 Zato, kaj smo prišličili, da sem tudi prišličen, ki so vse zelo vse, in ki so vse zelo vse, začelično, kvalitivno, početno. Zato, začel, ko sem prišličen, kako se prišli, je to vse izgledanje in izgledanje in tako. Všeč, kako smo izgledali na to, kako smo izgledali nekaj. Tako, izgledanje pri mene, nekaj z taj nekaj izgledanje prišli do taj, kako se prišli, kako se prišli, kako se prišli, vseč na bolje, da je tudi tradiacija terapija. To, ko sevedaja, če biš možete vzveči všeč nekičči o toh. Zdaj, da v apartmenti tezvorejte, da biš načo in počut, da nekaj sevedajte. V tako kliniku je začina ta nekaj umreženje in da je srečen, nekaj je loviti o toh, maybe you don't care about it, or somebody told you that you have to prepare a new type of treatment or an inspection, perhaps. Anybody of you made an error in your career in radiation therapy? One, two, three, four, five, come on, don't be shy. Five, okay. And the rest, you just don't want to admit to it, or you didn't do it long enough. Or maybe in your hospital, your chief of the hospital, the medical director, your boss, or somebody else, and says, well, we heard that something like this happened somewhere and it better not happen to you. Any of you had something like this happen to you? No, yeah? No. Or we were told that and so on. So there is a number of reasons why we are talking about this, basically. And each one of us may have one or two main motivations to discuss this. So I want to tell you a story, okay? Maybe a couple of stories, basically. Once upon a time, radiotherapy accidents or incidents were so rare and far between that we basically didn't hear about them. And when we heard about anything happening somewhere in the world, okay, we always thought that, well, is that all far away? I mean, except for Santiago that knows that this is in La Pampa, in Argentina, I mean, for you it may be like on the other side of the globe altogether. And usually, the circumstances were very unique, something very special that happened. I mean, so we were very surprised about what happened. But we say, well, that doesn't apply to us. I mean, in our circumstances, this is not going to happen, okay? Except that, in reality, that was rarely the truth. Because they were usually, for every accident that we know about, they had been something similar happened either somewhere else or even in the same place many times. And so, why do we talk about this now? Why should we talk about this now and not, you know, long time ago? Well, in many, for many years, medical physicists were working in Cognito. Hardly anybody knew about what medical physicists, even people in our hospitals didn't know what medical physicists do. Is that right? And all of a sudden, I mean, a few years ago, not too long ago, we became famous. I mean, the lights turn on, shining on us. I mean, you know, we were just becoming, everybody knew about medical physicists. Why? Because there was a series of articles in the New York Times, started in the New York Times in 2010, so it's only about five years ago. And this, Vald Bogdanovich, is an excellent reporter, and he did a series of articles in the New York Times that became national attention, not only in the United States, but everywhere in the world, basically. Any of you didn't hear about this? This series of articles? Really? Oh, I'm surprised, because, I mean, everywhere I talk to, I mean, knows about the articles in the New York Times. Well, I mean, the story is that in an institution, only a subway ride away from where I live, very respectable institution with a lot of medical physicists, board certified and so on, they had an accident where this man was being treated for head and neck cancer, and basically a terrible accident happened, and he was killed, basically, by this. They were treating him with IMRT. They left the MLCs open. There was a problem in the QA. They changed the plan. They didn't do the QA immediately. They didn't turn out right. They said, well, we'll do it next day. It was the end of the week. So they said, well, we'll repeat it on Monday. When they repeated, the results were bad. They said, well, we'll look into that further. By the time they looked into that again and found that there was a problem, the patient was treated three fractions, supposedly IMRT, but the MLCs were totally open. So they give ten times the dose in an area much bigger than intended. The result is that the patient died sometime later. So this is what started it. But two days after the first article, he writes another article in The New York Times, and he talks about medical physicists. They check the equipment, but who checks on them? Who knows what they are doing? Maybe they just don't know what they are doing. Well, it turns out that the medical physicists involved were very serious medical physicists, well-trained, very responsible, but there were a series of circumstances where this happened. As part of that series, Walt Bogdanic went into the state records of New York state and put a series of reports or took some information about different cases that happened in New York state. I mean, after all, he lives in New York. And this is the title of the article on January 24. New York state, the most