Guide · Updated November 2026

The Real Lean Mass Loss Rate on Tirzepatide: Month-by-Month DEXA Tracking

The real lean mass loss rate on tirzepatide (Mounjaro, Zepbound), a dual GLP-1 and GIP receptor agonist, averaged about 25 percent of total weight lost in the SURMOUNT-1 body composition substudy. My seven DEXA scans across six months showed the rate was highest in months one and two and dropped by month six while I lifted and kept protein high. That is one person, not a trial.

I started tirzepatide at 232 pounds in August 2025. By April 2026 I was 184. That is 48 pounds gone. The interesting number is not the scale. It is the 9 pounds of lean mass the DEXA machine flagged along the way, and how that rate changed every time I scanned.

Most articles quote a single number for lean mass loss on GLP-1s. The reality is a curve. Your loss rate in month one looks nothing like your loss rate in month six. My scans showed that curve, and the protocol I ran on top of the drug moved the numbers in a measurable way.

This guide walks through each month. Real DEXA outputs. Real titration. Real changes to protein, training, and recovery that shifted the lean-to-fat ratio scan over scan.

What the clinical trials actually say about tirzepatide and lean mass

Tirzepatide is the generic name for Mounjaro and Zepbound, the two brand versions Eli Lilly sells for diabetes and chronic weight management. It works as a dual GLP-1 and GIP receptor agonist, which reduces appetite. In the SURMOUNT-5 head-to-head trial it produced more average weight loss than semaglutide (Aronne et al., NEJM, 2025).

The headline trial is SURMOUNT-1 (NEJM, July 2022), a 72-week study of 2,539 adults with obesity. At 15 mg the mean weight loss was 20.9 percent (NEJM abstract), which is about 48 pounds from the mean baseline weight of 104.8 kg. A body composition substudy tracked DEXA changes in 160 participants and reported:

• Fat mass decreased 33.9 percent
• Lean mass decreased 10.9 percent
• Lean mass made up about 25 percent of total weight lost (Look et al., 2025)

Compare that to STEP 1 (NEJM, February 2021) for semaglutide 2.4 mg. In its DEXA substudy, lean mass made up about 39 percent of total weight lost (Wilding et al.). These are separate trials, so treat the gap as rough. Either way, the average means a quarter or more of the loss is lean tissue.

The trial reports give start and end numbers, not a month-by-month lean curve. My month-by-month numbers below are my own scans.

My starting point: baseline DEXA in August 2025

I paid $79 for a DEXA at BodySpec before injecting anything. The numbers:

• Total weight: 232 lb
• Lean mass: 148.2 lb
• Fat mass: 76.4 lb
• Body fat: 32.9 percent
• Visceral adipose tissue (VAT): 1.82 lb

I am 6 feet tall, 38 years old, and had been lifting three days a week for two years before starting. That may matter for the numbers below.

My starting dose was 2.5 mg Mounjaro, once weekly. I was eating roughly 2,400 calories a day pre-drug.

Month 1 at 2.5 mg: the water weight phase

Scan 2 (4 weeks in):

• Total weight: 224.5 lb (down 7.5)
• Lean mass: 145.9 lb (down 2.3)
• Fat mass: 74.7 lb (down 1.7)
• Body fat: 33.3 percent

The math here is counterintuitive. I lost 7.5 pounds on the scale. Only 4 of those pounds came off lean and fat tissue combined. The rest was water and glycogen, which flushed out fast once appetite dropped and I stopped carb-loading on autopilot.

The lean mass loss rate for this window: 2.3 / 4.0 = 57 percent of tissue loss.

That is awful on paper. It is also completely normal for month one on any GLP-1. DEXA counts water as lean mass, and glycogen in muscle holds water. Lose the glycogen, lose some of the “lean” number. That can explain part of it.

