FDA 510(k) Search - By Innolitics (SaMD and AI Experts)

The B-R-A-H-M-S PCT sensitive KRYPTOR® is an immunofluorescent assay using Time-Resolved Amplified Cryptate Emission (TRACE) technology to determine the concentration of PCT (procalcitonin) in human serum and EDTA or heparin plasma.
The B-R-A-H-M-S PCT sensitive KRYPTOR® is intended to be performed on the B·R·A·H·M·S KRYPTOR® analyzer family.
Used in conjunction with other laboratory findings and clinical assessments, B·R·A·H·M·S PCT sensitive KRYPTOR® is intended for use as follows:

to aid in the risk assessment of critically ill patients on their first day of ICU admission for progression to severe sepsis and septic shock,
to determine the change in PCT level over time as an aid in assessing the cumulative 28-day risk of all-cause mortality for patients diagnosed with severe sepsis or septic shock in the ICU or when obtained in the emergency department or other medical wards prior to ICU admission,
to aid in decision making on antibiotic therapy, for inpatients or patients in the emergency department with suspected or confirmed lower respiratory tract infections (LRTI) - defined as community-acquired pneumonia (CAP), acute bronchitis, and acute exacerbation of chronic obstructive pulmonary disease (AECOPD),
to aid in decision making on antibiotic discontinuation for patients with suspected or confirmed sepsis.

Device Description

The B·R·A·H·M·S KRYPTOR® compact PLUS analyzer is a fully automated system. The B-R-A-H-M-S KRYPTOR® compact PLUS analyzer is a closed system and can only operate utilizing special reagents provided by B.R.A.H.M.S GmbH.
The B·R·A·H·M·S PCT sensitive KRYPTOR® is a homogeneous sandwich immunoassay for detection of PCT in human serum or plasma. The measuring principle is based on Time-Resolved Amplified Cryptate Emission (TRACE®) technology, which measures the signal that is emitted from an immunocomplex with time delay.

AI/ML Overview

The provided text describes a 510(k) premarket notification for the B·R·A·H·M·S PCT sensitive KRYPTOR® device, which measures procalcitonin levels. The submission seeks clearance for expanded indications for use based on comparisons to predicate devices and meta-analyses of clinical studies.

Here's an analysis of the acceptance criteria and the study that proves the device meets them, based on the provided document:

1. Table of Acceptance Criteria and Reported Device Performance

The document doesn't explicitly state "acceptance criteria" in a tabular format with corresponding performance results for each expanded indication in a single place. Instead, it details various analytical and clinical performance characteristics from previous submissions (DEN150009 and K070310) and new meta-analyses to support the expanded claims.

However, we can infer acceptance criteria from the context of a 510(k) submission, which generally requires demonstrating substantial equivalence to a legally marketed predicate device. This often involves showing comparable analytical performance and equivalent safety and effectiveness for the intended use. For the expanded indications, the performance is demonstrated through meta-analyses of existing clinical trial data.

Inferred "Acceptance Criteria" (based on regulatory expectations for an IVD and expanded claims) and Reported Device Performance:

Acceptance Criteria (Inferred)	Reported Device Performance (from sections M.1 and M.2)
Analytical Performance (from previous clearances, largely not re-evaluated for this submission)
Precision (Reproducibility)	Internal Precision (from DEN150009) shown in table on Page 15:

%CV values for repeatability ranged from 0.58% to 12.31%.
Total Reproducibility %CV ranged from 2.57% to 14.88%.

