Now the sedating effect of the painkiller (or, well, probably mostly the anti-nausea side of that combination) is hitting him. Also Kalorm is really dizzy (though, slightly miraculously, still not especially nauseated). He's...maybe just going to lie here for a minute not moving. 

Yeahhhhh his exhaled CO2 levels are well below the floor of the normal range. Unsurprising, since they were pretty much over-ventilating him on purpose. Halthis wipes Kalorm's eyes and then waits for him to indicate that he's ready for anything else. 

Why does Kalorm's body see the need to respond to coughing by creating so much saliva. What exactly does it think he's going to do with it. 

He eventually remembers how to use his hands enough to attempt to communicate via gesture that he would appreciate if Halthis could Deal With That. 

Of course! While minimizing unpleasantness, even! Almost without thinking about it, she swaps in a thinner, softer, more flexible and rounded tip on the suction; it often won't cut it, but she did proper mouth care like five minutes ago, his saliva won't have had time to dry out. 

She watches his vital signs. 

O2 saturation still at 100%! 

Sheeee will test out easing the ventilator O2 concentration down to 35%. If she has to go back up during the minimum-settings trial, well, that's still only back to where it was this morning. And it really does look like something, most likely a combination of factors, improved quite a lot over the last four hours.

Kalorm looks awfully sleepy after the back-to-back bolus painkiller doses. She'll wait.  

Five minutes or so later, Kalorm feels a more normal degree of dizzy (his CO2 is back up, and his blood pH correspondingly back down toward normal) and a more normal level of alert. And getting impatient. He waves to try to get Halthis' attention and then in the direction of the ventilator. 

"Ready to go? Great. Let's give this a try." 

Halthis is not super expecting Kalorm to actually be ready to come off the ventilator at the end of the hour. That would be a pretty sudden large improvement. But she's set up everything she can to give him the best chance of making it to an hour. (Which actually makes it unrepresentative of how he would do in less ideal circumstances, of course.) He's propped up at 45 degrees, lying on his better side, on the best balance she can wrangle of pain control versus painkiller sedation, electrolytes optimized, as well-rested as he's realistically going to get...

Time to see how this goes. 

Uneventfully! At least at first! 

Kalorm can still tell that he's now doing most of the work of moving air in and out himself. He can still tell that something is pretty wrong with his lungs – he's resting comfortably, and in normal times, the effort required to breathe in that situation shouldn't even register. It's definitely registering. 

But it's a level of exertion that feels pretty sustainable? The first time, when Merrin had him lie flat, was really only comparable to healthy Kalorm's experience of a flat-out sprint; Kalorm is in excellent shape, and even in peak condition he can't sprint at literally top speed for more than a few minutes before his body is DONE. Earlier today, the trials felt more comparable to the effort of running moderately fast. Not fun to keep up for more than ten or fifteen minutes, but doable for longer. 

This...is not quite down to the effort level of a brisk walk. (In normal times, obviously, right now even a slow walk is literally impossible.) It's maybe like jogging. But Kalorm can maintain a jogging pace for a while before the effort is even that salient.

Halthis watches. His O2 sats fall to 95% a few minutes in, but stabilize there; she doesn't have to increase the oxygen concentration he's breathing. He's managing pretty reasonable breath volumes, if not as impressive as what he could do before with inspiratory pressure support. She can tell that he's using some accessory muscles, but his facial expression isn't one of someone who is having to particularly concentrate or exert willpower for a challenging physical task.

...She goes back to refreshing and reading through prediction market updates, watching him in her peripheral vision and with the vital signs up in a corner of her screen.

She lets him know when it's been fifteen minutes. "All good so far?" 

Everything is fine. Possibly everything is too fine, because Kalorm's current top complaint is that he's SO BORED. Breathing less-assisted is juuuuust enough effort and distraction that he can't actually fall asleep (also, he seems to be genuinely less drowsy?) but it's far from enough to occupy his attention. 

(He also noticed that he's having an easier time understanding what Halthis is saying to him lately? Though he can't tell if he's thinking better or if she's just run through most of the absurdly complicated explanations and questions she had and is now only needing to communicate straightforward things.) 

Kalorm is not sure how to convey this in gestures! Maybe the picture board has something. He can't remember what-all Khemeth put on it. It was a lot of things. At the very least, maybe the picture board will have a way to request Khemeth, who will take one look at him and instantly detect that he's bored out of his mind. (It helps that this is not, like, exactly a rare or weird issue for hospitalized patients to have.) 

When Halthis interrupts, Kalorm nods and makes an "okay" hand gesture and then points toward the box of smaller picture boards. 

They're each labelled by category, and Merrin set it up so that theoretically one could get the right one in two or at most three rounds of "X or Y?" questions. 

