Orofacial consequences of systemic sclerosis: A systematic review

Abstract

Orofacial involvement is common and often understated in the treatment clinical guidelines of systemic sclerosis. It impairs daily life by having repercussions on comfort, nutrition, aesthetics and self-confidence. This review aimed at describing exhaustively the different orofacial consequences of systemic sclerosis. A systematic search was conducted using four databases (PubMed, Cochrane Library, Dentistry & Oral Sciences Source and SCOPUS) up to December 2016 according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses. Grey literature and hand search were also included. To be eligible for the inclusion, studies needed to meet the following criteria: randomised controlled trials, cross-sectional studies, case-control studies, pilot studies or cohort studies and full text available in English or French, with abstract. The studies had to concern at least 30 patients suffering from systemic sclerosis and having clinical and radiological oropharyngeal examination. The diagnosis of systemic sclerosis had to be determined according to precise recommendations; the retrieved oropharyngeal manifestations had to affect hard or soft tissues of the mouth and/or pharynx and needed to be evaluated with clinical measures. Study selection, risk bias assessment (Newcastle–Ottawa scale) and data extraction were performed by two independent reviewers. The retrieved features were microstomia and xerostomia associated with real hyposialia, temporomandibular joint symptoms, high caries experience, periodontal diseases as well as an increased risk of oral cavity and pharynx cancer. Early diagnosis enabling early management, prevention and oral hygiene is the key to avoid complicated and invasive procedures. Studies with higher level of evidence remain necessary to create standardised protocols.

Introduction

Systemic sclerosis (SSc) is a rare autoimmune connective tissue disease characterised by vascular abnormalities and excessive collagen synthesis (1). The excessive production of extracellular matrix is due to abnormal interactions of different types of cells (endothelial cells, lymphocytes, monocytes and fibroblasts) in a setting of vascular hyperactivity and hypoxia (1, 2). Fibrosis typically affects the skin and soft tissues with an extension to face, lips and oral mucosa.

Orofacial abnormalities in SSc are frequent but understated (3). They may be left out when other serious systemic symptoms appear, such as pulmonary hypertension or renal crisis. Currently, the management of SSc mostly implies treatment of organ-specific complications but new strategies are progressively emerging, thanks to a better understanding of molecular pathways (4).

The most frequent oral features include dental, periodontal and bone repercussions (5). Poor oral health has adverse impacts on daily life, such as discomfort, pain and even malnutrition, which is responsible for 4% of SSc deaths (6). In addition, progressive SSc could be predictable by some orofacial manifestations, such as tightening of the perioral skin and ensuing microstomia (7). All these elements emphasise the importance of oral diagnosis.

This review aimed at (a) describing the different orofacial consequences of SSc and (b) giving keys to early diagnosis and management of these features.

Methods

Protocol and eligibility criteria

This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA statement) (8) and the PRISMA 2009 checklist.

To be eligible for the inclusion, studies needed to meet the following criteria:

• Randomised controlled trials, cross-sectional studies, case-control studies, pilot studies or cohort studies;

• Published from 1 June 2007 to 30 June 2017;

• Full text available in English or French, with abstract;

• At least 30 SSc patients presenting orofacial specificity or disease;

• Diagnosis of SSc according to precise recommendations, depending on the date of publication of articles;

• Orofacial manifestations affecting hard or soft tissues and evaluated with clinical measures (maximal mouth opening, periodontal measures, etc.).

No restrictions were applied concerning oral manifestations to obtain the most complete view. The exclusion criteria were comments, letters, editorials, reviews, systematic reviews, meta-analyses, case reports and cases series, opinions, abstracts without publication, publications without abstracts and studies, including less than 30 SSc patients and/or without clinical oropharyngeal examination or oral manifestations. In the same way, studies about treatments, comparisons or efficacy of treatments were not included.

Search strategy

Electronic and hand search, studies selection and risk of bias assessment were performed by two independent reviewers (R.S. and R.D.). An electronic search was conducted on four electronic databases through the access of PubMed (https://www.ncbi.nlm.nih.gov/pubmed/), Cochrane Library (http://www.cochranelibrary.com), Dentistry & Oral Sciences Source (DOSS; https://health.ebsco.com/products/dentistry-oral-sciences-source) and SCOPUS. Grey literature was searched via GreyLit (www.greylit.org). The search was restricted to the last 10 years (from 1 June 2007 to 30 June 2017).

