This story is included in the inaugural 2014 edition of Safety Partners’ publication, “Incidents, Accidents, and Near Misses in Laboratory Research.”
THE BACKGROUND
It was 2004. I was the director of biochemistry in our lab. We were studying a protein called decorin isolated from the cornea of cows. It was a while ago but my memories of the event are still very vivid since it happened the day before Game 1 of the World Series, Sox versus Cardinals.
Decorin regulates the organization of collagen bundles or collagen fibers in the eyes. Your cornea is principally made up of collagen but unlike skin or other tissues, the thickness of the fibers and their arrangement with respect to one another are precisely controlled, aligned in such a way that the cornea is allowed to be transparent.
If decorin is not present or doesn’t perform its function then the cornea actually becomes opaque. This is a huge problem in a variety of diseases where corneas become injured, scarred, or infected with different pathogens or even microorganisms, which causes a breakdown of the structure in the cornea; the cornea then becomes opaque.
The number one cause of blindness worldwide is corneal scarring. A scar on your skin is ugly but it doesn’t completely destroy the function of the skin. A scar on the cornea completely destroys its function, to let light refract as cleanly as possible. Understanding how the cornea can repair was ultimately the driving force to our project.
THE EXPERIMENT
We had ordered several hundred freshly enucleated cow eyes from a slaughterhouse out in Western Mass. A cow eye is bigger than a golf ball but smaller than a stickball ball. We would cut the cornea, which is bigger than a human cornea, maybe an inch or so in diameter, and dissolve it in extraction solution.
In the lab we worked up a procedure for isolating decorin and for purifying it, and we were scaling up to make a large amount so we could do a variety of studies on this protein.
We were doing this very large prep—it was a big investment for the lab to go through this process. After the initial extraction, you have a large volume of soluble corneal extract, probably about four liters of liquid with detergents and things called chaotropic agents like urea to extract the protein into solution.
We had filtered this big extraction solution through a very porous filter; all the particles were gone. The next step was to pump it over a chromatography column that would bind our protein and not other proteins.
Before we could do that, though, we had to degas the four liters of extract. Degas means to remove dissolved gases—in this case dissolved air—so they don’t clog the column as they pass through. It’s a pretty standard thing to do. If you have a glass of water, and you shake it, you’ll see some bubbles form. If we took that liquid and just pumped it onto a column, those bubbles might form as the pressure from the pump built up. And then, the bubbles can basically trap the plumbing. So we put the solution under vacuum. In an hour or so, we start to see the gas, the dissolved air, being released.
THE INITIAL CONDITIONS
It was a crisp October day—the kind of fall day that conjures up memories of the brilliance of New England foliage and warm cider. I remember it was after midday, around 2:00, because I had expected this degassing process to be started first thing in the morning. We really needed to get the four liters of extract onto the column because it might take a solid day just to pass it on the column, another day to wash it, another day to extract it.
I had two researchers working on this project. One researcher couldn’t come in that day. The other was a newer researcher, and he was dragging his feet a little bit on this. I remember getting back from lunch and the newer researcher reported to me that he didn’t know what to do.
I think I had a meeting to go to in another hour and probably a lecture to give later that day. There was probably also a November grant application deadline coming up, where we needed to produce some data in order to submit our proposal. I wanted to start and get this thing done.
I supervised the lab but had not been conducting much lab work myself. When the newer researcher came to me and said, “I’m not sure what to do,” I replied, “Just give me this thing. I’ll degas it to get the process started.”
THE INCIDENT
The extract was in a large, six-liter, regular glass Erlenmeyer flask, not a vacuum flask. Vacuum flasks are thicker, and they’re qualified to be able to withstand a certain amount of pressure from the vacuum. We had a six-liter vacuum flask in the lab that we use for this kind of thing, but I couldn’t find it quickly so I figured, “Well, I’ll just get this degassed. It’s only house vacuum. How strong can that be?”
The house vacuum goes through a small trap to collect anything that spills out, and then a large rubber stopper with a hole in it. You basically have a pipette that goes through the stopper. You place this on the top, turn the vacuum on, and it creates a seal. And then slowly, you’ll start to see gas coming out.
Normal procedure would be to let the solution sit in the vacuum flask. You might put a stir bar in it and let it sit on a stir plate, stirring slowly, and degas for an hour or two. But, I didn’t have an hour or two to wait. I wanted to get this started before the day was over.
