GAS GANGRENE AND GAS INFECTIONS. January 14, Section on Orthopaedic Surgery, The Mayo Clinic

Transcription

1 GAS GANGRENE AND GAS INFECTIONS BY RALPH K. GHORMLEY, M.D., ROCHESTER, MINNESOTA Section on Orthopaedic Surgery, The Mayo Clinic Any one who is called on to treat injuries incurred in such a manner as to permit the entrance of soil into the wound is probably more or less familiar with the condition commonly known as gas gangrene. The presence of the organism or organisms causing this condition should always be suspected, and, in the event of evidence of such infection in a compound fracture or other wound, it would seem wise to regard it as a possible etiological factor. It has been thought that the diagnosis in such cases could be made by the odor, which is said to be characteristic. In many instances, this is true ; however, in some cases, it is not true and, if one were to depend on odor alone, one might fail to make the correct diagnosis. The reports of two cases will serve to illustrate this point: CASE 1. A man, aged sixty years, caught his foot in a corn shredder. He suffered a compound fracture of the first and second metatarsals and of the phalanges of the first and fifth toes on the right foot. D#{233}bridement was performed and the fractures were reduced. Tetanus antitoxin (1500 units) was given. On the fourth day, the pulse rate and temperature increased and there was severe pain and discoloration of the toes. Smears from the wounds revealed organisms resembling Clostridium welchii, and the leg was amputated above the knee. The patient recovered. CASE 2. A man, aged fifty years, had fallen from a load of hay a week prior to his admission to the hospital and had injured his thumb, cutting the palmar surface. He thought that he had dislocated it. No improvement had resulted from treatment at home. On admission, there was a firm, oedematous swelling of the thumb with some yellowish discoloration at its tip, suggesting pus under the skin. Incision, however, failed to reveal any pus. The patient was given tetanus antitoxin. The oedema gradually spread to the hand and forearm and finally to the shoulder. Two blood cultures were negative. There was never any suggestion of gas in the exudate and no crepitus. There was no odor. The temperature on admission was 101 degrees; it dropped to normal on the second day and remained there during the eight days in which the patient was in the hospital, going below normal on the last day. The pulse rate was not much elevated until the day of the patient s death, when it rose sharply. At the end of the seventh day in the hospital, or about two weeks from the time of the injury, the patient rapidly went into collapse and died within twenty-four hours. Post-mortem examination revealed nothing remarkable except that cultures from the wound showed an organism of the Vibrion septique type. Thus we have two very different clinical pictures caused by very similar organisms,-that is, Clostridium welchui, or the gas bacillus, and Clostridium oedematis maligni, or Vibrion septique. Certain facts regarding the organisms may be of interest in correlating the clinical findings in these cases. For this reason, a group of anaerobic spore-bearing * Read before the American Academy of Orthopaedic Surgeons, New York, N. Y., January 14, VOL. XVII. NO. 4, OCTOBER

2 908 R. K. GHORMLEY bacilli (after Weinberg and Sguin ) have been tabulated (Table I). According to most bacteriologists, several of these organisms are often found in cultures from the same wound. Some of them are unquestionably pathogenic, but others are saprophytic. It is interesting to note that the gas is produced mostly by a non-pathogenic or saprophytic organism. Clostridium welchii, however, is the organism found most frequently and is one of the most pathogenic. Vibrion septique, or Clostridium oedematis maligni, while not found I SOME ANAEROBIC BACILLI (WEINBERG AND SIGUIN 1) 1. Bacillus perfringens-veillon and Zuber, 1897 (pathogenic). Bacillus aerogenes capsulatus-weich and Nuttall, Bacillus phiegmonis emphysematosae-eugene Fraenkel, Bacillus welchii-migula, Vibrion septique-pasteur, 1881 (pathogenic). Bacillus des malignen Oedem-Koch, Bacillus sporogenes-metchnikoff. Most frequent of the organisms in war wounds. Cause of fetidness (saprophytic). 4. Bacillus putrificus-bienstock, 1884, as Bacillus putrificus coli. Occurs in intestines of the cadaver (saprophytic). 5. Bacillus tertius -Henry. Found in war wounds,-that is, after Bacillus perfringens and sporogenes. 6. Bacillus bifermentans-tissier and Martelly, Found frequently in markets in meats in first stages of putrefaction (saprophytic). 7. Bacillus oedematiens-weinherg and S#{233}guin. Pathogenic for man and animals. 8. Bacillus fallax-weinberg and S#{233}guin. Found in some cases of gas gangrene. 9. Bacillus aerofoetidus-weinberg and SCguin. Found in four cases of gas gangrene. 10. Bacillus histolyticu.s-weinberg and S#{233}guin. Found in eight cases of gas gangrene (pathogenic). so frequently, is in many instances even more serious. One author states that infections by this organism cannot be cured and are 100 per cent. fatal. Another states that this organism is found in 10 per cent. of cases of gas gangrene. Whether or not this is true, it is important to know that infections by this organism alone may occur, and that in these infections the clinical picture is not at all the same as gas gangrene. Bacillus sporogenes is very commonly found in cultures. It is not, however, pathogenic, but is responsible largely for the putrid odor so commonly associated with these wounds. Bacillus oedematiens is a THE JOURNAL OF BONE AND JOINT SURGERY

