- Email: [email protected]
Correlation between diagnosis and results of serologic tests for antiglobulin antibodies
Correlation Between Diagnosis and Results of Serologic Tests for Antiglobulin Antibodies
MARION Richmond.
WALLER,
Ph.D
Virginia
From the Division of Connective Tissue Diseases, Department of Medicine, Medical College of Virginia School of Medicine, Virginia Commonwealth University, Richmond, Virginia. This investigation was supported in part by Research Grant AM 04549 from the National Institutes of Health. This is publication No. 58 from the Charles W. Thomas Arthritis Fund, Medical College of Virginia. Requests for reprints should be addressed to Dr. Marion Wailer, Department of Medicine, Medical College of Virginia, Richmond, Virginia 23219. Manuscript accepted January 25, 1973.
A study of 1,320 serum samples for rheumatoid factors and serum agglutinators revealed poor correlation between test results for rheumatoid factors and the presence of rheumatoid arthritis in hospitalized patients with a variety of different diseases. Although the correlation between test results for serum agglutinators and the presence of suppurative infection was good in hospitalized patients, correlation was very poor in less sick ambulatory patients. These studies reveal that correlation of serologic test results with specific disease states may be profoundly altered by severe systemic disease of diverse etiology. Serologic tests for the demonstration of antibody reactivity to heterologous antigens are sometimes useful as diagnostic aids. The presence of antibodies in low titer is usually characteristic of normal serum; elevated titers are frequently associated with a specific disease entity. The Paul-Bunnell test for infectious mononucleosis is probably the best known example of this type of test. associated with a renewed interest in In recent years, immunology, various tests have been developed that have proved valuable in confirming the presence of certain connective tissue diseases. The tests for antinuclear antibodies detect heterologous nuclear antigens that are widespread in nature, and the tests for rheumatoid factors detect antigens present on most mammalian immunoglobulin G (IgG) [1,2]. Rheumatoid factors are present in a variety of disease states and in normal persons, usually in low titer. High titers of rheumatoid factors are almost exclusively associated with rheumatoid arthritis [3]. A second group of antiglobulin antibodies, called serum agglutinators, distinct from rheumatoid factors has attracted considerable interest. Unlike the antibodies mentioned, these antibodies are not “humoral heterophils.” The humoral heterophils, as their name implies, are associated with diverse antigens in different species. The serum agglutinators are antiglobulin (anti-Fab IgG) antibodies presumably present in the serums of all mammals. High titers of these antibodies are not diagnostic for a specific disease but are closely associated with suppurative infection which is usually due to gram-positive organisms. These antiglobulin antibodies, unlike the rheumatoid factors, are species restrictive and will only bind to the Fab fragment of IgG of the same species or of closely related species [4].
June 1973
The American
Journal
of Medicine
Volume
54
731
SEROLOGIC
TESTS FOR ANTIGLOBULIN
Incidence Antibodies
TABLE I
of Positive
ANTIBODIES-WALLER
Tests
Test for Rheumatoid Factors Total No. Patients
Study Group 1 2 3
No. Positive
181 926 213
Totals
1,320
Per Cent Positive
for
Test for Serum Agglutinators No. Positive
Per Cent Positive
13 148 3
7.1 16.0 1.4 ~___
27 5 4
14.9
164
12.4
36
2.7
0.5 1.9
NOTE: Group 1 = hospital patients. Group 2 = patients attending the rheumatology clinic. Group 3 = patients attending the venereal disease clinic.
Tests for determining the titers of the serum agglutinators are used as diagnostic aids for hidden abscesses. In suppurative infection, and especially in gram-positive septicemia, titers are very high. Elevated titers have aided. in differentiating osteomyelitic abscesses from destructive tumor lesions of the bone. The titers of the serum agglutinators will fall when the suppurative lesion resolves; thus, falling titers of these antibodies help in evaluating the resolution of large lesions. In addition, the failure of the patient to respond with a rise in titers in the face of significant or multiple abscesses is usually a warning that the patient is not immunologically competent, such as patients with thymoma, lymphoma or myeloma. This prospective study of 1,320 persons whose serum was examined for the presence of both rheumatoid factors and serum agglutinators was undertaken to evaluate and compare the effectiveness of the tests for these two distinct antiglobulin antibodies as diagnostic aids in rheumatoid arthritis and suppurative infection, respectively. METHODS Selection
of
from three rheumatoid
Serum
separate factors
for
Testing.
