DOXYCYCLINE AND STAINING OF PERMANENT TEETH

Rocky Mountain spotted fever (RMSF) was first identified in 1873 and is the most common rickettsial disease in the United States, with an incidence in 1990 of 0.26/100 000 people.1 Children ages 5 to 9 are most often affected.2 Although the tetracycline group of antibiotics is effective for treatment of RMSF, it is well-known that they cause staining of the teeth when ingested during development of enamel and dentin. Tooth-staining properties among the various tetracycline analogs have been examined, and investigators have concluded that they produce different color staining of teeth.3, 4 Chloramphenicol is sometimes used as an alternative to tetracyclines in the treatment of RMSF, but it has potentially serious side effects, and no oral dosage form is available in the US. Other antibiotics (e.g. penicillin, aminoglycosides, cephalosporins) are not effective treatments for RMSF. Fluoroquinolones have been effective in vitro, but their safety and efficacy have not yet been established in the pediatric population.2 The American Academy of Pediatrics recently revised the treatment options for young children with RMSF. Tetracyclines remain contraindicated for treatment of common infections in children younger than 8 years of age, but doxycycline is recommended as the drug of choice for treatment of presumed or proved RMSF in children of any age.5 However, doxycycline presents a clinical dilemma because of inadequate data concerning the staining of developing teeth. The purpose of this retrospective, recall study was to determine whether doxycycline, given in therapeutic doses for the treatment of RMSF, stains permanent teeth when administered during tooth development. Methods. Children who received doxycycline for the treatment of RMSF from May, 1981, through August, 1988, were identified through a computer search of the medical records at an 800-bed community-based, academic medical center. The parents of these children were contacted, sent a letter explaining the study and asked to agree to their child's participation. A small stipend and a free oral prophylaxis (cleaning) were offered as incentive. Interested parents were sent a detailed description of the study, a consent form, and an appointment time. A control group of subjects matched for fluoride exposure, chronologic age and similar stage of dental development were solicited from the patient pool of the hospital dental service. Each study subject was matched with two control subjects. Exclusion criteria for both the control and exposure groups included the following: other exposures to tetracyclines, presence of orthodontic braces, a history of trauma to the teeth under evaluation, restoration of the teeth under evaluation with any restorative material or any clinically apparent dental fluorosis (white or brown staining), as judged by the principal investigator. On arrival to the clinic parents were asked to review and sign an institutionally approved informed consent form. A health history was obtained with a focus on past tetracycline use and fluoride exposure. Elements of the inclusion/exclusion criteria were reviewed by the principal investigator. The participant's teeth were then cleaned with a slurry of pumice and water applied with a slow speed handpiece and rubber cup. The teeth that were developing during the ingestion of the doxycycline were identified according to a standard table of dental development. These teeth were photographed using a Nikon® 35-mm intraoral camera with both ring and point flash, and Kodak 100® speed print film. The photographs of the teeth were compiled and mounted on dark blue 6- by 8-inch placards and coded. The study teeth were identified with an arrow. Photographs of the complementary teeth in the matched controls were similarly mounted and identified. Five residents in dentistry, blinded as to study groups, were used as independent raters. They were first trained and tested for interexaminer reliability with 12 placards of nonstudy subjects, which included normal subjects and those with varying degrees of tetracycline staining. The raters used an ordinal scale from 0 to 3 as follows: 0 = normal range of color (no color changes); 1 = mild discoloration (noticeable color difference upon close scrutiny); 2 = moderate discoloration (obvious color change); 3 = severe discoloration (extreme and requiring cosmetic treatment). The trial ratings were examined to validate this method for detection of tetracycline staining. After successful completion of training, the raters graded each of the 30 study photographs for presence and degree of staining. The high and low scores were thrown out and the median for the remaining three scores was calculated. A Wilcoxon's signed rank test was performed on the median of the color scores for the study and combined (two matched) control subjects. Results. Thirty-six individuals were identified by searching through medical records for the diagnosis of RMSF. Of these, 9 had incomplete medical charts and no demographic data could be found, 6 had received chloramphenicol, 4 had affected teeth that had not yet erupted, 4 could not be reached, 2 had died and 1 refused to participate in the study, leaving 10 study subjects. The age at exposure and the dose and duration of doxycycline exposure were recorded for individuals