In our study, we obtained detailed information on

– the path patients followed to diagnosis;

– the range, frequency, severity and impact of a comprehensive spectrum of oral symptoms and signs experienced;

– the patients’ utilization of ongoing dental care and the burdens associated with that care;

– the treatments (both conventional and alternative) that patients underwent;

– the effects of illness on patients’ general quality of life.

We used self-reported data collected by means of a mailed survey. To minimize the inherent weaknesses of such a design, we recruited subjects to the study through multiple sources with the aim of creating a population representative of people with PSS. We recruited patients with a physician-confirmed diagnosis of primary SS—according to criteria published by the 2002 American-European Consensus Group (AECG)¹⁹—through nine rheumatology and oral medicine practices that have a high volume of patients with SS and expertise in treating this condition. These practices either were known to us as having a high volume of patients with PSS or were recommended to us by the Sjögren’s Syndrome Foundation (SSF), Bethesda, Md. This process yielded a group with known, stringently diagnosed PSS. A large cross-sectional sample of patients with SS also was obtained through the SSF, which has a database of more than 8,500 registered patients with SS. Finally, to create a "healthy" control comparison group, we asked patients from the SSF to provide surveys to friends of an age similar to those of the patients but without a diagnosis of PSS. All subjects, patients and control subjects alike, came from communities throughout the United States.

	SUBJECTS, MATERIALS AND METHODS

TOP ABSTRACT SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION Reference

 
Participants and procedures. We recruited nine physicians at rheumatology or oral medicine clinics, whom either we or the SSF identified as dealing with a high volume of patients with SS, to participate in this study. We asked the nine physicians to identify from their records all patients classified as having PSS according to the 2002 AECG criteria.¹⁹ We asked physicians who had 100 or fewer patients with PSS to recruit all eligible patients for the survey. We asked physicians who had more than 100 eligible patients to select 100 patients at random and recruit them for the study. In total, staff members in the nine offices mailed surveys to 547 patients, a sample we refer to as having "physician-referred PSS" (PhysR-PSS). In addition, SSF staff members mailed surveys to 8,694 active patient-members of the SSF (a sample hereafter referred to as "SSF-PSS"). We asked, via the questionnaire instructions, one-half of these SSF patients to give a survey to a friend of the same age and sex as the patient but who did not have a diagnosis of SS; this group would become a demographically matched comparison group (control subjects).

The study was approved by the Western Institutional Review Board, Olympia, Wash., and, as needed, by the local institutional review boards with which the recruiting physicians were associated. The respondent-completed, paper-based questionnaire was unintrusive and completely anonymous. Additionally, the researchers did not have direct access to the sample, as third parties (staff members of physicians’ offices and the SSF) mailed the surveys to patients. Therefore, the institutional review boards deemed the survey appropriate for use with human subjects and did not require an informed consent form for approval by the Western Institutional Review Board or other university boards from which approval was obtained. We collected data between Jan. 1 and July 31, 2007.

Dry mouth lasting longer than three months was one of the sicca symptoms reported most often.

Survey items Participants completed the self-administered Assessment of Symptoms and Experiences of Sjögren’s Syndrome survey that we developed for this project. It included questions about medical and dental history and the following prevalidated instruments: the Profile of Fatigue and Discomfort–Sicca Symptoms Inventory (PROFAD-SSI),^20,21 Medical Outcomes Study 36-Item Short Form Health Survey (SF-36),²² the Functional Assessment of Chronic Illness Therapy Fatigue scale,²³ the Center of Epidemiologic Studies Depression Scale,²⁴ the Thinking Scale²⁵ and the modified Brief Pain Inventory short form.²⁶ In this article, we report data regarding the oral aspects of PSS, as well as selected information from the PROFAD-SSI and SF-36 results. Other data from the survey will be presented separately (B. Segal and colleagues, unpublished data, July 2007). (The survey instrument is available as supplemental data to the online version of this article [found at "http://jada.ada.org"].)

Data analysis We scored all of the prevalidated instruments according to their original scoring algorithms. We computed univariate analyses of variance (ANOVAs) for all mean score comparisons between the PhysR-PSS, SSF-PSS and control groups. We followed up significant ANOVAs with Fisher least significant difference tests to determine which specific groups differed from one another. We computed ² tests to determine differences between the three groups in the occurrence of events. Likewise, we performed separate ² tests comparing two groups at a time if we noted any significant (P .05) overall ² comparison of the three groups. Finally, to determine the relationship between oral sicca symptoms and quality of life, we conducted Pearson product moment correlation tests between the oral sicca domain of the PROFAD-SSI and the eight quality-of-life domains of the SF-36.

	RESULTS

TOP ABSTRACT SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION Reference

 
Of 547 surveys sent through participating physicians’ offices, respondents returned 281 (51 percent). Of these, four were duplicates, and we excluded them. Of the 8,694 surveys sent to SSF patients, 3,939 (45 percent) were returned. Control subjects returned 630 surveys.

