|Home | About | Journals | Submit | Contact Us | Français|
The aim of this review is to evaluate the principal clinical and conventional radiographic features of non-syndromic keratocystic odontogenic tumour (KCOT) by systematic review (SR), and to compare the frequencies between four global groups.
The databases searched were the PubMed interface of Medline and LILACS. Only those reports of KCOTs that occurred in a series of consecutive cases, in the reporting authors' caseload, were considered.
51 reports, of 49 series of cases, were included in the SR. 11 SR-included series were in languages other than English. KCOTs affected males more frequently and were three times more prevalent in the mandible. Although the mean age at first presentation was 37 years, the largest proportion of cases first presented in the third decade. The main symptom was swelling. Over a third were found incidentally. Nearly two-thirds displayed buccolingual expansion. Over a quarter of cases recurred. Only a quarter of all SR-included reported series of cases included details of at least one radiological feature. The East Asian global group presented significantly as well-defined, even corticated, multilocular radiolucencies with buccolingual expansion. The KCOTs affecting the Western global group significantly displayed an association with unerupted teeth.
Long-term follow-up of large series that would have revealed detailed radiographic description and long-term outcomes of non-syndromic KCOT was lacking.
The term odontogenic keratocyst was first used by Philipsen in 1956.1 This lesion was recently renamed by him as keratocystic odontogenic tumour (KCOT) and reclassified as an odontogenic neoplasm in the World Health Organization's 2005 edition of its histological classification of odontogenic tumours.2 According to this edition the KCOT has been defined as “A benign uni- or multicystic intraosseous tumour of odontogenic origin, with a characteristic lining of parakeratinized stratified squamous epithelium and potentially aggressive, infiltrative behaviour. It may be solitary or multiple. The latter is usually one of the stigmata of the inherited naevoid basal cell carcinoma syndrome (NBCCS).”2 To emphasise the essential parakeratotic feature of this new “tumour”, it adds “Cystic jaw lesions that are lined by orthokeratinizing epithelium do not form part of the spectrum of a … KCOT.”2
Although Blanas et al3 have performed a systematic review (SR) on the odontogenic keratocyst all their SR-included reports did not distinguish between the above KCOT and the orthokeratotic type, which is now recognized as an entirely separate lesion, the orthokeratinzed odontogenic cyst (OOC).4
Sackett et al5 defined an SR as a summary of the medical literature that uses explicit methods to search systematically, appraise critically and synthesize the world literature on a specific issue. This means that the SR, like any other form of primary research, will have a Materials and Methods, and a Results section.6
The SR has generally been applied to treatment and drug trials, but has also become a powerful tool when adapted to the clinical and radiological presentations of important oral and maxillofacial lesions.7
An important feature that helps to distinguish the KCOT is stated by White and Pharoah.8 KCOTs exhibit a “propensity to grow along the internal aspect of the jaws, causing minimal expansion” (see their Figure 21–15). Additionally, KCOTs associated with NBCCS occur earlier and exhibit a greater tendency to recur than non-syndromic KCOTs.8
The principal aim was to include as many reports or pertinent parts of those reports as possible and to evaluate the principal clinical and conventional radiographic features of non-syndromic KCOT by SR.
The primary research question for this SR is “Do other clinical and conventional radiographic features improve diagnosis of non-syndromic KCOT compared with no or minimal buccolingual expansion?” This follows the four-part format required for the SR's research question set out by MacDonald-Jankowski and Dozier.6 In order to include as many reports as possible a wide search of the literature was made, including non-English reports. In addition to a Medical Subject Heading (MeSH) search a “free-text” search was included supported both by hand-searching of journals, which are the natural destinations for reports on oral and maxillofacial lesions, and by reference-harvesting of those reports identified by database searches and hand-searching.
To assist in answering the above research question the SR-included reports will be divided into four global groups: Western, sub-Saharan African, East Asian and Latin American, broadly reflecting ethnic origin.
