Take lymphedema-prevention instructions with a grain of salt

If you have had breast cancer surgery that involved the removal of lymph nodes (i.e., axillary [armpit] lymph node dissection), you have almost certainly been given an advice sheet something like the one posted at breastcancer.org (excerpts below):

• Do moisturize your skin frequently and regularly. Use lotions such as Moisturel, Eucerin, Vaseline Intensive Care, or your own favorite brand to make your skin supple and prevent it from cracking.
• Do keep your hand and arm extra-clean, but don’t use harsh soaps such as Ivory (despite Ivory’s advertised image as a gentle soap) or Dial. Use Dove instead.
• Do use rubber gloves when you wash dishes or hand-wash clothes.
• Do wear protective gloves when you garden or do outside chores.

• Don’t go into high-heat hot tubs, saunas, or steam baths.
• Don’t apply heating pads or hot compresses to the arm, neck, shoulder, or back on the affected side. Also, be cautious of other heat-producing treatments provided by physical, occupational, or massage therapists, such as ultrasound, whirlpool, fluidotherapy, or deep tissue massage. Heat and vigorous massage encourage the body to send extra fluid into the compromised area.
• Don’t carry heavy objects with your at-risk arm, especially with the arm hanging downward.
• Don’t wear heavy shoulder bags on the affected side.

What most patients do not realize is that there is a huge disconnect between what is actually known in the scientific literature about risk factors for lymphedema, and the preventative measures that doctors (and web sites) routinely advise. Remarkably, this disconnect has persisted for at least the last decade (e.g., Petrek and Heelan, 1998). Although a lack of controlled, prospective studies still limits our knowledge greatly, most researchers have found that the primary predictors of lymphedema in breast cancer patients are: advanced disease at diagnosis, older age, higher body mass index (BMI), and longer follow-up.

The first of these presents a confound; more advanced cancer means that more lymph nodes are affected, which means more surgical involvement in the axilla. It is unknown whether the higher incidence of lymphedema in these cases is from the cancer itself or from the more intrusive surgery. Older age and higher BMI both are correlated with lower overall fitness level, which one might suspect is the true culprit, but fitness level has not been considered in any of the recent scientific papers or reviews. The last “predictor” simply means that the longer women are followed post-operatively, the more likely it is that more cases of lymphedema will be found. This is a statistical effect; women who have lymphedema after the end of a study with shorter follow-up will not be counted as having it.

Goshen and Smith (2006) say straight out what your surgeon will almost certainly never tell you, unless pressed:

Clinicians routinely suggest a number of other restrictions to patients after axillary dissection or radiation therapy to reduce the risk of lymphedema, although none of these techniques has been rigorously evaluated. They include avoiding injections, blood draws, and blood pressure monitoring, as well as constrictive clothing or jewelry on the affected arm. Because some episodes of lymphedema seem to be [emphasis added] triggered by infection of the arm or by repetitive or heavy arm exercise, patients are instructed to avoid skin breaks, observe meticulous nail care, and avoid heavy or repetitive arm activities. Some patients are cautioned to wear a compression garment during airplane travel. The efficacy of all of these precautions is unproven [emphasis added].

Following some of the typical precautions (e.g. avoiding injections, blood draws and blood pressure monitoring) makes sense even in the absence of actual data, because there is little inconvenience in doing so. But the recommendations regarding repetitive and heavy arm exercise are not only are much more inconvenient for a lot of us to follow, they were recently demonstrated to be completely wrong in a randomized study showing that women with lymphedema who lifted weights regularly actually had their condition improve (Schmitz et al., 2009):

There is evidence that exercise enhances the flow of lymph, and improves protein resorption and that the increased pulmonary work associated with exercise assists with lymph flow. It is also possible that increased muscle strength reduces the relative effect of common daily stresses to the limb.

Seemingly at odds with these data is another recent study that claims to have “confirmed that occupation is a risk factor” for breast cancer related lymphedema (Gur et al., 2009). These authors categorized occupations by presumed amount of activity, and found that women in occupations with more activity were more likely to have lymphedema. But there are several reasons why their conclusions are completely invalid.

Gul’s results are circular because their paper suggests that they categorized the patients in the study based on information about those specific patients, rather than a priori. The authors knew at the time the study began who had lymphedema, so the result is merely a self-fulfilled prophecy based on their assumptions of what causes lymphedema. It is astonishing that a paper using such a method could have ever been published.

Even if Gur’s categories were valid, the data are very weak relative to Schmitz’s because they are retrospective, rather than based on randomized trials. No potential covariates (either among the variables they chose, or those they did not, such as fitness level or daily exercise level) were accounted for in their analyses. Schmitz, on the other hand, took a group of women with known characteristics (eliminating as many covariates as possible by their selection of study subjects), randomized them to either weight train or not over several months, and then analyzed the results of that treatment. This is a much more powerful and accurate method for gathering data.

Secondarily, Gur’s categorizations were very general, and did not measure any particular level of activity directly. The way they analyzed their data actually just showed a correlation of occupation category with lymphedema, it did not at all establish causation, contrary to their assertion. So, their result and interpretation of it should not be taken as prescriptive advice, where as Schmitz’s could reasonably be.

Another commonly held belief is that infection to the affected arm is to be avoided at all costs; hence, the recommendations of electric razors, long sleeves and gloves while gardening, etc. In this case, too, the data simply do not exist yet to support this (Kocak and Overgaard, 2000).

A correlation between recurrent or late infections and arm edema has been reported in several studies , but it is difficult to accept this as a causative factor for lymphedema. The lymphedematous tissues are extremely sensitive to infections, and…the recurrent infections may be secondary to the edema, rather than the cause of the condition.

So few of the recommendations have been actually tested that there is no way to know which, if any, are valid. They are based on intuition, even though the intuition has turned out to be completely wrong in the case that was tested, heavy lifting and repetitive motion. We can say both from the weightlifting study and the risk factors that are fairly certain (age and BMI) that maintaining a high level of fitness is certainly a good way to reduce risk for lymphedema; after all, we know it is a good idea for better health in general. (Always keep in mind that you will have more success focusing on fitness level than on BMI, which was never meant to provide useful information for individuals.)

References

Golshan, M. and Smith, B. 2006. Prevention and management of arm lymphedema in the patient with breast cancer. Journal of Supportive Oncology 4:381–386

Gur, A.S., Unal, B., Ahrendt, G., Gimbel, M.L., Kayiran, O., Johnson, R., Bonaventura, M. and Soran, A. 2009. Risk factors for breast cancer-related upper extremity lymphedema: is immediate autologous breast reconstruction one of them? Central European Journal of Medicine 4(1): 65-70.

Kocak Z., Overgaard J., 2000. Risk factors of arm lymphedema in breast cancer patients. Acta Oncologica 39(3):389–392.

Petrek J.A., and Heelan M.C. 1998. Incidence of breast carcinoma-related lymphedema. Cancer 83(12 Suppl American):2776-81

Schmitz K.H., Ahmed R.L., Troxel A., Cheville A., Smith R., Lewis-Grant L., Bryan C.J., Williams-Smith C.T. and Greene Q.P. , 2009. Weight lifting in women with breast-cancer-related lymphedema. New England Journal of Medicine 361:664-73.

Tags: cancer, health, statistics

This entry was posted on Friday, October 9th, 2009 at 3:47 pm and is filed under cancer, health, statistics. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.