Download Final Report - CAE Users - University of Wisconsin–Madison

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts
Transcript
Device for Administration
of Intra-ocular Injections
Department of Biomedical Engineering –
University of Wisconsin-Madison
BME 400
Client: Barbara Blodi, MD
Group Members: Anthony Nelson, Mike Hallam,
Mike Swift, and Rajit Chakravarty
Advisor: Paul Thompson
12/13/02
Abstract:
This paper explored a number of different design possibilities for modifications to an
existing Becton-Dickinson 1cc syringe to allow for the user to depress the plunger with
their fingertips near the tip of the syringe. Both single use and reusable solutions are
presented. The two main mechanisms considered for applying force to the plunger are air
pressure and the use of a spring. A locking/releasing mechanism will be used to allow
the user to stop/start an injection at any time during the procedure. After considering
these possibilities, a spring driven reusable case device was chosen as the best solution
due to the ease of use and cost effectiveness.
Intra-ocular Injection 2
Problem Statement:
Our goal was to design an attachment to an existing 1cc syringe to allow for a
doctor to control the plunger from the tip of the syringe, allowing one person to
administer injections into the vitreous body of the eye.
Background information:
Doctors are currently treating many eye conditions that affect the posterior
surface of the eye, such as the retina, including age-related macular degeneration, ocular
inflammatory disease, branch and retinal vein occlusion, and diabetic macular edema.
In the past, medications were either given in pill form or as drops onto the surface of the
eye. However, these approaches often proved to be ineffective in delivering the correct
amount of medication to the back of the eye and exposed the body to possible side effects
corresponding to the particular medication. In order to target the back of the eye more
effectively, doctors began administering medications directly into the vitreous body of
the eye with a 1cc syringe. The patient is given a topical anesthetic and sterilizing eye
drop before the syringe is depressed into the eye 4 mm from the outside of the iris. The
doctor uses one hand to stabilize the eye and the other to hold the syringe similar to a
pencil. An assistant is then required to depress the plunger at the back of the syringe to
inject the medication in the eye. Typically between 0.1 and 0.3 cc of medication are
injected into the eye. These injections allow the doctor to place the medication directly at
the problem site and minimize the amount of side effects on the body due to the
medication. (Dr. Blodi 02/05/02, Final Report 05/02)
Intra-ocular Injection 3
Description of product function:
The problem with the existing procedure in administering intra-ocular injections
was that it requires the assistance of another person to inject the medication. This
practice brings about the possibility of miscommunication between the two people, which
could result in damage to the eye of the patient. The client is looking for a device that
will allow a single doctor to administer the injections with one hand, allowing their other
hand to be free to stabilize the eye for safe and effective medication injection. (Final
Report 05/02)
The product will inject medication into the vitreous body of the eye through
minimal movement of the hand supporting the syringe and without the previously
required help of an assistant. This product will use the existing 1cc syringe, as specified
by the client, with the typical injection volume of 0.1cc to 0.3cc. The modification to the
syringe should be easily operable with a single hand and will be easily reusable without
contamination.
The device should inject the medication at a slow and steady rate,
roughly 0.1cc per second, allowing the doctor to stop the injection at any time during the
procedure.
The medications being injected often differ in viscosity, with the most
viscous being similar to whole milk, so the device should be able to apply a large enough
force to push a number of different medications at an even rate. A complete list of the
product design specifications (PDS) is attached at the end of this report. (Final Report
05/02)
Intra-ocular Injection 4
Previous Work – BME 301
Alternative Solutions:
Hydraulic Pressure Pump:
Placed on the existing syringe, would be a small finger tip sized air bladder with a
small hose coming of the air sac and going back into an air tight chamber in the shaft of
the syringe, as seen below in figure 1. This airtight chamber will fill up with pressure by
the pumping of the bladder with the physician’s index finger and create a downward
force on the sliding portion that moves the medicine down the syringe. The faster one
pumps with his/her finger, the faster the injection will take place, allowing the doctor to
vary the rate of the injection during the procedure. This method helps to deliver the shot
at a variable speed. However, this method would require the attachment of an airtight seal
on the top of each of the syringes and thus making is un-reusable. Therefore, this design
was deemed unfeasible due to the amount of modifications that would need to be made to
the existing syringe as well as its lack of being a reusable solution.
