Search Results
Found 1 results
510(k) Data Aggregation
(207 days)
The Dimension® EXL™ LOCI® BRAHMS Procalcitonin (PCT) assay is an in vitro diagnostic test for the quantitative measurement of procalcitonin in human serum and plasma (lithium heparin, K2EDTA, and K3EDTA) using the Dimension® EXL™ integrated chemistry system with LOCI® Module.
The Dimension EXL LOCI® BRAHMS PCT assay is intended for use in conjunction with other laboratory findings and clinical assessments, as an aid in:
· The risk assessment of critically ill patients on their first day of Intensive Care Unit (ICU) admission for progression to severe sepsis and septic shock.
• 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 using percentage change in PCT levels over time.
· Decision making on antibiotic therapy for patients 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) - in an inpatient setting or an emergency department.
· Decision making on antibiotic discontinuation for patients with suspected or confirmed sepsis.
The Dimension EXL LOCI BRAHMS PCT assay is a homogeneous sandwich chemiluminescent immunoassay based on LOCI technology. The LOCI reagents include two synthetic bead reagents and one biotinylated anti-procalcitonin (anti-PCT) monoclonal antibody. The first bead reagent (Sensibeads) is coated with streptavidin and contains photosensitizer dye. The second bead reagent (Chemibeads) is coated with two anti-PCT monoclonal antibodies and contains chemiluminescent dye. Sample is incubated with biotinylated antibody and Chemibeads to form bead-PCT-biotinylated antibody sandwiches. Sensibeads are added and bind to the biotin to form bead-pair immunocomplexes. Illumination of the complex at 680 nm generates singlet oxygen from Sensibeads which diffuses into the Chemibeads, triggering a chemiluminescent reaction. The resulting signal is measured at 612 nm and is a direct function of the procalcitonin (PCT) concentration in the sample.
The provided document describes the performance characteristics of the Dimension® EXL™ LOCI® BRAHMS Procalcitonin (PCT) assay and its equivalence to a predicate device.
Here's a breakdown of the acceptance criteria and study information:
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria Category | Specific Criteria | Reported Device Performance |
|---|---|---|
| Precision | Within-lab %CV: ≤ 10.0% for QC1, QC2, QC3, Serum 2, 3, 4, 5; ≤ 15.0% for Plasma, Serum 1. | Lot FB1218: QC1 (4.3%), QC2 (2.2%), QC3 (2.5%), Plasma (4.0%), Serum 1 (4.0%), Serum 2 (2.7%), Serum 3 (2.0%), Serum 4 (1.5%), Serum 5 (2.7%). All Pass. Lot FC1218: QC1 (3.3%), QC2 (4.2%), QC3 (2.6%), Plasma (4.0%), Serum 1 (3.0%), Serum 2 (2.7%), Serum 3 (2.7%), Serum 4 (2.1%), Serum 5 (2.3%). All Pass. |
| Total %CV: ≤ 10.0% for QC1, QC2, QC3, Serum 2, 3, 4, 5; ≤ 15.0% for Plasma, Serum 1. | Lot FB1218: QC1 (7.1%), QC2 (3.5%), QC3 (3.2%), Plasma (6.0%), Serum 1 (5.0%), Serum 2 (4.1%), Serum 3 (2.7%), Serum 4 (3.0%), Serum 5 (5.5%). All Pass. Lot FC1218: QC1 (6.2%), QC2 (4.5%), QC3 (3.2%), Plasma (6.0%), Serum 1 (6.0%), Serum 2 (3.6%), Serum 3 (3.2%), Serum 4 (3.1%), Serum 5 (5.8%). All Pass. | |
| Reproducibility | Total Reproducibility (CV%) for various PCT levels. (No explicit criteria mentioned in the document, but results are provided indicating good reproducibility). | MDP1 (0.10 ng/mL): 7.0%, MDP2 (0.25 ng/mL): 4.4%, MDP3 (0.48 ng/mL): 3.8%, MDP4 (1.95 ng/mL): 3.8%, MDP5 (8.94 ng/mL): 4.2%, MDP6 (41.01 ng/mL): 9.3%. |
| Detection Capability | LoB < LoD, LoD ≤ 0.04 ng/mL, LoQ ≤ 0.05 ng/mL (using within-lab %CV of 20%). | LoB: 0.03 ng/mL, LoD: 0.04 ng/mL, LoQ: 0.05 ng/mL. All claims met. |
