510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY

ASSAY ONLY TEMPLATE

A. 510(k) Number:

K161810

B. Purpose for Submission:

To obtain a substantial equivalence determination for a premarket notification for the BD BACTEC™ Standard/10 Aerobic/F Culture Vials Soybean-Casein Digest Broth in a Plastic Vial

C. Measurand:

Aerobic microorganisms (bacteria and yeast) from blood

D. Type of Test:

Liquid culture medium for recovery of microorganisms from blood using fluorescent technology to detect the increased CO2 produced by the growth of microorganisms

E. Applicant:

Becton, Dickinson and Company

F. Proprietary and Established Names:

BD BACTEC™ Standard/10 Aerobic/F Culture Vials Soybean-Casein Digest Broth in a Plastic Vial

G. Regulatory Information:

1. Regulation section:

1

21 CFR 866.2560 Microbial Growth Monitor

2. Classification:

Class I

3. Product code:

MDB System, Blood Culturing

4. Panel:

2

83 Microbiology

H. Intended Use:

1. Intended use(s): BD BACTEC Standard/10 Aerobic/F culture vials (enriched Soybean-Casein Digest broth with CO2) are for aerobic blood cultures. Principal use is with the BD BACTEC fluorescent series instruments for the qualitative culture and recovery of aerobic microorganisms (bacteria and yeast) from blood.

2. Indication(s) for use:

Same as Intended Use

3. Special conditions for use statement(s):

Prescription use Limitations:

Recovery of SPS Sensitive Organisms from Blood Samples Because blood can neutralize the toxicity of SPS toward organisms sensitive to SPS, the recommendation of maximum volumes of blood (8–10 mL) can help to optimize recovery of these organisms. Some fastidious organisms, such as certain Haemophilus species, require growth factors, such as NAD, or factor V, which are provided by the blood specimen. If the blood specimen volume is 3.0 mL or less, an appropriate supplement may be required for recovery of these organisms. BD BACTEC FOS™, Fastidious Organism Supplement, may be used as a nutritional supplement. Recovery of Haemophilus parainfluenzae is strain dependent. General Considerations The default 5-day (120 hours) protocol was utilized for all analytical testing with the BD BACTEC Standard/10 Aerobic/F culture media and protocol lengths of >5 days have not been evaluated.

4. Special instrument requirements: BACTEC fluorescent series instruments BACTEC 9050, BACTEC 9240, BACTEC FX, and BACTEC FX40 using software versions listed below:

3

Instrument Software Version BACTEC 9050 V2.01 BACTEC 9240 V4.95 BACTEC FX V5.20

BACTEC FX 40 V2.51B

I. Device Description: The blood sample to be tested is inoculated into one or more vials which are inserted into the BACTEC fluorescent series instrument for incubation and periodic reading. Each vial contains a chemical sensor which can detect increases in CO2 produced by the growth of microorganisms. The sensor is monitored by the instrument every ten minutes for an increase in its fluorescence, which is proportional to the amount of CO2 present. A positive reading indicates the presumptive presence of viable microorganisms in the vial. Detection is limited to microorganisms that will grow in a particular type of medium.

J. Substantial Equivalence Information:

1. Predicate device name(s):

BD BACTEC Standard/10 Aerobic/F Culture Vials Soybean-Casein Digest Broth (glass)

2. Predicate 510(k) number(s):

K954921

3. Comparison with predicate:

Table 1: Comparison with Predicate Device Item Device

BD BACTEC Standard/10 Aerobic/F (Plastic)

Predicate BD BACTEC Standard/10 Aerobic/F (Glass)

K954921 Similarities

Intended Use Qualitative culture and recovery of aerobic microorganisms

(bacteria and yeast) from blood with the BD BACTEC

fluorescent series instrument

Same (from Name and Intended Use sections)

Sample Type Human blood Same Instrument BD BACTEC fluorescent series Same Detection

Technology Continuous monitoring;

measurement of CO2 increase; rocking agitation parameters

Same

Incubation 35ºC (± 1.5ºC) up to 120 hours Same

4

Item Device BD BACTEC Standard/10

Aerobic/F (Plastic)

