isolated from a furuncle a few months before the start of these experiments and

THE EFFECT OF PLEURAL EFFUSONS ON THE GROWTH OF STAPHYLOCOCCUS AUREUS N VTRO ALCE B. TOBLER AND MAX PNNER Division of Pulmonary Diseases of the Montefiore Ho8pital for Chronic Diseases, New York, New York Received for publication September 3, 1942 t was shown by Pinner and Voldrich (1932) that tuberculous effusions invariably contain a factor that behaves serologically like an agglutinin specific for Staphylococcus aureus. n the majority of non-tuberculous effusions, this factor could not be demonstrated or had a very low titer. They stated, too, that S. aureus had an agglutinative growth in pleural effusions. We wish to report further studies on the growth of S. aureus in tuberculous fluids in vitro, and on the possible factors in such effusions that modify the growth of S. aureus. SOURCE OF MATERAL AND EXPERMENTAL METHODS S. aureus, strain, was used in all experiments. Strain had been isolated from a furuncle a few months before the start of these experiments and had been subcultured many times in single colonies. t was a well pigmented strain. The strain was carried along on a 3 per cent nutrient agar. All pleural effusions were collected either in the Montefiore Hospital or in the Montefiore Country Sanatorium. The vast majority was obtained from patients with tuberculosis and artificial pneumothorax. Seven fluids only came from non-tuberculous patients. No fluids were used which contained secondary organisms. For all inoculations, a 24-hour broth culture of strain was diluted 1:5,- ;.2 ml. of this dilution was inoculated into 1 ml. of pleural effusion, into 1 ml. of nutrient broth, and was spread simultaneously on 3 per cent nutrient agar as a seeding control. This dilution of a 24-hour broth culture was chosen because.2 ml. of it,seeded on petri dishes,yielded isolated colonies, about 6 to 15 per plate. The inoculated tubes were incubated for 24 hours at 37 C. The pleural effusions contained varying amounts of agglutinated bacilli either in the bottom of the test tube or as a fine coherent network throughout the fluid. These tubes were then shaken in a shaking machine until all clumps were broken up. Then appropriate dilutions of the pleural effusions, ranging from 1:4 to 1: 1,, were made and.2 ml. of each dilution was plated on 3 per cent nutrient agar. The broth controls were diluted 1: 5, and were plated in the same way. The number of colonies on each plate were counted after 24 hours' incubation. The rate of growth was determined by inoculating 1 ml. of pleural effusion and 1 ml. of broth with.2 ml. of a 1: 5, dilution of a 24-hour broth culture;.2 ml. from each tube was plated every two hours on 3 per cent nutrient agar during the period of 24 hours. 341

342 ALCE B. TOBLER AND MAX PNNER Macroscopic agglutination tests were made with most of the pleural effusions. Clinical and laboratory data were collected and tabulated for each fluid. n the course of the experiments, three more strains of S. aureus (A. F., M. P., and W. L.) all isolated from furuncles and subcultured in a way similar to that described for, have been used occasionally in experiments parallel to strain. The ph of the pleural fluids was determined with Beckman's ph meter. *6w' z mj 'E *Z UJ 6 'U 12 16 16t 1t. U PLEURQAL BRPOT H EFrUsiON Downloaded from http://jb.asm.org/.z 4. i 1 3 4 5 6 LOGARTHMS OF NUMBER OF COLoNtEs _ =- 7 8 9 1 OF S. AUREUS PER CC. MG. 1. COMPARSON OF FREQQUENCY DSTRBUON OF GROWTH OF S. aureus N NuTRENT BROTH WiTH THAT N PLEURAL EFFUSONS on June 19, 218 by guest R"BULTS S. aureus showed regularly a much poorer growth in effusions than in nutrient broth. The range of the number of colonies obtained per ml. in broth and in pleural effusion after 24 hours' incubation is shown in the frequency distribution (fig. 1). (The difference in numbers between pleural fluid and broth is so large that a

