T H E SERUM COLLOIDAL GOLD TEST—AN AID TO DIAGNOSIS OF STRUMA LYMPHOMATOSA PHILIP SKIRPAN, M.D., ALFRED REICH, B.S., AND GEORGE CRILE, J R . , M.D. Departments of Clinical Pathology and General Surgery of The Cleveland Clinic Foundation, and The Frank E. Bunts Educational Institute, Cleveland, Ohio

MATERIALS AND METHODS

One hundred and forty-three persons were selected by the Department of Endocrinology, representing 4 clinical groups: (1) patients known to have struma lymphomatosa, (2) patients having toxic or nontoxic adenomatous goiter, (3) patients having nongoitrous diseases such as "collagen disease," hepatic disease, multiple myeloma, leukemia, Addison's disease, and (4) healthy persons. No clinical data were submitted to the laboratory with requests for the test; persons were identified only by name and number. In 19 patients a diagnosis of struma lymphomatosa was made by needle biopsy; 15 of these were diagnosed as struma lymphomatosa with eosinophilia (Hashimoto's type) and 4 as struma lymphomatosa with little or no eosinophilia. Nine patients with various diseases (group 3 above) served as positive controls because of the generally recognized related serum protein alterations that produce characteristic changes in colloidal gold solution. The serum colloidal gold test used was a modification of the method of Maclagan,8 differing by the substitution of commercially prepared colloidal gold solution and the quantities of serum and barbital buffer constituting the dilutions. Proportionally smaller quantities of serum and barbital buffer were used to reduce the amount of colloidal gold solution necessary for the test to 1 ml. for each tube. Reagents Four reagents are used to make the test: barbital buffer, colloidal gold solution, human serum, and sodium chloride solution. Received, May 27, 1955; accepted for publication June 22. Dr. Skirpan is a Fellow and Mr. Reich is a member of the Assistant Staff, Department of Clinical Patholog\ r ; Dr. Crile is a member of the Staff, Department of General Surgery. 1274

Downloaded from http://ajcp.oxfordjournals.org/ by guest on January 26, 2017

The report by Cooke and Wilder,2 in 1954, of a "persistent abnormality" of the serum colloidal gold test in 2 patients having struma lymphomatosa, suggested that abnormal blood chemistry might have pathognomonic significance. To evaluate this suggestion, serum colloidal gold tests were performed and plasma cholesterol levels were determined by the clinical laboratory of the Cleveland Clinic on a series of 143 persons, including 19 patients proved by needle biopsy to have struma lymphomatosa (primary thyroid failure with compensatory thyroid enlargement10). The serum colloidal gold test used was a modification of the method reported by Maclagan.8 The purpose of this paper is to present the results obtained in that series.

NOV. 1955

SEBUM COLLOIDAL GOLD TEST

1275

Equipment Test-tube rack Seven test tubes (13 by 100 mm.) Two 1-ml. pipets (graduated in hundredths) One 5-ml. pipet Careful cleansing of glassware is essential.7 Procedure The serial dilution technic was employed, the dilutions being made in 6 tubes, and tube 7 being the control tube. To tube 1 was added 0.36 ml. of barbital buffer; 0.20 ml. of barbital buffer was added to each of the 5 remaining tubes. To tube 1 was added 0.04 ml. of human serum, and the contents were mixed by drawing them back into the pipet and releasing them into the tube. A volume of 0.20 ml. of the mixture was transferred to tube 2 and the contents similarly mixed. The process was continued through tube 6, and 0.20 ml. was discarded. To control tube 7 was added 0.34 ml. of 1 per cent sodium chloride solution. To each tube was added 1 ml. of acidified colloidal gold solution. The tubes were then allowed to stand at room temperature for 24 hours. Grading of results Each test was read provisionally at 6 hours, and finally at 24 hours. Each tube was graded on the scale of 5 to 0, as is standard practice for colloidal gold chloride tests. Complete reduction, 5, is signified by a colorless supernatant fluid and marked precipitation of the colloidal gold; whereas, no reduction, 0, is indicated by the red color of the supernatant fluid and no precipitation of the colloidal gold. The supernatant fluid of the control usually has a bluer color than that of the corresponding positive serum-gold mixture. * Lunge's Colloidal Gold Solution, manufactured by Magar Chemicals, Inc., Cornwall Landing, New York, and by The Keystone Laboratory, Erie, Pennsylvania.