The approach follows the SR procedure set out in earlier SRs for other oral and maxillofacial lesions. The research question is described above and the search strategy, the strategy for sifting the literature and the interpretation of the data retrieved are set out below.
The search was based on the research question. The databases searched were the PubMed interface of MEDLINE (National Library of Medicine) and LILACS (Literatura Latino Americana e do Caribe em Ciências da Saúde) by BIREME (Latin American and Caribbean Center on Health Sciences Information). LILACS provided access to important Hispanic publications largely not indexed by MEDLINE and has been discussed in an SR on focal osseous dysplasia (FocOD).9
The two MeSH terms were “odontogenic cyst” and “odontogenic tumour”. Odontogenic cyst was defined by MEDLINE as “cysts found in the jaws and arising from epithelium involved in tooth formation”. This MeSH, first introduced in 1965, includes terms that are no longer used or are unknown to current dental and oral and maxillofacial practice. The MeSH definition contained keratocyst and keratocysts as two separate entry terms (synonyms). Odontogenic tumour was defined in 1980 as “neoplasms produced from tooth-forming tissues”. Its entry terms are odontogenic tumour; tumour, odontogenic; tumours, odontogenic; neoplasms, dental tissue; dental tissue neoplasms; dental tissue neoplasm; neoplasm, dental tissue; tissue neoplasm, dental; and tissue neoplasms, dental.
The free-text terms used were “odontogenic keratocyst” and “keratocystic odontogenic tumour”. These terms were also used to search LILACS. To include as many reports as possible the emphasis was placed on recall rather than precision, bearing in mind that the MeSH for dentistry and radiology is generally inadequate and that free-text searching may not hit the relevant article if the free-text term used was not included in the title or abstract. This strategy was further augmented by reference to the bibliographies (or citation lists) of all reports identified by the databases (reference harvesting) or hand-searching of journals listed in Table 1 of the SR on FocOD.9 Both database searches and hand-searches were last conducted on 2 September 2009.
The decision to include a report was generally made by reading the title and the abstract. As abstracts were used infrequently before 1979 it was anticipated that a call for the full paper, to determine whether it should be considered by the selection criteria, would be made more frequently for reports published before that year.
There were three inclusion criteria (1 to 3) and three exclusion criteria (A to C) for the SR. Each report passed through these criteria in strict sequence. Although a report may be excluded by more than one criterion, only the first criterion to exclude a particular report was entered in the Appendix. For the sake of brevity only those reports which cannot be readily included by reference to their title or abstract will be discussed and cited.
Information included in the SR was generally reduced to numbers and tabulated. This took account of the number of cases that passed the selection criteria for inclusion and, therefore, may differ from the numbers available in the original report. Information not given (ING) was applied wherever information on a particular feature could not be determined (either expressed or implied) from the original text. Inadequate information given (IIG) was applied to features in a partially included report, which had been compromised and could not be included in the SR. IIG was entered against those features which the reporters had not adequately quantified.
The aim was not only to include as many reports as possible, but also as many features of those reports as possible, and to include as many cases of each report as the selection criteria would allow. This required a separate strategy. Although the terms “exclusion”, “deletion” and “deduction” are normally synonyms, for the purposes of this SR they are specifically defined as follows: exclusion, the non-inclusion of a report in the SR and is used regardless of whether it is in relation to either inclusion or exclusion selection criteria; deletion, the removal of cases within a report that are not consistent with one or more of the selection criteria and have been reported in sufficient detail to permit their identification and removal from the report, therefore allowing the rest of the report to be included in the SR; deduction, is applied to those reports where deletion is not possible and the number of non-deletable cases exceeds 10% of the report (thus exceeds the “less than 10% allowance” for such non-conforming and non-deletable cases). For deduction to be permitted the report must include, for at least one feature, wholly inclusive words, such as “all” or “every” qualifying patient or case. “Almost all/every” or “the overwhelming majority”; for example, “…. of the reported lesions were well-defined” were construed to be over 90% (or less than the 10% allowance) and as a result were considered to approximate to 100% and considered to be wholly inclusive. On the other hand studies reporting “most” or “majority” of the cases were well-defined were not admitted (for an example refer to the SR on fibrous dysplasia),7 because these and similar phrases are construed to represent 51–89% and, therefore, well outside the 10% allowance. The strategy for minimizing the impact of non-conforming reports on the SR was exclusion, deletion and deduction (including the less than 10% allowance rule).
