Search Results
Found 5 results
510(k) Data Aggregation
(265 days)
Indications for Use: This device is intended for use in patients who are morbidly obese and have failed to lose weight with diet and exercise.
{{deviceDescription}}
Here's a breakdown of the acceptance criteria and study details for the Hypertension Prediction Index (HePI) Algorithm, based on the provided FDA 510(k) letter:
1. Table of Acceptance Criteria and Reported Device Performance
| Performance Metric | Acceptance Criteria | Reported Device Performance (Minimally Invasive) | Reported Device Performance (Non-Invasive) |
|---|---|---|---|
| Sensitivity (%) | > 80% | 99.7% [99.6, 99.9] | 99.6% [99.1, 100.0] |
| Specificity (%) | > 80% | 93.7% [92.9, 94.3] | 91.6% [90.7, 92.5] |
Note: The reported performance is for the overall datasets (N=1813 for Minimally Invasive, N=1351 for Non-invasive). All sub-categories (surgical/non-surgical) also met the acceptance criteria.
2. Sample Size Used for the Test Set and Data Provenance
- Minimally Invasive Sensor:
- US Patients: 1615 subjects
- OUS Patients: 198 subjects
- Total N: 1813 subjects
- Non-Invasive Finger Cuff:
- US Patients: 464 subjects
- OUS Patients: 887 subjects
- Total N: 1351 subjects
Data Provenance: The study used retrospective clinical data from multiple independent datasets. Data was collected from both US and OUS (Outside United States) patients.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
The provided document does not specify the number of experts used or their qualifications for establishing the ground truth.
4. Adjudication Method for the Test Set
The provided document does not specify an adjudication method. The ground truth definition of a "hypertensive event" is clearly stated (MAP > 115 mmHg for at least 1 minute or MAP increase of > 20% when current MAP > 95 mmHg), suggesting an objective, pre-defined criterion rather than expert consensus on individual cases that would require adjudication.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done
The provided document does not indicate that an MRMC comparative effectiveness study was done. The focus is on the standalone performance of the algorithm.
6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done
Yes, a standalone performance study was done. The results presented in Table 2 (AUC, Sensitivity, Specificity, PPV, NPV) are direct measures of the algorithm's performance in predicting hypertensive events based on retrospective clinical data, without human interaction.
7. The Type of Ground Truth Used
The ground truth used is defined by objective physiological measurements and thresholds:
A "hypertensive event" is defined as:
- Mean Arterial Pressure (MAP) greater than 115 mmHg for at least 1 minute OR
- A MAP increase of more than 20% when current MAP is greater than 95 mmHg.
8. The Sample Size for the Training Set
The provided document does not explicitly state the sample size for the training set. It mentions that "Algorithm performance was tested using retrospective clinical data" and "Prospective analyses of retrospective clinical data from multiple independent datasets...were analyzed to verify the safety and performance of the subject device," referring to the test sets.
9. How the Ground Truth for the Training Set Was Established
The provided document does not explicitly state how the ground truth for the training set was established. However, given the nature of the device and the ground truth definition for the test set, it is highly likely that the same objective physiological measurements and thresholds (MAP > 115 mmHg for at least 1 minute or MAP increase of > 20% when current MAP > 95 mmHg) were used to establish ground truth labels for the training data.
Ask a specific question about this device
(150 days)
The Edwards Acumen Hypotension Prediction Index software feature provides the clinician with physiological insight into a patient's likelihood of future hypotensive events and the associated hemodynamics. The Acumen is intended for use in surgical or non-surgical patients receiving advanced hemodynamic monitoring. The Acumen HPI feature is considered to be additional quantitative information regarding the patient's physiological condition for reference only and is not intended to make therapeutic decisions solely on the Acumen Hypotension Prediction Index (HPI) parameter.
The Acumen Hypotension Prediction Index parameter (HPI) provides the clinician with the likelihood that the patient may be trending toward a hypotensive event. The Acumen HPI feature is intended for use in surgical or non-surgical patients. By default, the software defines a hypotensive event as mean arterial pressure (MAP) < 65 mmHg for at least one minute in duration. This MAP target is adjustable and can be changed to 55, 60, 70, 75, 80 or 85 mmHg.
