The answer to the title of this Part 1 is in two parts. First, we know of no intrinsic relationship at all between scoliosis and breast-cancer. The second part is that for decades the management of scoliosis was done with frequent x-ray checks of the spine, in the worst possible way for breast irradiation. This led some of the women to develop breast-cancer from the radiation, not from the scoliosis. We shall be discussing some of the opportunities missed (for decades) to prevent such breast-cancer, and some of the elegant work done later to prevent many of the breast-cancers arising from the management of scoliosis.
Scoliosis --- What Is It?
Scoliosis is defined (from the Greek scolios, for curved) as a lateral curvature of the spine. There are several possible bases for lateral curvature of the spine, but there is one common entity which concerns us here. That is the entity known as Idiopathic Scoliosis or Habitual Scoliosis. The appellation, "idiopathic," simply means that the cause is unknown. Professor Arthur Steindler, Head of Orthopedic Surgery at Iowa State University, wrote the chapter on " Static Deformities of the Spine" in the 1942 Edition of Christopher's Surgery.
He wrote at p.463-464:
"The So-Called Habitual Scoliosis or Idiopathic Scoliosis: This term merely indicates the absence of a definite underlying cause. It implies a general or constitutional predisposition, such as anatomical build and general habitus, or inherited or acquired weakness or relaxation of the spine and its musculature ... The majority of cases occur in females." And (at p.466):
"So-called habitual or idiopathic scoliosis has always received considerable attention from the viewpoint of prophylaxis. It has been called `school scoliosis' because it is believed to develop during the school age; some investigators have found as high as 20 to 30 % of all school children afflicted with this `functional' scoliosis as a result of faulty posture. However, the usual prophylactic measures are, as a rule, belated. The critical age is betwen the third and fifth year, and it is at this time that prophylaxis should be practiced." (Recent investigators cite a later age-band for this disorder.)
The Early History of Treatment
From Dr. Steindler's discussion, it is clear he did not feel muscle training, external or internal splinting were too helpful. External splinting meant braces and internal splinting meant operative fusion of the spine. He certainly was not optimistic about the management of scoliosis. We quote his final conclusions (p. 469):
"This short outline of the treatment serves to emphasize the all-importance of the earliest possible recognition and timely prophylaxis of scoliosis. Until these vital facts have become generally appreciated, there is not likely to be any real progress in the management of cases of scoliosis."
Unfortunately --- during the months and years of the efforts to correct the scoliosis by braces, by casts, by musculature training --- the physicians continued to monitor the patients both with physical examination and with x-ray examinations of the spine. Most unfortunately, the method of taking the spine films in the early period was to have the beam enter the front of the body, and exit through the back (the anteroposterior, or AP, approach). From the point of view of breast irradiation, this was the worst possible approach for part of the x-ray examination. The lateral views which were taken were unfavorable but not as unfavorable as the antero-posterior view.
Apparently very little concern was shown for the steady accumulation of breast-dose from the repeated spinal exams, taken in this extremely unfavorable AP view.
So now the long-drawn out management procedures for scoliosis meant a seriously growing radiation dose to the breasts of adolescent and post-adolescent girls and young women. There was no comment whatever in the Steindler chapter about any hazard of the radiographic aspect of management. We did not expect to find any. The times were such that broadly, the medical profession did not believe that diagnostic x-rays spread over 3 to 6 years could be harmful. Let us be clear; this was a belief, not a demonstration.
A decade later, approximately, the 1950 Edition of Mitchell-Nelson Textbook of Pediatrics, 5th Edition, has the Orthopedic Pediatrics Chapter written by Dr. Charles C. Chapple. In his brief section on "Scoliosis," the part on Treatment in no way mentions the repeated x-rays associated with management of the disorder:
"Treatment. Nutritional improvement and postural exercises are usually sufficient in the treatment of mild cases, but in the more severe ones, since the scoliosis is liable to become more marked with increasing muscular development, correction by such orthopedic means as exercise and supporting devices is advisable in addition to the treatment of any underlying factor. Many require spinal fusion."
No mention is made of the numerous x-rays which accompanied the use of such corrective devices as braces or casts.
Two decades later, in 1968, Ambulatory Pediatrics edited by Dr. Morris Green and Dr. Robert Haggerty, commented succinctly on scoliosis as follows:
"Although complex, carefully fitted bracing may occasionally prevent progression of the deformity, there is increasing evidence that early spinal fusion is by far the most effective method of treatment. Complete correction of the curve is rarely possible, and the purpose of fusion is to stabilize the spine and prevent progression."
There is no mention whatever about the numerous spine films which were taken in the course of such management of scoliosis. Quite obviously the textbooks over three decades, at least, did not consider it worth mentioning that the many spine films were giving a very high dose to the breasts, and that such dosage was really unnecessary.