Figure 1 – Hydraulic Pump
Intra-ocular Injection 5
Spring Operated Administration:
Our second preliminary design system incorporated a spring driven syringe
plunger and a pivot displacement mechanism. A spring, located at the opposite side of
the needle of the syringe, would be attached to the plunger forcing it downward, as seen
in figure 2. This spring would have a high spring constant (k) so it could deliver a large
force to the plunger for the administration of the medicine. This plunger would be
stopped by a lever arm located on the outside of the syringe with an arrangement of
several teeth-like structures protruding to it; catching the plunger at each click. This
clicking mechanism is created by the pivot point operated by the fingertips on the outside
surface of the syringe (BME paper, 02). By pressing near the pivot point by the index
finger, it allows the teeth like structures to loosen its hold on the plunger and cause the
spring to push the plunger down. When the index finger takes the force off, the teeth like
structure goes back to its normal position and catches the top of the plunger, which can
vary the speed of the injection. This design is also not a reusable solution because of the
attaching of the spring to the existing syringe and the placement of the pivot on the side
on the syringe would need to be applied to each individual syringe.
Intra-ocular Injection 6
Figure 2 – Spring Driven Device
Reusable Case Mechanism:
To make it more feasible to be reusable, our third alternative solution came about.
A larger syringe or plastic tube is the starting point to this design alternative. The tube
would need to be cut down the longitudinal side of the shaft, creating two different
halves. A hinge would then be used to attach the two halves together and allow for the
tube to be opened and shut and a locking mechanism on the opposite side would allow
the tube to remain closed. This larger capsule will be large enough to encase the existing
syringe, as seen below in figure 3. The device will have a spring in the back that would
force the plunger, administering the medication. A button will be located at the tip of the
device that will be controlled by the index finger of the hand holding the syringe. A
locking mechanism will hold the spring in place and will be disengaged when the
fingertip depresses the button, allowing the medication to be administered into the eye of
Intra-ocular Injection 7
the patient. If the doctor releases the button at any time, the locking mechanism will reengage, stopping the plunger and thus stopping the injection, similar to the spring
operated method discussed earlier.
In order to use this device, the spring will first have to be “cocked” by locking the
spring plate (plate below the spring which transfers the force of the spring to the plunger)
in position. The loaded syringe will be placed in the larger tube. The larger tube will then
be shut and locked into place. The device is now able to administer the injection. Once
the injection is through, the physician will simply open the larger tube and remove the
used syringe and dispose of it. (BME paper, 02) However, the problem with this device is
the possible inability of the teeth like structures to properly catch the spring plate.
Another shortcoming of this design is the lack of room for the lever arm to operate under.
Therefore another form of catching technique is required.
Figure #4 – Proposed solution from BME 301
Intra-ocular Injection 8
Prototype:
Due to the complexity of the design, our group used last semester to test and show
that springs could be used to accomplish the specified injection. In order to do this, we
decided to postpone work on the locking. Wood blocks were used to house the syringe
and springs in holes drilled down the center of the blocks (unable to be seen in the
picture).
A dowel with a slit down the middle is used to push the springs into
compression. A pin is then inserted into one of the small holes that have been drilled in
the face of the blocks, holding the springs in compression. The two blocks, the front one
holding the loaded syringe and the back one holding the compressed springs, are then
held together either with a clamp or simply by hand. When the pin is removed, the spring
exerts force on the end of the syringe’s plunger, forcing the liquid to be expelled from the
needle.
The bulk of last semester was spent building a device to hold the syringe and
springs in position so that the action of the spring on the syringe could be observed. At
first, we attempted to determine the amount of force that would be needed to inject the
different volumes of liquid as well as determine the spring constant (k) values. However,
we did not have equipment that was precise enough to find accurate values. Therefore,
we chose to qualitatively determine which springs were needed and worked best for the
26 and 30 gauge needles. We tested the springs by seeing how long and at what rate the
injections occurred with each different spring and chose the most even injection time and
rate. We found that to inject using the 26-gauge needle, the two springs on the left side
of the wood block on the left are used. (See figure 5) They are compressed to the 0.3
mark on the device and the injection takes roughly 3 seconds to expel all the contents of
Intra-ocular Injection 9
the syringe. To inject using the 30-gauge needle, you can either use the large spring,
which is shown on the right of figure 5, or by compressing all three springs shown on the
wood block on the left. Due to the smaller diameter of the 30-gauge needle, more force
is needed to inject the contents of the syringe. The combination of springs on the left side
of figure 5 will most likely be used in the final construction since they have the potential
for injecting with both needle sizes. The user would just have to add the smaller spring to
the device if they wanted to inject using the 30-gauge needle. Also, these springs take up
less space, making the device more compact and easier to control.