The Edwards Acumen Hypotension Prediction Index (HPI) software feature was evaluated through algorithm verification and a usability study. The key aspects of the evaluation are as follows:
1. Acceptance Criteria and Reported Device Performance:
The document primarily focuses on verifying that the changes to the HPI algorithm (adjustable MAP targets) and the expansion to non-surgical patients did not negatively impact its safety and effectiveness. Specific numerical acceptance criteria for performance metrics (e.g., sensitivity, specificity, accuracy) are not explicitly stated in the provided text. Instead, the document states:
- "The results establish that the usage of the HPI algorithm with the adjustable Mean Arterial Pressure (MAP) targets for hypotension (55, 60, 70, 75, 80, 85 mmHg) did not adversely affect the safety and effectiveness of the subject device."
- "The usability study demonstrated that the intended users can perform primary operating functions and critical tasks of the system without any usability issues that may lead to patient or user harm."
Without specific performance metrics and their corresponding acceptance thresholds, a table of acceptance criteria versus reported performance cannot be fully constructed from the provided text. The overall reported performance is that the device meets the implicit acceptance criterion of not adversely affecting safety and effectiveness with the modifications.
2. Sample Size and Data Provenance for Test Set:
- Algorithm Verification: The algorithm verification was performed using "clinical data." For the expanded non-surgical indication, "non-surgical clinical data collected retrospectively" was used.
- Sample Size: The document does not specify the sample size for the clinical data used in the algorithm verification test set.
- Data Provenance: The data used for the non-surgical indication was "retrospectively" collected. The country of origin is not specified.
3. Number of Experts and Qualifications for Ground Truth:
The document does not provide information on the number of experts used or their qualifications for establishing the ground truth for the test set.
4. Adjudication Method:
The document does not specify any adjudication method used for the test set.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
The document does not mention a multi-reader multi-case (MRMC) comparative effectiveness study, nor does it provide any effect size of human readers improving with AI vs. without AI assistance. The HPI is an "Adjunctive Predictive Cardiovascular Indicator" providing "additional quantitative information" and is "not intended to make therapeutic decisions solely on the Acumen Hypotension Prediction Index (HPI) parameter," suggesting it is an aid rather than a replacement for human decision-making. However, the exact nature of human-in-the-loop studies, if any, is not detailed.
6. Standalone Performance:
Yes, a standalone (algorithm only) performance evaluation was done through the "Algorithm Verification" section. The verification confirmed that the modified algorithm, with adjustable MAP targets, did not adversely affect safety and effectiveness. This indicates an evaluation of the algorithm's performance independent of real-time human interaction.
7. Type of Ground Truth:
The ground truth used for algorithm verification appears to be based on observed "hypotensive events" defined by Mean Arterial Pressure (MAP) falling below user-defined thresholds (e.g., < 65 mmHg for at least one minute). The document mentions "clinical data" and "non-surgical clinical data," which implies real physiological measurements and outcomes from patients, rather than expert consensus on images or pathology.
8. Sample Size for Training Set:
The document does not specify the sample size for the training set. It mentions that the subject and predicate devices "have the same core predictive algorithm," but details on the original training of this algorithm are not provided in this submission summary.
9. How Ground Truth for Training Set was Established:
The document does not provide information on how the ground truth for the training set was established. It refers to the "core predictive algorithm" being the same as previously cleared devices (DEN160044 and K203687), suggesting that the original algorithm's development and training methods would be detailed in those previous submissions. This current submission focuses on verifying that modifications to the existing algorithm do not negatively impact its performance.
Ask a specific question about this device
(270 days)
The Edwards Lifesciences Acumen Hypotension Prediction Index feature provides the clinician with physiological insight into a patient's likelihood of future hypotensive events (defined as mean arterial pressure < 65 mmHg for at least one minute in duration) and the associated hemodynamics. The Acumen HPI feature is intended for use in surgical or nonsurgical patients receiving advanced hemodynamic monitoring. The Acumen HPI feature is considered to be additional quantitative information regarding the patient's physiological condition for reference only and no therapeutic decisions should be made based solely on the Hypotension Prediction Index (HPI) parameter.