EFFECT OF PL1URAL EFFUSONS ON GROWTH OF S. AUREUS 343 logarithmic scale had to be chosen.) The number of staphylococci in the 24- hour nutrient broth varied between 125,, and 75,,. t has not been determined whether this variation is actual or due to errors inherent in the counting method. The diagram (fig. 1) shows that the differences between growth in nutrient broth and that in effusions are significant and that the maximum growth in pleural fluids is far outside the range of variations in the nutrient broth. The pleural fluids yielded consistently less growth than the control broth. nhibition of growth, however, was irregular and varied from slight to marked. There was never complete inhibition of growth and there was apparently no actual destruction of staphylococci, because the number of staphylococci in the pleural effusions after 24 hours' incubation was never less than that in the inoculum. The ratio of growth between broth and effusions was 1 or more in 85 per cent of the effusions. The rate of growth in pleural effusions as compared to broth is shown in figure 2. The two outstanding features are a delay of multiplication in the pleural effusion and an increase at a lower rate. The maximum of growth in pleural effusion and broth is reached at about the same time (2-24 hours). The TABLE 1 Comparison of growth of S. aureus in 49 tuberculous pleural effusions with that in broth RATO OP NuMBER OF COLONES FROm 1 ML. OF PLEuRAL EFFUSON TO THAT OF BROTH NUMBER OF PLEURAL NFUSONS 1:29-1:99 2 1:1-1:999 5 1:1-1:9999 14 1:1,-1:99,999 22 1:1, and over 6 chart shows one pleural effusion which was a "good inhibitor" (ratio of inhibition was more than 1:1,) and one ind which the inhibiting power was slight (ratio of inhibition less than 1:1). The pattern in the two fluids is essentially the same. n the collection and tabulation of the laboratory and clinical data of the pleural fluids, the following relations were considered: 1. Ratio of inhibition and time interval between onset of effusion and withdrawal. This interval varied between three weeks and three years in 49 fluids (including repeated tapping). No correlation was found. 2. Ratio of inhibition and age of fluid after withdrawal. This age varied from two days to seven months in 49 fluids of 39 patients; 16 fluids were examined more than once following standing in the ice box from four days to three months. No trend was discernible. 3. Ratio of growth inhibition and bacteriological findings in effusion; 28 fluids contained tubercle bacilli and 21 did not. Seven additional fluids came from non-tuberculous patients. Of these, 5 had ratios from 1:2 to 1:485; the remaining 2 had higher ratios. The presence of tubercle bacilli showed no relation to growth inhibition.

344 ALCE B. TOBLER AND MAX PNNER 4. Ratio of inhibition and clinical course (acute, chronic, febrile, afebrile). No correlation was found. LUJ D. 1. i.j ci 9. n ui O 8!27, z o 6 L o 5. co ez + z,- 3- L J-. x 1. 1- PLEUQAL EFFUStON #81 "GOODNHBTORQ' 11...-... PLEULAL EFFUSTON#72 "POOPRNHBTOR." ---- NUTRENT BQOTH. 2 4 8 1 12 14 16 18 S 22 24 TTME fn HOUQl5 FG. 2. GROWTr OF S. aureus N PLEURAL EFFUSONS AS COMPARED TO THAT N NUTRENT BROTH TABLE 2 Relation between agglutination titer and inhibition ratio of 49 tuberculous effusion8 AGGLUTNATON TTER _ #. 1.e. #1. do -. -D de - - - - - NV31ON RATO *. H Below 5 5-9999 1,-499, 5, and over Below 15 2 2 1 15-799 4 6 4 8-4799 2 8 9 1 48 and over 2 7 1 The same lack of correlation exists between the agglutination titer and growth inhibition with the exception of 6 of the 7 non-tuberculous fluids, in which both agglutinative titer and growth inhibition were of a low order. Table

