Evaluation of dental erosion in patients with gastroesophageal reflux disease

Belinda L. Gregory-Head, BDS, MS, Donald A. Curtis, DMD, Lawrence Kim, MD, and John Cello, MD University of the Pacific School of Dentistry, and University of California, San Francisco, Calif. Statement of problem. The cause of dental erosion may be difficult to establish because of its many
presentations. Determination of the cause is an important aspect of diagnosis before extensive prosthodontic rehabilitation. Purpose. This cross-sectional study evaluated the association between loss of tooth structure as a result of dental erosion and gastroesophageal reflux disease. Material and methods. Twenty consecutive adult dentate subjects referred to the Division of Gastroenterology for investigation of gastroesophageal tract disease were also evaluated for signs of dental erosion. All subjects underwent a dental evaluation that included a patient history to determine potential etiologic factors responsible for dental erosion. Subjects were examined clinically to quantify loss of tooth structure using a Tooth Wear Index (TWI). Endoscopic examination and 24-hour pH manometry were carried out to determine which subjects met the criteria for gastroesophageal reflux disease (GERD). Scores for maxillary versus mandibular dentition and anterior versus posterior dentition were also compared. Data were analyzed with the Kruskal-Wallis test (P=.004). Results. Ten subjects were diagnosed with GERD and 10 subjects had manometry scores below the level indicating GERD. Overall, subjects diagnosed with GERD had significantly higher TWI scores compared with control subjects (mean difference = 0.6554; P=.004). GERD subjects had higher TWI scores in all quadrants, except in the mandibular anterior region where there was no difference. Conclusion. The results indicated that a relationship exists between loss of tooth structure, as measured by the TWI index, and the occurrence of GERD in this group of subjects. (J Prosthet Dent 2000;83: 675-80.)

CLINICAL IMPLICATIONS
Chronic acid exposure can have a profound effect on the dentition. Patients who exhibit significant dental erosion and require prosthodontic rehabilitation should be considered for investigation for GERD if other etiologic factors are not apparent.

ental erosion has been reported with varying prevalences in the population and may be as high as 42%.1 The impact of erosion on prosthodontic treatment planning is often unrecognized. Patients with severe erosion may require extensive dental rehabilitation. Unless the cause of the erosion is accurately determined, the prognosis of the dental treatment cannot be assured. Normal loss of tooth structure is usually due to functional microwear or attrition and proceeds at physiologic levels throughout life. It is related to dietary and oral hygiene habits.2 Normal vertical loss
Presented at the Pacific Coast Society of Prosthodontists Annual Meeting, Whistler, B.C., Canada, June 1997. aAssistant Professor, Department of Removable Prosthodontics, University of the Pacific School of Dentistry. bProfessor, Division of Prosthodontics, University of California. cResearch Fellow, Department of Gastroenterology, University of California. dProfessor and Chair, Department of Gastroenterology, University of California. JUNE 2000

of dental hard tissues has been estimated to be about 65 m per year.3 This may be 3 to 4 times higher in patients with bruxing habits.4 Smith and Robb,5 in a study of 1007 patients in the general population, reported greater loss of tooth structure on the palatal surfaces of maxillary anterior teeth than mandibular anterior teeth. They also noted greater wear of the mandibular occlusal surfaces than maxillary occlusal surfaces in all age groups. Pathologic loss of tooth structure occurs at an accelerated rate and may be due to 1 or more processes: abnormal attrition that is due to clenching or bruxing of one tooth surface against another; abrasion that is due to physical wear by extraneous objects such as toothbrushes; and erosion, which may be defined as chemical damage, mostly demineralization by acids other than those produced by bacteria. Some chelation of the organic phase of the matrix may also occur.6 The cause of dental erosion is associated with 1 or a combination of several factors, making diagnosis difficult. Reports in the literature have associated erosion
THE JOURNAL OF PROSTHETIC DENTISTRY 675