stringent regulator of radioactive medical devices in the United States. That's his statement. And he looked at the records in the state records of accidents and events that were reported to the state for a period from 2001 to 2009, about eight years, and he started reporting what were the things that were reported to the state. Well, let's go over a few of them just so you get an idea. Proced gland misidentified. That's one of the cases. What happens? The ultrasound machine that they used for the ultrasound implants, the ultrasound guided implants, was not performing very well. The vendor, they didn't have a consulting physicist to check the machine. The therapist that was helping the physician won the physician that something doesn't look right about the shape of the prostate or something. As a result, they place seeds, not in the prostate, or mostly not in the prostate. A patient treats a therapist, treats a patient on alternate dates, and it has to give two different treatments on alternate dates. He made the mistake. Two different treatments. The upper lung received one tenth of the prescribed dose, and the mediacenium goes ten times the prescribed dose. Big mistake, yeah? Well, the patient died sometime later. Radiative seed implanted in the wrong location, similar to that one. What was the thing? They didn't interpret correctly the CT scan to verify where the seeds were located after the implant. That's typically part of the process. You implant them on the ultrasound, and then you take a CT to verify the position. A patient receives the wrong treatment. Instead of treating patient A, they call patient B, maybe a similar last name. I don't know the circumstances. The dose that was supposed to be going to patient A was given to patient B. Radiative seeds measured incorrectly. They made a mistake in measuring the activity of the seeds. The physicist and the radiation oncologist, they had not done a treatment plan in six years for Iridium 192. They didn't have cases. In six years, they have a case. They do it, they make a big mistake. They miscalibrate or misinterpret the calibration of the seeds. They irradiate the wrong part of the body. They had body marks on the patient. The therapist had some parameters how to go perhaps from the marks to the correct place. You have the shift. Many of you do that. Misinterpreted probably there was a couch readout. The computer says, well, it's wrong. He says, don't mean. I know what I'm doing. Just override it. Anybody heard about that? Happened to you? A computer is over reading again. Oh, I'm sorry, that's the same slide. Wrong body part being irradiated. The setup notes that were on the computer correspond to the setup notes that were written on the chart. The radiation therapy had downloaded an older photograph where that patient was treated in the previous occasion. They put the wrong picture of the treated site in the computer, made a mistake. Errors of radiation overdoses. A therapist forgot to put a wedge on the machine. They treated without the wedge. They had some monitoring device, but they ignored the readout on the monitoring device. The therapist didn't inform neither the physician or the physicists that were involved. Why? Well, they didn't have a physicist full time. The physicist was maybe one day a week, 20% of his time. The physicist will tell him maybe next week or not. A patient was treated for breast cancer and it got 50% over those, over 10 treatments. Again, they left the wedge out. The physicist didn't do the first weekly check. The hospital had problems with staffing. Temporary works were coming, covering for others. They were not familiar with the treatment, routines, procedures. Error. Again, a wrong patient treated. I'm going to just go to that, because a lot of these, as you can see, are very repeated events, similar events. An old man was treated for the esophagus, was supposed to be treated twice daily. Physics, dosimetry, and the therapist, they all failed to catch an error. The state inspection came after the fact, because they had to report it to the state. They say it inspects and inspectors to investigate. Well, they were understaffed. There was not enough staff to cover everything. A computer error was not spotted. Look at the, read what this says. This is from the official record. A patient in his forties got massive overdose of radiation to his brain, because the device that shaped and modulated the beam was mistakenly left open. A physicist didn't have a check until the treatment, and the error was displayed on the treatment screen. But two therapists didn't notice it. What do you read from this? This is the case. This is this case. From reading that, I would never imagine that this was the case involved. You cannot understand what that report means, really. Correct? Well, more cases. Again, a wedge was left out. The therapist didn't notice the computer showed that the wedge was missing. They overrode the great things. The summary of what came out from this article on New York state mistakes, and over this period of nine years, or eight years, there were 621 radiation mistakes reported. It's interesting to notice that there were a total of, these are the different causes. I