The ACCESS Study
Characterization of Sarcoidosis in the United States

Michelle Freemer, M.D. and Talmadge E. King Jr., M.D.

University of California San Francisco, San Francisco, California

	ARTICLE

TOP ARTICLE REFERENCES

A recent consensus statement summarized current understanding of sarcoidosis (1). The level of evidence for the recommendations made in that statement was largely based on expert opinion developed by consensus because few large scale studies exist that clarify the epidemiology, cause, pathogenesis, genetic factors, diagnosis, and optimal treatment of sarcoidosis. In response to this paucity of reliable data, the National Heart, Lung, and Blood Institute has sponsored a multicenter studyA Case Control Etiologic Study of Sarcoidosis (ACCESS Study)to determine the etiology, socioeconomic status, and clinical course of patients with sarcoidosis (2). The study organization designated 10 clinical centers, a clinical coordinating center, specialized core laboratories, and a central specimen repository. In addition to study subjects, sites recruited controls (by random digit telephone dialing) matched to cases on age, sex, self-designated race, and geographic location.

In the current issue of the journal (pp. 1885-1889), Baughman and colleagues describe the association of race, age, and sex with organ involvement found in 736 newly diagnosed cases of sarcoidosis enrolled into the ACCESS study (3). Important deficiencies in previous studies were the lack of a precise, consistent case definition and variable methods of disease ascertainment. The ACCESS Research Group attempted to address these deficiencies by (1) establishing a case definition that requires recent tissue confirmation of granulomata and a compatible clinical course, and (2) by developing and implementing a standard instrument that provides criteria for diagnosing organ involvement (see Table E1 in online data supplement) (4).

How good is the case definition? The case definition is clearly defined. However, the requirement for tissue confirmation likely introduces a bias toward more severe disease. For example, less than 1% of the cases had Löfgren's syndrome (usually a clinical diagnosis in a patient with bilateral hilar adenopathy, erythema nodosum, and uveitis), whereas the literature reports its occurrence in 20 to 30% of white individuals (1). In addition, the case definition fails to account for the chronicity of the disease, a factor that affects disease severity and organ involvement.

How good is the ACCESS instrument at defining organ involvement? In a patient with biopsy-proven single organ involvement and no evidence of another potential cause, this instrument is intended to serve as a surrogate for tissue confirmation of other organ involvement. This tool defines features that support the involvement of organs and systems that are commonly affected by sarcoidosis (lung, skin, eyes, liver, and calcium metabolism), organs not usually involved but clinically important (nervous system, kidney, heart), and other sites (nonthoracic lymph nodes, bone marrow, spleen, bone/joint, ear/nose/throat, parotid/salivary glands, muscles). The tool depends upon the clinician's diligence in pursuing evidence for additional organ involvement (4). Consequently, it requires that clinicians pursue a standard method of initial investigation. The ATS consensus statement suggests performing a comprehensive initial evaluation including: history (occupational and environmental exposure, symptoms); physical examination; posteroanterior chest radiography; pulmonary function tests (spirometry and DLCO); peripheral blood counts (white blood cells, red blood cells, platelets); serum chemistries (calcium, liver enzymes [alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase], creatinine, blood urea nitrogen); urine analysis; electrocardiogram; routine ophthalmologic examination; and tuberculin skin test (1).

Is the ACCESS study population representative of sarcoidosis in the United States? Unfortunately, we cannot confidently address this question but it seems unlikely. As the authors noted, the ACCESS study was not designed to determine the incidence or prevalence of sarcoidosis. Cases were recruited from a variety of clinical settings, including inpatient services and outpatient clinics. However, the clinical setting at each center and the specific criteria for case selection are not clear. In addition, the study investigators were all pulmonologists, which essentially limited the patient population to those with lung involvement. The number of subjects who refused to participate in the study and their clinical characteristics is unknown. Importantly, no center was located in the far western or southwestern parts of the country (2). This leads to a potential bias in the recruitment of Hispanics, Asians, and Native Americans. Furthermore, some of the centers are in regions of the country with limited ethnic and racial diversity.

What are the important findings in this study? Previous studies have shown that sarcoidosis most commonly presents in patients 20 to 40 years of age. However, this study shows that there has been a shift of the peak age at presentation to patients over the age of 40 years, especially among women. The explanation for this "delay" in presentation is unknown. The investigators report that only lung involvement was independent of age, sex, and race. However, they did not perform further analysis of pulmonary disease severity based on age, sex, and race.

Sarcoidosis is reported to be more severe in African Americans, whereas white individuals are more likely to present with asymptomatic disease (1, 5). In addition, extrathoracic manifestations are said to be more common in certain populations, such as chronic uveitis in African Americans, lupus pernio in Puerto Ricans, and erythema nodosum in Europeans (1, 5). In the current study, extrathoracic lymphadenopathy, eye, liver, skin (other than erythema nodosum), and bone marrow involvement were more common in African Americans. Abnormal calcium metabolism (hypercalciuria or hypercalcemia) was more frequent in white individuals and in subjects older than 40 years of age. The frequency of erythema nodosum was higher in women but did not differ by race. Extrathoracic lymphadenopathy was more frequent in those individuals younger than 40 years of age.

The importance of the ACCESS Study in determining the etiology of sarcoidosis is obvious. The preliminary analysis of the subjects enrolled in the study shows that this is a potentially representative, though not ideal, cohort to study. The assessment instrument developed by this group appears useful and needs to be validated. If validated it may provide a robust instrument to standardize the identification of organ involvement and thus provide a method to monitor the course of the disease.

	Footnotes

	References

TOP ARTICLE REFERENCES

1. Hunninghake GW, Costabel U, Ando M, et al . . Statement on sarcoidosis. Am J Respir Crit Care Med 1999; 160: 736-755 [Free Full Text].

2. ACCESS Research Group. Design of a case control etiologic study of sarcoidosis (ACCESS). Journal of Clinical Epidemiology 1999;52:1173-1186.

3. Baughman RP, Teirstein AS, Judson MA, et al . . Clinical characteristics of patients in a case control study of sarcoidosis. Am J Respir Crit Care Med 2001; 164: 1885-1889 [Abstract/Free Full Text].

4. Judson MA, Baughman RP, Teirstein AS, Terrin ML, Yeager H, Jr., ACCESS Research Group. Defining organ involvement in sarcoidosis: the ACCESS proposed instrument. Sarcoidosis Vasc Diffuse Lung Dis 1999;16:75-86.

5. Newman LS, Rose CS, Maier LA. Sarcoidosis. N Engl J Med 1997; 336: 1224-1234 [Free Full Text].