ALBERT E. BULSON, Jr., B.S., M.D.. Editor and Manager

ORIGINAL ARTICLES

VOLUME III

SERUM DIAGNOSIS OF SYPHILIS*

J. P. SIMONDS, M.D. Superintendent Bacteriological Laboratory, Indiana State Board of Health, INDIANAPOLIS

The discovery of a method of serum diagnosis of syphilis furnishes a very striking illustration of the fact that one discovery in science forms a little later the foundation on which a more important discovery rests. The phenomenon of bacteriolysis, first demonstrated by Pfeiffer in 1894, gave to the study of immunity an importance it had not before possessed. Ehrlich three years later announced his side-chain theory of immunity, one of the most stimulating conceptions ever advanced. In 1900 Bordet1 found that

"following the introduction of the red blood-cells of one species into the organism of another, a hemolysin is formed which so injures the bloodcells of the first species that their hemoglobin goes into solution." He also showed that this hemolysis depends on the action of two substances in the hemolytic serum. These two substances are called respectively the hemolytic amboceptor and the complement. This discovery of Bordet. and a certain application of it discovered by Gengou are the basis of the serum tests for syphilis.

In the process of hemolysis, we have three factors: the red blood-cells to be hemolyzed or destroyed, the amboceptor which is specific and thermostable, and the complement which is nonspecific and thermolabile. The relation of these factors to each other in the process has been very aptly described by allowing a lock to represent the cell, a key to represent the amboceptor and the hand that turns the key, the complement. By

**Read before the Indiana State Medical Association at Fort Wayne, September 28, 1910.

1. Ann de l'Inst. Pasteur, 1900, xiv, 257.

NUMBER 12

giving a lock to a smith, we can get a key made to fit the specific lock exactly, but the key can be turned and the lock opened by any hand. When we inject the red blood-corpuscles of one animal. into another we get amboceptors which exactly fit corpuscles from the same species as those injected, but complement from any serum can produce the hemolysis. A member of this Society recently stated it in this way: A roof to be torn away represents the cell to be destroyed; the man who can tear away the roof represents the complement which destroys the cell; and the ladder by which the man reaches the roof represents the amboceptor by which the complement reaches the cell.

Gengou, in 1902, discovered that the phenomena just described applied not only to red bloodcells but also to bacteria and to the higher proteids. If serum from an animal immunized to typhoid bacilli, and therefore containing bacteriolytic amboceptor, be mixed with complement and a suspension of typhoid bacilli, and the mixture incubated for one-half an hour, little or no gross change can be detected. But it can be shown that the complement has been absorbed from this bacteriolytic system, by adding to it some red blood-cells and a corresponding hemolytic amboceptor. No hemolysis occurs because there is no complement available. It has all been absorbed or bound by the bacteriolytic system.

Wassermann and Brucks first applied this principle in the diagnosis of typhoid fever. Wassermann then conceived the idea that if he would use a syphilitic antigen this test could be applied to the diagnosis of syphilis. Spirochata pallida will not grow on any known culture media, and cannot be used as typhoid bacilli are. Hence, Wassermann used as his antigen an extract of the

2. Ann. de l'Inst. Pasteur. 1902, xvi, 734.
3. Deutsche Med. Wochenschr., 1906, p. 450.

centage of error with the Noguchi method was almost completely eliminated. In Noguchi's own hands1 and in the hands of those most experienced in its use, this over-sensitiveness does not exist. Its specificity appears to bear a direct relation to the skill and experience of the investigator.

The Wassermann test, on the other hand, in 8 or 9 per cent. of cases fails to give a positive result in patients that are known to be syphilitic. This is due to the fact that quite often human serum contains anti-sheep amboceptor. Since the amount of syphilitic amboceptor in the patient's serum is an unknown quantity it is possible that in some cases there may not be enough to completely absorb all the complement and thus a small amount may be left free. This small amount may be sufficient to cause hemolysis in the presence of such a large amount of antisheep hemolytic amboceptor, namely that added purposely plus that normally present in the patient's serum. For Morgenroth and Sachs14 showed that in the presence of larger amounts of amboceptor smaller doses of complement suffice for hemolysis. This variability in the amount of amboceptor and the impossibility of accurately measuring its amount form the chief objections to Wassermann's original method and to Hecht's and other modifications of it.

