Kybella is deoxycholic acid injection. It was approved under NDA 206333 on 29 April 2015 — approved, not cleared, because it is a drug rather than a device. That matters for this question, because an approved drug carries a dosing section, and the dosing section answers it.
The labelled numbers, all of them:
Dose: 2 mg per square centimetre of treatment area: The dose is tied to area, not to a fixed vial count.
0.2 mL per injection, placed on a 1 cm grid.
Maximum 50 injections and 10 mL — 100 mg — in a single session.
Up to 6 single treatments.
No less than 1 month between treatments.
A 30 gauge, 0.5 inch needle, into the fat in front of the platysma muscle.
One vial holds 2 mL, or 20 mg. So a maximum session is five vials, and a full six-session course at maximum dosing would be 600 mg. Almost nobody gets that.
The label does not say how many sessions you personally will need: It sets a ceiling and a spacing rule. What fills the gap between the ceiling and the answer is the trial data.
The two pivotal trials, REFINE-1 and REFINE-2, randomised 1,022 adults to deoxycholic acid or placebo and allowed up to six sessions about a month apart. The people in them had moderate or severe fullness under the chin by both clinician and patient rating, a mean age of 48.7 and a mean BMI of 29.3. People with excessive loose skin under the chin were excluded, as was anyone who had already had liposuction or another fat-reducing treatment there.
Here is what the dosing looked like in practice.
The first session was the biggest: Subjects received an average of 6.4 mL at session one — a bit over three vials — and those who went all the way to six averaged 4.4 mL at the last one. Mean volume per session was 5.4 mL, and the median number of injections per session was 27, not the permitted 50.
Across a whole course, the average total was 25.3 mL — about 253 mg out of a possible 600 mg: So the typical trial patient used well under half the maximum the label allows.
The volume needed fell at every successive session: The trial authors read that decline as direct evidence that fat was progressively being removed: there was less left to inject each time.
The median number of sessions to a composite response — one grade of improvement on both the clinician scale and the patient scale — was 3: The median to a response on either scale alone was 2.
Broken down by session:
**Clinician-rated 1-grade improvement: 52.2 per cent of treated patients after 2 sessions, 71.5 per cent after 4.**
**Patient-rated 1-grade improvement: 47.3 per cent after 2, 74.1 per cent after 4.**
The REFINE-1 paper puts it the same way: 55 per cent and 75 per cent of treated subjects showed a 1-grade clinician-assessed improvement after two and four treatments.
In an independent private practice, the pattern was shorter still. A prospective series of 100 consecutive patients across 195 sessions found 41 people had a single session, 36 had two, 14 had three, 6 had four, 2 had five and 1 had six. The author's conclusion was that two or more sessions were generally needed to reach the aesthetic goal. That author receives support from the manufacturer, which is worth knowing when reading the conclusion.
Three things show up in the data.
How much fat is there: The dose is area-adjusted, so a larger, deeper pad takes more injections per session and generally more sessions.
Whether you and your clinician both see the change: The trials required agreement between two raters for a "response". Patients and clinicians did not always agree, and the single-rater response came a session earlier than the joint one.
When you decide you are done: 41.1 per cent of treated patients received fewer than six sessions. 19.1 per cent stopped early specifically because they were satisfied or had no submental fat left to treat, against 3.9 per cent on placebo. Of those 98 people, 67.3 per cent had stopped by the fourth session.
The mirror image of that is also true: 59 per cent of trial subjects went through all six sessions the protocol allowed. The trial design encouraged continuing. That number is not a statement of clinical need.
It is worth setting the session count against the size of the result, because the two are usually quoted separately.
After up to six sessions, 66 to 70 per cent of treated subjects achieved a 1-grade improvement on both rating scales, against 18.6 to 22.2 per cent on placebo. But only 13.4 to 18.6 per cent achieved a 2-grade improvement on both scales — the stricter measure.
The only objective measurement in the whole programme was MRI, in a subset of 449 people, and 43 per cent of treated subjects achieved at least a 10 per cent reduction in submental fat volume, against 5 per cent on placebo. So after a full course, fewer than half of treated patients reached even a 10 per cent measured volume reduction.
A rating scale is not a measurement: The headline 66 to 70 per cent figure is a one-grade shift on a five-point scale, judged by eye. Adding sessions raises the chance of that shift; it has not been shown to convert a small change into a large one.
The label says no less than one month. The trials used 28 days plus or minus 5. There is no labelled maximum interval, and no published trial has tested whether spacing sessions further apart changes the result. A real-world series treating an off-label area averaged 108 days between sessions with results the author judged good, but that is a different site and a different question.
The practical reason for the minimum is that the swelling and firmness from one session need to settle before anyone can judge how much fat is left. Injection-site reactions in the pooled trial data had a median duration of about 10 to 11 days for swelling and 28 days for firmness, and numbness ran to a median of 43 days. Assessing a chin at two weeks tells you about the swelling, not the fat.
This is one of the better-documented findings and it changes how people plan a course.
Overall side-effect incidence was highest at session 1, affecting 94.4 per cent of treated patients, and fell at every session after, to 46.4 per cent at session 6: The pooled analysis states plainly that the incidence and the severity of swelling, pain and bruising declined after the first session. 80.9 per cent of all events were mild.
Time off follows the same shape. In a 12-month open-label study of 165 people, 13.3 per cent missed work in the seven days after the first treatment and 33.9 per cent missed social activities. After later treatments those figures fell to 2.4 to 6.0 per cent and 10.0 to 15.7 per cent.
7.4 per cent of treated patients in the pooled trials stopped treatment because of a side effect, against 1.2 per cent on placebo.
Say if you have ever had trouble swallowing. People with current or previous swallowing difficulty were excluded from every trial, and the label says to avoid the drug in them. The 2 per cent dysphagia figure comes from a population that had already been screened this way.
Agree in advance who assesses whether another session is needed, and at what point. The label ceiling is six; the trial median was three.
Bleeding risk, the screening the label asks for before a first injection, and the single contraindication of infection at the injection site are set out on the Kybella recovery page.
Difficulty swallowing or difficulty breathing: In the trials, difficulty swallowing happened alongside swelling and firmness of the area, and resolved on its own in a median of 3 days with a range of 1 to 81 days. It still needs to be assessed urgently, not at the next appointment.
Fever with spreading redness, warmth or pus from the treated area. Some injection-site infections have involved cellulitis and abscess needing intravenous antibiotics and drainage.
Your smile becomes uneven, or one side of your lower lip does not move properly. This is marginal mandibular nerve paresis, reported in 4 per cent of treated subjects in the trials and about 1 per cent of sessions. Every case in the trials resolved on its own, with a median of 44 days and a range of 1 to 298 days.
An open sore, drainage, or a darkening patch appears at an injection site.
Swelling, pain or firmness is getting worse after the first few days instead of settling.
Hair thins or disappears at an injection site. This is a labelled reaction, it occurred in 0.4 per cent of treated trial subjects, and the label says its duration varies and it may persist.
Numbness is still present well beyond six weeks, or is spreading.
You are being encouraged toward a fifth or sixth session and are not sure it is adding anything. The trial median was three.