Predicate BD BACTEC Standard/10 Aerobic/F

(Glass) Differences

Vial Material Plastic Glass Vial Weight Lighter than glass -- Vial Height 5.0 inches 5.6 inches

Sensor Adhesion Promoter

Yes N/A

Vial Sensor 2.6 gram per vial, specific for the plastic vial geometry

1.75 gram per vial

Sensor Components

BCP (indicator) - 6.5 mg/gram of sensor

Red dye- 4.0 mg/gram of sensor

BCP (indicator)- 1.8 mg/gram of sensor

Red dye- 1.9 mg/gram of sensor

K. Standard/Guidance Document Referenced (if applicable):

Not Applicable

L. Test Principle: The BD BACTEC Standard/10 Aerobic/F medium is an enriched soybean- casein digest broth, with each vial containing 40 mL of broth. The formulation is specifically designed to enhance the recovery of aerobic microorganisms from blood. Yeast extract is added to the soy casein digest base to supplement the B-vitamin concentrations. Menadione and hemin are added to fortify the medium; sucrose is added to the broth to enhance recovery of fastidious fermentative bacteria (Streptococci), Staphylocci and yeast. Sodium polyanetholesulfonate (SPS) is added to the medium as an anticoagulant and compliment fixation inhibitor.

Each BD BACTEC Standard/10 Aerobic/F blood culture medium vial contains a chemical sensor in a silicon rubber base that can detect increases in CO2 produced by the growth of microorganisms. Three to seven mL of blood is inoculated into the BD BACTEC Standard/10 Aerobic/F blood culture medium vial, which is inserted into the BD BACTEC Fluorescent Series instrument for incubation, agitation and periodic measurement. When microorganisms are present in the blood sample, they metabolize nutrients in the culture medium, releasing CO2 into the medium. A dye in the sensor reacts with the CO2, modulating the amount of light that is absorbed by the fluorescent material in the sensor. The instrument’s photo detectors monitor the sensor every 10 minutes and measure the level of fluorescence, which is proportional to the amount of CO2 present in the vial. Positivity of a vial is determined by algorithms resident in the instrument rack’s microprocessor. The algorithms use the rate of CO2 production as well as the absolute increase in CO2 to interpret the data.

Culture vials flagged as presumptively positive are removed from the instrument for subculture and Gram stain in order to identify the microorganisms for further evaluation and proposed patient treatment. Culture vials that are not flagged as positive remain in the

instrument until the test protocol has been completed and negative bottles are discarded at the end of protocol (120 hours).

M. Performance Characteristics (if/when applicable):

1. Analytical performance:

5

a. Precision/Reproducibility: The BACTEC Standard/10 Aerobic/F (plastic) vial was evaluated across three lots in the Time To Detection (TTD) and the Percent Recovery (Detection) studies. Different lots of key raw materials were used to manufacture each lot of culture media. The actual inoculum level of 0-1, 2-10, and >10 CFU were used in the analysis. The actual inoculum level was determined on the basis of growth observed on non-selective agar plate. The time to detection results are presented in Table 2.

Table 2: Precision/Reproducibility: Lot to Lot Difference Lot Bld Vol

(mL) MedianTime to Detection in Hours (95% CI)

0*-1 CFU 2 -10 CFU >10 CFU 1 3 33.34 (24.536, 49.901) 20.4 (17.086, 25.350) 18.14 (16.800, 20.126)

10 27.12 (17.559, 35.246) 19.91 (16.836, 25.574) 18.34 (16.717, 20.324) 2 3 28.91 (19.058, 49.107) 20.16 (17.491, 25.077) 18.63 (16.895, 20.465)

10 27.60 (19.342, 49.534) 19.72 (17.049, 26.088) 18.57 (16.873, 20.470) 3 3 24.54 (18.221, 40.566) 20.40 (17.200, 26.172) 18.05 (16.791, 20.043)

10 32.95 (22.702, 53.436) 20.84 (17.224, 28.034) 18.10 (16.667, 20.078) * No growth observed on non-selective media for inoculum verification due to sampling imprecision at low target inoculum level of 1 CFU/vial.