| Linearity | Regression statistics (deviation from linearity for non-linear pools) at all levels tested demonstrated a ≤20% deviation (≤ 0.04 ng/mL for the lowest sample) from the predicted linear fit. | The study demonstrated a linear range of 0.03 to 55.05 ng/mL, which supports a measuring interval of 0.05 to 50.00 ng/mL. Results met the acceptance criteria. |
| Dilution Recovery | Recoveries for all specimens ranged from 88 to 108%. | Ten native serum specimens diluted 1:20 with saline. Recoveries ranged from 88% to 108% (mean 98%). Pass. Manual dilution supports an extended measuring interval from 50.00 ng/mL to 1000.00 ng/mL. |
| Interference (HIL) | ≤10% interference from hemoglobin, bilirubin, and lipemia. | Bilirubin (Conjugated/Unconjugated, 40 mg/dL): 0% or 1% bias. Hemoglobin (1000 mg/dL): -5% or -7% bias. Lipemia (2000 mg/dL): 0% or 3% bias. All met ≤10% criteria. |
| Non-Interfering Substances | Bias due to substances ≤10% at PCT analyte concentrations of 0.25 and 2.00 ng/mL. | Most substances tested (Acetaminophen, Acetylsalicylic acid, Amoxicillin, Azithromycin, Caffeine, Cefotaxime, Celecoxib, Cetirizine HCl, Cholesterol, Dextran 40, Dextromethorphan, Dobutamine, Dopamine, Doxycycline, EDTA, Epinephrine, Ethanol, Fentanyl, Fluorescein, Furosemide, HAMA (various concentrations, except high levels), Heparin, Human Immunoglobulin (IgG), Human serum albumin, Human serum gamma globulin, Ibuprofen, Imipenem, Levofloxacin, Loratadine, Nicotine, Noradrenaline, Oxymetazoline HCl, Phenylephrine, Prednisolone, Rheumatoid Factor, Salmeterol, Tiotropium, Triglycerides, Vancomycin) showed ≤10% bias. |
| Interfering Substances | Biotin >10% interference at >1200 ng/mL. HAMA >10% interference at 32.5 mg/mL (HAMA 1) and 65.0 mg/mL (HAMA 1 and 2), Total Protein >10% interference at 15.0 g/dL. | Biotin: >10% interference at 1500 ng/mL (-19% bias at 1.88 ng/mL PCT) and 3510 ng/mL (-26% to -29% bias). HAMA 1: -10% to -12% bias at 65.0 mg/mL and -11% to -12% bias at 32.5 mg/mL. HAMA 2: -14% to -13% bias at 65.0 mg/mL and -11% bias at 32.5 mg/mL (at 2.19 ng/mL PCT). Total Protein: -18% bias at 15.0 g/dL. These led to specific labeling cautions. |
| Cross-Reactivity | No explicit criterion provided, only results of testing. | No significant cross-reactivity observed with Calcitonin (Human, Eel, Salmon), Katacalcin (Human), α-CGRP, and β-CGRP. For example, Human Calcitonin at 8 ng/mL showed 0.00% to -0.50% cross-reactivity. |
| Hook Effect | No hook effect for PCT concentrations up to 2000.00 ng/mL. | For patient specimens with PCT concentrations between 50.00 ng/mL and 2000.00 ng/mL the assay will report results as "Above Assay Range" (> 50.00 ng/mL). |
| Sample Carryover | No sample carryover from high to low samples. | Calculated to be 0.00 ng/mL. No sample carryover observed. |
| Method Comparison | Strong correlation (r), passing Deming/Passing-Bablok regression, high percentage agreement at clinical cutoffs with predicate device. | Measuring Interval (0.05-50.00 ng/mL): Lot FB1218: r = 0.958, Slopes 1.07, Intercepts -0.01. Lot FC1218: r = 0.963, Slopes 1.04, Intercepts 0.00 to -0.01. Positive % agreement: 96.1-97.8%, Negative % agreement: 89.1-97.5%, Overall % agreement: 95.9-97.5% across various cut-offs (0.10, 0.25, 0.50, 2.00 ng/mL). Extended Measuring Interval (0.05-1000.00 ng/mL): Lot FB1218: r = 0.988, Slopes 1.08, Intercepts -0.01. Lot FC1218: r = 0.991, Slopes 1.05, Intercepts 0.00 to -0.01. Positive % agreement: 96.5-98.0%, Negative % agreement: 89.1-97.5%, Overall % agreement: 96.1-97.6% across various cut-offs. These results demonstrate substantial equivalence to the predicate device. |