Downloaded from http://ajcp.oxfordjournals.org/ by guest on January 26, 2017

Barbital buffer (pH 7.6) Diethylbarbituric acid 0.55 Gm. Sodium barbitone 0.31 Gm. Phenol 0.20 Gm. Distilled water 100 ml. Colloidal gold solution. A commercially prepared gold solution is used.* The quantity of colloidal gold solution needed for each test is acidified at the time of use by adding the minimal amount of 1 per cent acetic acid that will completely reduce 1 ml. of colloidal gold solution in the presence of 0.34 ml. of 1 per cent sodium chloride solution. Once this quantity of 1 per cent acetic acid has been determined, the same amount may be used for each milliliter of colloidal gold solution of any given lot. Human scrum. A volume of 0.1 ml. of serum is more than sufficient to perform the test. Sodium chloride solution (1 per cent). The 1 per cent sodium chloride solution added to acidified colloidal gold solution serves as a control for each test.

1276

SKIRPAN ET

Vol. 25

AL.

In tests performed on patients with deranged serum proteins, complete reduction is obtained in tube 1 (serum dilution 2.86 X 10~2), and in tube 2 (serum dilution 4.2 X 10 -3 ). Tube 3 (serum dilution 6.0 X 10 -4 ) usually is moderately reactive (grades 2 to 4)- The remaining 3 tubes become progressively nonreactive. RESULTS

The serum colloidal gold tests of 15 of the 19 patients who had struma lymphomatosa showed significant reduction of colloidal gold solution in serial dilutions, with grades greater than 551-000. In comparison with the serum TABLE l R E S U L T S OF SERUM COLLOIDAL G O L D T E S T IN 19 P A T I E N T S H A V I N G STRUMA LYMPIIOMATOSA

Age

Sex

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

42 40 41 37 49 47 41 23 38 25 12 34 48 54 32 59 25 51 59

F F M F F F F F F F F F F F F F F F F

Serum Colloidal Gold Test Rating (Positive Greater than 551-000)

555-531 555-431 555-420 555-311 555-211 555-100 554-310 554-310 554-210 554-210 554-210 554-210 554-210 553-100 552-100 321-000 321-000 310-000 210-000

Cholesterol, mg./lOO ml. Plasma (Normal 150-250 mg./lOO ml.)

Initial B.M.R.

174 169 170 192 286 221 296 149 286 179 149 230 213 310 184 274 138 249 414

-13 -15 -29 -11 N o t done -8 -21 +8 -32 -29 -25 -15 -11 -17 -11 +5 +2 +4 -18

TABLE 2 R E S U L T S OF SERUM COLLOIDAL G O L D T E S T IN 9 P A T I E N T S H A V I N G N O N G O I T R O U S D I S E A S E S Case Number

Serum Colloidal Gold Test Rating (Positive Greater than 551-000)

Clinical Diagnosis

20 21 22 23 24 25 26 27 28

555-555 555-542 555-531 555-310 555-210 555-110 544-211 521-111 511-111

P o r t a l cirrhosis Histoplasmosis Infectious hepatitis, acute Multiple myeloma Brucellosis Addison's disease Idiopathic hypergammaglobulinemia Acute monocytic leukemia Lupus erythematosus

Downloaded from http://ajcp.oxfordjournals.org/ by guest on January 26, 2017

Case Number

Nov. 1955

1277

SERUM COLLOIDAL GOLD TEST TABLE 3

R E S U L T S OF SERUM COLLOIDAL G O L D T E S T IN R E L A T I O N TO SERUM P R O T E I N V A L U E S IN 4 P A T I E N T S H A V I N G STRUMA LYMPHOMATOSA

Case Number

1 6 16 18

Serum Colloidal Gold Test Rating (Positive Greater than 551-000)

555-531 (000-000)* 555-100 (211-000)* 321-000 310-000 Normal values

Serum Protein Values, Gm. per 100 ml. of Plasma (Electrophoretic Fractionation) Albumin

Gamma globulin

3.22 3.78 3.83 3.88 4.09 (mean) 3.72-5.11 (range)

1.53 1.27 0.97 0.83 0.77 (mean) 0.60-0.91 (range)

cholesterol determinations and the basal metabolic rates of those patients, the results of the serum colloidal gold test seemed to be a more sensitive indicator of struma lymphomatosa (Table I). The serum cholesterol was normal in 13 and elevated in 6. Table 2 shows a comparison of the results of the serum colloidal gold tests in patients of group 3 (those having various nongoitrous diseases); positive results similar to those in thyroid disease were obtained in most instances. The buffered, acidified, colloidal gold solution did not react with the serums of 75 healthy persons, nor with the serums of 28 patients having nontoxic adenomatous goiter and 10 having Graves' disease with hyperthyroidism. Positive reactions were obtained with serums from 2 patients having nontoxic adenomatous goiter. DISCUSSION

