Related ACE Reports
- ACE Review
- Published: Oct 2013
- ACE Report #4043
Spine: TDR improves ROM but provides similar outcomes to ACDF for pain and disability
|Not Reported||None disclosed|
|Industry funded||Synthes Spine||Royalties|
|Industry funded||Synthes USA Products.||None disclosed|
|Not Reported||None disclosed|
|Industry funded||Not listed||Other|
|Industry funded||Not listed||Other|
|Non-Industry funded||Chinese Medical Doctor Association||None disclosed|
|Industry funded||Not listed||None disclosed|
|Not Reported||None disclosed|
|Industry funded||Synthes Spine||Consultant|
|Not Reported||None disclosed|
|Industry funded||NuVasive, Inc.||Other|
|Industry funded||Medtronic Sofamor Danek||Other|
|Industry funded||Medtronic Spinal and Biologics.||Consultant|
Cervical total disc arthroplasty has been suggested as an excellent alternative to anterior cervical decompression and fusion (ACDF) in the treatment of degenerative disc disease and other cervical spine conditions. While ACDF has been proven to provide significant reductions in both pain and disability, it is associated with a reduction in cervical ROM at the operative level, which may lead to complications for adjacent vertebrae and the eventual progression of degeneration. Recent research findings have demonstrated that TDR can retain the cervical ROM at the operative level and may provide similar reductions in pain and disability, potentially preventing the progression of adjacent level degeneration. This review was necessary to assess the current high quality evidence available in order to determine if TDR does provide any significant improvements over the current standard, and if these differences are clinically significant.
What was the principal research question?
Is total disc replacement a safe and efficacious alternative to anterior cervical decompression and fusion for the treatment of single level cervical spine conditions? Additionally, does TDR provide any significant advantages when compared to ACDF?
|Report Characteristics:||18 reports were identified that assessed the use of total disc replacement (including Bryan Cervical Disc, Prodisc-C, Kineflex-C, Prestige ST Artificial Disc, PCM Cervical Disc, Mobi-C Cervical Disc, and Advent Cervical Disc) for cervical spine conditions. Two meta-analyses and 16 randomized control trials were included. The publication dates ranged from January 2006 - July 2013. A total of 2583 patients were identified from the RCTs.|
|Report Selection:||The following terms were used to search the OrthoEvidence database; "total disc arthroplasty', "total disc replacement", and "cervical disc arthroplasty". The search resulted in the identification of 42 reports for possible inclusion. Screening using the inclusion criteria led to the exclusion of 24 reports, leaving 18 reports for final inclusion. Inclusion criteria were: studies that (1) assessed the use of TDR in comparison to ACDF at a single level of the cervical spine, (2) had a minimum follow-up of 6 months, (3) and that reported one of pain, disability, range of motion, adverse events or adjacent level cervical disc degeneration.|
|Outcomes:||Outcomes assessed in the included trials were: the Neck Disability Index (NDI), Neck Pain and Arm Pain, neurological success, secondary surgeries at index and adjacent level(s), SF-36, range of motion, adverse events, and development of adjacent level degeneration and adjacent level disease.|
|Heterogeneity:||Heterogeneity was assessed using the I squared statistic for outcomes where pooling was acceptable.|
Report Details and Scores
|Randomization sequence adequately generated||Randomization sequence adequately concealed||Blinding outcome assessors||Blinding of patients||Loss to follow-up infrequent||Outcomes objective||Large enough sample size to ensure balance of outcomes|
|Anakwenze et al.|
|Anderson et al.|
|Burkus et al.|
|Coric et al.|
|Heller et al.|
|Jawahar et al.|
|Kelly et al.|
|Mummaneni et al.|
|Murrey et al.|
|Nabhan et al.|
|Nunley et al.|
|Park et al.|
|Phillips et al.|
|Sasso et al.|
|Zhang et al.|
|Zigler et al.|
|McAfee et al.|
|Yin et al.|
Yes Uncertain Not Applicable No
Pooling and Statistical Analysis
Statistical pooling was performed for 7 outcome measures: Neck Disability Index, Neck pain, Arm pain, neurological success, secondary surgeries at the index level, secondary surgeries at adjacent levels, and the SF-36 Physical Component Score. The remaining outcomes were not reported in a manner that allowed for statistical pooling.