The percent recovery results stratified by lot (combining blood volumes, inoculum levels, organisms, and instruments) are shown in Table 3.

Table 3: Percent of Positive Results by Lot

Lot Total Positive Percent Recovery (95% CI) 1 448 408 91.07% (87.94%, 93.47%) 2 448 407 90.85% (87.69%, 93.28%) 3 448 397 88.62% (85.22%, 91.33%)

Similar performance was observed among the lots and there were no statistically significant differences across all three lots in the precision study.

b. Linearity/assay reportable range:

Not Applicable

c. Traceability, Stability, Expected values (controls, calibrators, or methods):

Quality Control An internal validation study across three lots with inoculum at 10-100 CFU per vial

was conducted using the organisms listed below:

Alcaligenes faecalis ATCC 8750 Candida albicans ATCC 18804 Escherichia coli ATCC 25922 Haemophilus influenzae ATCC 19418 Neisseria meningitidis ATCC 13090 Pseudomonas aeruginosa ATCC 27853 Staphylococcus aureus ATCC 25923 Streptococcus pneumoniae* ATCC 6305 Streptococcus pyogenes ATCC 19615

*Specified in CLSI Standard, Quality Control for Commercially Prepared Microbiological Culture Media

There were a total of 150 replicates of these nine QC organisms for the study. Each organism was tested for either 15 (5 replicates x 3 lots) or 18 (6 replicates x 3 lots) replicates. All organisms were detected <72 hours, with mean ranges from 9.7 hours

6

(E. coli) to 26.1 hours (Neisseria meningitidis).

d. Detection limit:

Microbial Detection Limit (MDL, target inoculum level 0- 1, 1- 10 CFU/vial) The microbial detection limit study was conducted to assess the capability of the culture media to detect low numbers of organisms (expected target level of 0-1 and 1-10 CFU/vial) when present in blood. The study included 15 organisms tested at two blood volumes (3 and 10 mL), each with two inoculum levels of 0-1 and 1-10 CFU/vial and across three lots by two instruments (BACTEC FX and BACTEC 9240):

15 org x 3 lots x 2 blood vol x 2 inoculum levels x 2 instruments= 360

Results for target inoculum levels of 0-1 and 1-10 CFU/vial are provided in Tables 4.1 and 4.2 respectively. The percent recovery, difference between predicate and the new plastic bottles (henceforth referred to as Modified) was calculated with 95% two-sided bootstrap confidence intervals and stratified by organism for each target inoculum level. The combined performance of 0-1 and 1-10 is provided in Table 4.3.

Table 4.1: MDL Difference by Organism at Target Inoculum of 0-1 CFU/Vial

7

Organism Name Number

of Samples

Percent recovery

for Modified

Percent recovery

for Predicate

Difference between percent

recovery of

Modified and

Predicate

95% CI (low)

95% CI (high)

Candida albicans 12 41.67 66.67 -25 -71.91 21.91 Candida glabrata 12 75 75 0 -46.2 46.2

Cryptococcus neoformans 12 66.67 33.33 33.33 -1.95 68.62 Enterococcus faecalis 12 66.67 25 41.67 -12.3 95.63

Escherichia coli 12 0 33.33 -33.33 -68.62 1.95 Haemophilus influenzae 12 58.33 83.33 -25 -71.91 21.91

Haemophilus parainfluenzae biotype I

(ATCC 33392) 12 0 0 0 -23.1 23.1

Micrococcus luteus 12 16.67 25 -8.33 -44.55 27.88 Neisseria gonorrhoeae 12 58.33 33.33 25 -15.83 65.83 Neisseria meningitidis 12 100 91.67 8.33 -19.56 36.23

Pseudomonas aeruginosa 12 50 33.33 16.67 -34.12 67.45 Staphylococcus aureus 12 75 66.67 8.33 -27.88 44.55

Staphylococcus epidermidis 12 58.33 25 33.33 -1.95 68.62 Streptococcus pneumoniae 12 41.67 25 16.67 -28.56 61.89

