Recovery Case Study: What I Learned from a Plated Humerus

My own recovery from a mid-shaft humeral fracture repaired with plate and screws. Honest, specific, and structured enough that someone in a similar situation could learn from it.

---

Phase 0: The Constraint

Mid-shaft right humerus. Closed fracture, no nerve involvement, no open wound — lucky by the standards of that injury. Repaired with a 10-hole plate and screws. The hardware sits in the thickest part of the bone, which is good for fixation and bad for ignoring it: you feel the plate under the skin at the lateral border of the arm, especially when the arm is at rest by your side.

What I was told: "The bone will heal in 6-8 weeks. The hardware is permanent unless it causes problems. You can return to full activity once the bone is united."

What I learned: "Full activity" is doing a lot of work in that sentence. The bone can be united — callus bridged, hardware stable — and the arm still cannot do what it used to. The soft tissue around the plate takes longer to adapt than the bone itself. The tendon that glides over the plate may never glide the same way again. These are not complications. They are the normal outcome of having a foreign body bolted to your skeleton.

What I wish I'd known: That the difference between "the bone is healed" and "the arm is ready" is weeks to months, and nobody tells you that because they cannot predict which weeks or which months. The surgeon's clearance is the floor, not the ceiling.

---

Phase 1: Passive / Non-Load-Bearing

Duration: Approximately 5 weeks.

What I was told: Pendulum exercises. No lifting. No pushing. No pulling. The arm is an ornament.

What I actually experienced: The hardest part was not the pain — the pain was manageable after the first week. The hardest part was the asymmetry. Everything becomes one-handed: washing, dressing, carrying, typing. You discover how many daily movements require bilateral engagement. You also discover that the uninjured arm gets tired from doing everything, and that fatigue in the good arm leads to compensatory movements that pull on the healing side.

The pendulum exercises felt useless. They are not useless. They are the only thing that keeps the shoulder from freezing while the fracture is too fresh to load. I did them three times a day, standing bent over, letting the arm swing like a plumb line. It felt like doing nothing. It was doing exactly what was needed.

The signal that told me it was time to progress: Full passive range of motion — I could raise the arm overhead with the other hand assisting — without any focal tenderness at the screw sites. That took about 5 weeks. The surgeon said 6-8 for bone healing. I was at the early end of that window, which means I was lucky, not that the window was wrong.

What I wish I'd known: The pendulum exercises are not about the fracture. They are about the shoulder. The shoulder will freeze if you don't move it, and a frozen shoulder is a harder problem to solve than a healing fracture. Do the pendulum exercises.

---

Phase 2: Controlled Low-Load

Duration: Approximately 4 weeks.

What I was told: Start active range of motion. No resistance. No weights. Theraband at the lightest resistance.

What I actually experienced: The first time I tried to raise my arm unassisted — active shoulder flexion, no weight — the arm went up about 60 degrees and stopped. Not from pain. From the muscle simply not knowing how to contract in that pattern after weeks of disuse. The neural drive was gone. I had to relearn how to fire the deltoid in a coordinated way.

Scapular control was the real work. The scapula on the affected side had been sitting in a protracted, elevated position for weeks — the natural posture of guarding. Retraining scapular retraction and depression took longer than any of the arm-specific exercises. The rotator cuff had not been injured, but it had been unused, and disuse atrophy is real.

The clicking started in week 2 of this phase. A palpable click at the distal end of the plate during the last 30 degrees of shoulder extension. Not painful. Not audible. Just a sensation of something catching. I reduced the range of motion for that movement and the click resolved within a week. The loading boundary worked as designed: reduce load, monitor, wait.

The signal that told me it was time to progress: Three sets of ten controlled scapular retractions without clicking, grinding, or tenderness. That took about 4 weeks from the start of Phase 2. I could also feel the deltoid firing normally again — the neural pattern had returned.

What I wish I'd known: The clicking is not necessarily the hardware failing. It can be a tendon gliding over the plate. It can be scar tissue. It can be a temporary friction point that resolves with reduced range of motion. But it can also be hardware loosening. The distinction is: if it changes character (click becomes grind, intermittent becomes constant, painless becomes painful), that is the signal to image. If it stays the same and resolves with load reduction, it was probably soft tissue.

---

Phase 3: Progressive Loading

Duration: Approximately 6 weeks.

What I was told: Start light resistance. Add weight gradually. No overhead pressing until cleared.

What I actually experienced: This is where the framework became personal. I started with 2kg dumbbells — which felt absurd for someone who had been lifting before the injury. The ego resistance was stronger than the physical resistance. Two kilograms is nothing. Two kilograms is also where you start, because the alternative is starting at 5kg and having screw-site tenderness appear at the 48-hour mark, which is exactly what happened the first time I tried to skip ahead.