(The table provides detailed SD and %CV for various components like repeatability, between-operator, between-day, between-calibration, between-run, and between-lot for 12 different samples (P10-P16, QC1, QC2, HG1-HG3) with N=56 for most samples and N=54 for HG3.) |
| Linearity/Assay Reportable Range | From DEN150009 (Page 16):

Direct measuring range: 0.02 µg/L - 50 µg/L
Measuring range with automatic dilution: 0.02 µg/mL - 5000 µg/L |
| Detection Limit (LoB, LoD, LoQ) | From DEN150009 (Page 17):
LoB and LoD determined (same as predicate).
LOQ (lowest reported concentration level with bias ≤ 5%, % CV ≤ 15%, TE ≤ 30%) = 0.075 µg/L.
TE ≤ 20% at 0.25 µg/L (bias ≤ 5%, precision CV ≤ 10%)
TE ≤ 30% at 0.10 µg/L (bias ≤ 5%, precision CV ≤ 15%)
Performance at key cutoffs (table on Page 17):
0.10 µg/L: %CV 10.33, %BIAS 2.07, %TE 19.11
0.23 µg/L: %CV 5.04, %BIAS 1.85, %TE 10.17
0.27 µg/L: %CV 6.71, %BIAS 0.76, %TE 11.83 |
| Analytical Specificity/Interference | From DEN150009; supplementary interference studies performed for lower cut-offs (Page 17-18):
No interference up to specified concentrations for numerous endogenous substances (e.g., Hemoglobin 500 mg/dL, Triglycerides 22.5 mg/mL, Bilirubin 20mg/dL), cross-reacting substances (e.g., Human calcitonin 3.9 ng/mL), and drugs (e.g., Cefotaxim 90 mg/dL, Heparin 8000 IU/L, common asthma/COPD drugs like Budesonide, Albuterol, etc.). |
| Method Comparison (vs. Predicate) | Qualitative Agreement with VIDAS B·R·A·H·M·S PCT (PCT) on 203 samples (Page 19):
At 0.10 µg/L: Positive Agreement 86.5%, Negative Agreement 86.8%, Overall Agreement 86.7%, Kappa 0.7309.
At 0.25 µg/L: Positive Agreement 96.4%, Negative Agreement 98.0%, Overall Agreement 97.5%, Kappa 0.9380.
At 0.50 µg/L: Positive Agreement 95.6%, Negative Agreement 100.0%, Overall Agreement 99.0%, Kappa 0.9710.
At 2.00 µg/L: Positive Agreement 79.2%, Negative Agreement 100.0%, Overall Agreement 97.5%, Kappa 0.8702. |
| Clinical Performance (supporting expanded claims via Meta-analyses) |
| Aid in decision-making on antibiotic therapy for LRTI (reduce antibiotic use without negative outcomes) | Patient-level meta-analysis (13 RCTs, N=3142 total patients) for LRTI (Page 21):
19.2% reduction in relative antibiotic initiation.
38% reduction in overall antibiotic exposure (inpatients).
51% reduction in overall antibiotic exposure (ED/outpatients).
2.9 day reduction in antibiotic duration.
3.6 day reduction in total antibiotic exposure.
No negative effects in regards to mortality, complications, or length of stay.
Summary table (Page 21):
Antibiotic initiation: 88.4% (standard) vs. 71.4% (PCT guided)
Duration of antibiotics: 10 days (standard) vs. 7 days (PCT guided)
Total exposure of antibiotics: 9 days (standard) vs. 5 days (PCT guided)
30-day mortality: 7.4% (standard) vs. 6.7% (PCT guided)
Complications: 21.1% (standard) vs. 18.0% (PCT guided) |
| Aid in decision-making on antibiotic discontinuation for sepsis (reduce antibiotic use without negative outcomes) | Patient-level meta-analysis for Sepsis (5 RCTs, N=598 sepsis patients) (Page 22):
1.5 day reduction in antibiotic duration.
3.2 day reduction in total antibiotic exposure.
23% reduction in overall antibiotic exposure.
No negative effects in regards to mortality, hospital length of stay, or ICU length of stay.
Summary table (Page 22):
Total exposure of antibiotics: 12 days (standard) vs. 8 days (PCT guided)
30-day mortality: 23.8% (standard) vs. 19.9% (PCT guided) |

