ISSUE 013 · Oct 15, 2026 · 9 min read

What if kidney rescue began in the first 30 minutes of oliguria?

After cardiac surgery, use a smart catheter to read minute-level urine output, combine it with hemodynamic context and human cause confirmation, and route obstruction-, hypoperfusion-, or congestion-specific actions—then judge success by severe AKI, not alert volume.

The issue in one sentence

Diagram 01

From each drop to cause-specific rescue care

01Continuous output→

02Occlusion · quality→

0330-min risk window→

04Hemodynamic context→

05Human cause check→

06Targeted action→

07Kidney · fluid outcome

ANM proposed model. Low urine output is not an instruction to give fluid. Clinicians verify drainage quality and hemodynamics, choose a cause-specific action, and feed response and harm back into the record.

01Clinical problem · Evidence

Urine output can fall before creatinine rises—but charting is delayed and causes differ.

AKI after cardiac surgery raises the risk of dialysis, prolonged stay, and death. Manual hourly measurement is delayed and error-prone, while identical oliguria may reflect tubing obstruction, hypoperfusion, venous congestion, medication, or true kidney injury. Context-free alerts can cause inappropriate fluid and alert fatigue.

02Technology stack · Proposal

Bind a smart Foley and AI to verify-before-action care.

The sensor captures minute-level urine output, drainage flow, and temperature, with optional intra-abdominal pressure. AI summarizes weight-adjusted oliguria, possible obstruction, and trends in MAP, vasopressors, fluid balance, and central venous pressure without diagnosing. Nurse and physician verify catheter position, bladder distension, bleeding, and hemodynamics before choosing flushing, medication review, ultrasound, a fluid challenge, or decongestion.

03Research question · Hypothesis

Can better cause-specific action—not faster numbers alone—reduce severe AKI?

Test whether a verification-gated pathway combining minute-level oliguria with hemodynamic phenotyping reduces KDIGO stage 2–3 AKI within seven postoperative days versus standard hourly monitoring using the same sensor in silent mode.

04First study · Proposal

Eighteen cardiac-surgery ICUs, 2,400 people, a silent-sensor randomized trial.

Randomize adults requiring a Foley after cardiac surgery 1:1 on ICU arrival. In the intervention arm, 30 minutes of sustained oliguria or an abrupt fall triggers a quality gate, cause checklist, and physician-authorized action; control data are stored only on the research server while usual hourly observation continues. The single primary endpoint is KDIGO stage 2–3 AKI within seven days. Also measure MAKE30, dialysis, fluid overload, arrhythmia, ventilation and length of stay, action per alert, and nursing time.

05Defensible data · Hypothesis

The moat is not the urine curve—it is action and physiologic response.

Link raw flow, occlusion, temperature and pressure with MAP, CVP, vasopressors, fluid, diuretics, ultrasound, action timestamps, 30- and 120-minute response, creatinine, dialysis, fluid overload, and infection. Randomized counterfactual data can reveal which actions help—or harm—specific phenotypes.

06Venture thesis · Assumption

Sell a kidney-rescue-ready ICU bed-day—not a sensor.

Initial customers are high-volume cardiac-surgery hospitals and integrated ICU networks. Price the sensor and disposable, EHR integration, 24/7 cause-verification protocol, nurse training, and alert-to-action outcomes audit per monitored bed-day. Offer shared savings on AKI and length of stay only after randomized evidence shows both clinical benefit and net cost reduction.

Evidence map

The gap between measurement accuracy and patient benefit

+19.9 mL

Mean hourly overestimation in nurse charting among 44 post-cardiac-surgery patients

