3 paul p. bourbeau and brandi lynn swartz 4 geisinger...
TRANSCRIPT
First Evaluation of the WASP, a New Automated Microbiology Plating Instrument 2
Paul P. Bourbeau*
and Brandi Lynn Swartz 3
Geisinger Medical Center, Danville, PA. 4
5
6
7
8
*Corresponding author. Mailing address: Division of Laboratory Medicine, Geisinger Medical 9
Center, Danville, PA 17822-0131. Phone: 570-271-7467. E-mail:[email protected] 10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
ACCEPTED
Copyright © 2009, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved.J. Clin. Microbiol. doi:10.1128/JCM.01963-08 JCM Accepts, published online ahead of print on 21 January 2009
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
2
ABSTRACT 25
26
Many laboratories are experiencing growing shortages of trained microbiology technologists 27
and technicians. Consequently, there is considerable interest in new automation that could 28
potentially lessen labor demands for specimen processing. In this study, we present the first 29
published evaluation of a new microbiology instrument, the WASP (Walkaway Specimen 30
Processor) manufactured by Copan, Inc. in which we evaluated cross-contamination, accuracy of 31
plating, and quality of results. Absence of cross-contamination was demonstrated by plating a 32
total of 200 alternating inoculated and sterile specimen tubes. The ability of the WASP to 33
subculture enrichment broths was evaluated with 106 LIM broths, with results identical to 34
routine testing. Plating of urine specimens with the WASP was compared to plating with the 35
Dynacon Inoculab instrument. 300 specimens were plated in duplicate with both instruments 36
using 1-ul loops and 293 specimens were plated in duplicate on both instruments with 10-ul 37
loops. Results were compared between duplicate plating with the same instrument (replicate 38
plating) and for consensus agreement between the two instruments. Replicate plating results 39
were comparable for both instruments while the WASP had more specimens with significant 40
results than the Inoculab with the 1-ul loop only. Lastly, for plating of 113 ESwab specimens, the 41
manual method and WASP plating each yielded 90 potential pathogens. In summary, we report a 42
first evaluation of a new microbiology specimen-plating instrument, the WASP, which offers 43
opportunities to automate the plating of microbiology specimens to an extent that has not been 44
possible to date. 45
46
47
48
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
3
49
50
INTRODUCTION 51 52
Clinical microbiology laboratories have largely been bypassed in the advances in automation that 53
have benefitted other areas of the clinical laboratory in recent years. Continuously monitoring 54
blood culture systems as well as automated microbial identification and susceptibility testing 55
systems are widely utilized. However, the specific areas of specimen processing and culture 56
workup remain manual tasks with few changes in the recent past. While some larger labs utilize 57
urine-plating instrumentation, most microbiology laboratories have no automation in their 58
processing areas. In this report, we present the results of a preliminary evaluation of a new 59
microbiology plating instrument that offers the potential to automate the plating of a variety of 60
liquid-based microbiology specimens. 61
62
MATERIALS AND METHODS 63
64
Overview of WASP. The WASP (Walk Away Specimen Processor) is a new instrument 65
manufactured in Italy for Copan Diagnostics (Murrieta, CA) that is designed to plate liquid 66
specimens from a variety of different transport devices (See Figures 1 and 2). The WASP utilizes 67
two Toshiba SCARA (selective compliant assembly robot arm ) robots. The first robot moves 68
specimens and plated media and takes specimens to the decapping device while the second robot 69
does the actual inoculation and plate streaking. Barcode readers on the WASP scan the specimen 70
tube, and a printer prints specimen information and a barcode on a label that is placed on the 71
plated media. A 9-silo carousel holds 342 standard BD plates (Becton Dickinson Microbiology 72
Systems, Cockeysville, MD) or 378 standard Remel plates (Remel, Lenexa, KS). Only transport 73
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
4
devices with the specimen in a liquid phase can be processed on the WASP. With the exception 74
of large urine cups (120 ml cups in our lab), all specimens are loaded on the WASP using special 75
Teflon pallets that contain holes that are sized for specific containers. For example, one pallet 76
holds 12 Vacutainer urine transport tubes, a different pallet holds 12 ESwabs, and a third type of 77
pallet holds 6 enteric transport media tubes. Up to 6 pallets can be loaded on to the instrument at 78
one time, resulting in a maximum load of 72 Vacutainers or ESwabs at one time. The WASP 79
also has a vortex and a spinner/centrifuge that can be used to prepare specimens for plating. 80