Streptococcus sanguinis 12 41.67 75 -33.33 -81.42 14.75 Total 180 50.00 46.11 3.89 -5.77 13.55

Table 4.2: MDL Difference by Organism at Target Inoculum of 1-10 CFU/Vial

8

Organism Name Number

of Samples

Percent recovery

for Modified

Percent recovery

for Predicate

Difference between percent

recovery of

Modified and

Predicate

95% CI (low)

95% CI (high)

Candida albicans 12 100 100 0 -23.1 23.1 Candida glabrata 12 100 100 0 -23.1 23.1

Cryptococcus neoformans 12 100 91.67 8.33 -19.56 36.23 Enterococcus faecalis 12 100 100 0 -23.1 23.1

Escherichia coli 12 91.67 100 -8.33 -36.23 19.56 Haemophilus influenzae 12 100 100 0 -23.1 23.1

Haemophilus parainfluenzae biotype I

(ATCC 33392) 12 0 0 0 -23.1 23.1

Micrococcus luteus 12 83.33 91.67 -8.33 -36.23 19.56 Neisseria gonorrhoeae 12 100 83.33 16.67 -14.61 47.94 Neisseria meningitidis 12 100 100 0 -23.1 23.1

Pseudomonas aeruginosa 12 100 100 0 -23.1 23.1 Staphylococcus aureus 12 100 91.67 8.33 -19.56 36.23

Staphylococcus epidermidis 12 100 100 0 -23.1 23.1

Streptococcus pneumoniae 12 100 83.33 16.67 -14.61 47.94 Streptococcus sanguinis 12 100 100 0 -23.1 23.1

Total 180 91.67 89.44 2.22 -1.21 5.65

The MDL study demonstrated that the modified plastic device performed equivalently when compared to the predicate glass device at low target inoculum levels. However, none of the 24 paired sets of Haemophilus parainfluenzae ATCC 33392 were detected in the modified plastic or the predicate glass vials. The culture medium of BD BACTEC Standard/10 Aerobic/F does not support the growth of H. parainfluenzae at low target inoculum of 0-1 or 1-10 CFU/vial. Information about this organism was noted as a limiatation in labeling.

Table 4.3: MDL Difference at Target Inocula of 0-1, 1-10 CFU/Vial, and Combined

9

Target CFU

Number of

Samples

Percent recovery for

Modified

Percent recovery for

Predicate

Difference between percent

recovery of Modified and

Predicate

95% CI (low)

95% CI (high)

0-1 180 50.00 46.11 3.89 -5.77 13.55

1-10 180 91.67 89.44 2.22 -1.21 5.65

0-1, 1-10 360 70.83 67.78 3.06 -2.01 8.12

In summary, the data sets demonstrated that the modified plastic device performed equivalently when compared to the predicate glass device.

e. Analytical specificity:

Not applicable

f. Assay cut-off:

Not applicable

2. Comparison studies: Performance of the BD BACTEC Standard/10 Aerobic/F (plastic) blood culture vials was evaluated in seeded internal analytical studies to demonstrate comparable performance to the predicate device, the BD BACTEC Standard/10 Aerobic/F (glass) blood culture vials. Comparison results were acceptable. The comparisons were made using the following parameters: time to detection, percent recovery, false negative rate, and false positive. For statistical analysis, 95% two-sided bootstrap confidence intervals for differences between the modified and the predicate devices were used.

a. Method comparison with predicate device:

Instrument Time to Detection (TTD) study The TTD (in hours) was recorded as part of the combined Percent recovery and the Microbial Detection Limit studies using the BD BACTEC FX and 9240 instruments at three inoculum levels, across two blood volumes (3 mL and 10 mL) over three media lots. The Percent Recovery study represented by the standard inoculum of 10 to 100 CFU per vial and the Microbial Detection Limit study represented by the challenge inoculum levels of 0 to 1 and 1 to 10 CFU per vial.

The TTD data was also analyzed by target inoculum level, blood volume or lot numbers. The results are shown in Tables 5.1- 5.4. The results for all data combined are shown in Table 5.5

Table 5.1: Summary of TTD Study Results by Target Inoculum Levels and Blood Volumesa

10

Target CFU

# of Org.

Bld. Vol.