The Acumen Hypotension Prediction Index feature consists of software running on the Edwards Lifesciences HemoSphere Advanced Monitoring Platform paired with the Acumen IQ extravascular blood pressure transducer (K152980) and a peripheral arterial catheter. The monitoring system includes the Acumen Hypotension Prediction Index (HPI), and graphical user interface features displaying hemodynamic parameters relevant to assessing the root cause of a potential hypotensive event. The Acumen Hypotension Prediction Index is an index related to the likelihood of a patient experiencing hemodynamic instability defined as a hypotensive event* within 15 minutes, where zero (0) indicates low likelihood and one hundred (100) indicates a hypotensive event is occurring. The Acumen Hypotension Prediction Index parameter (HPI), should not be used exclusively to treat patients. A review of the patient's hemodynamics is recommended prior to initiating treatment. * A hypotensive event is defined as mean arterial pressure (MAP) < 65 mmHg for one minute in duration
Here is the extracted information regarding the acceptance criteria and the study that proves the device meets them:
1. A table of acceptance criteria and the reported device performance
| Acceptance Criteria (Primary Effectiveness Endpoint) | Reported Device Performance (HPI Study) |
|---|---|
| Reduce the mean duration of intraoperative hypotension (defined as MAP < 65 mmHg for at least 1 minute) (IOH) by at least 25% in surgical patients that require advanced hemodynamic monitoring, compared with a historic retrospective control group (MPOG). The incidence of IOH in the MPOG group was 88%. An episode of IOH was defined as a mean arterial pressure (MAP) below 65 for three (3) or more consecutive 20-second events. | The HPI Study met its primary effectiveness endpoint. The HPI Pivotal Subjects (full analysis set, n=406) experienced a mean IOH duration of 11.97 ± 13.92 minutes. This represents a reduction of 57.6% compared to the MPOG historical control mean IOH of 28.20 ± 42.60 minutes (p<0.0001). When considering zero episodes of IOH, there was a 65% reduction (p<0.0001). |
| Secondary Effectiveness Endpoint: Determination of total area under the curve (AUC) of the time and MAP for all time periods for which MAP < 65 mmHg in each Subject. This endpoint is correlated with duration, and a descriptive analysis was presented. | AUC results were provided for various subject groups: - All pivotal Subjects (n=457): Mean AUC = 46.38 MinmmHg - All pivotal Subjects with at least one episode (n=328): Mean AUC = 64.63 MinmmHg - All pivotal Subjects with ≥3 hours surgery duration (n=406): Mean AUC = 47.07 MinmmHg - All pivotal Subjects with ≥3 hours surgery duration and at least one IOH episode (n=293): Mean AUC = 65.23 MinmmHg |
| Primary Safety Endpoint: Percentage of serious adverse events (SAEs) to include perioperative events, postoperative complications, and device-related serious adverse events. | - No Subjects had events adjudicated to have any relationship to the Acumen HPI Feature. - No ADEs or SADEs adjudicated as related to the Acumen HPI Feature. - No unanticipated ADEs (0%) related to the HPI Feature. - No deaths occurred that were related/unrelated to the HPI Feature. |
| Secondary Safety Endpoint: Composite measure of complications (Post-operative non-fatal cardiac arrest, In-hospital death, Stroke, Acute Kidney Injury (AKI) within 30 days, Myocardial Injury in non-cardiac surgery (MINS) within 30 days). The study aimed to determine if the guidance provided by the Acumen HPI Feature reduced this composite measure. | In the Completed Cases (CC) population (n=400), the composite event rate was 4.75% (19 events [95% CI: 2.88, 7.32]). Specific events: one non-fatal cardiac arrest (0.25%), 16 AKI (4.00%), 3 MINS (0.75%). No in-hospital deaths or strokes were reported. In the ITT population (n=460), there were 3 (0.66%) MINS and 17 (3.7%) AKI incidents. |
Study Proving Device Meets Acceptance Criteria:
The study proving the device meets the acceptance criteria is titled:
"A Prospective, Single-Arm, Open-Label, Multicenter Study of the Hypotension Prevention and Treatment in Patients Receiving Arterial Pressure Monitoring with Acumen Hypotension Prediction Index Feature (HPI Study)."
2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)
-
Test Set (HPI Study):
- Sample Size: 485 eligible Subjects (460 pivotal with an additional 25 roll-in cases). The Full Analysis Set (FAS) for primary effectiveness included 406 subjects.
