RadPy is based on plugins so that anyone can contribute their analysis code to the suite. Most physicists have software tools they have written to manage the data generated in their clinical practice. If the code I have written is any example, these tools are not very user friendly. They also are usually not tested by independent users. The goal of RadPy is to provide a usable framework to distribute and test these tools. By making the tools open source, they can be checked for bugs and the overall reliability of these programs should be increased.

Anyway, I’ve been holding off on posting while I went on a binge of RadPy coding. Originally, I wanted to post here at least three times a week. I probably will not come anywhere near that now, but I will still post items as they strike my fancy. They probably won’t be on a grand scale like the “What is Radiation Therapy?” series, though.

So if you are reading this, sorry about the extended absence. I hope you stick around for the new era of sporadic updates.

In the face of the Mo-99 supply crisis, important diagnostic procedures often relating to life-threatening conditions such as heart disease and cancer are being postponed or cancelled because of the decreased volume of Tc-99m available to the nuclear medicine community. In addition, clinicians appear to be turning to older nuclear modalities with potentially less diagnostic certainty and more patient risk. Clinicians may also be foregoing nuclear medicine completely, opting for more invasive, more expensive, higher risk, surgical procedures. The nuclear medicine community seems widely affected by the supply crisis and appears to be adopting a variety of strategies to try to conserve the Mo-99 which is available.

The lack of controversy at the hearing is a good sign that hopefully indicates that the committee will report the bill to the whole House soon. On a personal note, I happened to be on an elevator yesterday when a courier was carrying a shipment of Tc-99m to a cardiologist’s office. I said to him, “I hear that you’re having a shortage of that.” “Yeah,” he replied, “in fact this is my last shipment of this. They’re having to switch over to Thallium.” As this presentation on the benefits of Tc-99m makes clear, that will result in poorer image quality and higher radiation dose for their cardiac imaging patients. It will take years for a domestic source of Mo-99 to become operational. We need to get the ball rolling right now.

The third strategy is use substitute isotopes whenever possible. The article mentions several alternatives, such as thallium-201 for heart imaging studies. However, these alternatives can be more expensive and, as is the case with thallium-201, give the patient a higher dose of radiation. In addition, even if the alternatives cost the same, apparently Medicare is refusing to cover them. The Center for Medicare Services is considering changing its position, but no decision is expected for at least 6 months.

Meanwhile the American Medical Isotopes Production Act languishes in the House Committee on Energy and Commerce. Even if it passes, it will be several years before a domestic source of Mo-99 is viable. When the Petten reactor comes back up in a few weeks, it will only be in service for 6 months before it shuts down for maintenance for an additional 6 months. It’s looking more and more like this shortage is going to be more than a temporary inconvenience.

(Via the tireless John Jacobus. You can assume that for any post of mine, there is at least a 75% chance he was the one who brought it to my attention.)

For some reason, every dosimetrist likes to work with all of the lights turned off. So, if you didn’t take the opportunity yesterday, make sure to drag your dosimetrist out of their dark, dank dungeon and into the light and congratulate them on a job well done.

Figure 1. A single 18 MV beam has a lower maximum dose than a 6 MV beam.

Figure 2. Parallel opposed beams.

Figure 3. Adding another beam directly opposed from the first limits the maximum dose.

Figure 4. Using higher energy beams will lower the dose to the normal tissue.

Figure 5. A four field beam arrangement.

So which arrangement is best? The answer is patient specific. Usually, simpler is better. The more fields there are, the longer it takes to treat the patient. As treatment time increases, the more likely it is that the patient or the patient’s internal anatomy will move and take the tumor away from the beam of radiation. When we get to patient immobilization, I will show some strategies we use to minimize this factor, but it can never be completely eliminated. Also, adding more beams may decrease the maximum dose, but it also exposes more normal tissue to radiation. Therefore, the answer is to use as few beams as it takes to treat the tumor effectively while sparing normal tissue. That answer is definitely vague, and this is where treatment planning becomes an art.

