Evaluating Device Design and Cleanability of Orthopedic Device Models Contaminated With A Clinically Relevant Bone Test Soil
Evaluating Device Design and Cleanability of Orthopedic Device Models Contaminated With A Clinically Relevant Bone Test Soil
net/publication/282658405
CITATIONS READS
9 718
4 authors, including:
All content following this page was uploaded by Anne D Lucas on 05 July 2018.
RESEARCH
Evaluating Device Design and
Cleanability of Orthopedic Device
Models Contaminated with a
Clinically Relevant Bone Test Soil
Anne D. Lucas, Srinidhi Nagaraja, Edward A. Gordon, Victoria M. Hitchins
complex orthopedic devices that are used to However, this method did not account for
remove bone and cartilage pose a serious orthopedic material that may be trapped in or
challenge. To demonstrate that the recom- adhered to the internal surface of a medical
mended cleaning procedure consistently device nor did it provide information on the
works to predetermined specifications, locations within the models.
manufacturers must conduct validation As a follow up to the Haugen study,11
studies. 1,4
identical model devices were used, but were
Patient soil found on reusable orthopedic made with clear acrylic. These transparent
devices includes bone and cartilage, which do model devices were soiled with bone cement,
not contain large amounts of protein or water- allowed to dry, and then minimally cleaned
soluble carbon. The primary component in to mimic the worst case scenario of orthope-
bone is hydroxyapatite, Ca10(PO4)6(OH)2.8 dic debris trapped in or adhering to the These transparent model
Recently, insoluble tissue debris was found inside of a device. Methods to detect the bone devices were soiled with
in arthroscopic shaver handles, an orthopedic cement were developed. Alizarin red S dye
device used to shave cartilage and bone.2,9 provided a rapid, qualitative indication that
bone cement, allowed to
This incident demonstrated the importance bone cement was present. Micro-computed dry, and then minimally
of evaluating medical device designs for tomography (μCT) was used to accurately cleaned to mimic the
cleanability of solid debris. Using solid assess the total amount and spatial location worst case scenario
debris, such as bone and cartilage, in a of bone cement in our model devices. The
of orthopedic debris
cleaning validation assay may prove problem- μCT uses X-rays to produce non-destructive
atic because commonly used tests to measure high resolution, 3D images of the debris in trapped in or adhering
protein and total organic carbon (TOC) are these models. to the inside of a device.
Figure 1. Device design models used in this study. Direction of flow indicates the direction the bone cement was
pulled through. To better compare the design feature as it relates to cleanability, the shaded areas in each of the
device design models show where the local μCT analysis was done. The length of A1 and E1 was 1 cm, B1 and C1
was 0.25 cm, B2 and C2 was 0.75 cm, D1 and E2 was 1.5 cm. The length of the models is 10.16 cm. The bend in
the lumen was 60° for D and 90° for E. The amount of bone cement was normalized to the total volume in which
the bone cement was quantified.
to 6 ml citrate solution to facilitate working sucrose and then thoroughly crushing the
with the slurry before it hardened; after solids using a mortar and pestle. Serial
adding the liquid to the bone powder, stirring dilutions of pulverized bone cement with
with a Teflon spatula began immediately and sucrose were created by taking a known
continued for 30 seconds at which time the weight of the crushed bone cement in sucrose
bone cement was used. To simulate the use and adding it to a known weight of the
of a device with a complex design such as an sucrose followed by thorough crushing with
arthroscopic shaver, immediately after the a mortar and pestle. The limit of detection
30 sec mixing period a house vacuum (-0.8 to for the bone cement with 1 mL of 0.01%
-0.9 bar) was used to pull the bone cement Alizarin Red S dye was 10 μg bone cement.
slurry through the device. This same mixture
was made fresh for each and every soiling of Application of Bone Cement to Devices
Device Models A through E (Figure 1). Note Application of bone cement to model devices
that the arrow in Figure 1 indicates the (Figure 1) was accomplished by mixing bone
direction the bone cement was pulled cement and pulling it through the lumens of
through the device. The model devices were the various models using a house vacuum.
