Application of a Combined Protocol for Rational Request and Utilization of Antibody Assays Improves Clinical Diagnostic Efficacy in Autoimmune Rheumatic Disease
By Tampoia, Marilina; Brescia, Vincenzo; Fontana, Antonietta; Zucano, Antonietta; Et al
* Context.-Because of a marked increase in the number of requests for antinuclear antibodies, anti-extractable nuclear antigen antibodies, and anti-double-stranded DNA antibodies for the diagnosis of autoimmune rheumatic disease, guidelines have been proposed for their appropriate use.
Objective.-To evaluate in terms of clinical efficacy and cost- benefit ratio the outcome of applying a protocol for the diagnosis of autoimmune rheumatic disease.
Design.-A diagnostic protocol for the rational utilization of second-level tests (anti-extractable nuclear antigen antibodies and anti-double-stranded DNA antibodies) was applied at Hospital Polyclinic beginning January 2004. The appropriateness of 685 consecutive requests received at the clinical pathology laboratory from January to June 2004 was assessed. Patients who underwent these laboratory tests were followed up for 12 months after blood sample drawing.
Results.-Introduction of the protocol led to a signifi- cant reduction in the number of second-level tests prescribed (27.9% vs 49.5% for anti-extractable nuclear antigen antibodies; 27.5% vs 56.6% for anti-double-stranded DNA antibodies). After the period of observation, none of the 163 patients who had negative results on the firstlevel test and were asymptomatic, for whom second-level tests had not therefore been performed, were found to have autoimmune rheumatic disease. In 90.5% (77/85) of patients positive for the second-level tests, clinical confirmation of autoimmune rheumatic disease was obtained.
Conclusions.-Not only did application of the diagnostic protocol reduce the number of second-level tests performed but it also increased their specificity. Our data thus indicate that the use of shared guidelines by clinical and laboratory specialists yields satisfactory results.
(Arch Pathol Lab Med. 2007;131:112-116)
Serologic tests for autoantibodies are essential in the diagnosis and treatment of patients affected by autoimmune rheumatic disease. Those most commonly prescribed include antinuclear antibodies (ANA), anti-extractable nuclear antigen antibodies (anti-ENA), and anti- double-stranded DNA antibodies (anti-dsDNA).1,2 Antinuclear autoantibodies performed by indirect immunofluorescence is the first- level test advised for use in patients with symptoms and signs suggestive of autoimmune rheumatic disease.1,2
In the presence of positive ANA indirect immunofluorescence results, the direct specific antibody titers against the different nuclear antigens (anti-ENA and anti-dsDNA) are assayed and thus constitute the second-level tests.1,2
Anti-ENA titers are used for the diagnosis of systemic lupus erythematosus (SLE), neonatal lupus, Sjgren syndrome (SS), mixed connective tissue disease (MCTD), polymyositis/dermatomyositis (PM/ DM), and progressive systemic sclerosis (PSS), with variable sensitivity and specificity.3
Anti-dsDNA is useful for the diagnosis and monitoring of SLE.1,2 However, in approximately 1% of patients affected by autoimmune disease, especially SS and PM/DM, the ANA indirect immunofluorescence test can yield negative results, and so a well- founded clinical suspicion could justify the request for the higher level tests.4,5
In recent years, the number of requests for such tests has markedly increased, unfortunately, partly because of inappropriate prescriptions, often related to an inadequate knowledge of the analytical performance of the tests, poor compliance to the guidelines, and some laboratory organizational aspects (forms favoring the prescription of blocks of tests, preset test profiles, delayed response times, lack of communication/sharing of available evidence, insufficient consultation activities).6,7
Innumerable strategies have been proposed to regulate the demand: continual education and the dissemination of guidelines, modification of the request forms, and change in economic policies for laboratory tests.6 Nevertheless, no relevant data are available on long-term clinical-diagnostic assessment of such strategies.
The aim of our work was to apply a diagnostic algorithm for autoimmune rheumatic diseases, complying with previously published guidelines, and to appraise any advantages observed with its use, in terms of clinical ef- ficacy and cost-benefit ratio.
MATERIALS AND METHODS
Patients
Beginning in January 2004, a diagnostic protocol formulated on the guidelines for the diagnosis of autoimmune rheumatic diseases1,2 was made operative at Hospital Polyclinic. It aimed to optimize the request and utilization of second-level tests.
The appropriateness of the requests for second-level tests was assessed in 685 consecutive request forms received at the clinical pathology laboratory from January to June 2004, arriving both from specialist rheumatologic wards (university and hospital rheumatology clinics, internal medicine wards treating immunologic diseases) and from specialist nonrheumatologic wards (dermatology, nephrology, neurology, gastroenterology, ophthalmology, pneumology, hematology).