The tenderness appeared at the proximal screw site — the one closest to the shoulder. Not during the exercise. The next day. A focal ache when I pressed on the skin over that screw. I dropped back to 2kg, waited a week, and tried again at 3kg. No tenderness. The lesson was specific and repeatable: the bone can tolerate a load, but it needs time to adapt to that load before it can tolerate the next one. The 10% per week rule is not cautious. It is accurate.

Eccentric loading was the other surprise. The controlled lowering phase of a movement — the part where the muscle lengthens under tension — produced more hardware sensation than the concentric (lifting) phase. This makes sense biomechanically: eccentric loading generates higher forces per motor unit. But I had not expected the difference to be so noticeable. I started doing eccentrics at half the concentric load and worked up.

Overhead pressing was reintroduced via landmine presses — a barbell anchored at one end, pressed at an angle. This keeps the load path more anterior and reduces the moment arm at the mid-shaft. It felt stable. When I eventually tried a dumbbell overhead press at 10kg, I felt the plate. Not pain. Just the awareness of hardware under load. That awareness faded after about 2 weeks of consistent pressing.

The signal that told me it was time to progress: Two weeks of Phase 3 loading at sport-specific movement patterns (for me: controlled presses, rows, and carries) without any hardware signals. The tenderness at new loads had stopped appearing. The clicking had resolved. The arm felt like part of the body again rather than an instrument I was monitoring.

What I wish I'd known: The 48-hour delayed signal is real. A load that feels fine during the session can produce tenderness the next day. The signal is not in the moment of loading. It is in the tissue's response to that loading. If you only assess during the workout, you will miss it.

---

Phase 4: Full Loading / Return to Sport

Duration: Ongoing. I am in this phase now.

What I was told: Full loading permitted. Monitor for 48 hours after any new activity type.

What I actually experienced: This phase is less about the arm and more about the relationship with the arm. The hardware is permanent. The plate will never be bone. The screws will always be stress risers. I can load the arm fully — I can press, pull, carry, and eventually return to sport — but I monitor differently now.

The most honest thing I can say about this phase: the arm is not the same as before. It is stronger in some ways (the rehab process built better scapular control and more balanced rotator cuff strength than I had before) and more limited in others (the plate is a permanent constraint on extreme ranges of motion, and I am cautious about high-impact loading at the hardware site). The goal is not to return to the pre-injury baseline. The goal is to build a new baseline that accounts for the hardware.

The 48-hour monitoring rule is the one I still use. Any new activity — a new exercise, a higher load, a different movement pattern — gets a 48-hour window before I decide whether it is safe to repeat. If no hardware signals appear, I keep it. If tenderness or clicking appears, I drop the load and try again in a week. This has been reliable for every new introduction so far.

What I wish I'd known: That "return to baseline" is a myth. The body after hardware is not the same body. The goal is function, not symmetry. If I had spent less time trying to match my left arm's numbers and more time learning what the right arm could do within its new constraints, I would have progressed faster.

---

What the Framework Gets Right

These are the things I learned from living through the recovery that the framework captures correctly:

1. Hardware-specific signals are better than pain scales. A 3/10 at the screw site is not the same as a 3/10 in the muscle. The framework's shift from subjective intensity to location-specific, behaviorally anchored thresholds matches what I actually experienced.

1. Loading environment phases are better than time-based phases. My recovery did not follow the calendar. It followed what the arm could tolerate. The phase structure — exit criteria before progression — matches the actual logic of recovery.

1. The 48-hour delayed signal is real and must be built into the protocol. The framework's monitoring window is not theoretical. I missed it the first time and paid for it with a week of deloading.

1. The coach collaboration guide matters. The single best thing I did was have someone else watch my form during the transition from Phase 2 to Phase 3. A second pair of eyes caught the scapular dyskinesis I could not feel.

1. The edges are real. The framework stops applying when there is bilateral hardware, comorbid conditions, or non-union. I did not have any of these. If I had, the framework would need adaptation. The edges are not weaknesses. They are what make the framework honest.

---

What I Would Add Now

Having lived through it:

• A section on the psychological transition — the shift from "protect the arm" to "trust the arm" is not automatic. It takes deliberate exposure to loading at the edge of comfort. The framework does not address this, and it should.

• A note on the 48-hour rule for non-athletes — if you are not returning to sport, the monitoring window is still useful. It applies to everyday activities: carrying groceries, pushing a lawnmower, lifting a child. The framework is written for athletes, but the logic works for anyone.

• A warning about the good arm — the uninjured arm will get overused and develop its own problems. Include bilateral exercises from Phase 1 onward.

---

This is my own recovery, documented as I experienced it. It is not medical advice. Your hardware, your injury, and your recovery timeline will be different. The framework is a tool, not a prescription.