Excellence in Equipment Documentation

Penelope Trunk has an interesting post on Jake Breeden's Tipping Sacred Cows which lists sacred cows in corporate life that we should reconsider:

Balance: Disguising indecision as a bland compromise that attempts to achieve many things but ends up accomplishing nothing

Collaboration: Creating a culture of learned helplessness with little individual empowerment and accountability

Excellence: Spending too much energy producing perfect work instead of developing the quick-and-dirty solution needed now

Fairness: Keeping score and evening the score to make sure no one gets more than their “fair share”

Passion: Racing down a path seeking success only to find burn-out and misbehavior instead

I think Excellence is a controversial sacred cow, so I wanted to use an example from my medical device factory.  We have a comprehensive equipment program, whenever you have a piece of equipment it will take you at a minimum two to three weeks to get it qualified.  The two to three weeks process time applies to off the shelf equipment we probably already have 15 of that we're already using. 

21CFR820.72 and 21CFR820.70(g) cover equipment requirements, 820.72 is mainly calibration, 820.70(g) is as follows:

(g)Equipment. Each manufacturer shall ensure that all equipment used in the manufacturing process meets specified requirements and is appropriately designed, constructed, placed, and installed to facilitate maintenance, adjustment, cleaning, and use.

(1)Maintenance schedule. Each manufacturer shall establish and maintain schedules for the adjustment, cleaning, and other maintenance of equipment to ensure that manufacturing specifications are met. Maintenance activities, including the date and individual(s) performing the maintenance activities, shall be documented.

(2)Inspection. Each manufacturer shall conduct periodic inspections in accordance with established procedures to ensure adherence to applicable equipment maintenance schedules. The inspections, including the date and individual(s) conducting the inspections, shall be documented.

(3)Adjustment. Each manufacturer shall ensure that any inherent limitations or allowable tolerances are visibly posted on or near equipment requiring periodic adjustments or are readily available to personnel performing these adjustments.

These requirements can be summarized as the equipment must be sustainable and qualified.  These requirements can generally be satisfied by information in the equipment manual and the process testing that you have to do anyway. 

However, as mentioned above, at my work we go far beyond the requirements, we must release a custom drawing of the equipment, custom maintenance procedure and form, - this information is in the manual, but we like to copy it into our own forms.  All of these are held to the internal standards, even though they are absolutely worthless, if I want to do any maintenance work on the equipment, I'm going to reference the manual, not the drawing an engineer threw together to meet a requirement.

A software evaluation must be completed even if the equipment obviously has no software, along with forms for installation qualification (IQ) assessments: line voltage, environment, EMF, safety, calibration, etc.  While it is necessary to perform and document an IQ, the company culture has developed tribal knowledge requirements to do so, if you don't justify the need to not validate the non-existent software properly, well you'll just have to do it again, of course the templates contain no guidance on these.  You can't justify out of measuring line voltage even though your soldering iron clearly works fine.  Operation qualifications are sometimes performed when only installation qualification is needed because justifying out of them has become difficult.

At all steps you need appropriate sign offs, which generally consist of four or five people.  While this is someone's version of excellence, it really accomplishes nothing that isn't included in the manual for an off the shelf piece of equipment. 

A review of warning letters from the FDA reveals the most common issue with equipment is not performing required preventive maintenance or calibration.  In fact, as far as I can tell, no one has ever been cited for not finding a calibrated volt meter and checking the voltage before plugging a piece of equipment in.

All the time making excellent equipment documentation is time spent not working on further understanding of the production process.  If you're spending your energy on getting approvals for a drawing you made of box oven #12, then you are not improving something meaningful.  

Tech Talk - Medical Device Particle Testing Part 3

Note: This is the final part of the medical devices particle testing tech talk, see part 1 and part 2.

Once the particulate test method has been validated, it is appropriate to start product testing.  The FDA guidance documents suggest testing finished devices subjected to sterilization, performing testing on the extremes and an appropriate intermediate size for the product matrix, and assessing both inter- and intra-lot variability.  A common way to meet these requirements is to perform testing on samples from design verification, aging, and three lots of process qualification.

The best practice would be to also test lots produced under worst case coating process conditions, which is the thickest allowable coating applied using the minimum cure time, although the FDA did not mention this.  If you do particulate testing as part of lot release testing, it is in your best interest to test the worst case coating process conditions.

It is desirable to finish as much testing as much as possible in one day; this makes the results more consistent and minimizes the amount of time spent cleaning.

A typical test format is:

1. Perform test on water with glassware
2. Perform test on water through the model without test device
3.  Perform test on water through the model after test device is cycled

To test, first ensure the water and glassware to be used are acceptably clean.  For these examples it is assumed the validated particle method used 50 ml of water.  For example, if the test includes using a syringe to inject water into your model then collecting the effluent in a beaker, use the syringe to inject 50 ml of water into the sample collection beaker and test it.  The result should show a small number of particles in the 10+ um bin and very few (i.e. 0, 1 or 2 per ml) in the larger bins.  If necessary, clean your test glassware some more and then retest.

Next, get baseline results.  This can be done by injecting 50 ml of water through the model, collecting it in the test collection container, and then performing the particulate test.  This is the baseline and should be subtracted from your test device results.  Typically, the baseline has more particulates than the glassware test, but it should still not be that many.  If you see more than 5 large particles (i.e. 50+ um), I would rinse the model with water and perform the test again.  The baseline test may be performed before every sample test, per sample group, or per day.  Any of these methods is defensible.  You should also re-determine the baseline if a test condition changes, such as a new bottle of water is used. 

Then you’ll perform your test to typical use conditions. 

