The Asterisk

The patient had been shot seven times.

This was, statistically speaking, above average. Most gunshot victims who made it to Dr. Tzu Yu's clinic had been shot between one and three times. Seven was ambitious. Seven suggested either a very determined assailant or a very unfortunate series of coincidences.

"Well, actually," Dr. Tzu Yu said, examining the wounds with clinical detachment, "only five of these are significant. The other two are superficial—barely grazed the dermis. Medically speaking, I'd call this a five-shot patient with mild abrasions."

The patient—a young woman in tactical gear, breathing through gritted teeth—stared at him.

"Does that... matter?"

"It affects the invoice." The Doctor's thermal printer whirred. A receipt began emerging from his belt. "I charge by penetration depth. Fairness in billing is essential to the patient-provider relationship."

The clinic existed in the back of what had once been a noodle shop. The neon sign outside still advertised "BEST RAMEN IN SECTOR 7" despite the fact that no ramen had been served here in years. The Doctor found the sign reassuring. It attracted a certain type of customer—people who knew that the best places never looked like what they were.

"You came recommended," the patient said as Tzu Yu prepared his instruments. "El Money said you were the best."

"El Money is generous. I'm merely the most available surgeon willing to operate without questions at 3 AM in a condemned building." He paused, considering. "Though I should mention—I'm not technically a surgeon."

The surgery took forty-seven minutes.

Dr. Tzu Yu narrated the entire procedure. This was not optional.

"The first penetration entered at approximately a 34-degree angle, passing through the trapezius and lodging against the scapula. Medically speaking, this is fortunate—two centimeters to the left and we'd be discussing subclavian artery repair, which would add roughly 4,700 credits to your invoice..."

"Suture application, reinforced polymer, 0.3mm gauge... Hemostatic compound, standard application... Neural pathway stabilization, minor..."

By the time he finished, the receipt had reached the floor and begun pooling like a paper snake.

"Your total comes to 12,847.33 credits." He tore the receipt free and examined it with satisfaction. "I should note that this reflects a 15% discount for El Money referrals. Standard rate would be 15,114.51. I find transparency in billing essential to building trust."

The patient, still groggy from anesthesia, tried to focus on the receipt. The itemization was exhaustive. Impossibly exhaustive.

"You charged me for... individual sutures?"

"Real-time surgery requires real-time billing. You know exactly what each stitch costs. It's the most transparent medical care in the Sprawl."

Fourteen hours later, the boy arrived.

He was carried in by two men who didn't give their names. Dregs runners—you could tell by the boots, the scarring around the jaw from cheap neural jacks, the way their eyes never stopped moving. They set the boy on the table and stepped back and the bigger one said, "Fix him."

The boy was maybe sixteen. Maybe younger. Hard to tell in the Dregs, where childhood is a variable-length condition. He was conscious but not tracking. His eyes were open, fixed on the ceiling, pupils dilated asymmetrically. The left one drifted. The right one didn't move at all.

"When did this start?" Tzu Yu asked, already scanning.

"Two days ago. Got a black-market neural jack installed. Worked fine for a day. Then he started seizing." The runner paused. "The guy who installed it is dead now. That's not related to the jack. Probably."

Tzu Yu examined the neural jack—a crude piece of work, fourth-generation knockoff hardware spliced into the boy's occipital interface port. The installation was rough but functional. The jack itself wasn't the problem.

The problem was the swelling.

"The jack is pressing on the cerebral cortex," Tzu Yu said, tilting the boy's head. "The tissue is inflamed. In a canine patient, this would be a standard presentation of post-implant edema—very common with occipital-lobe-adjacent installations. In racing hounds, I would administer a targeted corticosteroid to reduce the swelling, then reposition the jack by approximately 2.3 millimeters to relieve the pressure on the surrounding tissue."

He paused.

"This is essentially the same procedure."

He was already reaching for the instruments.

The boy's name, one of the runners eventually mentioned, was Dael. He'd been running data packages across Sector 9 for the past year. The neural jack was supposed to let him carry encrypted payloads directly—no physical media, no intercept risk, higher pay. The investment was supposed to pay for itself in three runs.

"Common enough motivation," Tzu Yu said, examining the scan results more carefully now. "The desire for economic mobility through augmentation. Medically speaking, it's the primary driver of unlicensed cybernetic installation in the Dregs, accounting for approximately 67% of—"

"Can you fix him or not?"

"Yes. The procedure is straightforward." He printed the pre-operative estimate. "The corticosteroid protocol will reduce the swelling within forty minutes. I'll reposition the jack during the same session. Total procedure time: approximately one hour and twelve minutes. Cost: 3,200 credits, inclusive of materials and anesthesia."

The runners looked at each other.