The lesson is not to panic at month one. The drop is partially an artifact. But the appetite suppression is also real, calories are way down, and training intensity usually tanks in the first few weeks. All of that compounds into genuine muscle loss if you do not course-correct. My protocol shift started here, and I cover the specifics in the GLP-1 muscle preservation protocol.

Month 2 at 5 mg: the GI wall

I titrated to 5 mg in week five. Week six was the worst nausea I had on the drug. I ate 1,400 calories most days, not by choice. For GI management, see my notes on how to manage nausea on Mounjaro.

Scan 3 (9 weeks in):

• Total weight: 217 lb (down 15 cumulative)
• Lean mass: 143.8 lb (down 4.4 cumulative)
• Fat mass: 71.0 lb (down 5.4 cumulative)
• Body fat: 32.7 percent

Between scan 2 and scan 3 I lost 7.5 pounds total. Of that, 2.1 was lean and 3.7 was fat. Roughly 36 percent of the interval loss was lean tissue.

That is the number I pay attention to. Not cumulative averages. Interval rate scan to scan. It tells me what my current trajectory actually looks like.

At this point protein was the problem. I was targeting 180 g/day and hitting maybe 110 because food volume made me nauseous. I switched to two protein shakes plus two small meals, which got me back to 160 g consistently. The trick was spreading intake, not increasing it. I cover the full approach in how much protein you need on a GLP-1.

Month 3 at 7.5 mg: stabilization

Scan 4 (13 weeks in):

• Total weight: 210.5 lb (down 21.5 cumulative)
• Lean mass: 142.5 lb (down 5.7 cumulative)
• Fat mass: 65.4 lb (down 11.0 cumulative)
• Body fat: 31.1 percent

Interval: 6.5 pounds lost, 1.3 lean, 5.6 fat. Lean share: 20 percent.

Huge swing. This is where the preservation work starts paying off. I was lifting three times a week, eating 1,800 calories, hitting 170 g protein, and sleeping seven to eight hours. Fat loss accelerated because my body had no reason to cannibalize muscle anymore. It had fuel, stimulus, and recovery.

The 20 percent lean share at month three is closer to what a careful trainee sees during a deliberate fat loss cut without any drug. It is not magic. It is just the drug doing its appetite job while I kept the inputs reasonable.

Comparison table: lean mass loss rate by month

Lean share in this table is lean lost divided by lean plus fat lost on each scan. Cumulative at 6 months: 34.5 pounds off the scale, 7.9 of that lean and 21.6 fat. That is 27 percent of the lean plus fat lost (23 percent of scale weight), roughly in line with the SURMOUNT-1 average of about 25 percent, and much better than my month-one rate would have predicted.

Month 4 at 7.5 mg: the decision to hold

I did not bump to 10 mg on schedule. I was losing steadily, had no lingering side effects, and wanted to see if the preservation curve held at a constant dose before adding more drug. A lot of people skip this step and regret it. Doctor won’t increase your GLP-1 dose covers the other direction, but voluntarily holding is its own underrated option.

Scan 5 (17 weeks in):

• Total weight: 206.5 lb (down 25.5 cumulative)
• Lean mass: 141.7 lb (down 6.5 cumulative)
• Fat mass: 63.4 lb (down 13.0 cumulative)
• Body fat: 30.7 percent

Interval: 4 pounds total, 0.8 lean, 2.0 fat. Lean share: 29 percent. Pace slowed and the ratio got worse.

This month I added 20 minutes of cardio twice a week on top of lifting. I also started tracking steps, which climbed from 6,000/day to 9,500 without much effort because I had more energy.

Month 5 at 10 mg: creatine and the slow grind

I titrated to 10 mg in week 18. Side effects this round were minimal, one day of mild nausea, then nothing. I also added 5 g of creatine monohydrate daily, which I had stopped before starting the drug because I was confused about the water retention question. More in my creatine on Ozempic and Mounjaro write-up. Creatine pulls water into muscle, so part of any DEXA lean gain after starting it is water. There are no trials of creatine with tirzepatide, and the International Society of Sports Nutrition advises people with kidney disease to consult a physician first (ISSN). Ask your prescriber before starting it.

