Biolaser • Galway Laser Therapy Clinic, 98 Lower Salthill, Galway, Ireland. T: 091 582883 E:info@biolaser.ie

General
Arthritis
Musculoskeletal
Nerve Regeneration
Tendinitis
Wound Healing

General

Clinical Reviews

1. Are all the negative LILT studies really negative? Jan Tuner, Lars Hode

The authors of this book have performed an analysis of a number of frequently cited studies on the effects of low-power-laser therapy. In many of these studies, analysis uncovered one or more reasons for the negative findings reported, the most common being the use of extremely low doses. Other reasons are: faulty inclusion criteria, inaccurate control group definition, ineffective methods of therapy, inadequate attention to systemic effects and tissue condition, and low power density. A weakness often encountered in these studies is their failure to provide sufficient data on laser parameters. Since negatively inclined studies such as these are often quoted as "proof" of the ineffectiveness of LLLT, it is important that they be subjected to a proper critical analysis. 1.400 articles were reviewed for this analysis, the emphasis being on double-blind studies. Of the 135 localised double-blind studies, 85 reported positive findings.

Though important, the critical examination of scientific literature is decidedly unglamorous. It involves hours and days of searching through a wide variety of different sources, and by no means all information is yet available on-line. There are numerous pitfalls, too, especially for those who opt to read abstracts only - criticism of sources is impossible unless an article can be studied in its entirety. Basing an opinion from abstracts obtained from e.g. Medline is risky. In addition, only a minority of the early LLLT research reports are available from the major databases.

In the following analysis of the available literature, we have chosen to analyse those studies unable to demonstrate the effectiveness of LLLT. Although priority was given to double-blind studies, non-double-blind studies were also included in certain typical cases. Certain studies were also included merely on the grounds that they are among the most frequently cited.
The 1.400 articles reviewed for this analysis are now being stored in computerised form.

"I heard it through the grapevine"

A recognisable pattern is often distinguishable in the bibliographies accompanying scientific reports. The manner in which these patterns arise goes something like this: Researcher A is the man or woman behind some pioneering achievement and is therefore extensively quoted by researcher B, as well as by C, D, E and several others. Researcher K, however, is content to read what E has written about A and B, while researcher Z treats the work of A and B as a simple historical reference point previously described by researcher P. And, like a rumor, word spreads: everyone knows about A, B and C, but no-one has actually read their published work. Although generally known, therefore, older studies are not always relevant and it may sometimes be rewarding to go back and review them in detail. Often, especially in the light of new findings, the impression given is quite unlike that suggested in later, second-hand reports.

Positive from negative

Having traditionally concentrated on studies positive to LLLT, over the last few years we found ourselves becoming more and more interested in those studies with a negative spin: provided they have been properly carried out, they may be able to show us the parameters that do not appear to work. Naturally, negative reports must always be taken seriously, but the fact that a given study has been unable to demonstrate the effectiveness of LLLT does not necessarily mean that the method studied is incapable per se of producing results within the indication in question. All that it shows is that the parameters selected for the study were not sufficiently effective. Therefore, it is illogical to conclude that LLLT is ineffective simply because no effect was reported in that particular study. A number of studies reporting negative results are marred by such startling illogicality.
Negative from negative

LLLT is a relatively young science that has only just emerged from its Sturm und Drang period, and it might perhaps be unfair to criticise the earlier negative studies. Many medical researchers then had - and indeed still have - a rather diffuse knowledge of physics, and qualified books on the physics of laser therapy were long in appearing. In many cases, the only information available to researchers on doses, methods of treatment and suitable indications came from the manufacturers or agents, while over-optimistic, ignorant salesmen often laid traps that would ensnare both themselves and the researchers.

Many studies will come in for criticism in the following paragraphs, although the researchers involved need not always take this to heart. As often as not they pioneered new territory, seeking either to retain an open mind towards the examination of new methods or reacting to what they perceived as a lack of objectivity. The purpose of our criticism is described in the Introduction - we wish to draw the reader's attention to the fact that many negative studies are poorly structured and are therefore largely irrelevant, even though they constantly feature in bibliographies and reading lists and are cited as "evidence" that LLLT does not work. Few people appear to have actually read them. As we see it, it is high time they were weeded out so that they can no longer function as traveller's tales in the future.

Important parameters

A. Wavelength
That biological effect is significantly related to the wavelength of the light emitted by the laser has been demonstrated in numerous studies. Today, the wavelengths most commonly used for therapeutic purposes are 633 nm (HeNe lasers), 635 nm, 650 nm, 660 nm, 670 nm (InGaAIP lasers), 780 nm, 820 nm, 830 nm (GaAIAs lasers), 904 nm (GaAs lasers), and 10600 nm (CO2 lasers). Except for GaAs and CO2 lasers, all these lasers usually produce a continuous beam but may also be pulsed.