In addition, hand search, including the international journals such as Rheumatology; Arthritis Care & Research; Journal of Dental Research and Oral Surgery, Oral Medicine, Oral Pathology, and Oral Radiology from June 2007 to June 2017, was performed on 1 July 2017 to enable a thorough screening process. Reference lists of included articles and relevant reviews were also manually searched at the same date.

The following search strategies were used on 1 July 2017:

• PubMed: (((‘Oral Manifestations’[Majr]) OR ‘Oral Health’[Majr]) OR ‘Mouth diseases’[Mesh]) AND ‘Scleroderma, Systemic’[Mesh].

• Filters activated: Abstract, published in the last 10 years, English, French.

• Cochrane Library: ‘systemic sclerosis’ AND ‘mouth’.

• DOSS: ‘systemic sclerosis’ AND ‘oral diseases’.

• The reference manager Zotero^® was used to organise the studies.

Screening and selection of articles

Duplicates were removed. All titles and abstracts of all identified articles were screened. Full-text articles of studies meeting the inclusion criteria were included in the systematic review. Two authors (R.S. and R.D.) independently reviewed the full texts. The final decision about the eligibility of all studies was made by mutual agreement.

Data extraction and quality assessment

The extraction of data was performed by R.S. using a standardised table. The following details were extracted: study and year, country, number of SSc patients, age, percentage of females, classification criteria for SSc, percentage of diffuse subtype, disease duration, specific oral features and results of the assessment of oropharyngeal involvement.

Quality assessment and risk of bias within studies were assessed using the Newcastle–Ottawa scale and completed with the study of sampling, measurement and attrition biases as well as examiners’ calibration.

Risk of bias across studies was assessed by studying publication bias and selective reporting within studies (reporting bias).

Results

Study selection

All searches provided a total of 431 citations published from 1 June 2007 to 30 June 2017: 257 from electronic search and 174 from manual searches. After duplicates removal, 391 citations were screened using titles and abstracts. After screening of title and abstract, 37 articles were obtained for full-text review.

At this step, 20 articles were excluded:

• 11 studies included less than 30 SSc patients;

• 2 described other autoimmune diseases;

• 3 were reviews;

• 1 did not treat about oral manifestations of SSc;

• 1 was a meta-analysis.

Finally, 17 articles met the inclusion criteria. The PRISMA flowchart of screening and selection processes is presented in Fig. 1.

Fig. 1.

PRISMA flow diagram.

Characteristics of included studies

The demographic characteristics of included studies are shown in Tab I. Among the 17 selected studies, 2 were retrospective cohorts (9, 10), 7 were case-control studies (11-17) and 8 were cross-sectional studies (5, 18-24). The retrieved studies came from different continents: 4 from North America (18-20, 24), 1 from South America (13), 5 from Asia (9, 10, 12, 15, 21), 6 from Europe (5, 11, 14, 16, 17, 22) and 1 from Oceania (23). Concerning the diagnosis criteria for the selection of SSc patients, most studies used international guidelines and classifications: American College of Rheumatology (ACR) (5, 13, 17, 22, 23), ACR/European League Against Rheumatism (EULAR) (11, 16, 18-20), LeRoy and Medgser (5). Four studies did not mention the classification used for the diagnosis of SSc (9, 10, 12, 24). The distinction between ‘diffuse’ and ‘limited’ cutaneous scleroderma (LeRoy classification) and disease duration were defined in most studies.

Table I.