Median of TTD for Modified (95% CI)

Median of TTD for Predicate (95% CI)

Median of TTD differenceb

(95% CI) 0 to 1 15 3 26.17 (21.798, 43.939) 25.57 (21.519, 32.781) -1.04 (-6.530, 2.166)

10 21.27 (18.405, 36.353) 25.05 (19.553, 32.271) -3.25 (-8.668, 0.665) 1 to 10 15 3 21.49 (19.231, 24.814) 23.97 (20.853, 29.144) -2.42 (-4.167, -1.500)

10 22.68 (19.604, 27.832) 26.12 (21.422, 30.900) -2.25 (-3.751, -1.000) 10 to 100

41 3 18.10 (17.232, 19.313) 20.31 (19.015, 21.435) -1.25 (-1.501, -1.001) 10 18.30 (17.297, 19.418) 20.17 (18.986, 21.262) -0.75 (-0.917, -0.584)

a All lots combined b Median of the difference between Modified and Predicate devices

Table 5.2: Summary of TTD Study Results by Target Inoculum Levelsc Target CFU

Median of TTD for Modified (95% CI)

Median of TTD for Predicate (95% CI)

Median of TTD difference (95% CI)

0 to 1 23.52 (21.001, 30.806) 25.32 (22.341, 29.872) -2.00 (-5.613, 0.332) 1 to 10 21.97 (20.008, 24.617) 24.93 (22.047, 28.770) -2.34 (-3.418, -1.501)

10 to 100 18.21 (17.509, 19.043) 20.24 (19.362, 21.031) -0.92 (-1.085, -0.834) c All lots and blood volumes combined

Table 5.3: Summary of TTD Results by Blood Volumesd Blood

Volume (mL)

Median of TTD for Modified (95% CI)

Median of TTD for Predicate (95% CI)

Median of TTD difference (95% CI)

3 19.10 (18.141, 20.018) 21.00 (20.011, 22.015) -1.34 (-1.668, -1.085) 10 19.06 (18.103, 20.015) 21.00 (19.926, 21.917) -0.83 (-1.083, -0.669) d All lots and target inoculum levels combined

Table 5.4: Summary of TTD Resutls by Lotse Lot # Median of TTD for

Modified (95% CI)

Median of TTD for Predicate (95% CI)

Median of TTD difference (95% CI)

Lot 1 19.00 (17.852, 20.094) 19.00 (17.852, 20.094) -0.92 (-1.168, -0.750) Lot 2 19.26 (18.110, 20.374) 21.06 (19.818, 22.305) -1.17 (-1.584, -0.917) Lot 3 19.00 (17.823, 20.174) 21.06 (19.842, 22.308) -1.17 (-1.584, -0.917)

e All blood volumes and target inoculum levels combined

Table 5.5: TTD Median Difference for All Data Combined Median of TTD for Modified

(95% CI) Median of TTD for Predicate

(95% CI) Median of TTD difference

(95% CI) 19.08 (18.405, 19.749) 20.98 (20.266, 21.667) -1.08 (-1.251, -0.918)

The TTD study demonstrated that the modified plastic device performed equivalently when compared to the predicate glass device when stratified by target inoculum level, blood volume or lot numbers.

It was observed that the median TTD for the modified plastic vial and the predicate glass

vial was 19.08 hours and 20.98 hours respectively. The median TTD difference was -1.08 hours (95% CI, -1.25, -0.92), favoring the modified plastic vial.

Percent Recovery (Detection) Study The percent recovery (detection) was evaluated in a study of 984 paired sets at the standard inoculum level of 10 to 100 CFU/vial on four instruments across three lots using a diverse set of microorganisms frequently isolated in blood.

41 org x 3 lots x 2 blood vol x 4 instruments= 984 paired sets

The performance of each organism at inculum of 10 to 100 CFU per vial is shown in Table 6.1.