Finally, in 1979, there was a massive change in the tide.Part 2. A Spectacular Achievement in Dose Reduction for Scoliosis Patients
Nash, Gregg, Brown, and Pillai (1979), all at Case Western Reserve University, called attention to the serious problem of excessive radiation dosage accumulated by female scoliosis patients during the course of one of the disorder's common treatments, known as the Milwaukee Brace treatment. As Nash and co-workers pointed out, adolescent girls would be monitored with some 20 x-ray films, plus retakes, or a total of 22 radiographs during the average three-year treatment period. Most films were taken as AP shots. Having made their own measurements, Nash and co-workers estimated an average entrance dose of 1.13 roentgens for the AP shots used, and 1.78 roentgens for the lateral shots. From their estimates of dose to organs, they concluded (p.371) that "given an average of twenty-two roentgenograms over a three-year Milwaukee Brace-treatment program," the risk of breast-cancer due to x-ray radiation rose "from 140 to 290 per million (110 percent)."
Even worse than the conclusion of Nash and co-workers that the monitoring of scoliosis therapy was more than doubling the patient's risk of later breast-cancer, was their additional conclusion that "many scoliosis patients may require more roentgenograms than the average scoliosis patient postulated in this study."
Since these types of therapy had been going on before Steindler's writings in 1942, there is good reason to consider that the entire 1920-1960 period was characterized by this sort of therapy for scoliosis.
Details of the Steps Which Reduced Breast-Dose
Nash and colleagues decided to do something about this situation. They proved in a series of scoliosis patients that shifting beam direction from anterior-posterior to posterior-anterior could reduce the breast-cancer risk by a factor of 28 ! As for the image quality, the Nash team reported that the x-ray films obtained from the PA beam direction proved technically satisfactory for routine scoliosis care.
Additional work of real merit on reducing the risk from x-rays for scoliosis patients has been done by Gray, Hoffman, and Peterson at the Mayo Clinic. In an extensive study (1983), they report on how they used the following combination of measures to reduce dose:
- A posterior-anterior beam direction;
- Specially designed leaded acrylic filters;
- A high-speed screen-film system;
- A breast-shield;
- Additional filtration in the x-ray tube collimator.
Dr. Gray and co-workers reported at page 5 that the following reductions in breast dose occurred:
"...For the breasts there was a sixty-nine fold reduction from 344 to less than five milliroentgens for the postero-anterior radiograph and fifty-five fold reduction from 277 to less than five milliroentgens for the lateral radiograph."Part 3. Determining (1920-1960) Breast-Doses in Scoliosis Management
Hoffman, Lonstein, Morin and co-workers (1989) have conducted a followup study of a series of women who had multiple diagnostic x-rays in the course of scoliosis management. We shall make use of some of their data to provide entries for the Master Table, Column H.
They regard scoliosis as a relatively common condition. The prevalence in the general population has been estimated to be as high as 8 %. However, a major determinant of the prevalence reported is the degree of spine curvature used to diagnose scoliosis. The disease is more prevalent in women than in men. Hoffman and co-workers provide the following [not totally consistent with their estimate above of prevalence of the disorder]:
"Adolescent idiopathic scoliosis afflicts nearly 2 % of the U.S. population, and school jurisdictions in over 50% of the United States have screening programs for children 9-14 years of age. Girls require treatment more often than boys. Once detected, scoliosis requires periodic x-ray monitoring of the spine, especially during adolescence, when the bones are growing at an accelerated rate. The times of breast development and menarche are clinically important because they indicate that the growth spurt has begun or will begin soon. Patients commonly receive anteroposterior and lateral spinal x-rays to monitor the progression of spinal curvature and the effect of treatment."
These school programs certainly are addressing an older group than that suggested worth studying in the Steindler paper, which suggested the critical age to be between the third and fifth year.
We shall use the breast-dose estimates provided by Hoffman et al. in this study (1989).
o - Item 1: This study was done in Hennepin County, in the State of Minnesota. The study population consisted of 1,030 women with a confirmed diagnosis of scoliosis or kyphosis who were seen in one of three hospitals or in one clinic in Minneapolis-St.Paul, Minnesota area. We were not given any assurance that this study included all the cases of scoliosis being treated in Hennepin County during the period under investigation. It is possible that such inclusion existed, but since we do not know, we caution the reader that breast-exposure estimated nationally from these data may well be an underestimate because of the absence of an unknown fraction of the total number of cases.
o - Item 2: The 1,030 women who qualified for inclusion were followed up for 26 years, on average.
o - Item 3: The estimate of dose to the breasts was a mean value of 12.8 total rads (medical rads), acquired over a period of 8.7 years, on average. The examinations were given in the period, 1935 to 1965.
o - Item 4: Since the acquisition of cases occurred over a period of 30 years (1935 -1965), the number of acquistions per year in Hennepin County must have been, on average, 1,030 / 30, or 34.3 cases per year.
o - Item 5: Since the average age at diagnosis was 12.3 years, we should assign some of the cases to age groups below 12.3 years and some to age groups above 12.3 years. We shall arbitrarily estimate the distribution of the annual 34.3 cases as in the following tabulation.