As before, a typical test might be:

a.       Fill model with 10 ml water, collect any effluent in sample container

This step ensures the model is hydrated prior to use, very few endovascular procedures are performed with a system that is not hydrated.  If the system is not hydrated the devices will likely generate extra particulates.

b.      Fill guide catheter with 1 ml water, collect any effluent in sample container

This step ensures the guide catheter interior is hydrated prior to use, for the same reasons as listed above.

c.       Perform simulated use with your device which takes 4 ml of water (obviously varies by device volume), leave device in model, collect any effluent in sample container

This step is the meat of the test.  Simulated use should match the IFU and typical use.  For example, if you have a guide wire and the IFU states to hydrate it for 30 seconds, you should hydrate it for 30 seconds prior to insertion into the RHV, through the catheter and into the model (the water used to hydrate is not used in the test).  Continuing the guide wire example, the guide wire should be advanced to a clinically relevant position in the model, and then retracted, the advance and retractions should be performed a clinically significant number of times.  For a PTCA catheter, the FDA guidance suggests inflating to the maximum labeled diameter.

d.      Flush guide catheter with 10 ml of water, remove device from model, collect any effluent in sample container

This is a typical example; the guide catheter is often flushed during endovascular procedures.  Using the guide wire example, you would flush through the guide catheter because it is standard practice and you will capture any particles removed from the outside of the guide wire.  Flushing through the guide catheter ensures you collect the most particles.  Alternatively you can perform a flush through the model with the guide wire in place, but the particles generated by the guide wire in the guide catheter will not be captured.  One could also perform both flushes to be conservative.

e.      Flush model with 25 ml of water, entirely empty model into sample collection container

Flushing after the device is removed from the model ensures that any particles generated during device removal are captured.

f.        Perform particulate count matching the validation conditions

Perform the test using the method previously validated.

g.       Flush the model with water

To ensure the model is clean for the next test, flush with water.  You can determine how much water is required by testing the effluent after a flushing, or you can perform a baseline test prior to every test as mentioned above.

h.      Identify particulate (as necessary)

Identifying the type of particulate can be done to determine the source of the particulates.  It is generally only attempted when an unexpected number of large particles are detected.  TIR42 lists typical methods for particulate matter determination.  To identify the particulate you have to retain the remainder of the sample, or collect it from the particle counter effluent.  Collecting the sample from the particle counter effluent can be challenging due to the particle counter volume.

To analyze your results, subtract the baseline the sample test results.  If the baseline had a higher result than the test (resulting it a negative number) it is generally acceptable to change that bin to zero, how to deal with this situation should be discussed in the protocol.  Finally it is generally desirable to convert the results to a per device basis and determine if the results met the specification.

Tech Talk – In Vitro Medical Device Verification Testing in Blood

I previously discussed testing medical devices in blood here (in 2007!), but I think I did a poor job of it and I’d like to revisit it.

Why do you test in blood? Well for one, blood is hard to simulate, it’s a non-Newtonian fluid, and using glycerin and water don’t really do it justice, but these can work depending on the application. For another, a common blood test is to check for hemolysis, sure this is tested during a biocompatibility test, but biocompatibility tests are not performed during actual use conditions. Hemolysis may also be a part of an animal safety study that you want to check out beforehand.

Where do you get the blood from? At a slaughterhouse of course, if you can find a smaller or craft meat location in your area, they’ll probably work with you, one used to sell to us for $40 a week, and we’d take a couple gallon buckets and their workers would fill them up while we waited. They only slaughtered on certain days, so call ahead. You’ll probably find cows easier to find and work with, but there isn’t really a reason you couldn’t use pig blood.

Before we leave for the slaughterhouse we’ll set up a water bath at 37ºC to be ready when we get back. Then we’ll add anticoagulant to the blood collection bucket. We’ll use either heparin or Acid Citrate Dextrose (ACD).

Once we get the blood, we mix the bucket to ensure the anticoagulant is distributed in the blood. Heparin is prescription drug, so hit up your vet consultant or animal lab for some ahead of time. ACD you can make based on USP guidelines from commonly available chemicals (water, citric acid, dextrose, and sodium). We used around 10,000 to 20,000 units of heparin per liter of blood. Of note is heparin is used clinically (on people) more in the U.S. and ACS is used in Europe, so you could maybe argue for the use of one over the other, but you’re using animal blood, so I’m not sure if that really matters. I’ll assume we are using bovine blood for the rest of this post. If you don’t use an anticoagulant, you’ll end up with a clot bucket when you get back to the lab, just throw it away if this happens, it is not recoverable.

Time is generally of the essence so don’t stop by Chili’s on your way back to the lab. Also, just be aware that water will damage your blood cells, so it is preferable to rinse your lab ware with a bit of saline before use.

When we get back to the lab we first check the blood pH and temperature, ideally the pH is between 7.2 and 7.4. We then take a hematocrit (hct) measurement by collecting blood in a capillary tube with clay sealant to stopper the bottom (get blood before using clay). Then we centrifuge the capillary tube for a few minutes at high rpm. Once centrifuged, the capillary tube will look like this:

You’ll need a hematocrit chart. Below is a simple representation of how to measure hematocrit, you put the capillary tube on the chart, line up the clay on the baseline, you move the tube left or right until the fluid level matches the top line, then you find the line where the red blood cells stop and follow it over to read the percent hematocrit, in this case 50%.

We generally take two hematocrit measurements and average; you need two capillary tubes to balance the centrifuge anyway. 38 to 42% hct is a typical range for a study like this one, although it will vary depending on how much the animal drank before it was slaughtered, I’ve seen it come in in the low 20s, so don’t worry about the initial hematocrit too much.

We then pump the blood from the collection bucket through a saline primed arterial filter (pediatric filters have lower priming volumes) and line to remove hair and large clots and into a carboy with a plugged outlet at the bottom. We’ll set up a circuit from the bottom of the carboy to the top (through the filter) with a peristaltic pump to keep the blood circulating. Place the outlet in the blood and not above it or you’ll get a bunch of foam. At this point we’ll add saline to the blood to get the hematocrit where we want it, usually around 22% to 32%. Keeping hematocrit consistent is better than not. We’ll measure the hematocrit and adjust until we’re good, using the following formula:

 S = [(H/F)-1]xV

 Where:

 H is the initial hct,
 F is the desired hct,
 V is the original volume of blood, and
 S is the volume of saline to be added.

Once we get the hct where we want, we’ll measure pH and temperature again. At one point we were centrifuging the entire sample to remove the buffy coat layer between the serum and the cells, then mixing it back together but this proved pointless and didn’t really benefit our results or affect our testing any and it was a major pain, so I don’t recommend it.