THE JOURNAL OF PROSTHETIC DENTISTRY

GREGORY-HEAD ET AL

with dietary factors such as citrus fruits and carbonated drinks.7 Occupational exposure to proteolytic enzymes8 and chlorinated swimming pool water9 has also been implicated. The chronic vomiting associated with alcoholism and bulimia nervosa has been extensively documented as associated with dental erosion.10 Acid movement from the stomach into the esophagus is usually controlled by an antireflux mechanism involving the lower esophageal sphincter. A failure of this mechanism can result in the chronic condition known as gastroesophageal reflux disease (GERD).11 GERD has recently been shown to be associated with dental erosion under some circumstances.12-14 The diagnosis of GERD is usually based on history and endoscopic examination. A 24-hour pH manometry test can be completed to determine the severity of the reflux of gastric contents into the esophagus.15 Bartlett et al16 evaluated the relationship among oral pH, GERD, and dental erosion. Bartlett measured the oral pH and esophageal pH simultaneously and found that a pH measurement of less than 5 at either site was correlated with palatal erosion. This supports the theory that gastric acids entering the esophagus can travel superiorly and enter the oropharynx to affect the teeth. Because the results from both monitors were correlated to palatal erosion, the choice of only 1 monitor in our study is justified. The nasogastric tube monitor developed by Johnson and Demeester17 is the diagnostic tool used routinely for determination of GERD and has a sensitivity and specificity of nearly 90%; thus, it was selected for use in our study. Salivary flow rates and pH were not measured in this study but may contribute to the relative severity of erosion lesions in individual subjects. Gudmundsson et al18 and Woltgens et al19 reported lower salivary buffering capacity or unstimulated flow rates in the erosion patients they examined. Because dental erosion is often a multifactorial phenomenon, the additional insult of poor buffering capacity in a subject with GERD is likely to cause erosion lesions that are more severe and occur more quickly than would otherwise be the case. Similarly, GERD patients who also brux may experience a more rapid loss of tooth structure. These complicating factors represent areas for further investigation. If reflux is suspected as the underlying cause for a patients dental erosion, certain preventive measures can be taken immediately. These include dietary control of reflux-causing foods such as peppers and onions. Some authors have advocated the use of antacids, which may be dissolved completely in the mouth before swallowing,20 or the use of bicarbonatecontaining chewing gums to buffer the intraoral environment.21 Toothbrushing should be avoided during active reflux episodes because the acid-weakened outer enamel layer is more easily removed by vigorous brushing. The use of stents with 1.1% neutral sodium
676

fluoride gel to remineralize the damaged enamel may be considered. There is in vitro evidence that the use of a titanium tetrafluoride solution on human enamel will create an acid-resistant surface layer.22 This study evaluated 20 subjects who were treated in the department of gastroenterology at San Francisco General Hospital. Subjects were examined for signs of dental erosion and reduced pH in the esophagus. The purpose of this study was to determine whether a relationship exists between dental erosion and GERD in this group of subjects.

MATERIAL AND METHODS

An initial cohort of 46 consecutive new patients was examined on presentation to the department of gastroenterology. The first 20 adult subjects (age range 18 to 69 years) with at least 6 uncrowned teeth in each arch were then further evaluated and included in the study. Each subjects dentition was examined by a prosthodontist and assigned a Tooth Wear Index score (TWI) using the protocol developed by Smith and Knight.23 The TWI index has been used by several other investigators in this field and utilizes a numerical scale to indicate clinical levels of tooth structure loss.24 The scoring index is based on clinical presentation and ranges from no loss of any surface characteristics (score 0) to complete loss of enamel with exposure of secondary dentin or pulp chambers (score 4). Buccal, lingual, occlusal, and incisal surfaces were assigned a score for each tooth examined. A mean total TWI score was calculated for each tooth for each subject. The dental evaluation included clinical examination, standardized intraoral photographs, diagnostic casts, and a questionnaire. The questionnaire detailed the subjects possible exposure to etiologic factors associated with dental erosion, including ingestion of citrus fruits, vinegar, and carbonated drinks. It also detailed exposure to occupational aerosols such as acids from a car battery manufacturer.20 The TWI scores were reevaluated 1 month later to assess examiner reliability. The scores given at the clinical examination and those recorded 1 month later using only diagnostic casts and photographs remained the same for each subject. After the dental evaluation, the same subjects were evaluated by 1 gastroenterologist for the presence of GERD using a 24-hour pH manometry test. A 2-channel pH probe was placed transnasally to a position 5 cm above the lower esophageal sphincter (LOS). The probe consisted of 2 pH-sensitive electrodes, one at the tip and the other 15 cm proximal to it. Information regarding the pH at the 2 electrode sites in the esophagus (5 cm and 20 cm above the LOS) was gathered every 4 seconds and stored on a portable data recorder that was kept with the patient. Data were recorded over a 24-hour period during
VOLUME 83 NUMBER 6

