Community Dent Oral Epidemiol 2001; 29: 260–3 Copyright C Munksgaard 2001Printed in Denmark . All rights reserved

ISSN 0301-5661

Joseph Katz1,2, Gavriel Chaushu1,4 andBenjamin Peretz3Recurrent oral ulcerations1Israeli Defense Force, Medical Corps,2Department of Oral Diagnosis and3Department of Pediatric Dentistry, Oralassociated with recurrent herpes Medicine, Oral Radiology, The HebrewUniversity Hadassah School of DentalMedicine, Jerusalem, 4Department of OralSurgery, Sheba Medical Center, Tellabialis – two distinct entities? Hashomer, Israel

Katz J, Chaushu G, Peretz B: Recurrent oral ulcerations associated with recurrentherpes labialis – two distinct entities? Community Dent Oral Epidemiol 2001;29: 260–3. C Munksgaard, 2001

Abstract – We investigated the prevalence of recurrent aphthous ulcerations (RAU)and recurrent herpes labialis (RHL) in a young adult population. The study groupconsisted of 20 689 soldiers (aged 18–21 years, mean 19.8∫1 years). There were14 991 men and 5698 women. Every patient was asked to complete a self-reporthealth questionnaire which requested information about the occurrence of eitherRAU or RHL. Data were analyzed using the chi-square test, and the odds ratio wasalso determined. The prevalence of RAU and RHL was 6.2% and 5.4%, respectively.The prevalence among women was significantly higher (P,0.001), than among menfor both RAU (7.1% and 5.8%, respectively) and RHL (6.8% and 4.9%, respectively).

Key words: recurrent aphthous ulcerations;Among women, the chance of having RAU for those who were RHL positive was recurrent herpes labialis; prevalence6.88 greater than for those RHL negative. Among men, the chance of having RAU

Dr. Joseph Katz, Dept. of Oral Medicine,for those who were RHL positive was 12.37 greater than for those RHL negative. InP.O. Box 9618, Jerusalem, Israel

women, the chance of having RHL for those who were RAU positive was 6.88 Tel: π972 2 6785934greater than for those who were RAU negative. In men, the chance of having RHL Fax: π972 2 6785126

e-mail: katzdr/internet-zahav.net.ilfor those who were RAU positive was 12.37 greater than for those who were RAUnegative. It is concluded that a similar underlying pathological process or unknown Submitted 15 December 1999;cofactor may be involved in both RAU and RHL. accepted 31 October 2000

Recurrent aphthous ulcerations (RAU) are recur-rent painful ulcerations on the moist mucous mem-branes of the oral cavity. The great morbiditycaused by the lesions and the high prevalence inthe entire population make the understanding ofthis phenomenon of great importance.

Epidemiological investigations of RAU revealeda wide range of occurrence (5–60%), depending onthe characteristics of the study group and the modeof evaluation (1–11). The prevalence of RAU in thegeneral population has been estimated to be nearly20% (12).

The specific etiology of RAU is not clear. Hormon-al changes, stress, trauma, hypersensitivity to foodproducts, hematological disorders, vitamin B1-B12

and folic acid deficiencies and malabsorptiondiseases (such as celiac disease and Crohn’s disease)have been associated with a higher prevalence of

260

RAU. The exact pathogenesis is not completelyunderstood (13). A strong genetic predisposition hasbeen demonstrated (14).

Some investigators have reported an associationbetween RAU and reactivation of varicella-zostervirus and cytomegalovirus (15). An association be-tween measles, Crohn’s disease and RAU has alsobeen suggested (16).

Some studies point towards a defective immuno-logic system as the most dominant etiologic factor(17–19). One study has raised an hypothesis aboutan increased expression of class I and class II majorhistocompatibility complex antigens in the epitheli-um of RAU patients, which are recognized asforeign bodies, and attacked by the immune sys-tem (18).

Several studies have shown that the prevalenceof recurrent herpes labialis (RHL) was higher in

Recurrent oral ulcerations associated with RHL

individuals with RAU than in the entire population(2, 7, 20).

RHL is characterized by recurrent appearances offluid-filled blisters on, or adjacent to, the vermilionborder of the lips. Precipitating factors includetrauma, prolonged exposure to sunlight, psycho-logical stress and menstruation (2). The prevalenceof RHL, as obtained from self-reported question-naires, ranges between 15% and 33% in differentstudies (1, 2, 7, 8).

The purpose of the present study was to assessthe prevalence of RAU and RHL in a population ofyoung adults and to determine the possible interre-lationship between those two phenomena.

