Here's the thing about the human body that nobody warns you about. It can look at a perfectly ordinary meal, a sandwich, a bowl of rice, and decide to respond by making you look six months pregnant.
And for half the people that happens to, the standard answer they get is gas.
Which is wrong. And Daniel has sent us a whole prompt about exactly that, and about what he's been living with since his gallbladder came out. Here's what he wrote. He's been documenting his post-surgical health issues for a while now, and somewhere in that collection of diagnoses is one that's hard to say out loud: abdominophrenic dyssynergia. APD is the shorthand, which he and everyone else prefers.
Understandably.
What it means practically is this. Thirty minutes to an hour after a normal meal, he distends. He says he looks and feels pregnant, and he's a man. The breathing gets laboured at the same time, and that detail matters, because it points straight at the diaphragm. He also brings up the teaching point that's been circulating in gastroenterology, the landmark finding that undid decades of assumption: in a study of patients with visible distension and APD, virtually all of them had normal gaseous volumes.
That's the one that reframed everything.
Then he asks two things. First, mechanism. This all started after abdominal surgery. He'd never had mysterious bloating before. So how does a physiological stressor like surgery make your body forget how to coordinate the muscles it uses to eat? He'd always assumed nerve damage, and he's wondering whether that's actually the explanation. Second, and he flags this as the more useful question, treatment. Specifically, thoracoabdominal wall motion guided biofeedback. What does it involve, what does availability look like worldwide, is there a standardised and validated protocol clinicians can actually deliver, and can any of it be done remotely, without face to face instruction? He mentions a researcher in Barcelona. He'd be willing to travel there, but it wouldn't be his first choice. Local would be better. So.
So.
Let's start with what APD actually is, and why it breaks the standard bloating narrative.
The name tells you almost everything. Abdominophrenic: abdomen and diaphragm. Dyssynergia: muscles that should be working together, working against each other. What should happen when you eat is that the diaphragm relaxes and the abdominal wall contracts, so the contents have somewhere to go without the pressure spiking. In APD, it's inverted. The diaphragm contracts downward, the abdominal wall relaxes outward, and the whole thing pushes out.
So the belly isn't inflating because something inside it expanded.
There's nothing extra in there. The wall itself is the problem. Which is why the gas thing was such a stubborn error for so long. Bloating and distension got treated as the same word, and the assumption underneath was that visible distension means measurable excess gas. Then the CT volumetry work came along and actually measured it. Patients with APD and visible distension, normal gaseous volumes. Not slightly normal. Normal.
And that's in the Rome IV criteria now, APD as its own entity rather than a flavour of irritable bowel syndrome.
Recognised as a functional bowel disorder, distinct from IBS with bloating, which matters clinically because the treatment pathways diverge. If you're chasing gas, you're chasing the wrong thing. And chasing gas has consequences. People end up on restrictive elimination diets for years, cutting out entire food groups, because if the assumption is that something you're eating is producing the gas, then the logical move is to stop eating it. Daniel's probably been through some version of that.
He doesn't say so explicitly, but the way he frames the question suggests he's been given the dietary runaround before he got to the actual diagnosis.
Which is the standard arc. You get told it's gas, you get told it's diet, you get told it's stress, and by the time someone says abdominophrenic dyssynergia you've already lost two years and developed a complicated relationship with bread.
Which brings us to Daniel's first question. How does surgery do this.
So his instinct is nerve damage, and it's a reasonable instinct. It's just not the main story. Surgical injury to the vagus or the phrenic nerve is possible, and it does happen, but it's relatively rare, and when it happens the picture looks different. You'd expect more constant symptoms, not something that shows up thirty to sixty minutes after a meal every time.
Right, the timing is the tell. Nerve damage doesn't wait for lunch.
That meal-triggered pattern is what points you somewhere else. The leading explanation is what I'd call maladaptive learning. Viscerosomatic reflex adaptation, in the literature, but the plain version is this. Surgery is a big sensory event. There's inflammation, there's pain, there's a change in intra-abdominal pressure, and the nervous system is receiving all of that while it's also trying to keep you breathing and digesting. So it recalibrates. It finds a coordination pattern that works under those conditions, and then the conditions resolve, and the pattern doesn't.