when possible (Table 1). The mean age of the study subjects was 13.7 years (range, 11 to 19 years), and the average age at exposure was 5.1 years. In 4 cases the median color score of the study subjects was higher than the median color score of the combined controls, indicating that the study teeth were perceived to have more staining (Table 1). In 3 cases the median color score was higher for the combined controls than for their respective study subjects. There was no difference in the median values of the color score for the 3 remaining study subjects and combined controls. An analysis of these data with the use of median values indicated that there was no statistically significant difference (P = 0.38) between study and control subjects in the incidence or degree of staining of the teeth.TABLE 1Reliability between raters was examined for the individually rated teeth. For 11 (15%) teeth there was no difference in the color scores assigned to all the teeth by all five raters. In 47 (63%) teeth, there was a 1-point difference, and in 17 (23%) of the teeth, there was a 2-point difference in color grading. Therefore within the group of five raters, the highest and lowest color scores assigned to the teeth were off by one point or less in 78% of the teeth graded, indicating good reliability of grading. Discussion. Despite the volume of literature on the topic of tetracycline and staining of teeth, there are no recent studies that examine factors influencing the degree of clinical staining induced by the tetracycline, particularly doxycycline. Wallman and Hilton6 concluded that total dose of antibiotic is more important than duration of treatment. This was refuted in 1968 when Brearley and Storey7 concluded that there is an increasing chance of dental staining with increased courses of antibiotics. They noted that even one course of antibiotic can produce a band of tetracycline staining, which can induce severe discoloration. This staining is particularly acute if the antibiotic uptake is located near the dentin-enamel junction. Grossman et al.8 also concluded that increased number of exposures to tetracycline increased the potential for tooth staining. The relatively low incidence of RMSF and the small number of children treated with doxycycline limited the number of potential subjects in this study. The lag time from the drug administration, its incorporation into the enamel and the subsequent eruption of the affected teeth creates a problem with patient recall. Although the small number of subjects affected the statistical power of our study, this is the largest number of such individuals examined for doxycycline staining to date. Several studies dating back more than 20 years have examined various populations of patients exposed to doxycycline, but all are flawed in their design. For example, one group examined 25 premature infants who received doxycycline for 6 to 17 days. They examined the teeth 1 year later and found that only one child exhibited a positive color change that fluoresced under a Wood's lamp.9 Poloczek10 examined the teeth of 282 children at 14 days, 4 weeks and 1 year after administration of doxycycline. However, the investigators in both of these studies could not have examined the affected teeth, as they would not have erupted and been visible within this 1-year time period. We attempted to evaluate photographs of the subjects' teeth utilizing standard dental shade guides but found this method to be unreliable, because shade guides are intended for intraoral use. We chose to use an ordinal scale (0 to 3) which lacks precise intervals, but we agree with Brearley's statement that clinical examination was the best criterion of tetracycline tooth discoloration.7 In 23% of the rated teeth there was at least a 2-point color difference between two raters. However, in 78% of the cases there was either a 1-point difference or no difference in scores. For this reason we chose to ignore the highest and lowest scores when calculating our median score. A prospective, randomized, double blind study comparing doxycycline with tetracycline would be necessary to make a definitive statement about the decreased incidence of dental staining with doxycycline. The Committee on Infectious Diseases of The American Academy of Pediatrics has revised the RMSF treatment options for young children. However, this revision was developed without the benefit of new information from controlled clinical trials. As a result clinicians may delay treatment of this life-threatening illness out of concern for the potential tooth-staining properties of the tetracyclines. Despite the shortcomings of our study our results provide useful and unique information that suggests that the short courses of doxycycline used to treat RMSF do not cause clinically significant staining of teeth. Acknowledgments. We thank Anne C. Gentry, R.D.H., and Louise Kent, R.N., for their invaluable assistance with this study and H. James Norton, Ph.D., for his help with statistical analysis of the data. This project was supported by a grant from the Charlotte-Mecklenburg Hospital Authority Health Services Foundation. Margaret E. Lochary, D.D.S., M.S. Peter B. Lockhart, D.D.S., F.D.S., R.C.S.Ed. William T. Williams Jr., M.D. Departments of Dentistry (MEL, PBL) and Medical Education (WTW); Carolinas Medical Center; Charlotte, NC