laid horizontally to dry overnight at room Although one can make hydroxyapatite cement
temperature before cleaning as described by with commercially available chemicals12,
Haugen et al.11 One set of model devices was it was convenient to purchase the prepared
soiled and cleaned one time; a second set was powder that was well characterized chemi-
soiled and cleaned 10 times to simulate cally. After allowing it to harden overnight,
possible accumulation of soil after multiple the model devices were gently tapped on the
uses and cleanings. lab bench to displace any loose debris. The
Models: Five individual samples for each of model devices were cleaned mimicking a
the five models, A–E (Figure 1), were used for worst case scenario as previously described,11
each series of soiling and cleaning. The clear rinsed 10 seconds with tap water (about
acrylic models were custom fabricated at the 23°C), brushed 5 times with appropriate
FDA machine shop where identical devices sized brushes, and rinsed again for 10
in aluminum were made for Haugen.11 The seconds with warm tap water. A Parafilm-
dimensions are detailed in Figure 1. oated stopper was firmly placed in the 5 mm
end of the model opening (Figure 1 the left
Dye Assay side of the device) and 0.01% Alizarin Red S
Detection of the bone cement was made was placed in the open end of the lumen and
possible by using a dilute solution of Alizarin allowed to react with the bone cement for
Red S dye (0.01% w/v in water). The presence about 15 minutes at room temperature (about
of the bone cement was estimated by meas- 23°C). Following this, the dye was removed
uring the absorption at 525 nm.13 A standard by placing the end of the device that had not
curve for bone cement was created by taking been plugged over a 15 ml test tube and
0.1 g of hardened bone cement and 0.9 g removing the plug. The absorbance of the
Table 1. Dye Assay: The average amount of bone cement (μg/ml) remaining in each of five different device designs
(see Figure 1), as determined by the qualitative dye assay for the entire device. One set of device design models was
soiled and cleaned once (left) and soiled and cleaned ten times (right).
reacted dye was read at 525 nm with the concen- model was examined. Second, once seg-
tration extrapolated from the standard curve. mented, the volume of bone cement was
Because the model devices held different quantified locally in distinct regions within
volumes of dye, the concentration of bone each model (see Figure 1). For device
cement was normalized per ml of dye used. model A, a 1 cm length in the middle of the
channel was analyzed. For device models B
Microcomputed Tomography and C, a 1 cm length at the transition
Microcomputed tomography (μCT) 100, between the large and small diameters was
Scanco Medical, Basserdorf, Switzerland) was examined. For device model D, bone cement
used to obtain high resolution (34–48 µm/ in a 1.5 cm length along the angled portion of
voxel) three-dimensional images of each the channel was quantified and a 1 cm region
device model post soiling and cleaning. at the 90 degree bend in device model E was
These gray-scale images were set to threshold investigated. Bone cement volume was then
levels with a constant value to observe the normalized by the total volume in each
spatial distribution of bone cement for two region resulting in a bone cement volume
separate analyses. First, the entire acrylic fraction (mm3/cm3).
Figure 2. Device design models C (Figure 1), one with no residual bone cement (1), one soiled and cleaned once
(2), and one soiled and cleaned ten times (3). The Alizarin Red S dye changes from yellow (no bone cement,
λmax 425nm) to red (with bone cement, λmax 525nm). The μCT image of model C is shown in column 4.
Table 2. μCT Entire model: The average amount of bone cement (mm3/cm3) remaining in each of five different
device design (see Figure 1, five replicates for each model each soiling and cleaning set) as determined by μCT for
the entire device. One set of device design models was soiled and cleaned once (left) and soiled and cleaned ten
times (right). Student T-test results show any significant difference of the various device design models relative to A.