The body doesn't forget. It learns something wrong and then never gets corrected.
That's the better way to put it, and it's why I'd push back gently on the word forget. Forgetting implies the information is gone. What actually seems to happen is that a new motor programme gets laid down and reinforced every time the person eats. Which is why it entrenches.
And the gut-brain axis piece runs alongside it.
Central sensitisation, altered autonomic regulation. Post-surgical functional disorders generally aren't structural damage stories. The signal processing changed. And there's a durability argument for that: cholecystectomy patients report persistent bloating years later in somewhere between ten and forty percent of cases, depending on the study. That's not post-operative inflammation fading slowly. That's a stable new state.
Ten to forty is a wide range.
It is, and it reflects how loosely the outcome is defined across studies. But the floor of it is ten percent, and that's a lot of people with a durable change.
Is there a clean analogy for the learning piece? Because "your nervous system learned a bad habit during surgery" sounds almost too neat.
The one I use is a limp after a leg injury. You sprain your ankle, and for a few weeks you walk with a shortened stride on that side because it hurts. The ankle heals. The limp stays, because the motor pattern got reinforced over hundreds of steps while the injury was active. Physios see this constantly. Nobody's ankle is still sprained, but the gait is still wrong, and it stays wrong until someone retrains it deliberately.
And the retraining is the hard part because the person can't feel the asymmetry.
They can't. That's the cruel bit. The pattern feels normal because it's what the body has been doing for months or years. You need an external signal to show the person what they're actually doing.
Which is a nice bridge to the treatment, but let's stay on mechanism for one more beat. How solid is it, really? I want to know how much of this is settled and how much is a story we tell because it's tidy.
Honest answer: mostly observational. There's no trial where you randomise people to gallbladder removal and watch their diaphragm coordination for two years. What you have is a consistent clinical pattern, a plausible physiological pathway, and the CT work ruling out the competing explanation. So the gas hypothesis is dead, which is a real advance, and the maladaptive learning hypothesis is the best fit for what's left. But I don't want to overstate it. The exact mechanism is still an open question.
It is, because it eliminates the thing everyone assumed for decades. When you rule out the easy answer, the harder answer gets more credible by default.
There's a practical implication buried in this that's worth pulling out. If this is a learned pattern rather than a broken nerve, then it's potentially unlearnable. Nerve damage you can't train away. A motor programme you might be able to.
That's exactly the logic that leads to the treatment, and it's why the mechanism question and the treatment question aren't really two questions. They're the same question from different ends. If the problem is wrong coordination, the solution is retraining coordination. Which is where Daniel's second question comes in.
Thoracoabdominal wall motion guided biofeedback.
It's retraining with a mirror. You put sensors on the thorax and the abdomen, and they pick up what the muscles are actually doing in real time, and that gets displayed on a screen. Then the patient breathes and activates muscles while watching the display, and the goal is to consciously produce the coordination pattern that's supposed to happen automatically.
So the person is learning to feel a thing they can't feel, by looking at it instead.
Better than they can feel it, yes. Because the internal sensation is unreliable. It's the same logic as standing on a force plate to fix a gait problem. You can't feel the asymmetry. You can measure it, though, and then you learn the corrected pattern until it becomes automatic.
What's the feedback modality? You said sensors. Is it muscle electrical activity, or movement?
Both exist. Electromyography is the classic one, surface electrodes picking up muscle activation. Motion sensors or a barostat measuring actual wall displacement is the other route. Azpiroz's group in Barcelona has used an electronic barostat alongside EMG, so they're measuring the actual movement of the abdominal wall, not just the muscle signal.
That's the name Daniel was reaching for. Barcelona.
Fernando Azpiroz at Vall d'Hebron Research Institute. His group has published the key work on this. What they showed is that with the feedback in place, patients could produce the correct pattern, and with repetition the distension and the symptoms improved. It's a small literature, and the protocols aren't fully standardised across centres, but the effect is there.
Give me the honest evidentiary read. How many patients are we talking about, and how would you characterise the strength of it.
