Gastrointestinal involvement in systemic sclerosis: An update

Abstract

Purpose of Review:

This review provides important updates in systemic sclerosis (SSc)-related gastrointestinal (GI) disease, specifically focusing on the most recent literature.

Recent findings:

In the past year, several studies were published that present interesting insights into SSc and GI disease. Studies focusing on newly identified risk factors, novel approaches to diagnosis and assessment of disease activity, survival and quality of life demonstrate progress in our understanding of the challenging area. Additional data on specific SSc GI-related topics, such as the link between GI and pulmonary disease, nutrition, and the microbiome, are also now available.

Summary:

SSc GI disease is heterogeneous in its clinical presentation, which presents a challenge in diagnosis and management. In the past year, several studies have evaluated risk factors and clinical features associated with specific GI complications in SSc. Objective GI testing may help to identify specific SSc GI subgroups and provide diagnostic accuracy to guide targeted therapies. Survival in very early SSc is affected by the severity of GI involvement. Other important GI subsets exist, including patients with esophageal disease and interstitial lung disease, should carefully be considered when developing a management plan for this patient population.

INTRODUCTION

Gastrointestinal (GI) involvement is the most common internal organ complication of systemic sclerosis (SSc), affecting over 90% of patients. GI clinical manifestations in SSc are variable, as regions from the esophagus to the anorectum may be affected. This clinical heterogeneity often leads to several key challenges for the treating physician. These include: (1) identifying patients at high risk for progressive severe GI disease; (2) determining whether immune-mediated disease activity or existing damage is causing symptoms; (3) determining whether early initiation of pro-motility agents or other GI medications may prevent complications; (4) determining whether there is a role for immunosuppression in preventing GI complications. GI complications in SSc also significantly impact healthcare costs and quality of life (1-3).

In the past year, several manuscripts were published that begin to address some of these areas of critical need in SSc GI disease. Here the data and clinically relevant information are provided which may help physicians diagnose and manage this complicated group of patients.

TEXT OF REVIEW

Methods:

The following initial search terms (01/01/2018-04/01/2019) were used in PubMed: ‘scleroderma gastrointestinal’ (denominator=20); ‘SSc gastrointestinal’ (denominator=28), ‘SSc gastrointestinal treatment’ (denominator=18), ‘SSc GERD’ (denominator=6), ‘SSc gastroparesis’ (denominator=0), ‘SSc gastric antral vascular ectasia’ (denominator=1), ‘SSc SIBO’ (denominator=3), ‘SSc bowel’ (denominator=29), ‘SSc colon’ (denominator=2), ‘fecal incontinence SSc’ (denominator=2), and ‘SSc nutrition’ (denominator=12). The search was specific for human studies. There were 35 original research papers of interest identified based on their relevance to SSc GI disease. Case reports were excluded from this review.

RESULTS

Clinical, serologic, and genetic risk factors associated with GI complications in SSc

Clinical and demographic risk factors

Clinical features and other risk factors associated with severe GI dysmotility in SSc were recently reported in a retrospective cohort study (Table 1) (4). Patients with severe GI disease requiring total parenteral nutrition, and patients with mild GI disease alone were compared, and it was determined that severe GI disease was significantly male sex (OR 2.47, 95% CI 1.34-4.56), myopathy (OR 5.53, 95%CI 2.82-10.82), and sicca symptoms (OR 2.40, 95% CI 1.30-4.42), even after adjusting for potential confounders. Patients with higher skin scores were also found to have an increased the risk of significant GI dysmotility (5), and a disease duration of greater than 5 years was associated with a high risk for developing small intestinal bacterial overgrowth (SIBO), (OR 9.38; 95%CI 1.09-80.47) (6).

Table 1.