B. Dose
The most important parameter in LLLT is always the dose, often referred to as "fluence". By dose (D) is meant the energy (E) of the light directed at a given unit of area (A) during a given session of therapy. The energy is measured in J (joules), the area in cm2, and, consequently, the dose in J/cm2. Mathematically, this may be expressed as follows:
E
D = ---- [J/cm2]
A
Assuming that the power (P) output of the laser probe remains constant during treatment, the energy (E) of the light will be equal to the power multiplied by the time (t) during which the light is emitted. The dose may then be calculated as follows:
P t
D = ---- [J/cm2]
A

Sometimes, however, the power output is not constant, such as when the laser is pulsed or modulated, which may be achieved in several ways. The preferred method of pulsing a HeNe laser is to use some form of mechanical switching device or shutter, such as a rotating pierced disc, the useful proportion of the time during which light is emitted by the laser normally being fixed at a given value (duty cycle), most often 50%. In other words, light is permitted to pass through the disc for 50% of the total operating time (and is blocked for the remaining 50%). This enables use of the concepts of mean power (Pm) and maximum power. In the example given here, the mean power is 50% of the maximum power. If the laser is pulsed at mean power, the above formula will apply, giving:
Pm t
D = ------- [J/cm2]
A
GaAs lasers always pulse, the duration of each pulse being extremely short, and in these lasers the maximum power is always much, much greater than the mean power. This type of pulsing is often referred to as super-pulsing. In GaAs lasers, the duration of the pulse is normally in the region of 100-200 ns (nanoseconds) and the maximum power is typically 1 - 20 W (watts). Assuming, for example, that the duration of the pulse is 150 ns and that the maximum power is 10 W, each pulse emitted by the laser will have an energy of 1.5 µ J (microjoules).

If the laser emits 100 such pulses per second (a pulse frequency of 100 Hz), its mean power output will be 0.15 mW (milliwatts). A pulse frequency of 1000 Hz gives a mean output of 1.5 mW, etc. In other words, the mean power output varies with the number of pulses emitted per second.

By applying these relationships, it is often possible to obtain doses or other parameters not explicitly stated in the article under review.

C. Power density
Power density, indicating the degree of concentration of the power output, has also increasingly proved to play a major role. It is measured in watts per square centimeter (W/cm2). If, for example, a circular area having a diameter of 5 mm (approx. 0.2 cm2) is illuminated with a laser operating at a power output of 100 mW, the biological effects are quite different from those produced by illuminating a circular area of 5 cm diameter (approx. 20 cm2) with the same laser. In the first case, the power density is 100 times greater than the second. Some studies have concluded that the power density may be of even greater significance than the dose. This parameter is very seldom indicated in the articles we have studied. It must also be remembered that the power density varies within the area illuminated - normally, it will be greatest at the centre.

Typical traditional laser instruments

Whenever possible, we also reviewed the brochure or brochures describing the instrument used for the study. The power output of the first commercial therapeutic lasers was very low. HeNe instruments often achieved an output of 1 - 2 mW at the laser tube, while the losses sustained in the optical fibers were frequently 50% or more. Further-more, the laser was sometimes pulsed (usually switched to produce a duty cycle of about 50%), thereby reducing the power/mean output power sent to the tissue by another half.

Brochures describing GaAs lasers often only specify the maximum pulse power, whereas the mean output power, which is of the greatest significance for LLLT, is often not named at all.

In the following we were often obliged to make a number of assumptions, since it was only very seldom that all the parameters were indicated in the studies. For example, unless otherwise stated, we have assumed that the fiber incurred a 50% loss; if the light was pulsed we assumed a duty cycle of 50%; and, unless otherwise indicated, we assumed that the mW values quoted envisage power output at the laser rather than at the fiber opening. For GaAs lasers, we sometimes had to make several assumptions at once, since here there are more parameters to be taken into account and they were often incompletely reported.

During the eighties, considerable discrepancies between actual outputs and those stated in manufacturers' brochures were not unusual. Only a handful of authors stated whether they themselves measured actual output at the tissue or whether they merely relied on the figure quoted in the brochure. Clearly, we must expect the dark figure to be quite large here.

Dose development

A number of early positive reports on the clinical effects of very weak HeNe lasers suggested that there was cause for some optimism - and skepticism, too. Among them are Walker (1983) [E1] (calculated at approx. 0.005 J per point) and Snyder-Mackler (1988) [E2, 3] (calculated at approx. 0.01 J per point), reporting on the effect of very weak HeNe lasers.

It must be remembered that Mester had been working with doses of around 1 J as far back as the early seventies. Later, in an article published in 1971 [E4] he recommended a dose of 1.5 J/cm2 as conducive to wound healing. The HeNe laser he used had an output of 25 mW at the laser. For a long time Mester's papers attracted little attention in the West, since they were published in relatively unknown journals. Later, in 1981, Kana [E5] published a study on the healing of open skin wounds in which he presented an analysis of the biological effect of 4, 10 and 20 J therapy.

The instrument he used was an HeNe laser producing an output of 25 mW from the laser tube. Mester's and Kana's experience of doses suitable for wound healing still hold good today. Although HeNe lasers with a power output of 25 mW were extremely expensive at the time, it cannot be held that information on suitable doses was not then available. It should be noted, too, that the treatment of pain requires larger doses than does the healing of open wounds. It seems that a large proportion of the negative studies concentrated mainly on testing the reliability of studies such as [E1, 2 and 3] without regard to existing knowledge of reasonable doses.

Pitfalls

1. Low outputs
In the following we review some of the studies in which low dose can plainly be identified as the most significant negative factor. We have also listed the parameters that we consider should always be specified in studies of this nature. It is not unusual for an author to criticise previous studies for inadequate specification of parameters, then himself to be found guilty of the same sort of omission.

In the following examples, the parameters are summarised in tabular form. It should be noted that the power output is here to be understood as mean output on pulsing, since this is the figure required in order to calculate the dose.

