Search Results
Found 1 results
510(k) Data Aggregation
(63 days)
The SenDx 100™ pH, Blood Gas and Electrolyte Analysis System (SenDx 100™) is intended to measure blood oxygen, carbon dioxide, pH, sodium, potassium, chloride, ionized calcium and hematocrit in whole blood.
The SenDx 100™ system can be used in the laboratory, near patient, or in point-of-care settings. It is intended for the measurement of blood gases, electrolytes and hematocrit in arterial or venous whole blood samples.
The SenDx 100™ system consists of the analyzer, a multi-use, disposable sensor cassette and a calibration solution pack. A bar code label containing lot and sensor-specific identifying parameters accompanies each new sensor cassette and calibration solution pack.
The SenDx 100™ line is available in two versions:
Model SD 100 B with modem
Model SD 100 IB without modem for international use
The SD100B is sold in the United States and is UL certified. The unit is battery powered and requires an external battery charger that is provided by SenDx Medical, Inc. Use only Jerome Industries power supply model WSZ118M which requires 100 to 250 VAC, 50 to 60 Hertz with these units. It also has a modem.
The SD1001B is sold internationally and is certified by TUV Rheinland. This unit is also battery powered and requires an external battery charger that is provided by SenDx Medical, Inc. Use only Advance Power Solutions power supply model AD-740U-1180 which requires 100 to 240 VAC, 50 to 60 Hertz with these units. This model is not equipped with a modem.
The SenDx 100™ analyzer utilizes microprocessor controlled electronics. The microprocessor controls the touch screen display, the analog electronics for processing sensor output, and the integral thermal printer. The printer provides a hard copy of the measured and calculated values. Samples are introduced into the analyzer via an aspiration stylus.
The SenDx 100™ sensor incorporates microelectrode technology for the measurement of blood oxygen (pO2), carbon dioxide (pCO2), pHx, sodium (Na+), potassium (K+), chloride (Cl-), ionized calcium (iCa++), and hematocrit (Hct) in a multi-use, disposable cassette assembly.
The sensors are analogous to traditional electrode methodologies for the measurement of blood gases and electrolytes. A Clark cell for pO2 measurement and potentiometric ion-selective electrodes for the measurement of pH, pCO2, Na+, K+, Cl- and iCa++ have been miniaturized for placement on a 1-3/8" x 3/8" sensing "chip". Hematocrit is measured by using conductivity.
The sensors are contained in a low volume flow-through cell. The sensor cassette contains reference electrodes for the potentiometric sensors and an integral temperature sensor and heating element for precise temperature control. Calibration and analysis are carried out at 37.0°C. The SenDx 100™ analyzer allows for patient temperature correction over the range of 12-45°C.
All SenDx 100™ sensor cassettes are 100% tested at the factory with multiple precision reference solutions.
Calibration of the system is accomplished using a calibration solution pack which contains two levels of precision tonometered electrolyte solutions packaged in gas tight disposable cartridges. Precise pH, blood gas, and electrolyte calibration values for each pack are provided on an attached calibration bar code label.
The SenDx 100TM performs two point calibrations at automatic intervals. Throughout the calibration and sample analysis phases, sensor signals are continuously monitored. If any abnormal conditions are detected, an error message is generated.
Blood analysis and the subsequent system calibration and flush take approximately 50-75 seconds. After analysis, pH, pCO2, pO2, Na+, K+, Cl-, iCa++, and Hct measurement results are displayed on the analyzer touch screen. The measured values and derived values (HCO3-, BEb, BEecf, TCO3, SBC, Hb, and %O2 Sat) are automatically printed.
The SenDx 100™ system can be interfaced via the RS232 serial data port with PC based or LIS/HIS data management systems.