Once prepared, we can expect the blood to last for five or so hours before it gets questionable. If we’re testing an endovascular device, we’ll pump the blood around a tubing circuit (using a peristaltic pump) and then place the device in the tubing. Preferably the tubing is a similar inner diameter to the artery or vein the device will be used in. You probably want to place the blood reservoir above the test set up and the pump after the test area. Putting the blood reservoir above the test set up ensures a more consistent blood flow. A simple set up is shown below.

We can measure the device performance in blood directly, or we may be interested in something like how much does the device damage the blood, we’ll check the serum and see how red it is in simple terms. If it gets worse over time, then we’re damaging the blood. In this case, for an accurate comparison we need to run a control at the same time. For example, if we have an elaborate pumping system, we’ll run our pump system on one closed circuit and the control (with no device) in another closed circuit and track the hemolysis of both over time.

Tech Talk – Anatomy Models and Medical Device Testing

One important consideration to take into account while designing and testing medical devices is how they’re going to be used. In the case of vascular devices, they will be used in arteries and veins and I’ll describe some of the test considerations. While performing your verification testing, you want to ensure that you have a reasonable clinical model to perform testing under.

For example you might have a device that is intended to be used in the Right Coronary Artery (RCA) as shown below (image from Wikipedia):

How do you ensure your device works correctly without using it on a person? You find an appropriate model.

Models of various anatomies are available from Elastrat and Shelley Medical Imaging Technologies. ASTM F2394 also contains a schematic of a two dimensional (2D) model recommended by the FDA for some applications. Although it is best to ensure you pick a model based on the vessel characteristics that your device will likely see (such as vessel diameter and tortuosity). Using the ASTM or another off the shelf model can get you into trouble if for example you’re testing a guide catheter and pushing it into the smallest vessels when it is only intended to sit in the aortic arch.

A 2D model is easy to construct and use, but not necessarily the most accurate, it generally consists of just two pieces of machined plastic (bottom with a clear top) with a piece of tubing inserted. The easiest model consists of just a radius as shown below (by the way, I did the diagram myself in MS PowerPoint!).

Ask your clinical source what the tightest radius your device is likely to see in-vivo and simply machine a curve of that diameter. Insert a piece of silicone tubing into that radius to simulate a vessel wall more accurately than hard plastic and you’re ready to get started (you can use pig vessels to line your tortuous pathway, but this is no fun to set up). It is easy to create 2D models of various tortuous pathways. Creganna (Medical Device Technology, May 2006) shows an example tortuous pathway in some of their coronary block model simulated use testing:

You can see the simulated aortic arch as the large looping arch (start from the top most pathway) and then into tighter arteries as the model continues.

A 2D model obviously lacks some realism as it is possible that your device will need to turn multiple planes during use. Although it may be argued that the worst case scenario is actually the 2D model as it stresses only those two dimensions, adding the third dimensions allows additional flexibility in most devices as they are generally concentric. In the case of a non-concentric model, you can perform your testing to the worst case in the 2D model, setting your weakest side up to take the most abuse. The 2D model is also arguably tougher on your device than clinical use as the give is limited to the wall thickness of the tubing you insert.

For 3D models, you’re probably better off buying one from the companies listed above, although they can get expensive, and can be damaged so you need to be careful. An example heart model from Elastrat is shown below.

Once you have your anatomy model, you generally want to further simulate clinical conditions, either by flowing 37C saline through the model, or a mixture of glycerin and water to simulate blood flow.

Now you want to run whatever tests you can using the model. For example, a guide wire or catheter turns to failure test is simple enough, just hold the distal end and turn the proximal end until the device breaks. However, to really simulate use, you should put the device through the model, and while the device is still in the model, hold the distal end then turn the device until it breaks. The addition of the model adds a bit of complexity to the test, but is a better test system.

I have seen the FDA question models, so document your clinical justification why the model you use is appropriate in the protocol. If you’re using a model that you bought, you should ask them to provide you with the justification.

Tech Talk – Guide Wires

I sometimes enjoy The Oil Drum’s tech talks, so I thought I’d give one a try.

Guide wires, or guidewires, are used in the vasculature to act as a guide for other devices, an example video shows guide wire and stent.  Guide wires are generally more flexible and steerable than devices used to treat or diagnose patients.  Typically, once access to an artery is gained, the guide wire is inserted and steered under fluoroscopy to the location of interest.  Then one or more devices (usually catheters) are delivered over the guide wire to diagnose and treat the condition.

Guide wires usually come in diameters of 0.010” to 0.035” with 0.014” being the most common (0.013” is equivalent to one french or 1/3 of a mm – so the typical guide wire is slightly over 1 french in size).  Guide wire lengths vary up to 400 cm, depending on the anatomy you want to reach and the work flow.  The flexible distal tip portion is usually 3 cm long, the slightly less flexible portion is usually 30 to 50 cm long, the less flexible proximal portion makes up the balance.

The user requirements of a guide wire are can it quickly get to where it needs to go and then can I reliably deliver devices over it without patient safety issues.

A typical guide wire construction is shown below:

In this diagram, the blue proximal section is the hypotube, similar to a syringe needle (minus the sharp end).  The hypotube is generally coated in PTFE- but other coatings are also used.  The PTFE coating allows devices to more easily slide over the guide wire and is probably the major component in device “deliverability” on the guide wire side.  It is easier to slide a catheter over a guide wire with PTFE than it is to slide a catheter over a guide wire without PTFE.  The PTFE coating of the hypotube can be a tricky part of manufacturer and is usually outsourced.  Just as your pans vary in quality of the non stick surface, so do guide wires and different variations in processing and materials can affect how well your guide wire delivers devices.

The core wire material is usually nitinol or stainless steel and tapers from the proximal end to the tip.  At the tip it is generally flattened.  The core wire affects the torquability of the device; you want the tip of your guide wire to turn in a 1:1 ratio with the proximal end.  The torquability affects the steerability of the device, can it get to where you want to go.  Nitinol core wires are harder to kink, but can lose some torquability.  The core wire is attached to the proximal end of the hypotube using solder or adhesive.

Alternative guide wire construction can look like this:

In this design there is no hypotube, the core wire is the proximal portion of the device.  Coating is applied directly to the core wire and you have better torquability because you are turning the core wire directly, instead of turning the hypotube, which turns the core wire.