GREGORY-HEAD ET AL

THE JOURNAL OF PROSTHETIC DENTISTRY

Table I. Comparison of mean Tooth Wear Index scores for control and GERD groups
GERD group Test for comparison N Mean SD SEM N Reflux group Mean SD SEM Mean difference P value

Overall All maxillary surfaces Maxillary anterior surfaces Maxillary posterior surfaces All mandibular surfaces Mandibular anterior surfaces Mandibular posterior surfaces

9 9 10 9 10 10 10

0.9539 0.9411 1.0445 0.7216 0.8751 0.8238 0.9256

0.545 0.584 0.728 0.671 0.568 0.646 0.633

0.182 0.195 0.230 0.224 0.180 0.204 0.200

10 10 10 10 10 10 10

0.2985 0.2708 0.3070 0.2346 0.3262 0.3472 0.3052

0.282 0.306 0.347 0.343 0.275 0.251 0.379

0.089 0.097 0.110 0.109 0.087 0.079 0.120

0.6554 0.6703 0.7375 0.4870 0.5489 0.4766 0.6212

.004* .005* .010* .059 .013* .043* .016*

GERD = Gastroesophageal reflux disease. One subject did not have sufficient maxillary posterior teeth to be scored, but did have sufficient anterior teeth for inclusion. *P<.05. Anterior teeth included incisors and canines, remaining teeth placed in the posterior group.

which the patients were asked to keep a log of meal times and their bodily position, either supine or upright. At the end of the observation period, the data were downloaded to a personal computer. The readings were compared with established norms and given a score, according to a standardized protocol developed by Johnson and DeMeester.25 The Johnson and DeMeester scoring system is the usual method for reporting the various parameters of gastroesophageal reflux. Reflux was defined as occurring when the pH was less than 4. Six components were measured and obtained from the data recorded: percentage time pH was less than 4 for 24 hours, percentage time for supine and upright positions, number of single episodes, number of episodes greater than 5 minutes, and the time of the longest episode. The 24-hour score was determined by calculating the number of standard deviation equivalents in each measured value of the 6 components starting at a fixed reference point and placing 2 standard deviations below the respective measured mean value in the group of control patients (asymptomatic). This scoring system allows comparison to controls and to other investigations and established norms. On the basis of the results of the pH manometry studies, the patients were identified and placed into either the control group (results within normal limits) or the GERD group (irrefutable evidence of pathologic reflux over the 24-hour period). The study was approved by the Committee on Human Research Subjects at UCSF. The procedures were explained to the subjects and their written consent obtained. The prosthodontist and the gastroenterologist were blinded to each others findings. Subjects were assigned a code and their names were withheld from the examiners. Finally, the control group and GERD group were identified and the TWI scores for each evaluated.
JUNE 2000

Statistical analysis
TWI scores were averaged across the teeth for each arch in both the control group and the GERD group. The data distribution followed a normal pattern. The mean and standard deviation of TWI scores were calculated for the GERD and control subjects. The mean and standard deviations of TWI scores were also calculated for the following groups in both the GERD and control subjects: all maxillary tooth surfaces, all mandibular tooth surfaces, all maxillary anterior tooth surfaces, all maxillary posterior tooth surfaces, all mandibular anterior surfaces, and all mandibular posterior surfaces. A power analysis determined that an effect size difference of one unit on the TWI scale could be identified with 2 groups of as few as 4 subjects. After evaluation of the pH manometry studies, it was found that there were 10 subjects with reflux and 10 subjects in the control group. It was therefore determined that statistical analysis could proceed with a planned unpaired t test. Significant differences between subsets were tested using the Kruskal Wallis test. The mean TWI scores for each group were also plotted graphically and related to position in each dental arch. P values of .05 or less were considered significant.