Serum
groups of persons by three different
specimens
were tested tests and
for for
serum agglutinators by four different tests. All positive tests were repeated one or more times. Group 1. Hospital patients: Serum specimens from 181 marily
hospital
patients
to be tested
were
sent
to our
for the presence
laboratory
of serum
pri-
agglutin-
ators. Infection was suspected in some of these patients. Many of the patients presented diagnostic problems or had fever of undetermined origin. Group 2. Ambulatory patients: Serum specimens from 926 patients attending the connective tissue (rheumatology) clinics were sent to our laboratory to be tested for the presence of rheumatoid factors. Group 3. “Normal” persons attending the city venereal disease clinic: Serum specimens were also obtained from 213 persons attending the venereal disease clinic. They were primarily young adults, some with and some without complaints of urethral or vagi-
732
June 1973
nal discharge. Neither antiglobulin test was ordered, and these were used as a control group.
Antiglobulin
The American Journal of Medicine
Antiglobulin tests were performed before the patients’ diagnoses were ascertained. The charts of all hospital patients (group 1) were completely reviewed, and a serum electrophoretic pattern was obtained in all. Among the ambulatory patients (group 2), only the charts of patients with positive tests were reviewed. In the patients from the venereal disease clinic (group 3). all those with positive tests were interviewed and their tests repeated. Methods of Testing. (1) The presence of rheumatoid factors (anti-Fc IgG) was determined by using the following three tests: slide latex test, sensitized human cell test and sensitized sheep cell test. In this study a test for rheumatoid factors was considered positive if two of these three tests were positive and if the titer was 160 or above by the sensitized human cell test. (2) Serum specimens were tested for the following serum agglutinators (anti-Fab IgG): papain, subtilisin, trypsin and chymotrypsin. A test for serum agglutinators was considered positive if three of these four tests were positive with a titer of 160 or above. Methods for performing these antiglobulin tests have been reported in detail elsewhere [5-71. (3) Serum electrophoretic patterns were obtained for the serum specimens from all 181 hospitalized patients. (4) Reduction of immunoglobulins ( IgM) by 2-mercaptoethanol, and serum chromatography by DEAE cellulose were performed on serum as described in the text. RESULTS
Table I shows the results of the tests for rheumatoid factors and serum agglutinators in 1,320 patients. The incidence of positive tests for serum agglutinators was highest in the hospitalized patients (group 1) whereas the incidence of positive tests for rheumatoid factors was highest, understandably, in the patients attending the rheumatology clinic (group 2). In the patients attending the venereal disease clinic (group 3), the incidence for both types of antiglobulin antibodies was only 1 or 2 per cent. Table II shows the variety of diseases among the 181 hospital patients. The association between positive tests for serum agglutinators and suppurative infection is apparent. In patients with pneumonia, suppuration is variable; in addition, there is natural drainage, and the patients receive early therapy. Thus positive tests are not usual. Both patients with granulomatous infection and positive tests had sarcoidosis, a disease not usually associated with high titers of serum agglutinators. However, one of these patients had a cavity in the upper lobe of the left lung; the other had diffuse myositis of unknown etiology. The incidence of positive tests in patients with localized infections depends upon the severity of the lesion. In neither malignancy nor metabolic disease are these antibodies present in high titer. The exception in this
Volume 54
SEROLOGIC
instance was a patient with chronic alcoholic cirrhosis, hematuria and a hemolytic anemia of unknown etiology. Among the patients with connective tissue diseases was a two and a half year old child with lupus erythematosus and a young boy with
juvenile
ters
of
rheumatoid
serum
agglutinators.