EFFECT OF PLEURAL EFFUSONS ON GROWTH OF S. AUREUS 2 shows this lack of relationship. The slight trend which the table seems to indicate is not statistically significant. t was noted that in general the agglutination titers did not reach quite the high values obtained by Pinner and Voldrich. Since agglutination and growth inhibition were not parallel, further studies were made to investigate what other factor in the pleural effusions might account for the inhibitory action. The next question to be answered was whether an active inhibitory factor could be shown, or whether the effusions constituted merely a poor medium. TABLE 3 Effect of inactivation on growth of S. aureus in pleural effusions 345 NVSER OF PLEURL ER OF COLON PER CC. NUMBER OF COLOMS PER CC. RAO EFUSON N NoN-NACTVAT u N NACTVAT UD ()/() 59 75,5, 7,5, 1:1. 21 7,4, 7,4, 1:1. 61 2, 3,4 1:1.7 62 7 1,45 1:2. 58 11 2 1:2. 81 1,6 4, 1:2.5 72 7,25, 2,, 1:2.5 66 1, 3, 1:3. 62 1, 3,4 1:3.4 68 4, 1,5, 1:3.7 67 2, 9, 1:4. 6 825 4,15 1:5. 76 5, 28, 1:5.6 79 14, 85, 1:6. 92 28, 2, 1:6. 54 4 3,2OO 1:8. 6 1, 88, 1:8. 63 75 6, 1:8. 34 5,6 43, 1:8. 69 1, 96, 1:9. 64 4 4, 1:1 74 6, 6, 1:1 78 9, 168, 1:18.6 77 3, 6, 1:2 82 2,4 6, 1:21 8 15 8, 1:5 nactivation A number of pleural effusions were inactivated and compared with the nonheated effusion and broth. Ten ml. lots of several pleural fluids were inactivated for 3 minutes at 56 C. in a water bath. They were inoculated in the usual way with S. aureus simultaneously with the non-inactivated fluid, a broth control, and a seeding control on agar. nactivation diminished the growth inhibition slightly but persistently. This definite trend is shown in table 3. The degree of decrease of the inhibitory

346 ALCE B. TOBLER AND MAX PNNER power bears no relation to the degree of inhibitory power in the non-inactivated fluid. Broth controls were run with these experiments with essentially the same results reported above. Shaking t seemed possible that the agglutinative growth prevented free multiplication of staphylococci in effusions. n a number of effusions, agglutination during growth was therefore prevented by shaking the culture tubes for 24 hours, starting immediately after inoculation. This was done at a fairly constant room temperature. One broth control was shaken with the pleural fluid, one broth tube and one pleural fluid tube were kept as controls at room temperature, and a second set of broth and pleural fluid controls were incubated at 37 C. Enhancement of growth was produced when the pleural effusions were shaken for 24 hours. Some of the results of these experiments are shown in table 4. TABLE 4 Effect of shaking on growth of S. aureus in pleural effusions and in nutrient broth PLEURAL EFFUSONS NUNT BROTH ncubated at Shaken at room Not shaken at room ncubated at Shaken at room Not shaken at room 37C temperature temperature 37C. temperature temperature 1' 4, 1,5 2 1} 27,, 5,, 1, 45, 1,5 16,, 6, 25, No growth 28,, 387,, 15,, 55, 18, 2 1,, 1,7, No growth 256,, 25,, 32,, 5, 185, 9 As these figures indicate, there was consistent increase of growth in the shaken broth as compared with the control at room -temperature. t was thought that this increase in growth was due to a rise in temperature in the shaken tube, which was about 2C. in 24 hours. t can be seen that the figures for shaken broth and incubated broth are not significantly different. n the pleural fluids, there is not only a marked difference between the shaken and the non-shaken tube at room temperature, but all shaken tubes show a definitely better growth than the incubated and not-shaken pleural effusions. t would then appear that the agglutinative growth probably plays some role in the inhibitory action of pleural effusions. Nutritive quality of pleural effusions Further experiments were done in order to find out whether the inhibitory action of effusions is due to the nutritive qualities of these effusions. t was argued that under the assumption of poor nutritive qualities 1. Dilution of effusions will yield an even poorer growth; 2. The addition of a minimal amount of growth-sustaining broth to pleural