RESULTS
Subjects diagnosed with GERD had significantly higher TWI scores (P=.004). There were significant differences between TWI scores in the GERD group and control group for mandibular surfaces (P=.013) and maxillary surfaces (P=.005). There were several significant differences identified between the control group and the GERD group (Table I). There was no significant difference between the 2 groups when the TWI scores for the mandibular incisal edges were compared. When the mean TWI scores for each tooth in each group were represented graphically, distinct patterns of
677

THE JOURNAL OF PROSTHETIC DENTISTRY

GREGORY-HEAD ET AL

Fig. 1. Subject C. Maxillary arch showing severe lingual erosion of incisors and generalized loss of enamel on occlusal surfaces.

Fig. 3. Graphic representation of mean TWI values for each tooth in maxilla and mandible in GERD and control groups.

the similar distribution patterns were evident but, with the GERD group, the distribution of the graph shifted 1.5 to 2.0 units higher on the TWI scale (Fig. 3). Some subjects in the GERD group also showed loss of buccal cervical tooth structure to the level of dentin (Figs. 4 and 5).

DISCUSSION
There appears to be a distinct pattern of tooth structure loss that is unique to each arch. The results of this study confirm the earlier work by Smith and Robb.5 They show a similar distribution of loss of tooth structure in the control group as in the general population. This study shows that the pattern is maintained but more severe in subjects with GERD. Other investigators also found that the maxillary lingual surfaces are the most severely affected in patients with reflux of gastric contents, the mandibular anterior surfaces remaining largely unaffected.26 The mandibular anterior teeth may be protected by the tongue or be more constantly bathed in saliva and therefore protected by its buffering action. Skeletal relationships and occlusal schemes were not recorded for each subject and may represent a limitation of the study. Because the first 20 subjects who met the inclusion criteria were chosen, it was assumed that selection bias toward one particular scheme would be equal in both groups. Positive results with 24-hour pH manometry in this study show a relationship with pathologic loss of tooth structure as measured by the TWI. In patients with GERD, further tests may be required to determine the cause. A hiatus hernia or incompetent gastric sphincter may be corrected surgically and in those patients in whom there is an overproduction of gastric acid, medication may be all that is required. The usual regimen is cimetadine 400 mg
VOLUME 83 NUMBER 6

Fig. 2. Subject C. Mandibular arch showing generalized loss of tooth structure on posterior occlusal surfaces with preservation of mandibular incisal edges.

tooth wear were observed in both maxillary and mandibular arches. The maxillary arch scores showed a pattern of higher TWI scores on the anterior teeth than for the premolars and molars. The TWI scores peaked at the midline for the GERD group. This pattern was more pronounced in the GERD group than the control group. Anterior maxillary teeth in the GERD group often showed significant loss of lingual tooth structure (Fig. 1). The mandibular arch scores showed a different distribution pattern with the lowest scores arising from the mandibular anterior teeth and the highest scores being recorded for the premolars and molars (Fig. 2). The distribution of erosion lesions for the mandibular arch was similar for the GERD and control group but the lesions were significantly more severe in the GERD group. When the graphs of both arches for the control group and the GERD group were plotted together,
678

GREGORY-HEAD ET AL

THE JOURNAL OF PROSTHETIC DENTISTRY

Fig. 4. Frontal view of subject F. Buccal cervical and erosion lesions are evident on right and left sides.

Fig. 5. Subject F. Left lateral view illustrating loss of enamel surface characteristics and cervical erosion lesions to dentin.

