Don't Disregard Ultrasound Yet–The Jury Is Still Out

To the Editor: I am writing to provide my viewpoint on the article by Robertson and Baker titled “A Review of Therapeutic Ultrasound: Effectiveness Studies” (July 2001). I commend them on a very rigorous study. They reviewed 35 studies performed between 1975 and 1999, and then deleted 25 studies due to small sample size, nonclinical condition, and so forth. Of the 10 remaining studies, only 2 studies showed that active ultrasound performed better than placebo ultrasound. The authors drew the conclusion that active ultrasound is no more effective than placebo ultrasound. As I studied their report, I discovered flaws in the 8 studies from which they drew their conclusion. I base my opinion on a decade of laboratory research establishing correct ultrasound parameters.1–3 Ultrasound crystals are not completely uniform, and some areas transmit sound better than others. The effective radiating area (ERA) is the amount of the crystal that transmits the sound wave. The crystal is housed inside the soundhead and is slightly smaller than the applicator faceplate.1–3 To obtain optimal ultrasound benefits, the treatment size should be no more than 2 times the size of the ERA of the crystal, or roughly twice the size of the soundhead. Only one of the studies deemed acceptable by Robertson and Baker included the actual ERA; the rest were estimates. The ERA is always smaller than the applicator faceplate. Lundberg et al4 treated an area 5 times larger than the faceplate, McLachlan's5 treatment area was 10 to 25 times larger than the faceplate, and Falconer et al6 treated an area 10 times the size of the applicator faceplate. The researchers in all 3 of these studies used continuous ultrasound to increase temperature. Based on previous studies testing large treatment areas,3 no temperature increase would have been attained in these 3 studies. Ter Riet et al7 used pulsed ultrasound to treat an area up to 10 times the size of the soundhead. There is no evidence suggesting that treating large areas with nonthermal ultrasound will heal tissue. If ter Riet et al had limited the treatment size to twice the size of the soundhead, the outcomes might have been different. High-frequency ultrasound is absorbed in superficial tissues.2 Thus, a frequency of 3 MHz is typically used for tissues 0.5 to 2 cm deep, whereas a frequency of 1 MHz is used for tissues 2 to 5 cm deep.2 Lundberg et al4 used a 1-MHz frequency to treat a superficial structure (epicondylagia). If a frequency of 3 MHz had been used (and the treatment area was 2 ERA), the target tissue would have absorbed more energy, and the results might have been different. Appropriate treatment duration is essential in the use of thermal ultrasound. Falconer et al6 used ultrasound for 3 minutes to test joint mobility. Based on previous calculations by Draper et al,2 1-MHz ultrasound at 2.5 W/cm2 for 3 minutes (2 ERA) would increase tissue temperature only 1.2°C, not enough to increase mobility. In addition, their treatment size was 12 times the ERA; thus, no heating took place.3 Of the 10 studies deemed acceptable by Robertson and Baker, 7 studies used ultrasound pulsed at 25%. Although the duty cycles were consistent, treatment time was not, as it varied from 2 to 15 minutes. Evidently, no one has established the appropriate treatment time for ultrasound pulsed at various duty cycles. I find it interesting that the 2 studies that showed active ultrasound was superior to passive ultrasound had the longest treatment times (15 minutes). Had the treatment times of the other 5 studies been longer, their results might have been different. After studying the article by Baker and Robertson, I am convinced that the original researchers mistakenly compared placebo ultrasound with placebo ultrasound. What I mean by this is that a study is flawed to begin with if correct parameters are not used. The jury is still out in the debate over the effectiveness of ultrasound. Although the ideal situation might be data-based research on patients with clinical conditions, such research is difficult to perform. Institutional review boards struggle with it, and it often takes years to obtain enough subjects. For those who want to test ultrasound in clinical conditions, appropriate parameters must be used. I have had positive results on a case-by-case basis when I used appropriate ultrasound parameters.8 I thank Dr Draper for his comments on our article and agree with many of his points on dosage. His laboratory has significantly contributed to this discussion by demonstrating the extent to which ultrasound increases the temperature at different depths of soft tissue and the factors that affect this change. This provides a rational basis for selecting dosages to test the clinical effectiveness of ultrasound. Such research has to be done using a structured series of cases or more methodologically rigorous multi-center randomized controlled trials. Case reports of individual practitioners are not sufficiently rigorous at this stage to assist the “jury” in its decisions. Dr Draper is right about the jury considering therapeutic ultrasound–it is still out on many issues, as it has been for decades. The exception is ultrasound used for fracture healing. On that, the clinical evidence is convincing, and the dosage being used is clearly effective.1,2 This suggests the basis of an explanation as to why we are still discussing the clinical effectiveness of ultrasound after decades of use. If we accept that the effects of ultrasound are small, any benefits accrued by using it during a 1- to 3-week healing period may be indistinguishable from the natural course of healing. This, in addition to the points Dr Draper made on dosage, would help explain inconsistent research findings. Only over a longer time frame and where the natural course is well understood–as for fracture healing and possibly chronic tendon lesions–are the small effects of ultrasound sufficient to change the clinical outcomes. If this idea is correct, the clinical “benefits” of traditionally used ultrasound will remain elusive or be shown to be no greater than any other form of local deep heating. In the meantime, Dr Draper is correct when he suggests that the jury is still out on the traditional uses made by physical therapists of ultrasound and that many are little more than placebo. Valma J Robertson, PhD Associate Professor School of Physiotherapy La Trobe University Bundoora, Victoria 3086, Australia