- The results from all trials indicated that both treatments provided significant improvements in pain and disability compared to preoperative values.
- Four studies were pooled for Neck Disability Index outcomes at 24 months (n=1155). The analysis indicated no significant difference between groups (MD -1.34 (95%CI -3.07 to 0.39); P=0.13). Moderate heterogeneity was noted (I-squared=35%). Six studies were additionally analyzed qualitatively. Sasso et al. reported significantly lower NDI scores after 24 month in the group treated with TDR compared to ACDF (P=0.005). However, 2 studies (Phillips et al. and Coric et al.) observed no significant difference between groups after 24 months. At longer-term follow-up (4-5 years), Jawahar et al. and Zigler et al. reported similar NDI outcomes for both groups. In contrast, Burkus et al. observed significantly lower NDI scores in TDR patients after 5 years (P=0.022). In 4 of 5 studies which reported immediate short-term outcome (6 weeks - 3 months), NDI scores of the TDR groups were reported to be significantly lower than those in the ACDF groups at one or both of these follow-up periods.
- Four studies were included in the pooling for neck pain scores. Two studies were pooled at 6-12 months (n=142), and indicated that there was no significant difference between TDR and ACDF in this time frame (MD 3.16 (95%CI -6.16 to 12.49); P=0.51). Significant heterogeneity was observed (I-squared=96%). Three studies were pooled at 24 months (n=954), and indicated a significant difference between treatments in favour of TDR (MD -2.62 (95%CI -4.85 to -0.40); P=0.02). Heterogeneity was not significant (I-squared=17%). An additional six studies provided data on neck pain outcome, although were ineligible for pooling. Sasso et al. illustrated significantly lower pain scores in those treated with TDR after 24 months (P=0.005), while in contrast 2 studies (Murrey et al. and Coric et al.) displayed no significant difference between treatments after 24 months. Over longer-term follow-up, Zigler et al. found significantly lower pain scores with TDR treatment after 5 years (P=0.0122), and Burkus et al. demonstrated lower neck pain scores for TDR treatment, although the level of significance was not indicated. Jawahar et al. reported no significant difference between TDR and ACDF after follow-up of 48 months.
- Four studies were included in the pooling for arm pain scores. Two studies were pooled at 6-12 months (n=142), and indicated that there was no significant difference between TDR and ACDF in this time frame (MD -1.38 (95%CI -4.51 to 1.75); P=0.39). Significant heterogeneity was observed (I2=86%). Three studies were pooled at 24 months (n=954), and indicated there was no significant difference between groups (MD -1.15(95%CI -2.62 to -0.32); P=0.12). Heterogeneity was not significant (I2=0%). An additional five studies reported arm pain outcome that were not included in the pooled analysis. Murrey et al. and Coric et al. observed similar arm pain scores between groups after 2 years, and both Burkus et al. and Zigler et al. found no significant difference after 5 year follow-up. However, Sasso et al. reported significantly lower arm pain level in the group treated with TDR compared ACDF (P=0.014).
- Results from 5 of the included trials were pooled regarding neurological success (n=1387). A total of 693/742 patients who received TDR treatment achieved neurological success at 2 years, compared to 567/645 patients who received treatment with ACDF. The analysis indicated that treatment with TDR lead to significantly greater neurological success at 2 years when compared to treatment with ACDF (p=0.001, I-squared=0%). However, it should be noted that pooled results from 2 studies, with 5 year follow up periods, demonstrated no significant difference between the two treatment methods regarding neurological success (202/216 TDR vs. 169/188 ACDF) (p=0.39, I-squared=39%).
- Results from 4 studies regarding secondary surgery at the index level were pooled (n=1283). A total of 13/665 patients who received TDR underwent surgery at the index level compared to 40/618 patients who underwent ACDF. The analysis indicated that significantly less secondary surgeries at the index level occurred in patients receiving TDR within 24 months of treatment (p=0.0005, I-squared=0%). Pooled results from 2 separate studies (n=750), reporting their findings within 60 months, produced similar results to those in the 24 month analysis, once again showing that the TDR treatment resulted in few secondary surgeries at the index level (21/379 TDR vs. 65/371 ACDF) (p<0.00001, I-squared=0%). The overall finding, regarding secondary surgery at the index level, was in favour of the TDR treatment (p<0.00001, I-squared=0%).
