View larger version (81K): [in this window] [in a new window]   Figure 2. Example of alveolar ridge keratosis presenting as a keratotic plaque on an edentulous area of the mandibular alveolar ridge.

Therefore, we conducted a retrospective consecutive-case review of ARK cases submitted to our biopsy service during a 10-year period to characterize the clinical and pathological features of these lesions. In addition, we aimed to determine whether including ARK in the OL category would have a substantial impact on the proportion of OL cases exhibiting dysplasia or carcinoma. In our discussion, we will examine how certain clinical warning signs may suggest OL mimicking ARK, though biopsy and microscopic examination are the only certain ways to exclude the possibility of dysplasia or carcinoma.

	METHODS

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
We conducted a retrospective review of all cases submitted to the oral pathology service of the Medical University of South Carolina College of Dental Medicine, Charleston, from 1995 through 2004. The study design and methods were approved by the university’s institutional review board. We reviewed the case history forms completed by submitting clinicians and pathology reports to identify cases of ARK and OL.

For the purposes of this study, we defined ARK as a white patch or plaque without erythema or ulceration and limited to the retromolar pad or edentulous alveolar ridge. We considered cases to be white patches or plaques if described as such or as "hyperkeratoses," "keratoses" or "leukoplakias." We excluded cases that extended beyond the retromolar pad or edentulous alveolar ridge to involve contiguous areas. For each case fulfilling the ARK case definition, we recorded basic clinical and demographic data, including patient age at biopsy, sex, ethnicity, lesion size (maximum diameter in centimeters), anatomical location and microscopic diagnosis. In addition, we noted whether the clinician expressly identified potential sources of local trauma, and we recorded what types of trauma, if any, were specified (for example, trauma from an ill-fitting denture or other appliance, occlusal trauma from an opposing tooth, masticatory trauma from eating hard foods). Since information regarding tobacco and alcohol use was not provided in the majority of cases, we were unable to perform any analysis of these factors. Two board-certified oral and maxillofacial pathologists (A.C.C. and B.W.N.) reviewed the slides of each case to ensure concurrence regarding the histopathologic diagnosis. The microscopic diagnostic categories were as follows: hyperkeratosis without dysplasia, hyperkeratosis with mild-to-moderate epithelial dysplasia, hyperkeratosis with severe epithelial dysplasia or carcinoma in situ, and squamous cell carcinoma. The criteria for histopathologic grading of dysplasia used were described by the World Health Organization.⁶ The pathologists included in the category of hyperkeratosis without dysplasia any cases that exhibited hyperkeratosis and epithelial features most consistent with reactive epithelial atypia.

In addition, we reviewed each case submitted to our laboratory during the 10-year study period to identify cases of OL. For this part of the study, we considered two OL definitions (Figure 3):

View larger version (51K): [in this window] [in a new window]   Figure 3. Methodology for oral leukoplakia (OL) case definitions 1 and 2.

– definition 1: OL excludes ARK cases;

– definition 2: OL includes ARK cases for which no specific source of trauma was expressly described by the clinician.

We designed OL definition 2 to simulate the viewpoint of an investigator who is not familiar with the concept of ARK; we would expect that such an investigator unwittingly might include in the OL case definition ARK cases for which no specific source of trauma (other than that incurred during the course of normal function) was evident. For both OL definitions, we included cases if they met the following clinical criteria: presence of lesions described as "a white patch," "white plaque," "hyperkeratosis," "keratosis" or "leukoplakia"; lesions involving intraoral mucosal sites; lesions with no identifiable cause reported by the clinician (such as local trauma, denture irritation or biting). We excluded lesions with an erythematous component (that is, those described as "red," "erythroleukoplakia" or "erythroplakia"). Because we limited our review to intraoral cases, we excluded lesions involving the lip vermilion, skin and oropharynx. We then reviewed the microscopic diagnoses of cases meeting these criteria to make a final determination regarding classification as OL. We excluded cases exhibiting histopathologic features of the following definable lesions: morsicatio, candidiasis, nicotine stomatitis, tobacco pouch keratosis, oral hairy leukoplakia, verruciform xanthoma, fibroma, giant cell fibroma, squamous papilloma, warty dyskeratoma, Fordyce granules, lymphoepithelial cyst, erythema migrans and granular cell tumor.