"We've got 1,800."

"I accept payment plans," Tzu Yu said. "Billing is a matter of principle, not circumstance."

Here is what Dr. Tzu Yu knew: in canine patients, post-implant cerebral edema responds reliably to localized corticosteroid injection. The swelling reduces. The tissue stabilizes. The neural interface seats correctly. He had performed this procedure hundreds of times on racing hounds, show dogs, and Senator Vance's prized retriever. Success rate: 98.3%.

Here is what Dr. Tzu Yu did not know: in human patients, the cerebral cortex responds to localized corticosteroids differently than in canines. Specifically, in a subset of cases involving adolescent neural tissue that is still undergoing myelination—still forming the insulating sheaths around nerve fibers, a process that in humans continues until approximately age twenty-five—the corticosteroid does not merely reduce swelling. It disrupts the myelination process itself. The drug treats the symptom and attacks the substrate.

In a dog, this distinction does not exist. Canine neural myelination is complete by eighteen months.

In a sixteen-year-old human, it matters enormously.

This information was available in any standard human neurological pharmacology reference. It was not available in any veterinary pharmacology reference, because it was not relevant to veterinary practice.

Dr. Tzu Yu administered the injection at 4:47 AM.

At 5:03 AM, the boy began seizing.

The seizure lasted ninety-one seconds. Tzu Yu's thermal printer recorded this.

The boy's vitals crashed. Heart rate spiked to 189, then dropped to 42. Neural activity—monitored through the very jack that had brought him here—showed cascading signal failure across the motor cortex. The myelination disruption was propagating.

Tzu Yu worked with the same steady hands he always had. He administered anticonvulsants. He stabilized the cardiac rhythm. He applied a neural dampening field to slow the cascading demyelination. These interventions were not veterinary. These were protocols he had developed himself over seventeen years of treating human patients, iterating on a body of knowledge assembled one case at a time, adjusted and readjusted with each new outcome.

The boy stabilized.

His vitals normalized.

At 6:14 AM, Dael opened his eyes. Both pupils tracked. He asked where he was. He asked for water. He moved his fingers when asked.

He could not feel his left hand. He could not feel his left forearm. When Tzu Yu pressed a diagnostic pin to the skin of his left wrist, Dael stared at it with an expression of confused recognition—he could see the pin, he understood what it was, he could not locate the sensation of it touching him.

"Localized sensory neuropathy," Tzu Yu said, making notes. "The demyelination appears to have affected the contralateral somatosensory pathways. Medically speaking, this is consistent with—"

He stopped.

He consulted his scan results again. He cross-referenced the dosage he had administered against the boy's weight, age, and neural development markers. He recalculated.

"The dosage was calibrated for a mature neural architecture," he said, quietly. Not to anyone in the room. To the thermal printer, perhaps, which recorded everything. "The myelination variance in adolescent patients would require a 40% reduction in concentration, with a modified delivery vector targeting the inflammation pathway without interacting with the—"

He stopped again.

"I see," he said.

The runners came back for the boy at noon. Dael could walk. He could talk. He could carry data packages. He could not feel his left hand, and the prognosis for recovery was, in Tzu Yu's precise estimation, "uncertain, with a 60% probability of partial restoration over six to eighteen months, contingent on the degree of remyelination that occurs naturally, which in adolescent patients tends to be—"

"Will he be able to work?" the bigger runner asked.

"Yes. The neural jack is functional. The sensory deficit is limited to the left hand and forearm."

"Then we're good."

They left. The boy looked back once, at the doorway. He was flexing his left hand open and closed, watching the fingers move, trying to reconcile what he saw with what he couldn't feel.

The invoice was completely sincere. Every line item was accurate. The adverse event management fee was lower than the initial consultation fee because seizure response requires fewer materials. The protocol revision was free because Dr. Tzu Yu considered it a cost of his own professional development, not a service rendered to the patient.

He filed the invoice. He filed the case notes. He updated his dosage protocols for adolescent patients, adding a new column to his spreadsheet: "Myelination Status (estimated)." He annotated the corticosteroid entry with a flag: "REDUCE 40% FOR INCOMPLETE MYELINATION. SEE CASE 2184-0917-DAEL."

He had now treated one case of adolescent corticosteroid-induced demyelination.

His adjusted protocol was based on a dataset of one.

Kira Vasquez came by the clinic three days later. She did this sometimes—not socially, exactly, but in the way that two people operating in the same illegal margin of the same broken system occasionally check on each other. She brought her own supplies. She never asked for anything. She sat on the counter and drank tea from a thermos and talked about her patients in the way that soldiers talk about their deployments: specific, clipped, with the weight carried in what wasn't said.

"I heard about the boy," she said.