mean, after every case that is reported, we try to identify what were the problems, what were the reasons, and there is essentially double the number of reasons than the number of mistakes. So, on average, there is at least two reasons for any mistake that happened. Well, this is New York state. Now, New York state, in 2001, enacted a law that requires every medical physicist to have a license to practice medical physics, the same thing as a radiation therapist or a doctor. And this is one of four states in all the United States that requires a license. I don't know about any country that requires a license. There are some regulations that say, in order to do A, B, and C, you have to have board certification, you have to pass a course, et cetera. But to have a license means that if you do malpractice and do something which is against the code of ethics of your profession, your license can be taken away. If you drive a truck on the highway and you just have an accident and it turns out that you were drunk while you were driving, they will take your license away. You cannot drive anymore, or you cannot drive for five years or something. In the real state, in that particular period of time, probably have more reported cases than many other places where you are not licensed. I know that if I had something happen to us, I would make sure that I report it, because if they find that I didn't report it, I lost my license, I lost my job. So let's talk about a couple of conceptual beliefs. I'm not saying that usually accidents in radiotherapy are very rare, and I think they are rare. It's perhaps true. People think that the majority happened long time ago in the developing world, in the under developing world. Any excuse, all kinds of reasons. And sometimes people say, well, it's linked for the low technology equipment. It's not sophisticated enough. Others say, well, it's the high technology equipment that causes the accident. It's too complicated. People don't know how to use it. Well, I want to go back, because I have a series of articles that appeared in the Ohio plane dealer. It's a paper, not as famous as the New York Times, but in 1992, how many years that is, it's almost 10 years before the New York Times. No, I'm sorry, almost 20 years in the series of New York Times. The plane dealer published a series of accidents 20 years before. There was a big uproar, and look at the titles. Dangerous medicine, deadly mistakes. People were killed. So very high profile report. I want to just go to a couple of these, because these are the subtitles on the different parts. 40 people killed, doesn't know it. NRC is the Nuclear Regulatory Commission in the United States. It regulates radioactive materials, cobalt units, HDRs, LDRs, et cetera. Doesn't regulate linear accelerators. It's a federal agency. The linear accelerators are regulated by the states, some more, some less. So there were a couple of things published in that series, and look at the subtitles. Human tragedies, the officials cover up. The government is not doing their job, et cetera. Lies, people are convicted or are not convicted, and nobody goes to jail. And there were hearings in Congress. After this series, people were in Congress saying, well, something has to be done. And there were decisions. We'll set up a commission. We'll do this. Exactly the same after the New York Times. Well, I want to just cover two accidents that were reported in that series. Does any of you heard, because I'm sure that most of us were not even born in 1974, at least the majority of us, of an accident in Riverside, Ohio. They are Ohio plane dealers. I mean, wanted to obviously cover that accident. That was, goes back to 74 years before I even went into medical physics. I think you probably didn't go into medical physics. Don't do that. Oh, my God. OK. So what was the accident in Riverside, Ohio? OK, first of all, don't expect to see anything fancy or something like that. It happened with a cobalt unit. You saw this picture on Monday, OK? A cobalt unit, very simple mechanism. You know, the dosimetry we went through that on Monday. What happens with that cobalt unit? They had a physicist for many years there. Old timer, he knew these routines, he had his own equipment, etc. No administration came. They decided to bring a new physicist on board. They hired a physicist Joel Axt, which had a PhD in physics and they had a small amount of training I think in California. They brought him to Ohio. He's the new guy on the block because they were going to install a linear accelerator. One of the earliest ones. And the previously medical physicist George Kalendine. George Kalendine, was it? George, yeah. He had a long reputation, but apparently there was not good communication between the two. The new one came to basically take his job away. So cobalt unit, this is the dosimetry planning system if you want. We have this on Monday. Everything depends on this output. The output 100 centigrade per or how many centigrade per minute you get from the machine. So everything revolves around that number and that number obviously will change with the half life of the cobalt unit. And you can basically calculate it from, you know, with your calculator