Summary of clinical and serologic features associated with significant GI disease

GI Outcomes	Associated variables	Point estimates or other measures	PubMed number
Severe vs. mild GI disease (Medsger GI = 4 vs. Medsger GI ≤1)	Male sex	OR 2.47 (95% CI 1.34-4.56)	PMID: 29193842
	Myopathy	OR 5.53 (95%CI 2.82-10.82)
	Sicca symptoms	OR 2.40 (95% CI 1.30-4.42)
Severe vs. mild to moderate GI disease (Medsger GI ≥ 3 and/or ≥ 10% weight loss vs. Medsger GI < 3)	Myositis	OR 4.68 (95% CI 1.65, 13.24)	PMID: 30517716
	Telangiectasia	OR 2.45 (95% CI 1.19, 5.04)
	Higher modified Rodnan skin scores	OR 1.03 (95% CI 1.01, 1.07)
	Death	HR 2.27 (95% CI 1.27, 4.09)
	Worse health-related QOL	β = −2.37 (SE 1.04)
Pseudo-obstruction vs. none (Medsger GI = 3 vs. Medsger GI < 3)
	Older age	HR 1.02; 95%CI 1.00-1.04	PMID: in press
	Male sex	HR 1.75; 95%CI 1.42-2.43
	Diffuse cutaneous disease	HR 2.52; 95%CI 1.59-3.99
	Myopathy	HR 1.83, 95%CI 1.09-3.08
	Opioid use	HR 2.38; 95%CI 1.50-3.78
Moderate to severe vs. mild GI disease (Medsger GI ≥ 2 vs. Medsger GI <2)	More symptoms of autonomic dysfunction
	- Orthostatic intolerance	median 10.0 vs. 0; p = 0.006	PMID: 29907667
	- Secretomotor dysfunction	median 6.4 vs. 4.3; p = 0.03
	Anti-RNPC3 antibody positive	OR 3.8 (95% CI 1.0-14.3)	PMID: 30242973
	Anti-RNA polymerase 3 antibodies, subunit RPC155 vs. both RPC155 and RPA194 subunits	51% vs. 26%; p = 0.043	PMID: 30888702
Higher UCLA GIT 2.0 scores	Carriers of the C-allele compared to the G-allele of the IL-6 gene	85% vs. 50%; p < 0.05	PMID: 29948346
	UCLA GIT 2.0 bloating and distention scores	1.4 ± 0.9 vs. 0.78 ± 0.8; p = 0.05

Medsger GI = Medsger GI severity score; UCLA GIT 2.0 = UCLA GI tract 2.0 patient reported outcome survey where higher scores = more severe symptoms; QOL = quality of life; GIT = GI tract; PMID = PubMed ID; OR = odds ratio; HR = hazard ratio

The incidence, predictors and outcomes associated with severe GI disease in SSc were also explored by a group in Canada. Patients with early disease (disease duration <2 years) were included and severe GI disease was defined as a Medsger GI severity score of ≥ 3 (pseudo-obstruction +/− malabsorption) and/or ⩾10% weight loss in association with the use of antibiotics for bacterial overgrowth or esophageal stricture (7). The probability of developing severe GI disease was estimated at 9.1% at 2 years and 16.0% at 4 years. Severe GI disease was associated with myositis (OR 4.68, 95% CI 1.65, 13.24), telangiectasia’s (OR 2.45, 95% CI 1.19, 5.04) and a higher modified Rodnan skin score (OR 1.03, 95% CI 1.01, 1.07) in the adjusted model. Severe GI disease was also associated with a more than 2-fold increase in the risk of death (HR 2.27, 95% CI 1.27, 4.09) and worse health-related quality of life [Short Form Health Survey physical (β = −2.37, P = 0.02) and mental (β = −2.86, P = 0.01) component summary scores].