Study #1

Author: Waylonis G.W. et al: Ref no: [E6]
Title: Chronic Myofascial Pain: Management by Low-Output Helium-Neon Laser Therapy.
Published in: Arch Phys Med Rehab. 1988; 69: 1017-1020.
Laser type: HeNe-laser (633 nm) Output: Not specified
Pulsing: Not specified Pulse frequency: Not specified
Dose: Not specified
Power density: Not specified Treatment distance: Not specified
Laser model: Dynatron (model 1120), with fiber optics
Treated area: All together 12 acupuncture points
Treatment time: 15 sek per point No of patients: 62
No of treatments: 2 x 5 (6 weeks inbetween) Time between treatm: Not specified

Our comments:
This study is frequently quoted. No dose is specified. However, other sources state that the tube output of the HeNe laser (Dynatron 1120) is less than 1 mW. Assuming that losses in the fiber-optic set-up reduce this to 0.5 mW, and given an irradiation time per point of 15 seconds, the dose will be 0.5 mW x 15 sec = 0.0075 J. Since a normal dose today is 0.5 - 2 J per acupuncture point and 1 - 4 J per trigger point, it is hardly surprising that no significant effect was observed. And since the instrument used can be pulsed, the dose and the effect may actually have been reduced still further.

The study is said to have been double-blind, although there is no description of how this was achieved. This would, in fact, have been valuable information, since double-blind studies are normally quite difficult to carry out with HeNe lasers - they use red, visible light that is immediately distinguishable from conventional red light by its characteristic laser speckles.

Study #2

Author: Jensen H. et al: Ref no: [E7]
Title: Is Infrared laser effective in painful arthroses of the knee?
Published in: Ugeskr L¾ ger. 1987; 149: 3102-3106.
Laser type: GaAs-laser (904 nm) Output: 0,3 mW
Pulsing: Yes. 200 ns puls width Pulse frequency: 190-250 Hz
Dose: Not specified
Power density: Not specified Treatment distance: Not specified
Laser model: Space Laser IR CEB
Treated area: All together 4 points per knee
Treatment time: 180 sek per point No of patients: 29
No of treatments: 5 Time between treatm: 1 day

Our comments:
Although the dose is not explicitly stated, approximate figures may be calculated from other data. The power output is given as 0.3 mW, although in Space's instruments (as in many other GaAs lasers), the output is directly proportional to the pulse frequency. At 1000 Hz, these Space instruments usually produce an output of 1 mW. The pulse frequency interval is stated as being 195 - 250 Hz. On the basis of the power output stated, the dose may be estimated as 0.0003 W x 3 x 60 sec = 0.054 J. Four points were treated on each knee, giving a total dosage per session of 0.2 J. This dose is totally inadequate for a part of the body as large as the knee. This was a double-blind cross-over study.

Study #3

Author: Basford J R et al: Ref no: [E9]
Title: Low-energy Helium Neon laser treatment of thumb osteoarthritis.
Published in: Arch Phys Med Rehab. 1987; 68: 794-797.
Laser type: HeNe-laser (633 nm) Output: 0.9 mW
Pulsing: Continuous Pulse frequency: -
Dose: Not specified
Power density: Not specified Treatment distance: Not specified
Laser model: Dynatronics (modell not specified), via fiberoptics
Treated area: 4 different points around 3 joints (All together 12 points)
Treatment time: 180 sek per point No of patients: Not specified
No of treatments: 9 Time between treatm: Not specified

Our comments:
Assuming that the fibre loss is about 50%, the dose will here be 15 sec x 0.9 mW x 0.50 = 0.007 J per point. No obvious effect can be expected from such a low dose. This was a single-blind study.

Study #4

Author: Taube S et al: Ref no: [E10]
Title: Helium-neon laser therapy in the prevention of postoperative swelling and pain after wisdom tooth extraction
Published in: Proc. Finn Dent Soc. 1990 (86) 1: 23-27
Laser type: HeNe-laser (633 nm) Output: 8 mW (tube)
Pulsing: Pulsed Pulse frequency: 50 Hz
Dose: Not specified
Power density: Not specified Treatment distance: Not specified
Laser model: Biotronical Laser MC-8
Treated area: Not specified
Treatment time: 120 sek before suturing and day 2
No of patientes: 17
No of treatments: 2 Time between treatm: 24 hrs

Our comments:
Assuming a 50% fibre loss and a 50% pulsing loss, the total dose will be 2 mW x 120 sec x 2= 0.48 J. This is a low total dose for such major surgery. Also the number of treatments are low.

Study #5

Author: Lundeberg T, Haker E, Thomas M Ref no: [E11]
Title: Effect of laser versus placebo in tennis elbow
Published in: Scand J Rehab Med. 1987; 19: 135-138.
Laser type: HeNe-laser (633 nm) Output: 1.56 mW
Pulsing: Continuous Pulse frequency: -
Laser type2: GaAs-laser (904) Output: 0.07 mW
Pulsing: Pulsed Pulse frequency: 73 Hz
Dose: 0.09 J/point (HeNe), 0,004 J/point (GaAs)
Power density: Not specified Treatment distance: 1 mm
Laser model: Modell was not specified, nor if fiberoptics was used
Treated area: 10 different acupuncture points through a 1 mm transparent plastic disc
Treatment time: 60 sek per point No of patientes: 82
No of treatments: 10 per point Time between treatm: 2 treatm / week

Our comments:
The doses are so low that significant effects can hardly be expected.

