The April 28 issue of The Journal of the American Medical Association featured an article, "Antepartum Dental Radiography and Infant Low Birth Weight," by Dr. Philippe P. Hujoel and colleagues,1 that is a very well-written and designed population-based case control study.
In the study, 1,117 women with low-birth-weight infants (less than 2,500 grams) were compared with 4,468 randomly selected control pregnancies with normal-birth-weight infants to determine if antepartum dental radiography was associated with low-birth-weight offspring. The study attempted to take into consideration 16 different, potentially confounding, elements. However, they did not consider thyroid gland function or potential exposure to nondental radiographic procedures.
Dental radiographic exposures were divided into three groups: 0 milligray, or mGy—no known radiographic exposure; 0.1 with 0.4 mGy; and higher than 0.4 mGy. Fetuses were not exposed, only the mothers. Odds ratios of low birth weight and term low birth weight were calculated. Women exposed to less than 0.4 mGy compared with 0 exposure were associated with an odds ratio, or OR, of 2.27 (P = .03) and exposures higher than 0.4 mGy exhibited an OR of 3.61 (P = .005).
The article concludes that dental radiography during pregnancy is associated with low birth weight, specifically term low birth weight. The study specifically identifies low-level exposure to the thyroid gland as the potential etiology for low birth weight, since the hypothalamus-pituitary-thyroid "axis" is significantly involved during pregnancy. It is suggested that the thyroid gland may be especially sensitive to radiation during pregnancy.
Statistical associations identified higher exposures in women who smoked, who had inadequate prenatal care or who were of different ethnic origins. The higher doses were in the range of 0.01 to 0.05 mGy. There was no significant association between diagnostic radiation doses higher than 0.4 mGy with very low-birth-weight infants and preterm low-birth-weight infants.
However, the authors found a strong association between term low birth weight and radiation doses higher than 0.4 mGy. The authors estimate that eliminating dental radiographs during the first trimester could reduce term low-birth-weight infants by up to 5 percent if the association with thyroid gland exposure is "causal."
Following are our comments on the article’s relevance to dentistry. Since the first dental radiograph in 1896, dentistry has made huge strides in reducing radiation exposure to the patient by several orders of magnitude. Nonetheless, when an article such as this JAMA article appears, the popular and lay press seize upon it, causing extreme angst for patients, and even leading them to avoid necessary dental care. Therefore, it is prudent to consider the issues raised in this article, since mainstream news outlets have given the article a great deal of publicity.
For example, the May 10, 2004, issue of Time magazine has a brief article with the headline, "Do You Dare See a Dentist?," 2 along with pictures of a protruding, very pregnant, female abdomen and a bitewing radiograph.
One cannot criticize the competence of the authors and the details of the statistical analysis used to determine a possible correlation between low birth weight and dental radiography. The problem lies in determining a mechanism to explain the correlation.
There are several authors3–5 who have reported thyroid doses from full-mouth radiographic surveys ranging from 0.4 to 0.63 mGy, compared with the 1.6 mGy cited by the authors. However, the rest of the absorbed doses for the other radiographic surveys used by these authors are within the "ballpark" of other studies not referenced in the original article, but available.6–14
The JAMA article attributes a high level of mathematical significance to a variety of variables, such as the higher doses associated with cigarette smoking (0.01 mGy, P = .14), inadequate prenatal care (0.05 mGy, P = .13) and ethnicity. Ethnicity was associated with a very high statistical probability of P = .01 for Asian (0.05 mGy), black (0.04 mGy) and white (0.03 mGy).
What was never brought forth in the article is that these dose levels represent a level of radiation intensity equivalent to only three to five days of normal "background" radiation. Currently, the average background radiation (including radon) in the United States is approximately equal to 3 mGy per year, or 0.01 mGy per day. The authors are thus indicating that their method is sufficiently sensitive to distinguish between zero/no radiation dose and the equivalent of three to five days of "background" radiation. The obvious next question: Is this a reasonable conclusion?