3 GAS GANGRENE AND GAS INFECTIONS 909 pathogenic organism of importance, as is also Bacillus histolyticus. When injected into animals, the latter causes extensive destruction of the soft tissues down to the bone, and it may be regarded as responsible for much of the destruction of tissue when it is found in wounds. Other organisms of less importance occur ; some are pathogenic, some saprophytic. Thus it will be seen that, if this subject is to be approached from a scientific standpoint, it must be remembered that there are several or- II CASES OF GAS GANGRENE AND INFECTIONS WITH GAS BACILLI (CL05TRIDIUM WELCHII) ENCOUNTERED IN FIVE-YEAR PERIOD Causative (or Preceding) Factors No. of Cases Following trauma: Compound fractures 9 Injuries to soft parts 7 16 Metastasis, primary source unknown 3 Following amputation: Thrombo-angiitis obliterans 3 Arteriosclerosis (with diabetes) 2 5 Following operation: On stomach 3 For intestinal obstruction. 2 Colostomy and resection 1 Cystostomy 1 For perinephritic abscess 1 8 Abortion 1 Total 33 Patients recovered : 19 (57.5 per cent.) Patients died: 14 (42.5 per cent.) ganisms to be dealt with, some of which are pathogenic and some of which are not. There may be cases in which infection is produced by this group of organisms without the presence of the supposedly characteristic odor, and again there may be sonic infections in which there is no evidence of gas (Table II). These thirty-three cases of gas gangrene and gas infection which were encountered at The Mayo Clinic in a five-year period represent a variety of causative factors. As would l)e expected, compound fractures produced the largest number of infections, although wounds of the soft parts were not far behind. Either of these may be said to be the commonly known types of trauma from which gas infections may develop. Any one doing many amputations for gangrene due to thrombo-angiitis obliterans, or for gangrene due to arteriosclerosis with diabetes, has probably seen the VOL. XVII, NO. 4. OCTOBER 1935

4 910 R. K. GHORMLEY occasional instance of gas infection in the stump. It is a disturbing cornplicationwhichmaytake placefrequentlyenough tojustifyadministration of a prophylactic dose ofga.s-gangrene a ntitoxin before amputation is undertaken. Besides these groups, two cases of spontaneous abscess in the soft parts and one case of pneumonia, in which Clostridium welchii was isolated from the sputum, are of considerable interest. In the first two cases, there was no history whatever of injury. In one case, there was an abscess of the buttocks of a man who was very ill with an upper abdorninal complaint that was diagnosed and later proved to be the result of a ruptured gallbladder. Whether or not this was the primary source of the infection with gas bacilli cannot be proved, although it must be presumed that the probability is that the infection entered the blood stream from the gastro-intestinal tract in some way and thus reached its destination. The abscess in the other case appeared in the thigh of a man who did not have any history of gastro-intestinal disturbance. It had apparently been present for some time; at least, symptoms had been noted for about two months. When the abscess in this case was opened, a mixed culture wa found, but the presence of the gas bacillus was clearly demonstrated bacteriologically. These two cases illustrate the possibility of the occurrence of metastatic abscesses which must have been implanted by infection of the blood stream and yet septicaemia did not develop. No definite primary source could be established. In this connection, the fact should be pointed out that 2 found Clostridium welchii in five of seventyfive gallbladders which were cultured after surgical removal for cholecystiti.s. The organisms were oflow virulence and spores were not demonstrated. Finally, the series includes at least two cases in which recovery followed infection of the blood stream, which had been demonstrated by positive blood cultures of Clostridiurn welchii. One of these patients was a man, aged twenty-three years, who suffered a compound comminuted fracture of the left elbow thirty-six hours before admission to the Clinic. There was also a compound fracture of both bones of the left forearm and a fracture of the middle portion of the shaft of the humerus. Tetanus antitoxin or gas-gangrene antitoxin had not been given. At the time of the patient s admission to the Clinic, his pulse rate was about 130 beats per minute; his temperature, 103 degrees. Cultures were taken from the blood as well as from the wound. Roentgenograms disclosed gas in the wound, and this finding was borne out by inspection of the wound. Multiple incisions were made; the arm was not amputated because the muscles appeared to be in good condition. One therapeutic dose of polyvalent antitoxin was given intravenously; then, in five hours, another dose intramuscularly. The patient rallied from an almost moribund state and in twelve hours was conscious. His pulse rate dropped to 90 beats per minute and his temperature to about 99 degrees. The arm went on to a fully developed gangrene below the site of the fracture in the humerus and was amputated after about forty-eight hours. The patient made an uneventful recovery. Thus it will be seen that these organisms may produce not only the local changes, due to their activity, but metastatic abscesses as well, and in some cases infection of the blood stream also. Without the extensive use of bacteriological cultures, accurate figures as to the incidence of infections of the blood stream and metastatic infections are almost impossible. THE JOURNAL OF BONE AND JOINT SURGERY