Figure #5 – Prototype for testing of springs
Intra-ocular Injection 10
Current Work – BME 400
Using the information we gathered last semester through testing the springs, we
began this semester by building a prototype that would be roughly the same size as the
finished product. We used one inch diameter PVC pipe for the outside of the case and
cut the pipe in half so that the loaded syringe could be placed inside. Once the loaded
syringe has been placed in the case, the case is closed and the spring is loaded from the
end opposite of the needle. A pin inserted through drilled holes in the side of the case
was used as a locking mechanism. A cap is then used to compress the springs against the
locking pin. When the pin is removed, the spring pushes the contents of the syringe
through the needle.
Figure #6 – Layout of Prototype with Simple Release
Due to the number of loose pieces in this prototype, we decided to enclose the
locking mechanism and spring in a case at the rear of the device. Also, an improved
locking mechanism was designed to be incorporated into the next prototype. These
changes are outlined in the next portion of the report.
Intra-ocular Injection 11
Releasing Mechanism:
The locking mechanism consists of two main components. The first component
transfers the force from the fingertip button to the internal plastic piece running the length
of the mechanism. This action ultimately displaces the rear wedge holding the locked
driving mechanism. This driving mechanism is responsible for transferring the force of
the compressed spring to the syringe plunger. The action of this component is similar to
the action of a mechanical pencil forcing lead through its tip. The button’s sloped edge is
directly in contact with the complementary sloped edge of an internal plastic piece. As
the button is pressed downward, the contact between the two sloped edges causes the
internal plastic piece to slide toward the driving mechanism. A close-up of this junction
is shown in Figure 7.
As the internal plastic piece moves towards the driving mechanism, it pushes a
shaft that runs the length of the case along with it. At the end of this shaft is another
sloped wedge that is in direct contact with a spring-loaded pin that holds the spring in its
locked position. This is the second component of the releasing mechanism. As the
wedge is pushed towards the pin, the pin is forced away from the spring, releasing the
spring’s potential energy and slowly empties the syringe’s contents. A close-up of the
pin/spring mechanism is shown in Figure 8. Notice the multiple notches located in the
spacer that the pin locks into. These multiple notches allow for the user to initially lock
the spring at any one of the notches settings, and it allows the user to stop administering
the drug at any time if he/she does not wish to inject the entire amount of medicine at one
time. To stop the administration of the drug, the user simply has to release the button
Intra-ocular Injection 12
he/she initially held down, causing the shaft and wedge to move away from the pin, and
the spring-loaded pin to stop the motion of the spring against the syringe plunger.
Figure #7 – Release mechanism for the device.
Release Mechanism Model:
To test our theory of previously discussed releasing mechanism, a wood model
demonstrating the feasibility of this idea was constructed. The button piece was able to
slide down against a bottom piece with complementary angles to the button. The bottom
piece was attached to a metal beam with another wood block with a gradual slope. When
the button was pressed down, the beam was forced away from the button and thus with
Intra-ocular Injection 13
the gradual slope of the other wood block was able to push a wedge piece upwards. This
wedge piece was forced down by a spring on top to hold a block, representing the driving
mechanism from figure 7, in the cocked position. When the wedge piece was forced
upwards, the cocked spring was allowed to release the block, which in turn would depress
the plunger of the syringe. This model helped verify the feasibility of our proposed
releasing mechanism. This model can be seen below in figure 8.
Figure #8 – Releasing mechanism model.
The angles used on our prototype were chosen to provide the required
displacement of the wooden wedge releasing the locking mechanism according to the
design of this model. Before the final plastic releasing mechanism is constructed, the
appropriate angles for the contact points and forces of the springs involved need to be
calculated. We did not perform these calculations because we were uncertain of the
materials we will use in the final design. The actual materials used will depend on: cost
and availability to the ME shop.
