Automated Versus Manual Oxygen Titration With the EMILY.AI Device in Hospitalized Patients With Acute Respiratory Failure
Functional, Clinical, and Predictive Validation Study of the EMILY AI Device in Patients With Acute and Chronic Respiratory Failure: Randomized Substudy in a Conventional Hospital Ward
This randomized study evaluates EMILY.AI, an investigational (non-CE-marked) smart oxygen valve that automatically adjusts the oxygen flow delivered to a patient, second by second, according to oxygen saturation (SpO2), heart rate and respiratory rate measured by a pulse oximeter.
Adults admitted to the pulmonology ward of Hospital Universitario Fundación Jiménez Díaz (Madrid, Spain) with acute respiratory failure are randomly assigned to receive oxygen titrated automatically by EMILY.AI or oxygen managed as usual by the nursing team, from randomization until hospital discharge. All participants wear the device; in the usual-care group it only records data.
The study compares the time patients stay within their target saturation range, episodes of low and high oxygen, nursing workload, oxygen use, length of stay and patient experience. This substudy is part of clinical investigation FJD-EMILY-23-02, whose other substudies are registered in separate records.
연구 개요
상태
상세 설명
Background On conventional hospital wards, oxygen is usually given at a fixed flow and saturation is checked about once per nursing shift, so periods of hypoxaemia or hyperoxaemia can go undetected. EMILY.AI combines a medical-grade oxygen valve with an edge-computing fuzzy-logic algorithm that sets the oxygen flow every second, with continuous logging of physiological signals and delivered flow.
Design Prospective, single-blind (participant), randomized, parallel-group study. Participants are randomized 1:1 to the EMILY arm (automatic titration) or the MANUAL arm (usual care with nursing monitoring every 8 hours and a fixed oxygen prescription), using permuted blocks of 4 stratified by main admission diagnosis (COPD exacerbation, pneumonia, interstitial lung disease, other) and prior long-term home oxygen therapy (yes or no). Allocation takes place after eligibility is confirmed and informed consent is signed.
Intervention and follow-up All participants are connected to the EMILY.AI device continuously from randomization until hospital discharge. In the EMILY arm the device titrates oxygen autonomously to keep SpO2 within the target range for the type of respiratory failure (hypoxaemic 91-97%; hypercapnic 91-94%). In the MANUAL arm the device works in fixed mode, recording SpO2, heart rate and respiratory rate without changing the flow, and nurses adjust oxygen as usual. Conventional back-up oxygen is available at all times.
Sample size 100 participants: 80 planned for the main comparison and approximately 20 added by protocol amendment (CIP v8.0) to further validate an adjustment of the algorithm parameter that sets the minimum oxygen flow in hypoxaemic patients. All 100 are randomized in the same way. As a pilot functional validation, no formal sample size calculation was performed.
The protocol FJD-EMILY-23-02 also includes a substudy in the Respiratory Intermediate Care Unit and an exercise substudy (six-minute walk test), registered separately.
연구 유형
등록 (추정된)
단계
- 해당 없음
연락처 및 위치
연구 연락처
- 이름: Alba Naya Prieto, MD
- 전화번호: +34 915 50 48 00
- 이메일: alba.naya@fjd.es
연구 연락처 백업
- 이름: Sarah Beatrice Heili Frades, MD, PhD
- 전화번호: +34 915 50 48 00
- 이메일: SHeili@fjd.es
연구 장소
-
-
-
Madrid, 스페인, 28040
- Hospital Universitario Fundación Jiménez Díaz
-
연락하다:
- Sarah Beatrice Heili Frades Zimmermann MD, PhD
- 전화번호: +34 915 50 48 00
- 이메일: SHeili@fjd.es
-
연락하다:
- Alba Naya Prieto, MD
- 전화번호: +34 915 50 48 00
- 이메일: alba.naya@fjd.es
-
수석 연구원:
- Sarah Beatrice Heili Frades Zimmermann, MD, PhD
-
-
참여기준
자격 기준
공부할 수 있는 나이
건강한 자원 봉사자를 받아들입니다
연구 대상 성별
설명
Inclusion Criteria:
- Men and women aged 18 years or older.
- Admitted to the general pulmonology ward (Unit 67) of Hospital Universitario Fundación Jiménez Díaz.
- Acute respiratory failure, defined as baseline arterial PaO2 below 60 mmHg or the need for supplemental oxygen to maintain SpO2 above 90%. Patients on long-term oxygen therapy are eligible if a clinical deterioration requires a higher-than-usual oxygen flow to maintain SpO2 above 90%.
- Signed informed consent.
Exclusion Criteria:
- Anatomical abnormalities preventing correct placement of the pulse oximeter.
- Pregnancy or breastfeeding (women of childbearing potential require a negative pregnancy test at screening).