In both designs, the next section is a more flexible distal section, usually consisting of a wire coil.  The connection is made to the hypotube or core wire by solder or adhesive at this point.  The coils are more flexible than the hypotube and the distal portion of the device is suited to navigate to the desired portion of the anatomy.  There may be several types of coils attached together or one long coil where the coil spacing changes.  The more proximal section of the coil is generally made of stainless steel for cost savings.  The distal tip coil of the guide wire is made of radioopaque metal, generally a platinum alloy, but a palladium alloy may also be used.  Some hospitals recycle these religiously for the $4 in platinum they can get in the tips.

At the very distal tip, the wire coil is soldered to the core wire using tin - silver solder, although in some older wires the core wire may be attached using adhesive.

The tip itself is rounded into a ball shape, which the solder or adhesive naturally forms when applied to the tip.  This allows the guide wire to follow the curve of the vessel and is generally called an atraumatic tip.  From a manufacturing point of view, manufacturers are very careful to not allow any burrs in the tip assembly for fear of vessel damage, but I have not seen any direct reports of vessel damage caused by badly made tips.

The tip itself is always shaped by the user to aid in steerability, usually to around 45 degrees.  Usually the tip is shaped by using an introducer or needle.  Here is a video of a shaped tip, although this method isn’t encouraged.  Some manufacturers offer tips pre-shaped.  One notable alternative tip construction is shown below:

In this case an extra wire has been attached to the core wire and the tip to allow better tip shapeability and can give a softer tip.  Depending on where you want to go in the vasculature different tip softness is desired, for example in the cerebral vasculature usually the softer the better.  Tip softness can be measured as tip load, the amount of force (or load) it requires to bend the tip a certain amount.  Obviously you want the tip to flex instead of damage the vessel wall.

The distal portion of the wire is coated with a lubricious coating, generally a hydrophilic coating, older wires used hydrophobic coatings, but they are becoming rarer.  The hydrophilic coating is a proprietary coating that manufacturers generally outsource and pay a considerable amount of money for.  There is some skill in developing a coating that does not come off of the guide wire, and retains its lubricity over time.

When building a guide wire you want to start at one end and work your way down, starting at the tip and working to the proximal end.  Making guide wires is not that labor intensive with the most skill you need is to be decent at solder and soldering cleaning.  You need to work at keeping them undamaged in manufacturing due to their small size and the less flexing of the wire you do the better.

There are other variations on guide wires, such as plastic coatings and different combinations of coils and hypotubes, but I’ve covered the major ones.  Guide wires are basically a commodity at this point and only the regulatory barriers to entry and the fact that physicians aren’t typically price sensitive are keeping it as a reasonably attractive market to pursue.

When performing performance verification testing on a guide wire, some common tests are turns to failure (you clamp the distal tip and turn the device until it breaks), particulate / adhesion tests (of the coating), tip tensile strength, torquability (which you turn the proximal end of the guide wire a set amount and measure how much the distal end turns), tip load, radioopacity, and deliverability (where you measure the force it takes to move a catheter along the guide wire).  DDL has uploaded this video on testing a guide wire to ISO 11070, but I’ve never used that particular test.

For validation testing you generally request feedback on navigating a guide wire to the desired location in a bench or animal model and delivering a device over the guide wire.

I was unable to find the size of the guide wire market, but every endovasculature procedure uses at least one guide wire, many times more.  Many companies outsource the manufacture of guide wire to a contract manufacturer; it is difficult to enter the market fresh at this point.  However, as a company, you want to sell a guide wire so you can get your 40% margin on an outsourced guide wire instead of the competitor getting 60%.  Also, it is generally beneficial to sell a whole suite of devices when selling to a hospital, if you sell only catheters you will have a harder time getting shelf space unless you can bundle it with other products.

More information is available at:  PCI equipment: Guidewire selection.  And the FDA chimes in with helpful guide wire tips on device safety-  always read those IFUs.

The Attribute Gage R&R

There is not that much information about the attribute gage (gauge) R&R readily available online (that I was able to find), so I thought I'd cover what I've done.  First the basics, an attribute is something you can't quantify, generally a visual inspection- is this part "red" or something like that.  While it generally seems simple, especially to the technical leads, operators can often get tripped up trying to pass or fail parts based on a description or a couple pictures.  In the red example, can operators compare against the Pantone effectively and is this repeatable?  You want to have your acceptance criteria and how you are proposing to test it with trained operators set up beforehand.

Your gage R&R should mimic what you do on the line and should be set up that way.  If the operator does an inspection on the attribute before passing the part, then a final QC does that same inspection, you have two inspections and this should be part of your testing.  You should present this as the entire package when possible.  QA types tend to freak out when they hear you would accept a 10% possibility of passing a bad part, when in reality its 0.1%.  Having the two (or more) inspections will be really helpful when you get to the acceptance criteria portion.

Test Method: You need to determine the number of operators, parts and trials.  Trials is easy, just use three, everyone does, obviously more is better, but three is generally good enough and you don't want to be looking at parts all day.  For operators, you need two, but if you have three or more lines or shifts, you can include those easily enough.  The number of parts is where it gets tricky and you're going to have to make a judgment call.  Generally medical device companies rely on some sort of confidence and reliability based on the severity or RPN of the potential failure, however, in the case of gage R&R everyone seems to follow auto industry guidelines which are usually a smaller quantity, 30 is generally defensible either way.

The Acceptance Criteria:  The key one medical device companies are concerned with is the probability of a miss, which is defined as:

• Probability of a Miss = (# times a bad part was passed) / (# of opportunities) [i.e. number of inspections]

You need to base this on what is acceptable and preferably tie this back into your risk analysis (you may have to increase sample size).  The good news is, if you're doing two of the same inspection, you can set it up so that your acceptance criteria can be something like:

• Probability of a Miss (2 inspections) = (# of times a miss by one operator that were missed by another operator) / (# of opportunities)

In that way you can argue that different types of misses will be caught by different operators, but you are only concerned with the fact that it is caught at some point in the process.  Setting it up this way significantly lowers the likelihood of a miss and allows you some flexibility.