Criterion 1. Consistency with the WHO classification. The lesions had to be, at least, consistent with the histopathology established by the WHO's 2005 edition of its classification of odontogenic tumours.2 Although this described only the parakeratotic type as KCOT, it excluded the orthokeratotic type, or OOC, considering it a separate lesion. To include only data pertaining to KCOT it excluded reports if:
It should be noted that although the mixed parakeratotic and orthokeratotic cases were not considered by the WHO's 2005 edition,2 they were not expressly excluded. It is reasonable to surmise the very presence of a parakeratotic element would merit their inclusion. Nevertheless, adequately identified mixed cases were removed. Their removal or retention is indicated in the comments column of Table 1.
Criterion 2. Non-syndromic cases. KCOT is a feature of NBCCS also known as Gorlin-Goltz syndrome.2 Although there is no doubt that these lesions are KCOTs, they not only present earlier in life, but are also likely to recur after surgery.8 To enhance our understanding of the more common solitary form the following strategy was employed, it excludes reports if:
Criterion 3. A complete collection of KCOT cases. The study should represent a complete collection of cases of KCOT, arising within a particular community, occurring in the reporter's caseload. Reports that were merely a selection of cases, such as case reports and those studies primarily concerned with specific investigations or a discrete age group, such as children or a particular jaw, were excluded.
Criterion A. Excludes data already reported and included in the SR. It prevented double reporting of the same clinical cases, by excluding those reports in which the data had already been reported and included in the SR, either by the same or different authors, unless the degree of overlap did not exceed 50% and there was at least one statistically significant different feature between them.
Criterion B. Excludes cases that recurred after primary treatment given elsewhere and/or earlier than the range in years of the study. It excludes reports if:
Criterion C. Excludes referred cases. It reinforces Criterion 3 by minimizing dilution of the data arising primarily within a specific community. It therefore excludes reports that include referred cases from outside that community. This is because they may possess unusual features that may skew the profile of KCOT within that community, which would, in turn, skew the SR. To include only data pertaining to the jaws, it excluded reports if:
Definitions of parameters, such as number of years a report covered, number of KCOTs per year, division of each jaw into sextants and radiologically apparent boundaries between the basal and alveolar processes for each jaw, are the same as the recent SR on FocOD.9
The term “radiolucency” could be implied from the reference to the radiological shape of lesions as unilocular or multilocular and these implied that the lesion was a radiolucency.
In order to determine deeper patterns within the SR, the reports were divided into four global groups broadly based on ethnicity. These were East Asian (predominantly represented in this SR by Chinese and Japanese), sub-Saharan African (predominantly black African, including Jamaica), Western/Caucasian (North American and European (including Turkey), Middle Eastern, North African and Indian) and Latin American (including Cuba). Although the Western group was predominantly white (Caucasian, classically European and Middle Eastern) it contained significant non-white minorities, particularly from sub-Saharan Africa. The population of the USA was at the last census 69.1% white.10 Reports from the Indian subcontinent are included in the Western/Caucasian group, because 95% of Indians are Caucasian (Indo-Aryans and Dravidians). Although these four global groups are cartographically represented by four almost discrete regions, they are not primarily regional, because variable socio-economic and other ethno-cultural factors also play important roles that affect the availability and provision of diagnosis and treatment; for example, the South Asian nations, including India, although largely Caucasian, are still developing their economies, along with many of those in sub-Saharan Africa. Although Africa itself is divided between a Caucasian north and a substantially black sub-Saharan south, it is the latter which constitutes the bulk of the population of the African continent and the African diaspora (Jamaica is 90% of sub-Saharan African origin). Previous SRs have shown that nearly all SR-included African reports are from south of the Sahara.11–16 The important point of this global distribution is to determine the number, size and quality of the SR-included reports to determine which communities are well reported and which are underreported.