Study #6

Author: Masse J-F et al Ref no: [E12]
Title: Effectiveness of soft laser treatment in periodontal surgery
Published in: Internat Den J. 1993; 43: 121-127.
Laser type: HeNe-laser (633 nm) Output: 0.27 mW
Pulsing: Continuous Pulse frequency: -
Laser type2: GaAs-laser (904 nm) Output: 0.8 mW
Pulsing: Pulsed, 200 ns pulse width Pulse frequency: 47.5-3040
Dose: Not specified
Power density: Not specified Treatment distance: 1 mm
Laser model: Stomalaser, independent measuring of power
Treated area: Not specified
Treatment time: 2 min 30 sek No of patientes: 28
No of treatments: 1 Time between treatm:

Our comments:
In this report, the authors studied the effect of combined HeNe/GaAs therapy on bilateral free autogenous gingival grafts and, commendably, performed independent measurement of the output specified by the manufacturer. The HeNe output, specified as 4 mW, proved actually to be 2 mW and a mere 0.27 mW after sustaining heavy losses in the fibre-optic rig. The maximum peak power output of the GaAs laser, given as 2 watts, was found to be only 0.8 watts. The size of the area treated is not specified, but assuming it was 1 cm2, the dose will be 0.0022 J/cm2 GaAs, plus 0.04 J/cm2 HeNe, that is, a total dose of 0.0422 J/cm2. Further, a single treatment is not likely to give significant results.

Study #7

Author: Smith R J et al Ref no: [E13]
Title: The effect of low-energy laser on skin-flap survival in the rat and porcaine animal model
Published in: Plastic and Reconstructive Surgery, 1992; 89 (2): 306-309
Laser type: HeNe-laser (633 nm) Output: 2.75 mW
Pulsing: Continuous Pulse frequency: -
Dose: Not specified
Power density: 310 mW/cm2 at probe tip Treatment distance: 1 mm
Laser model: Biostim 2000
Treated area: Four dorsally based skin flaps with distal demarcation of necrosis
Treatment time: 30 sek/cm2 No of patientes: 82
No of treatments: 5 Time between treatm: 24 hours

Our comments:
This study specifies just about everything but the dose, although this may be calculated as being 0.0825 J/cm2 per day. Five sessions of treatment were given before the skin flaps were prepared, five afterwards. Therapeutic treatment carried out before surgical invasion of healthy tissue is probably of questionable value. The total dose per flap will therefore be 5 x 0.0825 J/cm2 = 0.4125 J/cm2. This dose is quite low. The control procedure may also be called into question since symmetrical flaps were prepared on the right and left sides of the animal and only one side was irradiated. This procedure ignores the systemic effects of laser treatment (see below).

Study #8

Author: Klein R G et al Ref no: [E14 ]
Title: Low-energy laser treatment and exercise for chronic low back pain: double-blind controlled trial.
Published in: Arch Phys Med Rehab. 1990; 71: 34-37
Laser type: GaAs (904 nm) Output: 10 diodes of each 0.4 mW
Pulsing: Pulsed Pulse frequency: 1000 Hz
Dose: Stated : 1.3 J/cm2 per point. Calculated: 0.1 J/cm2
Power density: Not specified Treatment distance: Not specified
Laser model: Omniprobe
Treated area: Not specified
Treatment time: 4 min per point No of patientes: 20
No of treatments: 12 Time between treatm: Three times per week

Our comments:
The authors state that a GaAs laser was used to produce a point dose of 1.3 J/cm2, the indication being the heterogeneous diagnosis of "low back pain". However, analysis of the parameters given show that the dose was in fact only 0.1 J/cm2 (Pm = 2 W x 2 x 10 -7 sec x 1000 Hz = 0.4 mW; t = 240 sec; D = Pm x t = 0.1 J/cm2) and that the total dose was 5 J. In our experience, this recalcitrant indication calls for 2 - 4 J/cm2.

Study #9

Author: Seichert N. et al: Ref no: [E8]
Title: Wirkung einer Infrarot-Laser-Therapie bei weichteilrheumatischen Beschwerden
Published in: Therapiewoche, 1987; 37: 1375-1379.
Laser type1: GaAs-laser (904 nm) Output: (each of 5 diodes): 1.2 mW
Pulsing: Yes. 200 ns pulse width Pulse frequency: 1200 Hz
Laser type2: HeNe-laser (633 nm) Output: 6.5 mW
Pulsing: Continuous Pulse frequency: -
Dose: Not specified
Power density: Not specified Treatment distance: 15 cm
Laser model: Space Laser MIX 5
Treated area: Circular area, 6 cm diameter = 28 cm2
Treatment time: 10 min = 600 sek No of patientes: 18
No of treatments: 5 Time between treatm: Once per day

Our comments:
Although the author claims that his instrument is a GaAlAs laser, it is clear from the wavelength (as from the brand) that it is actually a GaAs laser. The dose is not explicitly stated, but, for the GaAs laser, can be calculated to 600 x 0.0012 x 5/28=0.128 J/cm2. On top of this comes the NeHe dose, which is more or less the same (0.139 J/cm2). See below for the purported double-blind procedure.

