Abstract
Objective. A subset of patients presenting with sicca features suggestive of primary Sjögren’s syndrome (pSS) do not fulfill diagnostic or histopathological criteria. This presentation was previously designated as dry eyes and mouth syndrome (DEMS) or sicca asthenia polyalgia syndrome (SAPS). We sought to define the underlying clinical, laboratory, and histological features of these patients.
Methods. The study population consisted of 27 consecutive patients with DEMS/SAPS; 54 patients with pSS served as controls. Medical charts were retrospectively evaluated for clinical and serological data and frozen sera were tested for the presence of antibodies against HIV, hepatitis C virus, and thyroid antigens. Immunohistochemical analysis of paraffin embedded tissues was also performed.
Results. Sicca symptoms and nonspecific musculoskeletal pain were the commonest clinical features of patients with DEMS/SAPS; positive titers of antibodies against thyroid peroxidase was the main underlying abnormality found in 16 out of 27 (59.2%) of patients with DEMS/SAPS compared to 11 out of 54 (20.4%) of pSS controls (p = 0.0009). Histological analysis of the minor salivary gland (MSG) biopsies of patients with DEMS/SAPS disclosed a mild inflammatory infiltration of the interstitial tissue with a predominantly perivascular distribution.
Conclusion. Patients with DEMS/SAPS present with sicca features and nonspecific musculoskeletal complaints, have high prevalence of antithyroid antibodies, and their MSG biopsies demonstrate a mild interstitial lymphocytic infiltration with a predominantly perivascular distribution. In the setting of clinical practice, we propose that in the presence of DEMS/SAPS testing for antithyroid antibody should be performed.
Symptoms of dry eyes and/or mouth, the so-called sicca syndrome, are among the most common symptoms for rheumatology consultation in everyday clinical practice. The differential diagnosis of dry eyes and/or mouth is quite extensive and includes drug side effects, diabetes mellitus, viral infections [human immunodeficiency virus (HIV), hepatitis C virus (HCV)], sarcoidosis, lymphoma, head and neck irradiation, and Sjögren’s syndrome (SS)1.
A subset of patients with sicca features without fulfilling the diagnostic or histopathological criteria for primary SS has been described and designated as dry eyes and mouth syndrome (DEMS) or sicca asthenia polyalgia syndrome (SAPS)2–4. Patients with DEMS/SAPS are characterized by the presence of nonspecific clinical features resembling those found in fibromyalgia or chronic fatigue syndrome, antinuclear antibodies (ANA), hypothyroidism, and moderate lymphocytic infiltration in minor salivary glands (MSG)2–4. Due to the diversity of symptoms/signs it is highly likely that such patients constitute a heterogeneous group rather than a single disease entity.
Given that antithyroid antibodies are frequently encountered in ANA positive individuals5, and sicca features have been previously described in patients with autoimmune thyroiditis6, we hypothesized that thyroid autoimmunity might be the underlying abnormality at least in a subset of patients with DEMS/SAPS. In this context, we sought to present the clinical features, immunological profile (including antibodies against thyroid antigens), and histopathological features of patients with DEMS/SAPS and compare them with those of an age-and-sex-matched group of patients with pSS.
MATERIALS AND METHODS
Patients
The study population (DEMS/SAPS cases) was recruited between 1998 to 2007 in our outpatient Rheumatology Clinic according to the following criteria: (1) presence of sicca symptoms; (2) presence of mild lymphocytic infiltrates in MSG biopsies without fulfilling histopathological criteria for pSS7; (3) absence of another underlying autoimmune or inflammatory disorder; and (4) absence of diagnostic SS criteria7. In addition, 54 consecutive patients with pSS, according to the American-European classification criteria7, who were matched for age, sex, and disease duration with the DEMS/SAPS cases, composed the control group. Both DEMS/SAPS cases and pSS controls were followed in the same period of time and were investigated according to a consistent protocol (described below). Frozen sera from patients with DEMS/SAPS and pSS, as well as paraffin embedded MSG tissues from 27 patients with DEMS/SAPS and 2 healthy volunteers were available for further evaluation. The prevalence of antithyroid positivity was also tested in a cohort of 75 healthy controls. Patients and controls provided informed consent prior to participation in the study.
Clinical and laboratory evaluation
Medical charts were evaluated for the presence of symptoms and signs related to systemic autoimmune disorders. Sicca features were elicited by validated questionnaires8, and objective evaluation of ocular dryness (Schirmer I test, Rose Bengal stain of the conjunctivae) was recorded. Conditions that could contribute to sicca symptomatology, such as diabetes mellitus, drug side effects, preexisting lymphoma, previous head and neck irradiation, or sarcoidosis were not present in any of the patients7.