Table 6.1: Percent Recovery (10-100 CFU/Vial) Summary by Organism

11

Organism Name

# of Samples

Percent Recovery

for Modified

Percent Recovery

for Predicate

Difference between percent

recovery of

Modified and

Predicate

95% CI (low)

95% CI (high)

Abiotrophia defectiva 24 100 100 0 -11.55 11.55 Acinetobacter lwoffii 24 100 100 0 -11.55 11.55 Aerococcus viridans 24 100 100 0 -11.55 11.55

Aggregatibacter actinomycetemcomitans 24 100 100 0 -11.55 11.55

Alcaligenes faecalis 24 100 100 0 -11.55 11.55 Bacillus subtilis 24 100 100 0 -11.55 11.55

Candida albicans 24 100 100 0 -11.55 11.55 Candida glabrata 24 95.83 100 -4.17 -18.21 9.88

Cardiobacterium hominis 24 100 100 0 -11.55 11.55 Corynebacterium jeikeium 24 100 100 0 -11.55 11.55 Cryptococcus neoformans 24 100 95.83 4.17 -9.88 18.21

Eikenella corrodens 24 100 100 0 -11.55 11.55 Enterobacter cloacae 24 100 100 0 -11.55 11.55 Enterococcus faecalis 24 100 100 0 -11.55 11.55

Escherichia coli 24 100 100 0 -11.55 11.55 Granulicatella adiacens 24 100 100 0 -11.55 11.55 Haemophilus influenzae 24 95.83 100 -4.17 -18.21 9.88 Haemophilus influenzae

biotype I 48 100 97.92 2.08 -4.96 9.13

Haemophilus parainfluenzae biotype I

(ATCC 33392) 24 0 4.17 -4.17 -18.21 9.88

Kingella kingae 24 100 100 0 -11.55 11.55 Klebsiella pneumoniae 24 100 100 0 -11.55 11.55 Leuconostoc citreum 24 100 100 0 -11.55 11.55

12

OrganismName

# ofSamples

Percent Recovery

for Modified

Percent Recovery

for Predicate

Difference between percent

recovery of

Modified and

Predicate

95% CI(low)

95% CI (high)

Micrococcus luteus 24 100 100 0 -11.55 11.55 Neisseria gonorrhoeae 24 100 87.5 12.5 -5.06 30.06 Neisseria meningitidis 24 100 100 0 -11.55 11.55

Pediococcus acidilactici 24 95.83 83.33 12.5 -8.52 33.52 Proteus mirabilis 24 100 100 0 -11.55 11.55

Providencia stuartii 24 100 100 0 -11.55 11.55 Pseudomonas aeruginosa 24 100 100 0 -11.55 11.55

Rothia mucilaginosa 24 100 100 0 -11.55 11.55 Staphylococcus aureus 24 100 100 0 -11.55 11.55

Staphylococcus epidermidis 24 100 100 0 -11.55 11.55

Stenotrophomonas maltophilia 24 100 100 0 -11.55 11.55

Streptococcus agalactiae (Strep. group B) 24 100 100 0 -11.55 11.55

Streptococcus pneumoniae 96 100 100 0 -2.89 2.89 Streptococcus pyogenes

(Strep. group A) 24 100 100 0 -11.55 11.55

Streptococcus sanguinis 24 100 100 0 -11.55 11.55 TOTAL 984 97.26 96.75 0.51 -0.26 1.28

The results from 984 paired sets are shown in Table 6.2:

Table 6.2: Percent Recovery (10- 100 CFU/Vial) Summary Predicate

Positive Negative Modified Positive 948 9 957

Negative 4 23 27 952 32 984

Of the 984 paired sets, 948 sets recovered organisms in both the modified plastic vial and the predicate glass vial and 23 sets were negative by both vials. The 23 paired sets were Haemophilus parainfluenzae not detected at actual inoculum level of 82 CFU. In addition, the modified plastic vial had four detection failures for the following organisms: one detection failure for Candida glabrata, Pediococcus acidilactici, Haemophilus influenzae type b, and Haemophilus parainfluenzae at 66, 26, 64, and 82 CFU respectively. The culture medium of BD BACTEC Standard/10 Aerobic/F does not support the growth of H. parainfluenzae at target inoculum of 10-100 CFU/vial.

