylectomy For Lower Lumbar Spinal Tumors In Posterior-only Approach: Feasibility And Clinical Assessment

Part I.Radiographic study of total en bloc spondylectomy for lowerlumbary spinal tumorsObjective.The main objectives of this study were to analyze and evaluate spinopelvicmorphology and its relationships with lower lumbar spine, and to establishi a reference oflower lumbar spine parameters in normal adults, and to study the feasibility of total en blocspondylectomy for lower lumbar spine in a posterior-only approach.Methods.From October2012to August2013，162patients with low back pain, including81males and81females were included in this study. Spinopelvic parameters were analyzedon anteroposterior and lateral X-rays in a standard standing position. The followingspinopelvic parameters were measured by software, including sacral slope, the verticaldimension between lower edge of L4and L5vertebra body and level of iliac crest vertex(showed as AB and BC).Results.The average AB, BC and AB/BC in males were (6.26±8.81）mm（,29.07±7.07）mm,（0.28±0.46）, respectively. The mean values of AB, BC and AB/BC in females were（14.40±9.26）mm,(18.40±6.96）mm,（1.21±1.51），respectively. There was significantdifference between the male and female groups(P＜0.001). The average SS was （32.33±7.25）°in males and (33.38±7.52)°in females. No significant difference was foundbetween two groups in SS(p＞0.05).Conclusions.The distance between L4verytebral body and the iliac crests wassignificantly higher in female adults compared with the male adults(P＜0.001). The blocingeffect on the lower lumbar spine in females is less than that in males. The feasibility of TESfor lower lumbar spine is higher for females compared with the males. PartII.Anatomical feasibility study of total en bloc spondylectomy forlower lumbar spine in a posterior-only approachObjective. To study the feasibility of total en bloc spondylectomy (TES) for lowerlumbar spinal tumors and provide anatomical basis.Methods. Fourteen human cadavers was used to carry out simulated surgical operationsof TES for lower lumbar spine. The cadavers included7males and7females. They weredivided into two groups, including L4surgical group (6cadavers) and L5surgical group (8cadavers). The vertical dimension between lower edge of L4vertebra and level of iliac crestvertex, the maximum distance of L3, L4, and L5nerve roots were measured. Therelationships of iliac creast, nerve roots, segmental arteries, major vessels and lower lumbarspine were recorded to evaluate their influence on the surgery.Results. The average of the vertical dimension between lower edge of L4vertebra andlevel of iliac crest vertex in male and female was（-3.56±12.30）mm，（11.61±11.12）mm，respectively, with significant statifical differences between the two group（P＜0.05）.Total enbloc spondylectomy was performed successfully in all cases in L4surgical group. Nerve rootstretch was encountered in5cases. Avulsion of unilateral L4nerve root and cerebrospinalfluid leak were observed in one case, respectively. Of8cases in L5surgical group, TES wassuccessfully achieved in all female cases and only2in male cases. Of4female cases, L5nerve root stretch was found in3，avulsion of L5nerve root and thecal sac in1. Of the2malecases, nerve root stretch and avulsion of L5nerve root and thecal sac were observed in1,respectively. The vertebral bodies of the other two cases can not be rolled out safely from oneside of the dural sac because of the block of the lumbar nerve roots and the iliac wings.Conclusions. The key points of TES for lower lumbar spine were the block of lumbarnerve root, lumbosacral plexus and ilic crest. It is feasible that TES for L4can be acieved in aposterior-only approach. TES for L5was significantly influenced by gender, nerve roots andiliac crest. Female patients may be more suitable for such surgery. One the contrary, it is moredifficult for male patients to undergo the procedure. Part III.Biomechanical analysis of spinal reconstruction following total enbloc spondylectomy for lower lumbar spineObjective. To evaluate and compare the mechanical stability of alternativereconstruction methods following total en bloc spondylectomy in the lower lumbar spine.Methods. Six adult fresh cadaveric lumbosacral spines (L1-S1) were used. Allspecimens were divided into2groups randomly. Group I was short segment instrumentation,and group II long segment instumentation. After intact testing, total spondylecotmy of the L4vertebra was performed. Eight designed reconstruction techniques were tested for the range ofmotion of the spine. Group I:(1) expandable artifical vertebral body and posterior shortsegment (L3-L5) fixation (SP),(2) expandable artifical vertebral body posterior short segment(L3-L5) with additional antreolateral fixation (ASP).