Use of cervical stand-alone cages

We have read with great interest the paper of Gercek and coworkers on their first experiences with implanting a stand-alone titanium cervical cage [Gercek E, Arlet V, Delisle J, Marchesi D (2003) Subsidence of stand-alone cervical cages in anterior interbody fusion: warning. Eur Spine J 12:513–516]. The authors describe the results of their first eight patients in whom they used the SynCage C, filled with autologous bone, to obtain fusion after anterior cervical discectomy. Mean follow-up was 15 months, and the authors found a cage subsidence of at least 3 mm in five of the eight patients. One cage fractured and was removed after this patient had recurrent symptoms of cervical radiculopathy. The authors concluded from their observations that cage insertion should not be carried out without additional anterior plating, and they interpret their findings as a serious warning to spine surgeons who want to use stand-alone cages.

As neurosurgeons who have inserted more than 400 different stand-alone cages in the past years without feeling guilty about doing something wrong, we think that this paper requires several comments. As the authors themselves clearly point out, the reported series is small. In our view, this low patient number does not allow any clear conclusions at all regarding the use of stand-alone cages. There exist larger studies in the recent literature on this issue that revealed no outcome differences between patients receiving a stand-alone cage or an autologous bone graft [5, 7].

The authors focus on the radiological results in their series, and we miss the description of the clinical outcome. Clinically, the patient with the broken cage was obviously the only patient with recurrent symptoms, although this is not clearly stated in the paper. Furthermore, the fusion rate is not reported in this work, but from the statement that “all the patients had functional radiographic evaluation, and no segmental motion or kyphosis was observed in any of these cases,” we conclude that fusion occurred in all patients.

Summarizing, in this small series, the authors obviously achieved a fusion rate of 100% and a clinical success rate of 87.5%, which compares favourably to all reported series on anterior cervical disc surgery, including the so-called “gold standard” of autologous bone graft insertion. We believe that not the radiological result but the clinical success of a surgical procedure should be the major target of reporting a clinical series. In this setting, it would have been of interest to know the further course of the patient who was re-operated because of a broken subsided cage. As shown in Fig. 4 of the paper, the cage was already broken at the 6-month follow-up. It remains unclear why the re-operation was performed 9 months later, although the symptoms recurred at the 6-month follow-up, as stated by the authors. Were there only slight symptoms after 6-months? Did revision surgery relieve the patient’s recurrent symptoms? Was the cage examined after removal to rule out mechanical damage at the time of insertion or a primary production failure? We are convinced that the cage, although broken, was completely embedded in bony fusion, and we would have performed a posterior lamino-foraminotomy in case of a true foraminal stenosis leading to the recurrent symptoms. In doing so, we would obviate the risk of a secondary fusion failure and also the risks associated with an anterior revision procedure.

Another point of debate is the relevance of subsidence. In our opinion, cage subsidence as well as bone grafts is physiological to a certain degree. It occurs due to the normal process of bone resorption and remodelling until fusion is established. As others have found, a subsidence of approximately 2 mm is generally to be expected [1, 6]. The use of constrained anterior plates does probably hinder bony fusion in some cases because settling cannot occur, and this has led to the recent development of dynamic plates, which is an excellent idea in our view [1].

The high rate of subsidence with 3 mm or more settling in the series of Gercek et al. has probably another reason. As they remarked, careful preparation of the endplates, which is an important surgical step, was performed. It is, however, obvious that the authors used relatively high cages in their patients (mainly 7–10 mm heigh). No data is given regarding the pre-operative disc height in these patients. We assume that in most cases the chosen cage height was too large owing to intra-operative over-distraction. Higher distractive forces are known to cause higher compressive forces after removing the distraction, thus eventually leading to excessive loads with the consequence of graft fracture and/or subsidence [8].

In our patients, we have generally used 4.5 to 6-mm cage heights and have not seen any fracture so far. These heights correspond better to the typical disc heights in humans [4]. With the use of the microscope in anterior discectomy, which was obviously not used by the authors, over-distraction is less likely due to better visualisation of the deep structures. Even a foraminal decompression, which is essential in our view even in cases of soft herniations, is carried out much simpler with the aid of the microscope.

Cages have the advantage of high primary stability [2], and they eliminate donor-site complications when placed unfilled or filled with bone from the operation site [5, 7]. Also, operation time is shortened. A brace is not necessary with cages in our view, as we have not seen dislocations in our patients although not using braces. The main disadvantage is still the price of these spacers. However, shortened operation time, shortened hospital stay, and no need for bracing do compensate much of the high costs of cages. Alternatively, in times of low budgets in health care, using PMMA as a spacer is very cost effective. Although not highly accepted in the Anglo-American segment of the spine community, PMMA spacers have a long tradition, especially in Germany, and are proven as effective as bone grafts in the treatment of cervical disc disease [3, 6].