- Data Provenance:
- Country of Origin: United States (11 study sites).
- Nature: Prospective, single-arm, unblinded study.
-
Comparison Group (Historical Control):
- Sample Size: 22,109 patients. The population for the primary effectiveness comparison that experienced IOH was 19,445 subjects.
- Data Provenance:
- Country of Origin: United States (from hospitals across the United States).
- Nature: Retrospective historical control group from the Multicenter Perioperative Outcomes Group (MPOG) academic consortium. Dates of data treatment were between January 1, 2017, and December 31, 2017.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)
The document does not explicitly state the number of experts or their qualifications for establishing the ground truth related to the MPOG historical control data or the real-time monitoring data from the HPI Study.
However, the "ground truth" for hypotension (MAP < 65 mmHg for at least one minute) is a directly measurable physiological parameter captured by arterial pressure monitoring, which is a standard of care. The definition of IOH (MAP < 65 mmHg for three or more consecutive 20-second events) is based on established clinical understanding and recorded data rather than expert adjudication of an image or subjective finding.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
The document does not mention an adjudication method (like 2+1 or 3+1) for establishing the ground truth or evaluating the outcomes in the HPI study. The primary outcome (duration of IOH) and secondary outcomes (AUC, adverse events) were based on directly measured physiological data (MAP) and recorded clinical events, rather than subjective interpretations requiring multiple expert adjudicators. Serious adverse events (SAEs) were "adjudicated," but the specifics of this adjudication process (e.g., number of adjudicators, their roles) are not detailed.
5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance
A multi-reader multi-case (MRMC) comparative effectiveness study was not explicitly stated or described as typically understood in the context of image interpretation or diagnostic performance comparison between human readers and AI.
Instead, the study investigated the effectiveness of the Acumen HPI Feature as a decision support tool for clinicians in reducing the duration of IOH. This is a comparative effectiveness study comparing a cohort where clinicians had access to the HPI feature's guidance (HPI Study arm) versus a historical control cohort where clinicians did not have this specific AI-driven predictive guidance (MPOG).
The "effect size" can be considered as the 57.6% reduction in mean IOH duration observed in the HPI arm compared to the MPOG historical control, and a 65% reduction when considering instances with zero IOH episodes. This indicates a significant improvement in patient outcomes (reduced IOH) when clinicians utilize the Acumen HPI Feature for guidance.
6. If a standalone (i.e. algorithm only, without human-in-the-loop performance) was done
The document describes the Acumen HPI Feature as a "decision support tool" providing "physiological insight" and "guidance" for clinicians. The primary effectiveness endpoint specifically evaluates the use of the Acumen HPI Feature to guide intraoperative hemodynamic management. This indicates that the study assessed human-in-the-loop performance, where clinicians were receiving and acting upon the device's output. There is no mention of a standalone algorithm-only performance study.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
The ground truth used for effectiveness endpoints was physiological measures and clinical outcomes data:
- Intraoperative Hypotension (IOH): Defined as Mean Arterial Pressure (MAP) < 65 mmHg for at least one minute (for the general definition) or for three or more consecutive 20-second events (for the study's specific IOH episode definition), which is directly measurable physiological data from arterial pressure monitoring.
- Area Under the Curve (AUC): Calculated from MAP values below 65 mmHg over time.
- Adverse Events/Complications: Such as cardiac arrest, death, stroke, AKI, and MINS, which are clinical outcomes recorded from patient data.
8. The sample size for the training set
The document does not specify the sample size or details of a training set for the Acumen Hypotension Prediction Index feature algorithm. It mentions that the "Acumen Hypotension Prediction Index (HPI parameter) algorithm as implemented on the HemoSphere Advanced Monitoring Platform (K201446, October 1, 2020) is identical to Acumen Hypotension Prediction Index (HPI parameter) algorithm granted in DEN160044." This suggests the algorithm itself was developed and cleared previously, and the current submission focuses on additional clinical data and labeling modifications.
9. How the ground truth for the training set was established
Since the document does not provide details of a training set for this specific submission, it also does not describe how the ground truth for any training set might have been established. Based on the nature of the device (predicting hypotension), it is highly likely that any prior training would have used large datasets of continuous arterial pressure waveforms and corresponding MAP values as ground truth for hypotensive events, similar to the definition used in the clinical study.