The clinical and demographic factors associated with pseudo-obstruction in SSc was also reported in a longitudinal cohort study using the Johns Hopkins Scleroderma Center Database. Patients with a modified Medsger GI score of 3, who had evidence of pseudo-obstruction as confirmed by chart review, were compared to patients with no history of pseudo-obstruction (Medsger GI <3) (8). Factors predictive of pseudo-obstruction were older age (HR 1.02; 95%CI 1.00-1.04), male sex (HR 1.75; 95%CI 1.42-2.43), diffuse cutaneous disease (HR 2.52; 95%CI 1.59-3.99), myopathy (HR 1.83, 95%CI 1.09-3.08), and opioid exposure (HR 2.38; 95%CI 1.50-3.78). Autoantibodies to RNA polymerase-3 were negatively associated with pseudo-obstruction (HR 0.34; 95%CI 0.17-0.66). Avoidance of opioids when possible in high-risk patients may reduce pseudo-obstruction events in SSc.

These studies all provide evidence that male sex, skeletal muscle involvement, and higher modified Rodnan skin scores and/or diffuse cutaneous disease were associated with more severe GI disease. Studies aimed at understanding the association between the GI motility (perhaps smooth muscle atrophy), SSc skeletal myopathy, and severe cutaneous disease may provide insights into disease mechanism.

Symptoms of autonomic dysfunction

As the autonomic nervous system plays an important role in controlling GI motility, it is interesting that symptoms of autonomic dysfunction were found to associate with GI severity in SSc (9). SSc patients (n=104) were recruited during routine clinical visits and asked to complete the COMPASS-31 questionnaire, a validated tool to assess symptoms of autonomic dysfunction. Patients with more severe GI disease had significantly higher scores across several COMPASS-31 subdomains, including orthostatic intolerance and secretomotor dysfunction. There was also a dose-response relationship between GI disease severity and autonomic symptom burden, suggesting that autonomic dysfunction may play an important role in SSc GI dysmotility. Interestingly, another study found that SSc GI patients with autonomic symptoms had more emotional distress on the GIT 2.0 survey than patients with minimal autonomic symptoms. This suggests that autonomic dysfunction in SSc may not only affect GI motility, but may also play a role emotional distress (10).

Autoantibodies

The presence of specific autoantibodies are reported to identify high-risk SSc GI patients (11). Two large SSc cohort studies evaluated the association between anti-RNPC3 antibodies and SSc GI disease. Anti-RNPC3 antibodies were associated with moderate to severe GI disease, defined by Medsger GI score ≥2 (OR 3.8, 95% CI 1.0-14.3), even after adjusting for relevant covariates. An association between severe GI disease and antibodies targeting only the RPC155 unit of RNA polymerase-3 rather than both protein subunits, RPC155 and RPA194 (51% vs. 26%; p = 0.043), was also reported (12), suggesting that specific autoantigen subunits targeted in SSc can identify patients with a high-risk for severe GI disease.

Genetic factors

Given the observation that polymorphisms in the IL-6 gene are important in the susceptibility to SSc, the clinical manifestations associated with the 174 C/G of the IL-6 gene polymorphism were evaluated in SSc patients (n=102) and healthy controls (n=93) (13). Carriers of the C-allele compared to the G-allele of the IL-6 gene, showed higher UCLA GIT 2.0 total scores (0.85 vs. 0.5, p < 0.05) and higher bloating/distension scores (1.4 ± 0.9 vs. 0.78 ± 0.8, p = 0.05), suggesting that the IL-6 gene variant −174 C/G is associated with greater risk for severe GI symptoms. The study did not explore objective measures of GI function in that no objective GI imaging was reported.

Updates in SSc GI disease: Diagnostic testing

Assessment of gastroesophageal reflux disease (GERD)

The degree of esophageal exposure to gastric acid, despite PPI therapy, was reported in patients with SSc (14). Investigators performed a case-controlled retrospective analysis, including 38 SSc and 38 non-SSc patients who underwent esophageal pH testing matched for PPI formulation and dose, hiatal hernia size, age, and gender. SSc patients had significantly longer acid exposure times, longer median bolus clearance, and lower nocturnal impedance values, suggesting that adjunctive therapies such as a pro-kinetic drug is needed for more adequate GERD control.

