RF plus electromagnetic muscle stimulation: a sham control and an ultrasound endpoint — rare conditions in this layer

A 2022 trial. The authors describe it as the first sham-controlled randomised trial of a device delivering synchronised radiofrequency and HIFEM.

• 72 patients randomly divided — 48 active (mean age 45.5), 24 sham (mean 44.6)
• Both groups received three treatments on the abdomen, once a week
• Active intensity was set to the maximum tolerable level; sham to 5%
• Ultrasound before treatment and at 1, 3 and 6 months measured subcutaneous fat and rectus abdominis thickness
• Photographs, satisfaction and therapy comfort were also assessed

The ultrasound endpoint is what makes this trial valuable.

On most procedure pages here the endpoint is a photograph score or a self-rating. Both pass through human eyes and expectation — on the PRP page, blinded doctors and participants reached opposite conclusions inside one trial.

Ultrasound does not go through that channel. Thickness does not change with expectation.

Ultrasound has its own limit, though. As the cryolipolysis page notes, the reading varies with probe pressure. So repeating measurements on the same person under the same conditions matters, and this trial did that.

The reason this is not 'strong evidence' is simple: it is one trial. This site's rule requires two or more distinct randomised trials for that badge. Deoxycholic acid had two, and even there we did not raise it, because both were manufacturer-funded.

The ultrasound results in the active group.

At one month:

• Adipose tissue thickness down 20.5% (4.8 ± 2.6 mm), P < .05
• Rectus abdominis thickness up 21.5% (2.0 ± 0.8 mm)

At three months:

• Adipose tissue down 28.3% (7.6 ± 3.7 mm)
• Rectus abdominis up 24.2% (2.3 ± 0.9 mm)

Improvements were maintained at six months, and the sham group showed no change.

Two things are worth noticing here.

First, the two measures moved in opposite directions. Fat down, muscle up. That matches the claim that the device does two things at once. This is not simply 'the abdomen looked slimmer' — what got thinner and what got thicker were captured separately.

Second, the three-month figures were larger than the one-month ones. Treatment finished within three weeks, and the result kept growing after that. It reads as tissue being cleared and muscle adapting over time.

And the conditions to attach.

How body weight was handled is not in the abstract. If participants' weight changed over six months, subcutaneous fat thickness moves with it. That the sham group did not move eases this worry somewhat, but whether weight change was equal between groups is not established.

Allocation was 48 active to 24 sham, a 2:1 split. With the sham group at half the size, 'the sham group showed no change' is correspondingly less precise.

The sham group in this trial did not simply have no device applied. The same device was applied with the intensity set to 5%.

The active group was at the maximum tolerable level.

Why it was designed that way is guessable. Electromagnetic muscle stimulation makes the abdomen contract visibly. At zero intensity a participant would know immediately they were in the sham group. Five per cent is a point where something is felt while falling far short of treatment intensity.

So the design is a reasonable choice for preserving blinding. But two things go on the record.

One, whether 5% really does nothing is not established. It differs in character from controls that can be taken as inert, such as saline (PRP) or a placebo cream. If there were even a small effect at 5%, the gap against the active group would have come out smaller than the true effect.

Two, whether participants were actually blinded is not in the abstract. The experience of the abdomen contracting hard at maximum tolerable intensity may differ considerably from 5% — especially for anyone who has had this procedure before.

That is why this item is filed as 'not established'. What is not established is whether blinding actually held.

Even so, by this site's standards a 5% sham is a far better condition than device A against device B. The problem we flag repeatedly on other procedure pages — that it cannot rule out neither working — is largely solved here.