Significant differences in frequencies were analysed using χ2 test with P < 0.05. Significant differences in age were analysed using Student's t-test with P < 0.05.
Many of the reports were automatically rejected because it was clear from reading the title or abstract that they were single case reports or review articles. Figure 1 outlines the process and disposal of the reports considered for a call of the full paper. Figure 1 includes all reports except reference 67 because the present author was a co-author, and the paper was already known to the present author/reviewer prior to conducting the SR.
The 49 SR-included consecutive case series contained within the 51 reports17–67 are set out in Tables 1 – 4 and the 102 consecutive case series in 104 reports excluded under specific exclusion criteria are set out in the Appendix. It should be noted that, in order to be as clear as possible, when a case series rather than a specific report is intended then case series will be used. Although the source of the literature (MEDLINE, LILACS, reference-harvested or hand-searched) did not differ between the reports excluded or included, reports identified solely by LILACS and published before 1990 were significantly more likely to be excluded than those published on or after 1990 (Table 2). The Hong Kong report67 was excluded from the analysis of the source in Table 2 because the present author was one of its co-authors.
The language of publication, when grouped into English, other European and East Asian languages, had no significant effect on inclusion.
Proportionally more Western/Caucasian reports were excluded from the SR than included; however, this was not significant. Western reports were significantly less likely to be excluded after 1990 whereas those from Latin America were significantly more likely to be excluded after 1990 (see Table 2).
82% of reports were first excluded under criterion 1. There was no significant difference regarding reports excluded under criterion 1 published prior to 1990 and those published on or after 1990 (χ2 = 3.59, 1 degree of freedom (df), P > 0.05).
Although not an exclusion criterion, the exclusion of the mixed (parakeratotic and orthokeratotic) cases occurred in 13 reports without further deduction of the details. This was largely owing to all the necessary deduction having been already undertaken to exclude the purely orthokeratotic (or OOC) cases, which was a fundamental exclusion criterion (criterion 1).
49 series of consecutive cases, in 51 reports (2 communities were each represented by 2 reports), were included in the SR (Table 1).17–67 The clinical features decades, age distribution in decades at first presentation and radiological details extracted from each of the series are shown in Tables 1, ,33 and and4,4, respectively.
The MeSH “odontogenic tumour” and/or the MEDLINE textword “keratocystic odontogenic tumour” were more effective at recalling KCOT reports from 2007 onwards, whereas the MeSH “odontogenic cyst” and MEDLINE textword “odontogenic keratocyst” were, until 2007, extremely effective in recalling KCOT reports; however, they displayed a marked reduction in effectiveness after that year.
Although all four global groups were represented, the sub-Saharan African group was only represented in three reports. Just under half of the reports were derived from the Western group. Reports from the East Asian group were derived from a wide range of nations. The mean range in years covered by the East Asian reports was significantly larger than that of the Western reports.
The number of KCOTs per year is shown in Table 1a; however, this was not determinable for four reports. The number of KCOTs per year fell from 7.68 (SD 6.57) KCOTs per year in the 12 series of cases reported prior to 1990, to 5.17 (SD 5.13) KCOTs per year in the 33 series of cases reported after and including 1990; this was not significant (t = 1.20, 43 df, P > 0.05).
Although the mean number of cases ranged widely between the global groups (Table 1b), this was not significant.
Most case series originally included the patients' ethnic origin, sex, age, site affected and presenting clinical features; however, these could not be included in the SR. The extent of the deletions is evident from the distribution of IIG in Table 1a. Six SR-included reports exhibited widespread deletion.