Top level is "question versus care request". For the first, the next division in the tree is between information on past or present events/test results/etc, versus future information (future information then breaks down into "prediction market updates versus scheduled care and planned treatments". For the second, the division is "communicate a symptom or complaint for the medical team to troubleshoot" vs "directly request something listed as a standard care item"; on the last one, since there are a LOT of items that Khemeth and Merrin thought were worth listing, there's another level of division between medical intervention (drugs, suctioning, etc) versus comfort-focused basic care.

It's not a perfect system, and some of the category boundaries are weirdly drawn, but it lets them cram in around fifty "vocabulary words" without Kalorm ever having to visually parse more than ten on the same page.

Dath ilanis being the way they are, "I am bored out of my mind and need something entertaining and also within my basic capabilities" is definitely an item on the comfort-focused basic care board. 

Halthis brings over the box. Fairly quickly ascertains that it's a care request, and from there, basic comfort not a medical treatment. 

She shows Kalorm the board. 

...It's more obvious now that he's still really impaired, because aaaaughh. That is still too many pictures. Put some of the pictures back.

He can, somewhat effortfully, manage to look at one of them at a time and parse what it is. Some are harder than others; it takes him, like, twenty seconds to recognize the black and white vector graphic image of a bedpan. Most of them turn out to be things that he's in the interim figured out how to convey just fine via empathic mime, at least with attentive medtechs who are doing a lot of the guesswork on their end. Although it had honestly not occurred to him that he could request lip balm. His lips are really dry. He's worried that he won't remember to find the thing he actually wanted if he stops to point at that one, though. 

There are images for "listen to audio" and "watch visual media" (the image is of a screen; the hospital does have a basic VR-goggles setup for patients, but it's almost certainly going to be way too overwhelming for Kalorm right now). There's also an image that is maybe probably supposed to be a puzzle game. 

Kalorm likes puzzle games normally - well, as much as he likes any games, he really prefers things that aren't fake - but even thinking about the concept right now makes his head hurt, and his hands aren't super working anyway. He would also prefer not to need to keep his eyes open. 

He manages to point at the headphones graphic for "listen to audio". (Halthis is holding the board steady for him, comfortably within reach.) And then remembers to point at the lip balm one. What a good idea. 

(If Merrin were awake and present, she would feel terrible that she never thought to offer that, and then defensive about how many she WOULD have if it wasn't during a day-night-flip shift when the first shift had been ALMOST SEVENTEEN HOURS and she was VERY FRIED and also, like, most of her training for the last two and half years has been for weird emergencies that did not involve prioritizing lip balm.) 

...Huh. The fact that he's apparently with it enough to be bored, and want more rather than less stimulation, is probably a really good sign? The predicted outcomes on permanent brain damage have been updating in drips and drabs all day, mostly not large numbers, a mix of positive and negative updates, but she's going to be eagerly refreshing it after this. 

She isn't actually sure what to suggest, though, because she's pretty sure the only audiobooks they have that Kalorm will have any chance of keeping up with are, like, literally aimed at two-year-olds, which he is predictably going to hate for different reasons. (And she's sort of worried about what happens if he uses up all of his verbal processing capacity on listening to a children's audiobook and spends the rest of the shift incredibly irritable about being asked questions.) Maybe Khemeth left a note about this– oh, good, perfect. 

"Do you want to listen to nature sounds?" It doesn't sound very interesting to her but it is, in fact, the top recommendation. Khemeth has suggested some tracks. 

Kalorm is SORT OF MAD at...Khemeth, presumably...for something that he is neither able nor especially inclined to break down any further than "it's my own business whether I listen to nature sounds." 

He nods, though. Nature sounds and frustration is better than no nature sounds. He can close his eyes and pretend he's somewhere else that isn't here.

Halthis will get him set up with the very comfortable (not especially well secured) headphones again, and start playing Recording: Forest Sounds Beside A Creek. 

It's still not VERY much like jogging in a forest beside a river, rather than trapped on a hospital floor and mildly short of breath for stupid reasons. There is still a lot of horribleness. But Kalorm will admit (sort of grudgingly, even to himself) that as many as one or two things are not horrible. And if he can convince them he's ready to breathe on his own, then he can get rid of one more horrible thing, but a big thing. 

He listens to nature sounds and breathes and is completely oblivious to Halthis' gentle attempts to note when 30 minutes pass...45 minutes...

Coming up on an hour now. It's nearly 6:30 pm. Four and a half hours into her shift. It's not the most stressful or exhausting four and a half hours of a shift that Halthis has experienced, but it's up there. 

