In testimony before an administrative law judge, the boy’s father, who during the scans was standing at the foot of the CT table to calm his son, recalled his growing concern as the scanning stretched on. “I said, ‘Stop this!’ ” Padre Roth recalled, noting that Knickerbocker finally stopped the scan only after he became angry.

Within a few hours, the child developed a bright red ring around his head from the massive overdose of radiation. Photographs of the left side of the boy’s face show a clear line extending from the infraorbital ridge backward through the ear and nape of the neck; a similar line extends from the infraorbital ridge through the ear on the right side.

In off-the-record comments, some state officials called it the worst case of radiation overdose of a child in the U.S.

How could something like this happen? Apparently the tech involved pressed the scan button 151 times, averaging 25 seconds between each press. Afterward, she could give no satisfactory explanation for the event. She gave various excuses including motion of the patient, mechanical failure and distraction by the parents. None of these are any excuse for intentionally irradiating a patient that many times. Her supervisor could only describe it as a “rogue act of insanity”.

The consequences were dire.

A report by the hospital’s medical physicist calculated that the boy’s absorbed radiation dose was 2.8 Gy (2,800 mSv) and possibly as high as 11 Gy (11,000 mSv). The dose the boy received compares to a range of 1.5-4.0 mSv for a normal pediatric CT study of the entire spine, according to pediatric imaging experts.

Using relevant material from the article “Estimated Risks of Radiation-Induced Fatal Cancer from Pediatric CT,” published in the American Journal of Roentgenology (February 2001, Vol. 176:2, pp. 289-296), a report by the hospital’s medical physicist concluded the child had a lifetime increased risk of a fatal cancer of 39%.

Hopefully the hearing will shed more light into what exactly went wrong. There should be a written procedure in place on how to respond to instances where the CT scanner behaves incorrectly. Often, however, scans like this are seen to be too routine to worry with a formal procedure. Technicians are expected to be able to adapt to unexpected events. If the technician does not have the necessary understanding of the system, though, the consequences can be catastrophic. In his book The Human Factor, Kim Vicente describes how a lack of understanding caused the Chernobyl accident.

The problem was that the plant designers hadn’t paid enough attention to the human factor – the operators were trained but the complexity of the reactor and the control panels nevertheless outstripped their ability to grasp what they were seeing. [The reactor operator] didn’t completely understand the effects his actions were going to have until it was too late – with devastating consequences.

As a friend of mine once told me, “the secret to life is being smarter than the machines you work with.”

Undoubtedly, the technician in this case will have her license revoked, and rightfully so. The fact that a scan that should have taken a few minutes instead took more than an hour was a warning sign that any reasonable person should have heeded. However, that won’t make the problem go away. If the CT console software is too complex for a trained technician to understand, then the software needs to be simplified or there need to be hard and fast guidelines as to when someone with a greater understanding needs to be called in.

The bill also, due to weapons proliferation concerns, will phase out the export of highly enriched uranium for the production of medical isotopes. The press release cites a study from the National Academy of Sciences showing that production of Mo-99 using low enriched uranium is technically feasible and that switching would not involve tremendous cost.

Considering the magnitude of numbers that are being thrown around in Washington these days, a little over $30 million a year hopefully won’t raise any eyebrows. Biting the bullet and switching to LEU production will lead to greater reliability in supply for Mo-99 without the attending proliferation uncertainties.

[Hat tip to auntminnie.com]

Hospitals were already rationing tests due to the shortage. In the next month or so, expect the shortage to become even more severe.

The American College of Radiology has posted an article from Reuters describing some of the ways that US hospitals are coping with the loss of needed medical isotopes. Basically, it involves wait lists, rationing or switching to more expensive or less accurate tests.

The Society for Nuclear Medicine has put together a clearing house of information and recommendations here.

One person subscribes to the notion that the Chinese symbol for crisis is also the symbol for opportunity.

In an unrelated story, Orac celebrates Homeopathy Awareness Week by pointing out a story about a British store that sells homeopathic plutonium.

Molybdenum-99 is made using neutrons from a nuclear reactor. When these neutrons strike uranium-235 nuclei, the nuclei undergo fission. Around 6% of the time, this fission results in an atom of Mo-99. This process requires large amounts of highly enriched uranium, with all the nuclear weapons proliferation issues that represents. Other processes involving neutron capture and photoneutron production are being researched. These would have the advantage of being able to use low-enriched uranium. Presently, though, all Mo-99 production is done with highly enriched uranium.

