A SYSTEMATIC REVIEW?

This is an interesting work in terms of its remarkable lack of coherence. The methodology embraced by the authors is a “systematic review” that systematically ignored a substantial portion of the relevant literature. Specifically, all studies that used biometrics and imaging such as EMG, T-scan, MRI and CBCT, devices that measure the biomechanical and physiologic aspects of the relationship between occlusion and TMJ pathology were excluded. This is an egregious example of the classical fallacy Cerasus Legend (suppressed evidence).

There is a huge lack of coherence when they conclude that; “…there seems to lack ground to further hypothesise a role for dental occlusion in the pathophysiology of TMD.” This conclusion was arrived at after all of the studies using the tools that measure pathophysiology were stripped out of this review, even though they should have fulfilled their criteria to be included in the review.

This is a forced use of the very well-known fallacy, Argumentum ad ignorantiam (argument from ignorance), in which ignorance represents “a lack of contrary evidence.” Its classical form asserts that a proposition is true because it has not yet been proven false or a proposition is false because it has not yet been proven true, the latter being the case in this review. The applied fallacy´s structure is the following:

Conclusion: “Occlusion is not a cause for TMD because the reviewed papers did not prove/show it.”

It is a problem that they carefully omitted all of the articles that demonstrate it, because of their selection bias, thereby incurring the fallacy of Sensus est Imperfecta Collatio (incomplete comparison).

The supposed contrary evidence in this review cannot be attributed to causality, (considering the authors’ backgrounds), but also makes it an inductive fallacy of generalization where a small sample, insufficient or imperfect enumeration, appears to support their pre-biased conclusions.

The question guiding this “Systematic Review” was stated in the first sentence of the abstract: "Is there any association between features of dental occlusion and temporomandibular disorders (TMD)?" The same idea was also expressed in the title of the review. An interesting fact is that, in the discussion section, the authors themselves discredit their own conclusion of no relationship between occlusion and TMD once they suggest that TMJ disorders may cause occlusion problems!

“…dental literature has predominantly been directed toward the view of dental occlusion as the cause of TMDs, the inverse relationship may even be more plausible and should have been considered to explain the occasionally-described association between cross-sectionally observed phenomena.”

Worse than that, they also present a specific association that is contrary to their conclusions; “…the absence of a disease-specific association…,” as they admit osteoarthrosis may cause an open bite!

“Similar suggestions have been proposed also for the purported relationship between anterior open bite and osteoarthrosis with the former being the consequence, rather than the cause, of the latter.”

It turns out there is an association between features of dental occlusion and temporomandibular disorders, another example of the lack of coherence in this review. Of course, this is also another inductive fallacy where the hypothesis is confirmed only by the arguments that best serve the interlocutor, while the others are discarded. Then, based upon that unsubstantiated conclusion, the authors proceed to an absurd extrapolation: “…there is no ground to hypothesize a major role for dental occlusion in the pathophysiology of TMDs.”

However, since pathophysiology refers to the study of abnormal changes in body functions that are either the causes, consequences or concomitants of a disease processes,¹ it does not matter if occlusion is an element of the cause or a consequence of TMD, it does play a major role in the pathophysiology. Now, more than in any other era, we have the tools (biometrics, imaging, etc.) and the scientific knowledge (from the literature carefully omitted by these authors) to measure and correct or manage as well as to better understand the occlusion/TMD relationship.

Thus, at this point the methodologic flaws of this incomplete Systematic Review may be summarized as:

1. A flawed premise that a strong association must be present for any etiologic factor

2. An incredibly biased selection from the TMD and occlusion literature

3. Incoherent conclusions designed to favor their psychosocial theory

4. The recognition of association followed by an incoherent disregarding that contrary data without any justification

5. An improper extrapolation of their conclusion.

The methodology and logical fallacies are not the only problems of this review. After selecting 25 papers out of 1670 initial citations, the authors applied a classification of quality assessment giving eight stars (out of nine possible) for only two papers, which deserves further analysis . One is a German 2004 paper that actually found a correlation between open bite and signs of TMDs and the other a 2005 paper from Hirsch et al, that was only looking for a relationship between overbite/overjet and “joint sounds.” The first actually denies the major premise of this Manfredini et al review (the lack of association between TMJ/occlusion) and the second, which relied on joint sounds to diagnosis TMJ internal derangement (according to the invalid RDC-TMD criteria), only tested for a correlation just between “joint sounds” and overbite/overjet among all of the occlusion features.

By themselves, “joint sounds” are not a sufficient tool to diagnose all TMJ internal derangements because some conditions are quiet. This fact was belatedly recognized by the original promoters of the RDC-TMD protocol,^2,3 requiring them to update the protocol (DC/TMD) and requiring the incorporation of MRI for more accurate diagnoses.^4,5 Curiously, Manfredini himself had previously recognized this, as he had stated this already in his 2008 paper,⁶ but seems to have forgotten it in this 2017 review:

“The presence of temporomandibular joint click sounds is not an accurate predictor of a magnetic resonance diagnosis of disk position.”

So, the two best eight-star papers, as selected by Manfredini and colleagues; 1) contradict their conclusion of no TMJ/occlusion association, 2) used invalid methods for the diagnosis and 3) also fell into the same fallacy of trying to determine causation by association or the lack thereof.

Considered altogether, this “systematic review” appears more as a constructed narrative, adjusted to fit the authors personal bias rather than a scientific analysis of the complete literature. This 2017 “systematic review” lacks coherence and follows a sequence of previous papers that carefully avoid contending with their psychosocial theory of the etiology of TMD. A theory in which one of its main proponents has been already recognized as “seriously flawed” nineteen years earlier!⁷

On the other hand, it also reveals that the authors recognize the need for an understanding of the pathophysiology of the occlusal/TMJ relationship, which they state is now possible because one era is ending, but a new one is emerging, However, the new era is Measured Dentistry and its capacity to help us understand the biomechanics and neurophysiology of both the TMJ and the occlusion, thru biometrics, imaging and many other technological devices⁸ that science has made available. The era that is over is the denial era.

A final observation with respect to this review is central to a larger problem in dental science today. It is the false assumption, the fallacy of Causa Ambiguam (confusing association with causation), that an association indicates a causal relationship. In the Manfredini et al review, the fallacy was the inverted, assuming that the lack of a strong association denies a causative relationship. This is a very common problem in many dental research studies today.

When considering TMD, one must acknowledge that there are at least 40 distinct TMD conditions, according to the Taxonomic Classification of the DC/TMD⁴ and others. Therefore, it is unreasonable to expect that any one factor could represent the etiology of all 40 conditions. A given etiological factor may only apply to one or just a few of the 40 conditions and therefore should not have a strong association with all TMD.

Since association does not equal causation, the systematic review conclusion that the lack of a strong association denies occlusion any credibility as a factor in the etiology of TMD was without substance and also without the understanding the nature of causal inference.⁹ However, the authors did acknowledge this fallacy when referring to the documented association between medio-trusive interferences and TMD: “Such association does not imply a causal relationship and may even have opposite implications than commonly believed (i.e., interferences being the result, and not the cause, of TMD).” From this statement, one can only conclude that they clearly understood the fallacy of Causa Ambiguam but chose to intentionally use it to mislead the profession. Therefore, the Manfredini et al’s 2017 systematic review cannot be considered to have contributed to anything positive to the advance of the science of the occlusion/TMD relationship.