arthritis; In
both
had
neither
high of
ti-
these
diseases are the tests usually positive. However, it was the first hospital admission for both children, and the possibility that titers may be high in early and explosive disease and then decline in the chronic state cannot be ruled out. Neither child had a positive test for rheumatoid factor or for lupus erythematosus. Table I II shows the correlation (or lack of it) between the positive test and the disease for which the test was ordered. The serum agglutinator test was ordered for the hospitalized patients (group l), the rheumatoid factor test was ordered for the patients attending the rheumatology clinic (group 2) and neither test was ordered for the patients attending the venereal disease clinic (group 3). In 13 of the 181 hospitalized patients, the tests for rheumatoid factors were positive, but only 6 of the 13 had rheumatoid arthritis. The other seven patients had a variety of disease states: one had pulmonary hemosiderosis, one Wegener’s granuloma, one sarcoidosis, one Aspergillus infection of the lungs, two heart failure and one a pulmonary TABLE
II
Tests for Serum Agglutinators in 181 Hospital Patients (Group 1) with a Variety of Diseases
Infection Gram-negative septicemia Pyelonephritis Pneumonia Granulomatous infection Subacute bacterial endocarditis Large abscesses (empyema, osteomyelitis) Infections associated with immunologic deficiency (lymphoma, thymoma, myeloma, transplantation) Localized infection (gonorrhea, pyoderma tonsillitis, infectious arthritis, wound infection, decubitus ulcers) Malignancy Connective tissue diseases (rheumatoid arthritis, lupus, gout) Undiagnosed (fever of undetermined origin) Metabolic disease (diabetes, nephrosis, toxic drugs, muscular dystrophy, alcoholism, cardiovascular disease) Totals
Total
TABLE
III
--..
Study Group
1 1 2 3 12 ...
25
4
1 2 3 Totals TABLE
IV
-
14 19
Presence of Rheumatoid Arthritis or Suppurative Infection in Patients with Positive Tests for Rheumatoid Factors or Serum Agglutinators Rheumatoid RheuFactor matoid Per Cent Test Arthritis CorrelaPositive Present tion
..
5
2 1
54
1
181
27
Study Group 1 2 3 Totals
June 1973
Serum Agglu- Suppuratinator tive Per Cent Test Infection CorrelaPositive Present tion
13 148 3
6 145 0
46.2 98.0 None
27 5 4
164
151
92.0
36
False-Positive bodies
..
15
ANTIBODIES-WALLER
infiltrate of unknown etiology. Thus, the correlation between positive test and disease state was only 46.2 per cent. However, among the same patients, 21 of 27 with positive tests for serum agglutinators had suppurative infection. Of the six patients who did not have suppurative infection, one had juvenile rheumatoid arthritis, one lupus erythematosus, one sarcoidosis with a cavity in the left lung, one cirrhosis, one sarcoidosis with a diffuse myositis, and one rheumatic fever with periosteal elevation of undetermined etiology. In the patients attending the rheumatology clinic (group 2) there was a very high correlation between the positive test for rheumatoid factor and the presence of rheumatoid arthritis. Of these 926 patients, 5 had a positive test for serum agglutinators (0.5 per cent), but none had a suppurative infection. Of these patients, one had hyperthyroidism, two rheumatoid arthritis, one connective tissue disease of unknown etiology and one osteoarthritis. Serums from these patients were reduced with 2-mercaptoethanol and chromatographed on DEAE cellulose, and in three of these patients serum agglutinators were IgM rather than IgG. Of the 213 patients attending the venereal disease clinic (group 3) three had a positive test for rheumatoid factor (1.4 per cent). Only one of these three had gonorrhea. None had rheumatoid
Patients (no.) Positive Test
TESTS FOR ANTIGLOBULIN
Total No. Patients 181 926 213
for
77.7 None 25.0
23
63.0
~~ ~~___
Antiglobulin
Anti-
Rheumatoid Factor Serum Agglutinator Test Test ____~~ _~__ _____-__ No. Positive No. Positive Without Per Cent Without Per Cent Rheumatoid FalseSuppurative FalseArthritis Positive Infection Positive --__ 6 3.3 7 3.8 0.5 5 0.3 3 3 1.4 1.4 3 13
1,320
The American
Tests
21 0 1
Journal
0.9
of Medicine
1.0
14
Volume
54
733
SEROLOGIC TESTS FOR ANTIGLOBULIN
ANTIBODIES-WALLER