Ask a specific question about this device
(146 days)
The Edwards Lifesciences Acumen Hypotension Prediction Index feature provides the clinician with physiological insight into a patient's likelihood of future hypotensive events (defined as mean arterial pressure < 65 mmHg for at least one minute in duration) and the associated hemodynamics. The Acumen HPI feature is intended for use in surgical or non-surgical patients receiving advanced hemodynamic monitoring. The Acumen HPI feature is considered to be additional quantitative information regarding the patient's physiological condition for reference only and no therapeutic decisions should be made based solely on the Hypotension Prediction Index (HPI) parameter.
The Acumen Hypotension Prediction Index Feature (DEN160044) consists of software running on the Edwards Lifesciences EV1000 Clinical Platform (DEN160044) and HemoSphere Advanced Monitoring Platform (K180881) paired with the FloTrac IQ or Acumen IQ extravascular blood pressure transducer (K152980) and a radial arterial catheter. The software includes the Acumen Hypotension Prediction Index (HPI), the Dynamic Arterial Elastance Parameter (Eadyn), the Systolic Slope Parameter (dP/dt), and additional graphical user interface features. The Acumen Hypotension Prediction Index is an index related to the likelihood of a patient experiencing a hypotensive event (defined as mean arterial pressure < 65 mmHg for one minute in duration) within 15 minutes, where zero (0) indicates low likelihood and one hundred (100) indicates high likelihood. The Acumen Hypotension Prediction Index, HPI, should not be used exclusively to treat patients. A review of the patient's hemodynamics is recommended prior to initiating treatment.
Here's a breakdown of the acceptance criteria and study information for the Acumen Hypotension Prediction Index based on the provided text:
1. Acceptance Criteria and Reported Device Performance
The document does not explicitly present a table of acceptance criteria with specific numerical targets. However, the overall goal of the device, as described in the Indications for Use and Device Description, is to predict hypotensive events (MAP < 65 mmHg for at least one minute in duration) within 15 minutes. The Performance Data section states that "Clinical performance data were provided to demonstrate substantial equivalence of use of the Acumen™ Hypotension Prediction Index software."
The core performance attribute is the ability of the HPI to indicate the likelihood of a hypotensive event. The HPI values range from 0 (low likelihood) to 100 (high likelihood). While specific performance metrics like sensitivity, specificity, or AUC are not provided in this excerpt, the study aims to show that the performance in non-surgical patients is substantially equivalent to its established performance in surgical patients (as the predicate device, DEN160044, already had a grant decision).
| Acceptance Criteria | Reported Device Performance |
|---|---|
| Prediction of hypotensive events (MAP < 65 mmHg for at least one minute in duration) within 15 minutes in non-surgical patients. | Demonstrated "substantial equivalence of use" in non-surgical patients compared to surgical patients. (Specific metrics of prediction accuracy are not detailed in this document but would have been part of the original predicate device's clearance and likely referenced in the full submission). |
| HPI value range, with 0 indicating low likelihood and 100 indicating high likelihood. | The HPI device is designed to provide an index from 0 to 100 representing the likelihood. |
2. Sample Size Used for the Test Set and Data Provenance
The provided text does not specify the exact sample size for the test set. It only mentions that "Clinical performance data were provided."
Regarding data provenance:
- Country of origin of data: Not explicitly stated.
- Retrospective or prospective: Not explicitly stated.
3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications
The document does not specify the number of experts used or their qualifications for establishing ground truth.
4. Adjudication Method for the Test Set
The document does not specify any adjudication method.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
The document does not mention a Multi-Reader Multi-Case (MRMC) comparative effectiveness study. The focus of this 510(k) submission is on expanding the indications for use to non-surgical patients, demonstrating substantial equivalence to the predicate, rather than comparing AI-assisted performance against human readers.
6. Standalone (Algorithm Only) Performance Study
Yes, a standalone study was done. The Acumen HPI is described as software that "provides the clinician with physiological insight" and "is considered to be additional quantitative information...for reference only and no therapeutic decisions should be made based solely on the Hypotension Prediction Index (HPI) parameter." This implies that the algorithm generates an output (the HPI score) independently, and its clinical performance data (as mentioned under "Performance Data") refers to this standalone algorithm's ability to predict hypotension.