Actual plating is done by three metal loops incorporated in a triquetra (three cornered) loop 81
inoculation tool (Figure 3). The loops are available in 3 sizes: 1 ul, 10 ul, and 30 ul. The WASP 82
contains multiple sensors to verify proper operation including one sensor that is connected to a 83
camera that verifies that the loop contains specimen after it is dipped into the specimen. A touch-84
screen computer facilitates selection of inoculation protocols, media, and streaking patterns, as 85
well as other functions, and guides operation of the WASP. 86
This evaluation consisted of a preliminary evaluation of the first production version of a 87
WASP in a clinical microbiology laboratory. At the time that this evaluation was performed, the 88
WASP contained software to process only urine Vacutainer specimens, UriSwabs and ESwabs. 89
Consequently, this evaluation was limited to those specimen transport devices. 90
Cross-Contamination Studies. To determine whether cross contamination occurs when 91
streaking sequential specimens using the WASP instrument, studies were performed with both 92
Vacutainer Urine C&S Preservative Plus Plastic Tubes (B.D.) and ESwabs (Copan). A fresh 93
subculture of Escherichia coli (ATCC # 35218) was used to prepare a suspension equivalent to a 94
0.5 MacFarland standard in sterile saline. Dilutions were performed to obtain ~105 CFU/ml and 95
~106
CFU/ml organism suspensions. 96
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
5
Four ml of sterile saline was added to each of 50 Vacutainer tubes, and 4 ml of the 97
105 CFU/ml E. coli suspension was added to each of an additional 50 Vacutainer tubes. The 98
tubes were then placed into the appropriate WASP pallet by alternating a tube filled with the 99
E. coli suspension and a tube filled with sterile saline. The pallets were placed on the WASP 100
instrument, and a standard urine-streaking pattern was selected that included a Blood Agar Plate 101
(BAP) and MacConkey (MAC) Agar Plate for each of the 100 Vacutainers. All Vacutainer tubes 102
were plated first using a 1µL loop and then retested using a 10µL loop. This yielded a total of 103
200 sets of plates: 100 sets inoculated with sterile saline and 100 sets inoculated with the E. coli 104
suspension. Plates were incubated at 37°C and examined at 24 h for growth. 105
For testing of ESwabs, 0.1 ml of the106
CFU/ml E. coli suspension was added to each ESwab 106
tube (ESwabs contain 1 ml of liquid Amies transport media). The tubes were placed into the 107
appropriate WASP pallet by alternating a tube inoculated with the E. coli suspension and a tube 108
filled with only the transport media. The pallets were placed on the WASP instrument, and a 3-109
quadrant streaking pattern was selected that included a Blood Agar Plate (BAP) and MacConkey 110
(MAC) Agar Plate for each of the 100 ESwab tubes. All ESwab tubes were processed first using 111
a 10-µL loop and then retested using a 30µL loop. This yielded a total of 200 sets of plates: 100 112
sets inoculated with sterile saline and 100 sets inoculated with the E. coli suspension. Plates were 113
incubated at 37°C and examined at 24 h for growth. 114
Enrichment broth subculture. To verify the accuracy of the WASP for subculture of 115
enrichment broths, testing was performed with LIM (Remel) broths. Routine LIM broths 116
inoculated for prenatal group B strep screens from the Geisinger Microbiology Laboratory were 117
used for this testing. Vaginal/rectals specimens are incubated in LIM broth for 18-24 h before 118
subcultured for routine testing. After the routine testing was completed, LIM broth tubes were 119
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
6
vortexed, and 1 ml was transferred to an empty, sterile ESwab tube. The ESwab tubes with LIM 120
broth were subsequently subcultured to neomycin nalidixic acid (NNA) agar plates (BD) and 121
BAP plates on the WASP using the 10-ul loop. These cultures were read and worked up at 24 h 122
and 48h, and the results were compared with the routine culture results. 123
Urine cultures. The Dynacon Inoculab (Dynacon Inc., Mississauga, ON, Canada) is used for 124
routine plating of urine cultures submitted in Vacutainer tubes to the Geisinger Microbiology 125
laboratory. Plates on the Inoculab can be inoculated with either 1-ul or 10-ul loops. For this 126
validation, routine urine specimens received in BD Vacutainers were first plated on BAP and 127
MAC agar plates using a 1-ul loop on the Inoculab for the routine microbiology lab culture, and 128
these cultures were worked up and reported using standard Geisinger Microbiology Laboratory 129
protocols. All specimens were then plated a second time with the Inoculab, and those plates were 130
labeled for study purposes. The same specimens were then subsequently plated using a 1-ul loop 131
by the WASP on BAP and MAC plates in duplicate and identified for the study. This resulted in 132
4 total platings (2 times on Inoculab and 2 times on WASP). Additional specimens were then 133