tion. This study, however, confirms the reported association of these antiglobulin antibodies with severe suppurative infection. These antibodies are usually produced as IgG, but when they are produced as IgM, titers may be spuriously high because the IgM molecules give higher agglutination titers than the IgG molecules. Among the 181 hospitalized patients, 27 (15 per cent) had elevated titers of serum agglutinators. This reflects the interest in infection in this group of patients since the test ordered was for the serum agglutinators. In 8 of these 27 patients the test for rheumatoid factors was usually positive only with the slide latex test. However, in two of these patients (one with an Aspergillus infection of the lungs and the other with sarcoidosis), the rheumatoid factor titer was high enough to be calculated as positive for this study. Positive tests for serum agglutinators are unusual in nonhospitalized patients, and as a rule those who do have positive tests do not have severe suppurative infection. However, in many instances, these tests may be helpful in differentiating a hidden abscess from a malignancy. One patient had a lesion in the upper lobe of the left lung which was thought to be a possible malignancy. Results of the tests for serum agglutinators were just below the level considered to be positive. Repeat tests gave the same results. At surgery the lesion was found to be a chronically infected lobe due to obstruction from aspiration. Sometimes a well walled off abscess will cause only modest elevations in the titers of the serum agglutinators. However, when the abscess is pierced, the inevitable spillage of its contents will stimulate the serum agglutinators and the titers will rise. As resolution of the lesion occurs, the titers will fall. This study lends credence to the postulate that rheumatoid factors are humoral heterophil antibodies that may be stimulated by a variety of immunologic events. As stated, the correlation of positive tests for these antibodies with rheumatoid arthritis is only about 50 per cent in very sick hos-
arthritis. Of these 213 patients 4 had a positive test for serum agglutinators (1.9 per cent). Only one had gonococcal urethritis. If the results of tests for rheumatoid factors and serum agglutinators, as standardized in this study, are evaluated for correlation of the test with specific disease, the percentages of false-positive tests among the hospitalized patients (group 1) are 3.8 and 3.3 per cent, respectively; among the patients attending the rheumatology clinic (group 2), 0.3 and 0.5 per cent, respectively; and among patients attending the venereal disease clinic (group 3) 1.4 and 1.4 per cent, respectively (Table IV). These percentages of false-positive tests are much higher when titers lower than those used as the standard in this study are accepted as positive tests. COMMENTS The requirements for a positive test for rheumatoid factors in this study are more stringent than those used in other studies. Among the 181 hospital patients (group l), 17 had only positive slide latex tests or a positive slide latex test and a sensitized human cell titer below 160. These patients were not considered to have positive tests for rheumatoid factor. Even so, among this group of patients, only 46.2 per cent of those with rheumatoid factor titers of 160 or above by the sensitized human cell test had rheumatoid arthritis. The sensitized sheep cell tests were positive for rheumatoid factor in three of the hospitalized patients in the absence of rheumatoid arthritis. The diagnoses in these were Wegener’s granuloma, chronic heart failure and sarcoidosis. Thus, it is necessary to conclude that among very sick patients positive tests for rheumatoid factors only correlate with rheumatoid arthritis about 50 per cent of the time. However, in patients in whom there is a high degree of suspicion of the disease, the correlation is about 98 per cent. It is also obvious that among ambulatory patients attending a clinic for a completely unrelated disease, positive tests for rheumatoid factors are not apt to correlate with rheumatoid arthritis. The tests for serum agglutinators (anti-Fab IgG) have only recently been described and will no doubt require further study and standardiza-
pitalized
patients.
ACKNOWLEDGMENT I am grateful to Mrs. Nellie Curry and Mrs. Jean Mallory for invaluable technical assistance.
REFERENCES 1.
2.
3.
734
Meischer P, Fauconnet M: L’absorption du facteur L. E. par des noyaux cellulaires isol&. Experientia IO: 252, 1954. Winbland S, Hansen A, Jensen R: The reaction of rheumatoid factor with immunoglobulin G from various animal sources, Human Antihuman Gamma Globulins (Grubb R. Samuelsson G. eds), New York, Pergamon Press, 1971, p 78. Waller M: Present status of the rheumatoid factor. Crit Rev Clin Lab Sci 2: 173, 1971. June 1973
The American
Journal
of Medicine
Volume
4.
5. 6. 7.
54
Wailer M. Duma RJ: Increased antibodies to IgG fragments: correlation with infection due to gram-positive bacteria. J Infect Dis 125: 45, 1972. Waller M, Decker B, Toone EC, lrby R: Evaluation of rheumatoid factor tests. Arthritis Rheum 4: 579, 1961. Waller M, Toone EC, Vaughan E: Study of rheumatoid factor in a normal population. Arthritis Rheum 7: 513, 1964. Wailer M, Blaylock K: Further studies on the antiglobulin factors in human serum to the pepsin digested fragment of Ri anti-Rh antibody. J lmmun 97: 438, 1966.