Sample eligibility All participating physician sites confirmed the identification and recruitment of patients with a diagnosis of PSS according to the 2002 AECG classification criteria.¹⁹ Therefore, we considered all 277 patients with PSS to be eligible for the study. Of the 3,939 SSF patients who returned surveys, we classified 1,225 (31.1 percent) as being "likely" to have PSS on the basis of patients’ reported positive labial minor salivary gland biopsy results, anti-SS-A (Ro) or SS-B (La) antibody test results or both, and lack of concurrent diagnoses of other rheumatic conditions (rheumatoid arthritis, systemic lupus erythematosus, mixed connective-tissue disease, myositis or scleroderma). We confirmed 60 percent of the 547 patients with PhysR-PSS as having positive test results for anti-SS-A (Ro) antibodies, anti-SS-B (La) antibodies or both, as determined by the recruiting physicians, and 65 percent of the SSF-PSS group reported having anti-SS-A (Ro) antibodies, anti-SS-B (La) antibodies or both. These demographics are generally similar to expected frequencies of these autoantibodies in patients with this disorder and to results of previous reports regarding cohorts of patients with PSS.^1,13,27–31

Finally, among the 630 surveys received from peer control subjects, we excluded from further analyses surveys from 24 subjects who reported having received a diagnosis of SS or another rheumatic condition. Thus, we based our analyses on data from 606 control subjects, 277 patients with PhysR-PSS and 1,225 SSF-PSS patients.

Demographics Patients with PhysR-PSS, SSF-PSS patients and control subjects were largely similar in terms of demographics (Table 1), and the profile of patients is consistent with that reported in prior large case series.^13,15–18 Patients with PhysR-PSS were, on average, 62 years old; SSF-PSS patients and control subjects were, on average, 61 years old (P > .05). At least 90 percent of the members of each group were women. However, patients in both the PhysR-PSS and SSF-PSS groups were significantly less likely to be currently employed than were control subjects (patients with PhysR-PSS, 38 percent; SSF-PSS patients, 41 percent; control subjects, 49 percent; P < .05).

View this table: [in this window] [in a new window]   TABLE 4 Association between oral sicca severity (PROFAD-SSI*) and quality of life (SF-36) in affected subjects.

	DISCUSSION

TOP ABSTRACT SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION Reference

We obtained data from 277 subjects who had a diagnosis from their physicians of PSS on the basis of the AECG criteria. These are the most widely accepted classification criteria and are considered a stringent means of diagnosis for clinical studies. Additionally, we classified 1,225 patient-members of the SSF who completed the survey as having PSS on the basis of self-reported criteria that corresponded to those used in the AECG. Although we did not confirm the diagnosis of the members of the survey group by examining their medical records, we believe that the marked similarity in responses between the physician-diagnosed cohort and the SSF member cohort argues that the SSF-PSS group represents a "true" population with PSS and one comparable with the PhysR-PSS population. We have reported the data separately for the groups, but they did not differ in terms of any clinically significant variables. One always must view data collected via survey methods and patients’ self-reports with caution; however, as noted above, we did not find marked differences in the two patient groups, either demographically or clinically, compared with information from published case series or prospective or retrospective studies. In particular, age, sex, prevalence of oral complaints and laboratory findings in our study all were similar to previously reported findings.

We recognize that patients who respond to surveys may differ from those who choose not to participate. We have no information on any differences that may exist between the respondents and the nonresponders in this study. However, the response rates (uncorrected rate for patients with PhysR-PSS, 51 percent; uncorrected rate for SSF-PSS patients, 45 percent) were high for mail-survey studies, and the clinical and demographic data generally were as expected for a population with PSS. Furthermore, the data are similar between the larger SSF-PSS group and the more selective PhysR-PSS group. This finding provided us with some confidence that the responding subjects were a fair representation of people with PSS in the United States.

The use of a subject-recruited healthy control group ensured that this cohort was well-matched with the patient groups in terms of demographic (and geographic) variables. Although we excluded potential control subjects who reported having a rheumatic disorder, we included control subjects with any other medical condition and did not control for other potential comorbidities. In other words, we made no attempt to recruit a "super-healthy" group of control subjects. The control subjects had a low prevalence of dry mouth, as was the case in other published studies (a range of prevalence of dry mouth in the general population between 10 and 29 percent^32,33). However, the responses and subsequent prevalence rates for xerostomia depend highly on how the question is asked and how the definition of "dry mouth" is applied. By using the AECG format, according to which we defined xerostomia as "dry mouth lasting for more than three months," we believe we used a more stringent criterion than was used in many other studies. This factor may be an explanation for the relatively low prevalence of xerostomia found among the control subjects.