For OL definition 1, we considered ARK a definable lesion and, thus, excluded it from the OL category. For both case definitions, we excluded cases that, on the basis of patient history, clinical description and microscopic features, appeared to represent lichen planus or other potential lichenoid conditions (for example, lichenoid drug eruption or lichenoid reaction to amalgam). The microscopic criteria we considered characteristic of lichen planus or oral lichenoid lesions were based on those previously described by the World Health Organization⁴ and Krutchkoff and colleagues⁷ (that is, hyperkeratosis or hyperparakeratosis, sawtooth rete pegs, superficial infiltrate of lymphocytes, basal cell liquefaction, absence of dysplasia). For the purposes of this study, we chose to exclude any cases for which there would be potential controversy or uncertainty regarding final classification as OL. Such cases included lichenoid lesions with dysplasia (or so-called lichenoid dysplasia), lichenoid mucositis with some but not all features of lichen planus and leukoplakia with associated candidiasis or candidal colonization. We classified the remaining cases as OL on the basis of clinical and histopathologic features. We recorded the microscopic diagnoses of cases meeting the criteria for the two OL case definitions.

We compared the percentage of cases with dysplasia or carcinoma among the cases classified as meeting OL definitions 1 and 2. In addition, we calculated the statistic for the level of agreement that would be attained by applying OL definition 1 versus OL definition 2 to each of the cases clinically identified as leukoplakia. That is, we assessed whether each case clinically identified as OL would be categorized as OL by both definitions, as not OL by both definitions, as OL by definition 1 and not OL by definition 2, or as not OL by definition 1 and OL by definition 2). We used ² analysis to compare the distributions for sex, ethnicity and microscopic diagnosis between our ARK cases and OL definition 1 cases. For each of these parameters, we performed ² analysis for comparison of two categorical variables (degree of freedom = 1) as follows: male versus female, white versus not white ("not white" included African-American, Hispanic and other) and hyperkeratosis without dysplasia versus any dysplasia or carcinoma. We performed a two-tailed unpaired Student t test to compare the mean age at biopsy and mean lesion size for the subjects with ARK versus the subjects with OL as per definition 1. To compare the mean lesion size for the nondysplastic versus dysplastic ARK cases, we performed a two-sided Wilcoxon rank sum test. For the ², t and Wilcoxon rank sum tests, we considered a P value less than .05 statistically significant. We managed data using Microsoft Access 2003 (Microsoft, Redmond, Wash.) and performed statistical analysis using SAS statistical software (version 8.2, SAS Institute, Cary, N.C.).

	RESULTS

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
Table 1 provides a summary of data collected regarding sex, age, ethnicity, anatomical location, source of trauma and microscopic diagnoses for all subjects with ARK identified from 1995 through 2004. We identified 477 ARK cases, constituting 1.5 percent of the total number of cases accessioned (n = 31,264) by our laboratory during the 10-year period examined. In 22 patients, lesions were present bilaterally on the retromolar pad. The majority of ARK lesions (97.9 percent) exhibited hyperkeratosis without dysplasia, although we found dysplasia in 10 cases (mild-to-moderate dysplasia [n = 7], severe dysplasia [n = 3]). None of the ARK cases exhibited carcinoma in situ or squamous cell carcinoma. Among the 10 dysplastic cases, all but one exhibited one or more of the following features: history of tobacco use (positive history of tobacco use [n = 5], negative history of tobacco use [n = 2], tobacco history not stated [n = 2]), history of alcohol consumption (positive history of alcohol consumption [n = 1], alcohol history not stated [n = 9]), marked verrucous or papillary clinical appearance (n = 4), history of oral squamous cell carcinoma (n = 2), concurrent or prior OL involving areas other than the retro-molar pad or edentulous alveolar ridge (n = 3).

– the proportion of male subjects in the ARK group (66.4 percent) was significantly greater than that in the OL definition 1 group (55.9 percent) (P = .0002, ² = 13.76);

– among cases for which ethnicity data were provided, the slightly greater proportion of whites in the OL definition 1 group (94.1 percent) compared with the ARK group (90.6 percent) was statistically significant (P = .0073, ² = 7.21);

– the proportion of hyperkeratosis without dysplasia was significantly greater in the ARK group (97.9 percent) than in the OL definition 1 group (75.2 percent) (P < .0001, ² = 120.82);

– a two-sample t test showed that the mean age at biopsy for the ARK cases (53.82 years) was significantly younger than for the OL definition 1 cases (56.0 years) (P = .0042, t = 2.87).