Small trials. On the order of tens of patients per study, not hundreds. Which means I'd want to see replication before anyone calls it settled, and I'd want to know what the comparison arm was, and who funded it. But the direction is consistent, and the mechanism is coherent, and there isn't a competing treatment with a better record. That combination is more than most things in this space have.
What does a session actually look like, though? Because I'm imagining someone wired up like a science experiment, and I suspect the reality is more mundane.
It's more mundane. You're sitting in a chair, typically reclined, sensors taped at a couple of points on the chest and abdomen. There's a screen in front of you showing two traces, or a shape that changes as you move. The therapist asks you to breathe in a particular way, or to gently push your belly out or draw it in, and you watch what the trace does. The first few sessions, most people are wrong in ways they can't predict. They think they're doing the thing and the screen says otherwise.
And then it clicks.
It clicks for some people in a session or two, and for others it takes a dozen. And the clicking isn't the endpoint. The endpoint is doing it without the screen, at home, at meals, until it's the default again. That's the transfer phase, and it's the part that's least standardised across protocols.
The word "standardised" is doing a lot in Daniel's question, though. He's asking whether clinicians around the world can deliver a standardised and validated protocol. And the answer seems to be mostly no.
Mostly no is right. There's no single published protocol that everyone follows. Different centres use different feedback modalities, different session counts, different treatment durations. So the technique exists and it works in the hands of people who do it, and it isn't yet a thing you can walk into any motility clinic and receive.
So where is it actually available.
Specialised motility centres, and not many of them. Barcelona is the obvious one given the publication record. There are centres in the United States and the United Kingdom that do biofeedback for functional GI conditions, though APD specifically is a narrower subset. The training pathway for clinicians isn't standardised either, which is the deeper problem. Even if a hospital wanted to offer it, there isn't a clear curriculum to follow.
That's the part that frustrates me about this whole area. The technique is described in the literature, the effect size is real, and yet there's no pathway from paper to clinic.
It's not unique to APD. It's a recurring pattern in functional GI medicine. The interventions tend to be low-tech and effective, and they don't attract the commercial infrastructure that would scale them. A drug gets a sales force. A biofeedback protocol gets a paper and a footnote.
Then there's the remote question, which is the one Daniel actually cares about. He'd travel. He'd rather not. Can this be done without being in the room with someone.
No validated remote protocol for APD specifically. Nothing published, nothing trialled. But the precedent exists in adjacent territory, and it's good precedent. Pelvic floor biofeedback for faecal incontinence has been delivered with home devices and telehealth coaching, and it's been shown effective in randomised trials. That's the same fundamental structure. Sensors, a display, a trained person interpreting and coaching. There's no obvious reason the model wouldn't transfer.
So what would it actually take.
A clinician willing to adapt it, a way to get the sensors to the patient, and some form of video coaching to interpret the data. All three of those exist as commodities. What doesn't exist is a validated protocol saying do it this way, this many sessions, with these endpoints.
Which means the honest answer to Daniel is: maybe, but you'd be the experiment.
You'd be doing it off-label, effectively. And I'd say that carefully rather than dismissively, because this is how a lot of these things get figured out. Pelvic floor remote training went from off-label to trial-supported over about a decade. Someone has to go first.
That's if he wants to go first.
Which he's been clear he'd rather not. He'd rather have it locally. Remote is a second choice, and travelling is a third.
So the state of play is: the treatment works, the mechanism makes sense, the evidence is small but real, and the delivery is stuck in a handful of centres on two or three continents. And the person who needs it lives in Jerusalem.
That's the accurate summary and I don't love it.
Here's what I keep circling. The mismatch between how specific Daniel's description is and how general the medical response is. He can tell you the exact timing, thirty to sixty minutes. He can tell you the breathing is involved. He can tell you it started after surgery and never happened before. That's a precise clinical picture. And the answer he gets is essentially stop eating the thing that triggers it.
Which is the answer that treats him as the problem rather than the physiology as the problem.
And he's been photographing his own abdomen to document it, which tells you how seriously he's taking the tracking.