Because distal esophageal baseline impedance (BI) levels reflect the esophageal mucosal integrity in reflux disease, investigators sought to determine whether BI levels identify esophageal involvement in SSc (15). Approximately 100 patients with non-erosive reflux disease (NERD) or SSc, and 50 healthy controls were prospectively evaluated. Median BI values were lower in both SSc and NERD patients compared to healthy controls (p<0.01). Measurement of BI may be used as an objective marker of SSc esophageal involvement.

Assessment of esophageal motility

The clinical diagnosis was determined in a cohort of 207 patients with absent esophageal contractility defined by high-resolution esophageal manometry studies (16). Systemic autoimmune rheumatic diseases were identified in 81% (n=169) of cases, including 64% (n=132) with SSc. The remaining patients (20%; n=38) had non-rheumatic diseases, demonstrating that the absence of esophageal contractility is not specific for SSc.

The association between esophageal dysmotility and abnormal gastric transit was assessed using esophageal high-resolution manometry (HRM) and liquid and solid gastric emptying scintigraphy (GES) data. GI patient data (n=482), which included a subset of SSc patients (n=33) were reviewed and analyzed. Patients with esophageal dysmotility were more likely to be older (OR 1.013), have abnormal gastric transit (OR 2.14), SSc (OR 6.29), and dysphagia (OR 2.63), while patients with abnormal gastric transit were more likely to have esophageal dysmotility (OR 2.11), autonomic dysfunction (OR 2.37) and other findings. The strong association between esophageal dysmotility and abnormal gastric transit provides evidence that a common pathogenic mechanism, such as autonomic dysfunction, is present in both conditions (17).

Evaluating GI disease activity in SSc with a novel tool

A novel 18F PET-MRI with T1 MOLLI mapping was utilized to screen 16 SSc patients and five healthy controls for inflammation and fibrosis in the bowel (18). A significant increase in mean T1 values in the large (p<0.001) and small bowel (p=0.02) were identified between cases and controls, and the percentage of non-fibrotic and non-inflamed tissue was significantly lower in SSc patients than controls for the large bowel (p=0.03). However, none of these findings correlated with the GIT total or subdomain scores. The clinical relevance of the abnormalities observed by this tool are therefore not defined.

Treatment of GI complications in SSc

Dietary modification in SSc GI disease

The evidence to support dietary modification in the management of SSc GI symptoms was examined by a systematic literature review (19). Though some improvements in patient-reported outcome assessment of GI symptoms were identified after either the initiation of probiotic therapy or the initiation of a low-FODMAP diet, the overall level of evidence was weak and could not fully support dietary modification for treatment of GI involvement in SSc.

Risk factors associated with malnutrition in SSc were examined in a French population (20). SSc patients who had vitamin C, Se and/or thiamine levels checked during a 5-year period were included. Overt malnutrition was present in 14 (17%) patients, and deficiencies in Se (35%), vitamin C (31%) and/or thiamine (6%). Malnourished patients had significantly lower hemoglobin (10.6 vs. 12.9 g/dL, p< .0001) and vitamin C levels (3.6 vs. 10.6 mg/L, p= 0.003). Vitamin C deficiency also associated with esophagitis or Barrett's mucosa (OR 4.05, 95%CI 1.27-13.54; p = 0.02), modified Rodnan skin score ≤ 14 (OR 0.33, 95%CI 0.11-1; p= 0.05), and pulmonary artery hypertension (27% vs. 0%; p= 0.0006). Vitamin C testing may identify important complications in malnourished SSc patients. The potential determinants of malnutrition (2015 European Society of Clinical Nutrition and Metabolism (ESPEN)) was explored in a large SSc cohort (21), where outpatients (n=141) were enrolled and body composition was analyzed by densitometry. Patients with malnutrition (9.2%) had a significantly lower forced vital capacity (FVC) and more overall severe disease.

Probiotics in SSc GI disease

A randomized placebo-controlled trial examined the effects of probiotics in SSc. Patients with moderate to severe total scores on the UCLA GIT 2.0 (n=73) were included and followed over 8 weeks (22). The primary endpoint was improvement in the total GIT score. While there was no difference in GIT scores between groups, the trial did not enrich for patients with distention and bloating, which is the indication for which probiotics have previously been found to be helpful in SSc (23). As the total GIT score is a composite of a variety of GI symptoms, this small heterogeneous population may have been underpowered to detect the benefit of probiotics between important GI subgroups.