The lesion size data for the dysplastic versus nondysplastic ARK cases were as follows:

– dysplastic ARK cases: the median was 1.5 centimeters, the mean was 1.63 cm, the standard error of the mean (SEM) was 0.38 cm, the standard deviation (SD) was 1.15 cm, and the size data were not available for one case;

– nondysplastic ARK cases: the median was 0.7 cm, the mean was 0.88 cm, the SEM was 0.032 cm, the SD was 0.64 cm, and the size data were not available for 81 cases.

Table 2 shows the lesion size data for the OL definition 1 and 2 cases. A two-sided Wilcoxon rank sum test showed that the greater median lesion size for the dysplastic ARK cases (1.5 cm) compared with that for the nondysplastic ARK cases (0.7 cm) was not statistically significant (P = .072). The small number of dysplastic ARK cases appeared to be a limiting factor for the power of the size analysis. According to a two-sample t test, we found that the greater mean lesion size for the OL definition 1 cases (1.05 cm) compared with that for the ARK cases (0.90 cm) was statistically significant (P = .0002, t = 3.74).

	DISCUSSION

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
On the basis of our study results, we propose that ARK constitutes a distinct clinicopathologic entity that investigators should consider excluding from the OL category. We hypothesize that ARK develops as a response to frictional trauma during mastication and, thus, most likely does not represent a premalignant condition. In support of this hypothesis, we found that most (97.9 percent) lesions fulfilling our ARK case definition exhibited hyperkeratosis without dysplasia at microscopic examination. However, we recognize that prospective studies with follow-up information would be necessary to confirm a lack of premalignant potential among these apparently benign hyperkeratoses.

We caution that there were a small number of cases (n = 10) that met the criteria for our ARK case definition yet exhibited variable degrees of epithelial dysplasia microscopically. We believe that this subset of cases represented true OL mimicking ARK. All but one of these 10 cases exhibited one or more of the following notable features: patient history of tobacco or alcohol use, marked verrucous or papillary clinical appearance, history of oral squamous cell carcinoma, and concurrent or prior OL involving areas other than the retromolar pad or edentulous alveolar ridge. The one dysplastic case that did not exhibit any of these clinical warning signs was described by the submitting clinician as a flat white lesion found beneath the base of a removable partial denture and involving the crest of an edentulous area of the posterior maxillary alveolar ridge in a 76-year-old man. Additional historical information regarding tobacco use, alcohol use or history of OL or oral squamous cell carcinoma was not indicated on the patient history form, although the possibility that one or more of these factors may have been present cannot be excluded entirely.

The results of this study suggest that in most cases, a white patch or plaque without erythema or ulceration and limited to the retromolar pad or an edentulous area of the alveolar ridge represents a benign frictional keratosis. However, the possibility of OL’s mimicking ARK cannot be eliminated entirely, and, in practice, the possibility of dysplasia or carcinoma can be excluded only by means of biopsy and microscopic examination.

A potential caveat regarding this list of clinical warning signs is the case of lesions with a verrucous or papillary appearance. We noted that a common clinical and microscopic feature of many typical ARK cases was a slightly irregular or corrugated surface architecture. In contrast, among the subset of exceptional cases exhibiting epithelial dysplasia, there were cases exhibiting a prominent verrucous or papillary clinical appearance corresponding to marked verrucous epithelial hyperplasia microscopically. Although there was a distinct difference between the slight corrugations of typical ARK cases and the more prominent verrucous hyperplasia of the atypical dysplastic cases, we recognize that the potential for cases with borderline verrucous features could be problematic in making this distinction.

The greater median lesion size of the dysplastic ARK cases compared with that of the nondysplastic ARK cases was not statistically significant, and the size range for the dysplastic ARK cases (0.7–3.0 cm) was within the size range of the nondysplastic ARK cases (0.2–5.0 cm). Thus, lesion size would not appear to be a reliable indicator of whether a lesion clinically manifesting as ARK may harbor dysplastic changes. Hypothetically, if a lesion manifesting as ARK were of particularly large dimensions, one might be concerned about the possibility of proliferative verrucous leukoplakia (PVL). PVL is an extremely high-risk form of OL characterized by the development of multiple OL lesions exhibiting persistent growth, often a creeping pattern along the gingival mucosa as well as other oral sites, progression through various stages (including flat or homogeneous leukoplakia, verrucous hyperplasia, verrucous carcinoma and conventional squamous cell carcinoma) and resistance to therapy with recurrence as the rule.^2,8 The slightly corrugated or verruciform surface architecture of many ARK lesions potentially could mimic an early stage of PVL. However, a frankly verrucous growth pattern and persistent creeping growth beyond the confines of an edentulous ridge or retromolar pad would be more characteristic of PVL than of ARK. To our knowledge, none of the lesions in the ARK group arose in patients with PVL. Nonetheless, we recognize that early stages of PVL are notoriously difficult to diagnose and that often PVL is a diagnosis made in retrospect only after a patient has developed multiple lesions and, ultimately, carcinoma. Thus, the only way to exclude the possibility of early PVL among the ARK group entirely would be to conduct a prospective clinical study.