Demographic characteristics of selected studies

Study	Country	Number of SSc patients	Mean age (years ± SD or range)	Females (%)	Diagnosis of SSc	Diffuse SSc (%)	Disease duration (years ± SD)
Baron et al. (18)	Canada	163	56.20 ± 10.56	89.6	ACR/EULAR criteria	27.5	13.9 ± 8.5
Baron et al. (19)	Canada	163	56.20 ± 10.56	89.6	ACR/EULAR criteria	27.5	13.9 ± 8.5
Chu et al. (12)	China	42	54.0 ± 12.2	97.6
Crincoli et al. (11)	Italy	80	56.28 ± 12.96	92.5	ACR/EULAR criteria	82.5	12 ± 8.39
Dagenais et al. (20)	Canada	163	56.31 ± 10.43	90.6	ACR/EULAR criteria	28.3	13.7 ± 8.4
Ferreira et al. (13)	Brazil	35	41.8 ± 11.6	100.0	ACR	60.0	10.08 ± 8.97
Knaś et al. (14)	Poland	55	35–65	100.0	ACR/LeRoy	63.6	10.08 ± 8.97
Kobak et al. (21)	Turkey	180	48.3	91.5	ACR		8.2
Kuo et al. (9)	Taiwan	2053		77.0	SSc certificate (ICD-9 code: 7101)
Leung et al. (15)	China	36	50.06 ± 11.7	97.2	LeRoy	30.5	12.3 ± 9.8
Matarese et al. (22)	Italy	30	52.4 ± 8.5	83.3	ACR
Pischon et al. (16)	Germany	58	51.1 ± 12.9	75.9	ACR/EULAR criteria	62.1	6.2 ± 5.1
Swaminathan et al. (23)	Australia	193	54.63	82.9	ACR	24.4
Tseng et al. (10)	Taiwan	2217	48.41 ± 14.76		SSc certificate (ICD-9 code: 7101)
Vincent et al. (5)	France	30	58.6 ± 13.6	87.0	ACR or LeRoy and Medsger	33.3	8.0
Yuen et al. (24)	United States	48	50.7 ± 13	79.2		41.7	7.6 ± 6.1
Zalewska et al. (17)	Poland	35	54 (36–70)	100.0	ACR/LeRoy	48.0	10.29 (1–30)

ACR = American College of Rheumatology; EULAR = European League Against Rheumatism; SD = standard deviation; SSc = systemic sclerosis; ICD = International Statistical Classification of diseases.

Concerning recruitment, SSc patients came from rheumatology and internal diseases departments, while healthy controls came from dentistry departments. Six studies used specific databases:

• Both studies of Baron et al. (18, 19) as well as Dagenais et al. (20) were based on the same sample from the Canadian Scleroderma Research Group (CSRG) registry, a multi-site cross-sectional study conducted between 2008 and 2011.

• Kuo et al. (9) and Tseng et al. (10) used the National Health Insurance Research Database (NHIRD) and most specifically the Registry of Catastrophic Illness Database which included information of all 23 million Taiwanese citizens.

• Yuen et al. (24) used the Connective Tissue Disease Database which reports SSc patients who received consultation and/or treatment at the Medical University of South Carolina.

Studies sizes varied from 30 to 2053 SSc patients. In case-control studies, healthy controls were often age- and gender-matched.

Risk of bias within and across studies

Quality assessment of selected studies was done according the Newcastle–Ottawa Assessment Scale and summarised in Tab II. Among cross-sectional studies, five did not provide a control group (5, 21-24).

Table II.

Newcastle–Ottawa Scale quality assessment

Study	Type of study	Selection	Comparability	Outcome	Total
Baron et al. (18)	Cross-sectional	**	**	**	6/9
Baron et al. (19)	Cross-sectional	**	**	**	6/9
Chu et al. (12)	Case-control	***	**	***	8/9
Crincoli et al. (11)	Case-control	**	**	**	6/9
Dagenais et al. (20)	Cross-sectional	***	**	**	7/9
Ferreira et al. (13)	Case-control	****	**	*	7/9
Knàs et al. (14)	Case-control	**	*	**	5/9
Kobak et al. (21)	Cross-sectional	**	**	*	5/9
Kuo et al. (9)	Retrospective cohort	***	**	**	7/9
Leung et al. (15)	Case-control	***	**	**	7/9
Matarese et al. (22)	Cross-sectional	**	**	*	5/9
Pischon et al. (16)	Case-control	***	**	**	7/9
Swaminathan et al. (23)	Cross-sectional	**	**	*	5/9
Tseng et al. (10)	Retrospective cohort	****	**	**	8/9
Vincent et al. (5)	Cross-sectional	**	**	*	5/9
Yuen et al. (24)	Cross-sectional	**	**	*	5/9
Zalewska et al. (17)	Case-control	***	**	**	7/9

^*= Quote given to the studies. it’s a quote out of 4* for selection, 2* for comparability and 3* for exposure. Maximum value given to a study by the sum of all categories is 9.