In collaboration with the hospital rheumatologic specialists, an additional request form was set up and distributed to all clinicians working for the hospital (Figure 1).
In addition to specifics for identifying the sample and reasons for the request (diagnosis or monitoring of disease), the form posed a clinical inquiry about the presence of the signs and symptoms included in the clinical criteria classifying the principal autoimmune rheumatic diseases: SS,8 SLE,9 PSS,10 MCTD, and DM/PM.11
Before applying the protocol it was judged necessary to hold meetings with the directors and operators in the wards involved in the study to illustrate all the possible advantages and overcome the inevitable resistance because of the need to include a greater quantity of clinical data.
The protocol prescribed the utilization of second-level tests (anti-dsDNA and anti-ENA) for all samples positive to the firstlevel tests (ANA) and, using the additional request form, for negative samples only in the presence of signs or symptoms suggestive of autoimmune rheumatic disease, that is, those included in the criteria classifying the principal diseases. For samples negative for the ANA test lacking the additional form and/or with a form indicating only systemic inflammatory symptoms and signs (ie, fever, fatigue, elevated erythrocyte sedimentation rate, elevated C reactive protein), second-level tests should not be performed.
Assessment of the reported signs and symptoms resulted in the categorization of 3 groups of requests:
1. Only systemic inflammatory symptoms and signs (ie, fever, fatigue, elevated erythrocyte sedimentation rate, elevated C reactive protein).
2. Only 1 symptom or sign included in the classification criteria.
3. Two or more symptoms and/or signs.
Finally, after 1 year of observation, the laboratory and clinical data of the patients in whom second-level tests had not been performed were reviewed to verify any cases of onset of autoimmune rheumatic disease.
Methods
The search for ANA was conducted by indirect immunofluorescence, using API-Plus slide preparers and HEp-2 cells as substrate (Euroimmun, Lubecca, Germany), starting with an initial dilution of 1:80, and adding conjugated fluoresceinated human anti- immunoglobulin G.
The search for anti-dsDNA was conducted with the immuno- fluoroenzymatic method, completely automated on a UNICUP 100 analyzer (Pharmacia Diagnostics, Freiburg, Germany), that uses sensitized polystyrene microwells with plasmid doublestranded recombinant DNA antigen. The search for anti-ENA was conducted with the immmunoenzymatic method, on the ETIMAX 3000 analyser (DiaSorin, Saluggia, Italy), that uses microplates coated with the following antigens: SS-A, SS-B, Sm, Sm- RNP, Scl-70, and Jo-1 extracted from calf thymus and recombinant CENP B baculovirus.
Statistical Analysis
Statistical analysis was conducted using the chi-squared test for intergroup comparison. A value of P β .05 was taken to be statistically significant.
RESULTS
Analysis of the clinical data reported in the 685 request forms showed that the symptoms most commonly reported were joint pain (34%) and systemic inflammatory symptoms (32%) (Table 1).
Distribution of the Different Request Types
There was a statistically significant difference in the distribution of the 3 request groups according to whether the request was made by a specialist rheumatologic ward or a specialist nonrheumatologic ward. In fact, a lower percentage of requests indicating only systemic inflammatory symptoms and signs was made by specialist rheumatologic wards (group 1: 16% vs 53%; P < .001) (Figure 2, A) and a higher percentage of requests indicating 2 or more specific symptoms and/or signs (group 3: 62% vs 20%; P < .001) (Figure 2, B); no significant difference was observed between the 2 types of ward with regard to requests indicating only 1 specific symptom or sign (group 2: 20% vs 27%; P = .24) (Figure 2, C).
Positivity to First-Level and Second-Level Tests
The percentage of positivity to the first-level test (ANA) was not significantly different according to the request group (P = .43) coming from specialist rheumatologic wards; instead, it was significantly higher in req\uests indicating 2 or more specific symptoms and/or signs (P = .002) coming from specialist nonrheumatologic wards (Figure 3, A).
The percentage of positivity to second-level tests increased to a significant degree in requests indicating 2 or more specific symptoms and/or signs, coming both from specialist rheumatologic wards (P = .001) and specialist nonrheumatologic wards (P < .001) (Figure 3, B).
Assessment of the 685 requests analyzed showed that the percentage of positivity to the first-level test (ANA) was higher in requests reporting 2 or more specific symptoms and/or signs (53.6%) than in those indicating a single specific symptom (33.3%) or only systemic inflammatory symptoms and signs (30%). The percentage of positivity to second-level tests showed a similar distribution (49.6% positivity for requests reporting 2 or more specific symptoms and/or signs, 34.5% for requests with a single sign or symptom, and 35.7% for requests with systemic inflammatory symptoms) (Figure 4).