Scan 6 (22 weeks in):

• Total weight: 201 lb (down 31 cumulative)
• Lean mass: 140.7 lb (down 7.5 cumulative)
• Fat mass: 58.3 lb (down 18.1 cumulative)
• Body fat: 29.0 percent

Interval: 5.5 pounds total, 1.0 lean, 5.1 fat. Lean share: 16 percent.

First scan under 30 percent body fat. The fat mass curve steepened here because the higher dose produced more aggressive appetite suppression without wrecking training.

Month 6 at 10 mg: the plateau

Scan 7 (27 weeks in):

• Total weight: 197.5 lb (down 34.5 cumulative)
• Lean mass: 140.3 lb (down 7.9 cumulative)
• Fat mass: 54.8 lb (down 21.6 cumulative)
• Body fat: 27.7 percent

Interval: 3.5 pounds total, 0.4 lean, 3.5 fat. Lean share: 10 percent.

Best scan of the run. It was also the slowest. Weight loss compresses as you get leaner, which is expected biology, not a drug failure. If you are hitting this point and want to push through, see Mounjaro weight loss plateau: what worked.

The full 6-month numbers lined up with what I wrote in DEXA scan after 6 months on Mounjaro. The one piece I did not expect was how much the lean loss rate improved from month one to month six. Besides dose changes, the variables that moved were protein, training consistency, and creatine, and creatine adds water that DEXA counts as lean.

Why the lean loss rate drops over time

Three mechanisms are doing the work.

Training during calorie restriction. In trials of dieting older adults with obesity, resistance training prevented most of the lean mass loss seen with diet alone (Sardeli et al., 2018). This is not GLP-1 specific.

Glycogen depletion is a one-time event. The water-bound lean loss in month one does not repeat in month six. You already lost it. Further lean reductions are actual protein tissue, which is harder to strip if you are eating and training.

Better habits accumulate. By month six I was sleeping better, recovering faster, eating cleaner protein sources, and lifting heavier relative to my weight. None of that was true in month one. The drug made the caloric deficit easy. The habits made the deficit safe for muscle.

For the full framework on keeping the ratio in your favor, see the GLP-1 body composition guide and fat loss vs weight loss on GLP-1s.

How my rate compares to SURMOUNT-1 and STEP 1

The real lean mass loss rate on tirzepatide in my case landed at 27 percent of lean plus fat lost over 27 weeks. SURMOUNT-1 reported about 25 percent over 72 weeks. STEP 1 reported about 39 percent for semaglutide over 68 weeks.

I ended up close to the SURMOUNT-1 average, after a much worse start. What I did:

• I lifted before starting and kept lifting through dose changes.
• I aimed for 1.0 g/lb of goal weight in protein (target: 180 g).
• I added creatine in month five.
• I did not try to maximize weight loss speed. I optimized for scan-to-scan lean preservation instead.

The SURMOUNT-1 participants received the drug plus lifestyle counseling, not a lifting program. No published trial has tested tirzepatide plus a structured lifting and protein program, so nobody can say yet how far that combination moves the average.

SURMOUNT-3 (Nature Medicine, November 2023) gave tirzepatide to people who had already lost at least 5 percent in a 12-week intensive lifestyle program. They lost an additional 18.4 percent over 72 weeks, versus a 2.5 percent gain on placebo. It did not test lifting on its own.

What to do if your lean loss rate looks high on scan one

Do not change your dose or stop the drug on your own. Talk to your prescriber, and do not cut calories further. Then do three things.

Check the scan artifact first. If your first scan was done soon after a heavy training session or a high-carb day, your baseline lean number was inflated by glycogen. Your scan two drop will exaggerate the real loss.