Material and methods

The study population consisted of 20 689 soldiersfrom 22 clinics. There were 14 991 men and 5698women. Ages ranged from 18 to 21 years with amean of 19.8∫1 years. Each participant was asked tocomplete a health questionnaire that included twoquestions, similar to those used by Ship et al. (1):1. Have you ever had ‘‘canker sores’’ or other re-

curring ulcers inside your mouth?2. Have you ever had recurrent blisters on your

lips?Subjects were asked to respond in a yes/no man-

ner to those questions. Subjects who reported posi-tively were considered to have a history of eitherRAU positive or RHL positive. Data from thehealth questionnaire was analyzed using a chi-square test. The odds ratio was also determined ac-cording to (P1/P2)¿(Q2/Q1) where p is the risk inci-dence and QΩ1-P (21).

Results

The prevalence of RAU and RHL in the entirepopulation by gender is presented in Table 1. Thetotal self-reported prevalence of RAU and RHLwas 6.2% and 5.4%, respectively. The prevalenceamong females was significantly higher (P,0.001),than among males for both RAU (7.1% and 5.8%,respectively) and RHL (6.8% and 4.9%, respec-tively).

Since an interaction among the variables wasfound (difference between odds ratios in men andwomen), the statistical model that was chosen asthe appropriate was the loglinear model in theform of:

pijkΩlxi πly

j πlzkπlxy

ij πlxzjk πlyz

jk πlxyzijk l

261

The association between RAU positive and RHLpositive among women was different than amongmen. The association between RAU positive andgender in RHL positive was different than found inthe group that was RHL negative. The associationbetween RHL positive and gender in RAU positivewas different than found in the group that wasRAU negative. Thus, there exists an association be-tween RAU positive and RHL positive, which dif-fers with gender (non-homogenic odds ratios).

Table 2 shows the odds ratio for RAU in RHLpositive (compared to RHL negative), and for RHLin RAU positive (compared to RAU negative).Among women, the odds for RAU among thosewho were RHL positive was 6.88 times greater thanfor those RHL negative. Among men, the odds forRAU for those RHL positive was 12.37 timesgreater than for those who were RHL negative. Asfor RAU positive, in women, the chance for RHLfor those RAU positive was 6.88 times greater thanfor those RAU negative. In men, the odds for RHLfor those RAU positive was 12.37 times greaterthan for those RAU negative. It should be notedthat the odds ratios for RAU and RHL are identical.

Discussion

The oral mucosa is considered as one of the body’smore important initial lines of defense, acting as it

Table 1. Prevalence of RAU and RHL, in the study sampleaccording to gender

RAU positive RHL positivePopulationsample Total no. n % n %

Women 5698 404 7.1 388 6.8Men 14 991 875 5.8 731 4.9P value ,0.0001 ,0.0001Total 20 689 1279 6.2 1119 5.4

Table 2. Odds ratios for RAU in RHL positive (compared toRHL negative), and for RHL in RAU positive (compared toRAU negative)

Lower Upperconfidence Odds confidence P

limit ratios limit value

RAUWomen 5.35 6.88 8.85 ,0.0001Men 10.41 12.37 14.7 ,0.0001

RHLWomen 5.35 6.88 8.85 ,0.0001Men 10.41 12.37 14.7 ,0.0001

Katz et al.

does as a protective barrier against the invasion ofmicroorganisms and noxious agents (22). Ulcera-tive conditions such as RAL and RHL are commonand are a great nuisance to the affected subjects (1–11). Since the exact etiology of RAU is unknownand there is no available predictable treatmentmodality, the further exploration on its potentialpredisposing factors and pathogenesis is of contin-uing interest.

The total prevalence of RAU and RHL (6.2% and5.4%, respectively) in the present study can be re-garded as within the lower limits of the reportedprevalence in the literature. There may be somecontributing factors to this finding: a) the phrasingof the questions used to elicit subject’s responsemight be responsible for the relatively low values;b) soldiers (as our population) may tend to com-plain less than the civilian population; c) our popu-lation was homogeneous both in age and medicalcondition. Epidemiological studies are based on avariety of methodologies, populations and ages,which may explain some of the differences inprevalence between the studies. The prevalence re-ported in most studies has been based on self-re-port questionnaires. A clinical survey gives theprevalence of RAU and RHL in the study popula-tion, in a real time (6). The prevalence found inthose studies was low. Their disadvantage was thatthey did not allow a complete evaluation of thetotal number of subjects suffering from the studiedconditions. Such a study can help in assessing thenumber of subjects requiring treatment in a definedperiod of time. Another alternative is to evaluatethe short-term history (11). The advantage is thatmost of the individuals with such a history willprobably remember its occurrence and mention itin the study questionnaire. The disadvantage isthat it does not encounter the total lifetime preva-lence. The search for the total lifetime prevalencegives the best estimation of the total prevalence ofa given condition, yet the reliability of the answersis lower since one has to rely on the subject’s ownmemory and correct interpretation of the question-naire. The present study attempted to evaluate thelifetime prevalence in a population whose ageranged between 18 and 21 years. We found a rela-tively low prevalence. While there is a reason tobelieve that the prevalence of RAU and RHL in thisstudy is underestimated, it is also possible thatsome of the available studies have overestimatedthe prevalence.