7. Type of Ground Truth Used
The ground truth used for identifying a hypotensive event is clearly defined: mean arterial pressure (MAP) < 65 mmHg for at least one minute in duration. This is a physiological measurement, not expert consensus or pathology.
8. Sample Size for the Training Set
The document does not specify the sample size for the training set. The submission is focused on demonstrating equivalence for an expanded indication, so information about the original training of the algorithm (which likely happened for the predicate device DEN160044) is not detailed here.
9. How Ground Truth for the Training Set Was Established
The document does not explicitly state how the ground truth for the training set was established, but it can be inferred that it would have used the same physiological definition of hypotension (MAP < 65 mmHg for at least one minute) that is used for the device's indications for use and performance evaluation. This would involve continuous monitoring of MAP in a patient population.
Ask a specific question about this device
(536 days)
The Edwards Lifesciences Acumen Hypotension Prediction Index (HPI) feature provides the clinician with physiological insight into a patient's likelihood of future hypotensive events (defined as mean arterial pressure < 65 mmHg for at least one minute in duration) and the associated hemodynamics. The Acumen HPI feature is intended for use in operating room (OR) patients receiving advanced hemodynamic monitoring. The Acumen HPI feature is considered to be additional quantitative information regarding the patient's physiological condition for reference only and no therapeutic decisions should be made based solely on the Hypotension Prediction Index (HPI) parameter.
The Acumen Hypotension Prediction Index Feature ("the device") consists of software running on the Edwards Lifesciences EV1000 Platform (previously cleared under K100709, K110597, K131892. K140312, and K160552) paired with the FloTrac IQ extravascular blood pressure transducer (K152980) and a radial arterial catheter. The device includes the Hypotension Prediction Index (HPI), the Dynamic Arterial Elastance Parameter (Eagyn), the Left Ventricular Contractility Parameter (dP/dt), and additional graphical user interface features.
HPI is an index related to the likelihood of a patient experiencing a hypotensive event (defined as mean arterial pressure (MAP) <65 mmHg for one minute in duration) within fifteen minutes, where zero (0) indicates low likelihood and one hundred (100) indicates high likelihood. The EV1000 Platform initiates a high priority alarm (red) when HPI exceeds 85. The Hypotension Prediction Index, HPI, should not be used exclusively to treat the patients. A review of the patient's hemodynamics is recommended prior to initiating treatment.
HPI is not defined by a single equation. HPI uses features extracted from FloTrac IQ measurements, some compared to an initial base value determined over the first 10 minutes of the patient monitoring session, to a data-driven model developed from retrospective analysis of an arterial waveform database collected from ICU and OR patients containing annotated hypotensive (defined as MAP <65 mmHG for at least 1 minute) and non-hypotensive events.
Eadm is a physiologic parameter related to left ventricular afterload relative to the left ventricular elastance. It is calculated by dividing Pulse Pressure Variation (PPV) by Stroke Volume Variation (SVV), or Eadyn = PPV/SVV.
dP/dt is a physiologic parameter related to the change in contractility of the left ventricle. It is calculated as the maximal first derivative of the arterial pressure waveform with respect to time.
Additional graphical user interface features include the HPI High Alert Popup, which displays when HPI exceeds 85 for two consecutive 20-second updates or reaches 100 at any time, and the HPI Secondary Screen, which consolidates hemodynamic parameters into a single screen.
1. Table of Acceptance Criteria and Reported Device Performance
| Metric | Acceptance Criteria (Implicit from Clinical Validation) | Reported Device Performance (N=52 Study) | Reported Device Performance (N=204 Study) |
|---|---|---|---|
| Sensitivity | High enough to be clinically useful | 83.7% [81.5, 86.0]% | 65.8% [63.7, 67.9]% |
| Specificity | High enough to avoid excessive false positives | 99.8% [99.4, 100.0]% | 99.4% [99.2, 99.7]% |
| AUC | High enough to indicate good discrimination | 0.95 | 0.88 |
| Positive Predictive Value (PPV) | (Not explicitly stated as AC, but evaluated) | 99.9% [99.7, 100.0]% | 98.3% [97.6, 99.0]% |
| Negative Predictive Value (NPV) | (Not explicitly stated as AC, but evaluated) | 75.1% [71.9, 78.4]% | 84.9% [83.9, 86.0]% |
Note: The document does not explicitly state numerical acceptance criteria for sensitivity, specificity, and AUC. However, the reported performance metrics from the clinical validation studies demonstrate a level of accuracy deemed acceptable by the FDA for de novo classification. The high specificities and AUC values, along with the detailed performance table for different HPI ranges, suggest that the device's ability to predict hypotension within the 15-minute timeframe was considered sufficient. The acceptance criteria for usability testing (at least 80% of participants agree or strongly agree) are explicitly stated in the Usability Testing section.