Additional Testing a) Candida spp. A study was conducted to evaluate TTD of Candida species other than C. albicans. The study included two strains each for Candida tropicalis, C. parapsilosis and one strain each for C. glabrata, Cryptococcus neoformans at target inoculum level of 10-100 CFU with two blood volumes (3 and 10 mL) tested on four instruments (9050, 9240, FX, and FX40). The TTD results for all data combined are shown in Table 6.3.

Table 6.3: TTD Median Difference at 10-100 CFU for All Data Combined

13

Median of TTD for Modified (95% CI)

Median of TTD for Predicate (95% CI)

Median of TTD difference (95% CI)

28.71 (27.560, 30.045) 30.67 (28.680, 51.510) -1.33 (-2.490, -1.000)

The study demonstrated that the modified plastic device performed equivalently when compared to the predicate glass device.

b) Haemophilus parainfluenzae Five well-characterized H. parainfluenzae strains were tested to evaluate recovery in both the predicate glass and the modified plastic device. The isolates were tested in triplicates with three blood volumes (3, 5, and 10 mL), two instruments (9240, FX) at 10-100 CFU/vial, resulting in 90 paired sets. The result is provided in Table 6.4.

5 strains x 3 bld vol. x triplicates x 2 instruments= 90 paired sets

Table 6.4: Haemophilus parainfluenzae Percent Recovery with Well-Characterized Strains at Target Inoculum 10-100 CFU

Strain ID Modified Plastic Predicate Glass ATCC 9796 94% (17/18) 100% (18/18) ATCC 33392 38.9% (7/18) 33.3% (7/18)

Strain #3 0% (0/18) 6% (1/18) Strain #4 0% (0/18) 6% (1/18) Strain #5 0% (0/18) 6% (1/18)

The recovery of H. parainfluenzae varied from 0% to 94% and 6% to 100% for modified plastic and predicate glass respectively. The modified plastic device performed equivalently when compared to the predicate glass device.

The text below was added to the existing limitation for Haemophilus species in the Recovery of SPS Sensitive Organisms from Blood Samples section of the package insert:

“Recovery of Haemophilus parainfluenzae is strain dependent.”

Further, the study result was also described in the Performance Characteristics section of the package insert:

“In an additional seeded study with five Haemophilus parainfluenzae strains tested with blood volumes 3, 5, and 10 mL at target inoculum 10–100 CFU, recovery rate was strain dependent and varies from 0% to 94% for plastic vial and from 6% to 100% for glass vial.”

False Positive Rates (Instrument-positive, subculture-negative) False positive was defined as instrument positive but subculture negative in the evaluation. False positivity was assessed with vials inoculated with fresh human blood of 2, 4, 6, 8, and 10 mL, but no organisms were added to the vials. There were a total of 240 pair sets across three lots using BACTEC FX and BACTEC 9240 and completed at the default 5-day protocol.

40 vials x 3 lots x 2 instruments= 240 paired sets

There was no positive observed in the predicate glass vial; one positive (instrument positive and subculture positive) was observed in the modified plastic vial on BACTEC FX at 16.98 hours. The organism was viridans group streptococcus (Streptococcus salivarius) which likely was a contaminant introduced during the inoculation process. The performance is shown in Table 7.

Table 7: Summary of False Positive Study

14

# of paired

sets

Percent Recovery for

Modified

Percent Recovery for

Predicate

Defference be tween Percent Recovery of

Modified and Predicate

95% CI

(low)

95% CI

(high) 240 0 0.42 -0.42 -1.83 1.00

The modified device performed equivalently when compared to the predicate device in the false positive study.

False Negative Rates (Instrument-negative, subculture-positive) All inoculated paired sets (984 from the Recovery study + 360 from the Microbial Detection Limit study =1344) that were instrument negative at the end of default protocol (120 hours) were subcultured onto appropriate culture media plates. This combined data set was evaluated for the false negative rates. A false negative is a vial that was instrument-negative at the end of protocol yet contains viable organisms upon subculturing onto appropriate culture media. There were a total of 91 paired sets where both the modified and the predicate devices were negative at 120 hours; there were 41 sets where the predicate device only detected (i.e., 41 modified plastic vials subcultured) and 57 vials where the new device detected (i.e., 57 predicate glass vials subcultured).