Other Information:  I recommend keeping it simple and only requiring a certain probability of a miss in your protocol (maybe effectiveness as well).  You'll want the rest of the information you can collect documented, but that is a business decision, not a safety decision.  You can cover:

1. Effectiveness - (# of parts correctly identified) / (# of opportunities) [a low effectiveness indicates your process is probably not robust and will give you trouble over time, greater than 70% is generally acceptable]
2. Probability of a false alarm = (# times a good part was rejected) / (# of opportunities) [waste]
3. Repeatability = (# agreements) / (# parts inspected) [calculate per operator and total, if an operator has low repeatability, less than 80% or so, he or she needs retrained]
4. Reproducibility = (# agreements among all operators) / (# parts inspected)
5. Bias = (Probability of a false alarm) / (Probability of a miss) [calculate per operator and total]

The Test Setup:  For this you'll need two experts in the attribute being inspected, the experts will sort out the good parts from the bad, label them in some fashion, and randomize them.  In my experience you should have at least 25% bad parts, even though your process isn't likely to have 25% reject rate (hopefully).  It is nice to include some very marginal parts, but those can be hard to find and agree on.  Don't have the operators performing the test make the parts if you can avoid it.

The Test:  You want to set it up so an operator makes a determination and someone else records it, don't let the operators know the sample being given to them, previous results, or talk amongst themselves.  Do the testing on the line and try to keep the production pace.  During a gage R&R I find the operators tend to err on the cautious side.

The Results:  There you go, you can now say your test method is qualified and have the data to back it up.  You can also make operator decisions based on the results, maybe move one around to catch things earlier in the process, which one is the go to person, etc.  I find the attribute gage R&R easier to perform than the variable one, yet it is generally more important than a dimensional one from a safety perspective because there are more things I can't measure easily.

Why do clinical trials in the US?

I'm searching for a reason and here is the one I came up with:

• The world expert resides in the US and you really need her on your team

I couldn't come up with any more concrete reasons. Maybe you don't know the regulations oversees, but that can be easily taken care of, search the web or buy a book, it will save you in the long run. Even if most of your sales are in the US, I don't think a foreign trial hurts you (as long as your sponsor is halfway respected of course).

If you have a standard device, almost anywhere will do. If you have a new device, find a well respected doctor in Europe and see what kind of magic you can work there at half the regulatory burden and less lawsuit liability.

(Photo from Karen Horton)

Patching Microsoft Windows on medical devices

Law Firm IT has an interesting post on installing the latest Windows patches on medical devices, including this:

"...because the machines were running an unpatched version of Microsoft's operating system used in embedded devices they were vulnerable.

Normally, the solution would be simply to install a patch, which Microsoft released in October. But the device manufacturer said rules from the U.S. Food and Drug Administration required that a 90-day notice be given before the machines could be patched."

I'm not sure if that is completely correct about a required 90 day notice to the FDA before patching Windows. I'll leave that to a regulatory expert.

I do know that any changes to Windows and the medical device software has to be revalidated, at least the potentially affected parts anyway. The revalidation is obviously going to take time and effort and the rewards are often low, especially if you label your device to not be connected to the internet. You want your software guys putting in new features, not screwing around with Windows compatibility for people who are using the device off label. Additionally, the Windows patch needs to be installed, no small feat for a device that is not supposed to be connected to the internet. All of this time adds up and the bottom line is you're never going to be a step ahead of software virii on a medical device, which is why they almost all say do not connect them to the internet.

That being said, Windows with the help of one of several off the shelf software programs, such as Clean Slate, Rollback Rx, and Deep Freeze, can be fairly easily configured so that the chances of a virus are minimized, meaning that you wouldn't have to update Windows with every patch. In fact, this seems like a halfway decent mitigation (along with the aforementioned labeling) to your "device connected to the internet" hazard as part of your risk management.

(Picture by Alexander Fediachov)

Biocompatibility testing on a formerly radioactive medical device

Patrick in the comments asked a while ago:

We have a radioactive device, and the act of charging with radiation has an impact on the surface chemistry, so we need to conduct biocompat tests. Now, our device has a very rapid decay rate, and with a month or so it is effectively not radioactive any more...but testing house are not willing to test it because of the stigma of having once been radioactive. Any idea how these types of devices should be tested?

I don't understand the device and when exactly it would be used clinically, but I think you'll probably have to find some way to test to ISO 10993. Assuming the device is used clinically after the radiation has dissipated, I'd write up a nice summary report saying as much with relevant radiation references and call every testing lab I could with it. Talk and offer incentives to as many people as possible at each lab until I got a definite no from management, you might be able to get a smaller lab to do it, or even offer to do the extractions yourself or under their supervision at your facility if you have a local lab, non-GLP would be better than nothing. You could try hiring a consultant with good contacts at a testing lab as well, they might be able to persuade the testing lab better than you can. Other than that depending on what device category you fall under maybe you could do the testing yourself, although I admit this is not ideal. I don't have any other ideas beyond those, good luck.

Moved the company

We moved our happy medical device company down the road and I've been spending my time working on projects related to moving. Thrilling stuff, like changing the address on labeling and other documentation, not to mention tracking down various missing items. The cleanroom was moved and the floor done, electricity hooked up, next the whole cleanroom will be scrubbed, the filters turned on and all the furniture will be cleaned and moved in. Then it will be given a few days before its certified and the bioburden checked.

Once that is done we can get started doing installation qualifications (IQ) on the manufacturing equipment. After that we are planning on redoing any validation that involves a manufacturing procedure, but not revalidating things that involve material properties for example. Since the validations require sterile product made at the new facility, away we go. I've written enough validations the past six months I think I could work them up in my sleep. It actually feels like running in place and I think company morale is low due to lack of cash bonuses.

I have a few other projects I work on when I have time, but it is going to take a while to get back to normal after the move. I can only imagine what a nightmare it would be moving a mid-sized or larger company. Although overall it is a good thing as it means the company is doing well enough to expand and grow.

I am web published!

Medical device blogger addresses technical questions

The picture looked less geeky when I picked it out.

Our packaging design went fairly smoothly and was validated. There are couple cosmetic problems related to workmanship that need ironed out, and a few tweaks to the packaging could help that, but nothing serious. Of course, I blamed all of the problems on manufacturing not following the drawing precisely and QA for not inspecting correctly. Just kidding, I'll make the changes when I have time.

