Evidence based therapies and their pitfalls

By Jan Tunér

Many systematic reviews on LLLT fail to understand the importance of analysing the actual laser parameters, thus mixing apples and bananas. One systematic review overcoming such shortcomings is the excellent review by Petrucci et al. on the use of LLLT for temporomandibular disorders (TMD). The abstract reads as follows:

Petrucci A, Sgolastra F, Gatto R, Mattei A, Monaco A. Effectiveness of low-level laser therapy in temporomandibular disorders: a systematic review and meta-analysis. J Orofac Pain. 2011 Fall;25(4):298-307.

Aim: To assess the scientific evidence on the efficacy of low-level laser therapy (LLLT) in the treatment of temporomandibular disorders (TMD). Methods: The databases of PubMed, Science Direct, Cochrane Clinical Trials Register, and PEDro were manually and electronically searched up to February 2010. Two independent reviewers screened, extracted, and assessed the quality of the publications. A meta-analysis- was performed to quantify the pooled effect of LLLT on pain and function in patients with chronic TMD. Results: The literature search identified 323 papers without overlap between selected databases, but after the two-phase study selection, only six randomized clinical trials (RCT) were included in the systematic review. The primary outcome of interest was the change in pain from baseline to endpoint. The pooled effect of LLLT on pain, measured through a visual analog scale with a mean difference of 7.77 mm (95% confidence interval [CI]: -2.49 to 18.02), was not statistically significant from placebo. Change from baseline to endpoint of secondary outcomes was 4.04 mm (95% CI 3.06 to 5.02) for mandibular maximum vertical opening; 1.64 mm (95% CI 0.10 to 3.17) for right lateral excursion and 1.90 mm (95% CI: -4.08 to 7.88) for left lateral excursion. Conclusion: Currently, there is no evidence to support the effectiveness of LLLT in the treatment of TMD. J Orofac Pain 2011;25:298-307.

In the manuscript the authors have performed a qualified analysis of applied dosages and also suggest that some of the negative studies have used dosages outside of those recommended by the World Association for Laser Therapy.

So why is the outcome of the analysis negative and what is the problem with this kind of analyses? The Cochrane style of analysis requires randomized clinical studies. These are to be analysed and evaluated acc. to strict requirements. Quite often such analyses come to the conclusion that the available evidence is insufficient.

Systematic reviews present the core of our present knowledge and, if performed competently, contribute to improved treatment standards. But a poor score in a systematic review does not necessarily mean that the therapy evaluated is not working. It may only be due to few and inconclusive studies. In fact, a recent analysis on endodontics (root canal treatment), published by The Swedish Council on Health Technology Assessment (SBU) came to the conclusion that the evidence is very poor and in fact had the lowest score of all performed reviews. Still, root canal treatment has been performed for 100 years and with considerable success. The scientific problem is that studies are small, cover short periods, use different medicaments, use different techniques and so forth. This means that there are too few RCTs, and indeed a placebo root canal therapy is not ethically possible.

LLLT share some of the problems elucidated in the SBU report. Although there are some 4000 studies, they cover a vast array of medical indications, leaving but a few for each indication. Only a handful have enough qualified studies to back them up as evidenced based therapies, such as LLLT for prevention of mucositis. In the Petrucci analysis the existing RCTs were using different energies and different treatment protocols. The material was just not sufficient for any firm conclusions – fair enough.

The Cochrane style is of course very justified and highly recommended. But even the SBU warns medical professionals to rely too heavily on what is evidenced based. Just going by such guidelines not only eliminates therapies such as endodontics and other therapies which are known to work well. And it also puts a dead hand over therapies that are promising but not yet well documented. Such as LLLT. In the new National Guidelines for Swedish dentistry, LLLT for TMD is listed as “Not to be performed”. This means that it is not even qualified as a 10 on the used 1-10 scale! “Not to be performed” must mean that there are risks for the patient. But there are none. It also must mean that there is no scientific background. But there is. Not overwhelming, but at least worth to be included in the 1-10 scale. Here, the strict adhering to Evidence Based principles actually hampers the development of a very useful dental tool. This is unfortunate.

The beauty of LLLT is its global effect. What is working in the nose also works in the toe – the basic cellular effects are the same. So if we get out of the Evidence Based chains and look at, for instance, TMD, we need to look at studies outside of the label TMD. And look at studies on pain, inflammation, oedema, wound healing etc. These studies support the general effect of LLLT and can be intellectually transferred into many different medical areas.

LLLT science needs to improve in many ways. The goal is to reach an Evidence Based level in several areas. In the meantime, LLLT is qualified as an “emerging science” and due to its potential deserves to be treated accordingly. Millions are dying from pharmaceutical side effects, multiresistant bacterias are on the move, the number of chronics is increasing and dissatisfied patients are too often trying their luck with crystals, tarot cards and astrology. Many of these could be helped by LLLT instead.