Only two reports, one sub-Saharan African18 and the other East Asian67, indicated the duration of the patients' pre-existing awareness of their lesions prior to first presentation. Although the period between first awareness of the lesion and first presentation for the east Asian report67 was shorter than that for the sub-Saharan African, report it was not significant; t = 0.87: 27df: P < 0.05.
Males predominated in the East Asian, Latin American and Western global groups, but this was significantly lower for the Latin American group (Table 1c). The mean age at first presentation was higher for the Western global group than for the other three global groups; this was not significant (Table 1b). Table 3 is a comparison between 17 reports of the distribution according to age in decades. KCOTs first present most frequently in the third decade overall, and for the East Asian, Latin American and Western global groups. Females predominate in the first decade, while males predominated slightly from the second to the seventh decades.
Swelling and pain at first presentation were significantly more frequent in East Asians, whereas KCOT discovered as an incidental finding was more frequent in the Latin American reports (Table 1c). Numbness was the greatest proportionally for the sole small Latin American report.17 Cases presenting with numbness were also quantified in three East Asian reports.41,47,67
Location of the lesions by quadrant or sextant was specified in only 15 case series for the mandible and 13 for the maxilla, whereas location by jaw was specified in 25 (Table 1c). The mandible was affected twice as frequently as the maxilla and significantly more for the East Asian (77%) and Latin American (73%) compared with the Western global group (63%). There was no difference between the groups for either the mandibular or maxillary sextants; all four global groups display a predilection for the posterior sextants of both jaws.
Table 4a shows the radiological features and contains 12 SR-included reports covering four global groups; the sub-Saharan African group was represented by only two small reports. The general paucity of radiological detail is illustrated by the frequent use of ING. There was a significant predilection for a multilocular shape (Table 4b) in the East Asian global group (36.7%), in contrast with the Latin American (24.2 %) and Western (20.2 %) global groups.
64% of KCOTs were well-defined. One East Asian report (Hong Kong Chinese)67 displayed significantly better marginal definition, cortication and buccolingual expansion than those from a Western report.32 Cortication occurred in 60% cases synthesized from these two reports. Only one report67 revealed downward displacement or erosion of the lower border of the mandible (71%), involvement of the maxillary antrum (100%) and tooth displacement (32%). Root resorption occurred in 23% of a synthesis of two reports.32,67 A significant association with unerupted teeth occurred more frequently in the Western global group.
The overall recurrence rate was 28%. The recurrence rate was significantly higher in the Western global group (33%) compared with the East Asian global group (24%), which also synthesized a similar number of cases (Table 1c).
The number of the orthokeratotic type (or OOC), when compared with the number of the parakeratotic type (now KCOT) accounted for 10% of the combined OOCs and KCOTs overall. The proportion of OOCs was significantly greater for the Western global group than for the East Asian and Latin American groups, which also included more than one case series (Table 1c). For the Western, East Asian, Latin American and sub-Saharan African global groups the proportion was 11%, 8%, 8% and 7%, respectively.
13 case series permitted the exclusion of mixed cases and 3 did not. The Latin American global group contained proportionately significantly fewer excludable mixed cases (2%) than the other global groups (East Asians, 9%; sub-Saharan Africans, 11%; and Western, 8% (Table 1c)). The overall percentage of mixed cases excluded and included in the SR were 7% and 5% respectively.
The percentage of syndromic cases for the East Asian, Latin American and Western global groups, containing more than one case series, were 6%, 11% and 6% respectively; the higher predilection for syndromic cases in the Latin American global group was significant (Table 1c). The overall mean was 7%.
