Oral Health in Children and Adolescents With Juvenile Idiopathic Arthritis

A Systematic Review and Meta-analysis

Marit S. Skeie; Elisabeth G. Gil; Lena Cetrelli; Annika Rosén; Johannes Fischer; Anne Nordrehaug Åstrøm; Keijo Luukko; Xieqi Shi; Astrid J. Feuerherm; Abhijit Sen; Paula Frid; Marite Rygg; Athanasia Bletsa


BMC Oral Health. 2019;19(285) 

In This Article


Nineteen articles met the inclusion criteria, ten from Europe and nine from countries outside Europe with Brazil in the lead, see Table 1. The age range of children and adolescents with JIA was from two to four years in two studies[22,31] and up to 20 years of age in one study.[30] Altogether, the included articles covered topics such as dental caries, oral hygiene (dental plaque and calculus accumulation), periodontal disease (gingivitis included), enamel defects, tooth calcification (dental maturation) disorders, TMJ arthritis, TMJ involvement, TMD, oral ulcerations, and OHRQoL. Beyond these topics, information about inflammatory mediator measurements in blood samples and gingival crevicular fluid, was reported.

Eighteen studies were cross-sectional in nature, only a study by Miranda et al.[32,33] had a prospective cohort study design. At baseline, adolescents with and without JIA were examined for clinical and immunological variables of periodontal inflammation, and two years later a subgroup, eighteen adolescents with JIA and fourteen without JIA, were re-examined. Another study of Lehtinen et al.,[24] distributed coded radiographs at random between different examination sessions, so the only examiner was blinded for the information of whether the radiographs belonged to participants with JIA or healthy controls.

All included studies reported age of children and adolescents with JIA and of those without JIA. However, the degree of matching varied. Although no study of true case-control design was included in the review, two studies showed controls matched for age, gender and ethnicity.[22,39] Another characteristic of included studies was a distinct variation of sample sizes. In some studies, the sample size was too low to justify any result evidence. Beforehand sample size calculation was uncommon, as only one article[29] described this. In most studies the number of examiners was low, usually one. With some exceptions,[22,24,31,33] no description of calibration of examiners or reliability values were included. Bitewing radiographs were reported by only two research groups.[25,26,32]

Dental Caries

Eight of the included articles described dental caries, but with divergent results. Both Ahmed et al. and Welbury et al.[22,31] documented that a significantly larger proportion of children with JIA had untreated caries compared with healthy peers. Welbury et al.[31] also documented that individuals with JIA, had a higher burden of caries than individuals without JIA; among children, more primary teeth decayed, filled or extracted and among adolescents, predominantly more dental decay (D: decayed component in the DMFT). In contrast, the study of Santos et al.[29] revealed caries in primary teeth to be more frequent among healthy children than among children with JIA. Five articles[23,25,27,30,39] did not show any significant difference between the children and adolescents with and without JIA when subgroups were not included. The way of reporting caries varied from untreated caries, dmft, DMFT, D to caries prevalence of affected individuals. The use of diagnostic tools also varied. Both the World health Organization (WHO) criteria[45] and the British Association for the Study of Community Dentistry (BASCD) standardized criteria[80] were used, while some studies did not report the caries diagnostic tool used. Only one study by Leksell et al.,[25] reported enamel caries.

Quantitative Synthesis

Three cross-sectional studies (three publications) were included in the analysis to evaluate the association between caries in primary teeth and JIA (71 children with JIA and 141 total participants). We observed no difference in summary mean dmft scores between JIA and those who did not experience JIA (− 1.16, 95%CI, − 3.02–0.71, I2 = 87.9%, p heterogenity = < 0.0001) (Figure 1).

Figure 1.

Mean differences in dmft indices in adolescents and children with Juvenile Idiopathic Arthritis (JIA) compared with those who did not experience JIA

Six cross-sectional studies (three same publications as used above as they comprised of data from both primary- and of permanent teeth, and three other publications) were included in the analyses to evaluate the association between caries in permanent teeth and JIA (162 children and adolescents with JIA and 320 total participants). We observed no difference in summary mean DMFT score between children and adolescents with JIA and those who did not experience JIA (− 0.08, 95% CI, − 0.42 to 0.26, I2 statistic = 0.0%, %, p heterogenity = 0.95) (Figure 2).

Figure 2.

Mean differences in DMFT indices in adolescents and children with Juvenile Idiopathic Arthritis (JIA) compared with those who did not experience JIA

No evidence of publication bias with Egger's test (P dmft = 0.27, P DMFT = 0.78) or with Begg's test was found (P dmft = 0.98, P DMFT = 0.45) (Additional file 6: Figure. S2 and Additional file 7: Figure. S3). However, because of the small number of studies and small sample size of included studies, the results from these formal tests should not be inferred with great reliability.

Figure S2.

Funnel plot for assessment of bias in the mean difference of dmft of primary dentition studies between children with JIA and controls (n = 3 studies)

Figure S3.