- Five studies were pooled for incidence of secondary surgeries at adjacent disc levels (n=1437). A total of 12/733 patients who received TDR underwent reoperations at an adjacent level, compared to 23/704 ACDF. The analysis indicated a trend toward significant difference in favour of TDR treatment, however significance was not reached (OR 0.52 (95%CI 0.25-1.07); P=0.08). Heterogeneity was not present in the analysis (I2=0%).
- Pooled results from 2 of the included trials indicated no difference between TDR and ACDF treatments regarding SF-36 physical component scores, 24 months after treatment (p=0.22, I-squared= 0%).
- Range of motion (ROM) at the index level was reported in 7 trials. The results from all of the trials indicated that TDR led to maintenance of the preoperative range of motion at the index level, whereas ACDF led to a significant decrease in ROM at the index level.
- Data on adjacent segment degeneration and adjacent segment disease were reported in three of the included studies. Phillips et al. reported development of adjacent level degeneration at either the superior or inferior segment in 39.1% of TDR patients and 49.2% of ACDF patients after 2 years (P=0.111). Jawahar et al. (n=93 patients) reported development of adjacent level disease in 18% of TDR patients and 15% of ACDF patients, which was not a significant difference (P=0.885). In a serial study by Nunley et al. (n=182), the rates were reported to be 17% and 14%, respectively. Concurrent lumbar degenerative disc disease was reported to significantly increase the risk of developing adjacent segment disease (P=0.02).
- Eight studies reported adverse events. Seven of these studies indicated no differences between procedures. One study reported slightly more adverse events in the TDR group relating to surgery, while there were more serious complications and re-operations in the ACDF group.
- The findings from the two meta-analyses indicated similar findings to those observed in this report.
Forest Plot: Neck Disability Index at 24 months
Forest Plot: Neck Pain Scores
Forest Plot: Arm Pain Scores
Forest Plot: Neurological Success
Forest Plot: Secondary Surgeries at Index Level
Forest Plot: Secondary Surgeries at Adjacent Level(s)
What should I remember most?
The results from this review have demonstrated that both TDR and ACDF lead to significant improvements in pain and disability scores in patients with single level cervical spine conditions when compared to preoperative levels. Disability measured using the NDI was not significantly different between TDR and ACDF from pooled results of outcome at 24 months, while qualitatively there was contrasting evidence of the comparative efficacy between the two treatments, with both statistical significance and non-significance reported. It remains unclear if differences in disability between these two procedures were clinically relevant. Pooled pain assessment illustrated that TDR provided a significant reduction in neck pain compared to ACDF at 24 months, although qualitative analysis displayed contrasting evidence, both at short- and long-term. It remains to be determined if the differences observed were clinically significant There did not appear to be any significant difference between treatments regarding postoperative arm pain. Neurological outcome appeared to be significantly better with TDR treatment, and resulted in significantly fewer secondary surgeries at the index level. The difference between TDR and ACDF did not reach statistical significance regarding secondary surgeries performed at adjacent segments. Range of motion at the index level was preserved in the TDR patients in comparison to ACDF treatments which may have provided superior long term benefits and reduce the prevalence of adjacent level disc degeneration. However, the rates of developing adjacent segment degeneration and disease were found to be similar between treatments, although this was reported in just three studies in which the follow-up duration was relatively short.
Implications for patient treatment and future research:
The results from this review indicate that both treatments provide significant improvements in pain and disability for patients with TDR allowing for the maintenance of range of motion at the operative level. However, further long-term assessments of high methodological quality are needed to determine if this preservation in ROM can decrease the incidence of adjacent level disc degeneration. The results from this review additionally demonstrate the need for standard outcome reporting guidelines within the literature to improve comparisons between trials.
ACE Reports in this review:
The authors responsible for this critical appraisal and ACE Report indicate no potential conflicts of interest relating to the content in the original publication.