For studies with a healthy control group, sampling bias of controls could be pointed out because they were recruited from patients attending dental clinics or rheumatology departments for other diseases (mechanical joint disease and osteoarthritis). Only three studies had community controls (9, 10, 13), providing external validity.

Concerning the measurement bias, measures and questionnaires were most of the time done and checked by a single examiner in a standardised manner. Due to typical physical features of the disease, blinding of the examiner was not possible. When implying more examiners, four studies assessed calibration by calculating intra-observer and inter-observer reliability (Kappa coefficient) and provided the detailed results (12, 18-20). Outcome measures (such as maximal mouth opening) were most of the time taken the same way in all articles (by measuring the interincisal distance) according to international guidelines. Three studies performed multivariate regression analyses, adjusting for possible confounders (18, 19, 24).

Attrition bias was difficult to assess because no study mentioned loss of participants during studies. Concerning risk of bias across studies, no reporting bias was found because all authors mentioned significant as well as non-significant findings. Publication bias tried to be avoided by exhaustively screening literature, including different sources and grey literature.

Results of individual studies

To enable clearer results, manifestations were classified into the five following sub-groups:

• Skin and mucosa involvement: 7 studies (5, 11-13, 18, 19, 21);

• Subjective dryness and modifications of saliva: 11 studies (5, 10-12, 14, 17-19, 21, 23, 24);

• Bone resorption and temporomandibular joint (TMJ) involvement: 6 studies (5, 11-13, 16, 20);

• Dental repercussions: 12 studies (5, 12-20, 22, 24);

• Risk of oral cancer: 1 study (9).

Skin and mucosa involvement

The main reported feature was a reduced mouth opening measured by interincisal distance and ranged from 33.29 to 40.1 mm (Tab III) (11, 12, 18, 19). Three other studies reported microstomia (interincisal distance <40 mm) in 60%–91.4% of their SSc patients (5, 11, 13). A typical skin feature retrieved was the presence of telangiectasia and concerned 42.3%–81% of patients (5, 12, 21). Crincoli et al. (11) looked for mucosa repercussions and reported significantly increased prevalence of oral ulcers (38.8%), petechiae (22.5%) and erythematous tongue (46.3%).

Table III.

Skin and mucosa involvement

Study	Interincisal distance (mm ± SD)	Microstomia (patient %)	Telangiectasia (patient %)	Oral ulcers (patient %)	Petechiae (patient %)	Erythematous tongue (patient %)	Fibrous tongue (patient %)
Baron et al. (18)	37.68 ± 8.36*
Baron et al. (19)	37.68 ± 8.36*
Chu et al. (12)	40.1 ± 6.5*		81%	0%
Crincoli et al. (11)	33.29 ± 7.97*	80%*		38.8%*	22.5%*	46.3%*	26.3%
Ferreira et al. (13)		91.4%
Kobak et al. (21)			42.3%
Vincent et al. (5)		lSSc: 60% dSSc: 80%	lSSc: 70% dSSc: 70%

SD = standard deviation; dSSc = diffuse systemic sclerosis; lSSc = localised systemic sclerosis.

Results with * were statistically significant according to authors (p < 0.05).

Subjective dryness and modifications of saliva

Among subjective features, xerostomia and xerophthalmia concerned 29.2%–80% and 30%–65%, respectively (5, 11, 14, 17, 21, 23, 24). These patients’ complaints were completed by objective measures that showed significant low-unstimulated flow rate (12, 14) and Schirmer-I test (Tab IV) (17). Sjögren’s syndrome (SS) was found in 10.4%–33.9% of SSc patients (10, 21, 24). For the diagnosis of SS, most studies (5, 14, 17, 21, 23) used the criteria from the American-European consensus Group (AECG) (25). The other studies did not describe in details the diagnosis criteria; the occurrence was reported by past medical records and the use of the International Statistical Classification of diseases (ICD-9 code 7102). According to AECG classification, in patients with another connective tissue disease, the presence of item I (ocular symptoms) or II (oral symptoms) plus any two from among items III (ocular signs), IV (histopathology) and V (objective evidence of salivary gland involvement) may be considered as indicative of secondary SS (25). We noticed the absence of item VI which corresponded to the presence of autoantibodies (SSA, SSB or both). Three studies performed biopsies on SSc patients, none on healthy controls because of a lack of ethics committee consent (14, 17, 21).