2. Sample Size Used for the Test Set and Data Provenance

Analytical Performance Test Set (Method Comparison):
- Sample Size: 203 frozen banked samples.
- Data Provenance: Retrospective. Samples were from the "ProRESP trial bank of consecutive patients with clinically suspected COPD, acute bronchitis and CAP," analyzed for concordance with a predicate device.
- Country of Origin: Not explicitly stated but the publication referenced, "Schuetz P, et al. Clin Biochem. 2010," suggests European origin (Switzerland is common for Müller and Schuetz research groups).
Clinical Performance Test Set (Meta-analyses for expanded indications):
- LRTI Antibiotic Decision Making:
  - Study-level Meta-analysis: 11 Randomized Controlled Trials (RCTs), 4090 patients.
  - Patient-level Meta-analysis: 13 RCTs, 3142 patients.
  - Data Provenance: Retrospective, aggregated from previously published RCTs.
  - Country of Origin: Not specified, but given the list of publications, these would be multi-national clinical trials.
- Sepsis Antibiotic Discontinuation:
  - Study-level Meta-analysis: 10 RCTs, 3489 patients.
  - Patient-level Meta-analysis: 5 RCTs, 598 patients (after excluding non-sepsis patients).
  - Data Provenance: Retrospective, aggregated from previously published RCTs.
  - Country of Origin: Not specified, but likely multi-national.

3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications

Analytical Performance (Method Comparison): The "ground truth" here is the measurement by the predicate device (VIDAS B·R·A·H·M·S PCT). No human experts were involved in establishing the ground truth for this direct measurement comparison.
Clinical Performance (Meta-analyses): For PCT-guided therapy trials, the ground truth for patient outcomes (e.g., diagnosis of LRTI, sepsis, mortality, complications, length of stay, antibiotic duration) would have been established by the clinical teams and investigators involved in each of the original RCTs. The document does not specify the number or qualifications of these individual experts for the initial studies, as it relies on published, peer-reviewed clinical trial data as its basis. The meta-analyses themselves are statistical syntheses of these existing ground truths.

4. Adjudication Method for the Test Set

Analytical Performance: Not applicable for a quantitative assay method comparison. Results are compared directly between two devices.
Clinical Performance: For the individual RCTs that form the basis of the meta-analyses, adjudication methods for clinical endpoints would have been defined in their original protocols. The meta-analyses themselves don't involve a separate adjudication process beyond the data synthesis. The text states "Each meta-analysis used random-effects models and calculated point estimates, differences, odds ratios (OR), interquartile ranges (IQRs) and 95% confidence intervals as appropriate," which describes the statistical method, not human adjudication.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

This is an In Vitro Diagnostic (IVD) device (a blood test for procalcitonin), not an imaging AI device. Therefore, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study, which is typically relevant for evaluating medical imaging interpretation (human readers with/without AI assistance), was not conducted. The "comparative effectiveness" demonstrated here is about the utility of the PCT biomarker (measured by the device) in guiding clinical decisions (like antibiotic therapy) when compared to standard care, based on clinical trial outcomes.

6. Standalone Performance

The device is an in vitro diagnostic (IVD) that provides a quantitative measurement of PCT. Its "standalone performance" is implicitly covered by the analytical performance studies (precision, linearity, detection limit, interference). It's not an "algorithm only" device in the sense of an AI interpreting medical images; it's a lab instrument that measures an analyte. Its performance is the accurate measurement of PCT levels.

The "standalone" statement regarding its clinical use (Page 6) is a warning/precaution: "B·R·A·H·M·S PCT sensitive KRYPTOR® is not indicated to be used as a stand-alone diagnostic assay and should be used in conjunction with clinical signs and symptoms of infection and other diagnostic evidence." This clarifies that the clinical decision-making should not solely rely on PCT even if the device itself accurately measures PCT.

7. Type of Ground Truth Used

Analytical Performance: The ground truth for the method comparison was the quantitative result provided by a legally marketed predicate device (VIDAS B·R·A·H·M·S PCT).
Clinical Performance: The ground truth for the expanded indications was based on outcomes data and clinical diagnoses established in prospective, randomized controlled trials. These outcomes include antibiotic initiation and duration, total antibiotic exposure, 30-day mortality, complications, and length of hospital/ICU stay.

