Adult spinal deformity surgery: how bad is reoperation?
For most patients, having surgery to relieve the symptoms of adult spinal deformity (ASD) is enough of an undertaking. The impact of adult spinal deformity in terms of quality of life leaves some patients in a poorer position than many patients with chronic medical illness. Furthermore, corrective surgery carries with it a risk of complications reported in the literature at between 20% and 71%. For some unfortunate patients, things can get even more difficult if they are one of the 15% to 20% who need a second operation. The need for reoperation can be due to a complication, or might be from a new problem that rapidly follows the first – either way, it’s probably bad news. But how bad is not entirely clear. A group from Barcelona (Spain) looked at the impact of second procedures from ASD on the healthcare quality of life (HQoL), and specifically the impact of unplanned reoperation within the first year of adult patients being treated for adult spinal deformity.1 This study reports a retrospective analysis of 280 patients, most of whom had been diagnosed with idiopathic scoliosis who underwent corrective surgery in one of six European centres in four countries. The authors included any patient aged 18 years or more who underwent ASD corrective surgery with a preoperative Cobb angle > 20°, thoracic kyphosis > 60°, or sagittal vertical axis > 5 cm with a two-year minimum follow-up. There were 280 patients who met the necessary criteria, with idiopathic scoliosis being the most common diagnosis. Of the initial cohort, 43 patients underwent 46 revision procedures (18%), most commonly due to implant-related complications; 43.5% of these were within the first month and 67% within the first three months following the primary procedure. Patients were more likely to undergo revision surgery if they were older, had a higher American Society of Anesthesiologists (ASA) grade, showed a greater sagittal deformity, or had a worse starting HQoL. At one year, the Oswestry Disability Index (ODI) and the Scoliosis Research Society outcome score (SRS22) were significantly better in the group who did not require revision surgery; however, the overall improvement in HQoL was no different between the two groups at two years. Importantly, the SRS22, ODI, and 36-Item Short-Form Health Survey (SF-36) outcomes improved by the minimal clinically important difference (MCID) only in the non-revision surgery group. In addition, the mental health component of the outcome scores was significantly adversely affected by revision surgery and took longer to recover to preoperative norms. So, are these findings unsurprising? Perhaps. The improvement in HQoL appears to be the same, but the changes in some of the HQoL metrics for patients likely to need revision surgery don’t reach the MCID in contrast to the non-revised group. Furthermore, revision surgery has a significant effect on patient mental health. As with all things spinal, information is power, and now we have an idea of how revision surgery impacts on a patient’s postoperative course, improving the information we offer during the consent process and informing patient decisions.
Ligament augmentation in proximal junctional failure
The soft tissues are the key to successful outcomes in all branches of orthopaedic surgery, and this is no more evident than in spinal surgery. Soft-tissue complications are the source of most woe in complex patients, and attention must be paid to the preservation of soft-tissue structures and function if stability is to be maintained. There is a contradiction between the requirement for exposure and the consequent exposure of the spinous processes and the interruption of the associated soft tissues during a posterior fusion procedure. This is furthered by the basic desire to reduce soft-tissue injury to maintain soft-tissue function, and this seems to have struck a group from San Diego, California (USA).2 Their study aims to examine the potential benefits of augmentation of the posterior ligament structures following fusion in an effort to prevent post-surgical proximal junctional failure (PJF). The group took 200 consecutive adult patients who underwent posterior instrumentation for deformity correction (after excluding neuromuscular and neurodegenerative deformity): 100 prior to the introduction of a surgical posterior ligament augment and 100 afterwards. The precise technique is described in the paper, but generally involves weaving an elastic material through the base of the spinous processes and under alternate lamina from the level above the uppermost instrumented vertebra (UIV) to the level below in an effort to reduce the junctional stresses. The reported results make encouraging reading; 67% of patients included in the intervention group were female, with degenerative scoliosis being the most common indication for surgery. Of those reported, 51% of procedures were revision, 38% required both an anterior and posterior approach, and 40% required three column osteotomies. The mean number of instrumented levels was 11. The mean increase in post-junctional angle was 6° in the intervention group, compared with 14° in the control group (p < 0.001). This subsequently led to fewer revision procedures for pseudoarthrosis (14 in the control group, three in the intervention group). On first reading, this series looks great, but there are a few inevitable caveats. First, the cause of PJF is unknown. Theories abound: associated degenerative changes, facet joint failure, vertebral fracture, posterior longitudinal ligament failure, disc disease, and so on. One of the difficulties in establishing the cause of PJF is that there isn’t even a strict definition of what constitutes PJF. The therapies employed differ as well. The authors in this study comment that they tend to use vertebroplasty alongside the index surgery to reduce the risk of postoperative fracture in vulnerable patients, that they use sublaminar hooks at the UIV (because of the evidence of reduced PJF when these are used), and that they use in situ contouring to reduce proximal stresses. Some of these techniques are not universally applied and so could confound the results; however, ligament augmentation looks like it could be useful in the battle against PJF.