Study #10

Author: Mulcahy D et al Ref no: [E41]
Title: Low level laser therpy: a prospective double blind trial of its use in an orthopaedich population.
Published in: Injury. 1995; 26 (5): 315-317.
Laser type: Not stated Output: 35 mW
Pulsing: Not stated Pulse frequency: Not stated
Dose: 1 J/cm2 “of skin”
Power density: Not specified Treatment distance: Not specified
Laser model: Not stated Treated area: Not specified
Treatment time: Not stated No of patientes: 20
No of treatments: 8 Time between treatm: 2-4 days

Our comments:
Very little is known about the parameters, such as wavelength, pulsing/continous and treatment technique. The indications are plantar fasciitis, trochanteric bursitis, tendonitis, lateral epicondylitis, knee pain, cervical pain and lumbar pain. If applied in the compressive mode, 1 J/cm2 may be a reasonable dose (but of the low side) for some of the indications but certainly subclinical for indications such as cervical pain and lumbago. The list could be made much longer.

For example, Krikorian [E15] used a HeNe laser and a dose of 0.05 J/cm2 to study wound healing in rats. Zarkovic [E16] used a GaAs laser and a dose of 0.0004 J/cm2 to study pain perception in mice. Using a HeNe laser and a dose of 0.004 J/cm2, Jarvis [E17] could (naturally) find no evidence of stimulation of thermoreceptors in humans.
It is interesting to quote the abstract of the negative study by Siebert [E40] and then compare it to the analysis made by Baxter [E35].

Siebert: “The efficacy of "athermic" lasers (HeNe/GaAs) in the treatment of tendinopathies was investigated in a randomized double-blind study. On 10 consecutive days, 64 patients (32 therapy, 32 placebo) were treated for 15 minutes each with a switched-on or switched-off laser under otherwise identical conditions. The extent of movement in involved joints (neutral 0 method) and rating on a pain scale for rest in pain, movement pain, and pressure pain before treatment, after treatment, and 2 weeks after conclusion of therapy, as well as infrared thermography, served to check therapy.
After the end of therapy, a significant reduction (P = < 0.001) of 50% was shown for resting pain as well as reductions of 30% for movement and 30% for pressure pain. This result was identical in the therapy group and in the placebo group. There was also no indication of a different result of therapy between the therapy and placebo groups with regard to the thermographic control and the extent of movement.
The breakdown of the data in terms of age, sex, and duration of disease did not provide any indications of different results for placebo or therapy. It was striking that the patients who reported sensations during or after the treatment (irrespective of whether pleasant or unpleasant) had a greater reduction of pain than the patients without sensations. This laser therapy thus did not show any effect above and beyond that in the untreated group”.

Baxter: “Despite being highly critical of the standard of previous laser research, these investigators employed a non-contact technique in their trial, irradiating the patients’ skin from a distance of 10 cm. Given the beam divergence of clinical LLLT apparatus, the use of such a distance would appear to be inappropriate, producing minimal power and energy densities on the irradiated tissue. This, coupled with the apparent inaccuracies in calculation of doasge (by a factor of 10), casts serious doubts upon the reliability and validity of the reported findings”.

2. Inclusion criteria
In study by Hansen [E18], 40 patients suffering from various "chronic orofacial pains" were tested. The pain had on average lasted 4.9 years (0.5 - 42 years). Twenty-eight patients suffered from “The burning mouth syndrome” (oral dysesthesia), five from toothache in one single tooth, three from tension headaches, etc. There were no objective pathological findings. X-ray examination had revealed nothing. In the headache group, acrylic splints had proved ineffective. A GaAs laser was used, and the initial dose of 2.4 J/cm2 was increased to 4.8 J/cm2 if no effect was observed initially. According to the literature (quoted in the study),The burning mouth syndrome is considered to be either multi-factorial, psychosomatic or purely psychogenic. It is therefore safe to assume that the tissue treated for this study was healthy in every respect. That no pain relief was obtained and that there was an absence of 5-HIAA in the patients' urine is therefore only to be expected.

3. Lack of proper control groups
Seichert [E8] (part1) used a combined GaAs/HeNe laser in a study of patients suffering from various rheumatic complaints. The probes housed five GaAs diodes producing an output of 1.2 mW per diode, while the HeNe laser produced an output of 6.5 mW. One group was given GaAs + HeNe, another HeNe with the GaAs diodes switched off (the placebo group). As both the "verum group" (0.258 J/cm2) and the "placebo group" (0.134 J/cm2) were actually both treated by laser, there is in fact no control group here. The comparison made is not between laser and placebo but between a combination GaAs + HeNe laser and a HeNe laser.

4. Therapeutic technique
Moustsen [E19] did not find any effect of LLLT in sinuitis. This is understandable, because a 30 mW 830 nm laser was used, applying 3J on four skin points close to the nose. While on the low side, such therapy would reduce the sinui-nasal obstruction, but not affect the sinuitis in itself. Successful therapy requires irradiation intraorally over the projections of the sinus, 8-10 J per point, 4-6 points depending on condition and repeated daily for 2-4 days.

5. Systemic effects
The systemic effect of therapeutic laser light has been described by many researchers, such as Braverman [E20], Rochkind [E21], Airaksinen [E22], Inoue [E23] and Schindl [E24]. Essentially, a systemic effect is one such that treatment of a given complaint at one site will also tend to affect a similar complaint elsewhere. It is therefore important to observe caution in interpreting the result of studies in which parts of the test person's/ /animal's body has been treated by laser and another part of the same body has been used as a control, especially in small animal studies.