Age National Females Cases Mean Person- Avg. Dose Group Women in in in Dose Rads Rads, for Age Grp. Hennepin Age Grp. Med. Rads Hennepin 9 889589 4181 4.9 12.8 62.72 0.01500 10 889390 4180 4.9 12.8 62.72 0.01500 11 889209 4179 4.9 12.8 62.72 0.01501 12 889028 4178 4.9 12.8 62.72 0.01501 13 888829 4177 4.9 12.8 62.72 0.01501 14 888585 4176 4.9 12.8 62.72 0.01502 15 888277 4175 4.9 12.8 62.72 0.01502
o - Item 6: We need to know the number of female persons in each age group during the 1920-1960 period.
The population of Hennepin County in 1960 is 842,854 persons (County and City Data Book 1962). The population of the USA in 1960 is 179,333,000 persons. To calculate the number of female persons in each age group, we multiply (842,854/179,333,000) by the number of women nationally in each age group (from the Master Table Col.A) in the average year of 1920 - 1960. This was done and the entries are given under "Females in Hennepin." in the preceding tabulation.
o - Item 7: We know only that there were, on average, 34.3 cases acquired per year, but we do not know the age distribution of the cases. We approximate this by assigning cases to 7 age-years. So (34.3 / 7), or 4.9 cases are assigned per age-year.
o - Item 8: We need to estimate the person-rads of breast-pair exposure for each year. This is the product of (persons irradiated) x (average dose). Since all the cases are taken to be 4.9 per year, and the mean dose of 12.8 rads is used for all cases, the person-rads for all entries is (4.9 persons) x (12.8 rads), or 62.72 person-rads.
o - Item 9: We have the dose to individuals (12.8 rads), and we have the person-rads represented by the cases. we need, and have not yet calculated, the average population dose for these groups of women. That number is the person-rads injected into the population / the persons in the population. So, we have average dose = (62.72 person-rads) / (number of females in Hennepin, for each age group). This, for the first entry into our tabulation, is (62.72 person-rads / 4,181 persons), or 0.01500 rads.
o - Item 10: We must consider the issue of duration before making entries into the Master Table. We have data for 1935 to 1965. We can regard those data as adequately close for 1930 to 1960. The question before us is what we do about the 1920-1930 period for which we have no data.
We suspect, but can not prove, that the dose was higher in that earlier period than in 1930-1960. We, therefore, take the conservative approach of eliminating any dose for 1920 to 1930. The final dose adjusted very conservatively for duration is obtained as follows:
((30 years x 0.01500 rads) + (10 years x 0 rads)) / 40 years, or 0.01125 rads per age-year. This value of 0.0112 is transferred to the Master Table, Column H, for age-years 9 through 15.
We believe the true dose was higher, but using our general rule of conservatism when we can not prove it to be higher, we choose the lowest dose possible, zero dose, for the 1920-1930 time-period in question.
# # # # #
What motivated Dr. Joel Gray and his Mayo Clinic colleagues to achieve spectacular reductions in breast-dose (see Part 2) for scoliosis patients? They seem to have answered this question, when they wrote (Gray 1983, p.12): " ... no level of exposure to ionizing radiation is entirely safe ..."
o - Big Consequences from Low-Dose Carcinogens:
A Warning from the National Cancer Institute (NCI) in 1990
This book (Preventing Breast Cancer) states that low radiation doses do indeed count with respect to breast-cancer totals in the population.
Women will hear from some quarters that when the risk is very low, one can simply dismiss such a risk. Thus, for a given low-level radiation exposure, someone might calculate that radiation will cause a breast-cancer in only one woman out of every 10,000 women receiving that dose. For risks of this size, "Just forget it" is often the advice.
Good advice, or deadly?
The National Cancer Institute (NCI) provides the following warning at page 7 of its 1990 booklet:
"In the human population, large numbers of people are exposed to low doses of chemicals, but the total impact may not be small at all. For example, a carcinogen might cause one tumor in every 10,000 people exposed to it, which may not seem great. But exposure of 230 million Americans would result in 23,000 cancers --- a public health disaster."
Suppose ionizing radiation is the carcinogen for which the NCI is calculating the causation of one breast-cancer in every 10,000 women exposed to a specified "low dose." The female population in 1994 is approximately 133 million, so the number of breast-cancers induced would be 13,300 per year, if the "low dose" is an annual dose.
The nation would be giving only "lip-service" to the goal of preventing breast-cancer, if various ways to prevent 13,300 cases per year were each dismissed because a woman's personal risk "may not seem great" at 1 per 10,000. The NCI makes a good statement (p.11):
"Individually and together, we must make every effort to reduce or eliminate human exposure to carcinogens." And, of course, the NCI starts its booklet at page 1 with the statement:
"Cancer-causing agents also include X-rays, sunlight, and certain viruses." And at page 5, "radiation and radioactive materials" are listed again as proven human carcinogens. The NCI advises at page 12: "Don't ask for an X-ray if your doctor or dentist does not recommend it. If you need an X-ray, be sure X-ray shields are used if possible to protect other parts of your body."
Source: National Cancer Institute (USA), "Everything Doesn't Cause
Cancer," 12-page booklet, March 1990 (NIH Publication 90-2039).