Read part 1 of this interview: “Your DXA Questions Answered by Larry Jankowski CBDT”
Biographical Sketch - Larry Jankowski, CBDT
Larry Jankowski is the chief DXA technologist and a research study coordinator for Illinois Bone and Joint Institute, a physician group practice with 19 locations, 81 orthopedic surgeons, and 7 rheumatologists in Chicago, and its north and westerns suburbs. He has been performing DXA scans for over 25 years. He is certified by the International Society for Clinical Densitometry (ISCD) since 1996 as a CBDT. He also is an instructor for the ISCD/IOF Osteoporosis Essentials courses in both the United States and internationally. Larry was the first DXA technologist elected to the ISCD Board of Directors in 1996, and has held the office of treasurer from 1997 to 2000, and again from 2010-2013. He served as an expert panelist for several ISCD Position Development Conferences, and has presented posters at bone meetings including the ISCD, NOF, ASBMR, and AAOS.
He is also on the editorial board for the Journal of Clinical Densitometry, is a member of the ISCD Facility Accreditation Council, and a co-editor of the ISCD "Case of the Month" e-learning project and also volunteers as a moderator for the NOF/Inspire Osteoporosis Support Group.
In his spare time, which he claims he has less and less of, he enjoys fishing, carpentry/woodworking, baseball, and being a grandfather. He is still happily married to his wife Carol for 32 years (and counting).
Q: When a patient takes over-the-counter strontium citrate for the treatment of osteoporosis, can you get an accurate reading of bone mineral density?
A: In the periodic table of the elements, strontium appears in the same column and just below calcium, and just above barium. That means they react similarly in chemical reactions, but the lower you are on the table, the weaker the reaction. Barium is used in x-ray exams because it is almost completely opaque to x-rays because it is so dense. Strontium is halfway between barium in calcium and density in its appearance to x-rays including DXA scanners. Because it is chemically similar to calcium, people who take strontium in their diet in amounts that are hundreds if not thousand times higher in concentration than what occurs naturally, that strontium will replace some small percentage of the calcium atoms in the bone matrix, and make it appear more dense to the DXA scanner than it really is. Remember, these scanners were calibrated against cadaver bones of subjects who were not exposed to high doses of trace metals like strontium. So yes, it does make it impossible to get an accurate reading of true density unless you know exactly how much strontium has replaced calcium in a particular bone. To do that in a clinical setting is well beyond our current technology. One would have to scan the patient, kill them, then harvest the bone in question, cremate it, and then do chemical analyses on the ashes in order to determine how much of the strontium was in that bone and how much correction would be applied to the DXA results for accuracy. It makes it difficult to do follow-up studies using that technique when no one shows up for the second scan or volunteers for the first. (Humor alert!!)
Q: Can you tell if a medication treatment is working by comparing DXA scores over time? Do different types of bone react differently to osteoporosis medications?
Yes you can if a medication that increases bone density is working by serial scanning. (With the exception of those on strontium treatments). However, this gets back to our earlier conversation about the error in the measurement. The changes in bone density over time has to be larger than the LSC of the test. Even the most effective bone density treatments we have will only increase bone density at the spine between 3 and 8% in one year. Changes at the total hip are only 3 to 5% in a year, and changes at the femoral neck are only 1 to 2% at most an entire year. The very best centers have precision and LSC values that are small enough to detect changes at the spine and total hip in one year. Most centers will need at least two full years of treatment before the bone changes are larger than their LSC. And at poor quality centers the changes are never going to be large enough to be measured with any degree of confidence.
_One other caveat: you should never compare T-scores over time, but only the absolute density measurement itself in grams per square centimeter. It should also be whenever possible done on the same exact machine and if possible even the same technologist in order to measure the increase in BMD with any degree of certainty. The T-score’s can change if the software on the device has been updated with a new set of normal curves. The underlying bone mineral density value will not change when such upgrades are performed. _
Q: If you have osteoporosis of the hip and spine, does this also mean you may have it at other locations within the skeleton?