Revise, revise, and revise again: complications and revisions in 50 000 disc procedures
Anterior cervical decompression and fusion (ACDF) and anterior cervical decompression and arthroplasty (ACDA) are two of the most rewarding operations to perform in spinal surgery, and often have enough of a functional benefit that the procedure can literally turn a patient’s life around. ACDF is the benchmark surgical intervention for single-level degenerative myelopathy. However, ACDF has historically been associated with complications, including adjacent segment ossification and degeneration in up to 3% of patients, which can create more symptoms. An answer to this was ACDA, where the cervical disc is replaced with a device maintaining motion at this segment, hence relieving the stresses on adjacent motion segments. This, of course, adds the attendant risks and complications of an arthroplasty procedure. However, it is not clear how the two techniques compare with regards to reoperation. A team from St. Louis, Missouri (USA) has looked at a state-wide database of all patients aged between 18 and 65 years who underwent ACDF or ACDA between 2003 and 2010, and investigated patient outcomes for the postoperative five years (as far as they can be established in a database study).3 Patients were excluded if they were operated on for infection, neoplasia, or pathological fracture. The authors identified 52 395 cases, of which 1469 were ACDA. There was a significantly higher short-term readmission rate for ACDF when compared with ACDA, although ACDA showed a higher risk of vertebral artery injury. No correlation existed between the procedure and infection. Reoperation within the short term was higher in ACDF group (3.55%) when compared with ACDA (2.04%) (p = 0.015), although long-term reoperation rates were equivocal. These results look favourable for ACDA. However, the higher incidence of vertebral artery injuries is probably misleading due to the disparity in numbers within each treatment group and the low absolute incidence. Early studies of ACDA were more supportive of the technique, and the authors comment that this may be due to the bias that tends to influence commercially funded studies. Furthermore, any study using large databases as a source of data have inherent weaknesses around clerical accuracy and loss to follow-up. Patient selection for each procedure again is likely to influence outcome, particularly in light of widespread spondylosis being a contraindication to ACDA. A useful technique then emerges and, in light of this paper, ACDA seems to be as safe as ACDF with regards to reoperation, accepting the inevitable learning curve. The difficulty with studies like these is that they are inherently biased, and those selected for the procedure are the patients who the surgeon feels will do well with one or other treatment. All this study can say is that, in the reported population, ACDA is no more likely to result in a reoperation than ACDF.
Annular closure in lumbar microdiscectomy: a randomized clinical trial
The risk of reherniation following a discectomy for disc herniation can be up to 18% within two years of initial surgery with the risk being greater the larger the defect in the annulus. There is little evidence as to which procedures, if any, can be employed to reduce this risk. This study reported from Innsbruck (Austria) aims to evaluate one potential way to address this risk.4 These authors report a multicentre randomized study conducted across Europe with the aim of assessing a bone-anchored mesh-based annular closure device. The study population involved patients who at discectomy were found to have a large annular defect (6 mm to 10 mm width). They were then randomly allocated either to the closure device (n = 276) or to undergo the discectomy alone (n = 278). The study was designed to evaluate the co-primary endpoints of recurrent herniation and a composite endpoint of patient-reported, radiographic, and clinical outcomes. These were assessed over a two-year follow-up period. In short, the results of this well-conducted study favoured the annular closure device; there was less overall recurrent herniation (50% vs 70%, p < 0.001), and greater composite endpoint success (27% vs 18%, p = 0.02). The perhaps much more clinically relevant measures of frequency of reoperations to address reherniation was also significantly different: 5% with the annular closure device and 13% in controls, which was mirrored by the reported symptomatic reherniation rates (12% vs 25%). There were, however, no differences reported between the two groups in terms of back pain, leg pain, ODI, and health-related quality of life over the two-year follow-up, which is a curious observation given the differences in reherniation rates seen. The intervertebral disc is known to have a limited healing potential and especially so when faced with a large annular defect. This closure technique is potentially an ideal non-fusion option to reduce the reherniation rates seen in lumbar discectomy. The concern for most surgeons is that there is the potential for a mesh to come loose or erode into the dura, causing significant harm. Only larger studies with longer follow-up can address this concern. There is, as they say, plenty of food for thought here.