(3) expandable artifical vertebral bodyand posterior long segment(L2-S1)fixation (MP),(4) expandable artifical vertebral bodyposterior long segment (L2-S1) with additional antreolateral fixation (AMP). Nondestructivebiomechanical testing was performed on each construct under load control. Range of motion(ROM) for each construct was obtained by applying pure moments in flexion, extension,lateral bending, and axial rotation.Results. In flexion, extension and lateral bending, all the reconstructed constructssignificantly decreased (P＜0.01) the ROM compared to the intact. The ROM of specimenswith anterolateral fixation was less when compared to the ones without additional fixation (P＜0.05). No significant difference was found in each group (P＞0.05). In axial rotation, theROM was significant decreased only in the circumferential support constructs (P＜0.01).Short segmental instrumentation increased the ROM compared to the intact, but no significantstatistical difference (P＞0.05).Conclusions. The stability of the reconstructed construct after total en blocspondylectomy is mainly inflenced by the segmental length for posterior fixation. Longsegmental fixation provided more stability to the construct than short one. Conmpared to thespecimens without additional anterolateral fixation, circumferential fixations showed a higherstability. Multilevel segmental instrumentation with circumferential fixation provided themost stability of the consturct. Part Ⅳ.Total en bloc spondylectomy for tumors in lower lumbar spine in aposterior-only approach: clinical application and outcomesObjective. To study the feasibility of total en bloc spondylectomy for tumors in lowerlumbar spine and evaluate the clinical outcomes.Methods. From March2011to October2013,20patients who had undergone total enbloc spondylectomy (TES) in posterior-only approach were enrolled in this study. Thepatients included8males and12females, with a mean age of51years (range:21-65years).Tumors involved in the fourth lumbar spine were16cases and5in the fifth lumbar vertebra.Tomita classification: type1in2case, type2in5cases, type3in6cases, type4in6cases,type5in1cases. Surgical modified instruments have been developed for TES, including wiresaw with a diameter of0.54mm, wire saw guide,“Z”-notched pedicle osteotome, curvedvertebral spatulas. We applied Tomita’s TES surgical technique to the lower lumbar spine andmodified the surgical procedures. Preoperative evaluation according to clinical, imaging andpathologic features was performed meticulously to select patients. The pedicles are cutbilaterally with a “Z”-notched osteotome or combined “Z”-type osteotome and wire saw.Discetomy was performed anterior-posteriorly using wire saws until and then discetomyknives were used to cut the intervertebral discs.Results. Total en bloc spondylectomy was performed successfully in15patients with L4spinal tumors and4patients with L5spinal tumors. Even though one patient with L4Ewing’ssarcoma was achieved complete resection, tumor residual was also detected after surgerybecause the tumor involved in the prevertebral soft tissue. Therefore, this case was notconsidered as ture TES. One L5giant cell tumor can not be rolled out safely from one side ofthe dural sac because of the block of the lumbar nerve roots and the iliac wings. The L5vertebral body was removed in two pieces in sagittal resection. Average operative time ofeach and estimated blood loss were299min (range,225～445min) and2435ml (range,1200～4100ml), respectively. The mean follow-up time was18months (range,6～37mons). Allpatients except one encountered nerve roots stretch and9patients showed lower extremetiesneurological dysfunction. Of the9patients,5were seen in L4and4in L5. However, all of thepatients improved in4weeks postoperatively and recomvered completely at6months.Cerebrospinal fluid leak was found in3patients. One patient with Ewing’s sarcoma in L4hadtumor recurrence because of tumor residural. Six patients with metastatic tumors were alive with disease at the last follow-up. One patient with liver cancer had evidence of othermetastatic sites. Titanium mesh cage subsidence was observed in3patients, including one inL4and2in L5. No implant failure was occurred in the last follow-up.Conclusions. Total en bloc spondylectomy for tumors in the lower lumbar spine in aposterior-only approach is feasible. However, the indications for TES are extremely limited,expecially for spinal tumor of L5. Detailed preoperative evaluation and surgical staging ismandatory. The adverse event profile of these surgeries is high in neurological complicationseven at experienced centers. Therefore, experienced, multidisciplinary teams should performthese complex surgeries.