5 GAS GANGRENE AND GAS INFECTIONS 911 DIAGNOSIS FFh diagnosis of gas infections must depend, therefore, not only on one s ability to judge clinical findings, but on the laboratory aids as well. If one were to tabulate, in the order of their importance, these diagnostic aids, they would be as follows : (1) pain, which is the most common symptom; (2) swelling, which is the most common sign; (3) elevation of the pulse rate; (4) bacteriological findings,-that is, smears from the wound, cultures from the wound, and cultures of the blood; (5) discoloration; (6) the presence of crepitus in the tissues or of gas in the exudate (not constant) ; (7) a bad odor, which is said to be characteristic, but which again is not a constant sign ; (8) elevation of temperature, which at times, however, is not important ; and (9) the presence of gas bubbles in the roentgenogram of the affected part. Nearly all who have written on this subject agree that pain of severe degree is probably the earliest, if not the most common, symptom found in cases of gas infection or gas gangrene. Accompanying this pain is swelling, which is usually of a firm type, without much fluctuation, until necrosis is well established. Probably one of the most significant of the early signs is elevation of the pulse rate. While this is, of course, not diagnostic of gas infections alone, yet it is one of the earliest and most definite signs of a change in the patient s condition. As a rule, elevation of the pulse rate is out of proportion to the elevation of temperature. When this is observed to be the case, some infection should be suspected, and a smear and culture of the wound should be taken. If the smear shows organisms of suspicious appearance to one familiar with the examination of such smears, the author feels that institution of treatment by administration of gas-gangrene antitoxin is justified. The report of the culture will necessarily be delayed and one should not wait to receive it before commencing treatment when gas infection is suspected. Discoloration of the skin is at first reddish ; it later becomes grayishyellow, and finally, cyanotic. Crepitus is one of the signs looked for most frequently and considered so characteristic. It is characteristic of gas infections, but may not be present in all anaerobic infections, as it has already been pointed out that some of them are not gas producing and yet are as pathogenic. Bubbles of gas in the exudate or pus are usually looked for and, in most instances, are found, but again it should be emphasized that, in spite of the name gas gangrene, there are allied conditions just as pathogenic which do not produce gas. The odor is said to be characteristic; it is interesting to read the various attempts to describe it. One author says the odor is that of rotting meat ; another declares It smells something like a mouse ; a third states The odor is putrefactive, offensive, almost indescribable, but once encountered will never be forgotten ; and still another characterizes it as mousy. It is obvious that an attempt to describe the odor so that any one can recognize it is out of the question. From what has already been said, it must be clear that in some cases the odor may not be so characteris- VOL. XVII. NO. 4. OCTOBER 1935