plated twice each on the Inoculab and the WASP using 10-ul loops. 134
All plates were placed in a non-CO2 incubator for a minimum of 16 hours. One set of the 135
plates from the Inoculab inoculated with a 1-ul loop was read and reported by a Geisinger 136
Microbiology Technologist and the other sets of plates were read by one of us (BLS). Standard 137
Geisinger microbiology laboratory procedures were used in the workup of all cultures. A senior 138
technologist or one of us (PPB) reviewed all discrepant results. 139
ESwabs. Routine cultures from a variety of sources submitted to the laboratory in ESwabs 140
were plated using the WASP. After routine manual plating was completed, specimens were 141
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
7
loaded onto the WASP and inoculated with a 3-quadrant streaking pattern, using BAP, MAC, 142
and CHOC agar plates. The WASP 30-ul loop was used to inoculate all plates. 143
All cultures plated by the WASP were worked up independently of the manually plated 144
cultures. Results were compared after all testing was completed on the manually plated and 145
WASP plated specimens. 146
RESULTS 147
Cross-Contamination Studies. 50 inoculated and 50 sterile Vacutainer tubes were alternately 148
loaded on the WASP. They were plated with both 1-ul and 10-ul loops for a total of 200 149
cultures. No colonies were observed from the sterile tubes, and consistent streaking patterns were 150
noted from the inoculated tubes. Similar results were observed with ESwab tubes for both the 10-151
ul and 30-ul loops with no growth from any of the sterile tubes and consistent streaking patterns 152
from the inoculated tubes. 153
Enrichment broth subculture. 106 LIM broths were subcultured with the WASP. Preliminary 154
studies (data not shown) indicated that the use of the 30-ul loop did not result in a satisfactory 155
number of isolated colonies. The use of the 10-ul loop, however, produced consistent isolated 156
colonies. Using the 10-ul loop, there was 100% concordance with the routine culture method 157
with each detecting 20 positive and 86 negative test results. 158
Urine cultures. A total of 300 specimens were processed in duplicate on the WASP and 159
Inoculab using 1-ul loops, while 293 specimens were processed in duplicate on both instruments 160
using 10-ul loops. The same specimens were not necessarily used for the 1-ul loops and the 10-ul 161
loops so results were analyzed separately by loop size. Culture results were divided into those 162
considered likely significant and those considered likely not significant. 163
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
8
For specimens plated with 1-ul loops, single or multiple isolates with < 1000 CFU/ml and 164
cultures with 3 or more organisms each with >10,000 CFU/Ml (multiple flora) were considered 165
not significant as were isolates of Micrococcus spp. and Lactobacillus spp. in any quantity. For 166
the purposes of this study, coagulase-negative staphylococci and viridans streptococci were 167
considered likely significant if present in a concentration of ≥10,000 CFU/ml when present as a 168
single pathogen or with no more than one other organism. For tabulation purposes, if a culture 169
contained ≥10,000 CFU of a potential pathogen (e.g. E. coli) and < 10,000 CFU/ml of one or two 170
other organisms, the E. coli isolate was included with the significant isolates and the other 171
organism(s)was included with the not significant isolates. For specimens plated with 10-ul loops, 172
results obtained were interpreted in a similar manner, accounting for the 10-fold difference in the 173
dilution factor. Results obtained with the Inoculab and the WASP were evaluated in two ways. 174
First, the replicate results were compared for each instrument. For example, the result obtained 175
on specimen #1 plated with the WASP using the 1-ul loop was compared with the second plating 176
of specimen #1 on the WASP using the same size loop. Second, the recovery of significant 177
isolates was compared between the two instruments. 178
The results of the replicate results for specimens with significant isolates are summarized in 179
Table 1. Both instruments yielded similar results in replicate plating for significant isolates for 180
both sized loops. Overall, for quantitation (same CFU/ml range, e.g. both replicates >105
181
CFU/ml), there were no significant differences between the results for the two instruments. 182
For cultures with non-significant results, there was agreement for the Inoculab between paired 183
plates for the 1-ul loop for 184/223 results and 219/220 for the 10-ul loop. For the WASP, for 184
specimens with non-significant results, there was agreement for the 1-ul loop for 192/220 results 185
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
9
and 218/220 for the 10-ul loop. Most of the discrepant results involved specimens with no 186
growth on one set of plates and < 10 actual colonies on the paired culture plates. 187