47 min

Mean delay in hourly urine-output documentation in the same study

2,208 people

2024 randomized AKI alert-plus-bundle trial; no patient-centered benefit

No clinical data

No clinical data required for the additional catheter sizes in FDA K221020

Evidence

The sensor is more accurate—but an alert alone is not enough

Established: in a prospective study comparing 187 hours among 44 post-cardiac-surgery patients, nurse charting overestimated hourly urine output by a mean 19.9 mL and was entered a mean 47 minutes late. The sensor differed from the reference by 2.29 mL on average, not significantly. A 2026 non-contact optical prototype achieved R² 0.9889 and mean error of −0.1 mL in its final version across 285 hours of animal-experiment validation, but it did not test human clinical benefit. More importantly, a 2024 randomized trial of 2,208 adults found that a creatinine-based AKI alert plus care bundle changed behavior and diagnosis but did not improve kidney function or patient-centered outcomes. FDA K221020 specifies urine-output, temperature, and intra-abdominal-pressure monitoring—not AKI prediction or treatment effectiveness—and required no clinical data for the added catheter sizes. Reduced severe AKI, nursing time, and cost therefore remain hypotheses.

44-person automated-monitoring study 2,208-person AKI alert trial 2026 optical flow-sensor validation Automated urine-output trial registry FDA smart-Foley 510(k)

Failure conditions and safety, ethics, and regulatory risks

Clinical failure condition

Do not scale unless seven-day KDIGO stage 2–3 AKI falls by at least 20% relatively with directionally consistent MAKE30, dialysis, and length-of-stay findings. Faster charting and alert response cannot rescue clinical failure.

Inappropriate fluid

Never equate oliguria automatically with hypovolemia. When congestion, right-heart failure, or high intra-abdominal pressure is plausible, withhold reflex fluid and require ultrasound or hemodynamic review; independently adjudicate pulmonary edema and arrhythmia.

Occlusion and false alerts

Kinks, air locks, clots, and patient position can create false oliguria. Put the quality gate and bladder scan before treatment, and suspend the algorithm if false alerts exceed two per patient-day.

Catheter harm

Track urinary infection, urethral injury, and unnecessary dwell time. Never insert a Foley solely for the sensor or delay removal; reassess indication daily.

Alert fatigue and accountability

Deduplicate alerts, assign a named responder, and distinguish unreviewed alerts from model error in audits. Prohibit automated orders, fluid boluses, or diuresis.

Regulatory and business boundary

A 510(k) for urine-output monitoring is not authorization for AKI prediction or treatment advice. Disease-risk scores and action recommendations may require separate review, so pursue early FDA feedback plus cybersecurity, interoperability, and human-factors validation. Contract on patient outcomes and fewer catheter-days—not alert volume.

Bibliography

References

Established facts are separated from research and venture hypotheses that still require validation.

01
Minor J, Smith A, Deutsch F, Kellum JA. Automated versus manual urine output monitoring in the intensive care unit. Sci Rep. 2021;11:17429.
A prospective observational study of 44 post-cardiac-surgery patients and 187 hours; automated sensing was more accurate and manual charting delayed, but AKI reduction was not tested.PMID 34465821 · DOI 10.1038/s41598-021-97026-8
02
Li T, Wu B, Li L, et al. Automated Electronic Alert for the Care and Outcomes of Adults With Acute Kidney Injury: A Randomized Clinical Trial. JAMA Netw Open. 2024;7:e2351710.
A 2,208-person randomized trial in which an alert plus care bundle changed clinician behavior but did not improve kidney function or patient-centered outcomes.PMID 38241047 · NCT03736304
03
Hota P, Shyamala Pandian A, Domínguez RE, et al. Development and Validation of an Optical Sensor-Based Automated Urine Flow Meter for Real-Time Patient Monitoring. Sensors. 2026;26(3):849.
Engineering validation across 285 hours during animal experiments; it supports prototype accuracy, not human AKI outcomes.PMID 41682367 · DOI 10.3390/s26030849
04
ClinicalTrials.gov. Automated System for Monitoring Urine Output in Intensive Care Patients.
A registered study testing whether strict automated urine-output monitoring affects AKI and fluid balance; registration itself is not evidence of benefit.NCT05232045
05
U.S. Food and Drug Administration. Accuryn Monitoring System and SmartFoley IAP UO Temp Tray System. 510(k) Premarket Notification.
Primary regulatory record for urine output, core temperature, and intra-abdominal pressure monitoring. It does not authorize AKI prediction or treatment advice, and no clinical data were required for the added sizes.K221020 · Decision Jan 13, 2023

The next issue publishes every day at 9 AM KST.All issues