In addition, even a lesion clinically consistent with ARK that microscopically exhibits hyperkeratosis without dysplasia potentially may represent true OL mimicking ARK. As described by Natarajan and colleagues,³ ARK has a tendency to exhibit the following histopathologic features: hyperorthokeratosis, hypergranulosis, acanthosis, minimal to absent inflammation, and corrugated or slightly verruciform surface architecture. These characteristic microscopic features may provide additional guidance in distinguishing ARK from true OL. We are working on an objective, quantitative analysis of these microscopic parameters in our ARK and OL cases.

Other notable clinical observations included a statistically significantly younger mean age at biopsy, greater proportion of male subjects, a slightly smaller proportion of white subjects and a smaller mean lesion size among the ARK cases compared with findings in the OL definition 1 cases. Although purely conjecturing, we surmise that younger male subjects perhaps are especially prone to greater traumatic forces during the course of masticatory function. The smaller mean lesion size of the ARK cases may reflect the limits of these lesions to traumatized areas of the alveolar ridge, whereas OL cases of premalignant potential are not bound by such limits.

Although in many cases clinicians indicated a clinical diagnosis of oral leukoplakia, the microscopic features often revealed lesions that would not be expected to exhibit a clinical appearance of the condition.

Twenty-two patients in our study exhibited ARK arising bilaterally on the retromolar pad. It seems reasonable that the mechanical forces favoring a hyperkeratotic response of the retromolar pad mucosa at times may exhibit a bilaterally symmetric distribution. Precisely why the retromolar pad was involved more often than edentulous areas of the maxillary and mandibular ridges is uncertain. This predilection simply may reflect that the presence of a retromolar pad is ubiquitous, whereas not every patient has edentulous areas of the alveolar ridge. In addition, during mastication the anterior teeth provide an initial incising action, whereas the posterior molar teeth provide the bulk of subsequent mechanical forces necessary for grinding and fragmentation of foods. During swallowing, the tongue propels the formed bolus through the fauces during entry into the oropharynx. Thus, we speculate that posterior regions of the oral cavity, including the retro-molar pad, are subject to prominent mechanical trauma during mastication, bolus formation and swallowing.

In our analysis, we used the statistic to demonstrate that the level of agreement between OL definitions 1 and 2 was good but less than excellent ( = 0.6128). The statistic is a method commonly used for measurement of interrater reliability. Calculation of the statistic provides an assessment of the level of agreement beyond that which would be expected to occur by chance. This statistical method provides more of a descriptive measure than a measure of statistical significance.⁹ Various authors have proposed guidelines for interpreting . Our assessment of "good but less than excellent" agreement is based on criteria previously published by Byrt¹⁰ ( 0, no agreement; 0.01–0.20, poor agreement; 0.21–0.40, slight agreement; 0.41–0.60, fair agreement; 0.61–0.80, good agreement; 0.81–0.92, very good agreement; 0.93–1.00, excellent agreement). Applying similar guidelines proposed by Landis and Koch¹¹ ( 0, poor agreement; 0.01–0.20, slight agreement; 0.21–0.40, fair agreement; 0.41–0.60, moderate agreement; 0.61–0.80, substantial agreement; 0.81–1.00, almost perfect), one also could consider the level of agreement between OL definitions 1 and 2 substantial but less than almost perfect.

One limitation of our study was that in assessing each case for criteria regarding our ARK and OL case definitions, we had to rely on a retrospective review of clinical information provided by the submitting clinician. Although in many cases clinicians indicated a clinical diagnosis of OL, the microscopic features often revealed lesions that would not be expected to exhibit a clinical appearance of OL (for example, fibroma, giant cell fibroma, papilloma, Fordyce granules, granular cell tumor). This observation suggests that making a clinical diagnosis of OL depends partly on one’s clinical experience and familiarity with such lesions. This finding also highlights the importance of microscopic examination in cases in which the clinician is uncertain regarding the etiology and nature of an observed lesion on the basis of the clinical findings.