The results present a compelling argument for the importance of dental health professionals’ increasing their understanding of SS. Dry mouth was the initial symptom experienced by more than one-third of the patients in our study (Table 2). This was the second most common initial complaint, close in frequency to the most commonly cited symptom of dry eyes. In addition to xerostomia, a substantial number of patients with PSS reported tooth or gingival problems or swollen salivary glands as initial complaints. In total, more than one-half of the patients experienced an oral symptom or sign as the first manifestation of the disorder. Increased recognition among practitioners that the oral cavity, head and neck are the primary sites of early symptoms of SS could lead to inclusion of SS in the differential diagnosis for these complaints and could lead to earlier diagnosis.

The need for health professionals to improve their recognition of SS is demonstrated clearly by the length of time required to obtain a diagnosis (Table 2). On average, at least seven years elapsed between the patient’s initial experience of symptoms or signs and the establishment of a definitive diagnosis. Greater awareness of PSS among oral health care practitioners could reduce this delay in diagnosis. As noted above, oral signs and symptoms often are the initial manifestation of PSS. Furthermore, patients saw their dentists more frequently than did control subjects (Table 5). Therefore, dentists have a unique opportunity to start the diagnostic process. The fact that clinicians commonly use oral tests (questions about xerostomia, salivary flow test, lip biopsy) (Table 2) in diagnosing PSS emphasizes the important role that dental professionals could—and should—play in diagnosis of this condition. It is possible that the low percentage of dentists responsible for a diagnosis reported in Table 2 is an underestimation. Subjects were asked which health care provider made the diagnosis, not which health care provider suggested the subject might have PSS. Primary care practitioners may refer patients to specialists (rheumatologists or oral medicine specialists) to make the final diagnosis.

The results presented in Table 3 show the effect of PSS on the oral cavity and oral functions. Virtually all patients with PSS experience oral symptoms, with about 90 percent reporting persistent xerostomia. The majority of patients with PSS also had difficulties with eating, swallowing, speaking and caries. The pervasive effects of xerostomia on oral comfort and functions are well-recognized,³¹ and for most patients with PSS these are persistent and significant. That is borne out by the results of the PROFAD-SSI. The significant association of the PROFAD-SSI with all domains of the SF-36 (Table 4) demonstrates that the effect of PSS is widespread and severe. In this light, the significantly lower employment rate among patients with PSS (Table 1) was an interesting finding and may be a reflection of the burden that the illness places on them.

Recently, Stewart and colleagues⁵ reported about quality of life in a group of 39 patients with SS and the condition’s effect on their general health. The investigators found pervasive effects of these patients’ oral involvement, and they concluded that SS had major effects on systemic health and quality of life. Results from our study greatly expand on and reinforce these findings, because we examined a broader range of signs and symptoms in a larger patient group.

PSS symptoms are of sufficient severity that the majority of patients seek treatment for their oral condition. As shown in Table 5, almost one-half of patients with PSS reported using, at the time they completed the survey, oral comfort agents or supplemental fluoride. The availability of agents to reduce dry-mouth symptoms by increasing salivary output has had an effect on PSS treatment. More than one-half of patients had tried a prescription parasympathomimetic secretagogue, and more than one-third reported that they used this therapeutic option at the time they completed the survey. There also was widespread use of alternative and complementary therapies, suggesting that allopathic options are not adequate for many patients. Such widespread use also points out the importance, when taking a medication history, of asking patients specifically about their use of any alternative or complementary therapies.

A further measure of the effect of PSS is the data regarding dental care utilization in Table 5. Dental care for patients with PSS is sought frequently and is expensive. To attempt to maintain oral health and comfort, patients saw their dentists about twice as often as did control subjects. In comparison with control subjects, they had experienced about three times the amount of dental disease in the preceding year and expended about three times as much for their care. This is a continuing problem and burden. To our knowledge, these data are the first to document the economic impact of PSS, which is manifested through increased oral health care costs and reduced employment. A surprisingly low percentage of both the patients and the control subjects reported ever having experienced tooth loss or dental caries. Subjects may have responded to these questions on the basis of their experiences in the previous year, about which the survey also inquired, and not of their lifetime experience. We do not have a clear explanation for these data but note that they are consistent internally between groups.

	CONCLUSION

TOP ABSTRACT SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION Reference

 
Investigators have published numerous reports of various oral aspects of SS.^5–12 However, most have examined relatively small groups or focused on a particular aspect of the condition. We believe our report is the first to evaluate a large number of subjects (> 1,000) and the first to attempt to capture the full spectrum of the condition’s oral signs and symptoms, as well as its effect on patients’ general health and well-being. Our results clearly demonstrate the marked effects of PSS and also present a clear challenge to oral health care professionals. The results suggest that dentists often see patients with PSS early in their disease course; indeed, they may be the first health care practitioners from whom a patient with PSS seeks care. Dentists have an opportunity to recognize the possibility of SS early and facilitate a more rapid diagnosis of the condition.

Along with earlier diagnosis comes the possibility of earlier intervention. With careful management and use of appropriate preventive measures, many of the negative health consequences of PSS can be minimized or eliminated. This, in turn, could reduce the economic and general health burdens seen in patients with this condition. Greater awareness and recognition of SS in the dental office are critical.