Table IV.

Subjective dryness and modifications of saliva

Study	Xerostomia (patient %)	Xerophthalmia (patient %)	Unstimulated saliva flow rate (mg/min or mL/min ± SD or ranges)	Schirmer-I test (mm/5 min) median (ranges)	Sjögren’s syndrome (patient %)	Sjögren’s antibodies (anti-SSA and anti-SSB; patient %)	Specific peroxidase activity in the unstimulated saliva (median IU/100 mg)
Baron et al. (18)			147.52 ± 95.07 mg/min			16.9%
Baron et al. (19)			147.52 ± 95.07 mg/min
Chu et al. (12)			0.18 ± 0.17 mL/min*
Crincoli et al. (11)	52.5%*
Knàs et al. (14)	lSSc: 45%* dSSc: 51.4%*	lSSc: 60%* dSSc: 45.7%*	lSSc: 0.2 (0.14–0.5)* dSSc :0.21 (0.12–0.32)*	Left eye:  lSSc:18 (13–35)  dSSc: 20 (0–35) Right eye:  lSSc: 21 (1–33)  dSSc: 24 (8–35)	0%	0% for both groups	lSSc: 0.00006* dSSc: 0.00005*
Kobak et al. (21)	29.5%**	26.9%**			33.9%	SSA: 10.3%** SSB: 5.1%
Swaminathan et al. (23)	lSSc: 59% dSSc: 49%					4.1%
Tseng et al. (10)					16.7%
Vincent et al. (5)	lSSc: 80% dSSc: 40%*	lSSc: 65% dSSc: 30%				lSSc: 20% dSSc: 20%
Yuen et al. (24)	29.2%				10.4%
Zalewska et al. (17)	SScN: 59.1%* SScH: 59.1%*	SScN: 92.3%* SScH: 61.5%*	SScN: 0.33 (0.21–0.46) SScH: 0.11 (0.002–0.2)*	Left eye:  SScN: 20 (6–35)*  SScH: 6 (0–33)* Right eye:  SScN: 18 (2–35)  SScH: 6.5 (1–35)		0%	SScN: 0.00011* SScH: 0.00003*

dSSc = diffuse systemic sclerosis; lSSc = localised systemic sclerosis; SD = standard deviation; SS = Sjögren’s Syndrome; SScH = systemic sclerosis patients with hyposalivation; SScN = systemic sclerosis with normal unstimulated salivary flow rate.

Results with * were statistically significant compared to healthy controls according to authors (p < 0.05).

Results with ** were statistically significant compared to the SS–SSc group according to authors (p < 0.05).

Biochemical modifications of the saliva were pinpointed by a statistically significant decrease in the specific peroxidase of the unstimulated saliva (ranging from 0.00003 to 0.00011 IU/100 mg) (14, 17).

Bone resorption and TMJ involvement

Bone resorptions and erosions ranged from 2.4% to 20% of patients and concerned more particularly condyles (5, 12, 13, 20). TMJ symptoms remained a major issue with 92.5%–94.8% SSc patients being affected (Tab V) (11, 16). Most reported TMJ changes were bone erosion associated with crepitation and muscle pain on chewing and dysfunction (5, 11-13, 16, 20). It was suggested that unlike other rheumatologic diseases, such as rheumatoid arthritis, TMJ changes were due to increased skin thickness and decreased skin elasticity and not to joint inflammation (11, 13). Although arthropathy is common among SSc patients, it seems difficult to separate lesions due to TMJ involvement from those due to perioral tissues fibrosis (lips, skin and subcutaneous tissues) (26). None of the studies mentioned subluxation.