6. Tissue condition
Persson [E25] observed no effect on angiogenesis in a study of experimental human gingivitis. Although this finding is confirmed in a study by Kusakari [E26], this author does report LLLT as causing an increased flow of blood, which was not studied in Persson's work. Both studies, however, are marred by the fact that healthy humans/animals were used as test subjects. Gingivitis was, for example, induced in young, completely healthy individuals whose immune system was in excellent order. A study by Kozlov [E27] reported that LLLT had a moderate effect on slight periodontitis, a good effect on more manifest periodon-titis, and little effect on advanced periodontitis. The immunological con-dition of the tissue and of the individual is a significant factor in the effectiveness of LLLT, and a "genuine" clinical condition cannot be achieved in a study based on healthy volunteers. This may explain the discrepancy sometimes noted between clinical work and scientific tests. The significance of the condition of the tissue can clearly be seen in an experiment by Steinlechner [E28] in which keratinocytes, present in 1% and 5% solutions of fetal calf serum, were irradiated with laser light. The cells in the less nutritious solution were stimulated most. The same observation has been made by Yamamoto [E36], irradiating human fibroblasts. There is reason to beleive that the outcome of many in vitro studies has been just as influenced by the nutrient conditions of the cells as by the laser parameters.

7. Power density
According to a study by van Breughel [E29], power density appears to be more important than total dose in wound healing. Lundeberg [E30] used a 6 mW HeNe laser to treat venous leg ulcers. 4 J/cm2 was said to be given to the ulcers. Ulcer size ranged from 3-32 cm2. Treatment technique is not stated. Regardless of technique, it would take between 36 minutes and 6 hours to achieve the stated dose, per wound and session. Using a sweep technique with a focused beam, the power density would be around 0.15 W/cm2. If a defocused beam was used to cover the entire wound (32 cm2), energy density would be around 0.00019 W/cm2, which is lower than the energy density of the normal illumination in an operatory, which is extremely low.

A dose miscalculation is probable but the authors of the study have been reluctant to reveal the parameters used. In the absence of such parameters, this study cannot be properly evaluated, but very low power density is a probable reason for negative results. In another study on venous ulcers by Malm [E37] and Lundeberg, GaAs was employed. 4 mW was used for 10 minutes on ulcers ranging from 4 to 52 cm2, regardless of ulcer size. The 4 cm2 wound would thus receive 0.6 J/cm2 and the largest wound 0.046 J/cm2, not the 1.96 J/cm2 stated by the authors. Energy density as well as dose for larger wounds are thus low. Treatment technique is not indicated. "The laser was held perpendicular to the surface of the wound". This is not a sufficient description of the treatment method. There is a great difference between following the outer border of the wound (active healing area) and spreading the beam over the open wound area. The shape of the invisible GaAs beam is also important to know. The distance between diode and wound is not indicated.

8. Mixed parameters
A study by Hall [E31] is confidently entitled "Low level laser therapy is ineffective in the management of rheumatoid arthritic finger joints". Such a title suggests that all reasonable parameters (wavelengths, doses, pulsing, etc.) have been checked and are kept properly under control. One of the probes used is a so-called cluster probe having a 15 mW GaAlAs laser diode surrounded by 30 non-coherent light-emitting diodes of three different wavelengths. Which wavelength was effective, which was ineffective, and did any particular wavelength have a detrimental effect on the overall result?

Blending coherent and non-coherent light while giving therapy with light of different wavelengths may possibly result in some clinical improvement but may also result in less improvement than if the therapy had been more exactly controlled. Tina Karu, for example, has showed that in cell cultures first irradiated with laser light and showing a clearly demonstrable biological effect, the effect is reduced practically to zero if the cells are then irradiated with broad-band (non-monochrome, incoherent) light [E35]. The main objection to mixed parameters is that they do not result in unequivocal new knowledge.

9. The influence of ambient light
The influence of ambient light is not a clinical problem. In the laboratory, however, it may influence the outcome of a study. As mentioned above Karu [E35] has shown that the effect of laser light may be partly or completely washed out by broadband light. In a study by Lundeberg [E38] the effect of HeNe and GaAs laser on the generation of signals in a sensory receptor was studied. The stretch receptor of Astacus fluviais (crayfish) was mounted over a glass platform. HeNe 1.56 mW and GaAs 0.07 mW, 73 Hz was used, dose not indicated. Disecting the stretch receptor and mounting it is done under the microscope, using very bright broadband ligth. This disturbing factor must be taken into account when evaluating similar trials.

10. Premature conclusions
Some researchers have used titles or conclusions such as "laser therapy is ineffective in...". This is an unscientific approach, since these reports have solely investigated a few of the many possible parameters. The language used reflects an unscientific bias. An example [E42] of this is: "Our results indicate that the analgesic effects reported in humans with similar modes of LLLT might be due to placebo" (comparison of response time on tail-flick between laser acupuncture, morphine and electrical stimulation).

11. Meta-analyses
Two meta-analyses [E32, E33] of the clinical effects of LLLT have been performed. In one, various studies were graded by points on the basis of a number of quality criteria. By this means, a negative study may achieve a high grade even if one of the parameters - though it may be essential to the result of the study - is entirely wrong. In Beckerman's study [E32], for example, Basford's study [E9] has been given the highest score (18 out of a possible 25) in spite of the fact that the dose used is not actually therapeutic. A comparison might be made with a chain: it is no use having 19 perfectly sound links if the 20th is open. The chosen method is important. Bjordal [E41], using a systematic review system, found a higher validity for the same studies than Gam did.