Intra-ocular Injection 14
Preliminary Design Solution:
The next design we considered was an enhanced version of an earlier prototype
concept. The major improvements in this design focus on the syringe casing and the
releasing mechanism. The casing and releasing mechanism were the primary areas that
we focused on improving from the earlier design in order to better accommodate Dr.
Blodi’s needs specified earlier. The enhanced design is pictured below in Figure 9.
Figure #9 – Diagram of the preliminary design solution
The syringe casing needed to be altered for two fundamental reasons.
The
diameter of the case needed to be relatively small and the casing needed to be convenient
to reload. The device must be held in a similar manner to the way one would hold a
pencil. Therefore, the diameter of the casing needed to be no bigger than the diameter of
an EXPO marker (approx. 0.75”). The smaller diameter aided in the grip of the unit and
allowed for better control while making the injection. The case also only has one-half of
its length on a hinge that can be opened. This is the area where the old syringe can be
Intra-ocular Injection 15
disposed of and where a new syringe can easily be placed for the next treatment. The
inside of this portion of the case is lined with a spongy material so that the syringe is fit
snuggly ensuring that the needle remains stationary throughout the injection process. The
second half of the case is sealed aside from a “screw top” end that can be opened only
when needed. This end is not designed to be opened often because it houses the spring
and locking mechanism that drives the plunger of the syringe forward upon the user
pressing the release button. We felt that this area should remain closed for the majority
of its lifetime in order to ensure that the spring did not get dislodged or that foreign
substances did not get into the driving mechanism thus decreasing its effectiveness. The
closed driving mechanism also allows for easy reloading for the new procedure. After a
new syringe is placed into the hinged half of the case the spring can easily be pushed
back and locked into place; ready for the user to press the releasing mechanism and
administer the drug (this locking action and release mechanism is described later). The
“screw top” end is incorporated into the design allowing the user to access the driving
mechanism if desired for sterilization or maintenance purposes.
Along the backside of the case (side opposite the hinged door) is a slight ridge
with a button located near the fingertip of the user holding the device like a pencil. This
ridge encloses the releasing mechanism the transfers the button press down the shaft of
the case, to the enclosed driving mechanism. A diagram of this releasing mechanism is
shown in Figure 7.
Before beginning construction on a final prototype, our group met with Dr. Blodi
to discuss the current design on November 22, 2002. Dr. Blodi specified a number of
changes that she wanted to be changed for the final design. First of all, she wanted the
Intra-ocular Injection 16
front portion of the device to have a smaller diameter to allow for better control. Also,
she indicated that she would be injecting a maximum of 0.3 ml. Therefore, we are able to
make the back portion of the device that holds the spring considerably shorter. This
reduction in length also allows the device to be lighter, making the device easier to
control. Lastly, she stated that she would prefer to have increments of 0.05 ml rather than
the original 0.1 ml. This change allows greater flexibility of volumes being injected and
allows the user to stop the injection quicker.
Final Design Solution:
Our final design solution is a modified version of the preliminary design solution,
with the specified changes requested by Dr. Blodi in our meeting on November 22.
These proposed changes incorporated were: smaller diameter of front portion of device ~
less than 1 inch, a smaller volume being dispensed and smaller increments of 0.05 cc
rather than 0.1 cc. With these suggestions in mind, we constructed scaled drawings using
Auto-CAD program. These drawings can be seen in figure 10. Figure 10(a) is a top view
of the device where the syringe would be placed. Figure 10(b) is an end-view, looking at
the end where the needle would extend. Figure 10(c) is a side view, showing how the
releasing mechanism would be positioned. Not pictured in this figure is the top portion
which would act as a lid to enclose the syringe, creating a continuous cylinder to allow
for easy gripping of the device.
Intra-ocular Injection 17
Intra-ocular Injection 18
Potential Problems:
One potential problem with a reusable device is making sure that the device does
not become contaminated or transmit disease, bacteria, and/or viruses between patients.