Table V.

Bone resorption and temporomandibular joint involvement

Study	Bone resorption/erosion (patient %)	Condylysis (patient %)	TMJ disorders/symptoms (patient %)	TMJ dysfunction (patient %)	TMJ tenderness/pain (patient %)
Chu et al. (12)	2.4%				7%
Crincoli et al. (11)			92.5%*		91.2%
Dagenais et al. (20)	14.47%*	6.29%*
Ferreira et al. (13)	14.2%	28.6%	Symptom-free (Ai0): 20%* Mild symptoms (AiI): 17.2%* Severe symptoms (AiII): 62.8%*	Normal (Di0): 8.6%* Mild (DiI): 48.6%* Moderate (DiII): 22.8%* Severe (DiIII): 20%*
Pischon et al. (16)			94.8%*
Vincent et al. (5)	lSSc: 0% dSSc: 20%				lSSc: 20% dSSc: 40%

A_i = anamnestic Helkimo index; D_i = clinical Helkimo index; dSSc = diffuse systemic sclerosis; lSSc = localised systemic sclerosis; TMJ = temporomandibular joint.

Results with * were statistically significant compared to healthy controls according to authors (p < 0.05).

Dental repercussions

Number of missing teeth ranged from 3.7 to 7.96 but did not seem statistically significant compared to controls (16, 18-20, 24). Mean number of decayed, missing or filled teeth (DMFT) ranged from 10.5 to 23.5 (Tab VI) (12, 14, 16, 17). Periodontal diseases’ indices were the most detailed using evaluation of the tissue loss around teeth (clinical attachment loss; probing depth), inflammation (bleeding on probing, gingival index and plaque index) as well as the community periodontal index (CPI) or community periodontal index of treatment needs (CPITN) which gave treatment needs. All indices were increased (14-16, 22) and Matarese et al. (22) evaluated the mean CPITN at 3 which requires advanced periodontal treatment (deep scaling, root planning and surgical procedure). Crincoli et al. (11) was the only study reporting bruxism and concerned 81.2% of SSc patients. Widening of the periodontal ligament, a usual feature among SSc patients, was reported by three studies and concerned almost one patient out of three (5, 15, 20).

Table VI.

Dental repercussions

Study	Number of missing teeth (mean ± SD or ranges)	Number of decayed teeth (mean ± SD)	Caries experience: DMFT (median, ranges or SD)	AL/CAL (mean full mouth in mm, ranges or SD)	BOP (mean % ± SD)	PI (mean % ± SD)	GI (median, ranges)	CPI/CPITN	PD (mean full mouth in mm, ranges or SD)	Bruxism (patient %)	Number of teeth with widening of the periodontal ligament (mean ± SD) or patient % with a widening of the periodontal ligament	Number of teeth periodontal disease (mean ± SD) or patient % with periodontal disease	Mean periodontal ligament space (mm ± SD)
Baron (18)	7.90± 9.44	0.88 ± 1.82*										5.23 ± 5.63
Baron (19)	7.90± 9.44											5.23 ± 5.63*
Chu (12)		2.1 ± 2.4	10.5 ± 7.8					CPI 0: 0% patients CPI 1: 3% CPI 2: 20% CPI 3: 40% CPI 4: 37%
Crincoli (11)										81.2%
Dagenais (20)	7.96 ± 9.46										1.08 ± 2.57*
Knàs (14)			lSSc: 18 (8–28)* dSSc: 22.5 (10–28)*				lSSc:0.5 (0–1) dSSc: 1 (0–1)		lSSc: 2 (1–3) dSSc: 2.5 (1–3)
Leung (15)				3.19 ± 0.94	78.4 ± 19.6*	69.3 ± 18			2.52 ± 0.58*				0.36 ± 0.06*
Matarese (22)				4 (4.25–4.78)	1 (0.91–1.27)	2 (1.44–1.7)		Mean CPITN: 3 (2.57–2.85)	5 (4.53–4.98)
Pischon (16)	5.50 (2.0–10.0)		17.66 ± 6.00	4.01 ± 1.04*	0.29 (0.2–0.45)	0.88 (0.46–1.73)*	0.16 (0.07–0.40)*		2.99 ± 0.59
Vincent (5)											lSSc: 35% dSSc: 30%	lSSc: 10% dSSc: 40%
Yuen (24)	3.7 ± 4.0
Zalewska (17)			SScN: 20.5 (9–28) SScH:23.5 (16–28)*				SScN: 1 (0–2.5) SScH: 1 (0–2.5)		SScN: 2 (1–3) SScH: 2 (0.5–3)