Results from WASP and Inoculab for specimens with significant results not in agreement are 188
summarized in Table 2. From 74 specimens with significant results that were plated with 1-ul 189
loops, there were 11 specimens with discrepant results. For 5 of the specimens, one of the 190
Inoculab results was not in agreement with the 3 other results while for 1 specimen, one WASP 191
result was not in agreement with the 3 other results. Four significant isolates were detected only 192
by the two WASP cultures. Lastly, for one specimen, there was a difference in CFU/ml counts 193
between the two WASP cultures and the two Inoculab cultures. 194
From 65 specimens with significant results that were plated with 10-ul loops, there were 7 195
specimens with discrepant results. One each of the WASP and Inoculab specimens was not in 196
agreement with the 3 other results. For five specimens, there was a difference in CFU/ml counts 197
between the two WASP cultures and the two Inoculab cultures. 198
ESwab cultures. A total of 113 specimens that were collected in ESwabs were plated on the 199
WASP. These included 13 vaginal specimens and 100 routine cultures of mixed types (wounds, 200
drainages, fluids, nares, skin, upper respiratory and others). All plates were inoculated using a 201
30-ul loop and read at 24 and 48 h of incubation. Manual and WASP plating both yielded 90 202
potential pathogens. The yield of normal flora (skin, vaginal, and respiratory) was also the same. 203
One of the WASP cultures grew one colony of mold that was not detected on the manually 204
plated culture. 205
206
207
208
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
10
DISCUSSION 209
Many laboratories are experiencing growing shortages of trained microbiology technologists 210
and technicians. This has been exacerbated not only by growth in routine testing but also by the 211
demand for testing performed for epidemiological purposes, such as for MRSA (3). 212
Consequently, there is considerable interest in new automation that could potentially lessen labor 213
demands for specimen processing (7). 214
Current instrumentation available for microbiology processing includes streaking and plating 215
instruments. There are 3 instruments currently available that can perform plate inoculation and 216
streaking: The Dynacon Inoculab (LQ and LQH models); the bioMérieux MicroStreak 217
instrument; and the WASP. 218
Inoculab instruments are designed to plate urine specimens of one specimen type. They plate 219
specimens from one type of container that is selected at the time of instrument purchase. The 220
Dynacon model LQH that we have in our lab holds a total of 38 uninoculated Remel plates, 221
giving it a capacity of 19 specimens without reloading if two plates are used for each culture or 222
38 specimens if one plate is used. Since the LQH model has only one silo for uninoculated 223
plates, if you use more than one type of plate for a culture, the plates must be intercalated in the 224
stack. Inoculab instruments can also be used in a streak-only function 225
The MicroStreak was released in 2008 (2). The MicroStreak requires a plugged disposable 226
pipette tip for each specimen and one disposable plastic applicator for each plate. It can plate 227
either single plates or biplates, and the applicator spreads the inoculum over the entire area of a 228
standard 100 mm plate or a biplate. The smallest pipette that the MicroStreak can use is a 10-ul 229
pipette and currently, the specimen top must be removed at the time the specimen is placed on 230
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
11
the instrument. If the residual specimen is to be saved, the specimen must be recapped after plate 231
inoculation(2). 232
This study was designed to be a preliminary evaluation of a new instrument named the 233
WASP. It was restricted to two container types-Vacutainers and ESwab tubes. Software that was 234
not available when this study was performed will permit sampling of other specimen containers. 235
We compared inoculation of urine Vacutainer specimens by the WASP and the Inoculab LQH 236
instrument. We did not perform manual plating for this study as our in-house validation of the 237
Inoculab instrument in indicated that the Inoculab results were more reproducible than our 238
manual plating method. Lue has also shown that the Inoculab was more accurate than manual 239
plating in her lab using a 1-ul loop (5). Moreover, variability in manual specimen volume 240
transfer for urine specimens has been demonstrated (1).We also evaluated the accuracy of the 241
WASP to subculture Lim broths. Lastly, we compared manual plating of routine specimens 242
collected in ESwabs with plating performed by the WASP. 243
Overall, the results obtained with the Inoculab and WASP for plating of urine specimens were 244
comparable for specimens plated with 1-ul and 10-ul loops. Interestingly, there were 4 specimens 245
plated with the 1-ul loop that grew a significant pathogen at a concentration of 104-10
5 CFU/ml 246