In formulating our OL case definitions, we excluded hyperkeratotic lesions for which the clinician expressly described a potential source of trauma as the underlying cause. Ideally, one would remove a potential source of trauma and then monitor for resolution of the lesion. However, because of the retrospective nature of this study, removal of sources of trauma and follow-up examination were not possible to confirm whether such lesions should be considered traumatic in etiology. Instead, we had to assume a probable traumatic etiology and thus excluded such lesions from our OL case definitions. We hypothesize that for most ARK cases in which the submitting clinician did not specifically describe a potential source of trauma (such as an ill-fitting denture or other appliance, occlusal trauma or masticatory trauma), a benign hyperkeratosis developed simply as a result of trauma incurred during normal function. Thus, it may not always be feasible to remove sources of trauma as part of the clinical evaluation of a suspected leukoplakic lesion. Finally, we cannot entirely exclude the possibility that traumatic lesions may have been included inadvertently among the OL cases because of a lack of adequate clinical data.

Nevertheless, the clinical and microscopic features of cases satisfying the criteria for our OL case definitions appeared to be valid because they were comparable with those reported in previous studies of OL. In particular, in their landmark study regarding the clinicopathologic features of OL, Waldron and Shafer¹² found that 19.9 percent of OL specimens examined exhibited "potentially serious pathology" (that is, dysplasia or carcinoma) microscopically. This percentage is similar to the 20.2 percent of cases in OL definition 2 exhibiting dysplasia or carcinoma in our study. When we excluded ARK from the OL case definition (OL definition 1), the percentage of cases with dysplasia or carcinoma increased from 20.2 to 24.8 percent. This 4.6 percent difference represents a 23 percent increase; we believe that this increase is remarkable in illustrating how a single modification of the OL case definition may change the diagnostic profile. If one were to consider additional refinements of the OL case definition, one could potentially isolate a group of lesions that would have a substantially greater likelihood of exhibiting dysplastic or already malignant histopathologic features compared with those of lesions included in more traditional OL definitions.

The results of our study suggest that ARK is a fairly common lesion. ARK constituted 21 percent of cases included in OL definition 2 and 1.5 percent of all cases our laboratory accessioned during the 10-year study period. Although specific references to ARK in population-based studies regarding the prevalence of various oral lesions are lacking, a few such studies reported the prevalence of frictional keratoses in general. In their study of oral mucosal lesions in U.S. adults based on data from the Third National Health and Nutrition Examination Survey, Shulman and colleagues¹³ reported the prevalence of frictional white lesions to be 2.67 percent and the prevalence of OL to be 0.38 percent. In contrast, in two published reports of oral mucosal lesions in an adult Swedish population, Axéll^14,15 reported a prevalence of approximately 5.5 percent for frictional white lesions and 3.37 to 3.6 percent for OL. In a mass screening examination of 23,616 white adult Americans, Bouquot¹⁶ noted a prevalence of lesions with clinically evident white or keratotic surface changes to be 27.8 per 1,000 people and the prevalence of OL to be 29.1 per 1,000 people. Although the observed differences in prevalence of frictional keratoses and OL among these studies may be due partly to differences in the populations studied, differences in diagnostic criteria also may account for this variation. On the basis of the methods described by these authors, it is not possible to determine whether ARK was included among frictional keratoses or OL.

Precise definitions of OL are crucial for achieving greater consistency and enabling meaningful comparisons across studies regarding oral precancers. Proper evaluation of epidemiologic features, treatment results and malignant transformation rates of OL is not possible at present because of the considerable variations among authors in defining and applying the term "OL."¹⁷ For many decades, there has been much interest in refining the OL case definition. In practice, many clinicians consider a white patch or plaque of the oral mucosa that cannot be wiped away to represent OL. Since OL cannot be wiped away, wiping can be used as a clinical test to aid in distinguishing OL from other oral white lesions that can be wiped away, such as pseudomembranous candidiasis. Nonetheless, because there are a number of white lesions that cannot be wiped away and do not necessarily constitute OL, a more precise definition is needed.