Table II. Medical therapies for GERD

Dosage Manufacturer

H2-receptor antagonists Cimetidine (Tagamet) Famotidine (Pepcid) Nizatidine (Axid) Ranitidine (Zantac) Proton pump inhibitors Lansoprazole (Prevacid) Omepraxole (Prilosec) Prokinetic agents Cisapride (Propulsid)
GERD = Gastroesophageal reflux disease.

400 mg bid 20 mg bid 150 mg bid 150 mg bid 30-60 mg daily in morning before food (adjusted to suit individual needs) 20 mg daily in morning before food 10 mg bid

SmithKline Beecham Johnson & Johnson/Merck Lilly Glaxo Wellcome Takeda-Abbott Pharmaceutical Astra/Merck Janssen Pharmaceutica

twice daily or omeprazole 20 mg daily. Cimetadine is a histamine H2 receptor antagonist and omeprazole is a proton pump inhibitor, both consequently reduce gastric acid secretion. Omeprazole has recently been associated with dry mouth, therefore it may be prudent to consider the alternatives.27 A list of standard therapeutic alternatives for GERD is given in Table II. The treatment of GERD is both safe and predictable. If a referral for 24-hour pH manometry is considered, it may be made directly to a gastroenterology department or through the primary care provider. Medical insurance plans usually cover the procedure and the fee is approximately $200 to $400. Communication with the gastroenterologist regarding the need to rule out GERD before dental reconstruction is important. Because there is a strong correlation between loss of tooth structure and positive pH manometry results, it may be appropriate to test for GERD in this way to determine the cause of severe erosion if no other factors are evident. A positive 24-hour test can be followed with safe and predictable treatment for the reflux phenomenon.
JUNE 2000

One of the goals of prosthodontic care is long-term predictability. The establishment of a stable, healthy oral environment before beginning definitive treatment is a predictor of success. Although there is still a need for a careful history and examination as part of the diagnostic process, and other tests may be appropriate, it is our conclusion that a 24-hour pH manometry study may be a useful tool in the diagnostic armamentarium for a difficult erosion case.

CONCLUSIONS
This study evaluated the relationship between GERD and dental erosion in a population of outpatients at San Francisco General Hospital. On the basis of the results of this study, the following conclusions were drawn: 1. A relationship was demonstrated between the 2 phenomena. 2. Subjects with GERD exhibited greater loss of tooth structure than those in the control group as measured by the Tooth Wear Index (mean difference of 0.6554 on a scale of 0 to 4; P=.004).
679

THE JOURNAL OF PROSTHETIC DENTISTRY

GREGORY-HEAD ET AL

3. Both groups demonstrated a similar distribution of loss of tooth structure to that which has been found in the general population, but the loss was more severe in the GERD group especially in the maxillary anterior teeth (mean difference of 0.7375; P=.01).
REFERENCES
1. Lussi A,Schaffner M, Hotz P, Suter P. Dental erosion in a population of Swiss adults. Comm Dent Oral Epidemiol 1991;19:286-90. 2. Westergaard J, Moe D, Pallesen U, Holmen L. Exaggerated abrasion/erosion of human dental enamel surfaces: a case report. Scand J Dent Res 1993;101:265-9. 3. Lambrechts P, Vanherle G, Vuylsteke M, Davidson CL. Quantitative evaluation of the wear resistance of posterior dental restorations: a new three-dimensional measuring technique. J Dent 1984;12:252-67. 4. Xhonga FA. Bruxism and its effect on the teeth. J Oral Rehabil 1977;4:6576. 5. Smith BG, Robb ND. The prevalence of toothwear in 1007 dental patients. J Oral Rehabil 1996;23:232-9. 6. Robb ND, Cruwys E, Smith BG. Is lingual surface attrition of the maxillary teeth (LSAMAT) caused by dental erosion? Am J Phys Anthropol 1991;85:345-51. 7. Pindborg JJ. Chemical and physical injuries. In: Pindborg JJ. Pathology of the dental hard tissues. Philadelphia: WB Saunders; 1970. p. 312-25. 8. Westergaard J, Moe D, Pallesen U, Holmen L. Exaggerated abrasion/erosion of human dental enamel surfaces: a case report. Scand J Dent Res 1993;101:265-9. 9. Centerwall BS, Armstrong CW, Funkhouser L, Elzay RP. Erosion of dental enamel among competitive swimmers at a gas-chlorinated swimming pool. Am J Epidemiol 1986;123:641-7. 10. Spigset O. Oral symptoms in bulimia nervosa. A survey of 34 cases. Acta Odontol Scand 1991;49:335-9. 11. Bartlett DW, Evans DF, Smith BG. The relationship between gastrooesophageal reflux disease and dental erosion. J Oral Rehabil 1996,23: 289-97. 12. Jarvinen VK, Rytomaa II, Heinonen OP. Risk factors in dental erosion. J Dent Res 1991;70:942-7. 13. Meurman JH, Toskala J, Nuutinen P, Klemetti E. Oral and dental manifestations in gastroesophageal reflux disease. Oral Surg Oral Med Oral Pathol 1994;78:583-9. 14. Barlett D, Smith B. The dental relevance of gastro-oesophageal reflux: part 2. Dent Update 1996;23:250-3. 15. Eckardt VF, Dilling B, Bernhard G. The impact of open access 24-h pHmetry on the diagnosis and management of esophageal reflux disease. Am J Gastroenterol 1999;94:616-21.