After the Tc-99m is produced, it can be attached to a molecule that has properties desired for the imaging procedure. This creates a substance that is called a tracer. For example, to image cardiac bloodflow, a tracer called 99mTc-tetrofosmin or 99mTc-sestamibi is injected into the body. It then collects in the heart, with more of it collecting where blood flow is strongest. As the Tc-99m decays, it gives off a gamma ray photon that can be detected by a camera outside of the patient. By analyzing the amount of photons detected as the camera moves around the patient, the three dimensional distribution of Tc-99m can be reconstructed. Any area of low blood flow or cardiac defect can then be seen.

A large number of nuclear medicine imaging studies rely on Tc-99m. In 2008 of 22.5 million doses of diagnostic radiopharmaceuticals were given to patients in the United States, 18.5 million or 82% of these were Tc-99m. Any disruption in the supply would therefore have a dramatic effect on diagnostic imaging. When the Chalk River reactor closed in 2007, many tests had to be postponed. Exacerbating the problem this time is that three of the other four reactors that produce Mo-99 are currently down for maintenance. The future of the reactor is in doubt as there are conflicting reports from the Canadian government. A CTV report states that some officials say the reactor will be down for three months and others say that the reactor cannot be repaired. In addition, the Canadian government has announced plans to privatize the nuclear reactor business of Atomic Energy of Canada Ltd., a Canadian Crown, or government owned, corporation.

As my headline suggests, the real problem here is not the Canadian government’s spotty record at operating the Chalk River plant. Instead it seems incomprehensible that a radioisotope that is so vital to so many procedures is only available in North America from a single source. The cost of building and operating a nuclear reactor and handling highly enriched uranium is just too large compared to the potential profit from radiopharmaceuticals, I suppose. A report from the Society of Nuclear Medicine identified several reactors that could be converted to Mo-99 production, but all of them had technical and regulatory hurdles to overcome. Even if the effort were to start today,production from any source would be at least three to four years away. Private companies have also announced plans to develop Mo-99 production capability. That effort is also years away from fruition.

The extent of the shortage of Mo-99 is still unknown, but all signs point toward a significant disruption of medical imaging procedures in the United States. We need to make it a priority in the United States to have at least two sources of medical isotopes in production as soon as possible. Until then, we are at the mercy of equipment failures at reactors all over the world.

The article at lewrockwell.com engages in similar arguments; downplaying the efficacy of chemotherapy and radiation therapy. It also mentions a paper that appeared in the International Journal of Radiation Oncology, Biology, and Physics in February of this year. Mr. Sardi states in his article in regards to radiation therapy for Hodgkin’s lymphoma that the paper “shows major deficiencies in reports which substantiate this type of treatment.” In fact, the article does not show major deficiencies, but only inconsistencies in the way that these reports are published.

The paper by Drs. Bekelman and Yahalom from Sloan-Kettering is titled “Quality of Radiotherapy Reporting in Randomized Controlled Trials of Hodgkin’s Lymphoma and Non-Hodgkin’s Lymphoma: A Systematic Review.” They looked at 61 clinical trials on the effectiveness of radiation therapy in the treatment of lymphoma. What they found was that some of the published reports lacked data needed to accurately replicate the treatment in the clinic or in future research. They looked at six pieces of information needed to ensure consistency in patient treatment. These six measures were listed by the Radiation Therapy Oncology Group, a group that coordinates clinical trials across multiple institutions, to “ensure consistency and accuracy of treatment specifications and to minimize potential variations in the conduct of trials.” The six measures were target volume definition, radiation dose specification, fractionation specification, radiation prescription point specification, QA process, and QA process adherence.

A target volume is the anatomical site that needs to be treated. This can include the tumor, adjacent lymph nodes or other nearby tissue. For example, in treating prostate cancer sometimes we will include the seminal vesicles in the target volume since the cancer often will spread to them. We then design the treatment so that the target volume is irradiated to the prescribed dose. Obviously, the definition of the target volume can have large effects on the patient’s outcome. Too small a volume could lead to local failure, where the tumor is not completely killed and begins to regrow. Too large a volume will increase the amount of normal tissue that is irradiated and can cause unnecessary side effects.

In my first two posts explaining radiation therapy I talked about the issues surrounding how much dose to give and in how many fractions. I showed how changing the fractionation scheme can have dramatic effects on patient outcome. Therefore, the total dose and fractionation scheme used is critical information needed to reproduce the treatments from a clinical trial. The radiation point prescription indicates where in the body that the doctor prescribes the dose. Without getting too much into the details of treatment planning, the location of the point (or volume) of prescribed dose can raise or lower the total dose given to the target volume, and is therefore critical to applying results from clinical trials.