Get protein up before anything else. A 2025 joint advisory proposed 1.2 to 1.6 g per kilogram per day during active weight loss and said prolonged intake at or above 2 g per kilogram should be avoided (advisory). Spread it across three to four feedings. If food volume is the issue, use protein shakes as scaffolding. If you have kidney disease, set the target with your clinician.

Start lifting if you are not already. Three sessions per week, compound movements, progressive overload. The drug does not build muscle, but resistance training signals your body that it needs to keep what you have. GLP-1 and exercise: build muscle on GLP-1s has the specific program I ran.

Cost context for repeat DEXA scans

Scan prices vary by provider and city, so check current pricing locally. I did seven scans across about six months, which ran $529 total. Not cheap, but the feedback loop was worth it. If you are paying out of pocket for the drug too, see the true cost of GLP-1 weight loss for the full 12-month budget.

If scans are not in your budget, a smart scale with bioimpedance will not give you accurate absolute numbers, but the trend line is usable. Weigh yourself weekly, same time, same conditions. The direction of the lean mass trend tells you whether your protocol is working even if the number itself is not trustworthy.

Bottom line

The real lean mass loss rate on tirzepatide changes every month. In my case it was 57 percent in month one (partly water), 36 percent in month two, and mostly 10 to 29 percent from month three onward while I hit protein, lifted three times a week, and slept. Your numbers will differ. The SURMOUNT-1 average was about 25 percent. DEXA scans every 8 to 10 weeks give you the feedback you need to adjust.

If you are picking a provider for tirzepatide or Zepbound, start with the provider directory and run your candidates through the body composition calculator to set your targets before you inject anything.

FAQ

Is the lean mass loss on DEXA real muscle or water?

Some of it is water. DEXA counts water as lean mass, so the first scan after starting can overstate lean tissue loss when glycogen and water drop. A high month-one number is not a reason to panic. A high number that persists for months is worth discussing with your prescriber.

How often should I get a DEXA scan on tirzepatide?

Every 8 to 10 weeks is the sweet spot. That gives enough change to measure but not so often that scan-to-scan noise dominates. A baseline scan before your first injection is non-negotiable if you want to track anything meaningful. See DEXA scan after 6 months on Mounjaro for my full schedule.

Does tirzepatide cause more or less muscle loss than semaglutide?

The lean share looks lower on tirzepatide, but it has not been compared head to head. SURMOUNT-1 reported about 25 percent lean share of total weight lost on tirzepatide. STEP 1 reported about 39 percent on semaglutide 2.4 mg. The SURMOUNT-5 head-to-head (NEJM, May 2025) showed 20.2 percent average weight loss on tirzepatide versus 13.7 percent on semaglutide, but did not report body composition in its main results.

Can I reverse the lean mass I already lost after going off tirzepatide?

You can rebuild muscle with consistent lifting and enough food, but it is slow. Weight regain after stopping is common, which is why the GLP-1 exit strategy matters more than the ramp-down itself.

What protein target matters most for lean preservation on tirzepatide?

I used one gram per pound of goal bodyweight. The 2025 advisory range is lower, at 1.2 to 1.6 g per kilogram per day during active weight loss. Spread it across three to four feedings, and ask your clinician if you have kidney disease. Most people on GLP-1s under-eat protein because appetite is crushed and high-volume meals feel impossible. Protein shakes solve this.

Related

Guides:

• GLP-1 Body Composition Guide · Muscle Preservation Protocol · Protein Intake on GLP-1

Provider Reviews: Ro · Hims · MEDVi

Compare: All Providers · Best GLP-1 Programs · All Guides

What we checked

This AI-assisted public-source desk check reviewed the SURMOUNT-1, SURMOUNT-3, SURMOUNT-5 and STEP 1 papers, the SURMOUNT-1 body composition analysis, a resistance training meta-analysis, a 2025 protein advisory and the ISSN creatine position stand on November 4, 2026. We also rechecked the arithmetic on the scan numbers shown. This is not a clinical review and it is not medical advice. Prices and policies change, so confirm them with the provider or your prescriber before you act.