16. Bartlett DW, Evans DF, Anggianish A, Smith BG. A study of the association between gastro-oesophageal reflux and palatal dental erosion. Br Dent J 1996;181:125-31. 17. Weiner GJ, Morgan TM, Cooper JB, Wu WC, Castell DO, Sinclair JW, et al. Ambulatory 24-hour esophageal pH monitoring. Dig Dis Sci 1988;33: 1127-33. 18. Gudmundsson K, Kristleifsson G, Theodors A, Holbrook WP. Tooth erosion, gastroesophageal reflux, and salivary buffer capacity. Oral Surg Oral Med Oral Pathol 1995;79:185-9. 19. Woltgens JH, Vingerling P, de Blieck-Hogervorst JM, Bervoets DJ. Enamel erosion and saliva. Clin Prev Dent 1985;7:8-10. 20. Lazarchik DA, Filler SJ. Effects of gastroesophageal reflux on the oral cavity. Am J Med 1997;103:107s-13s. 21. ten Cate JM, Imfeld T. Dental erosion, summary. Eur J Oral Sci 1996; 104(2 Pt 2):241-4. 22. Buyukyilmaz T, Ogaard B, Rilla G. The resistance of titanium tetra-fluoride-treated human enamel to strong hydrochloric acid. Eur J Oral Sci 1997;105(5 Pt 2):473-7. 23. Smith BG, Knight JK. An index for measuring the wear of teeth. Br Dent J 1984;156:435-8. 24. Barlett DW, Coward PY, Nikkah C, Wilson RF. The prevalence of tooth wear in a cluster sample of adolescent schoolchildren and its relationship with potential explanatory factors. Br Dent J 1998;184:125-9. 25. Johnson LF, DeMeester TR. Twenty-four-hour pH monitoring of the distal esophagus. A quantitative measure of gastroesophageal reflux. Am J Gastroenterol 1974;62:325-32. 26. Scheutzel P. Etiology of dental erosionintrinsic factors. Eur J Oral Sci 1996;104(2 Pt 2):178-90. 27. Teare JP, Spedding C, Whitehead MW, Greenfield SM, Challacombe SJ, Thompson RP. Omeprazole and dry mouth. Scand J Gastroenterol 1995; 30:216-8. Reprint requests to: DR BELINDA L. GREGORY-HEAD DEPARTMENT OF REMOVABLE PROSTHODONTICS SCHOOL OF DENTISTRY UNIVERSITY OF THE PACIFIC 2155 WEBSTER ST SAN FRANCISCO, CA 94115 FAX: (415)929-6654 E-MAIL: belinda_head@yahoo.com Copyright 2000 by The Editorial Council of The Journal of Prosthetic Dentistry. 0022-3913/2000/$12.00 + 0. 10/1/107193 doi:10.1067/mpr.2000.107193

680

VOLUME 83 NUMBER 6