The final two measures refer to the quality assurance process needed to ensure that the dose is delivered as prescribed. Without testing, it is impossible to know whether the dose as specified by the trial protocol was actually delivered. This can include tests such as special measurements made by the medical physicist on test equipment to simulate an actual patient, measurements made on the treatment device to check output and performance, measurements of dose outside the patient to attempt to infer dose to the tumor and review of treated cases by a third party to check for errors. Without specification and proof of adherence to protocols, there is no way of knowing if the results are a valid test of those protocols.

The authors found that most of the reports studied had deficiencies in one or more areas. In particular, only seven of the reports (11%) documented adherence to a quality assurance process. However, the authors state that they did not have access to the original protocols and studied only the final report. Therefore, quality assurance was likely performed but not put into the final report. For clinical trials coordinated by the RTOG, for example, detailed quality assurance must be performed and audited by a third party in order to be accepted to the trial. Regardless, the point of the study was to evaluate the published reports, as these are all a clinical oncologist may have access to. It seems clear that groups which perform clinical trials need to improve reporting in order to assure the usefulness of their data.

Now that we know what the paper actually says, we can see that Mr. Sardi’s summary is misleading. Here is the entire paragraph from his article.

Ironically, the Hauser case comes just when radiation oncologists are questioning reports on the effectiveness of radiation treatment for Hodgkin’s lymphoma. A review of 61 trial reports that were published between 1998 and 2007, published in the February 1, 2009 issue of the Journal of Radiation Oncology Biology Physics, the official journal of the American Society for Radiation Oncology, shows major deficiencies in reports which substantiate this type of treatment. The question is, can cancer doctors assure with certainty that the treatment is science-based, something the new administration in Washington DC is pushing.

This is just incorrect. Nothing in this article says anything about the quality of the clinical data, only about the reporting of the protocols used. Without the including the recommended information, it is difficult to compare studies or apply them in the clinic. However, the results of the studies would still be valid if the studies are performed properly. The deficiencies are only in the protocol information included in the final published report, not in the trials themselves. If Mr. Sardi wants to attack the usefulness of radiation therapy for Hodgkin’s lymphoma, he will have to attack each of the many studies that have demonstrated a survival benefit. This paper does not make his argument for him.

I am sure that the debate over Daniel Hauser will continue, even if he gets the treatment he needs. There is a debate to be had over how much control the government should have over our health care. Personally, I think that if an adult wants to refuse treatment that is his or her right, but a child is not mature enough to make that decision. There is a gray area, though, if the treatment has only a small chance of success. How much painful treatment should a child have to go through if it is unlikely to improve the situation? This is an important question that I hope will be explored. However, that is not the question in this case. Daniel Hauser had a extremely good chance at survival before he ran off, and he hopefully still has a good chance. No matter what proponents of alternative medicine say, chemotherapy is the best treatment for Daniel and refusing it is child neglect.

BEKELMAN, J., & YAHALOM, J. (2009). Quality of Radiotherapy Reporting in Randomized Controlled Trials of Hodgkin’s Lymphoma and Non-Hodgkin’s Lymphoma: A Systematic Review International Journal of Radiation OncologyBiologyPhysics, 73 (2), 492-498 DOI: 10.1016/j.ijrobp.2008.04.058

www.youtube.com/watch?v=0v_85tAey9s

(H/T: The Amateur Scientist via Orac)

While not life threatening, xerostomia still has a large negative effect on a patient’s quality of life. The American Dental Association gives some of the problems a patient might face.

Some of the common problems associated with dry mouth include a constant sore throat, burning sensation, problems speaking, difficulty swallowing, hoarseness or dry nasal passages. Left untreated, dry mouth can damage your teeth. Without adequate saliva to lubricate your mouth, wash away food, and neutralize the acids produced by plaque, extensive decay can occur.

While there is some indication that salivary function may eventually recover, for most patients there is no cure; only symptom management. Anything that could help relieve these symptoms would make a profound difference in these patients’ quality of life. So it probably shouldn’t be a surprise that some groups are using acupuncture to attempt to stimulate the flow of saliva.

Many studies have looked at the effect of acupuncture on various maladies. Some small scale studies have shown a positive effect. However, when follow-up studies are performed, again and again the results are found to be consistent with the placebo effect. Things are no different in this case.