Laboratory data, including routine serological and immunological studies, were also available in all patients’ medical files. The latter included rheumatoid factor (RF), ANA, antibodies to extractable nuclear antigens (anti-Ro, anti-La, anti-Sm, anti-U1RNP), and IgG antibodies to double-stranded DNA. Additionally, patients’ and controls’ frozen sera were further tested for antibodies to thyroid peroxidase (anti-TPO) and to thyroglobulin (anti-TG) using commercially available kits (by radioimmunoassay and ELISA, respectively). Testing for thyroid stimulating hormone (TSH) (by radioimmunoassay; normal limits between 0.4–5.00IU/ml) and for HIV and HCV infection was also performed.
MSG biopsies
All H&E stained MSG biopsies were reassessed by an expert (FNS) blind to clinical data. Immunohistochemical analysis of paraffin embedded tissues — using specific monoclonal or polyclonal mouse antibodies against CD20 (B cell marker, DAKO), CD3 (T-cell marker, DAKO), CD8 (cytotoxic/suppressor T cells, Novocastra), CD4 (helper/inducer T cells, OPD4, DAKO) and CD68 (macrophages, DAKO) was performed by the avidin-biotin immunoperoxidase technique.
Statistics
Comparison of frequencies between the DEMS/SAPS and pSS group was performed using 2-sided Fisher’s exact test. Comparison of the antithyroid antibody levels between the DEMS/SAPS and pSS groups was performed by Mann-Whitney test.
RESULTS
Descriptive analysis of the DEMS/SAPS group
The DEMS/SAPS group consisted of 27 patients (2 men and 25 women). The mean age of the DEMS/SAPS group was 52.9 ± 12.8 years. None of them fulfilled the classification criteria for either primary or secondary SS7.
The clinical and laboratory profile as well as results of sicca evaluation are presented in Table 1. Coexistence of sicca oral and ocular symptoms was present in 21 out of 27 patients (77.8%), while 3 had xerostomia alone (11.1%) and 3 (11.1%) isolated dryness of the eyes. Abnormal Schirmer I test was evident in 9 out of the 24 patients tested (37.5%), while Rose Bengal stain of the conjunctiva, suggestive of keratoconjunctivitis sicca, was present in 9 out of the 23 patients tested (39.1%). Routine laboratory evaluation was unrevealing for the great majority of the patients, including viral testing for HIV and HCV antibodies. Immunology testing revealed ANA and RF positivity in 12 (44.5%) and 3 (11%) patients, respectively. Fifteen patients (55.5%) had positive titers of anti-TPO antibodies and 3 (11.1%) of anti-TG antibodies, while 6 of them had positive titers for both specificities (22%). Thyroid function was normal, as reflected by normal TSH values (data not shown).
Comparative analysis with pSS
Table 2 illustrates clinical and serological comparative data between DEMS/SAPS and primary SS groups. The prevalence of subjective oral and ocular dryness was higher in the pSS group (88.8% vs 100%, p = 0.03). While keratoconjunctivitis sicca as detected by Rose Bengal stain did not differ significantly between the 2 groups (39.1% vs 54.5%, p = 0.30), abnormal Schirmer I test was found less frequently in patients with DEMS/SAPS than in those with pSS (37.5% vs 86.0%, p = 0.001). In addition, no differences in the prevalence of joint involvement and Raynaud’s phenomenon were detected (48.1% vs 61.1%, p = 0.34 and 14.8 % vs 16.6%, p = 1.0, respectively). Extraglandular features, such as pulmonary, liver, and renal involvement as well as vasculitis and lymphoma were not present in the DEMS/SAPS group (data not shown). In regard to immunological profile, the prevalence of ANA and RF positivity was lower in DEMS/SAPS compared to patients with primary SS (44.5% vs 88.9%, p = 0.00003, 11.1% vs 42.6%, p = 0.005, respectively). Interestingly, patients in the DEMS/SAPS group had a significantly higher prevalence of anti-TPO antibodies (59.2% vs 20.4%, p = 0.0009), but not of anti-Tg antibodies (22.2% vs 18.5%, p = 0.77), while the corresponding values in a cohort of 75 tested healthy controls were 10.7% for both reactivities. In Table 3, comparison of the median and range of anti-TPO and anti-Tg antibodies in the DEMS/SAPS and SS groups is also presented.
