Author's response to reviews
Cervical spine kinematics after anterior cervical discectomy with or without implantation of a mobile
cervical disc prosthesis; an RCT
Authors:
Toon FM Boselie, MD 1; Henk v. Mameren, MD, PhD 2; Rob A de Bie, PT, PhD 2; Henk v. Santbrink,
MD, PhD 1
(T.Boselie@mumc.nl, or afmboselie@gmail.com)
Version: 2
Date: December 28th 2014
Author's response to reviews: see over.
Reviewer 1 (Xijing He)
Reviewer's report:
MAJOR REVISION
-
-
1.The primary outcome must be clear and unique. To ascertain whether a mobile cervical disc
prosthesis restores/preserves normal motion can not be a primary outcome. Order of
movement or segmental mobility is OK. Regarding the outcome measure: we agree on this
point, the primary outcome already was normal motion, defined as a normal sequence of
segmental contributions. The text has been altered to stress this more.
2.Data analysis didn’t mentioned in the manuscript. Regarding the data analysis: we agree
with this comment, a separate section has been added to the methods section.
MINOR REVISION
-
-
-
1.Inclusion criteria and exclusion criteria should be listed in the manufacture, unless the
journal has special rules. Although readers can find them on clinicaltrials.gov. Regarding the
inclusion and exclusion criteria, which are described in the text at p. 5, from line 19 on, they
are now also added as a table to be easier to find.
2.Every figure should have numbers and illustrations under them. Regarding the absence of
figure legends, they were present as a separate figure legends section, p. 15, line 12. A
legend for Table 1 has been added.
3.The authors talked lots about adjacent segment degeneration which give readers a sense
that they will do research on adjacent segment degeneration. Regarding adjacent segment
degeneration/disease: the reviewer is correct that we do not do research on ASD. The alleged
role of fusion in ASD is however the reason that cervical disc arthroplasty exists as a
treatment, which is why it is discussed in the introduction.
DISCRETIONARY REVISION
-
1.Flexion-extension radiograph is a classical method to mesure the ROM of cervical spine.
Using a software to track the vertibrae automatically is creative and avoid a high intra- and
inter- individual variability. But the mean center of rotation(MCR) and the instant center of
rotation(ICR) of the cervical spine changed after surgery. The influence to the angles by the
change of MCR and ICR should be concerned. At this time we focus on the sequence itself.
Future study will include further analysis of the data, to look for instance for determinants that
correlate with a prosthesis (placement) which shows a normal sequence of segmental
contributions. This will likely include the ICR, amongst others.
Level of interest:An article of importance in its field
Quality of written English:Acceptable
Statistical review:Yes, but I do not feel adequately qualified to assess the statistics.
Declaration of competing interests:
None.
Reviewer 2 (Huajiang Chen)
Reviewer's report:
Major Compulsory Revisions:
-
-
-
-
1. The first objective of this RCT study is to ascertain whether a mobile cervical disc
prosthesis restores/preserves normal motion when compared to healthy controls, in contrast
to patients undergoing simple discectomy. Since the fusion rate of simple discectomy can be
up to 80%, segmental range of motion (sROM) study of this group seem make no sense. And
objective of this RCT study can not be achieved by this two group comparation. Experimental
design needs to be reconsidered. Regarding the fact that sROM study in the simple
discectomy group makes no sense: we fully agree with this fact, which is why we already state
in the introduction that sROM are often used but have very little use in this type of study, and
therefore the focus of the study is sequence of segmental contributions (from p. 4, line 28 on).
SROM will be evaluated however as an outcome measure to evaluate the presence of fusion.
Additionally we added a section explaining why, even in the presence of a fused segment
(which a large part of the simple discectomy group will gain through surgery), analysis of the
sequence of segmental contributions is of benefit.
2. The second objective of this RCT study is to evaluate the clinical results of CDA compared
to simple discectomy. However the sample size is ten per group which is obviously too small
to obtain a credible conclusions. Sample size calculation should be more detailed in this
article. Regarding the small sample size, with respect to the secondary outcomes. We agree
that the sample size is too small to show a significant difference between the two groups.
However, we have previously shown in a Cochrane review, that a clinically relevant difference
is not to be expected in this type of surgery in the short term.
The primary focus of the study is to investigate the sequence of segmental contributions, the
clinical outcomes are merely taken into account to be able to see if results for this specific
prosthesis are in line with those described in literature.