- Need for non-invasive ventilation or high-flow nasal cannula at admission.
- Palliative care or limitation of therapeutic effort at admission.
- Expected hospital stay under 48 hours.
- Clinical situations expected to determine length of stay independently of oxygen therapy, such as intravenous antibiotics for more than 10 days, scheduled interventional or surgical procedures, active social care needs that may delay discharge, a non-respiratory main admission diagnosis, or acute or decompensated non-respiratory comorbidities.
- Known or suspected carbon monoxide poisoning or methemoglobinemia.
- Severe hypercapnia with decompensated respiratory acidosis (pH below 7.30), hemodynamic instability requiring vasoactive drugs, or severe sepsis at admission.
- Cognitive impairment, behavioral disorder or any circumstance preventing proper use of the device or compliance with study procedures.
- Simultaneous participation in another interventional clinical trial, or readmission of a patient previously included in this study.
공부 계획
연구는 어떻게 설계됩니까?
디자인 세부사항
- 주 목적: 치료
- 할당: 무작위
- 중재 모델: 병렬 할당
- 마스킹: 하나의
무기와 개입
참가자 그룹 / 팔 |
개입 / 치료 |
|---|---|
|
실험적: EMILY
Automatic oxygen titration by EMILY.AI 24 hours a day from randomization to hospital discharge, targeting SpO2 91-97% (hypoxaemic) or 91-94% (hypercapnic). |
Investigational smart oxygen valve with an embedded fuzzy-logic algorithm that adjusts oxygen flow every second from pulse-oximetry SpO2, heart rate and respiratory rate to keep SpO2 within a target range. Other names: EMILY.AI Smart Valve |
|
활성 비교기: MANUAL
Usual care: fixed oxygen prescription adjusted by nursing staff, with vital signs checked every 8 hours. The EMILY.AI device is connected in fixed mode and only records data. |
Usual care: oxygen flow set and adjusted by the ward nursing team according to standard practice. The EMILY.AI device stays connected in fixed mode, recording signals but not controlling delivery. |
연구는 무엇을 측정합니까?
주요 결과 측정
결과 측정 |
측정값 설명 |
기간 |
|---|---|---|
|
Percentage of time with SpO2 within the target range
기간: From randomization to hospital discharge, up to 30 days.
|
Percentage of monitored time with SpO2, measured continuously by the device's pulse oximeter, within the target range for the type of respiratory failure (hypoxaemic 91-97%; hypercapnic 91-94%). Compared between the EMILY and MANUAL arms. |
From randomization to hospital discharge, up to 30 days.
|
2차 결과 측정
결과 측정 |
측정값 설명 |
기간 |
|---|---|---|
|
Proportion of participants with SpO2 in range more than 95% of the time
기간: From randomization to hospital discharge, up to 30 days.
|
Proportion of participants in each arm whose SpO2 stays within their target range for more than 95% of the monitored time. |
From randomization to hospital discharge, up to 30 days.
|
|
Time with SpO2 below the target range
기간: From randomization to hospital discharge, up to 30 days.
|
Percentage of monitored time with SpO2 below the target range for the type of respiratory failure (hypoxaemia). |
From randomization to hospital discharge, up to 30 days.
|
|
Time with SpO2 above the target range
기간: From randomization to hospital discharge, up to 30 days.
|
Percentage of monitored time with SpO2 above the target range while oxygen is being delivered (hyperoxaemia). |
From randomization to hospital discharge, up to 30 days.
|
|
Nursing oxygen flow adjustments
기간: From randomization to hospital discharge, up to 30 days.
|
Number of oxygen flow changes and oxygen interface changes made by nursing staff per shift (morning, afternoon, night). |
From randomization to hospital discharge, up to 30 days.
|
|
Number of nursing vital sign measurements per shift
기간: From randomization to hospital discharge, up to 30 days.
|
Total number of individual vital sign measurements taken by nursing staff per 8-hour shift, where each recording of SpO2, temperature, blood pressure, heart rate or respiratory rate counts as one measurement. For each participant, the counts are averaged across all shifts of the stay and reported as the mean number of measurements per shift. |
From randomization to hospital discharge, up to 30 days.
|
|
Calls to the physician
기간: From randomization to hospital discharge, up to 30 days.
|
Number of calls to the physician related to desaturation, oxygen flow changes or abnormal vital signs. |
From randomization to hospital discharge, up to 30 days.
|
|
Total oxygen delivered
기간: From randomization to hospital discharge, up to 30 days.
|
Total oxygen delivered (litres) during the hospital stay. |
From randomization to hospital discharge, up to 30 days.
|
|
Mean oxygen flow
기간: From randomization to hospital discharge, up to 30 days.