There were 34 false negatives; 18 (12 at 10 to 100 CFU) that were observed in the predicate glass vials and 16 (9 at 10-100 CFU) in the modified plastic vial. The 16 false negatives included one each for Candida glabrata, Haemophilus influenzae, and Pediococcus acidilactici, two Cryptococcus neoformans and 11 Haemophilus parainfluenzae. The modified device performed equivalently when compared to the predicate device in the false negative study.

The false negatives of the modified plastic vial is noted in the Performance Characteristics section of the package insert:

“Sixteen false negatives across all inoculum levels were identified within the BD BACTEC Standard/10 Aerobic/F medium contained in a plastic vial: Candida glabrata (1) Cryptococcus neoformans (2), Haemophilus influenzae (1), Haemophilus parainfluenzae biotype I (11), and Pediococcus acidilactici (1). Of the eleven Haemophilus parainfluezae biotype I, six inoculated at 82 CFU and five inoculated at 2 CFU.”

BD BACTEC Instrument Compatibility Study The BACTEC instrument compatibility study was evaluated from the Percent Recovery study (i.e., 984 paired sets) dataset across four fluorescent- series instruments: BD BACTEC FX, FX40, 9240, and 9050. The study included 3 and 10 mL of blood at inoculum level of 10-100 CFU per vial. The BACTEC instrument compatibility study is shown in Table 8.

Table 8: Summary of Instrument Compatibility Study

15

Instrument Blood Volume

(mL)

Median Time to Detection (TTD) in Hours (95% CI) TTD Median for

Modified TTD Median for

Predicate Median of TTD

difference FX 3 18.08 (16.532, 20.101) 20.11 (17.812, 22.149) -1.17 (-1.668, -0.752)

10 18.40 (16.618, 20.817) 20.30 (18.049, 22.302) -0.67 (-1.002, -0.501) FX40 3 18.60 (16.962, 20.656) 21.11 (18.750, 23.580) -1.34 (-2.250, -0.752)

10 18.59 (16.726, 20.653) 20.42 (18.334, 22.510) -0.75 (-1.167, -0.501) 9240 3 17.51 (16.102, 19.903) 19.44 (17.096, 21.524) -1.33 (-1.834, -1.000)

10 17.66 (16.033, 20.044) 19.30 (17.157, 21.658) -0.75 (-1.167, -0.500) 9050 3 18.25 (16.584, 20.917) 20.58 (18.250, 23.167) -1.08 (-1.916, -0.750)

10 18.42 (16.667, 20.584) 20.58 (18.167, 22.834) -0.75 (-1.167, -0.500)

The study demonstrated that the four instruments performed equivalently and they are compatible with BD BACTEC Standard/10 Aerobic/F culture medium in plastic vials.

b. Matrix comparison: In seeded analytical studies, the performance of BD BACTEC Standard/10 Aerobic/F culture medium in plastic vial was compared to that in glass vial, with two human blood volumes, 38 common blood bloodstream bacteria and five yeasts across four fluorescent series instruments: BACTEC FX, FX40, BACTEC 9240, and BACTEC 9050.

3. Clinical studies: Not applicable; seeded analytical studies to compare the new plastic blood culture vials to the glass blood culture vials (predicate).

a. Clinical Sensitivity:

Not applicable

b. Clinical specificity:

Not applicable

c. Other clinical supportive data (when a. and b. are not applicable):

Not applicable

4. Clinical cut-off:

16

Not applicable

5. Expected values/Reference range: Seeded analytical studies demonstrated equivalent performance of the BD BACTEC Standard/10 Aerobic/F (plastic) blood culture medium when compared to the BD BACTEC /10 Aerobic/F (glass) blood culture medium.

N. Proposed Labeling: The labeling is sufficient and it satisfies the requirements of 21 CFR Part 809.10.

O. Conclusion: The submitted information in this premarket notification is complete and supports a substantial equivalence decision.