A_: Osteoporosis is defined as a systemic skeletal disease. That means in general terms that all the bones will lose density at the same time. Some bones may lose a little faster and some bones may lose a little slower depending on their metabolic activity. Some bones may be higher than the average person at one part of the skeleton, and lower than the average person in another part of the skeleton before bone loss starts. That is why it is not uncommon for us to have different T-score’s at different bones of the body. We all look different on the outside because we all have slightly different shaped bones on the inside. This is no different than noticing that not everyone who wears the same size shoe is the same height. _
Now there are some exceptions to this general rule. There is something known as disuse osteoporosis. If you are paralyzed in one limb and cannot perform weight bearing exercise on that side, you will develop osteoporosis at that site while the remainder of the skeleton is spared. We can see this quite dramatically in spinal cord injury patients where the arms are above normal, and the hips are severely osteoporotic. That is why the diagnosis should be made at the site with the lowest T-score provided it is a normal bone, and not subject to abnormal stresses such as patients with strokes or spinal cord injury.
Q: Do you feel that osteoporosis is over-diagnosed and if so, why?
Actually I believe just the opposite. Osteoporosis is under diagnosed in the population. The Center for Medicare services, CMS, estimates that less than one third of Medicare recipients have had a bone density test, even though it is part of the welcome to Medicare exam.
However, I do think that in those patients who have had a bone density test, and a T-score below -2.5, many of them are over-treated. This is where the FRAX score I think will become helpful. T-score is a measure of relative risk. For every T-score unit you are lower than someone else, all else being equal, your risk of fracture is approximately twice that of the other person. So if you have a T-score of -2 and they have a T-score of -1, your risk is double theirs. Most of the osteoporosis drugs that we have available today will reduce your risk of fracture by 30 to 50% in three years of treatment. But, if your risk is only 1% for fracture in the next 10 years, then taking the drug reduces it from 1% to 0.5%. The cost of the medication and the risk of side effects are the same regardless of your risk of fracture, and taking an expensive medication for three years to reduce your risk by half of a percent is not cost-effective nor do the benefits outweigh the risk of possible side effects. If you are older, and have a lot of other risk factors as determined by FRAX, that same T-score of -2 might equate to a risk of fracture of 50% in the next 10 years, taking the same drug at the same cost for the same period of time from 50% down to perhaps 25% and the same drug, same cost and same risk profile provides 25% of reduction, rather and 0.5% of absolute fracture risk reduction. That would be cost-effective, and the risk of side effects is far outweighed by the benefits of fewer future fractures.
Q: Are errors on DXA scans common? What would you do to rectify this, if you saw a scan with errors?
A: I give a full day course on how to correct common errors in DXA scans. But to give you a summary in a short answer, there are two types of errors. One type of error is an error in positioning the patient. If the hip is not adequately abducted or internally rotated, or if the spine is not centered properly in the scan field, there is not much you can do to correct it other than repeat the test. But if that test was done two years ago, you would first need to own a DeLorean with a 1.21 GW flux capacitor in the back seat, and some spare plutonium.
The other types of errors are errors in the analysis. I have seen spine scans were T12 was labeled as L1, or the intervertebral space markers were drawn in the middle of vertebral bodies rather than in the disk spaces between them. I have seen the femoral neck box placed incorrectly, I have seen the bone mapping (the outline surrounding the bone that is drawn by the software as a limit to what it considers bone and what is considered soft tissue) identifying kidneys as part of the spine. I have seen artifacts such as underwire bras and bra hooks called spine density, and buttons and rivets and sequins on the back pockets of jeans on hip scans, and more recently, navel piercings in spine scans. I have even seen hip scans reported were both hips have been replaced with metal and T-scores of +22 to +26 and read as normal. Some of these errors can be corrected by reanalyzing the scan correctly but many cannot.
It is much easier to prevent errors than it is to correct them. Bone density centers who use ISCD certified technologists, and physicians with specific training to interpret them, and especially centers that use dedicated technologists whose principle or only responsible is operating their scanners, tend to produce the best results with the least number of errors.