Kaplan15 urges very strongly that all sera be tested by the Wassermann and Noguchi methods simultaneously. In his opinion, “a fairly marked Wassermann reaction, ninety-nine times out of a hundred, means syphilis, and a negative Noguchi the same number of times means no syphilis. The two methods are very decisive, but in opposite ways; and used together, carry with them an assurance which no amount of thoroughness and precision will replace if only one method is used."

In some respects the serum test for syphilis seems simple enough, but any one who has had any experience with hemolysis work knows the abundant chance of error that may entirely vitiate the results obtained. At the State Laboratory we have done something like 300 tests according to Noguchi's modification, but we are far from

considering our results as possessing any degree of accuracy, and therefore we make no analysis of them. Kaplan did 1,390 tests but did not consider the first 390 sufficiently accurate to justify their use in drawing conclusions in regard to the value of the test. Those who have had

13. Jour. Am. Med. Assn., 1910, lv, 727.

14. Ehrlich's Collected Studies on Immunity, 1st. Ed., New York, 1906, p. 252.

15. Am. Jour. Med. Sc., 1910, cxxxix, 82.

most experience in the use of the reaction are the most emphatic in declaring that abundant experience and thorough understanding of the phenomena of hemolysis are absolutely essential to reliable results.

In the first place all the reagents will, in time, lose their potency. This occurs slowly when they are in the dry form, more rapidly when kept in a liquid state. On account of this gradual loss of strength it is necessary to restandardize all reagents at intervals of a few days to two weeks. The dosage is exceedingly important. Too little antigen will render impossible the binding of all the complement present and thus result in a negative reaction where it should be positive. If the hemolytic amboceptor has deteriorated there may not be enough present to cause complete hemolysis and a given reaction may be pronounced positive when it should.be negative. Too large a dose of hemolytic amboceptor may, as already stated, be activated by a subminimal dose of complement and hemolysis may thus occur in a test that should be positive. This illustrates the necessity of frequently restandardizing all reagents. In performing the test such minute quantities of the various substances are used that it requires no little practice to accurately measure out 0.1 e.c. instead of 0.12 c.c. A difference of 0.02 c.c. -about one-fourth of an ordinary drop-will not infrequently entirely change the result. The guinea-pig serum used as complement must be fresh. Serum that has stood for twenty-four hours even at refrigerator temperature is practically worthless. Even variations in the reaction. of the ingredients used in the test may change the result. Sachs and Altmann16 examined ten sera with respect to the effect of increased or diminished acidity on the syphilitic reaction. They found that by adding 1/800 to 1/3200 normal sodium hydroxid solution to positive sera, the syphilitic reaction could be completely abolished. Sera thus inactivated could be reactivated by adding a proper amount of hydrochloric acid. Non-syphilitic sera, however, could not be made to react positively by changing the reaction.

The syphilitic antibody which is the cause of the reaction exists in the blood serum and in the

lymph. It has been found in the milk of syphilitic mothers, in the urine,18 and in pericardial, pleural and peritoneal exudates. 19 Constantini20

16. Berl. kl. Wochenschr., 1908, S. 699. Abstract in Centralbl. f. Bakt., Referate, 1908, xlii, 301.

17. Bab, Zeitschr. f. Geb. u. Gyn., 1908. lx, H. 2. Abstract Centralbl. f. Bakt. Referate, 1908, xliii, 279.

18. Blumenthal and Wile, Berl. kl. Wochenschr., 1908, No. 22, Abstract in Centralbl. f. Bakt., Referate, 1908, xlii, 304.

19. Veszpremi, Centralbl. f. allge. Path., 1910, xxi, 203. 20. Policlinico Sez. Med., 1908, Nos. 5 & 6. Abstract in Centralbl. f. allge. Path., xx, 590.

mothers whose sera gave positive Wassermann tests thus proving beyond any reasonable doubt that these women are truly syphilitic.