Kalorm...does not actually look super ready to come off the ventilator. The markets are only putting a 40% chance that he would need to be reintubated in the next 12 hours if he pulled the tube out now, but 4 in 10 is kind of way higher than what you want, and a lot of that probability mass is on cases where a neurotypical dath ilani would definitely prefer to be on a ventilator and comfortably sedated rather than spending all night kind of struggling to breathe with the supplementary oxygen and a lot of emergency bronchodilators. 

He's definitely on a trend of rapid improvement, especially since the successful dialysis. At the start of her shift, the 20% / 50% / 80% spread on when he would be ready for extubation (with Treatment Planning approval as opposed to "Kalorm decides this himself") was 6 hours / 13 hours / 36 hours. Now, 4.5 hours later, it's updated to 2 hours / 5 hours / 9 hours. The 80% number has dropped a lot, because a new unexpected respiratory complication before Kalorm is ready to come off the ventilator is now considered less than 20% likely. Complications aren't impossible - the 95% interval, not usually displayed on the main summary screen but viewable if you tab in, is still at 30 hours - but they're most of the way to safe. 

Anyway. She's undeniably going to be handing over a patient in better shape than she started with. Assuming she can convince him to rest after this, see if one more sleep cycle with full ventilator support gets him to a point when the work of breathing is more sustainable for 12+ hours, or compatible with sleeping. She's pretty sure it's keeping him awake right now; his EEG hasn't even slipped into the early stages of sleep, and she reviewed some historical logs and confirmed he's never stayed awake for over an hour at a stretch before. (Well. Not during his hospitalization. Presumably he managed it regularly before the accident.) 

...She's not going to interrupt him just because it's now been a full hour. He shows no signs of actual distress and seems engaged in his nature sounds, he's not due to turn yet and actually doesn't sound like he needs suctioning again just yet, and she's sort of curious how long he'll actually go before noticing. Meanwhile, she has Diagnostic prediction updates to review! 

The markets are sparser than they were during the height of the emergency, but still very well funded, and drawing on a slightly distinct pool of experts, since longer-term critical care medicine is not exactly the same domain of expertise as emergency medicine. It's not as interesting a case to be involved in, and the sheer quantity of data coming in to feed to medical trading company algorithms is much smaller now that they've mostly dispensed with invasive sensors, but it's still pretty interesting, particularly as a case study of a situation that happens relatively rarely.

And there's money to be made for experts who can take in the human dimension, how Kalorm interacts with his medtechs, and feed that into outcome predictions. Halthis, who is vastly ahead of Merrin on ability and willingness to look at large quantities of numerical data, can go into each outcome measure and see how the predictions have evolved since Merrin looked at them eighteen hours ago. 

The probability spread on when Kalorm's pain will be sufficiently controlled that they can ease him off the high-dose sedating painkillers, and get a more meaningful neurological assessment, have tightened up considerably. If the original spread had somehow stayed unchanged, the current 20% / 50% / 80% would be 6 hours / 20 hours / 54 hours. You would basically never expect this to happen, though; as they get closer and more and more best-case or worst-case scenarios can be mostly ruled out, uncertainty decreases. The markets have updated on the fact that Halthis is already weaning the dose on the continuous painkillers, and that Kalorm is surprisingly unbothered by physical pain – but also the fact that he clearly is still in a lot of pain. The spread is now 9 hours / 16 hours / 30 hours. 

Kalorm starting physical rehab: there was a significant downward update on the predictions when he started proactively asking for passive range-of-motion exercises. This doesn't yet count for meeting that milestone; initial rehab will be bed-based, or rather floor-based given the givens, but Kalorm needs to be alert and strong enough - and motivated enough - to actively participate. He's clearly doing great on motivation, and seems physically stronger after successful dialysis, but but his fine motor abilities and coordination haven't shown any significant improvements. At this point, it's looking more likely that the problem is minor brain damage, or maybe peripheral nerve damage, rather than just drugs and exhaustion. Still, the new 20% / 50% / 80% spread is 12 hours / 28 hours / 48 hours.  

Predictions on Kalorm being able to eat or drink again: intervals are still pretty wide, and the optimistic 20% estimates are actually less optimistic now. Despite lower doses of painkillers and normalized magnesium, his gut motility is showing no sign of responding to the gentler drug treatment they started with, and he was experiencing significant nausea. His nasogastric tube is still draining bile, which under normal conditions shouldn't really be in his stomach at all, and indicates that the smooth muscle function of his stomach and upper intestines is still badly disrupted; this hints that, in addition to tissue damage to the membrane lining, the prolonged low-circulation, low-oxygenation period may have damaged the enteric nerves that regulate this process. Given the lack of other risk factors, it's very likely (>95%) that the damage will repair itself over time, but...more time.