or your graph. So in order to save time he was just plotting the output of the machine every month as a function of the date on a semi-lock paper. It's a straight line on a semi-lock paper. But then after he got to July 73, he ran out of semi-lock paper and he continued his plot on linear paper. Well, at the beginning nothing happens because the change is so slow that it doesn't quite make a difference. But after several months or after a year the two curves started diverging. So his output was calculated to be so much, but in reality was increasing and increasing and increasing in relation to his calculation. Well, the physician that was relatively new came to the hospital because they were about to install a linear accelerator and they started looking. I mean they were still doing mostly of the patient's cobalt and he felt that he was getting too many side effects for his patients. Too many complications. At the beginning I mean there was nothing to be allowed and Axe because he was so busy trying to plan for the accelerator and his cobalt was something so simple so trivial that he just forgot or didn't forget but he didn't do his monthly output checks. So he was just relying on that graph paper. Just that. Well, for about 27 months more than two years he didn't put a chamber to check his output. Initially he says he had the higher priority to put a new linear accelerator and look into all the things and it was new technology. So when the physician started seeing that there were clinical problems with his patients he confronts Axe and says are you sure that the output is okay? So he gave some excuse. Finally he got very upset. He says I want you to put a meter under the machine to verify the output. Okay? He put the meter, it turns out that the machine was producing about 30% more than calculated. Or something like that. Initially he didn't, but at the end when he was confronted he admitted that not only he made a mistake but he also covered it up. He was not honest about it. There were reports, they had to report it to the authorities, it was cobalt accelerator, the NRC needed to be notified and they just issue a fine. Nobody was given a big trouble, I don't remember what the fine was, but the director of the radiation oncology group the new physician that had to face all this inquiry from his hospital, from the NRC and so on. The day that he reported he died of a heart attack, he was 37 year old. Full coincidence, I don't think it's all related at all. Okay? It's possible that it's nothing to relate, but I mean so this is the case that when they looked into that they found that there were 28 patients that died of overdoses over the period of years. Some of them were not reported as a consequence of the overdose. For instance, a woman had a problem in her shoulder, she got her shoulder almost frozen, she was driving she got into a car accident after the treatments, maybe a year or two years after the treatment. She didn't die from overdose, she died from a car accident so it wasn't reported as a result of an overdose. Okay? So there were a number of cases these reporters went there and looked at this under the nuclear regulatory commission reports, there were only two people that were reported as dying from that accident. These reporters find about 26. So who was at fault here at Riverside? Okay? Asked? The new physicist? It's obvious, yeah? You all agree? He was really at fault. However, this was not his own fault, only fault. There was a team effort that made this happen. A lot of people contributed to that. The administration hired unqualified staff because he was not really qualified to do radiation therapy. There were conflicting priorities on the workload, the new Linux versus the routine work, the stuff that it's not so important. Okay? There were not enough staff, obviously, to do everything, to do one and the other. There were no external audits. There was nobody checking that the output of the machine independently was checked. There were no internal audits and there were no external audits. And there were no external audits about the complications, the clinical complication, not only about the dosimetry. And there were no review or peer review of the mortalities of the complications until something really got out of hand. And, obviously, there was not a very good quality control program because they were not using redundancy methods to check things. What would be the redundancy method in the case of the machine output? Speak up. Well, either two people to check the output of the calculation instead of with a graph paper or somebody or just putting the chamber. That's a redundancy method. You calculate and you measure. All right? The physician also ignored clinical science until it really became obvious. This was one case in that series of reports. There is another case in which I want to just spend a little bit of time. How much time do I have? That's all. OK. Time flies. OK. Well, I'll try to go over this relatively quickly. A center in Tyler in Texas there were four centers in Texas, in Georgia, in Canada and in Washington state. Diverse places in the United States were using a Terak 25 linear accelerator. Any of you heard about the