With the effects of ionizing radiation massively researched for over 100 years with billions of dollars in research funding, no other authors have been able to demonstrate biological effects at such a low level of exposure. The Health Physics Society’s position papers on "Radiation Risk in Perspective"15 and "Risk Assessment"16 indicate that effects of ionizing radiation have not been reliably observed at less than 100 mGy (10 rem).
As a point of reference, various localities around the world with large resident populations have background radiation levels averaging 15 to 40 mGy per year,17 or 0.04 to 0.11 mGy per day, with no known harmful effects, excluding Ramsar, Iran,18 where an individual was found to have an annual dose of 132 mGy per year, or 0.36 mGy per day!
The authors hypothesize that dental exposures to the thyroid gland ranging from 0.12 to 1.6 mGy, depending on the type of radiograph, may be associated with low birth weight. They hypothesize that absorbed doses equivalent to 12 to 40 days (cutoff point for increased radiation effect) of background equivalent radiation are of sufficient intensity to produce some vague, unspecified alteration in the hypothalamus-pituitary-thyroid axis, leading to low birth weight.
It is worth noting that the number of ionizing particles (fluence) deposited in each cell nucleus can be estimated,19 and is relevant because 1 mGy of 70 kilovolt peak dental X-rays will produce a fluence of about one ionizing particle per nucleus. Thus, the implication of the JAMA article is that thyroid glands whose cell nuclei receive from 0.12 to as many as 1.6 ionizing events can potentially initiate physiological and biochemical changes that ultimately produce low-birth-weight infants.
Since thyroid gland exposures associated with common dental diagnostic radiographs are in the range of a few days normal background radiation, this would suggest that their associated findings could be a chance finding, especially since the study did not include thyroid function measures and information on nondental radiation exposures, two variables that would appear to be critically important "control points," before postulating potential "causality" for dental diagnostic radiation exposures.
The authors place a great deal of importance on their observations that the OR with doses at the higher-than-0.4 mGy-level was higher than those between 0.1 to 0.4 mGy. From a biophysical standpoint, the difference in radiation exposure at the nuclear level would be one to three cell nuclei out of 10 receiving an energy-depositing event for a low effect (OR = 1.2) versus four cell nuclei out of 10 receiving an energy-depositing event to produce a "high" effect (OR = 2.27).
Based on the biophysical evidence, one might suspect a chance statistical association is in action rather than a special sensitivity of the thyroid gland at the onset of pregnancy.
Before jumping on the bandwagon of causality for dental diagnostic radiation exposures, it would be prudent to take into consideration thyroid function and nondental radiation exposure, and possible exposure of the antepartum women to low-level volatile toxins such as might be associated with the frequency of filling a vehicle with gasoline, or using common household cleaning agents and solvents of various kinds, along with potential substance abuse other than cigarettes and alcohol.
Finally, the authors suggest that reducing dental radiation exposure could potentially reduce term low-birth-weight offspring, or TLBWs, by as much as 5 percent. We learn from Table 2 in the article that there were 336 TLBWs, with 287 (85.4 percent) having no identifiable antepartum exposure. Thus, reducing radiation exposure would potentially affect 5 percent of the 49 dentally exposed women, or 3 infants. This would reduce the 1,117 identified low-birth-weight infants in the study to 1,114, a difference of only 0.3 percent. Thus, one might question the clinical significance of the findings. Is a change of three-tenths of one percent a meritorious clinical goal?
The potential low clinical impact of the findings, along with the P = .01 association of statistical significance for ethnicity at levels of radiation equivalent to only three to five days of background radiation, suggests a strong possibility that the authors have described a chance statistical association to which it is extremely difficult to ascribe a mechanism of action by which such a small number of radiation-depositing events at a cellular/nuclear level could produce an effect of the nature described in the article.
In the meantime, the article would appear to significantly raise angst within the medical profession and the general public toward dental radiography, when there are still many questions to be answered before any causal relationship is made between diagnostic dental exposures and low-birth-weight infants.
Dentists should continue to defer elective radiographs on pregnant patients. However, they should not let fear and uncertainty prevent them from ordering appropriate diagnostic radiographs when the patient’s dental care needs dictate for proper dental care.
TOP
REFERENCES