To speed up the degassing, I grabbed this six-liter flask and gave it a big swirl to help release the dissolved air. As I was swirling it, I was also giving it a close visual inspection from about six inches away to see if the dissolved air had been released. I didn’t have my lab safety glasses on. I gave it another little swirl and all of a sudden the entire flask imploded and exploded because of the pressure.
The glass and the liters of extract shot into my face and across the room. As it exploded, I knew instantly that it was the stupidest thing I’ve ever done in the lab. I closed my eyes, but not fast enough because I felt something go into my eyes.
Luckily, in September we’d just had a safety inspection in our lab, during which everyone had been shown where the eye wash station was—when you need the eye wash station, you’re in the most compromised position and generally aren’t going to be able to navigate your way there by sight.
Opening my eyes just a crack, I made my blurry way to our eye wash station, just outside the lab in the hallway. I could see the way you might see through water. I exhaustively rinsed my eyes with the eyewash for a solid ten minutes. I blinked a few times and realized my eyes and my vision were completely clear.
When I looked closely, I discovered I had 20 small cuts and a couple shards of glass in my forehead, cheeks, face, various places, but evidently, no glass in my eyes. I got lucky.
The extract had various chemicals in it. There was, to my memory, urea, which was at a fairly high concentration and can certainly have an effect on a mucous membrane, like an eye. I know there were some other agents in there. There were compounds or chemicals that are in there to inhibit protease activities, which are proteins that chew up other proteins and some chemicals to regulate ion concentration. They’re chemicals that are in things like normal consumer detergents, other detergent agents that break up the flow, but still you don’t want to get it in your eye and if you do and it’s a chemical in the lab, it should be reported. There should be formal SOPs for procedures like this, and a formal way to report and record incidents when accidents happen.
At the time, the occupational health system in place was not really as codified or robust as we have these days. I called down and reported that everything was apparently fine. I probably downplayed it and said the glass blew up, I felt a little something odd, washed it out, and it was fine. I didn’t actually go for an exam, and I didn’t actually call my own physician. In retrospect, and certainly with today’s policy at this lab, I would have had to seek medical attention.
THE CHAIN OF POOR JUDGMENT
The chain of events and decisions that led up to this accident were driven by the established culture of work pressure, my desire to get the work started, and the knowledge that I didn’t have more than an hour to spend actually helping in the lab. I let these things interfere with my judgment to find the appropriate vacuum flask, to put on the appropriate safety gear. I was trying to jump through a bunch of hoops and get this thing done in ten minutes instead of waiting an hour.
I made no conscious decision about not putting on my safety glasses. I didn’t even think about it.
I knew I should have used a vacuum flask. I knew we had a vacuum flask somewhere in the lab. I couldn’t find it. Someone probably had borrowed it. I made an active decision at that point to go ahead and use something else. That was a judgment driven by the desire to have it started before I had to go off to some other meeting.
I also reasoned that the house vacuum wasn’t what we would call very high pressure. I know there were times in the past that I put a non-vacuum-rated flask and glassware under this house vacuum and had no problem. I had a history with this process and the sense that, well, it should be fine. But, it certainly is a vacuum, and it is applying pressure to the glass. And with any piece of glassware there may be imperfections and minor nicks or fractures that you can’t see. With a larger vessel and surface area comes a greater probability of having an imperfection. All you need is one of those tiny points to break the surface tension and BOOM, it fractures, causing a catastrophic avalanche effect.
We should have had an established protocol in place. We knew we were starting this very important and urgent purification process. There was certainly a breakdown in preparation; it shouldn’t have come to a last-minute decision for something like whether to use the right flask. Our column was poured and ready to go. We should have anticipated needing the vacuum flask to do this step.
Maybe that was why the newer researcher was unable to get it started, because he couldn’t find some of the equipment he needed. When he came to me, a considerable part of my attitude was, “I don’t want to hear about why. Just get it started. Enough excuses. I want results.”
THE LESSONS LEARNED; THE LESSONS SHARED
This is a story I tell almost all of the new scientists and lab techs that join my lab. Ten years ago, I’d already had 20 years of research experience: a solid 15 years really in a lab, and another five or six years administering the lab. But even with lots of experience, even if you think you know what you’re doing, there’s a reason why we have certain safety procedures in place. I want all my researchers to understand that they can’t skip any steps when it comes to their safety.
To request a copy of Incidents, Accidents, and Near Misses in Laboratory Research, email [email protected]