Medications

Several studies in the past year evaluated the medical treatment of GI complications in SSc. A systematic review and meta-analysis examined the treatment of SSc-related SIBO (24). Treatment outcomes included symptomatic relief or SIBO eradication. Studies were generally of low quality and un-controlled. The effectiveness of octreotide, individual antibiotics or antibiotic combinations were examined. They concluded that antibiotics may eradicate SIBO in some patients, but that there is a paucity of data reporting the effectiveness of either prokinetics (e.g. octreotide) or probiotics in SSc.

Another study evaluated the side effects, medication adherence, and dose ranges for pyridostigmine in SSc patients with refractory GI symptoms. Pyridostigmine is not labelled for use in scleroderma GI disease. Patients were defined as responders if they remained on pyridostigmine for at least 4 weeks with documented clinical benefit. Of 31 patients treated with pyridostigmine for at least 4 weeks, 51.6% reported symptomatic improvement. Constipation was the most commonly improved symptom based on prevalence prior to therapy, and diarrhea was the most common adverse event. Pyridostigmine may hold promise in SSc GI disease, particularly in patients with refractory constipation, though controlled studies need to be done (25).

Minimally invasive surgery

The treatment of refractory GERD in patients with SSc with laparoscopic Roux-en-Y gastric bypasses (RYGB) was evaluated at Cleveland Clinic (26). SSc patients undergoing fundoplication (n=7) or RYGB (n=7) for the treatment of GERD (2004 to 2016) were identified. Of the patients who had assessment of their GERD symptoms at follow-up, all five patients in the RYGB group and only three (50%) patients in the fundoplication group reported symptom improvement, and no mortality occurred during the 30-day follow-up. While more studies are needed, laparoscopic RYGB may be a safe and effective alternative to fundoplication in a subset SSc patients with severe esophageal dysmotility

Impact of GI complications on healthcare costs and quality of life in SSc

Rates of SSc-related adult hospitalizations in the United States and factors associated with in-hospital mortality, longer length of stay and higher hospital costs was studied using the National Inpatient Sample (2012-13) (1). From the GI standpoint, acute bowel obstruction and aspiration (OR > 2.0 with P < 0.0001 for both) both predicted higher cost of hospitalization. This emphasizes the importance of defining risk factors for this subgroup and initiating specific interventions to reduce risk in such patients.

The correlation between health-related quality of life and the oropharyngeal manifestations of SSc was assessed in a systematic review (27). Oropharyngeal manifestations of SSc (i.e. maximal mouth opening, Mouth Handicap in SSc Scale) were significantly associated with an impaired quality of life. However, overall there was a low level of evidence among the included studies. .

The impact of malnutrition on quality of life (QoL) was studied in 129 patients with SSc who were screened with the Malnutrition Universal Screening Tool (28). All patients completed the Short Form 36 Questionnaire and the Scleroderma Health Assessment Questionnaire (SHAQ). The prevalence of malnutrition was 10.9%. All QoL scores (except bodily pain and self-reported health) were significantly impaired in malnourished patients, and the SHAQ, which assesses disease-specific QoL, was significantly higher in the malnourished patients. The authors proposed that standardized nutritional screening should be conducted in SSc to identify the risk of malnutrition and enable the initiation of multimodal treatment.

Functional disability and its predictors in SSc (EUSTAR) were assessed using patients from the prospective DeSScipher cohort with a completed SHAQ (29), and the effects of disability-related factors were analyzed. High SHAQ scores were associated with the presence of GI symptoms (esophageal, gastric or intestinal) in the multivariable model suggesting that SSc patients perceive GI complications as significant contributors to their level of disability.