on both WASP cultures (2 with Enterococcus sp. and one each with K. pneumoniae and B-247
hemolytic Streptococcus) while the Inoculab culture was mixed or had multiple flora. Similarly, 248
there was also one specimen plated with the 10-ul loop that grew Enterococcus sp. at a 249
concentration of 103-10
4 CFU/ml on both WASP cultures and mixed flora on the Inoculab 250
culture. 251
There are several possible explanations for the different results between the WASP and 252
Inoculab for these specimens: (1) random differences in CFUs near the breakpoints for reporting 253
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
12
organisms; (2) failure of a loop to pick up specimen (loss of meniscus); and (3) more 254
homogeneous specimen preparation with one instrument than the other. Differences between the 255
operation of the Inoculab and WASP operation support the latter two possibilities. The WASP 256
has a camera that detects the presence of a droplet inside the inoculating loop after the specimen 257
is sampled. If no specimen is detected, the loop returns to the tube a second time and, if 258
necessary, a third time. If the camera fails to again detect a droplet, the tube is moved to a 259
discard bin. The Inoculab has no sensor to detect the presence of specimen in the loop. A second 260
difference between the instruments is the preparation of the specimen prior to sampling. The 261
instrument manual for the Inoculab states that the specimen should be agitated prior to placing it 262
on the instrument but does not specify how that agitation should be accomplished. In contrast, 263
the WASP has a vortex that vigorously mixes the urine tube prior to sampling. The exact reasons 264
for the observed differences are clearly speculative on our part; nonetheless, the ability to verify 265
the presence of actual specimen in the inoculating loop as well as vigorous vortexing prior to 266
plating may offer more than a theoretical advantage for the WASP compared to the Inoculab 267
instrument. 268
Clearly, there is a need for studies that will assess the speed, throughput, and labor savings of 269
the WASP. At the time that this preliminary study was performed, the WASP required 27 min. to 270
plate 24 Vacutainer tubes to two plates each while the Inoculab required 24 minutes to plate 24 271
Vacutainer tubes to two plates. We anticipate collecting more data once the Sunquest interface is 272
fully validated and software upgrades have been installed. 273
There are also significant capacity differences between the Inoculab and the WASP. The 274
Dynacon model LQH, which was used in our laboratory, holds a total of 38 uninoculated Remel 275
plates compared to 378 Remel plates for the WASP. Since the Dyacon LQH model has only one 276
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
13
silo for uninoculated plates, if you use more than one type of plate for a culture, the plates must 277
be intercalated in the stack. This would not be necessary if a single plate was used. The Dynacon 278
Inoculab model LQ has 2 silos for uninoculated media that permits the use of one silo for each of 279
two types of media for each specimen. The WASP has nine silos each holding 42 Remel plates, 280
and one to nine types of media can be loaded at one time. More than one silo can be loaded with 281
the same type of media. For example, if a lab uses a BAP for most specimen types, they might 282
chose to load 3 silos with BAP and fewer silos with other types of media. 283
In our evaluation of the WASP for subculturing broths to plated media, we encountered no 284
problems with the WASP. Using a 10-ul loop, the WASP consistently produced isolated 285
colonies. For this study the Lim broths were transferred to empty ESwab tubes. We do not 286
anticipate that labs would want to do this as a standard practice. The resolution to this issue could 287
be the production of broths in smaller tubes either by Copan and or another manufacturer or 288
software changes to the WASP to accommodate taller tubes. 289
Although our evaluation of the WASP for plating of routine cultures was limited to 113 290
specimens, results obtained were comparable to the manually plated cultures. We developed a 3-291
quadrant streaking pattern for use with the 30-ul loop that was used for all of these specimens. 292
The WASP has the potential to process different size containers. The jaws on the robot have 293
three settings to accommodate 3 sizes of containers. There are three docking stations (the 294
docking station holds the bottom of the container during the uncapping process) that also 295
accommodate 3 sizes of containers. The docking station can be easily swapped out. There are 296
two decapping devices on the WASP. One is fixed and is designed for Vacutainer urine tubes. A 297
variety of other decappers can be placed in the second decapping station including an ESwab 298
decapper. For this study we used the decapper and docking station designed to hold ESwabs. 299
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