The total prevalence of both RAU and RHL isalso highly influenced by the characteristics of the

262

examined group. The prevalence of RAU rangesbetween 5% and 60% depending on the studiedpopulation (1–11). A higher prevalence has beenfound in professional persons and those in the up-per socioeconomic classes. The prevalence of her-pes infection varies according to ethnic origin,socioeconomic status and sexual behavior (23–26).In a previous study we have determined, in thesame study group, a low incidence of nearly 60%anti-HSV antibodies (26), compared to 90% in otherstudies (28). The characteristics of the presentstudy population (a sample representing the entireIsraeli population aged 18–21 years, and a relative-ly low incidence of anti-HSV antibodies and recur-rence rate) may also provide an explanation for therelatively low prevalence of RAU and RHL in thepresent study.

Several studies report an existing link betweenRHL and RAU, (2, 7, 21), which is not necessarilyetiological (27). In a study population of soldierswe would expect a high prevalence of RHL basedon stress as a precipitating factor. The relativelylow prevalence of RHL further emphasizes the truerelationship between RHL and RAU. In the presentstudy, men who were RAU positive presented thehighest prevalence of RHL. Ship et al. (2), in amuch smaller sample of 343 medical and dentalstudents and 242 hospital patients, demonstrated a3:2 chance that a person with RAU would alsohave RHL.

RHL represents the reactivation of a latent virus.It is still an enigma why subjects with antibodiespositive to HSV have recurrences. The pathophysi-ology of recurrence is unclear and has been relatedto a focal immune failure (28). Ship et al. (2) showedthat there was no significant difference in the preva-lence of RAU among HSV positive and HSV nega-tive subjects. The relationship was found only in therecurrent group. The emotional factor was also test-ed and was not found to be related. They suggestedcommon features in the pathogenesis of the two syn-dromes. An antigenic stimulus response was theirmost suitable speculation. It can be speculated thatRAU reside in a cascade, beginning with an antigen-ic stimulus (such as herpes labialis), continuing withchanges in the epithelial membrane receptors andending with reactivation of the complement systemleading to epithelial lysis.

It should be noted that high odds ratios may bepartly a function of the low prevalence observed.In this study it appears that most persons who re-sponded positively to the RAU question also re-sponded positively to the RHL question. This

Recurrent oral ulcerations associated with RHL

could indicate some common etiologic or underly-ing immune factor, or it could simply be that re-spondents could not very easily differentiate theintent of the question.

Although an association between RAU and RHLwas found in both men and women, it was strong-er among men (OR 12.37 and 6.88 respectively).The explanation for this is not entirely clear. In aprevious study, no gender predominance wasfound in antibodies for HSV-1 (26). With regard toaphthous ulcerations, most studies did not demon-strate gender predominance. Therefore our presentfinding of higher odds ratios for RAU and RHL inmen compared to women is rather unexpected. Itmight be related to other cofactors such as smokinghabits, using contraceptives, seeking medical care,periodontal health and genetic predisposition.

The strong association between RAU and RHLfound in this study, although substantially greaterthan reported in previous studies, points to theneed for additional research to investigate the pos-sibility of common underlying pathologic pro-cesses.

References

1. Ship II, Morris AL, Durocher RT, Burket L. Recurrentaphthous ulcerations and recurrent herpes labialis in aprofessional school dental population. III. Oral examina-tions. Oral Surg 1960;13:1438–44.

2. Ship II, Brightman VJ, Laster LL. The patient with recur-rent aphthous ulcers and the patient with recurrent her-pes labialis: a study of two population samples. J AmDent Assoc 1967;75:645–54.

3. Ship II. Epidemiologic aspects of recurrent aphthous ul-ceration. Oral Surg 1972;33:400–6.

4. Miller MF, Ship II, Ram C. A retrospective study of theprevalence and incidence of recurrent aphthous ulcers ina professional population, 1958–1971. Oral Surg 1977;43:532–7.

