Search Results

TRU RSV is a rapid, qualitative, lateral-flow immunoassay for the detection of Respiratory Syncytial Virus (RSV) antigens (fusion protein or nucleoprotein) in human nasal wash, nasopharyngeal aspirate, and nasal and nasopharyngeal swab samples. It is designed to test specimens from symptomatic patients aged 5 years or less. A negative result does not preclude RSV infection. It is recommended that all negative test results be confirmed by cell culture.

TRU RSV is a rapid, single-use, qualitative lateral-flow immunoassay screening test. The test kit includes a Test Strip attached to a plastic frame or holder, a Conjugate Tube containing a lyophilized bead of colloidal gold-linked monoclonal antibodies to RSV fusion protein and nucleoprotein, Sample Diluent/Negative Control, and plastic transfer pipettes. The Test Strip carries a nitrocellulose membrane with dried capture antibodies. The holder caps the Conjugate Tube during testing and disposal. The conjugate bead is rehydrated with Sample Diluent, sample is added and mixed, and the Test Strip is inserted. If RSV antigens are present, they bind to the conjugate and are captured at the Test Line, forming a pink-red line. An internal control line indicates proper flow.

Here's a summary of the acceptance criteria and study details for the Meridian Bioscience TRU RSV device, based on the provided text:

Acceptance Criteria and Device Performance

The document does not explicitly state pre-defined acceptance criteria metrics (e.g., "Sensitivity must be >X%, Specificity must be >Y%"). Instead, the study aims to demonstrate substantial equivalence to the predicate device and the "standard reference method - tissue culture." The "Correlation" metric appears to be the primary indicator of performance used in the individual site tables.

Therefore, the table below reflects the reported performance to demonstrate this equivalence. The overall results across all sample types and frozen/fresh conditions are used for the main reported device performance.

Note on "Overall" Reported Performance Calculation: The provided document breaks down performance by fresh/frozen and sample type. I calculated a weighted average for the overall device performance based on the totals from the provided tables (Fresh Wash/Aspirate Total, Frozen Wash/Aspirate Total, Fresh Swab Total, Frozen Swab Total).

• Positive (Overall): 72 (Fresh Wash) + 88 (Frozen Wash) + 13 (Fresh Swab) + 46 (Frozen Swab) = 219
• Negative (Overall): 111 (Fresh Wash) + 161 (Frozen Wash) + 107 (Fresh Swab) + 26 (Frozen Swab) = 405
• TRU RSV Positive (Overall): 85 (Fresh Wash) + 91 (Frozen Wash) + 19 (Fresh Swab) + 34 (Frozen Swab) = 229
• TRU RSV Negative (Overall): 98 (Fresh Wash) + 158 (Frozen Wash) + 101 (Fresh Swab) + 38 (Frozen Swab) = 395
• Total Tested (Overall): 183 (Fresh Wash) + 249 (Frozen Wash) + 120 (Fresh Swab) + 72 (Frozen Swab) = 624 (Note: The document states 625 samples, one was "invalid" in Fresh Wash/Aspirate Site 1.)

Using these counts for the combined data (excluding the invalid case):

• True Positives (TP): 64 (Fresh Wash) + 79 (Frozen Wash) + 12 (Fresh Swab) + 33 (Frozen Swab) = 188

• False Positives (FP): 21 (Fresh Wash) + 12 (Frozen Wash) + 7 (Fresh Swab) + 1 (Frozen Swab) = 41

• True Negatives (TN): 90 (Fresh Wash) + 149 (Frozen Wash) + 100 (Fresh Swab) + 25 (Frozen Swab) = 364

• False Negatives (FN): 8 (Fresh Wash) + 9 (Frozen Wash) + 1 (Fresh Swab) + 13 (Frozen Swab) = 31

• Sensitivity: TP / (TP + FN) = 188 / (188 + 31) = 188 / 219 = 85.8%

• Specificity: TN / (TN + FP) = 364 / (364 + 41) = 364 / 405 = 89.9%

• Correlation (Accuracy): (TP + TN) / (TP + TN + FP + FN) = (188 + 364) / (188 + 364 + 41 + 31) = 552 / 624 = 88.5%

It's important to note that the site-specific and fresh/frozen breakdowns show some variability, and the total sensitivity/specificity for each category were often presented with wide confidence intervals.