Q: If a patient has osteopenia and then sustains a fracture, does this change their diagnosis?
A_: Yes. If they sustained a fracture from a fall from standing height or minimal impact trauma that was less than what would be expected to cause that bone to break, then they have a clinical definition of osteoporosis. This is how we diagnosed osteoporosis before they invented DXA scanners and no one has abandoned it. You can have normal bone density and fragility fractures and be given a diagnosis of osteoporosis as well. Osteoporosis is not a T-score. A T- score is a risk factor for osteoporosis and we shouldn’t be treating T- scores but rather treating the patient and the goal is to prevent fractures not to improve T- scores. If you are in the emergency room with a heart attack, you have coronary artery disease, even if your cholesterol level is only "borderline" or even if it’s normal. Fractures override T-scores in the same manner._
Q: Does a DXA scan take into consideration the patients’ height and weight, and does this have an impact on the resulting T-score or diagnosis?
A: This is not a simple yes or no question. There are some brands of DXA scanners that use height and weight in order to determine what scan speed to use. Heavier patients would require a slower scanning speed and very thin patients would require a lower x-ray intensity setting, so that the correct amount of radiation will strike the radiation detector design limits. Too much or too little radiation would affect the accuracy and precision of the scan. Other brand scanners are relatively immune to the thickness of the patient being scanned. But there are two other considerations regarding the patient’s body habitus that are perhaps more important than just what machine settings are used in a given height and weight patient. Just one other note on radiation. Even at their highest settings, a DXA scan delivers less radiation than a single day’s worth of natural background radiation, or about one hour of being in an airplane at altitude. It should not be considered as a deterrence to getting a scan.
The first is a controversy based on the fact that DXA does not measure true density which would have units of mass per unit of volume (g/cm3), but rather, it measures something called "area density" which is mass per unit of area (g/cm2). Because of this a person with smaller bones will read lower than someone with larger bones of the same true density. Some would say that this means that smaller boned people will be over-diagnosed as having osteoporosis while heavier patients will be under-diagnosed. But the purpose of bone density testing is to determine what your risk of fracture is. Having once been a carpenter, I can tell you that a 2 x 2 spindle on a deck railing is not as strong as a 4x4 post on that same deck railing, even though both are made out of cedar and both have the exact same density. If we could scan the two pieces of wood on a DXA scanner, the post would read higher than the spindle and better reflect its true strength. The same is true with bones. Larger bones will be stronger than smaller bones even if they are composed of the same density material. In this respect DXA will give you a better assessment of the strength of the bone, precisely because it does not adjust for the size of the bone. FRAX also incorporates height and weight to estimate bone size and adjusts fracture risk because of this concept.
The second has to do with fat, which is an artifact in DXA scans. The fatter the soft tissues surrounding a bone the higher the bone density reading will be, especially in DXA of the hip. This error just happens to improve the fracture risk ability of DXA at the hip in women but interestingly, not in men. Heavier women tend to put their fat around their hips thus adding a layer of protection in the case of fall onto their sides. In men, the extra fat is typically deposited in the abdomen and very little padding is found over the hips. It’s a case where muffin-tops protect you from hip fracture, but beer-bellies do not!
As a DXA technologist I am much more concerned with patients who had a large change in weight between scans. This means that changes in bone density readings may be influenced by the amount of weight and fat change, rather than the amount of bone that has been gained or lost between the two scans. That is why we always measure patient’s weight just before we scan them, rather than taking the patient’s word for it. If the patient returns for a follow-up scan and has lost 20% of their body weight, but we see no change in the measurement of density itself, is quite likely that the patient has actually improved their bone density. It’s just that we cannot see it because of the influence of weight loss. The same would be true of someone who gains a large amount of weight between scans, but their bone density has not changed at all. It is quite likely that they are losing bone, but it is being masked by the increase in weight.
Thank you Mr Jankowski!