Tzu Yu was recalibrating an instrument. He didn't look up. "The outcome was suboptimal. I've revised the protocol."

"Tzu." She set the thermos down. "He lost feeling in his hand."

"Partial sensory neuropathy. Sixty percent probability of natural recovery within—"

"I know the prognosis. I'm asking you something different." She waited until he looked at her. "How did you calculate the dosage?"

"Standard protocol for post-implant cerebral edema. Corticosteroid concentration scaled to body mass, with adjustment for—"

"For what species?"

The clinic was quiet. The neon RAMEN sign buzzed outside.

"The protocol was developed for canine patients," Tzu Yu said. "The physiological response is—"

"Not the same."

"In 73.4% of parameters, the response is—"

"Not. The. Same." Kira's voice was not angry. That was the worst part. She sounded like someone explaining something to a person she respected, which made the explanation worse. "You used a vet protocol on a kid whose brain was still growing. The myelin sheaths aren't finished. You'd know that if you'd ever opened a human neuro-pharm textbook."

"I have read extensively on human—"

"You've read. You haven't been trained. There's a difference." She picked up the thermos again. Took a sip. Set it down. "I'm not saying you're bad at this, Tzu. I'm saying you have a blind spot the exact size and shape of your training, and you can't see it because it's a blind spot."

Tzu Yu was quiet for a moment.

"My outcomes are better than the alternatives available to patients in this sector," he said. "Across 2,847 procedures over seventeen years, my serious adverse event rate is 1.7%. The nearest comparable rate for unlicensed practitioners in the Dregs is approximately 12%. For corporate emergency services that actually respond to Dregs calls—which is to say, none—the rate is undefined, because the service does not exist."

"I know your numbers."

"Then you know the math."

"I know the math." Kira stood up. She was already at the door. "I also know that kid can't feel his hand. And I know that the next kid you treat might lose more than that. And I know you won't know which kid it is beforehand, because your dataset is one."

She paused at the doorway.

"'Better than nothing' is not the same as 'good enough,' Tzu. And the gap between those two things is where people get hurt."

Dr. Tzu Yu sat in the empty clinic for a long time after she left.

He understood, in a technical sense, what she was saying. The dosage error was real. The protocol gap was real. The revision he had made was based on a single case, extrapolated from first principles, without the benefit of clinical trials or peer review or any of the institutional infrastructure that existed, in theory, to prevent exactly this kind of error.

He understood all of this.

What he did not understand was the conclusion she seemed to be drawing from it.

He had made an error. He had identified the error. He had revised the protocol. The next adolescent patient would receive the adjusted dosage. If the adjustment itself proved wrong—if the 40% reduction was too much, or not enough, or if there was a secondary interaction he hadn't anticipated—he would identify that error too, and revise again.

This was how medicine worked. You treated patients. You observed outcomes. You adjusted. You treated more patients. The dataset grew. The protocols improved. The error rate decreased.

He could not understand why Kira seemed to think this process was the problem rather than the solution.

The boy could not feel his hand. This was true. The boy was alive, neurologically intact in all other respects, and able to work. This was also true. The boy would have died of progressive cerebral edema within approximately seventy-two hours if no one had treated him. This was also true.

The next patient might suffer a different adverse outcome from the adjusted protocol. This was possible. The next patient would also die without treatment. This was certain.

Dr. Tzu Yu picked up the thermos Kira had left behind. He poured himself a cup of tea. It was good tea—she always brought the good kind, the real-leaf stuff from the upper Sprawl markets, not the synthesized packets that tasted like someone had described tea to a machine that had never experienced flavor.

He drank the tea.

He opened his patient log. He reviewed the updated protocol. He added a second annotation: "MONITOR: Myelination assessment should precede all neural-adjacent pharmacological interventions in patients presenting under age 25. Imaging required. Estimated additional procedure time: 12 minutes. Estimated additional cost: 200 credits. Bill to patient."

He added the imaging cost to his standard pre-operative checklist. He updated the estimate template. He printed a test invoice to verify the formatting.

Tomorrow a patient would come. Or tonight. They always did. Someone shot, or cut, or seizing, or dying in a way the corporate system had decided was not its problem. They would come to the noodle shop with the sign that promised ramen and delivered surgery. They would lie on the table and Dr. Tzu Yu would examine them and the thermal printer would begin its accounting and he would fix what he could fix using what he knew, which was more than most people in the Dregs and less than what the work required.

He finished the tea.

He washed the cup.

He did not know if the protocol revision was right. He did not know which patient would reveal the next gap. He did not know if the next adverse event would be minor or catastrophic or if the patient would be sixteen or sixty or if they would lose a hand or a life.

He knew that the alternatives were worse.

He continued practicing.