5. Shapiro S, Olson DL, Chelemi SJ. The association be-tween smoking and aphthous ulcers. Oral Surg 1970;30:624–30.

6. Fahmy MS, Recurrent aphthous ulcerations in a mixedArab community. Community Dent Oral Epidemiol1976;4:160–4.

7. Embil JA, Stephens RG, Manuel FR. Prevalence of recur-rent herpes labialis and aphthous ulcers among youngadults on six continents. Can Med Assoc J 1975;113:627–30.

8. Kleinman DV, Swango PA, Pindborg JJ. Epidemiology oforal mucosa in United States schoolchildren: 1986–87.Community Dent Oral Epidemiol 1994;22:243–53.

9. Donatsky O. Epidemiologic study on recurrent aphthousulceration among 512 Danish dental students. Com-munity Dent Oral Epidemiol 1973;1:37–40.

263

10. Axell T. A preliminary report on prevalence of oralmucosal lesions in a Swedish population. CommunityDent Oral Epidemiol 1975;3:143–5.

11. Axell T, Henricsson V. The occurrence of recurrent aph-thous ulcers in an adult Swedish population. Acta Odon-tol Scand 1985;43:121–5.

12. Axell T. A prevalence study of oral mucosal lesions in anadult Swedish population. Odontol Rev 1976;27(Suppl36):1–103.

13. Woo S-B, Sonis ST. Recurrent aphthous ulcers: A reviewof diagnosis and treatment. J Am Dent Assoc 1996;127:1202–13.

14. Miller MF, Garfunkel AA, Ram CA, Ship II. Inheritancepatterns in recurrent aphthous ulcers: twin and pedigreedata. Oral Surg 1980;49:409–12.

15. Pederson A, Hornsleth A. Recurrent aphthous ulcer-ation: A possible clinical manifestation of reactivation ofvaricella zoster or cytomegalovirus infection. J Oral Pa-thol Med 1993;22:64–8.

16. Katz J, Findler M, Marmary Y et al. Measles, Crohn’sdisease and recurrent aphthous stomatitis. Lancet 1996;348:1250–1.

17. Savage NW, Seymour GJ, Kruger BJ. T lymphocyte sub-set changes in recurrent aphthous stomatitis. Oral SurgOral Med Oral Pathol 1985;60:175–81.

18. Savage NW, Seymour GJ, Kruger BJ. Expression of classI and class II major histocomplex antigens on epithelialcells in recurrent aphthous stomatitis. J Oral Pathol Med1986;15:191–5.

19. Savage NW, Mananonda R, Seymour GJ et al. The pro-portion of suppressor-inducer T lymphocytes is reducedin recurrent aphthous stomatitis. J Oral Pathol Med1988;17:293–7.

20. Young TB, Rimm EB, D’alessio DJ. Cross-sectional studyof recurrent herpes labialis. Prevalence and risk factors.Am J Epidemiol 1988;127:612.

21. Ipsen J, Feigl P. Evaluation of effectivity and risk. In: Ip-sen J, Feigl P, eds. Bancroft’s introduction to biostatistics,2nd edn. New York: Harper & Row; 1970. p. 143–52.

22. Squier CA. Barrier function of oral epithelia. In: Macken-zie IC, Squier CA, Dabelsteen E, eds. Oral mucosaldiseases: biology etiology and therapy. Iowa City: Laege-foreningens Forlag; 1987. p. 7–9.

23. Burnet FM, Williams SW . Herpes simplex – a new pointof view. Med J Aust 1939;1:637–43.

24. Greenberg MS , Brightman VJ , Ship II. Clinical and labo-ratory differentiation of recurrent intra-oral herpes sim-plex following fever. J Dent Res 1969;48:385–91.

25. Ades AE, Peckham CS, Dale GE, et al. Prevalence of anti-bodies to herpes simplex virus types 1, and 2 in pregnantwomen and estimated rates of infection. Epidemiol Com-munity Health 1989;43:53–60 .

26. Katz J, Gdalevitz M, Ashkenazi I, Shemer J. Seculartrends in the epidemiology of herpes simplex virus type-1 among young adult population. Infection 1999;27:295–7.

27. Lennette EH, Magoffin RL. Virologic and immunologicaspects of major oral ulcers. J Dent Assoc 1973;87:1055–73.

28. Regezi JA, Sciubba J. Vesiculo-bullous diseases. In: Re-gezi JA, Sciubba J, eds. Oral pathology, clinical-patho-logic considerations, 2nd edn. Philadelphia: WB Saun-ders; 1993.