14
During this evaluation, we encountered no problems opening and closing either Vacutainers or 300
ESwab tubes. Subsequent to the performance of this study, an interface was established between 301
the WASP and the Sunquest Laboratory Information System. While now operational, it has yet 302
to be fully validated. Unlike the MicroStreak, and similar to the Inoculab, the WASP requires no 303
disposable products for plating. The option of plating urine specimens with 1-ul or 10-ul 304
provides laboratories with flexibility to address clinical needs (4, 6). 305
The only mechanical (hardware) problem that we encountered during this study was a burnt 306
out incinerator that we replaced. There were several software problems that were systematically 307
addressed by Copan as they appeared. None of those problems persisted. 308
This study was designed to be a preliminary evaluation of a new instrument, the WASP. As 309
we expand the use of the WASP to other specimen types, we are working to optimize the 310
workflow pattern for each specimen type. In our routine use of ESwabs, specimens are manually 311
plated using a disposable pipette that is manufactured to deliver approximately 28 ul/drop. Each 312
piece of media receives one drop of the inoculum, and, for specimens requiring a Gram stain, 313
one drop is placed on a slide. We envision a similar workflow with the WASP, with the 314
preparation of any necessary slides occurring prior to loading containers onto the WASP for 315
plating. It is important to understand that the only type of swab specimens that can be plated by 316
the WASP is a swab that transfers the specimen to a liquid phase. ESwab is currently the only 317
swab with that capability. ESwabs are more expensive than traditional wound swabs, and, for 318
labs that are not using ESwabs routinely, this increase in cost should be included in the 319
operational budget for the WASP. 320
We anticipate that the relative amount of specimens that a laboratory can plate on a WASP will 321
be strongly influenced by specimen volume for a particular test and the media required for that 322
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
15
particular specimen/test. For example, if a laboratory plants an average of 5 specimens each day 323
of a particular specimen/test and that test requires media not used for other tests, adding and 324
removing the required media from the silos may not be an efficient use of time. Importantly, the 325
current operation of the WASP is designed for batch testing, not random-access testing. 326
In summary, we report a first evaluation of a new microbiology specimen-processing 327
instrument, the WASP. We were satisfied by the performance of the WASP and believe that it 328
offers opportunities to automate the processing of microbiology specimens to an extent that has 329
not been possible to date. Additional studies by this and other laboratories are merited to test the 330
full potential of the instrument. 331
332
333
334
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
16
335
336
337
338
339
340
341
342
343
344
345
346
347
348
REFERENCES 349
350
1. Albers, A. C., and R. D. Fletcher. 1983. Accuracy of calibrated-loop transfer. J. Clin. 351
Microbiol. 18:40-42. 352
2. Glasson, J. H., L. H. Guthrie, D. J. Nielsen, and F. A. Bethell. 2008. Evaluation of an 353
Automated Instrument for Inoculating and Spreading Samples onto Agar Plates. J. Clin. 354
Microbiol. 46:1281-1284. 355
3. Jones, G. A., and R. B. Matthews. 2008. Abstr. 108th General Meeting American 356
Society for Microbiology., Boston, MA. 357
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
17
4. Kass, E. 1956. Asymptomatic infections of the urinary tract. Trans. Assoc. Am. 358
Physicians 69:56-63. 359
5. Lue, Y. A., C. J. Krause, and A. B. John. 2002. Abstr. 102nd General meeting of the 360
American Society for Microbiology, Salt Lake City, Utah. 361
6. Stamm, W., G. Counts, K. Running, S. Fihn, M. Turck, and K. Holmes. 1982. 362
Diagnosis of coliform infection in acutely dysuric women. N Engl J Med 307:463-468. 363
7. Tilton, R. C., and R. W. Ryan. 1978. Evaluation of an automated agar plate streaker. J. 364
Clin. Microbiol. 7:298-304. 365
366
367
368
369
Figure Legends 370 371
Fig. 1. Front view of the WASP. The dimensions of the instrument are 75 inches wide, 75 inches 372
high and 43 inches deep. 373
Fig. 2. Top view of the WASP. The two SCARA robots are referred to as Tarzan and Jane. 374
Fig. 3. Triquetra Loop. From left to right, 10-ul, 30-ul, and 1-ul loops. 375
376
377
378
379
380
381
382
383
384
385
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
19
Figure 1 407
408 409
410
411
412
413
414
415
416
417
418
419
420
421
422
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
20
Figure 2 423 424
425
426
427
428
429 430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
21
Figure 3 447 448
449 450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
Table 1. Summary of replicate plating results for significant isolates.