6 912 R. K. GHORMLEY tic and he who diagnoses the condition on odor alone may come to grief. Elevation of temperature, as has already been pointed out, is not in proportion to the elevation of the pulse rate; in the case of infection by Vibrion septique, the temperature was elevated one day only and throughout the rest of the fatal course of one week remained normal or subnormal. The presence of gas bubbles in the roentgenograin of the affected part is regarded by some as of great value in diagnosis. This may no doubt be III MORTALITY IN CASES OF GAS GANGRENE ENCOUNTERED IN CIVIL PRACTICE BASED ON THE FIGURES OF MILLAR End Result No. of Cases Recovery (48.0 per cent.) 291 Death (47.2 per cent.) 287 Known outcome 578 Unknown (4.8 per cent.) Total 607 regarded as a valuable aid in diagnosis, but it cannot be pathognomonic. Air bubbles are often seen in the roentgenograms of cases of compound fractures, as well as in those of cases of interstitial emphysema. Unless roentgenograms are taken repeatedly and an increase in the number or size of the gas bubbles can be demonstrated, the observation cannot be said to carry much weight in making the diagnosis as far as traumatic lesions are concerned. In reviewing the literature on gas gangrene in civil practice, the available statistics, while small, seem to compare favorably with those at The Mayo Clinic, both as to incidence and types of infection. Millar has reviewed the greatest number of cases (607) and has presented a comprehensive summary of them. Table III is based on his figures. He stated that, according to the Surgeon General s office, the death rate for the American Expeditionary Force in France was per cent. in cases of gas infection. Of 128,765 wounds of the soft parts, gas gangrene developed in 1,389, or 1.08 per cent., and there were 674 deaths. In 25,272 cases which included fractures, gas gangrene developed in 1,329 cases, or about 5 per cent. Boland4 reported a series of cases from the Grady Hospital in Atlanta. Stone and Holsinger5 reported sixty-seven cases from the University of Virginia Hospital in a twelve-year period; the mortality in this series was 32.4 per cent. However, in those cases in which an adequate dose of antitoxin was given as part of the treatment, a mortalityof only 15.3 percent. was noted. THE JOURNAL OF BONE AND JOINT SURGERY

7 GAS GANGRENE AND GAS INFECTIONS 913 Larson and Pulford6, in an excellent review of the subject, reported a death rate of only 13 per cent. in a series of seven cases. Other smaller groups of cases have been presented. The later writers all deal with the importance of the use of gas-gangrene antitoxin in such cases. The consensus of opinion is that antitoxin is of value in the treatment of the infection if administered as soon as the infection is recognized, and the earlier it is used the better. As yet, the statistics are far from sufficient to permit us to base any great claims on them, but the remarkable similarity between the figures presented l)y the various writers and those obtained at The iviayo Clinic indicates a step toward improvement in the treatmeiit of this condition. TREATMENT The treatment. may be divided into folli phases, as follows: (1) recognition, (2) seruni therapy, (3) surgery, and (4) dressings. Recognition of the condition is, of course, not. part of the treatment, but. it is so essential to the correct treatment that for the sake of emphasis it is included here. With the present knowledge and methods of treatment, once the diagnosis is established the chances of recovery are certainly greater than they were at the time of the War. As to the actual treatment, once the diagnosis is established, the author believes that the first. thing is to give gas-gangrene antitoxin in therapeutic doses. The preferable antitoxin is one of the polyvalent types nw prepared. Most of these antitoxins are made to be effective against Clostridium welchii, Clostridium oedematis maligni, Clostridium oedematiens, Clostridium histolyticum, and Clostridium sporogenes. Some variation is found in strength and amount. of antitoxin and in the individual antitoxins. The intravenous method of administration is the most effective for reaching the affected tissues. Some reaction may be observed, following the intravenous use of this antitoxin, but, when it is administered with some saline solution. the effect of the reaction may be reduced. At the Clinic no serious results from these reactions have been observed, but they may be at times quite violent. In our series, an average of tw o doses was given in each case, and, in many instances, t.he intravenous dose was followed in a few hours by an intramuscular dose. It is questionable how many doses are necessary; it is the author s impression that those patients who reacted favorably to antitoxin did so after one dose was given. In the case of septicaemia from Clostridium welchii previously noted, it seems that there is no doubt as to the value of antitoxin, as the patient had a positive blood culture before antitoxin was given and was practically moribund, whereas after antitoxin was given the blood culture was negative and the condition of the patient rapidly returned to normal. No less important than the use of antitoxin is the use of the proper surgical procedures. The state of progress has not yet been reached where these patients can be cured by antitoxin alone, and one must not expect it.. VOL. XVII. NO. 4. OCTOBER 1935