AL = attachment loss; BOP = bleeding on probing; CAL = clinical attachment loss; CPI = community periodontal index; CPITN = community periodontal index of treatment needs; DMFT = mean number of decayed, missing or filled teeth; SD = standard deviation; dSSc = diffuse systemic sclerosis; GI = gingival index; lSSc = localised systemic sclerosis; PD = pocket depth; PI = plaque index; SScH = systemic sclerosis Patients with hyposalivation; SScN = systemic sclerosis with normal unstimulated salivary flow rate.

Results with * were statistically significant compared to healthy controls according to authors (p < 0.05).

Risk of oral cancer

Kuo et al. (9) reported a standardised incidence ratio (SIR) of oral cavity and pharynx cancers of 3.67 (1.83–6.56)* with a higher SIR for males (4.00 (1.86–7.60)* vs 3.00 (0.76–8.17); confidence interval (CI) of 95%; *p < 0.05). This ratio was calculated by dividing the observed numbers of cancer cases among patients with SSc by the expected number derived from the incidence rate among the general population (9). The CI was calculated by Kuo et al. (9) under the assumption that the observed and expected number of cases had Poisson distributions. A study by Derk et al. (27) reported as well a high risk of oropharyngeal cancer (SIR: 9.63) with a specific risk of tongue cancer which was 25-fold greater among SSc patients than that expected in an age-adjusted population. No specific risk factors except gender were reported, but authors discussed the impact of tobacco and smoking having a significant adverse effect on almost all vascular, gastrointestinal and respiratory outcomes on SSc patients (28).

Discussion

This systematic review enabled us to highlight the most frequent orofacial repercussions of SSc reported which were microstomia and xerostomia due to hyposialia (objectified by salivary flow measurements), TMJ symptoms, high caries experience, periodontal diseases as well as an increased risk of oral cavity and pharynx cancer. These findings came from the strongest observational studies on the subject and are in accordance with other reviews (29-31). Orofacial involvement in SSc has a crucial impact on global functions. For example, nutritional involvement is initiated by insufficient salivary secretion; the patient is unable to properly lubricate the food bolus and the oral cavity, increasing tooth loss (by raising the incidences of tooth decay, erosion and periodontal disease), ultimately leading to impaired mastication and complicated dysphagia (32, 33). Moreover, in 5%–10%, gastrointestinal disorders are the leading cause of death (34). Breaking this vicious circle is a priority.

Most selected studies had a medium- to low-risk bias. Reporting bias was limited since all studies mentioned significant as well as non-significant results and gave details about preliminary tests. Extending the search to grey literature and exhaustive manual search limited publication bias. SSc being a rare disease, most studies describing orofacial manifestations were case reports and case series with very low level of evidence. Focus on studies with at least 30 patients published within the last 10 years enabled to obtain more recent and stronger results. All identified studies were available and retrieved in full text.

Our findings are consistent with other studies such as the recommendations for the care of oral involvement in patients with SSc published in 2011 (35)

Prevention of mouth infections and tooth decay implies mouth hygiene education, periodontal maintenance and treatment of hyposialia (35-37). Since SSc is a multi-system disease, it requires a well-coordinated multidisciplinary management. Daily oral upkeep can be impaired and complicated when patients have sclerodactyly or depression. Indeed, Yuen et al. (38) showed that significant depressive syndromes induced daily oral care neglect (38). Early diagnosis is needed for an early management, avoiding invasive treatments as much as possible. However, if necessary, treatments can be realised even with severe microstomia (35, 39). Studies with higher level of evidence remain necessary not only to create standardised protocols and recommendations of oral management but also to strengthen the proof of the negative impact of orofacial manifestations on global health and quality of life of SSc patients.