MARION Richmond.
WALLER,
Ph.D
Virginia
From the Division of Connective Tissue Diseases, Department of Medicine, Medical College of Virginia School of Medicine, Virginia Commonwealth University, Richmond, Virginia. This investigation was supported in part by Research Grant AM 04549 from the National Institutes of Health. This is publication No. 58 from the Charles W. Thomas Arthritis Fund, Medical College of Virginia. Requests for reprints should be addressed to Dr. Marion Wailer, Department of Medicine, Medical College of Virginia, Richmond, Virginia 23219. Manuscript accepted January 25, 1973.
A study of 1,320 serum samples for rheumatoid factors and serum agglutinators revealed poor correlation between test results for rheumatoid factors and the presence of rheumatoid arthritis in hospitalized patients with a variety of different diseases. Although the correlation between test results for serum agglutinators and the presence of suppurative infection was good in hospitalized patients, correlation was very poor in less sick ambulatory patients. These studies reveal that correlation of serologic test results with specific disease states may be profoundly altered by severe systemic disease of diverse etiology. Serologic tests for the demonstration of antibody reactivity to heterologous antigens are sometimes useful as diagnostic aids. The presence of antibodies in low titer is usually characteristic of normal serum; elevated titers are frequently associated with a specific disease entity. The Paul-Bunnell test for infectious mononucleosis is probably the best known example of this type of test. associated with a renewed interest in In recent years, immunology, various tests have been developed that have proved valuable in confirming the presence of certain connective tissue diseases. The tests for antinuclear antibodies detect heterologous nuclear antigens that are widespread in nature, and the tests for rheumatoid factors detect antigens present on most mammalian immunoglobulin G (IgG) [1,2]. Rheumatoid factors are present in a variety of disease states and in normal persons, usually in low titer. High titers of rheumatoid factors are almost exclusively associated with rheumatoid arthritis [3]. A second group of antiglobulin antibodies, called serum agglutinators, distinct from rheumatoid factors has attracted considerable interest. Unlike the antibodies mentioned, these antibodies are not “humoral heterophils.” The humoral heterophils, as their name implies, are associated with diverse antigens in different species. The serum agglutinators are antiglobulin (anti-Fab IgG) antibodies presumably present in the serums of all mammals. High titers of these antibodies are not diagnostic for a specific disease but are closely associated with suppurative infection which is usually due to gram-positive organisms. These antiglobulin antibodies, unlike the rheumatoid factors, are species restrictive and will only bind to the Fab fragment of IgG of the same species or of closely related species [4].
June 1973
The American
Journal
of Medicine
Volume
54
731
SEROLOGIC
TESTS FOR ANTIGLOBULIN
Incidence Antibodies
TABLE I
of Positive
ANTIBODIES-WALLER
Tests
Test for Rheumatoid Factors Total No. Patients
Study Group 1 2 3
No. Positive
181 926 213
Totals
1,320
Per Cent Positive
for
Test for Serum Agglutinators No. Positive
Per Cent Positive
13 148 3
7.1 16.0 1.4 ~___
27 5 4
14.9
164
12.4
36
2.7
0.5 1.9
NOTE: Group 1 = hospital patients. Group 2 = patients attending the rheumatology clinic. Group 3 = patients attending the venereal disease clinic.
Tests for determining the titers of the serum agglutinators are used as diagnostic aids for hidden abscesses. In suppurative infection, and especially in gram-positive septicemia, titers are very high. Elevated titers have aided. in differentiating osteomyelitic abscesses from destructive tumor lesions of the bone. The titers of the serum agglutinators will fall when the suppurative lesion resolves; thus, falling titers of these antibodies help in evaluating the resolution of large lesions. In addition, the failure of the patient to respond with a rise in titers in the face of significant or multiple abscesses is usually a warning that the patient is not immunologically competent, such as patients with thymoma, lymphoma or myeloma. This prospective study of 1,320 persons whose serum was examined for the presence of both rheumatoid factors and serum agglutinators was undertaken to evaluate and compare the effectiveness of the tests for these two distinct antiglobulin antibodies as diagnostic aids in rheumatoid arthritis and suppurative infection, respectively. METHODS Selection
of
from three rheumatoid
Serum
separate factors
for
Testing.