The 20% / 50% / 80% spread for tolerating clear fluids (which, as Merrin clarified, means Kalorm tolerating juice or broth as a significant fluid source without nausea or vomiting, not just sucking on ice chips with the nasogastric tube helpfully removing anything he swallows from his stomach) is now 72 hours / 5 days / 8 days. For nutritious liquids – defined as Kalorm being able to meet at least half of his calorie and protein needs via liquid meal replacements – the new spread is 5 days / 9 days / 12 days. For eating solid food normally, the 50% estimate is now, like, a month.  

Assisted walking >5 m, 20% / 50% / 80%: now 4 days / 7 days / 12 days. 

Unassisted walking >50m: 8 days / 16 days / 25 days.

Off supplementary oxygen for a contiguous 24 hour period: 5 days / 9 days / 16 days. 

Hospital discharge: 14 days / 4 weeks / 10 weeks. It seems like they're vaguely succeeding at making the hospital environment more tolerable for Kalorm, and the experts (including Khemeth, who is apparently now awake enough to contribute to the expert groupchats and put in his bids on the markets, if not to actually get out of bed and come over here) think it's less likely that Kalorm will check himself out of the hospital and leave the moment he's physically capable of it.

(In particular, Khemeth thinks that if they can keep Merrin around - and not intolerably bored, she really isn't specced for rehab cases - then his brother will put up with staying in the hospital until he's at least past the high-risk-of-complications period.) 

Predicted likelihood of a significant complication (that they don't catch early enough to treat proactively and prevent from becoming seriously medically dangerous): 7%. 

Predicted likelihood that Kalorm will fall or otherwise injure himself: significantly down, partly because the floor is in fact safer for this. 14%. 

Predicted likelihood that Kalorm will specifically manage to hit his head: less than 5%. Odds of additional permanent neurological damage, less than 1 in 1000. 

Halthis has been staying pretty up to date on the evolving outcome prediction markets. She feels good about the recent updates, though. 

Aaaaand where are they at on the critical final outcome prediction, "likelihood of permanent neurological deficits." 

Down! 

Not, actually, that far down. 41%. 2 in 5. 

But even "minor brain damage" is a big category, that elides over a vast range of possible specific deficits. The markets are now breaking that prediction down in further detail, splitting out the estimated likelihood of various different possible futures that would all technically fall under "permanent minor neurological deficits", while importantly different in other ways. 

(There's still a lot of guesswork, here, because they can't really ask Kalorm for a detailed self-report on what was going on in his head while he appeared to slowly and effortfully work his way through ten simple line drawings, and they certainly can't put him through any formal tests.) 

Kalorm's most obvious and severe current deficits, relative to his baseline, appear to be around visual-spatial processing. More broadly, his reaction times are delayed and his raw processing speed, on just about any mental task, is way down. At this point, with inflammatory and repair processes still happening throughout his brain, this is something they would expect to observe either way. And there are other factors probably contributing. Pain and nausea are distracting; nobody (not even Keepers) functions at peak cognitive capacity while in serious physical pain. Kalorm's auditory processing, at least, does seem to have improved at least modestly since he first woke up. 

From the outside, he appears to have some kind of generalized-planning-and-execution impairment, but the problem could be in a variety of places: reduced working memory capacity, reduced attention span and attentional control, executive dysfunction and difficulty turning intentions into action, or even just a degree of drowsiness and brainfog that makes explicit reasoning aversive for him. And he's having to work much harder than usual for the basic building blocks of communication; that, combined with low energy, makes it unsurprising if he doesn't have a lot of capacity left over once he's managed to communicate his basic needs. 

(There's also whatever is going on with his fine motor skills, but the markets are actually very optimistic that any neurologically-caused deficits there will be resolved in 3-6 months. Rehab and occupational therapy are really good. Stroke patients who recover with minimal or no global cognitive deficits and only have local motor deficits are more likely to stick around for it rather than choosing cryo, so they have both a lot of historical case data, and major optimization effort having gone into the relevant rehab programs. Kalorm...seems to have some Merrin-like traits, actually, when it comes to both stamina and sheer stubbornness. The experts think he has 9 in 10 odds of achieving "no functional deficits.") 

The main other divide is whether Kalorm ends up with residual global-cognition-affecting deficits, in attentional control or executive function or working memory. The markets are calling a 25% likelihood of that. The remaining 15% of worlds are ones where Kalorm only has more specific residual deficits - most likely, that he never gets back to his baseline on visual-spatial processing. It was a pretty spiky baseline to begin with; he was somehow simultaneously well above median at the subskills relevant to, like, navigating mazes or keeping track of objects in his environment, and also couldn't read.