Anyway, I spent about 6 hours editing a software requirements document today and will spend about 6 more hours tomorrow because we thought it would be a good idea to let the software guy rewrite and make pretty the requirements before we submitted to the FDA. Instead of taking the already released document and editing it, he made a brand new one, and I get to reconcile the old with the new, both of which are missing major components. The joys.

End of the 510(k) line

I know how Creo Quality (sorry to post off of Creo so much, but I can't find many other device blogs) feels, the closer you get to your deadline, the more you think, we should (or think we should) have discovered these problems months ago. Our device incorporates an air detector to ensure air doesn't enter the patient's bloodstream, it is a fairly common accessory for any device with an extracorporeal circuit. We just discovered recently that the software was set up to check the air detector every half second, at our flow rates more than 50 ml of air could have gone by in half a second. Thats is why we do the validations. It is a relatively simple job to change the scan rate and then retest that it works correctly this time, probably takes about a week overall, but at this stage every delay is painful and they start to add up.

I've been fairly lucky this week though and haven't caught much flak, management has been occupied with self inflicted problems that I'm not part of. The 510(k) checklist hasn't had many tasks crossed out the last few weeks though, most of it is due to contractors or consultants having one problem or other. Unfortunately it ripples back and other stuff has to be put on hold. A big piece is due tomorrow (No MLK day break for us!), the final software V&V, we'll see how it looks. Until we have that we haven't been able to finish a lot of traceability documentation and a bit of testing that was waiting for the final software version to be signed off on.

Bench Testing

I've spent the last few weeks dealing with bench tests as we wait for other results to come in. Bench testing or performance testing gets stuck in section 18 of the 510(k), since it is in the back means they don't read it right? You will save yourself a lot of grief if you read the format guidance and make sure your protocols and reports line up nicely with their requirements, although you will need more than what is listed in the guidance, at least a scope.

For a small company there are several challenges to bench testing, all revolving around the number of people that have enough understanding to run the tests. My company has three plus one consultant that can run the majority of the tests, it is preferable to have employees sign off on everything so the consultant is out, and one of the three has the understanding to run the tests, but not the personality type to see it through. That leaves the two engineers, one of which is on vacation this week, so that leaves me for now. MD&DI sums up the who should do the bench testing very well.

The first problem is the protocol, which must be signed off before the test begins, the problem here is that no one besides the engineer authors are likely to really understand what is going on. This means no problems will be caught until the engineer testers try it for real. Sure, we've tested it some previously, but when everything is recorded things change. I wrote a protocol and discovered I couldn't hold a negative pressure I thought I could so had to change it up a bit. This means rewriting and walking around getting signatures to get it approved before I can start again. This is not much of a problem, unless it is after 3:30pm and QA has gone home for the day. Then I'm forced to wait around until they come in at 9 the next day. I have argued that by having my signature on it that the protocol has therefore been predefined and good to go, but I haven't gained much ground with that.

The next problem is that these tests take time, we are shooting for 24 hours of use. I rallied around testing for 26 hours but my boss vetoed that saying 1.5 times is standard, meaning 36 hour tests and every other day I have to come in at an awkward time (do not worry, I am getting my revenge- see below). I am amazed my wife hasn't accused me of cheating on her yet with the late night stops by work. The 1.5 times the maximum limit you're shooting for is a good rule of thumb, and appropriate here, but it doesn't work for everything, like negative pressures.

The last of my whining centers around the sample sizes that will not be high enough to make everyone happy. With limited product and limited resources, running a dozen 36 hour tests could take a month. Unless you are going to manufacture, sterilize, and shipping simulate a batch of samples yourself in the next week, complaining about sample size doesn't accomplish much. Do a reasonable job and if the FDA picks on it the most likely thing that will happen is they'll ask for more testing.

I mentioned in my previous post that the deadline slipped (still not my fault), this has given me time to come up with some additional bench testing to put in motion. I say put in motion because I was so confident I'd meet my part of the original 510k deadline that I planned a two week Hawaii trip starting one day before the deadline. Now all the loose ends will have to be tied up by my boss and the other engineer, I sorta feel guilty now, but about 20 minutes after landing it will be forgotten. I give the extra testing a 40% chance of not being done when I get back. I have to say though that the last year has been a blast and if you're an engineer with a good work ethic that can tolerate the risk of working for a smaller company then go for it.

Sterlization finally resolved

I posted previously on failing a natural product sterility (NPS) test in a lot release using ethylene oxide (ETO) sterilization, and we finally have everything somewhat resolved, well really a couple weeks ago we did, but I've been too busy to post. An ISO 11135 lot release requires a half cycle, then biological indicator (BI) and NPS testing which must come back sterile, then a full cycle and further BI testing which comes back sterile. Residuals and pyrogen samples are taken from the full cycle. In the previous post I mentioned we failed the half cycle NPS testing.

After that we made our full cycle the half cycle and tested the full cycle samples for NPS and doubled the sterilization time for the rest of the samples (this still confuses some people at the company). Unfortunately those NPS samples failed as well, or more precisely one of forty samples failed on Day 7 of 14. All BI samples were negative. After much debate and me being the ever eager middleman between management and the sterilizer we decided to double our sterilization time once again then proceed. However, we had to start all over with the samples because we had used so many in testing.

While we were building more samples we looked into the NPS testing. The most notable thing was that we discovered we were using twice as many samples as required, 40 instead of 20. 20 samples assumes you use 10 samples with aerobic media and 10 samples with anaerobic media. You can even use 10 samples if you use the membrane method instead of using the two types of media on the device. For us, each sample was tested individually and had their own large media jar. We made a couple minor, mostly for show changes to how the test was conducted.

So we built more samples and re-sterilized at a fairly ridiculous length of time for the half cycle and twice this ridiculous length of time for the full cycle, the length of time is so high it borders on what the FDA considers non standard sterilization. Hopefully we can lower the time some in an actual sterilization validation in the future. Oddly (IMO), the sterilizer doesn't charge by time in the chamber, so sterilizing for 1 hour costs approximately the same as sterilizing for 5 hours.

It took much begging, but we actually managed to get these sterilized half and full cycle in less than three weeks. We passed all the required testing and we are finally on our way. There is another part to our product sterilization that I'll get back to when its done. Talking to the 510(k) consultant afterwards he said sometimes companies don't file with sterility test results. I'm not exactly sure what they're filing with then.