Serum
groups of persons by three different
specimens
were tested tests and
for for
serum agglutinators by four different tests. All positive tests were repeated one or more times. Group 1. Hospital patients: Serum specimens from 181 marily
hospital
patients
to be tested
were
sent
to our
for the presence
laboratory
of serum
pri-
agglutin-
ators. Infection was suspected in some of these patients. Many of the patients presented diagnostic problems or had fever of undetermined origin. Group 2. Ambulatory patients: Serum specimens from 926 patients attending the connective tissue (rheumatology) clinics were sent to our laboratory to be tested for the presence of rheumatoid factors. Group 3. “Normal” persons attending the city venereal disease clinic: Serum specimens were also obtained from 213 persons attending the venereal disease clinic. They were primarily young adults, some with and some without complaints of urethral or vagi-
732
June 1973
nal discharge. Neither antiglobulin test was ordered, and these were used as a control group.
Antiglobulin
The American Journal of Medicine
Antiglobulin tests were performed before the patients’ diagnoses were ascertained. The charts of all hospital patients (group 1) were completely reviewed, and a serum electrophoretic pattern was obtained in all. Among the ambulatory patients (group 2), only the charts of patients with positive tests were reviewed. In the patients from the venereal disease clinic (group 3). all those with positive tests were interviewed and their tests repeated. Methods of Testing. (1) The presence of rheumatoid factors (anti-Fc IgG) was determined by using the following three tests: slide latex test, sensitized human cell test and sensitized sheep cell test. In this study a test for rheumatoid factors was considered positive if two of these three tests were positive and if the titer was 160 or above by the sensitized human cell test. (2) Serum specimens were tested for the following serum agglutinators (anti-Fab IgG): papain, subtilisin, trypsin and chymotrypsin. A test for serum agglutinators was considered positive if three of these four tests were positive with a titer of 160 or above. Methods for performing these antiglobulin tests have been reported in detail elsewhere [5-71. (3) Serum electrophoretic patterns were obtained for the serum specimens from all 181 hospitalized patients. (4) Reduction of immunoglobulins ( IgM) by 2-mercaptoethanol, and serum chromatography by DEAE cellulose were performed on serum as described in the text. RESULTS
Table I shows the results of the tests for rheumatoid factors and serum agglutinators in 1,320 patients. The incidence of positive tests for serum agglutinators was highest in the hospitalized patients (group 1) whereas the incidence of positive tests for rheumatoid factors was highest, understandably, in the patients attending the rheumatology clinic (group 2). In the patients attending the venereal disease clinic (group 3), the incidence for both types of antiglobulin antibodies was only 1 or 2 per cent. Table II shows the variety of diseases among the 181 hospital patients. The association between positive tests for serum agglutinators and suppurative infection is apparent. In patients with pneumonia, suppuration is variable; in addition, there is natural drainage, and the patients receive early therapy. Thus positive tests are not usual. Both patients with granulomatous infection and positive tests had sarcoidosis, a disease not usually associated with high titers of serum agglutinators. However, one of these patients had a cavity in the upper lobe of the left lung; the other had diffuse myositis of unknown etiology. The incidence of positive tests in patients with localized infections depends upon the severity of the lesion. In neither malignancy nor metabolic disease are these antibodies present in high titer. The exception in this
Volume 54
SEROLOGIC
instance was a patient with chronic alcoholic cirrhosis, hematuria and a hemolytic anemia of unknown etiology. Among the patients with connective tissue diseases was a two and a half year old child with lupus erythematosus and a young boy with
juvenile
ters
of
rheumatoid
serum
agglutinators.