Here's a breakdown of the acceptance criteria and the study information regarding the SenDx 100 Blood Gas and Electrolyte Analysis System, based on the provided document:
1. Table of Acceptance Criteria and Reported Device Performance
The document doesn't explicitly state "acceptance criteria" in a numerical or categorical format with pass/fail thresholds. Instead, it demonstrates the device's performance through correlation and precision studies against predicate devices and known control values. The implicit acceptance criterion is that the device demonstrates performance substantially equivalent to these predicate devices or acceptable precision for clinical use.
Here's a summary of the reported performance, categorized by study type:
A. Linearity of Recovery (Internal Study)
(Comparison against Corning 850 for pH, pCO2, pO2, Na+, K+, Cl-, iCa++; spun hematocrit for Hct)
| Analyte | n | Slope | r | Intercept | Std Error Sy.x |
|---|---|---|---|---|---|
| pH | 333 | 1.21 | 0.99 | -1.56 | 0.017 |
| pCO2 | 384 | 1.00 | 0.99 | 0.92 | 1.83 |
| pO2 | 321 | 1.01 | 1.00 | 0.51 | 3.0 |
| Na+ | 465 | 1.09 | 0.99 | -11.17 | 2.15 |
| K+ | 475 | 1.06 | 1.00 | -0.18 | 0.13 |
| Cl- | 414 | 1.03 | 0.99 | 1.10 | 1.66 |
| iCa++ | 600 | 1.01 | 1.00 | -0.04 | 0.04 |
| Hct | 747 | 1.12 | 0.98 | -2.42 | 1.69 |
B. Correlations (Internal Study)
(Comparison against Corning 850 and spun hematocrit using tonometered and spiked whole blood)
| Analyte | n | Slope | r | Intercept | Std Error Sy.x |
|---|---|---|---|---|---|
| pH | 472 | 1.23 | 0.99 | -1.65 | 0.013 |
| pCO₂ | 413 | 1.00 | 0.99 | 0.49 | 1.62 |
| pO₂ | 1677 | 0.97 | 0.99 | 3.81 | 5.56 |
| Na⁺ | 619 | 1.17 | 0.99 | -24.22 | 2.11 |
| K⁺ | 601 | 1.02 | 1.00 | 0.07 | 0.14 |
| Cl⁻ | 619 | 1.01 | 0.99 | -0.10 | 1.54 |
| iCa⁺⁺ | 621 | 1.33 | 0.99 | -0.4 | 0.05 |
| Hct | 1229 | 1.12 | 0.95 | -4.63 | 2.07 |
C. Precision (Internal Study) - Using RNA Medical Qualidata and Hct Levels
(Summary of Mean, S.D., %CV)
| Analyte | Mean | S. D. | %CV |
|---|---|---|---|
| pH | |||
| Level 1 | 7.177 | 0.006 | 0.089 |
| Level 2 | 7.416 | 0.003 | 0.047 |
| Level 3 | 7.612 | 0.006 | 0.073 |
| pCO₂ | |||
| Level 1 | 67.5 | 2.1 | 3.1 |
| Level 2 | 45.0 | 0.7 | 1.7 |
| Level 3 | 24.7 | 0.7 | 2.8 |
| pO₂ | |||
| Level 1 | 71.4 | 3.0 | 4.3 |
| Level 2 | 109.5 | 2.0 | 1.9 |
| Level 3 | 146.0 | 1.5 | 1.0 |
| Na⁺ | |||
| Level 1 | 107.6 | 0.7 | 0.6 |
| Level 2 | 130.1 | 0.5 | 0.4 |
| Level 3 | 157.4 | 0.6 | 0.4 |
| K⁺ | |||
| Level 1 | 1.85 | 0.02 | 1.30 |
| Level 2 | 4.39 | 0.03 | 0.78 |
| Level 3 | 6.46 | 0.06 | 0.92 |
| iCa⁺⁺ | |||
| Level 1 | 1.55 | 0.02 | 1.32 |
| Level 2 | 1.19 | 0.01 | 0.97 |
| Level 3 | 0.54 | 0.01 | 2.03 |
| Hct | |||
| Low | 35.0 | 0.3 | 0.7 |
| High | 58.2 | 0.5 | 0.8 |