That's actually clinically useful. Visual documentation of distension episodes is exactly the kind of data a motility specialist wants, and most patients don't have it.
Hilbert: The barostat sensor heads they used in those Barcelona sessions, you couldn't sterilise them between patients the way the manual said, because the seal on the pressure port degraded after about forty autoclave cycles. So the tech would swap the head and recalibrate, and the calibration took twenty minutes, and the clinic scheduled thirty minute slots.
Which means almost half the appointment was calibration.
Hilbert: They learned to do it before the first patient and after the last one. Two heads, rotating.
So the throughput was fixed by the hardware, not the clinician.
Hilbert: Everything was fixed by the hardware. I serviced those units out of a van, late eighties, the ones with the chart recorder and the strip paper, and the paper feed was the thing that failed, always, because the pinch roller would glaze. You'd pull it, rough it up with emery cloth, put it back. Five minutes if you knew. An engineer from the manufacturer would take the whole transport assembly out and charge for it.
Did the clinics ever get the remote version, in the end?
Hilbert: The home units came later and they were simpler, surface EMG, no barostat, no strip paper. But the patients who needed them worst were the ones furthest from a clinic that stocked the sensors, and that hasn't changed. I had one, a woman, she'd been told for eleven years it was gas, and the biofeedback was the first thing that helped her. Four hours each way to the clinic. She did it for two years.
Four hours each way. That's the whole availability question in one number.
And it's not a new problem. That's the part that bothers me. The technology to deliver a version of this at home has existed for decades, and the reason it isn't standard isn't technical. It's that nobody's built the pathway.
Hilbert: The clinic tried to set up a satellite location closer to her. Licensing for the equipment was per-site, and the fee didn't scale down for a small site, so it never happened.
So the thing that stopped it was a licensing structure.
Hilbert: It's usually a licensing structure or a lease. Either way it's a contract, not a machine.
Which means the barrier is commercial and organisational, which is actually encouraging in a bleak sort of way. Commercial barriers move when someone decides they should. Technical ones don't.
That's the note I'd take from Hilbert's story. The four hour drive isn't a science problem. It's a deployment problem.
And deployment problems are solvable in a way that mechanism problems aren't. We already have the mechanism answer, roughly. We don't have the deployment answer.
Which leaves Daniel where he started. He has a precise diagnosis, he knows what treatment he wants, he knows where it lives, and he's now weighing a flight to Barcelona against getting a local clinician to improvise.
If it were me, I'd ask a local motility specialist directly whether they'd be willing to run the protocol with remote guidance from a centre that does it. Worst case, they say no. Best case, you're the first patient in the country and they write it up.
Would you want to be the first patient in the country.
I'd want to know who was supervising before I answered that.
Here's where I land on the whole thing. If the mechanism is maladaptive learning rather than nerve damage, then this is a reversible condition in principle. The muscle coordination was learned wrong, so it can be learned right. That's the entire premise of the biofeedback, and the small trials support it.
The reversibility is the important word. It's the difference between managing a symptom and correcting a pattern.
Which makes the access problem harder to accept rather than easier. If this were a drug that only worked in thirty percent of patients with side effects, fine, the hesitancy makes sense. But this is a training protocol with a sensible mechanism and no pharmacology, and it's available in a handful of cities.
The infrastructure lags the science. That's the pattern, and it's not specific to APD. It's the same story in pelvic floor rehab, in gait retraining, in a dozen low-tech interventions that work and don't have a sales force behind them.
So the honest forward look is: as remote monitoring and telehealth keep expanding, does APD biofeedback get pulled along with them, or does it stay a niche that only people who can fly to Barcelona ever receive?
I'd bet on getting pulled along, eventually, on the strength of the pelvic floor precedent.
Which is the least satisfying place to end an episode. The science exists, the treatment works, and the distance between here and it is measured in licensing contracts and clinic schedules.
That's medicine. The evidence and the access are two different projects and they run on different clocks.
Thanks as always to Hilbert Flumingtop, who produces this show and who has apparently serviced a barostat with emery cloth.
This has been My Weird Prompts.
If you got something out of this one, a review and a subscribe help other people find the show. We'll be back soon.