All clinics in the United States have full time staff members known as dosimetrists whose job it is to find the optimal beam arrangement for each patient. (Other countries have equivalent positions, but with different job titles). Obviously, this can be a very time consuming process. Fortunately, most types of cancer can be treated in a standardized fashion. For example, most lung tumors can be effectively treated with a parallel opposed set of beams. Breast cancer can often be treated with two tangent fields that treat the breast from either side. Sometimes, however, extra fields must be added to effectively treat lymph nodes that are at risk of disease.

A lot goes into treatment planning, and beam arrangement is just one part. In the next installments, I will discuss how blocking the beam affects the dose distribution and we will move from there into more advanced techniques.

Generally the Procurator Fiscal will receive notification of a person’s death and will investigate any which appear suspicious or where investigation is mandatory regardless of the suspicion of crime. Where the death appears to be due to a criminal act the Procurator Fiscal will initiate investigations by the police or other appropriate public authorities to enable the identification of suspects and associated evidence to enable him to prosecute the case in the Sheriff Court or for an Advocate Depute to prosecute in the High Court of Justiciary. However, if the circumstances give rise to investigation, or if the death occurred while the deceased was in lawful custody the Fiscal may choose to examine matters in greater detail. Where the circumstances justify it in the public interest, or where there is a statutory requirement the Fiscal will intimate his intention to prepare evidence for a Fatal Accident Inquiry.

If I am reading this correctly, it appears as if the inquiry could end in criminal charges being filed against the health care workers responsible.

Criminal charges for medical errors in radiation therapy are not unprecedented. Two medical physicists in Panama were sentenced to four years in prison after an error in treatment planning software overdosed 28 people, killing at least 18 of them. Also in Costa Rica, a physicist was sentenced to six years in prison after miscalibrating a Cobalt-60 unit, killing 30 people and injuring 59.

In this case, I believe criminal charges would be a gross overreaction. Leaving aside the question of whether Lisa died from her cancer or the overdose, the error occurred at a clinic that was woefully understaffed and under-equipped for the patient load they were expected to treat. According to this article from 2006, 5000 patients a year were given radiation treatments at the Beatson Oncology Clinic. That is an incredibly busy center. It accounted for almost half of all of the radiation treatments in Scotland. At the same time, a report from the Scottish government stated,

207. Using current recommendations from IPEM (Institute of Physics and Engineering in Medicine) an establishment of 58.5 WTE radiotherapy physicists is required for Scotland. The current establishment is 42.5 WTE, a shortfall of 16 WTE posts. Also 8 WTE posts were vacant as at December 2004 and therefore only 34.5 WTE were in post, less than 60% of the recommended level.

The final report on Lisa’s death acknowledged the role that understaffing played in the error.

When a horrible incident like this occurs, the first instinct is to find out who is to blame. In this instance, I believe that the blame lies in a system strained to the breaking point. Given that workload, it was only a matter of time before a grave error was made. Holding people criminally accountable who were only trying to do an impossible job would send a chilling note throughout the radiation therapy community. Needless to say, many will be watching the outcome of this inquiry, myself included.

Dr. Kao also accused the NRC of making up the dose metrics it used to classify the procedures in question as sub-standard. He claimed that terms such as D90, the dose that 90% of the prostate receives, are nowhere in the NRC’s definition of a medical event. Again, this is superficially true. As I discussed before, the NRC does not have the authority to second guess the decision of a doctor. At least twice before the Philadelphia VA program was shut down, the NRC investigated a procedure but did not take action because Dr. Kao changed his written directive of the procedure’s goals to match what the patient actually received. To avoid this happening for all of the cases in question, the NRC developed certain dose metrics to determine if the patient was given an adequate implant. It is a little ex post facto, but the metrics were based on widespread clinical practice. If they were not met, the chance of a successful treatment were greatly reduced.

Rep. Phil Roe (R – Tenn), a practicing OB/GYN for 31 years, had the most pertinent question. Rep. Roe related a conversation he had with a radiation oncologist colleague who stated that in all the prostate brachytherapy procedures he had performed, he had never placed a seed in the patient’s bladder. He asked how Dr. Kao could reconcile his statement that seeds in the bladder are a recognized risk of the procedure. (Remember, one or two seeds may be a recognized risk, but Dr. Kao placed 40 out of 80 total seeds in the bladder during one procedure.) Dr. Kao did not have a direct answer to this, blaming poor image quality on the ultrasound for the error. When the obvious follow up was asked, why did he not stop the program then, Dr. Kao stated that to do so would have meant that some patients would not have received any care at all. However, even if the patients were unable to spend 8-9 weeks receiving external beam treatments, there are numerous clinics in the Philadelphia area who could have performed the brachytherapy procedure.