Since then we've been bench testing like crazy and I have been unsuccessfully trying to delegate tests, but it seems like we don't have any people that understand the product well enough to pass most tests on to. We did have a Dec 31 filing deadline, but its been moved. I am happy to say that my disposables part isn't the thing that is holding everything up despite a completely failed sterilization.

Handling samples

As part of our company growing up, and failing a sterility test, we've put the quality department in charge of samples instead of manufacturing. In most companies I've seen, quality generally handles various samples and installs things like biological indicators (spore strips) and other special sample conditions. For us, manufacturing was doing it because the QA/QC department was stretched thin, but now somehow we have acquired more quality personnel than manufacturing personnel. A situation which immobilizes manufacturing at least 2 out of every 5 days and has 50% of them threatening to quit regularly.

Anyway, I mentioned spore strips because I wanted to detail the fun of using them. First off, if you can, use the ones in the glassine envelopes. However, for some reason, they make smaller and different shaped spore indicators, but leave them in the large envelope. This means you may have to use a type that does not come in the envelope or open the envelope and insert the sample into your product. If you own a BI company I see a market opportunity, or maybe I just can't find them and everyone else knows about them.

Because nothing is ever easy, we of course have to use the non-enveloped indicators. My first instinct is that you don't want to do this in the clean room, that seems to be just asking for trouble. (ASFAIK not a problem with the enveloped ones) One of the afore mentioned "quality" personnel will drop one on the floor and then step on it and track it around the room and we'll have to spend the next week disinfecting. Instead, manufacturing is going to bag the product and remove it from the clean room and masked and gloved QA will insert the spore indicators under a fume hood with sterile instruments as needed. They will then rebag the product and seal it immediately outside of the clean room. I ruled out using a clean bench instead of a hood because the air blowing out (instead of up) is more likely to disperse bacteria, most likely right on to the person doing the indicator placement, although I'm not sure how much I should be worried about that. I'm not very happy with this procedure and might rework it on a whim as I think more about it. Moving the product out of the clean room for the prep is likely to add some bioburden to it, although it shouldn't be much and surface bioburden is probably easier to kill, its still better not to have it in the first place.

PS I might sometimes sound mean and sarcastic to readers, but those who know me know I'm only sorta joking and don't take myself too seriously!

Failed a natural product sterility test

One part of the ETO sterilization validation we're performing involves testing the natural product sterility (NPS). This entails sterilizing the medical device and then immersing the device in a media (you can also run the media through the device). Then removing the media and waiting around to see if it grows anything. We had 40 samples tested, which is standard, I'm not sure if they mix the media together or keep it separate while waiting for stuff to grow. By testing the sterility this way you're ensuring that the product is at least as easy to sterilize as known spore strips. Once your sterilization is fully validated you can skip the NPS testing.

Anyway, our natural product sterility sample failed on day 12 of 14, looks like a do-over with some more aggressive sterilization conditions. Even if it is a false positive (day 12!!), it doesn't matter, its a non-sterile load. Resterilization (or actually now just sterilization) is a tricky mix of old full cycles becoming the new partial cycles and the sample requirements cascade from that.

There is a silver lining, we did pass the bacteriostasis/fungistasis test! Actually, while a bit of a downer we had planned for something like this previously and we have to push some things a bit harder- we need to make up some samples, but overall it is not too bad and our timeline is largely intact. My boss has had these things happen with sterilization before so he had us prepared.

Medical device questions asked

I thought I'd answer some questions that people have attempted to find answers for on this blog, keep in mind that I'm no expert- and looking at everything from a USA point of view, you should double check everything I say.

If you validate a medical device with one sterilization method (ETO), do you have to revalidate if you change methods (to gamma)?

Yes, you will almost certainly have to revalidate almost everything. I can't think of any concrete exceptions now, if I do think of some or hear of some I'll add them at a later date. I think you may be able to justify your way out of redoing some performance studies depending on what you had to show.

Do you have to sterilize all medical devices?

No, the term "medical device" covers a wide range of products including things like defibrillators, ECG machines, and external pumps. Some classifications don't require sterilization, find your classification on the FDA website starting here and go from there. A good rule of thumb is if you're going to contact something not normally exposed to outside air, then you have to sterilize it.

What do I need to get a medical device ready for clinical trials?

Medical devices manufactured to GMP and a 510(k) if you're going down a 510(k) path. If you're following a PMA path in my experience you'll need to convince the FDA that you have a quality system and manufacture to GMP, biocompatibility validation, safety validation which may include a few animal studies, and a some decent test results- along with whatever else the FDA and the hospital review board say you need.

What is the accepted standard deviation for medical devices?

I have no idea if this is related to lab results or what, but there is no standard accepted standard deviation for medical devices, however, if you go less than two, you should be prepared to justify it. This does not mean that two is required, you just have to justify it, please keep this in mind if you are transferring from the drug industry!

How can you tell if a medical device needs FDA approval?

Check the FDA web site, it really is very informative. Find your device classification on the FDA website starting here and go from there.

What is a typical day for a medical device representative?

I talk with these guys from time to time, it generally involves driving around to doctor's offices and hospitals to try to sell and then complaining that the company doesn't have any good products in the pipeline.

How many people to hire for a medical device start up?

I think I've covered this before, but I'll go for it again, 1 QA, 1 manufacturing, 1 engineer and 1 management is a starting point to build from. You really do want all of your departments covered from the start or you'll have a big mess on your hands when you try to finalize things.

Are medical devices good or bad?

They are good!

How much money to start a medical device company?

If you have a cheap, simple, 510(k) device and are somewhat knowledgeable about the field, I bet you could do it for less than $1 million. If you have some drug coated, ground breaking, research intensive device you'll need many times that.

What is the medical device standard expiration date?

There is no standard expiration date, but unless there is a good reason otherwise, you can label for 6 whole months without validating the packaging and saying you'll "hand carry" everything to the site. Hand carrying involves someone flying on the plane with the boxes. This is useful because you don't want to validate your packaging until you have a final product. On the other side of things, it used to be you tried to label for a 5 year shelf life, but I think lately a lot of products and companies are content with less, 2 or 3 years is fine.