| Cl⁻ | |||
| Level 1 | 74.0 | 1.1 | 1.5 |
| Level 2 | 94.0 | 0.5 | 0.6 |
| Level 3 | 127.0 | 1.0 | 0.8 |
D. Clinical Study - Hospital Stat Lab (Representative Site)
(Comparison against Corning 865, Nova Stat Profile 5, Corning Co-oximeter, and spun hematocrit)
| Analyte | n | Slope | r | Intercept | Std Error Sy.x |
|---|---|---|---|---|---|
| pH | 120 | 0.89 | 0.99 | 0.841 | 0.013 |
| pCO2 | 114 | 0.84 | 0.99 | 5.93 | 1.03 |
| pO2 | 119 | 1.02 | 1.00 | -0.08 | 5.77 |
| Na+ | 120 | 0.98 | 0.89 | 3.04 | 2.48 |
| K+ | 120 | 1.01 | 0.99 | -0.15 | 0.07 |
| iCa++ | 118 | 0.85 | 0.94 | 0.17 | 0.03 |
| Hct | 102 | 1.11 | 0.93 | -4.03 | 2.55 |
E. Clinical Study - Hospital Operating Room Point-Of-Care (Representative Site)
(Comparison against Mallinckrodt Hem-Premier)
| Analyte | n | Slope | r | Intercept | Std Error Sy.x |
|---|---|---|---|---|---|
| pH | 112 | 0.78 | 0.99 | 1.66 | 0.01 |
| pCO2 | 106 | 0.80 | 0.96 | 4.96 | 1.56 |
| pO2 | 108 | 0.79 | 0.95 | 54 | 28.5 |
| Na+ | 112 | 0.85 | 0.90 | 17.1 | 1.98 |
| K+ | 112 | 1.06 | 0.98 | -0.4 | 0.20 |
| iCa++ | 97 | 0.88 | 0.95 | 0.12 | 0.06 |
| Hct | 111 | 1.03 | 0.99 | -2.35 | 1.85 |
F. Clinical Precision Study (Site to Site)
(Summarized data from four clinical sites for pH, blood gases, electrolytes and hematocrit)
- pH: Site to Site* Mean ranges from 7.192 to 7.602, S.D. 0.006-0.008, %CV 0.075-0.102.
- pCO2: Site to Site* Mean ranges from 23.3 to 66.4, S.D. 0.385-1.883, %CV 1.653-2.835.
- pO2: Site to Site* Mean ranges from 68.2 to 143.1, S.D. 1.328-2.546, %CV 1.236-3.667.
- Na+: Site to Site* Mean ranges from 107.4 to 156.2, S.D. 1.176-2.133, %CV 0.753-1.987.
- K+: Site to Site* Mean ranges from 1.81 to 6.43, S.D. 0.019-0.047, %CV 0.728-1.055.
- iCa++: Site to Site* Mean ranges from 0.49 to 1.52, S.D. 0.008-0.045, %CV 1.681-2.959.
- Hct: Site to Site* Mean ranges from 34.2 to 57.3, S.D. 0.546-0.797, %CV 1.391-1.594.
The conclusion states that the performance is "substantially equivalent" to predicate devices, implying these reported values meet the unstated acceptance criteria for equivalency.
2. Sample Size Used for the Test Set and Data Provenance
-
Non-Clinical Testing (Linearity and Correlations):
- Sample Sizes: For linearity, individual analyte 'n' values range from 321 (pO2) to 747 (Hct). For correlations, 'n' values range from 413 (pCO2) to 1677 (pO2). These represent the number of samples tested for each analyte.
- Data Provenance: The data was generated on-site at SenDx Medical, Inc. The samples used were "pooled whole blood" for linearity of pH, K+, Na+, Cl-, iCa++, and "tonometered and spiked whole blood" for correlations. This indicates the data is retrospective in nature, likely laboratory-controlled samples, not from real patients.