Sample size calculations are therefore based on the primary outcome, which is largely
dependent of the presence of fusion. The difference in amount of fusion will be large.
Unpublished data (which are in preparation) suggest sensitivity and specificity of our method
of analysis to be >80%, allowing for credible conclusions. We have added a section in the
discussion to clarify the questions about the sample size.
3. Statistical methods which is an indispensable part in study protocol have not mentioned in
this article. Regarding the absence of statistical methods in the manuscript: we agree with this
point, they have been added as a subsection to the methods section.
4. Both X axis Title and Y axis Title were not labeled in Figure2. Figure legend is not detailed
enough to express the meaning of Figure2. Readers may have difficult to fully understand the
meaning of this Figure. Regarding the labeling of the axes in Figure 2: the X- and Y-axis, and
legend have been altered to be clearer, to ‘sagittal rotation’ and ‘amount of flexion in C 4-C7’.
Level of interest: An article whose findings are important to those with closely related research
interests
Quality of written English: Acceptable
Statistical review: No, the manuscript does not need to be seen by a statistician.
Declaration of competing interests: I declare that I have no competing interests.
Reviewer 3 (Daniel Lubelski)
Reviewer's report:
Discretionary Revisions:
-
-
-
1. The authors describe use of a novel computer algorithm to determine the postoperative
mobility, which, from what is described seems like a thorough approach. Will these data be
analyzed (as well as defining the template areas, running the software, etc) in a blinded
fashion? Regarding the blinded analysis of the images: this is not possible, the templates are
selected by a user that can clearly see the prosthesis (if present). The analysis by the
software after that is automatic. To avoid occasional errors due to for instance ‘ghosting’ there
is the possibility to correct individual frames after the analysis, so strictly speaking that part is
not blinded as well.
2. With the small sample size, do the authors believe that their study is powered to identify
significant differences in complications, pain scale, or SF-36? Regarding the small sample
size: (also see the answer to reviewer 2) with respect to the secondary (clinical) outcomes we
agree that the sample size is too small to show a significant difference between the two
groups. However, we have previously shown in a Cochrane review, that a clinically relevant
difference is not to be expected in this type of surgery in the short term (even with extremely
large sample sizes). Sample size calculations are therefore based on the primary outcome.
We have included the secondary outcomes to be able to see if the clinical results are in line
with those for this type of surgery.
We have added a section in the discussion to clarify the questions about the sample size.
3. The Discussion may include further description of the novelty of this study. There are
dozens of prospective cohort and RCTs comparing outcomes of CDA vs ACD vs ACDF. Many
of these studies have longer followup and larger sample sizes. Many of these previous studies
have supported that CDA leads to improved segmental ROM, but much debate has revolved
around the fact that, nonetheless, clinical and quality of life outcomes are similar. In addition to
using a new implant, and a new method of analyzing the segmental range of motion, how will
the conclusions of this study advance clinical knowledge on this topic? The novelty of this
study is the fact that sequence of segmental contributions might be able to give an early
indication if a prosthesis facilitates normal motion (normal quality of motion, rather than
quantity), and if there is a possibility that it will reduce chances of ASD in the future. This
cannot be done through conventional use of sROM. In the future this method of analysis might
be applied to different prosthesis, or to investigate its use as a prognostic parameter.
Major Compulsory Revisions
-
1. In the methods, within sample size calculation, the authors performed a power analysis to
detect a 40% difference in fusion rate. Since the expected effect size is so large, they are
sufficiently powered with a small sample size of 10-12 patients per group. What is the previous
evidence that leads the authors to expect the effect to be so robust? Furthermore, why is
fusion rate the outcome by which the power analysis is based on…if the primary outcomes are
based on segmental range of motion? Regarding the sample size calculations: these are
based on the primary outcome, which is largely dependent of the presence of fusion, which is
why this is the basis for the sample size calculation. The difference in amount of fusion
between the groups at 1 year will be large (around 10% versus 80% based on literature).
Historical data show specific sequences to be consistently present in 90% of healthy controls.
Unpublished data (which are in preparation for publication) suggest sensitivity and specificity
of our method of analysis to be >80% (in differentiating between healthy controls and people
with a cervical radiculopathy). We therefore think credible conclusions can be drawn, even
with a small sample size.