| Matrix Comparison | No significant difference based on Passing-Bablok regression analysis between serum and plasma samples. | All specimen types (Serum (SST), Serum (RST), Lithium Heparin plasma, Sodium Heparin plasma, K2EDTA plasma, and K3EDTA plasma) showed high correlation coefficients (0.996-0.998) and regression equations close to y=x (slopes around 0.98-1.00, intercepts around 0.00-0.01 ng/mL) when compared to Serum (SST), indicating no significant difference. |
| Reference Interval Verification | Reference interval claim of <0.10 ng/mL verified. | The reference interval claim in the Instructions For Use is <0.10 ng/mL [<0.10 µg/L] verified using 33 apparently healthy individuals. |
| Reagent Stability | Real-time shelf life (12 months), Calibration interval (7 days), Unopened onboard stability (30 days), Opened onboard stability (3 days). | Shelf life: Two lots demonstrated 12 months of stability (ongoing). Calibration interval: Results support 7 days. Unopened onboard stability: Results support 30 days. Opened onboard stability: Results support 3 days. |
| Sample Stability | Storage and handling recommendations in IFU are supported (room temp 8h, refrig 24h, frozen -70C/90d -20C/90d, 6 freeze-thaw cycles, onboard 1h). | Studies supported all specified conditions for serum and plasma tube types (SST, RST, Li Hep, Na Hep, K2EDTA, K3EDTA) at various PCT levels, showing accurate results. |
2. Sample sizes for the test set and data provenance:
- Precision Studies: 80 replicates for each sample (QC and serum/plasma pools) for each lot (total 2 lots tested).
- Reproducibility Study: 225 replicates per sample ID (5 replicates/day x 5 days x 3 instruments x 3 reagent lots).
- Detection Capability: No specific sample size mentioned, CLSI EP17-A2 was referenced.
- Linearity: 16 samples.
- Dilution Recovery: 10 native serum specimens.
- Interference and Cross-Reactivity: Human serum pools, specifically "approximately 0.25 ng/mL and 2.00 ng/mL PCT". The exact number of individual samples is not explicitly stated.
- Hook Effect: Testing for PCT concentrations up to 2000.00 ng/mL. Number of samples not explicitly stated.
- Sample Carryover: 11 separate runs, each testing a pattern of high and low samples (21 tests per run).
- Method Comparison: 595 native human serum samples within the concentration range of 0.05-1000.00 ng/mL.
- Matrix Comparison: 76 matched sets (Serum, Lithium Heparin plasma, Sodium Heparin plasma, K2EDTA plasma, and K3EDTA plasma - except RST with 75 matched sets).
- Reference Interval Verification: Serum and plasma specimens from 33 apparently healthy individuals.
Data Provenance: The studies used human serum and plasma samples, as indicated by terms like "native human serum samples" and "patient-sourced internal standards." The document does not specify the country of origin, but it mentions Siemens Healthcare Diagnostics, Inc. in Newark, Delaware, USA, as the applicant. The nature of the studies (e.g., method comparison, stability) suggests these are retrospective analyses of collected samples or controlled laboratory experiments. It does not explicitly state prospective or retrospective data collection for all studies, but method comparison using "native human serum samples" implies existing clinical samples from a test set.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
This device is an in vitro diagnostic (IVD) test for quantitative measurement of procalcitonin (PCT). The "ground truth" here is not established by human readers/experts in the same way as, for example, a diagnostic imaging AI. Instead, the ground truth for this type of device is based on:
- Reference Method/Predicate Device: For method comparison, the predicate device (B·R·A·H·M·S PCT sensitive KRYPTOR) serves as the reference for comparison.