Confusion between groups
I (LH) took part in double-blind study several years ago. When the code was to be broken, the investigator having been entrusted with the envelope could not find it. There was certainly a significant difference between the two groups. However, it is not ethical to decide which group is supposed to be the placebo group just by looking at the outcome of the study, so the whole study had to be cancelled. Knowing how easy it is for codes to be mixed up, for documents to be mixed up and for misunderstandings to arise when several persons are involved in a study, it is not too unlikely for mistakes to occur. This may be the case in the following study by de Bie [E43]:

Study #11

Author: de Bie et al Ref no: [E43]
Title: Low-level laser therapy in ankle sprains: a randomized clinical trial
Published in: Arch Phys Med Rehabil 1998 Nov;79(11):1415-20
Laser type: GaAs Output: Not stated
Pulsing: Pulsed Pulse frequency: 500, 5 000 Hz
Dose: 0.5 and 5 J/cm2 “on skin”
Power density: Not stated Treatment distance: Not stated
Laser model: Not stated
Treated area: 1 cm2
Treatment time: Not stated No of patientes: 217
No of treatments: 12 Time between treatm: 2-3 days

The investigators got the following results:
Function was significantly better in the placebo group at 10 days (p = 0.01) and 14 days (p = 0.03).
Placebo group performed significantly better on days of sick leave (p = 0.02) and at some points for hindrance in activities in daily life and pressure pain.
Placebo group performed significantly better on subjective recovery (p = .05).
Total days of absence from work and sports were remarkably lower in the placebo group than in the laser groups, ranging from 3.7 to 5.3 and 6 to 8 days, respectively.
The total number of relapses at 1 year in the laser group (n = 35) was significantly higher than in the placebo group (n = 13) CONCLUSIONS: Neither high- nor low-dose laser therapy is effective in the treatment of lateral ankle sprains.

Our comments:
de Brie et al came to the conclusion that laser therapy has no effect on ankle sprains. This is an obvious misinterpretation of the presented results. The laser treatment had a clear NEGATIVE effect on all studied parameters. There are several studies showing that laser therapy, on the indication choosen and with the parameters used, had no better effect than placebo. However, among the 2000 studies known to us, no study has reported such a clear negative effect of laser therapy, except for a few studies using doses clearly in the inhibiting dose range. We have the feeling that the two groups might have been mixed up.
Conclusion:
In studying the therapeutic effects of laser light, there are many pitfalls along the way, and many is the researcher who has fallen in. Unfortunately, their work is still cited as evidence that LLLT does not work. In fact, many of these older studies should be disregarded in future discussions on this subject, since they are clearly irrelevant. Meta-analyses, too, may also prove meaningless unless pitfalls of this type can be avoided. In our analysis we have criticised a number of well-known negative studies, and it would be right to insist that a number of positive studies should be subjected to critical scrutiny as well. Positive studies have, however, been the subject of critical comment for many years, whereas only Baxter [E34] and Bjordal [E41] have hitherto made a detailed analysis of the parameters of negative studies.