According to Madras and Lapoint63 three factors led to the recharacterization of the keratocyst as KCOT. The KCOT exhibits locally destructive and highly recurrent behaviour; the histopathology of the KCOT reveals budding of the basal layer into the connective tissue and frequent mitotic figures; and, finally, the KCOTs are associated with an inactivation of PCHT, the tumour suppressor gene. The presence of a genetic component suggests that the patient's ethnic origin, which is “family history” written large, may have a role to play. The significant differences observed between the global groups in this SR suggest that ethnic origin of the KCOT patient is important.
The effectiveness of MEDLINE searches using MeSH and freetexting have reflected both the change in nomenclature of the parakeratinized type of odontogenic keratocyst to KCOT and its reclassification from cyst to neoplasm. Therefore, the most effective MeSH for searching for KCOT is now “odontogenic tumour” and the most effective textword is “keratocystic odontogenic tumour”. Latin American reports, identified by LILACS, were published significantly more frequently after 1990. This is due, most likely, to the recent origin of this database and its widespread application to the Latin American literature.
The significantly greater range in years of the East Asian reports compared with the Western reports may reflect their earlier commencement of record keeping of this particular lesion.
The paucity of reports on KCOT from the sub-Saharan African global group stood in marked contrast to the SR of the ameloblastoma, another odontogenic lesion.11,12 There were only three small reports: Nigerian,18 South African (of undisclosed ethnic origin)21 and Jamaican.59 Mosadomi18 reported that only 3 KCOTs presented in his Nigerian community over 5 years, whereas 19 ameloblastomas presented during the same time–period. Furthermore, Rachanis and Shear68 remarked on the lower frequency of keratocysts (this report was excluded under criterion 1 because the type of keratinization was not identified) affecting the South African black community in comparison with the white South Africans, the reverse was true for ameloblastoma. This would indicate that KCOTs occurred less frequently than ameloblastomas in at least two sub-Saharan African communities. This may explain why the three KCOTs in the Mosadomi18 report had been provisionally diagnosed as ameloblastomas.
It was clear from the significant proportion of OOC (or orthokeratotic type of the former odontogenic keratocyst) in the SR-included reports that they could have played a major role in the variability of outcomes in many reports, which did not distinguish between the types of keratinization. Such reports accounted for the vast majority of the SR-excluded reports. This reinforces the prior decision to deduct completely all features that would have also included OOC.
The removal of OOC cases, and those which are part of the NBCCS, was necessary because they are completely different lesions. The additional removal, whenever possible, of the mixed cases was desirable because they appear to be intermediate in behaviour between the KCOT and OOC. Although Crowley et al36 compared the recurrence rates of KCOT (42.6%), OOC (2.2%) and mixed cases (14.3%), the mixed cases were few in number. Nevertheless, the separate evaluation of the clinical, radiological presentations and treatment outcomes of these mixed cases should be considered in future reports.
Although the frequency of mixed cases was highest in the sole sub-Saharan African report21 and lowest in the sole Latin American report44,45 detailing this feature, this phenomenon may not survive the addition of subsequent additional reports to these global groups.
There was no significant difference between reports included and excluded from the SR on the basis of membership of a particular global group, language of publication and source. This indicated that any bias against the inclusion of reports on the basis of global group, language of publication and source is unlikely.
Although the East Asian global group presented with significantly more pain and swelling, this was not reflected in significantly more cases presenting with a purulent discharge at first presentation. This may be due to two possibilities: the reporters of this global group did not report such a discharge or KCOT in East Asians is more likely to provoke swelling and pain. The Hong Kong Chinese presented their KCOTs at a young age.67 Only Iranians in the report by Habibi et al56 were younger. This was supported by a South Korean report by Myoung et al,47 which reported a westernized East Asian community whose KCOTs presented at a young mean age of 31 years. Only 5% of the cases were discovered as incidental findings. These may suggest that in certain East Asian communities KCOTs first present with symptoms. The cause of such symptoms needs to be considered; it is possibly not due to size or at least size alone. The largest report by Myoung et al47 reported that each lesion affected a discrete sextant, either an anterior or a posterior sextant, but not both; therefore, the lesions were not unduly large. Nevertheless, KCOTs in one East Asian report (Hong Kong)67 were significantly associated with buccolingual expansion when compared with one small Latin American report.17
KCOT in Latin Americans presented significantly more frequently with a discharge and numbness. They presented more frequently in the mandible and in its posterior sextant, where purulent infection of the KCOT can affect the inferior dental nerve and cause numbness.