Terak 25 accident? Colin heard. I heard. No? Well, I'm surprised because the Terak 25 accident became a poster case for bad engineering process, not bad engineering, the bad engineering process. As a matter of fact, all this that I'm going to tell you about is published and you can download it for free. I mean there is a woman in Boston, I believe, either at MIT or some race, published a big report on all the process, all the things that happened there, what happened, what didn't happen, what should have been happening in the design and the fabrication of this linear accelerator. The company that fabricated this machine is ACA, Atomic Energy of Canada and now it's been privatized, but at the time it was one of the most advanced linear accelerator designs. It was a high energy machine independent joes with electrons, photons you could do a lot of things with it. I worked with the Terak 20, which was before this computerized design so it was semi-computerized. But there were six massive overdoses over a period of more than two years. More than two years. And it was considered the worst radiation accidents in the history of medical accelerators until that point. You can look at that, this is one of the cases reported, is a series of articles, the IAA has published accident reports for many years. This is one of them and it's included, I believe I'm sorry, I think this is not in this one, but you can download it, it's a case study of major accidental exposures and you can go to this site and download it. But basically what happens is the following. In the head of this linear accelerator you have for photons you have the target, under the target you have a flattening filter, now you have some machines that are flattening filter free but that's much, much more modern. And then you have the primary collimator and you have the Jaws. When you try to treat with electrons you have the bending magnet, but the target, which is here is taken away and instead of that you have a monitoring chamber and you have the diaphragm and basically the collimator trimmers to fix the field close to the patient. In the design of the machine there is what's called the carousel. It's like a plate that has either the flattening filter or a scattering foil or other devices that will move around depending on what mode of treatment you select. If you treat electrons, if you treat photons for electrons you will have different scattering foils for even for photons you may have different flattening filters. And there is some mechanism that will tell you is the correct piece in the right place before you can treat. Now you have to remember what is the electron beam current on the target in relation to the electron beam current when you treat with electrons. Any idea? I mean you have the answer there. For photons you use almost a thousand times more electrons hitting the target. And you know the reason. First of all a lot of that the generation of photons is not very efficient and second you have then underneath that it's going to attenuate the beam before it comes out. So you have to produce all this many more electrons hitting the target than you would use if you just didn't have the target and you treat with electrons per se. So a patient was treated with electrons for breast cancer jumped out of the couch basically saying you burn me when he was treated. Ok, well even the technologies that went to the room fell that the area which was treated was warm was too warm. Now if any of you is trying to boil water for tea in a linear accelerator you would know that a linear accelerator on radiation is very inefficient heating device. So for something to raise the temperature of the tissue so it can be felt you know you are talking about at least several 5-10 degrees which is many many times more than what you would get the heat deposited in an absorber. So there was a line of events because the first time that this happened they didn't know what happened. They knew that something strange would happen but they couldn't reproduce it. It went on and there were different communications between their staff with the company ACL they reported we had a problem the answer from the company was no it's not possible never happened it's designed that you cannot it's impossible what you described is impossible. Since they were not required that was in Texas and this was an accelerator not in a radioactive machine they didn't have to report it to the NRC so there was no report they didn't report it to the state to the federal government sometime later in Ontario Canada they had another overdose with the same type of machine and then they were suspecting that maybe a micro switch that tells where the carousel is failed so they replace the micro switch they say well that everything is fine now then there was a lawsuit against the ACL in december we are talking about half a year later they had a abnormal skin reaction severe skin reaction and they reported that they tried to think they contacted the ACL impossible and no other accident like this occurred so the company didn't even make connect the two cases just to make the things short there was a malfunction error appearing on the computer malfunction 54 nobody