8. Sample Size for the Training Set

The document does not directly mention a "training set" in the context of machine learning or AI algorithm development, as this is an IVD device measuring a biomarker. The analytical performance data (precision, linearity, detection limits, interference) are part of its fundamental characterization, not typically referred to as a "training set."

For the meta-analyses, the "training" analogous to that for an AI would be the collective body of clinical evidence from the published RCTs. The data from various studies (Ns provided in point 2) are "pooled" or combined within the meta-analysis framework. It's not a single "training set" for a new algorithm, but rather a synthesis of existing clinical evidence to support a new clinical claim for an established diagnostic test.

9. How the Ground Truth for the Training Set Was Established

Since there isn't a "training set" for a machine-learning algorithm in this context, the question translates to: How were the original clinical trial data (which underpin the meta-analyses) established?

The clinical data used in the meta-analyses (serving as the "evidence base" for the expanded indications) were derived from multi-center, prospective, randomized controlled clinical trials (RCTs). In these trials, patient outcomes (e.g., diagnosis, mortality, antibiotic use) were carefully collected and documented by the individual study investigators according to their original study protocols. This type of data from well-designed RCTs is considered high-quality evidence in medical research. The meta-analyses then systematically combined and statistically analyzed the results from these individual studies, effectively using their established "ground truths."

Summary

Regulation Number and Section

§ 866.3215 Device to detect and measure non-microbial analyte(s) in human clinical specimens to aid in assessment of patients with suspected sepsis.

(a)
Identification. A device to detect and measure non-microbial analyte(s) in human clinical specimens to aid in assessment of patients with suspected sepsis is identified as an in vitro device intended for the detection and qualitative and/or quantitative measurement of one or more non-microbial analytes in human clinical specimens to aid in the assessment of patients with suspected sepsis when used in conjunction with clinical signs and symptoms and other clinical and laboratory findings.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Premarket notification submissions must include the device's detailed Indications for Use statement describing what the device detects and measures, the results provided to the user, whether the measure is qualitative and/or quantitative, the clinical indications for which the test is to be used, and the specific population(s) for which the device use is intended.
(2) Premarket notification submissions must include detailed documentation of the device description, including (as applicable), all device components, software, ancillary reagents required but not provided, explanation of the device principle and methodology, and for molecular devices include detailed documentation of the primer/probe sequence, design, and rationale for sequence selection.
(3) Premarket notification submissions must include detailed documentation of applicable analytical studies, such as, analytical sensitivity (Limit of Detection, Limit of Blank, and Limit of Quantitation), precision, reproducibility, analytical measuring range, interference, cross-reactivity, and specimen stability.
(4) Premarket notification submissions must include detailed documentation of a prospective clinical study or, if appropriate, results from an equivalent sample set. This detailed documentation must include the following information:
(i) Results must demonstrate adequate device performance relative to a well-accepted comparator.
(ii) Clinical sample results must demonstrate consistency of device output throughout the device measuring range likely to be encountered in the Intended Use population.
(iii) Clinical study documentation must include the original study protocol (including predefined statistical analysis plan), study report documenting support for the Indications for Use(s), and results of all statistical analyses.
(5) Premarket notification submissions must include evaluation of the level of the non-microbial analyte in asymptomatic patients with demographic characteristics (
e.g., age, racial, ethnic, and gender distribution) similar to the Intended Use population.(6) As part of the risk management activities performed under 21 CFR 820.30 design controls, you must document an appropriate end user device training program that will be offered as part of your efforts to mitigate the risk of failure to correctly operate the instrument.
(7) A detailed explanation of the interpretation of results and acceptance criteria must be included in the device's 21 CFR 809.10(b)(9) compliant labeling, and a detailed explanation of the interpretation of the limitations of the samples (
e.g., collected on day of diagnosis) must be included in the device's 21 CFR 809.10(b)(10) compliant labeling.