Reduction of the Number of Tests
Application of the diagnostic protocol resulted in a decrease in the number of second-level tests performed but no change in the number of first-level tests (Figure 5). In fact, before introduction of the protocol (reference year 2002), ANA tests were included in 99.1% of the requests, anti-ENA tests in 49.5%, and anti-dsDNA tests in 56.6%; after the introduction of the diagnostic protocol (year 2004), despite the same number of requests for the ANA test (P = .56), a significant reduction was observed in requests for the anti- ENA (27.9% vs 49.5%; P = .001) and anti-dsDNA (27.5% vs 56.6%; P = .001) tests.
Clinical Follow-up
None of the 163 patients with negative results to the first- level test and without symptoms suggestive of autoimmune disease developed specific symptoms and/or signs of autoimmune rheumatic disease during the 1-year observation clinical follow-up period. Of the 194 patients with positive results to the first-level test and negative results to the second-level tests, 50 patients (25.7%) were diagnosed with autoimmune rheumatic disease (14 SLE, 9 SS, 10 PSS, 3 DM/PM, 14 undifferentiated connective tissue disease), 4 (2%) patients had rheumatoid arthritis, and 11 (5.6%) patients had organ- specific autoimmune disease (1 autoimmune thyroiditis, 5 type I autoimmune hepatitis, 5 primary biliary cirrhosis). A diagnosis of autoimmune rheumatic disease was made in 90.5% (77/85) of the patients with positive results to both first-level and second-level tests (27 SLE, 17 SS, 28 PSS, 4 MCTD) or with positive results to only the second-level tests (1 PM/DM) (Table 2).
In fact, of the 243 patients with negative results to the ANA test, but with a request form indicating a single specific symptom or sign, as well as those indicating 2 or more symptoms and/or signs who underwent second-level ENA/dsDNA tests, only 1 patient showed positive results to second-level tests, anti-ENA (anti-Jo-1) (0.41%).
The remaining patients, all with negative results to the second- level tests, underwent clinical follow-up.
COMMENT
Many studies have been published assessing methods aimed at obtaining a more appropriate use of laboratory tests and containing the number of requests for costly tests. It seems clear that some of the ways to reduce the number of unnecessary tests include promoting appropriate test prescriptions, fostering communication between clinicians and laboratory experts, developing decisionmaking support systems using shared guidelines, and ensuring continuous education.12
Our study assessed the clinical-management advantages of the application of a diagnostic protocol, developed by implementing guidelines reported in the literature, for appropriate use of laboratory tests to diagnose autoimmune rheumatic disease. The protocol required specification of clinical data referred on the request form, to help select samples in which second-level tests (anti-ENA and anti-dsDNA) were appropriate.
Assessment of the request forms showed that specialist nonrheumatologic wards prescribe first-level and secondlevel autoantibody tests even in patients with only systemic inflammatory signs that are not always related to autoimmune rheumatic disease. This approach could be interpreted as an attempt to avoid underestimation of symptoms and signs of onset of a severe autoimmune disease.
After application of the protocol, although the percentage of positivity to the ANA test (first level) was not significantly different according to the request group type, the percentage of positivity to the anti-ENA and antidsDNA tests (second level) in patients positive to the firstlevel test increased significantly in the group of requests indicating 2 or more specific signs and/or symptoms of autoimmune disease, coming both from specialist rheumatologic wards and specialist nonrheumatologic wards.
The clinical data reported in the form did not affect the utilization of the first-level test (ANA), implemented as an ”open door testing service,”13 because these were necessary to obtain a homogeneous diagnostic approach to the selection of patient samples in which to perform higher level tests. In absence of clinical data suggestive of disease and negativity to the first-level test, utilization of the anti- ENA and anti-dsDNA tests was not considered appropriate. This determination was implemented as a part of a ”gated testing service.”14
Instead, in patients with negative results to the ANA test, but with clinical signs and symptoms suggestive of autoimmune rheumatic disease, utilization of the secondlevel tests was considered necessary, and in a single case the second-level tests performed yielded positivity to anti- Jo-1, suggesting the diagnosis of dermatopolymyositis.
In reality, the low frequency of positive results to second- level tests among ANA-negative patients with clinical signs would not justify the performance of the tests. After determining that the request form lacked a clear association between the different autoimmune rheumatic diseases and their accompanying signs and symptoms, we revised the form to report the symptoms and signs subdivided by specific disease.