One question that was not asked was why the VA Information Technology department took over a year to install a network jack to permit post-implant dosimetry. As Dr. Silverstein feared, IT got a complete pass in this hearing. There were a few statements to the effect that these events could have happened in any hospital. This is unfortunately true; many hospitals have IT groups that are not responsive to vital patient concerns. However, this does not excuse the behavior of IT at the Philadelphia VA. As Dr. Silverstein writes, “multiple people failed here, including executives, physicians, safety officers, etc. I believe responsibility needs to be fairly assigned. However, IT needs to be included.”

I’m not sure where the debate goes from here. Probably, the NRC will come out with new guidelines and better standards for reporting of dosimetric data. I think more emphasis will be made on peer review, and accrediting organizations such as the Joint Commission and the ACR will expect to see evidence of that review. The VA will, I’m sure, take a hard look at the outside contractors brought in to provide treatment. However, in regards to the specific situation at the Philadelphia VA, I expect to see nothing more except the inevitable malpractice suits. Issues that should have been explored more thoroughly, like the failures of the IT department, will not get their fair hearing. I think that the radiation therapy community as a whole has learned a valuable lesson, though. One that should have already been apparent: triple check everything.

In other Philadelphia VA news, the full House Veterans’ Affairs Committee held their hearings on the issue this past Wednesday. I’ve been busy and haven’t had time to watch the webcast, but you can see it here, along with written statements from the witnesses.

So says Princeton University Archivist Dan Linke. While moving Princeton’s chemistry library to a new building, library staff found a cabinet full of files relating to Princeton’s involvement in the Manhattan Project. Linke stored the files in a safe place while the General Counsel’s office determined if the materials were still classified. He then received a call from Princeton’s radiation safety officer informing him that some of the notebooks were radioactive. The radiation was low level, and the metal in the file cabinet shielded it completely. Princeton decided to keep the uncontaminated paper files, and dispose of the notebooks with other low level radioactive waste. Still, not your ordinary day at the library.

While the number of particles in the beam decreases at it penetrates farther, we are more concerned about how the dose to the patient decreases. Since more particles will deliver a higher dose, they are related, but the relationship is more complicated than it appears. For the most part, DNA is not damaged by the beam particles directly, but by the particles with which they interact. For example, a photon beam with an energy in the range of about 1 to 20 megavolts is most likely to interact with an atom by a process called Compton scattering. This process transfers energy from a photon to an electron which can then interact with cells in the body and kill them. These electrons are mostly scattered forward (along the beam). When a photon beam enters a patient, the dose is not delivered right at the skin, but further in the body where the Compton scattered electrons, either directly or by creating free radicals, damage the DNA. This results in what is called skin sparing, where a high energy beam of radiation delivers less dose to the skin than it does to points deeper within the body. Note that this is a greatly simplified version of the true physics of the situation (even more so than usual), but gives the general flavor of what is going on.

So what does the dose delivered to a patient actually look like? We can get an idea by measuring the dose delivered to a tank of water known as a phantom. Dose measuring instruments known as ion chambers move within the tank and show us the dose distribution. (Obviously a patient is not a tank of water, but we can correct for that in our calculations.) This figure shows a curve of dose versus depth within the phantom for a 6 megavolt and an 18 megavolt beam. Note that the point of maximum dose (called dmax) is deeper for the 18 megavolt beam. This makes sense as the higher energy photons will give more energy to the scattered electrons and drive them deeper into the water. Also note that the higher energy beam delivers a higher dose as a percentage of the maximum dose than the lower energy beam. These curves are called depth-dose curves and are the basis for radiation treatment planning.

In the next post I will show how these depth dose curves are used to plan treatments, and why we need full-time dosimetrists to do it.