Funnel plot for assessment of bias in the mean difference of DMFT score of permanent dentition studies between children and adolescents with JIA and controls (n = 6 studies)

Plaque, Gingivitis and Periodontitis

The oral health descriptors most often reported were dental plaque and signs of periodontal inflammation (gingival bleeding and bleeding on probing, probing depth ≥ 2 mm, clinical attachment loss, pocket depths etc.). There were studies focusing on oral hygiene and dental plaque[23,25,28,31] showing a statistically higher Plaque Index (PI) or Simplified Oral Hygiene Index (OHI-S) in the JIA group compared with those without JIA. Other studies[22,26,27,29,30,39] did not find this association. Additionally, Leksell et al. found calculus to be significantly more prevalent in individuals with JIA compared with those who did not experience JIA.[25] Many articles also documented poorer periodontal status among children and adolescents with JIA; more gingival inflammation and gingival bleeding,[22,29,31] bleeding upon probing,[25] deeper probing depth[25,26] and periodontal attachment loss.[26,28] However, not all articles documented differences in periodontal status when comparing individuals with JIA with healthy counterparts.[23,30,39]

Developmental Enamel Defects and Ulcers

The only study reporting developmental enamel defect,[22] found the condition more frequent among children with JIA than among healthy peers, but the sample size was too small to draw any reliable conclusion. Another study focused on oral ulceration[25] and found five out of forty-one children with JIA to be affected, but only one out of forty-one children in the group without JIA.

Dental Maturation

Two of the included studies investigated the status of dental maturation and found divergent results. By examining orthopantomograms (OPG), Lehtinen A et al.[24] in 2000 documented more advanced dental development in children with juvenile rheumatoid arthritis (JRA) compared with healthy peers, while Ley et al.[40] nine years later assessed dental maturity in children and adolescents with JIA to be within the norms of healthy peers.


TMJ arthritis (active inflammation of the joint), TMJ involvement (osteoarthritis or growth disturbances as a result of TMJ arthritis)[81] and TMD were coherently reported more frequently among children with JIA than among healthy peers.[35–38]


In the study of Leksel et al.,[38] orofacial symptoms influenced more often the daily life in a group of children with JIA compared to the healthy individuals. Santos et al.[29] also compared oral health-related quality of life in children and adolescents with JIA and healthy peers. The instrument Brazilian SF:13- B-PCPQ instrument was used and consisted of thirteen items related to oral symptoms, functional limitations and well-being. In the different groups, most caregivers indicated that the oral health status of their children and adolescents had little or no effect on their well-being, and no significant differences between the groups were found.

For the present review Additional file 3: Table S3 presents the critical appraisal of the included studies while Additional file 4: Table S4 shows a completed 2009 PRISMA check list.

Influence by JIA Activity and Severity

The majority of the studies contained some clinical information about the JIA status of the participants. Examples of descriptors were JIA category, disease activity, anti-rheumatic medication, JIA onset and functional impairment. Pugliese et al.[27] showed that well-established JIA disease and validated activity scores (Juvenile Arthritis Disease Activity Score (JADAS), physician global assessment of disease activity visual analogue scale (PhysglobVAS) and parent/patient global assessment of well-being VAS (PglobVAS) were positively correlated with the DMT score. Other scores; Escola Paulista de Medicina Range of Motion Scale (EPM-ROM) and Child Health Assessment Questionnaire (CHAQ) were positively correlated with Gingival Index (GI), PhysglobVAS was correlated with PI, and Pediatric Quality of Life Inventory 4.0 (PedsQL) parents was correlated with Gingival Bleeding Index (GBI). Savioli et al.[30] found that a subgroup of children with polyarticular RF negative JIA, had a statistically higher GBI and DMFT index than controls. Also a subgroup of children with three to eight affected joints in upper extremities, had significantly higher bleeding index than controls.[30] Self-reported pain or weakness in the hand when tooth brushing was documented in the study of Leksell et al.[25] as a problem among children with JIA. A significantly higher proportion of children with JIA compared with children without JIA, also answered that they did not brush their teeth when they did not feel well. Additionally, Miranda et al. reported a significantly higher mean number of joints with limitation of movements (LOM) in adolescents with two mm or more attachment loss (AL) than in adolescents without AL.[26] It is also worth mentioning the findings of Miranda et al.[32] of increased serum IL-18 and IL-1β in adolescents of JIA subgroups with AL, suggesting that AL might be associated with systemic inflammatory response. Low sample sizes, however, made it difficult to draw conclusions.

For children and adolescents with JIA, medication constitutes a substantial part of life which in turn may impact on oral health. Leksell et al. showed that children taking anti-TNFα had a higher frequency of sites with increased probing depth compared to children not taking this medicine.[25] Reichert et al.,[28] comparing adolescents with JIA who took non-steroidal anti-inflammatory drugs (NSAIDs) with other JIA peers who did not take drugs, found a significant decreased mean value for modified sulcular bleeding index in the NSAID group. The frequency of cyclosporine medication, assessed by Pugliese et al.,[27] was found to be higher in JIA patients with gingivitis compared with those without gingivitis. It is important to bear in mind that it is very difficult to differentiate the effect of single drugs from the effect of the disease activity with associated systemic inflammatory response in these studies. Children on anti-TNFα or cyclosporine probably have a more severe disease than children without these drugs, and differences in oral health between groups with or without a certain drug may be due to the disease severity and not the drug itself. Miranda et al.[33] in a follow-up study of adolescents with JIA documented that anti-rheumatic treatment resulting in reduction in disease activity clearly and positively influenced the periodontal status. After two years, no clinical or laboratory differences in periodontal inflammation could be documented between the adolescents with and without JIA. Pugliese et al.[27] documented that adolescents with an increased C - reactive protein (CRP) showed a higher mean clinical attachment loss (CAL) compared with those with normal CRP values.

All these reports about medication shared the previously reported problem of lacking adequate sample sizes for evidence. The comparisons were also hampered by the fact that the disease status of children and adolescents with JIA taking a certain drug were not the same as children and adolescents with JIA who did not take the drug.