The difficulty of diagnosing clinically all forms of visceral syphilis, except possibly gumma of the brain, is too well known to require comment. In this field the serum reaction may be of very great value. It may be the deciding factor in differentiating gumma of the liver or one obstructing the bile duct from cancer. Butler37 Butler37 declares that in every disorder of the liver and in prolonged febrile disturbance in which the etiologic factor is uncertain, syphilis should be considered and a serum test made. The infrequency with which visceral syphilis is recognized clinically is shown by the fact that Lessers saw 30 gummas at post mortem that had not been discovered during life. He also found unmistakable evidence of syphilis in 9 per cent. of all males over 25 that came to autopsy in Berlin. It has been estimated that 25 per cent. of all males in Berlin over 25 years old have syphilis. Butler calls attention to the significant fact that the number showing demonstrable lesions of lues at autopsy is nearly one-half the total estimated number of syphilitics. "About 50 per cent. of cases of latent syphilis give a positive Wassermann reaction, i. e., the frequency of positive reactions in latent syphilis about equals the frequency of syphilitic processes in internal organs that are not recognized during life. No stronger argument than this could be introduced to prove that a positive Wassermann reaction indicates an existing syphilitic process in the organism." Neisser is quite emphatic in his opinion of the significance of a positive serum reaction. He declares that "When we find these so-called (syphilitic) antibodies, we have to do with a patient who has got syphilis, and I believe we may even say, with one who still harbors the syphilitic virus." A strong statement, that, from so great an authority as Neisser. It may be remembered, however, that not every ex-typhoid patient who gives a positive Widal reaction is a bacilli-carrier, and vice versa, not every bacillicarrier gives a positive Widal reaction. Reasoning by analogy, the presence of a positive Wassermann test does not absolutely prove that the syphilitic virus is still active, but it must be considered strong evidence to that effect. For this reason, the serum reaction may come to be of very great assistance in solving, in individual cases, the troublesome problem of syphilis and marriage.

37. N. Y. Med. Jour., 1909, lxxxix, 207.

38. Quoted by Butler, 1. c.

39. Brit. Med. Jour., 1908, ii, 1087.

At present there is no reliable method of determining just when it is safe for a syphilitic to marry. More or less arbitrary rules have been laid down, such as requiring three years of thorough treatment and at least one year without symptoms after treatment has ceased. But Lesser's post-mortem findings prove that syphilitic lesions may exist in various viscera without giving rise to any definite symptoms. Hence the question of syphilis and marriage cannot be settled by arbitrary rules in a manner that will be satisfactory in every case. Since in the vast majority of cases a positive Wassermann test indicates the presence in some part of the body of syphilitic virus at least potentially active, it is not improbable that further investigations may prove that no syphilitic whose serum gives a positive reaction should be allowed to marry until he has been given further treatment. While such a rule might not protect the wife any better than an arbitrary one, it would, no doubt, more thoroughly protect the off-spring both from hereditary syphilis and from the so-called syphilitic heredity characterized by the various dystrophies which give the child such an unequal chance in the struggle for existence.

The Wassermann reaction has practically settled many questions of etiology which, previous to its discovery, were mere matters of conjecture. The examination of serum from post mortems has yielded some interesting results particularly in reference to cases with cardiac and vascular lesions. Schlimpert obtained fifteen positive reactions in sixteen cases showing mesaortitis. Veszpremi1s obtained 100 per cent. positive reactions in serum from bodies showing "chronic fibrous aortitis," and 75 per cent. of positives in cases of chronic aortic endocarditis. The most striking results, however, have been obtained in cases of tabes and general paresis. McCampbell and Rowland have recently reviewed the literature and reported a number of results of their They came to the conclusion that "a positive Wassermann reaction with the blood serum occurs in 97.8 per cent. of those cases which, after varying periods of time, have been diagnosed as paresis. We have thus, a positive demonstration that paresis is a parasyphilitic or metasyphilitic disease."

own.

While the facts thus far known would not justify the use of the serum reaction as an absolute control for the treatment of syphilis, it has, however, been found to be of some value in this respect. The iodids seems to have no effect on

40. Verhandl. der deutsch. Path. Gesellschaft, 1909, 41. Jour. Med. Research, 1910, xxii, 185.

p. 96.