The device may need to be sterilized so the materials used should be able to withstand
such a treatment. However, the only part of the syringe that actually comes into contact
with the patient is the needle, which is a one-time use, sterile needle. Therefore, after
speaking with the client, we did not consider contamination to be a critical risk. (Final
Report 05/02)
Another concern is the ability for the injection to be stopped during the process of
an injection. If a problem arises during the injection, the doctor needs to be able to stop
the injection. To guarantee this, we will need to design a fail-safe locking mechanism
that is 100% successful. This may include a back up lock or some other sort of safety
measure. (Final Report 05/02) However, after discussing this issue with our client, we
feel that our new locking mechanism design will adequately address this concern.
One other aspect of our design that we need to address is the action of the spring
while it is exerting force on the plunger of the spring. We need to make sure that there is
absolutely nothing that the spring could be caught on while it is in motion, such as plastic
debris or any other debris that could alter the motion of the spring. If the spring does get
caught on something, it could potentially jump forward if moved and dislodged. This
movement could cause damage to the patient’s eye. One way to ensure this would be to
enclose the springs so that they will not be exposed to any debris. (Final Report 05/02)
After discussing with our client, we feel that the enclosed releasing mechanism design
will prevent exposure to debris.
Intra-ocular Injection 19
Future Work:
Since the capstone design sequence is intended to be a three semester endeavor,
our objectives were to thoroughly test each component of the design. For this reason, the
first semester was devoted to adequately prove the feasibility of spring driven
mechanism. This semester we set out to design and show the viability of a
locking/releasing mechanism. Now that we have shown the functionality of these main
components, we can integrate these two components into a working prototype by the end
of next semester. Our first objective next semester will be to talk with the ME shop and
present our latest CAD drawings to determine how much it will cost to fabricate a
working prototype. Once we have a cost estimate, we will work to secure the funding
necessary to have the prototype built by the ME shop. Once the prototype is built, we
will conduct testing to ensure the proper function of the device. We also plan to contact
the WARF to explore the feasibility of obtaining a patent. If our device fits these criteria,
then it would be sent for FDA clearance for human testing, where this spring driven
reusable case device would be considered for Class II regulation because it would
warrant more than just general controls
Intra-ocular Injection 20
Appendix 1 Anthony Nelson, Mike Swift,
Mike Hallam, Rajit Chakravarty
BME 400
12/13/2002
Product Design Specifications
Title: Device for administration of intra-ocular injections
Function: Develop a syringe that allows for a safe and efficient means of
injecting medication into the vitreous cavity of the eye.
Client requirements:




Allow for thumb and forefingers to be close contact to the tip of syringe
Inject medication at a slow and steady rate
Roughly the same size as the existing syringe
External and detachable from the existing syringe (multiple uses)
Design requirements
1. Physical and Operational Characteristics
a. Performance requirements: Enclose a 1cc-unmodified Becton and Dickinson
syringe and deliver medication at a slow and steady rate, roughly 0.1cc per
second. Device must be operable with one hand while the needle is being held
in the eye so the other hand can be use to stabilize the eye.
b. Safety: Must not affect the doctor’s ability to hold the syringe steady while
injecting medicine. Eliminate need for additional assistance while injecting.
Does not need to be sterile.
c. Accuracy and Reliability: Inject medicine at a slow, steady rate. Should allow
for injections up to 0.3cc, in increments of 0.05cc.
d. Life in Service: Multiple uses, not specified by client.
e. Shelf Life: N/A
f. Operating Environment: Normal medical clinic and hospital conditions.
g. Ergonomics: Device should be easy to use and require the use of only one
hand.
Intra-ocular Injection 21
h. Size: Attach to existing syringe, fitting easily in the palm of average hand.
i. Weight: Very light (less than one pound)
j. Materials: Materials should be non-hazardous, similar to existing syringes.
Client would prefer a clear plastic material.
k. Aesthetics, Appearance, and Finish: Should be professional looking
2. Production Characteristics
a. Quantity: One (prototype)
b. Target Product Cost: $200 – 300 for a reusable device
3. Miscellaneous
a. Standards and Specifications: Needs to be certified as safe to use in clinical
settings by the FDA
b. Customer: Ophthalmologists who give intra-ocular injections. Could also be
used for other applications that would be aided by the ability to inject accurately
with the use of only one hand
c. Patient-related concerns: Must allow doctor to use easily with one hand
d. Competition: None known of at this time