2. Sample Size Used for the Test Set and Data Provenance
The "test set" for the HPI algorithm's performance evaluation was derived from two retrospective patient databases:
- First Database (Edwards Lifesciences):
- Sample Size: 52 subjects (OR patients)
- Data Provenance: Global clinical sites, collected via prospective, IRB/EC approved clinical protocols with informed consent for each patient. (Retrospective analysis of prospectively collected data).
- Second Database (University Hospital):
- Sample Size: 204 subjects (OR patients)
- Data Provenance: From a university hospital, includes OR patients. (Retrospective analysis of an arterial waveform database).
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
The document does not specify the "number of experts" or their "qualifications" used to establish the ground truth for the test set.
Instead, the ground truth for hypotensive events was defined objectively: "mean arterial pressure (MAP) < 65 mmHg for at least one minute in duration." The databases used for validation contained "annotated hypotensive (defined as MAP <65 mmHG for at least 1 minute) and non-hypotensive events." This implies that the ground truth was established based on these objective physiological measurements rather than expert consensus on individual cases.
4. Adjudication Method for the Test Set
Not applicable. The ground truth was based on objective physiological measurements (MAP < 65 mmHg for at least one minute), not expert adjudication requiring consensus or multiple readings.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No, a multi-reader multi-case (MRMC) comparative effectiveness study was not reported. The device is intended as an adjunctive predictive cardiovascular indicator, providing quantitative information, not as a diagnostic aid that directly replaces or assists human readers in interpreting complex images or clinical scenarios requiring reader performance comparison. The study focused on the algorithm's standalone performance.
6. Standalone Performance Study (Algorithm Only Without Human-in-the-Loop Performance)
Yes, a standalone performance study was done. The reported sensitivity, specificity, and AUC values (83.7%, 99.8%, 0.95 for N=52 study; 65.8%, 99.4%, 0.88 for N=204 study) represent the performance of the HPI algorithm itself in predicting hypotensive events from extracted FloTrac IQ measurements. The device is explicitly stated as an "adjunctive predictive cardiovascular indicator" and "not intended to independently direct therapy," but its performance metrics are presented in a standalone context.
7. Type of Ground Truth Used
The type of ground truth used was objective physiological outcomes data, specifically:
- Hypotensive event: Defined as mean arterial pressure (MAP) < 65 mmHg for at least one minute in duration.
This objective definition was used to annotate the events in the retrospective arterial waveform databases.
8. Sample Size for the Training Set
The document states: "HPI uses features extracted from FloTrac IQ measurements... to a data-driven model developed from retrospective analysis of an arterial waveform database collected from ICU and OR patients containing annotated hypotensive... and non-hypotensive events."
However, it does not explicitly state the sample size of the training set. It only mentions that the model was "developed from retrospective analysis" of the overall databases, and the reported clinical validation results (N=52 and N=204 studies) were presumably used as test sets to evaluate the performance of this already developed model. The training and test sets might have been derived from the same larger pool of data.
9. How the Ground Truth for the Training Set Was Established
The ground truth for the model development (training) "was established" in the same manner as the validation ground truth:
- Hypotensive event: Defined as mean arterial pressure (MAP) < 65 mmHg for at least one minute in duration.
The training data came from "an arterial waveform database collected from ICU and OR patients containing annotated hypotensive (defined as MAP <65 mmHG for at least 1 minute) and non-hypotensive events." This indicates that the ground truth for both training and testing was based on these objective, physiologically defined annotations of hypotensive events.
Ask a specific question about this device
Page 1 of 1