What is a sample IQ/OQ/PQ?

I think I'll do a post on this later, until then search the MDI archives.

Moving?

The powers that be have recently started discussing moving the company down the street to a slightly bigger, nicer, cheaper place. If this happens I may have to be restrained from stabbing myself in the eye. While there are certainly good reasons to move a medical device company, the people working on validations never ever want to hear the words "We're moving." To them, this translates into "Everything you've done in the last year is now worthless." Softening the blow with a raise at that point is just good form, if anything a company moves because they're doing well, right?

Moving a medical device manufacturing facility basically means a revalidation of almost everything. Every piece of manufacturing equipment must undergo an installation, operation and performance (or process) qualification (IQ/OQ/PQ- terminology can vary from company to company) when you start to use it. These can range from simple to complex, with most taking less than a day to complete and document. Move a piece of equipment across the room and you are supposed to perform at least the IQ again (which can be very simple) - usually we do all just to be on the safe side. It can become tedious depending on how it is set up, a lefty moving a balance to the other side of the table can give you fits. (Although I don't think I've yet met a QA who kept track of exactly where a piece of equipment was placed, they've relied on Manufacturing to tell them of moves- sounds like something that should be added on the IQ form.) A move to a new building invalidates everything and on top of that you have to show that you can produce product of the same or better quality that you previously produced.

My company and I assume most other companies perform and document these qualifications in house. Engineers generally write them, a tech or engineer will perform them, and then QA will sign off on them and keep them in an equipment log. At my company its a relatively informal process and QA runs the entire show without management involvement.

The silver lining to moving the company is that you don't have revalidate sterility, biocompatibility and maybe some performance testing, but everything else I can think of now is fair game. And of course that raise.

PS: Really nice companies have downloadable sample IQ/OQ/PQs available.

Validating the packaging process

Medical Device Technology has put "How to Validate a Packaging Process" online which contains pretty thorough information. I've discussed packaging a bit before here. Medium sized companies generally have a packaging engineer while small companies get whoever they can to design packaging and hope.

All sorts of design issues can show up unexpectedly in packaging, a small wrinkle in a bag or pouch seal can mean a complete redesign or worse. My boss's blood pressure skyrockets whenever someone mentions the dye penetration test (ASTM F1929) mentioned in the article. We received a bunch of failed test samples back from a testing lab that were solid coated with dye which supposedly showed how we failed the test, except after the test is over dye coats everything so you can't tell. Ever since then he's hated the test with a passion and if you want to avoid a 15 minute rant its better to stay off the subject.

I was part of a PMA that the FDA specifically required the dye penetration test for to move forward, so you'll likely have to get through it (although you can avoid packaging validation in the early stages of a PMA if you hand carry). The dye penetration test basically consists of pouring a dye solution into the packaging and then observing its progression through the package seal in a certain amount of time. The problem is that the test requires judgment calls regarding wrinkles and has a wide time range. A small wrinkle in the seal can allow the dye to penetrate quickly through the seal.

Dye Penetration Test Results

How do you avoid this problem? There are a couple ways to avoid it completely. If you gamma sterilize you don't need Tyvek breathers and the bags won't leak if you seal them properly in the first place. It is slightly harder to seal completely plastic bags, you can end up with ridiculously large amounts of air in the bag if you're not careful. There is also a company that makes Tyvek windows and you seal the plastic bag, since the Tyvek is not on the edge, you're much less likely to have a problem (I forget the name of the company offhand, but I'll update later). The sterilization degassing time may be longer however. The other packaging option is to use a more expensive tray type instead of a bag.

How do you minimize the problem? Don't handle the packaging bags. It makes me cringe when manufacturing or QA handles the bags roughly. The other thing you can do is properly design the box to limit the movement inside- this is difficult with the larger your product is however.

Biocompatibility testing of medical devices

I spent today writing the biocompatibility testing protocol which is required for product validation, engineers writing testing protocols is what happens when you don't have dedicated regulatory or QA. I'm sure that is making some of you cringe- but I promise I write a better procedure than blog post.

For the testing, which should be done on the final device after final processing, you basically just round up ISO 10993 and look at the biocompatibility testing chart and then call around for some GLP quotes. The big three places to get biocompatibility testing done seem to be NAMSA, Apptec, and Nelson Labs. There are also regional players, in our neck of the woods its Pacific Biolabs (who have decent information on their webpage here). Most people who have been around for a while have their more liked lab that they use most often. I wouldn't call it a favored lab because the relationship is hot and cold depending on how long the lab is taking- a bit silly since most testing requires standard amounts of extraction, incubation or other times. I annoy them all depending on the whims of my betters, I probably request five quotes for every one test we actually perform, sorry if you're the one doing the quoting, but thats how it works here.

For the most part, our required tests are fairly straight forward, we do have an "Additional test which the FDA may consider applicable" note under one test. In cases like that, you can generally do the test, or spend just as much time and money justifying why you don't have to and then have the FDA tell you they'd really like to see the test done and why don't you just go ahead and do it. We're just going to go ahead and get the additional test done.

I spent a bit of time trying to convince people that Murine Local Lymph Node Assay (LLNA) was the way to go for sensitization testing. Its a newer test, that is quicker and supposedly more accurate, it costs the same, and is also supposedly better from an testing animal welfare standpoint. I made some progress, but we could revert back to the Guinea Pig Maximization test at any time someone gets antsy. We just last year switched to LAL tests from rabbit pyrogen tests.

There really isn't any trouble anticipated, we're using standard tubing, components and assembly methods for the disposable component. These materials have probably been tested thousands of times previously, which is a little depressing, but its less than $20K for the testing for our device which takes about two months, so its not a huge burden for a very nice little piece of mind. The rationale behind the repeated testing is no two processes are exactly the same so better to be safe than sorry. Once the samples get back from sterilization we're good to go and hopefully the remainder of the work required for biocompatibility validation is just sorting out test results into binders.

UPDATE: I just wanted to make clear that not all devices can get away with a relatively easy biocompatibility testing as we can. The longer the device is in contact with the patient, the more rigorous the testing and some of the tests (genotoxocity or implantation for example) can be $20K+ each and some can take 4+ months. I also clarified that the testing should be done on the final device after final processing as per the comments.