|
Mean oxygen flow (litres per minute) delivered during the hospital stay. |
From randomization to hospital discharge, up to 30 days.
|
|
Length of hospital stay
기간: From randomization to hospital discharge, up to 30 days.
|
Days from randomization to hospital discharge. |
From randomization to hospital discharge, up to 30 days.
|
|
Escalation to a higher level of respiratory support
기간: From randomization to hospital discharge, up to 30 days.
|
Proportion of participants requiring high-flow nasal cannula, non-invasive ventilation, or transfer to the Respiratory Intermediate Care Unit or ICU due to respiratory or hemodynamic deterioration. |
From randomization to hospital discharge, up to 30 days.
|
|
Device alerts
기간: From randomization to hospital discharge, up to 30 days.
|
Number of alerts generated by EMILY.AI related to oxygenation or vital signs per participant during the hospital stay. |
From randomization to hospital discharge, up to 30 days.
|
|
Percentage of device alerts corresponding to a true clinical change
기간: From randomization to hospital discharge, up to 30 days.
|
Percentage of EMILY.AI alerts confirmed by the clinical team as corresponding to a true clinical change. |
From randomization to hospital discharge, up to 30 days.
|
|
Change in arterial partial pressure of oxygen (PaO2)
기간: At randomization and at hospital discharge (up to 30 days).
|
Change in PaO2 (mmHg) measured by arterial blood gas analysis between randomization and hospital discharge. |
At randomization and at hospital discharge (up to 30 days).
|
|
Change in arterial partial pressure of carbon dioxide (PaCO2)
기간: At randomization and at hospital discharge (up to 30 days).
|
Change in PaCO2 (mmHg) measured by arterial blood gas analysis between randomization and hospital discharge. |
At randomization and at hospital discharge (up to 30 days).
|
|
Change in arterial pH
기간: At randomization and at hospital discharge (up to 30 days).
|
Change in arterial pH (pH units) measured by arterial blood gas analysis between randomization and hospital discharge. |
At randomization and at hospital discharge (up to 30 days).
|
|
Change in dyspnoea measured with the modified Borg Dyspnoea Scale
기간: At randomization and at hospital discharge (up to 30 days).
|
Change between randomization and hospital discharge in dyspnoea measured with the modified Borg Dyspnoea Scale, which ranges from 0 (no breathlessness at all) to 10 (maximal breathlessness). Higher scores mean a worse outcome. |
At randomization and at hospital discharge (up to 30 days).
|
|
Satisfaction with oxygen therapy measured with a 5-point Likert scale
기간: At hospital discharge (up to 30 days).
|
Participant-reported satisfaction with oxygen therapy at hospital discharge, measured with a single-item 5-point Likert scale ranging from 1 (very dissatisfied) to 5 (very satisfied). Higher scores mean a better outcome. |
At hospital discharge (up to 30 days).
|
기타 결과 측정
결과 측정 |
측정값 설명 |
기간 |
|---|---|---|
|
Intercurrent events
기간: From randomization to hospital discharge, up to 30 days.
|
Number of participants with falls with clinical consequences, non-respiratory nosocomial infection, acute cardiovascular events, major bleeding, decompensation of non-respiratory comorbidity, social problems prolonging stay, or unplanned procedures. |
From randomization to hospital discharge, up to 30 days.
|
공동 작업자 및 조사자
스폰서
수사관
- 수석 연구원: Sarah Beatrice Heili Frades, MD, PhD, MD, PhD, Hospital Universitario Fundación Jiménez Díaz
연구 기록 날짜
연구 주요 날짜
연구 시작 (추정된)
기본 완료 (추정된)
연구 완료 (추정된)
연구 등록 날짜
최초 제출
QC 기준을 충족하는 최초 제출
처음 게시됨 (실제)
연구 기록 업데이트
마지막 업데이트 게시됨 (실제)
QC 기준을 충족하는 마지막 업데이트 제출
마지막으로 확인됨
추가 정보
이 연구와 관련된 용어
기타 연구 ID 번호
- FJD-EMILY-23-02-WARD
- FJD-EMILY-23-02 (기타 식별자: Aether Tech S.L.)
개별 참가자 데이터(IPD) 계획
개별 참가자 데이터(IPD)를 공유할 계획입니까?
약물 및 장치 정보, 연구 문서
미국 FDA 규제 의약품 연구
미국 FDA 규제 기기 제품 연구
이 정보는 변경 없이 clinicaltrials.gov 웹사이트에서 직접 가져온 것입니다. 귀하의 연구 세부 정보를 변경, 제거 또는 업데이트하도록 요청하는 경우 register@clinicaltrials.gov. 문의하십시오. 변경 사항이 clinicaltrials.gov에 구현되는 즉시 저희 웹사이트에도 자동으로 업데이트됩니다. .