ANOVA analysis for the set soiled and cleaned once is P=0.0077, and for the set soiled and cleaned ten times is
P=0.0000.
model A. Student T-test comparison between diameter, 5 mm, could be used on either end
the five different device designs show that of the model. It is important to note that 2 of
models B–E were significantly different than the model D and 3 of the model E were devices
model A when soiled and cleaned ten times completely clogged and did not undergo a
(Table 2). There appears a trend from model full 10 soil and clean cycles. All other devices
A to B to C with increasing residual bone could be soiled and cleaned 10 times.
debris trapped in the lumen with increasing
complexity of these linear lumen models. μCT Localized Model Results
Model A is statistically significantly different Figure 3 shows the average amount of bone
than model B and C, but this is not true for cement (mm3) per volume of the model
model B compared to model C when soiled device (cm3) in the specified area in the
and cleaned ten times (p=0.116). However, lumen (denoted by the shaded area in
model D is significantly different than model Figure 1). Overall ANOVA analysis of the
E (p=0.028) when soiled and cleaned ten amount of bone cement per volume for those
times. Whether soiled and cleaned once or model device designs soiled and cleaned
ten times, model C retained on average the once, was p=0.790. However, when soiled and
most bone cement debris. After soiling, cleaned ten times, an ANOVA of the five
devices were laid down lengthwise and different device designs becomes significant
allowed to dry. This facilitated debris retained with a p=0.000. Individual student T-test
in the center of model C as the larger lumen comparison of the model devices that are
in the middle of model C could not be soiled and cleaned only once, show only C
cleaned because only brushes of smaller and D are significantly different than A
(p=0.048, p=0.011 respectively; data not
shown). However, when these model devices
One Time Ten Times P-value were soiled and cleaned ten times, the
A 0.02 25.79 0.0642 significant difference between A and all other
B 2.16 82.41 0.0000
model devices is apparent (Figure 3).
Additionally, both models D and E retain
C 9.21 154.28 0.0000
significantly more bone cement than B.
D 3.44 49.32 0.0000
Between models C, D, and E there is no
E 0.81 83.85 0.0001
significant difference as these designs had
larger internal lumens (C), and angles too
Table 3. Student T-test values for comparing the amount of
sharp to be easily cleaned with a brush. Two
bone cement (mm3/cm3) for the same model device soiled and
cleaned one time to that of soiled and cleaned ten times using of the model D and 3 of the model E were
μCT analysis for the entire device. Five individual samples for devices completely clogged and did not
each of the five models, A–E (Figure 1), were used for each undergo a full 10 soil and clean cycles.
series of soiling and cleaning.
All other devices were able to be soiled and This indicates that there is a build-up of
cleaned 10 times. bone cement with subsequent use and
cleaning for more complex device designs.
Cleaning Once or Ten Times Some areas beyond the scope of this study
The effect of soiling and cleaning these five include (1) disinfection or sterilization of
different model device designs once or ten accumulating debris within these models,
times for the entire model device was also (2) the effect of brushing that may alter There is a build-up
evaluated. Table 3 shows the student T-test internal surfaces (e.g., scratching), (3) use of
results for this comparison. Soiling and detergents and rigorous cleaning protocol,
of bone cement with
cleaning ten times results in statistically (4) the adherence of this specific bone test subsequent use and
significantly more bone cement deposition soil on a particular device material and cleaning for more
for all designs except A (p=0.064); as expected (5) the use of bone cement with blood, complex device designs.
A has a straight lumen and is the easiest to synovial fluid, and/or other common clini-
clean. Ten-fold to thousand fold increases in cal soils associated with orthopedic surgeries
bone cement were noted in the model may change the outcome of cleanability
devices soiled and cleaned ten times com- of devices, especially those with
pared to those soiled and cleaned once. complex designs.
Figure 3. μCT localized: Results of local analysis using μCT for the 5 device models soiled and cleaned 10 times;
5 individual models for each of the 5 designs were tested, error bars represent one standard deviation. Figure 1
boxed off area indicates where the localized analysis took place. Student T test was used to assess any significant
differences relative to model A.
Sterile Processing
Compare sterilization practices, prepare for accreditation
surveys, and more.