Litterature: (no cross reference with main index)
[E1] Walker J: Relief from Chronic Pain by Low Power Laser Irradiation. Neuroscience Letters. 1983; 43: 339.
[E2] Snyder-Mackler L et al: Effect of helium-neon laser irradiation on skin resistance and pain in patients with trigger points in the neck or back. Physical Therapy. 1989; 69:
[E3] Snyder-Mackler L et al: Effect of helium-neon laser on musculoskeletal trigger points. Physical Therapy. 1986; 66: 1087.
[E4] Mester E et al: Effects of laser rays on wound healing. Am J Surg. 1971; 122: 532-535.
[E5] Kana J S et al: Effect of low-power density laser radiation on helaing of open skin wounds in rats. Arch Surg. 1981; Vol 116: 293.
[E6] Waylonis G W et al: Chronic myofascial pain: management by low-output helium-neon laser therapy. Arch Phys Med Rehab. 1988; 69: 1017
[E7] Jensen H et al: Infrarø d laser - effekt ved smertende knæ artrose? Ugerskr Læ ger. 1987; 149.
[E8] Seichert N et al: Wirkung einer Infrarot-Laser-Therapie bei weichteilrheumatischen Beschwerden in Doppel-blindversuch. Terapiwoche. 1987; 37: 1375.
[E9] Basford J R et al: Low-energy Helium Neon laser treatment of thumb osteoarthritis. Arch Phys Med Rehab. 1987; 68: 794-797.
[E10] Taube S, Piironen J, Ylipaavalniemi P: Helium-neon laser therapy in the prevention of postoperative swelling and pain after wisdom tooth extraction. Proc Finn Dent Soc 1990; 86: 23-27.
[E11] Lundeberg T, Haker E, Thomas M: Effect of laser versus placebo in tennis elbow. Scand J of Rehab Med. 1987; 19: 135.
[E12] Masse J-F et al: Effectiveness of soft laser treatment in periodontal surgery. Int Dent J. 1993; 43: 121-127.
[E13] Smith R J et al: The effect of low-energy laser on skin-flap survival in the rate and porcine animal models. Plastic and Reconstruct Surg. 1992; 89 (2): 306-310.
[E14] Klein R et al: Low-energy laser treatment and exercise for chronic low back pain: a double-blind controlled trial. Arch Phys Med Rehabil. 1990; 71: 34
[E15] Krikorian D et al: Use of HeNe laser for treatment of soft tissue trauma: evaluation by Gallium-67 Citrate scanning. J Orthop Sports Phys Ther. 1986; 8 (2): 93-97
[E16] Zarkovic N et al: Effect of semiconductor GaAs laser irradiation on pain perception in mice. Lasers in Surgery and Medicine. 1989; 9: 63-66
[E17] Jarvis D et al: Electrophysiologic recording and thermodynamic modeling demonstrate that Helium-Neon laser irradiation does not affect peripheral A? - or C-fiber nociceptors. Pain. 1990. 43: 235-242.
[E18] Hansen H, Thorø e U: Low power laser biostimulation of chronic orofacial pain. A double-blind placebo controlled cross-over study in 40 patients. Pain. 1990; 43: 169.
[E19] Moustsen P et al: Laserbehandling af bihulebetæ ndelse i almen læ gepraxis vurderet ved en dobbeltblind kontrolleret undersø gelse. (In Danish). Ugeskrift Læ ger. 1991; 153(32): 2232. [E"Laser Treatment of Sinuitis in General Medical Praxis Evaluated in a double blind study"]
[E20] Braverman B et al: Effect of Helium-Neon and Infrared Laser Irrradiation on Wound Healing in Rabbits. Lasers in Surgery and Medicine. 1989; 9: 50.
[E21] Rochkind S. et al: Systemic Effects of Low-Power Laser Irradiation on the Peripheral and Central Nervous System, Cutaneous Wounds and Burns. Lasers in Surgery and Medicine. 1989; 9: 174.
[E22] Airaksinen, O et al: Effects of laser irradiation at the treated and non-treated trigger points. Proc. 4th Intern Symposium. Acupunc & Electrother Res. 1988; 13 (4): 238-239.
[E23] Inoue K et al: Suppressed tuberculine reaction in guinea pigs following laser irradiation. Lasers in Surgery and Medicine, 1989; 9: 271-275.
[E24] Schindl L et al: Influence of low-power laser irradiation on "arthus phenomenon" induced in rabbit cornea. Laser Therapy. 1994; 1; 23
[E25] Persson L, Rydé n H, Preber H, Bergströ m J: Effect of low-energy laser on gingival inflammation. Swedish Dental Journ. 1990; 14: 47.
[E26] Kusakari H, Orikasa N, Tani H: Effects of low power laser on wound healing of gingiva and bone. Laser Bologna '92, p. 49. Monduzzi Editore S.p.A., Bologna, Italy.
[E27] Kozlov, V et al: Lasers in diagnostics and treatment of microcirculation disorders under parodontitis. SPIE Proc. 1995; Vol. 1984: 253-264.
[E28] Steinlechner C, Dyson M: The effect of low level laser therapy on the proliferation of keratinocytes. Laser Therapy. 1993; 5 (2): 65.
[E29] van Breughel H et al: Power density and exposure time of HeNe laser irradiation are more important than total energy dose in photo-biomodulation of human fibroblasts in vitro. Lasers in Surgery and Medicine. 1992; 12: 528
[E30] Lundeberg T, Malm M. Low power HeNe laser treatment of venous leg ulcers. Ann Plast Surg. 1991; 27: 537.
[E31] Hall J et al: Low level laser therapy is ineffective in the management of rheumatoid arthritic finger joints. Br J Rheuma. 1994; 33: 142.
[E32] Beckerman H et al: The efficacy of laser therapy for mucoskeletal and skin disorders: a criteria-based meta-analysis of randomized clinical trials. Physical Therapy. 1992; 7 (72): 483
[E33] Gam A N et al: The effect of low-level laser therapy on musculo-skeletal pain: a meta-analysis. Pain. 1993; 52: 63-66
[E34] Baxter C D: Therapeutic lasers. Theory and practice. 1994. Churchill Livingstone.
[E35] Karu T: Photobiological Fundamentals of Low Power Laser Therapy. IEEE Journal of Quantum Electronics. 1987; QE23(10): 1703.
[E36] Yamamoto Y et al. Effect of low-power laser irradiation on procollagen synthesis in human fibroblasts. J Clin Laser Med Surg. 1996; 14 (3): 129-132.
[E37] Malm M, Lundeberg T. Effect of low power gallium arsenide laser on healing of venous ulcers. Scand J Plast Reconstr Hand Surg. 1991; 25: 249-251.
[E38] Lundeberg T et al. Low power laser irradiation does not affect the generation of signals in a sensory receptor. Am J Chinese Med. 1998; 16 (3-4): 87-91.
[E39] Siebert W et al. What is the efficacy of "soft" and "mid" lasers in therapy of tendinopathies? A double-blind study. Archives of Orthopaedic & Traumatic Surgery 1987;106 (6): 358-63
[E40] Mulcahy D. et al. Low level laser therpy: a prospective double blind trial of its use in an orthopaedich population. Injury. 1995; 26 (5): 315-317.
[E41] Bjordal JM, Greve G : "What may alter the conclusions of reviews ?". Physical Therapy Reviews. 1998; 3: 121-132
[E42] Lundeberg T, Hode L, Zhou J. A comparative study of the pain-relieving effect of laser treatment and acupuncture. Acta Physiol Scand. 1987; 131: 161.
[E43] de Bie RA, de Vet HC, Lenssen TF, van den Wildenberg FA, Kootstra G, Knipschild PG, Low-level laser therapy in ankle sprains: a randomized clinical trial. Department of Epidemiology, Maastricht University, The Netherlands. Arch Phys Med Rehabil 1998 Nov;79(11):1415-20