One of the earliest formal definitions of OL was proposed in 1978 by the World Health Organization Collaborating Centre for Oral Precancerous Lesions.⁴ These authors defined OL as "a white patch or plaque that cannot be characterized clinically or pathologically as any other disease"; however, they also commented that the "rather negative" nature of this definition likely would lead to future revisions and refinements.⁴ Further modifications were proposed by experts who convened at an international seminar on OL held in Malmö, Sweden, in 1983¹⁸ and by experts who met at an international symposium on oral white lesions held in Uppsala, Sweden, in 1994.⁵ Some of the proposed revisions resulting from these meetings included the following:

– inclusion of etiologic factors in the definition of OL (that is, OL should not be associated with any physical or chemical causative agent except tobacco);

– substitution of "whitish" for "white," because color can be subjective and variable;

– distinction between a provisional diagnosis of OL on the basis of clinical examination and a definitive diagnosis of OL made after persistence despite elimination of potential causative factors and after histopathologic examination of persistent lesions.

The retrospective nature of our study made it impossible to eliminate possible causative factors for two to four weeks to monitor for regression of a lesion.

In 2002, van der Waal and Axéll¹⁷ proposed a uniform reporting system for OL. This system included staging (based on size and presence or absence of dysplasia) and recognition of variable levels of certainty in making a diagnosis of OL (on the basis of results of clinical examination, follow-up after elimination of suspected etiologic factors and histopathologic examination). In addition, these authors listed a number of common definable oral white lesions to be excluded from consideration as OL. In particular, they highlighted two problematic issues:

– in the case of hyperplastic candidiasis, there is no consensus as to whether to regard these lesions as "candida-associated leukoplakia";

– tobacco-induced lesions such as smoker’s palate and snuff-related lesions are not traditionally described as leukoplakia, although some of them may transform into cancer.

Although many of the above investigators regard frictional lesions related to mechanical trauma (such as biting, vigorous toothbrushing and dental restorations) as definable lesions, none specifically mentioned or described ARK as an example of frictional keratosis.^5,17

In our study, we based our OL definition 1 largely on the OL definition described in 2002 by van der Waal and Axéll.¹⁷ Accordingly, we considered a clinical diagnosis of OL as "a predominantly white lesion of the oral mucosa that cannot be characterized as any other definable lesion"¹⁷ and considered van der Waal and Axéll’s list of common definable white lesions to be excluded from the OL category. In addition, these authors recognized the distinction between a provisional clinical diagnosis and a definitive one on the basis of elimination of suspected etiologic factors and histopathologic examination. Although we included histopathologic analysis in our evaluation of OL cases, the retrospective nature of our study made it impossible to eliminate possible causative factors for two to four weeks to monitor for regression of a lesion. As discussed above, in cases in which the clinician specifically described a potential source of trauma, we had to assume a probable traumatic etiology and, thus, excluded such lesions from both of our OL case definitions. In addition, because we had to rely on clinical descriptions provided on the biopsy requisition forms by the submitting clinician, clinical subclassification of lesions as homogeneous versus nonhomogeneous leukoplakia was not possible.

In rare cases, gingival hyperkeratosis may occur in association with inherited conditions, such as hyperkeratosis palmoplantaris and attached gingival hyperkeratosis syndrome, autosomal dominant tylosis and carcinoma syndrome, oral hyperkeratosis with congenital strictures and squamous cell carcinoma of the esophagus and pachyonychia congenita. In many of these inherited conditions, there is a tendency toward diffuse oral hyperkeratoses that may involve the gingiva and potentially other oral mucosal sites. However, some patients with these conditions have a particular predisposition for involvement of the traumatized area—for example, edentulous denture-bearing areas of the alveolar ridge.^19,20 The presence of other intraoral or extraoral associated findings and a family history may aid in diagnosing such conditions. In our study, none of the patients had a known history of inherited disorders associated with oral hyperkeratosis.

	CONCLUSION

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
We have described a distinct type of frictional keratosis for which we use the term "alveolar ridge keratosis." ARK manifests as a white lesion limited to the retromolar pad or an edentulous area of the alveolar ridge. In our study, ARK cases were characterized by a statistically significantly younger mean age at biopsy, a greater proportion of male subjects, a slightly smaller proportion of white subjects and a smaller mean lesion size compared with the OL cases. Microscopically, the majority of ARK lesions were benign hyperkeratoses. Although prospective studies are needed to confirm a lack of premalignant potential, the results of this study suggest that investigators should consider excluding ARK from studies on OL. Furthermore, certain features of the patient history or clinical examination—such as history of tobacco or alcohol use, history of multiple OL lesions, history of oral squamous cell carcinoma or verrucous clinical appearance—may alert clinicians to potential cases of OL mimicking ARK.