The reaction between the serum and the colloidal gold solution is based on the serum protein components.4' 9 The albumin component of normal serum serves to inhibit the reaction, as it does when the colloidal gold chloride test is performed with cerebrospinal fluid.5 Alterations in the relative proportions of beta globulin and gamma globulin also affect the results of the reaction.1 Kabat and his associates6 found that the colloidal gold reaction could be completely inhibited in most instances by adding electrophoretically separated albumin to gamma globulin in appropriate proportions. Skillern and associates* have found that the serums of the 15 patients having positive colloidal gold curves contained less albumin and more gamma globulin than normal subjects, and state that this is due to effects of thyroxine deficiency.10 In patients treated with desiccated thyroid, the serum colloidal gold test becomes nonreactive (Table 3). The deranged serum proteins account for the positive reactions with the acidified, buffered, colloidal gold solution. A similar mechanism controls the * Examples kindly furnished by t h e m in instances of positive and negative colloidal gold tests are listed in Table 3.

Downloaded from http://ajcp.oxfordjournals.org/ by guest on January 26, 2017

* After t r e a t m e n t with desiccated thyroid. D a t a in this table were furnished by Skillern and associates. 7

1278

SKIBPAN ET AL.

Vol. 25

positive reactivity of colloidal gold solution to the serums of patients having certain nongoitrous diseases (Table 2). Gray 3 demonstrated by electrophoretic studies that there is elevation of gamma globulin and decrease of serum albumin in hepatic disease. In those conditions in which there is elevation of the gammaglobulin fraction, or alteration of the albumin-globulin ratio, as in certain cases of collagen disease, hepatic disease, multiple myeloma, leukemia, and Addison's disease, the reaction of the colloidal gold solution is positive. Thus, the reaction lacks specificity. Nevertheless, the serum colloidal gold test proved to be of value as a clinical aid to diagnosis of thyroid disease, where struma lymphomatosa must be differentiated. The test is inexpensive and is easily performed. SUMMARY

REFERENCES 1. BERNSOHN, J., AND BORMAN, E . K . : Proteins in t h e colloidal gold reaction. J. Clin. Invest., 26:1026-1030,1947. 2. COOKE, R . T . , AND W I L D E R , E . : Hashimoto's struma lymphomatosa. (Letter to t h e Editor.) Lancet, 1:984-985,1954. 3. GRAY, S. J . : Mechanism of the colloidal gold reaction of blood serum in liver disease. Proc. Soc. Exper. Biol. & Med., 5 1 : 400-401, 1942. 4. GRAY, S. J.: T h e colloidal gold reaction of blood serum in diseases of t h e liver. Arch. I n t . Med., 65: 523-544, 1940. 5. HOFFMAN, W. S.: T h e Biochemistry of Clinical Medicine. Chicago: Year Book P u b lishers, I n c . , 1954, 681 p p . 6. K A B A T , E . A., H A N G E R , F . M . , M O O R E , D . H . , AND LANDOW,H.:The relation of cephalin

flocculation and colloidal gold reactions to the serum proteins. 563-568, 1943.

J. Clin. Invest., 22:

7. L A N G E , C , M I L L E R , J . K., AND H A R R I S , A. H . : E x a m i n a t i o n of cerebrospinal fluid and

blood. Simplified methods derived from the q u a n t i t a t i v e gold reaction. Am. J. Clin. P a t h . , 20: 872-876, 1950. 8. MACLAGEN, N . F . : T h e preparation and use of colloidal gold sols as diagnostic agents. Brit. J . Exper. P a t h . , 27: 369-377, 1946. 9. MACLAGAN, N . F . : Laboratory tests in t h e diagnosis of liver disease; a report on three procedures. Brit. M . J., 2 : 363-365, 1944. 10. S K I L L E R N , P . G., C R I L E , G., J R . , M C C U L L A G H , E . P . , H A Z A R D , J. B . , L E W I S , L. A., AND

BROWN, H . : Struma lymphomatosa: primary thyroid failure with compensatory thyroid enlargement. J . Clin. Endocrinol., in press.

Downloaded from http://ajcp.oxfordjournals.org/ by guest on January 26, 2017

A method of performing a serum colloidal gold test is described. The serum colloidal gold test was positive in 15 of 19 patients with proved struma lymphomatosa, and seemed to be more sensitive than serum cholesterol determinations in establishing a diagnosis of primary thyroid failure with goiter.