Observation of the cases and clinical review at 1 year did not reveal the onset of autoimmune rheumatic disease in any patient with negative ANA results in whom second- level tests had not been performed.
In accordance with reports in the literature6,15 we found that a combination of several factors (implementing guidelines, modifying the request form method, highlighting economic advantages) represents the most efficacious strategy for improving the appropriateness of laboratory tests.
The application of the protocol reduced the number of second- level tests performed, with obvious cost-benefit advantages for the laboratory budget in terms of both reagents and staff time, without reducing the diagnostic ef- ficacy of the tests. Accurate preselection of patients in whom higher level tests needed to be performed was able not only to contain the number of second-level tests performed but also to increase their specificity.
The clinicians involved appreciated the advantages resulting from the introduction of the protocol and have continued to use it after the end of the study period.
Interpretation and assessment of the clinical data provided to the laboratory expert must be considered as ”clinical-laboratory interface” activities that can improve the quality of care and medical outcome.
References
1. Kavanaugh A, Tomar R, Reveille J, et al. Guidelines for clinical use of the antinuclear antibody test and tests for specific autoantibodies to nuclear antigens. Arch Pathol Lab Med. 2000;124:71- 81.
2. Tozzoli R, Bizzaro N, Tonutti E, et al. Guidelines for the laboratory use of autoantibody tests in the diagnosis and monitoring of autoimmune rheumatic diseases. Am J Pathol. 2002;117:316-324.
3. Bizzaro N, Wiik AS. Appropriateness in anti-nuclear antibody testing: from clinical request to strategic laboratory practice. Clin Exp Rheumatol. 2004;22: 349-355.
4. Bylund DJ, Nakamura RM. Importance of detection of SS-A/Ro autoantibody screening immunofluorescence tests for autoantibodies to nuclear antigens. J Clin Lab Anal. 1991;5:212-218.
5. Menard HA. Antinuclear antibodies: the medium is the message. Clin Exp Rheumatol. 2000;18:429-430.
6. Walraven C, Naylor CD. Do we know what inappropriate laboratory utilization is? JAMA. 1998;280:550-558.
7. Hindmarsh JT, Lyon AW. Strategies to promote rational clinical chemistry test utilization. Clin Biochem. 1996;29:291-299.
8. Vitali C, Bombardieri S, Moutsopoulos HM, et al. Preliminary criteria for the classification of Sjogren’s syndrome: results of a prospective concerted action supported by the European Community. Arthritis Rheum. 1993;36:340-347.
9. Hochberg MC. Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum. 1997;40:1725.
10. Subcommittee for scleroderma criteria of the American Rheumatism Association diagnostic and therapeutic criteria committee: preliminary criteria for the classification of systemic sclerosis (scleroderma). Arthritis Rheum. 1980;23: 581-590.
11. Tanimoto K, Nakano K, Kano S, et al. Classification criteria for polymyositis and dermatomyositis. J Rheumatol. 1995;22:668-674.
12. Plebani M. The clinical importance of laboratory reasoning. Clin Chim Acta. 1999;280:35-45.
13. Lock RJ. Rational requesting or rationing testing. J Clin Pathol. 2004;57: 121-122.
14. Sinclair D, Saas M, Stevens JM. The effect of a symptom related ”gating policy” on ANCA requests in routine clinical practice. J Clin Pathol. 2004;57:31- 34.
15. Solomon DH, Hashimoto H, Daltroy L, et al. Techniques to improve physicians’ use of diagnostic tests: a new conceptual framework. JAMA. 1998;280: 2020-2027.
Marilina Tampoia, MD; Vincenzo Brescia, MD; Antonietta Fontana, MD; Antonietta Zucano, PhD; Luigi Fr\ancesco Morrone, MD; Nicola Pansini, MD
Accepted for publication May 26, 2006.
From the Department of Clinical Pathology I, Hospital Polyclinic of Bari, Bari, Italy (Drs Tampoia, Brescia, Fontana, Zucano, and Pansini); and the Department of Nephrology, University of Bari, Bari, Italy (Dr Morrone).
The authors have no relevant financial interest in the products or companies described in this article.
Reprints: Marilina Tampoia, MD, Unita` Operativa di Patologia Clinica I, Policlinico di Bari, Piazza G. Cesare, 11, 70100 Bari, Italia (e-mail: mtampoia@libero.it; m.tampoia@fastwebnet.it).
Copyright College of American Pathologists Jan 2007
(c) 2007 Archives of Pathology & Laboratory Medicine. Provided by ProQuest Information and Learning. All rights Reserved.