Parallel closing laminectomy rongeur
We perhaps pay less attention to our instruments in orthopaedic surgery and research than we should. Whilst it is common place to see studies including randomized controlled trials evaluating different implants from an efficacy perspective, it is rare to see even the lowest-evidence-grade study addressing the question of instrumentation design and its effect on clinical outcomes as a research question. The Kerrison rongeur is the most commonly used specialist instrument by spine surgeons and is used in the majority of decompression procedures. The instrument design has undergone little modification since its invention and is similar across most manufacturers. The authors of this study from St Gallen (Switzerland) have performed a biomechanical assessment of a modified Kerrison rongeur that features a parallel closing mechanism, comparing it with the traditional, classic design.5 Whereas the classic design involves two angled handles that are compressed like scissors, the modified version had a parallel handgrip. The study involved volunteers (including surgeons and office staff) using both instruments whilst shaft movement was measured with a stereoscopic, contactless image correlation system. The authors also produced force diagrams from both instruments. The authors undertook a mechanical study with 40 volunteer surgeons from a range of disciplines (orthopaedic surgery, n = 35; urology, n = 2; and neurosurgery, n = 3). Participants were randomized to either the older or newer instruments, and then undertook ten punches with the first instrument, subsequently using the other instrument. The main outcome measure of the study was instrument shaft movement, which was measured in 3D space using a stereoscopic, contactless, full-field digital image correlation system. The authors effectively showed that the modified version of the instrument had specific design advantages: its longer lever arm led to a smaller closing angle of the handles, meaning a shorter range of movement in the hand and thus greater accuracy. Also, for a given output, a smaller input force was required. These mechanical advantages were borne out by the shaft movement analysis: movement was much more precise, with less movement in all three dimensions. For the spine surgeon, these advantages are potentially valuable. Decompression surgery involves operating near delicate structures and multilevel surgery involving hard bone can be particularly fatiguing. Perhaps spine surgeons should give this new design a try?
Selective thoracic fusion for Lenke 1 adolescent idiopathic scoliosis
With a selective thoracic fusion, the challenge is predicting which compensatory lumbar curves will spontaneously correct, and which will not, following the fusion procedure. Getting this prediction wrong results in the risk of either under-correction, lumbar curve progression, and the need for further surgery to fuse additional levels (‘adding-on’), or fusing unnecessary lumbar segments. Most of the current decision-making strategies centre on coronal plane imaging including traction films and assessing which vertebrae are stable and lie partly within the path of the central sacral vertebral line. This study from Philadelphia, Pennsylvania (USA) assesses whether the patient’s sagittal profile influences spontaneous lumbar curve correction.6 The study involved analysis of the imaging studies of 63 patients, all with Lenke 1B and C curves, as well as imaging from 20 control subjects. The authors required biplanar x-rays pre- and post-surgery to generate a 3D reconstruction. The study involved two aspects. First, the authors assessed the rate of spontaneous lumbar curve correction following surgery and then divided patients for reporting purposes into two clusters depending on the amount of correction achieved (high amount of correction vs low correction). Second, the curves were assessed to determine the ratio of apical translation of the thoracic and lumbar curves from a central sacral line, and the relationships between these two aspects were investigated. The key finding here was that patients with a greater preoperative sagittal thoracic to lumbar apical translation ratio were associated with lower spontaneous lumbar correction. This is a novel approach to investigating an age-old problem. Whilst sagittal parameters are scrutinized in adult degenerative deformity, the focus in idiopathic scoliosis is assessing coronal features such as lumbar vertebral rotation, curve flexibility on side bending, and determining the stable vertebra. For the deformity surgeon, the ratio of thoracic to lumbar apical translation is a potentially useful parameter to guide decision-making. Nevertheless, a larger prospective clinical study is needed to demonstrate its reliability from plain radiographs.