Philipsen recently stated that root resorption was a rarity in KCOT.2 Two SR-included reports, one each from the Western32 and East Asian67 global groups, reported root resorption in 8% and 41% of the cases, respectively. Therefore, although root resorption is not common, it may not be rare in certain communities.
Although all global groups displayed a predilection for the mandible and for the posterior sextants of both jaws, the Western global group's significant association with unerupted third molars was unexpected, particularly as the largest of the two Western reports was an American report.36 An explanation is that the KCOTs in this report may not have been affected by the purportedly high incidence of routine prophylactic removal of third molars in this community.
All global groups displayed recurrence, which on average occurred in one out of every four KCOTs removed; however, this may be an underestimate as follow-up may have not been carried out for a long enough period. Long-term follow-up is required for KCOT as Stoelinga48 found that 5 cases recurred 6 to 25 years after enucleation. Recurrence in recent reports may be due to the type of treatment, for example decompression (marsupialization) may result in recurrence within 2 years. Pogrel69 now suggests that decompression should be supplemented, once the cavity has been sufficiently reduced in size, by aggressive curettage and treatment with liquid nitrogen. Madras and LaPoint63 state that the “WHO's reclassification … underscores the aggressive nature of the lesion and should motivate clinicians to manage the disease in a correspondingly aggressive manner.”
I wish to express my gratitude to Dr D Ruse of the Faculty of Dentistry and H Lin of the Faculty of Arts (Asian Studies) at the University of British Columbia, for their assistance with the Hungarian and Chinese texts, respectively.
|Selection criterion||Report First authors surname and date of publication||Global group||Language of publication||Database|
|3||Avelar et al 200870||L||English||Me|
|3||Cavalcante et al 200871||L||English||Me|
|3||Gonzlez Moles et al 200872||W||English||Me|
|3||Kuroyanagi et al 200873||E||English||Me|
|1||Tortorici et al 200874||W||English||Me|
|1||Ochsensius et al 200775||W||English||Me|
|1||Jones et al 200676||W||English||Me|
|1||Maurette et al 200677||W||English||Me|
|1||Varinauskas et al 200678||W||English||Me|
|1||Bornstein et al 200579||W||French, German||Me|
|1||Camisasca et al 200580||L||Portuguese||Li|
|1||Kim et al 200581||E||English||Me|
|3||Morgan et al 200582||W||English||Me|
|1||Bataineh et al 200483||W||English||Me|
|1||Koseoglu et al 200484||W||English||Me|
|1||Pippi Vitolo 200485||W||English||Me|
|3||Pogrel Jordan 200469||W||English||Me|
|1||Vallejos Briend 200486||W||Spanish||Li|
|1||Ali Baughmann 200387||W||English||Me|
|1||de Amorim et al 200388||L||Portuguese||Li|
|A||Li et al 200389||E||Chinese||Me|
|1||Barreto et al 200290||L||English||Me|
|1||da Silva et al 200291||L||English||Me|
|1||Nakamura et al 200292||E||English||Me|
|1||Tsukamoto et al 200293||E||English||Me|
|1||Zhao et al 200294||E||English||Me|
|1||Giardina et al 200195||W||English||Me|
|B||Schmidt Pogrel 200196||W||English||R|
|1||August et al 200097||W||English||Me|