arthritis; In
both
had
neither
high of
ti-
these
diseases are the tests usually positive. However, it was the first hospital admission for both children, and the possibility that titers may be high in early and explosive disease and then decline in the chronic state cannot be ruled out. Neither child had a positive test for rheumatoid factor or for lupus erythematosus. Table I II shows the correlation (or lack of it) between the positive test and the disease for which the test was ordered. The serum agglutinator test was ordered for the hospitalized patients (group l), the rheumatoid factor test was ordered for the patients attending the rheumatology clinic (group 2) and neither test was ordered for the patients attending the venereal disease clinic (group 3). In 13 of the 181 hospitalized patients, the tests for rheumatoid factors were positive, but only 6 of the 13 had rheumatoid arthritis. The other seven patients had a variety of disease states: one had pulmonary hemosiderosis, one Wegener’s granuloma, one sarcoidosis, one Aspergillus infection of the lungs, two heart failure and one a pulmonary TABLE
II
Tests for Serum Agglutinators in 181 Hospital Patients (Group 1) with a Variety of Diseases
Infection Gram-negative septicemia Pyelonephritis Pneumonia Granulomatous infection Subacute bacterial endocarditis Large abscesses (empyema, osteomyelitis) Infections associated with immunologic deficiency (lymphoma, thymoma, myeloma, transplantation) Localized infection (gonorrhea, pyoderma tonsillitis, infectious arthritis, wound infection, decubitus ulcers) Malignancy Connective tissue diseases (rheumatoid arthritis, lupus, gout) Undiagnosed (fever of undetermined origin) Metabolic disease (diabetes, nephrosis, toxic drugs, muscular dystrophy, alcoholism, cardiovascular disease) Totals
Total
TABLE
III
--..
Study Group
1 1 2 3 12 ...
25
4
1 2 3 Totals TABLE
IV
-
14 19
Presence of Rheumatoid Arthritis or Suppurative Infection in Patients with Positive Tests for Rheumatoid Factors or Serum Agglutinators Rheumatoid RheuFactor matoid Per Cent Test Arthritis CorrelaPositive Present tion
..
5
2 1
54
1
181
27
Study Group 1 2 3 Totals
June 1973
Serum Agglu- Suppuratinator tive Per Cent Test Infection CorrelaPositive Present tion
13 148 3
6 145 0
46.2 98.0 None
27 5 4
164
151
92.0
36
False-Positive bodies
..
15
ANTIBODIES-WALLER
infiltrate of unknown etiology. Thus, the correlation between positive test and disease state was only 46.2 per cent. However, among the same patients, 21 of 27 with positive tests for serum agglutinators had suppurative infection. Of the six patients who did not have suppurative infection, one had juvenile rheumatoid arthritis, one lupus erythematosus, one sarcoidosis with a cavity in the left lung, one cirrhosis, one sarcoidosis with a diffuse myositis, and one rheumatic fever with periosteal elevation of undetermined etiology. In the patients attending the rheumatology clinic (group 2) there was a very high correlation between the positive test for rheumatoid factor and the presence of rheumatoid arthritis. Of these 926 patients, 5 had a positive test for serum agglutinators (0.5 per cent), but none had a suppurative infection. Of these patients, one had hyperthyroidism, two rheumatoid arthritis, one connective tissue disease of unknown etiology and one osteoarthritis. Serums from these patients were reduced with 2-mercaptoethanol and chromatographed on DEAE cellulose, and in three of these patients serum agglutinators were IgM rather than IgG. Of the 213 patients attending the venereal disease clinic (group 3) three had a positive test for rheumatoid factor (1.4 per cent). Only one of these three had gonorrhea. None had rheumatoid
Patients (no.) Positive Test
TESTS FOR ANTIGLOBULIN
Total No. Patients 181 926 213
for
77.7 None 25.0
23
63.0
~~ ~~___
Antiglobulin
Anti-
Rheumatoid Factor Serum Agglutinator Test Test ____~~ _~__ _____-__ No. Positive No. Positive Without Per Cent Without Per Cent Rheumatoid FalseSuppurative FalseArthritis Positive Infection Positive --__ 6 3.3 7 3.8 0.5 5 0.3 3 3 1.4 1.4 3 13
1,320
The American
Tests
21 0 1
Journal
0.9
of Medicine
1.0
14
Volume
54
733
SEROLOGIC TESTS FOR ANTIGLOBULIN
ANTIBODIES-WALLER