Cervical discectomy and fusion with or without anterior plate fixation
There is somewhat of a dichotomy in clinical practice with regards to the use of anterior plates in anterior cervical spine discectomy and fusion. Whilst the removal of disc material and surgical approach is taken as a given, some surgeons would undertake a fusion without plate supplementation due to worry about space occupancy, and complications – specifically dysphagia and swallowing complications. Others would not undertake a fusion without a plate due to their worries about failure of fusion due to instability. This team based in Rochester, Minnesota (USA) undertook a thorough systematic review and, where possible, a meta-analysis to attempt to answer some of these questions.7 Their review aimed to compare these two treatment strategies with regards to postoperative surgical, radiographic, and patient-reported outcomes following anterior cervical discectomy and fusion (ACDF) with or without plate fixation. The study team were able to include a total of 15 studies (12 observational and three randomized controlled trials) in their review. These trials reported the outcomes of 893 patients overall. The bottom line from this study is that overall, ACDF with plate fixation was associated with significantly higher vertebral fusion rates (odds ratio (OR) 1.98), lower subsidence rates (OR 0.31), and more favourable visual analogue scale (VAS)-neck pain scores (mean difference 0.59) at last follow-up. On the flipside, ACDF without use of an anterior fusion plate had better long-term VAS arm pain scores (marginal mean difference 0.2). Perhaps most importantly, the authors found no differences in the rates of dysphagia (OR 1.21, 95% confidence interval 0.57 to 2.56). Overall, this meta-analysis, although based on limited data, supports the use of plates to achieve a more stable fixation with less subsidence and a higher fusion rate without the feared increase in dysphagia rates.
References
1. Nunez-Pereira S , Vila-Casademunt A , Domingo-Sabat M et al. . Impact of early unanticipated revision surgery on health-related quality of life after adult spinal deformity surgery. Spine J2018;18:926-934.CrossrefPubMed Google Scholar
2. Safaee MM , Deviren V , Dalle Ore C et al. . Ligament augmentation for prevention of proximal junctional kyphosis and proximal junctional failure in adult spinal deformity. J Neurosurg Spine2018;28:512-519.CrossrefPubMed Google Scholar
3. Kelly MP , Eliasberg CD , Riley MS , Ajiboye RM , Soohoo NF . Reoperation and complications after anterior cervical discectomy and fusion and cervical disc arthroplasty: a study of 52,395 cases. Eur Spine J2018;27:1432-1439.CrossrefPubMed Google Scholar
4. Thomé C , Klassen PD , Bouma GJ et al. . Annular closure in lumbar microdiscectomy for prevention of reherniation: a randomized clinical trial. Spine J2018 (Epub ahead of print). PMID: 29730458CrossrefPubMed Google Scholar
5. Külling FA , Ebneter L , Rempfler GS , Zdravkovic V . A new parallel closing mechanism for the laminectomy rongeur makes it significantly more precise: a biomechanical and mechanical comparison study. Eur Spine J2018;27:1172-1177.CrossrefPubMed Google Scholar
6. Pasha S , Flynn JM , Sankar WN . Outcomes of selective thoracic fusion for Lenke 1 adolescent idiopathic scoliosis: predictors of success from the sagittal plane. Eur Spine J2018. (Epub ahead of print). PMID: 29564609CrossrefPubMed Google Scholar
7. Oliver JD , Goncalves S , Kerezoudis P et al. . Comparison of outcomes for anterior cervical discectomy and fusion with and without anterior plate fixation: a systematic review and meta-analysis. Spine2018;43:E413-E422.CrossrefPubMed Google Scholar