The overall summary for each type in the table 6 summary are:

• Fresh Wash/Aspirate Total: Sensitivity 88.9%, Specificity 81.1%, Correlation 84.2%
• Frozen Wash/Aspirate Total: Sensitivity 89.8%, Specificity 92.5%, Correlation 91.6%
• Fresh Swab Total: Sensitivity 92.3%, Specificity 93.5%, Correlation 93.3%
• Frozen Swab Total: Sensitivity 71.7%, Specificity 96.2%, Correlation 80.6%

Study Details

1. Sample Size Used for Test Set and Data Provenance:

   • Test Set Size: 625 samples.
   • Data Provenance: Clinical samples from symptomatic patients aged 5 years or less, collected in the US.
     • 304 samples: Collected during the 2006-07 season and tested fresh.
     • 321 samples: Collected during the 2006 and earlier seasons and tested as frozen/thawed.
   • Specimen Types: Nasal wash, nasopharyngeal aspirate, nasal swabs, nasopharyngeal swabs.

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

   • The primary ground truth for the clinical study was established using tissue culture as the "standard reference method." The document does not specify the number or qualifications of experts interpreting the tissue culture results, but it implies this was done by each laboratory's established internal method.
   • For discrepant results between TRU RSV and tissue culture, Direct Specimen Fluorescence Assay (DSFA) or Polymerase Chain Reaction (PCR) assays were used as additional confirmatory methods. Again, the number of experts interpreting these results is not specified.

3. Adjudication Method for the Test Set:

   • The study used a hierarchical adjudication method for discrepant results. When the TRU RSV result differed from the tissue culture result, the samples were "tested by Direct Specimen Fluorescence Assay (DSFA) or by a Polymerase Chain Reaction (PCR) assay." The results of these confirmatory tests (DSFA or PCR) were then used to clarify the true RSV status of the sample, as indicated in the postscripts to Tables outlining discrepant results. This suggests an adjudication process where a third, more definitive test was used to resolve discrepancies between the index test and the primary reference standard.

4. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:

   • No, a multi-reader multi-case (MRMC) comparative effectiveness study was not explicitly described. The study focuses on evaluating the standalone performance of the TRU RSV device against a reference method (tissue culture). It does not appear to assess the improvement of human readers with AI assistance versus without.

5. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study:

   • Yes, this was a standalone performance study. The TRU RSV is a rapid, qualitative, lateral-flow immunoassay that provides a visual result (pink-red lines). The performance evaluation section describes the device's accuracy compared to tissue culture, independently of human interpretation variations (beyond the initial reading of the test, though the reproducibility study suggests consistency in reading).

6. Type of Ground Truth Used:

   • The primary ground truth used was tissue culture.
   • For discrepant samples, DSFA (Direct Specimen Fluorescence Assay) and PCR (Polymerase Chain Reaction) were used as confirmatory methods to establish a more definitive ground truth.

7. Sample Size for the Training Set:

   • The document does not detail a separate "training set" in the context of device development or machine learning. For an immunoassay like TRU RSV, the development process (e.g., antibody selection, reagent formulation) relies on laboratory optimization rather than a distinct "training set" of clinical samples. The clinical study described served as a validation or test set to demonstrate pre-market performance.

8. How the Ground Truth for the Training Set Was Established:

   • As there's no mention of a traditional "training set" in the context of machine learning, this question is not explicitly addressed. The development of such diagnostic assays typically involves extensive analytical studies (e.g., analytical sensitivity, specificity, interference) using characterized samples and isolates, rather than a clinical "training set" with established ground truth in the same way as AI/ML algorithms.

Ask a specific question about this device