Histopathology of the MSG in patients with DEMS/SAPS
Histological analysis of H&E sections of MSG biopsies of the patients with DEMS disclosed scattered lymphocytes and some occasional focal (< 50 lymphocytes) infiltrates. In contrast to pSS, which is characterized by periductal and lymphoepithelial infiltrates with focus score > 1.0, MSG biopsies from patients with DEMS/SAPS demonstrated a predominantly perivascular lymphocytic infiltration with focus score < 1.0 (Figure 1A, B).
Immunohistochemistry was performed in serial sections of 5 μm. The scattered lymphocytes were almost mainly CD3+ of the CD4+ subset. The perivascular focal infiltrates consisted in their great majority of T lymphocytes (mainly CD4+) and surrounded a smaller number of B cells. No abnormalities were noted in the MSG biopsies from the 2 healthy volunteers (Figure 2A–D). Macrophages (CD68+) were occasionally seen in areas of infiltrations.
DISCUSSION
In our study, we have attempted to identify distinctive clinical and serological features in patients with DEMS/SAPS. Sicca features, nonspecific musculoskeletal pain, and Raynaud’s phenomenon were the commonest symptoms, and antibodies against thyroid antigens were the most prevalent underlying finding, suggesting the autoimmune nature of this often-encountered clinical syndrome in rheumatology outpatient clinics. Although sicca features have been reported in the setting of autoimmune thyroid disease, it is unclear whether they are directly related to the disease itself or to concomitant SS5,9.
When compared with the pSS group, patients with DEMS/SAPS have shown a similar prevalence in the occurrence of subjective dry mouth symptoms and keratoconjunctivitis sicca, musculoskeletal complaints, and Raynaud’s phenomenon. However, severe extraglandular features, such as parenchymal involvement, vasculitis, and lymphoma known to occur in the setting of pSS, were not present in our cohort of patients with DEMS/SAPS1. Compared to the prevalence of antithyroid antibodies in SS reported by us and others (20–30%)10, patients with DEMS/SAPS had a significantly higher prevalence of anti-TPO antibodies; they were present in approximately two-thirds of patients. It should, however, be acknowledged that the very high frequency of thyroid autoimmunity in our cohort could be attributed to the pattern of referral (to a referral autoimmune disease center) or to possible ascertainment bias due to the retrospective nature of the study.
Nevertheless, these findings highlight the importance of testing for underlying thyroid autoimmunity in patients presenting with sicca manifestations after known causes, including SS, are cautiously excluded and provides insights for the underlying diagnosis in the great majority of patients with DEMS/SAPS. As far as the remaining 40%, which were mainly characterized by fibromyalgia-like features previously associated with sicca manifestations11, no underlying abnormality could be detected after extensive investigations. However, in these patients, the possibility of a “seronegative” subset of autoimmune thyroiditis, well recognized in the literature, could not be excluded since thyroid ultrasound, which can detect areas of thyroid hypoechogenicity relating to regional lymphocytic infiltration, was not performed. Thyroid status at the time of study entry was also assessed by measuring TSH levels. A great proportion of our patients had already been given thyroxine by their referring physician, giving a potential explanation for the normal TSH values in our patient cohort.
Histopathological analysis of MSG biopsies of patients with DEMS/SAPS disclosed the presence of perivascular infiltrates consisting mainly of cells of the T lymphocyte helper/inducer subset surrounding a smaller, but significant, number of B cells. This perivascular distribution is similar, although less intense, to that observed in those of SLE/SS overlap patients13. Of interest, brain autopsies from patients with Hashimoto’s encephalopathy had comparable findings14, suggesting a possible common vasculopathic denominator in these entities. However, such mild lymphocytic infiltration cannot adequately explain the occurrence of sicca features. It is possible that in the presence of an autoimmune condition, such as autoimmune thyroiditis, several natural autoantibodies15 with anti-epithelial cell or anti-muscarinic16 properties are upregulated and could cause significant impairment in the saliva production or secretion resulting in the development of sicca symptoms.
For the first time, we report a high prevalence of underlying thyroid autoimmune process in patients presenting with DEMS/SAPS. In the setting of clinical practice, we propose that in the presence of sicca features suggestive of SS, but without fulfilling specific classification criteria, antibody testing to thyroid antigens should be performed, after other secondary causes are carefully excluded.
Acknowledgments
We thank Dr. J. Tsonis for collecting the clinical/serological data and Mr. D. Liakos for technical assistance.
Footnotes
- Accepted for publication March 5, 2009.