1-ul Inoculum 10-ul Inoculum Total
a Same result
b
(same CFU range)c
Total Same result
(same CFU range)
Inoculab 80 80 (75) 71 71 (70)
WASP 81 80 (79) 70 70 (70)
a Total number of significant isolates for that instrument.
b Same result indicates same organism in replicate cultures with ≥10
4 CFU/ml for 1-ul
inoculum or ≥103 CFU/ml for 10-ul inoculum.
c Same CFU range indicates results of both replicates had similar quantitation (10
3 –
104 CFU/ml, or 10
4 – 10
5 CFU/ml or > 10
5 CFU/ml)
ACCEPTED
on June 21, 2018 by guesthttp://jcm
.asm.org/
Dow
nloaded from
Table 2: Results from WASP and Inoculab for specimens with significant results not in agreementa.
Inoculab WASP Loop Size Replicate 1 Replicate 2 Replicate 1 Replicate 2
Multiple Flora <10 4Mixed Flora 10
4-10
5Enterococcus
<10 4 2 other orgs.
104-10
5 Enterococcus
<104 2 other orgs
<104 Mixed Flora <10
4 Mixed Flora 10
4-10
5 Beta-strep 10
4-10
5 Beta-strep
<104 1 type <10
4 1 type 10-
410
5 Enterococcus 10
4-10
5 Enterococcus
<104 1 type No growth 10
4-10
5 Kl.
pneumoniae
<104 1 type
104-10
5 Kl
pneumoniae
<104 1 type
>105 E. coli 10
410
5 E. coli > 10
5 E. coli > 10
5 E. coli
>105 E. coli 10
410
5 E. coli > 10
5 E. coli > 10
5 E. coli
>105 Beta-strep 10
4-10
5 Beta-strep >10
5 Beta-strep > 10
5 Beta-strep
104-10
5 S. aureus 10
4-10
5 S. aureus >10
5 S. aureus 10
4-10
5 S. aureus
104-10
5 P. mirabilis 10
4-10
5 P. mirabilis >10
5 P. mirabilis >10
5 P. mirabilis
104-10
5 NLF > 10
5 NLF > 10
5 NLF > 10
5 NLF
1-ul
104-10
5 CNS
<104 1 O.T.
> 10 5CNS
<104 1O.T.
> 105 CNS
<104 1O.T.
> 105 CNS
<104 1O.T.
Multiple Flora Multiple Flora 10
3 – 10
4 CNS
<103 1 O.T.
Multiple Flora
<103 2 orgs. <10
3 2 orgs.
103 – 10
4 Enterococcus
<103 1 O.T.
103 – 10
4 Enterococcus
<103 1 O.T.
<102 Mixed Flora
103 – 10
4 CNS
<102 1 0.T.
10
3 – 10
4 CNS
<103 1 0.T.
103 – 10
4 CNS
<103 1 0.T.
103 – 10
4 E. Coli 10
3 – 10
4 E. coli > 10
4 E. coli > 10
4 E. coli
103 – 10
4 GBS 10
3 – 10
4 GBS > 10
4 GBS > 10
4 GBS
103 – 10
4 Enterococcus 10
3 – 10
4 Enterococcus > 10
4 Enterococcus > 10
4 Enterococcus
10-ul
103 – 10
4 Vir. strep.
<102 1 O.T.
103 – 10
4 Vir. strep.
<102 1 O.T.
> 104
Vir. strep.
<102 2 O.T.
> 104
Vir. strep.
<102 1 O.T.
a O.T.= other type; GBS= Group B Streptococci; CNS= coagulase negative staphylococci; NLF= unidentified non-lactose
fermenting Gram-negative bacilli.
ACCEPTED on June 21, 2018 by guest
http://jcm.asm
.org/D
ownloaded from