tion. This study, however, confirms the reported association of these antiglobulin antibodies with severe suppurative infection. These antibodies are usually produced as IgG, but when they are produced as IgM, titers may be spuriously high because the IgM molecules give higher agglutination titers than the IgG molecules. Among the 181 hospitalized patients, 27 (15 per cent) had elevated titers of serum agglutinators. This reflects the interest in infection in this group of patients since the test ordered was for the serum agglutinators. In 8 of these 27 patients the test for rheumatoid factors was usually positive only with the slide latex test. However, in two of these patients (one with an Aspergillus infection of the lungs and the other with sarcoidosis), the rheumatoid factor titer was high enough to be calculated as positive for this study. Positive tests for serum agglutinators are unusual in nonhospitalized patients, and as a rule those who do have positive tests do not have severe suppurative infection. However, in many instances, these tests may be helpful in differentiating a hidden abscess from a malignancy. One patient had a lesion in the upper lobe of the left lung which was thought to be a possible malignancy. Results of the tests for serum agglutinators were just below the level considered to be positive. Repeat tests gave the same results. At surgery the lesion was found to be a chronically infected lobe due to obstruction from aspiration. Sometimes a well walled off abscess will cause only modest elevations in the titers of the serum agglutinators. However, when the abscess is pierced, the inevitable spillage of its contents will stimulate the serum agglutinators and the titers will rise. As resolution of the lesion occurs, the titers will fall. This study lends credence to the postulate that rheumatoid factors are humoral heterophil antibodies that may be stimulated by a variety of immunologic events. As stated, the correlation of positive tests for these antibodies with rheumatoid arthritis is only about 50 per cent in very sick hos-
arthritis. Of these 213 patients 4 had a positive test for serum agglutinators (1.9 per cent). Only one had gonococcal urethritis. If the results of tests for rheumatoid factors and serum agglutinators, as standardized in this study, are evaluated for correlation of the test with specific disease, the percentages of false-positive tests among the hospitalized patients (group 1) are 3.8 and 3.3 per cent, respectively; among the patients attending the rheumatology clinic (group 2), 0.3 and 0.5 per cent, respectively; and among patients attending the venereal disease clinic (group 3) 1.4 and 1.4 per cent, respectively (Table IV). These percentages of false-positive tests are much higher when titers lower than those used as the standard in this study are accepted as positive tests. COMMENTS The requirements for a positive test for rheumatoid factors in this study are more stringent than those used in other studies. Among the 181 hospital patients (group l), 17 had only positive slide latex tests or a positive slide latex test and a sensitized human cell titer below 160. These patients were not considered to have positive tests for rheumatoid factor. Even so, among this group of patients, only 46.2 per cent of those with rheumatoid factor titers of 160 or above by the sensitized human cell test had rheumatoid arthritis. The sensitized sheep cell tests were positive for rheumatoid factor in three of the hospitalized patients in the absence of rheumatoid arthritis. The diagnoses in these were Wegener’s granuloma, chronic heart failure and sarcoidosis. Thus, it is necessary to conclude that among very sick patients positive tests for rheumatoid factors only correlate with rheumatoid arthritis about 50 per cent of the time. However, in patients in whom there is a high degree of suspicion of the disease, the correlation is about 98 per cent. It is also obvious that among ambulatory patients attending a clinic for a completely unrelated disease, positive tests for rheumatoid factors are not apt to correlate with rheumatoid arthritis. The tests for serum agglutinators (anti-Fab IgG) have only recently been described and will no doubt require further study and standardiza-
pitalized
patients.
ACKNOWLEDGMENT I am grateful to Mrs. Nellie Curry and Mrs. Jean Mallory for invaluable technical assistance.
REFERENCES 1.
2.
3.
734
Meischer P, Fauconnet M: L’absorption du facteur L. E. par des noyaux cellulaires isol&. Experientia IO: 252, 1954. Winbland S, Hansen A, Jensen R: The reaction of rheumatoid factor with immunoglobulin G from various animal sources, Human Antihuman Gamma Globulins (Grubb R. Samuelsson G. eds), New York, Pergamon Press, 1971, p 78. Waller M: Present status of the rheumatoid factor. Crit Rev Clin Lab Sci 2: 173, 1971. June 1973
The American
Journal
of Medicine
Volume
4.
5. 6. 7.
54
Wailer M. Duma RJ: Increased antibodies to IgG fragments: correlation with infection due to gram-positive bacteria. J Infect Dis 125: 45, 1972. Waller M, Decker B, Toone EC, lrby R: Evaluation of rheumatoid factor tests. Arthritis Rheum 4: 579, 1961. Waller M, Toone EC, Vaughan E: Study of rheumatoid factor in a normal population. Arthritis Rheum 7: 513, 1964. Wailer M, Blaylock K: Further studies on the antiglobulin factors in human serum to the pepsin digested fragment of Ri anti-Rh antibody. J lmmun 97: 438, 1966.