|1||Cabral et al 200098||L||Portuguese||Li|
|1||Chiapasco et al 200099||W||English||Me|
|3||de Paula et al 2000100||L||English||Me|
|A||Lam Chan 2000101||E||English||Me|
|1||LedesmaMontes et al 2000102||L||English||Me|
|A||Ngeow et al 2000103||E||English||Me|
|1||Oda et al 2000104||W||English||Me|
|1||de Quadros Oliveira Calvet2000105||L||Portuguese||Li|
|1||Soost et al 1999106||W||German||Me|
|1||Arotiba et al 1998108||A||English||Me|
|1||Meara et al 1998109||W||English||Me|
|1||Dammer et al 1997110||W||English||Me|
|3||Marker et al 1996111||W||English||Me|
|1||Minami et al 1996112||E||English||Me|
|3||Nakamura et al 1995113||E||English||Me|
|1||Daley et al 1994115||W||English||Me|
|1||Das et al 1994116||W||English||Me|
|2||Berrone et al 1994117||W||Italian||Me|
|1||Kreidler et al 1993118||W||English||Me|
|1||Utsumi et al 1990120||E||Japanese||Me|
|1||Gerlach et al 1989121||W||German||Me|
|3||Dominguez Keszler 1988122||L||English||Me|
|B||Jensen et al 1988123||W||English||Me|
|1||Kndell Wiberg 1988124||W||English||Me|
|A||Stoelinga Bronkhorst 1988125||W||English||Me|
|1||Forssell et al 1988126||W||English||Me|
|1||Kalusokoma et al 1987127||W||French||Me|
|1||Partridge Towers 1987128||W||English||Me|
|1||Weir et al 1987129||W||English||R|
|1||Woolgar et al 1987130||W||English||Me|
|1||Gspr et al 1986131||W||Hungarian||Me|
|A||Siar et al 1988132||E||English||Me|
|1||Ewers Hrle 1985134||W||German||Me|
|1||Hoffmeister Harle 1985135||W||German||Me|
|1||Lai Chen 1985136||E||Chinese||Me|
|1||Niemeyer et al 1985137||W||German||Me|
|1||ReffEberwein et al 1985138||W||German||Me|
|1||Spitzer Steinhusern 1985140||W||German||Me|
|1||Takita et al 1985141||E||Japanese||Me|
|3||Weerheijm van der Waal 1985142||W||Dutch||Me|
|1||Zachariades et al 1985143||W||English||Me|
|1||Arafat Lunin 1984144||W||English||Me|
|1||Chuong et al 1982146||W||English||Me|
|1||Sakamoto et al 1982147||E||English||Me|
|1||Voorsmit et al 1981149||W||English||Me|
|1||Buffetaud et al 1980150||W||French||Me|
|1||Jain Kherdekar 1980151||W||English||Me|
|1||Lechien et al 1980152||W||French||Me|
|1||Martinez et al 1980154||W||French||Me|
|1||Vedtofte Praetorius 1979155||W||English||Me|
|1||Hodgkinson et al 1978156||W||English||Me|
|1||Rachanis and Shear 197868||W||English||Me|
|1||Smith Shear 1978157||W||English||Me|
|1||Eversole et al 1975161||W||English||Me|
|A||Mosadomi et al 1975162||A||English||Me|
|1||Forssell et al 1974163||W||English||Me|
|1||Radden Reade 1973165||W||English||Me|
|1||Calonius et al 1972166||W||English||Me|
|1||Donoff et al 1972167||W||English||Me|
Global group A, subSaharan African L, Latin American E, East Asian W, Western Me, MEDLINE Database Li, LILACS R, reference harvesting H, handsearching.
Criterion 1. Giardina et al report95 was excluded because they did not disclose the type of keratinization. Nevertheless, if they had been included they would have been excluded later under criterion 3.
Criterion 3. It is clear that de Paula et al 10 uninflamed parakeratotic cases must have been selected.100
Criterion A. Although Avelar et al 2008 report70 was more recent, not only did it substantially report the detail reported by Antunes et al earlier report,53but the latter was more detailed and therefore retained as a SR-included report, the former was excluded.