Nineteen patients were given acupuncture following radiation therapy to the head and neck. Over the next 8 weeks, their salivary flow was measured and they were given questionnaires to assess their quality of life. The responses to the questionnaires indicated that quality of life had improved; the patient’s scores going from 40% to 55.6% over the 8 weeks (a higher percentage is better). However, measurements of their salivary flow indicated no statistically significant increase. The explanation:

The researchers also note that basal and salivary flow rates vary significantly among individuals, and therefore, subjective perceptions and objective measurements do not always correlate. A definitive threshold of increased saliva output that results in a clear clinical benefit has not been established, so even a small increase can provide relief to the patient.

That seems like a convoluted conclusion. The simpler conclusion would be that acupuncture did not actually cause any physiological effect, but caused the patient to believe that it would help. This is the very definition of the placebo effect. The authors even admit to this (at least somewhat).

It is also possible that the results could be attributed to a placebo effect, at least in some patients. However, the researchers do note that patients who have had major salivary glands irradiated do not spontaneously improve after 4 months, so the improvements in this trial “were most likely attributable to the acupuncture treatment rather than spontaneous recovery.”

So, who cares? Patients feel better. Even if it’s just a placebo, if it helps them out what’s the harm? There’s no real harm to the patients, aside from being unnecessarily stuck with needles. The harm comes from the diversion of resources from treatments that have a chance of leading to something better. This research was apparently funded by a grant from the National Cancer Institute. It probably was a cheap study to run, so the loss of NCI resources was not too great. However, ask anyone who depends on government funding to perform medical research; there is not enough to go around. Why continue to spend money on something that has been shown time and time again not to work?

I sympathize with the desire to try anything to help these people out, but too many promising research avenues have been cut short from lack of funding. If a private foundation wants to sponsor research into acupuncture, that’s fine. It’s possible that some positive physiological effect will eventually be found. Government money is hard to come by, though, and should be given to projects with a better chance of success.

Recently I came across a story about wait times for knee surgery in Quebec. I was unaware that there is an organization, the Canadian Institute for Health Information, tracking wait times for various procedures and documenting progress in reducing those wait times. I looked up the collected data and saw that among the statistics tracked are wait times for radiation therapy.

At the CIHI website, there are tables giving wait times and trends for 2007-2009. These wait times are broken out by province. Looking at the 2009 report gives values for the period of April to September 2008. Each province reports their results differently, but most report the percentile of patients that meet their benchmark of 28 days between the date the oncologist made a decision that the patient was ready to begin therapy and the date of the first radiation therapy treatment. Some provinces like PEI and Manitoba report that nearly 100% of their patients start treatment within that time period. On the other end, Newfoundland and Saskatchewan report that 85% percent or fewer of their patients meet that benchmark. Unfortunately, while the report indicates that wait times for other types of medical are decreasing, radiation therapy wait times are either steady or climbing.

There will always be some delay between the oncologist decision and the first treatment. The treatment needs to be planned by a dosimetrist and approved by both a physicist and the physician. For complex treatments, this can take longer than 7 days, but the vast majority of treatment plans can be completed in this time given adequate staffing levels. The clinics I am familiar with in the USA treat most patients within 7 days and virtually everyone in 14 days or less.

The report does not explain why wait times are so long in Canada, but surely some part of it is a lack of resources. A report from the Wait Time Alliance, an umbrella group of Canadian physician organizations, blames a shortage of equipment and staff. The report states that staff are “working late into the evenings”, but that can only go so far. A linear accelerator needs some down time for maintenance and safety checks from the physicist. Beyond a certain number of patients, the only solution is to buy another linac. However, linacs are expensive both in their upfront cost and in maintenance.

While it is clear that wait times in Canada are longer than those in the US, it is unclear that patient outcomes are worsened because of it. The two types of cancer that make up the bulk of radiation therapy patients are breast and prostate. Some forms of both prostate and breast cancers are non aggressive and those patients will probably see no difference in outcomes if treated a few weeks later.

An interesting study would be to look at outcomes in the US and other countries for various tumor types. The hard part in such a study is controlling for demographic differences (these differences are what lead to the US having such a comparatively high rate of infant mortality). However, this is the type of info we need in order to make an informed decision about the future of our health care system. Otherwise, we’re left with dueling anecdotes.

Their latest update, commenting on the White House briefing with Homeland Security and the CDC, is here.