-
Clinical Testing:
- Sample Sizes: Over 400 split-samples were evaluated across four clinical sites. Individual analyte 'n' values for representative sites range from 97 (iCa++ in OR POC) to 120 (pH, Na+, K+ in Stat Lab). This indicates the number of patient samples.
- Data Provenance: Conducted at four different geographical sites (hospitals: Stat Lab and Operating Room Point-Of-Care). These were prospective clinical studies using patient samples ("split-samples"). The country of origin is not explicitly stated but implied to be the US given the submission to the FDA.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
The document does not mention the use of "experts" to establish ground truth in the typical sense of human readers for images or diagnoses. Instead, the ground truth was established by measurements from established, legally marketed predicate devices (e.g., Corning 850, Corning 865, Nova Stat Profile 5, Mallinckrodt Gem-Premier, and spun hematocrit).
The qualifications of operators for these predicate devices or for collecting calibration data are not specified, but it's implied they are trained laboratory personnel ("trained technologists, nurses, physicians, and therapists" as per the SenDx 100's intended use).
4. Adjudication Method for the Test Set
Not applicable. The ground truth was established by instrumental measurements from predicate devices, not by human interpretation requiring adjudication.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No. This type of study is designed to assess the impact of an AI algorithm on human reader performance, typically for diagnostic imaging interpretation. This device is a blood gas and electrolyte analyzer, and its output is a direct quantitative measurement, not an interpretation by a human reader.
6. Standalone (Algorithm Only Without Human-in-the-Loop) Performance
Yes, this entire submission is essentially a standalone performance study. The SenDx 100 system itself is the "algorithm" (or rather, the automated analytical device) that measures the parameters. Its performance is compared directly against established laboratory reference methods and other commercially available automated analyzers without human intervention in the measurement process itself, beyond sample introduction and standard operating procedures. The data presented are the results from the SenDx 100 as an automated system.
7. Type of Ground Truth Used
The ground truth for both non-clinical and clinical testing was established by:
- Measurements from Predicate Devices: For all analytes, readouts from the Corning 850, Corning 865, Nova Stat Profile 5, and Mallinckrodt Gem-Premier served as the comparison for blood gases and electrolytes.
- Spun Hematocrit (Micro-centrifuge): For hematocrit measurements.
- Reference Materials: RNA Medical Qualidata levels 1, 2, 3 for blood gases and electrolytes, and RNA Medical hematocrit levels low and high were used in precision studies.
- Tonometered and Spiked Whole Blood: Used in non-clinical studies to create known concentrations for linearity and correlation.
This is best described as comparison against established reference methods and certified reference materials.
8. Sample Size for the Training Set
The document does not explicitly mention a "training set" in the context of an AI/machine learning algorithm. The SenDx 100 is an electrochemical and optical measurement device, not an AI algorithm that learns from data in that sense.
However, the "training" analogous to an AI algorithm would be the internal development and calibration processes. The document states:
- "All SenDx 100™ sensor cassettes are 100% tested at the factory with multiple precision reference solutions."
- The system performs "two point calibrations at automatic intervals" using "calibration solution pack which contains two levels of precision tonometered electrolyte solutions."
The specific sample size (number of runs, types of solutions) for these internal factory tests and ongoing calibrations is not provided, only that they are performed extensively ("100% tested," "multiple precision reference solutions").
9. How the Ground Truth for the Training Set Was Established
As explained above, there isn't a "training set" in the common AI sense. For the internal development and calibration of the instrument, the ground truth was established by:
- Known concentrations of precision reference solutions: Used for factory testing and calibration.
- Precision tonometered electrolyte solutions: Used in calibration packs for ongoing automatic calibrations.
These solutions would have their analyte values determined by highly accurate, often gravimetric or spectroscopically verified, methods.
Ask a specific question about this device
Page 1 of 1