The association between poor sleep and GERD was also evaluated in patients with SSc (n=287) who completed the UCLA GIT 2.0, the Pittsburgh sleep quality index (PSQI), the fatigue severity scale (FSS), and the multidimensional gastrointestinal symptom severity index (GSSI) (30). Poor sleep quality was identified in 194 (68%) patients and associated with significantly higher GIT Reflux scores (p<0.001), and moderate/severe heartburn on GISSI (p<0.001). The association between GERD symptoms and poor sleep remained in the multivariable model (OR 2.53, 95%CI 1.52-4.25; p< 0.001).

Updates in the microbiome and H. pylori in SSc

One systematic review and metanalysis characterized the association between gastric H. pylori infection and SSc (31). Eight observational studies (n=1,446 subjects) were examined and in the pooled results there was an increased prevalence of H. pylori infection in SSc patients compared to healthy controls (OR 2.10, 95%CI 1.57-2.82). Prospective studies exploring the timing of SSc onset and H. pylori infections would be helpful in further understanding this interesting association. Fecal microbiota and the plasma metabolome were characterized in SSc patients (n=59) and healthy controls (HCs) (n=28) (32). A model of nine bacteria was capable of differentiating HCs from SSc patients. It was determined that SSc gut microbiota have fewer protective butyrate-producing bacteria and more pro-inflammatory noxious genera, especially Desulfovibrio compared to HC’s. Interestingly, a multivariate model with 17 metabolite intermediates clearly distinguished cases from controls. This is consistent with prior studies suggesting that SSc intestinal microbiota are characterized by pro-inflammatory alterations that may promote intestinal damage and influence amino acid metabolism.

The association between esophageal disease and interstitial lung disease in SSc

Several cross-sectional international studies the past year have evaluated the association between esophageal dysfunction and interstitial lung disease (ILD) in SSc. Even after adjusting for potential confounders, the studies from Greece, Japan and Italy all confirmed an association between esophageal involvement and interstitial lung disease in SSc (33-35). A longitudinal study examined whether a causal relationship between esophageal disease and ILD severity, ILD progression and mortality exists (36). HRCT scans from 145 SSc-ILD patients were scored for fibrosis, esophageal diameter, and the presence of a hiatal hernia. Interestingly, for every 1 cm increase in esophageal diameter, a 1.8% higher fibrosis score and 5.5% lower forced vital capacity was identified (p≤ 0.001). Patients with a hiatal hernia had a higher fibrosis score (p=0.001). Though esophageal diameter predicted worsening fibrosis score over the subsequent year (p=0.02), this was not significant when adjusting for the baseline fibrosis score (p=0.16). The investigators ultimately determined that while esophageal diameter and hiatal hernia are independently associated with SSc-ILD severity and mortality, they are not associated with ILD progression. It was therefore concluded that it is unlikely that esophageal disease is a significant cause of SSc ILD progression.

Scleroderma Clinical Trials Consortium GI working group

The Scleroderma Clinical Trials Consortium continues to support collaborative research in SSc GI disease with the aim of to improving clinical trials and observational studies (37, 38). Ongoing international GI-focused studies are now further exploring the role of the microbiome in SSc GI complications using an international patient population.

CONCLUSION:

The risk stratification, diagnostic approach, outcomes, and management of SSc GI disease remain active areas of research. While there has been progress in the past year in each of these areas, there remains much work to be done. Developing further insights into disease mechanism, biomarkers of disease activity, and standardized, data-driven approaches to diagnosis and management remain high priorities in this field.

KEY POINTS:

• GI disease in SSc significantly healthcare costs and patient quality of life

• Several clinical and demographic features are associated with patients at high risk of developing severe GI complications in SSc

• Objective testing in SSc GI disease is important in understanding the underlying problems in patients with more severe or refractory disease, and may ultimately be useful in distinguishing between disease severity and damage

• While dietary modification and probiotics may play a role in the management of SSc GI disease, more data is needed to determine benefit and identify an optimal approach

• Further studies focused on the benefits of early interventions with promotility agents and the application of novel treatments in SSc GI disease will be important