understood what that meant nobody went to the books to try to interpret what malfunction 54 is and ACL claim that it's impossible they never had something like this they put back the machine in service they had another overdose and the physicist and the therapist managed to reproduce the error what it involved is that the physicist and I particularly I'm not in the computer era so I'm not too fast with my fingers I would have gone and find the keys and so on the therapist were so familiar with the machine they were doing the things very very fast like blind typing and when they were going from one mode photons to electrons they were typing and they selected the energy there were two energies that had the same number I think it was 16 MV and 16 MV and they changed from one to another too fast for the computer and the machine to react to that so the machine went into an unknown state that's what's called computer design and undefined state and it was able to so they were able to reproduce the error they just found the dose that they measured they put the dosimeter they put the dose that measured 4,000 times more than what they expected and truly that was because the dosimeter under responded because it was saturated ok so bottom line is that this case went on and until they pulled up that machine of service in the old world I mean they said drawing board but it took about two years to get this thing resolved two years so these are the characteristics of the report the carousel rotation, malfunction code you can read all of that and I suggest if you want to read a novel go read this because this is really important ok so who was at fault here any idea whose fault was it the manufacturer that's good you probably saw my presentation before so ACL obviously was at fault because they designed a machine that was capable of doing something like this however again this was a team effort ok the patient complaints were not investigated thoroughly thoroughly they were investigated they say well ok we don't know the clinical outcome didn't trigger a thorough investigation and inquiry three of the four clinics failed to investigate really until they found the problem and the facility didn't assume the responsibility for what happened they say well we call ACL they need to know there were not regulations for reporting this type of error and there was no communication between user groups ok I'm not going to go to all the rest of the things because these are another series of cases in that article but the mistakes were at institutions were small, large rural, academic then the biggest thing was who reports and who doesn't report and again most of the cases were not accelerators because the NRC doesn't require them now the key thing is these are the three reporters that wrote that series so how difficult was for these reporters to find out about these cases ok well they filed more than 100 requests under the freedom of information act this is a federal law that allows you as a private citizen to request from the government that they open a file on something that you are trying to get in information they filed more than 100 requests to get these things opened and they analyze like a million and a half NRC records ok these are other cases which are not in that series cases that you may have heard the IAA has given a lot of reports these are cases that reported in the IAA Oregon in Spain Costa Rica Panama France anywhere you want and these are the places where reports came out so it obviously it's a global issue and both the IAA and other world organizations have taken this very seriously and there is a publication on prevention accidental exposure in radiotherapy but the IAA I would like you to download that and read it and the World Health Organization recognizes that it's not just in radiation therapy these are accidents in medicine are relatively common so this is in accidents in general not just in medicine the IAA also publish this analysis of the causes and contributing factors so what did we learn from all this first of all that accidents happen ok when they happen there is more than one factor more than one reasoning in the thing and the other thing which we didn't learn from here but I will try to demonstrate that or show an indication of that that there is more in most accidents than the real ones that get published there is probably many many many more and there are some common factors some of them are related to training communications both internal within the institution external between institutions there are barriers there are questions of authority who can question what is being done or what happens or perhaps the lack of authority the lack of redundancies distractions numerous questions you can read all of that and the lack of clarity and the analysis on reporting of what happened because if we want to learn of what happened as you saw from even that new york state report on the accident you cannot learn anything from that report I mean that's almost useless for somebody else to learn the thing so coming soon to this theater in the next hour what can we do about this ok so we have coffee now and I continue after that or you continue ok alright so thank you for now