EFFECT OF PLEURAL EFFUSONS ON GROWTH OF S. AUREUS effusions will produce a better growth than that obtained in a comparable mixture of minimal amounts of broth in saline. Growth of S. aureus was observed in undiluted pleural fluid compared with that in various dilutions of pleural fluid in saline (1:2, 1:4, 1:1, and 1:2). The results were not quite uniform. n 7 out of 17 pleural effusions, growth was definitely enhanced when the fluid was diluted 1:2 or more up to 1:2; 7 other fluids showed practically no change in growth and in only three effusions did diminution of growth occur. These results are not uniform enough to allow any definite conclusions but they suggest the presence of an actively inhibitory factor. t was found that in nutrient broth diluted 1:1, 3, to 2, staphylococci were present per ml. following the usual inoculation 'and incubation. n 11 experiments, 1 per cent of broth was added to 1 ml. pleural fluid and the growth of S. aureus in this mixture was compared with that in 1 percent of broth in 1 ml. saline. n all experiments, growth was poorer in the effusion-broth mixture than in the saline broth mixture, the ratios varying from 1:2 to 1:29. n a twelfth fluid from a non-tuberculous patient, the ratio was reversed, being 2: 1. These results, again, are indicative of an active growth-inhibitory factor in tuberculous effusions. Three strains of S. aureus, used occasionally, in parallel experiments with strain gave in each instance similar results. The ph of 49 pleural fluids ranged from 7.2 to 9. and bore apparently no relation to the agglutination titers nor to the inhibitory power of the respective fluids. SUMMARY AND CONCLUSONS Forty-nine tuberculous effusions from 39 patients were tested in 88 experiments as culture media for Staphylococcus aureus with control series in nutrient broth. All fluids examined allowed, according to colony counts, a considerably poorer growth than nutrient broth. The ratio between the growth in tuberculous effusions to that in nutrient broth varied from 1:2 to more than 1:1,. Only three of these fluids had a ratio lower than 1:1. Out of seven additional non-tuberculous fluids, three had ratios as low as 1:2 or 1:3. The remaining 4 non-tuberculous fluids had ratios of between 1:4 and 1:1,. With the exception just stated, no correlation could be demonstrated between the degree of inhibition on one side, and the clinical and bacteriological characteristics of the effusions on the other side. The degree of inhibition is not related to the agglutination titer of pleural effusions for S. aureus. Growth inhibition in tuberculous pleural effusion is apparently due to an actively inhibitory factor and not to deficiency in nutritive qualities because: 1. By adding a small percentage of nutrient broth to effusions, the growth is poorer than in the same percentage of nutrient broth in saline; 347

348 ALCE B. TOBLER AND MAX PNNER 2. Dilution of effusions by salinie tends to decrease their inhibitory action; 3. nhibition is diminished by shaking the effusions during the period of growth; 4. nactivated effusions inhibit less. A small number of non-tuberculous fluids examined had a less marked inhibitory action. The results which were obtained in vitro seem interesting in view of the frequent development of fluid in patients treated with pneumothorax and the relatively rare occurrence of secondary infections with S. aureus. REFERENCES PNNER, M., AND VOLDRCH, M. 1932 Staphylococcus aureus agglutinins in tuberculous effusions. J. nfectious Diseases, 5, 143-145. PNNER, M., AND VOLDRCH, M. 1932 Derivation of Staphylococcus albus, citreus and roseus from Staphylococcus aureus. J. nfectious Diseases, 5, 185-22. Downloaded from http://jb.asm.org/ on June 19, 218 by guest