- Analytical Standards: For studies like linearity, detection capability, and precision, the "ground truth" or true value of PCT is established using known concentrations in control materials, calibrators, or by spiking samples with recombinant PCT. These are typically prepared and validated by qualified laboratory professionals following established analytical chemistry principles and CLSI guidelines.
- Defined Clinical Cut-offs: While the device aids in decision-making, the clinical cut-offs (e.g., 0.10 ng/mL, 0.25 ng/mL) are established within medical guidelines and clinical practice, not by individual experts for the purpose of validating this specific device's performance.
Therefore, the concept of "number of experts and their qualifications for establishing ground truth" as it applies to image interpretation AI is not directly applicable here. The "experts" would be the scientists and laboratorians who designed and executed the analytical validation studies according to CLSI guidelines and robust quality control practices.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:
Adjudication methods like 2+1 or 3+1 are typically used in studies involving human readers interpreting diagnostic images or clinical cases where disagreement needs to be resolved by a tie-breaker. This is an in vitro diagnostic assay with quantitative results. Therefore, adjudication in that sense is not applicable or done. The performance is assessed against quantitative metrics, reference methods, or predicate devices.
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:
An MRMC comparative effectiveness study is not applicable for this device. This is a standalone in vitro diagnostic assay that provides a quantitative measurement of a biomarker (PCT) in clinical specimens. It does not involve human readers interpreting cases with or without AI assistance, nor does it involve improving human reader performance.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:
Yes, the studies presented are all standalone performance studies for the Dimension® EXL™ LOCI® BRAHMS Procalcitonin (PCT) assay. The device measures PCT levels directly from patient samples, and its performance characteristics (precision, linearity, detection capability, interference, method comparison, etc.) are evaluated intrinsically without direct human interpretation influencing the measurement outcome. The role of the human is in operating the instrument, collecting samples, and interpreting the numerical results in a clinical context.
7. The type of ground truth used (expert concensus, pathology, outcomes data, etc):
The ground truth used for this quantitative in vitro diagnostic device is primarily based on:
- Assigned Values of Calibrators and Control Materials: For accuracy, precision, and linearity studies, known concentrations of PCT in calibrators and quality control materials serve as the ground truth. These values are established through rigorous analytical methods and traceability chains.
- Reference Method/Predicate Device Results: In the method comparison study, the results from the legally marketed predicate device (B·R·A·H·M·S PCT sensitive KRYPTOR) are used as the reference against which the candidate device's performance is compared to establish substantial equivalence.
- Spiking with Recombinant Procalcitonin: For studies like dilution recovery, interference, and cross-reactivity, samples are often "spiked" with known concentrations of highly purified recombinant procalcitonin (or interfering substances) to create samples with a known "true" concentration of the analyte for testing the assay's behavior.
8. The sample size for the training set:
This document describes the validation of an in vitro diagnostic assay, not a machine learning or AI algorithm in the context of typical "training" and "test" sets. Therefore, there isn't a "training set" in the sense of data used to train a predictive model. The development of such assays involves extensive R&D and analytical optimization, but a specific "training set sample size" as asked in the context of AI is not applicable.
9. How the ground truth for the training set was established:
As mentioned above, there is no "training set" in the AI/ML context. The development and optimization of an IVD assay like this involve:
- Immunoassay Design: Selection of specific antibodies (monoclonal anti-PCT antibodies) that bind to the analyte.
- Reagent Formulation: Optimizing concentrations of reagents (sensibeads, chemibeads, biotinylated antibody, assay buffer) and reaction conditions to achieve desired sensitivity, specificity, and dynamic range.
- Analytical Verification Studies: These studies (like those described in the document, e.g., precision, linearity, detection capability) are used iteratively during development to ensure the assay performs as intended. The "ground truth" during this phase would be known concentrations of purified analyte, reference materials, and comparison to established methods or predicate devices.
The "ground truth" is thus established through fundamental analytical chemistry principles, quantitative measurements, and comparison to internationally recognized standards or well-characterized reference materials and methods.
Ask a specific question about this device
Page 1 of 1