8 914 R. K. GHORMLEY Surgical judgment and execution must play a very important part in the treatment. In cases of severe compound fractures or infections of soft parts, the decision as to amputation must be made. If the extent of the infection is such as to leave doubt that the limb can be saved, it must be amputated at once. However, in cases of localized infection, without extensive involvement, many a limb can be saved. D#{233}bridement will usually be necessary, and, in doing such d#{233}bridement, dead tissue should all be excised. Three factors should be considered in deciding whether or not muscle is viable : (1) if it is red, it may be viable ; (2) if it contracts when pinched, it is usually viable; and (3) if it bleeds freely, it probably is viable. Abscesses should be drained as they appear, and thorough drainage must be established. When wide-spread areas of infection are found, multiple incisions must be made, and an entire muscle may be excised in an effort to stop the progress of the infection. Much diversity of opinion exists as to the relative value of various types of dressings. Those usually advised are potassium-permanganate packs, Dakin s solution, and hydrogen peroxide. It probably makes little difference which is used as long as the wounds are not packed tightly. The looser the bandage, the better. If possible, air should be allowed to reach the wounds for at least a part of the day. Amputations in severe cases should be of the guillotine type, so that there may be no danger of closing in flaps containing the organisms. The use of oxygen has been recommended, even to injecting it around the wound; such use, however, is deprecated as dangerous. RESULTS The results in this series of thirty-three cases are shown in Tables II, IV, and V. The findings indicate a mortality of 42.5 per cent. (Table II). This is somewhat below the percentage in the World War. However, in Table IV, the cases are presented in a manner to show the value of gasgangrene antitoxin. Excluding the group of patients with abdominal involvement, most of whom were hopelessly ill and in four of whom the con- IV CASES OF GA.s GANGRENE Tt INrc rxons WITH GAS BACILLI (CLOSTRIDIUM WELCHII) ENCOUNTERED IN FIVE-YEAR PERIOD End Results No. of Cases Per Cent. Condition recognized and antitoxin given: Recovered Died Condition recognized and antitoxin not given: Recovered Died Condition not recognized and antitoxin not given: Recovered 0 0 Died THE JOURNAL OF BONE AND JOINT SURGERY

9 GAS GANGRENE AND GAS INFECTIONS 915 ditioti was not (liagliosed as such but was recognized at necropsy, the percent.age of those who recovered following the use of the antitoxin is high (Table V). Others have reported similar results with the use of antitoxin. In general it may be said that, with recognition of the condition and a judicious combination of the use of antitoxin and surgery, a. mortality of approximately 15 per cent.. may be expected. V CASES OF GAS GANGRENE AND INFECTIONS WITH GAS BACILLI (CLOSTRIDIUM WELCHII) ENCOUNTERED IN FIVE-YEAR PERIOD End Results (Excluding Abdominal Cases) Per Cent. Antitoxin given: Recovered 86.6 Died 13.4 Antitoxin not given: Recovered 55.5 Died 44.5 As far as the prophylactic use of the antitoxin is concerned, there is little opportunity to give any significant figures as yet. In the present series, one patient had only prophylactic doses of antitoxin, and it was felt. that the infection was much mitigated by use of the antitoxin. The author feels that the only way in which we can arrive at any definite conclusions in this regard is to wait until a large enough series of patients who have had prophylactic doses of antitoxin has been studied to permit a satisfactory comparison with a large series of patients who have not been given prophylactic doses of antitoxin. CONCLUSIONS 1. Gas gangrene and gas infections must be diagnosed early if good results are to be obtained. 2. The multiplicity of anaerobic organisms, with variation in the clinical picture, must be remembered. 3. With the judicious use of polyvalent gas-gangrene antitoxin and surgery, the mortality in such cases should be reduced to approximately 15 per cent.. REFERENCES 1. WEINBERG, M., ET SGU1N, P.: La gangrene gazeuse. BactCriologie, reproduction exp#{233}rimentale, s#{233}roth#{233}rapie. Paris, Masson & C, THORSNESS, E. T.: Bacteriology of Cholecystitis. The Virulence and Spore Formation of Clostridium Welchii. Surg. Gynec. Obstet., LIX, 752, MILLAR, W. M.: Gas Gangrene in Civil Life. Surg. Gynec. Obstet., LIV, 232, BOLAND, F. K.: Gas Gangrene in Compound Fractures, Ann. Surg., XC, 603, STorr1, C. S., JR., HOLSINGER, H. B.: The Diagnosis and Treatment of Gas Bacillus Infection. Virginia Med. Month., LXI, 200, LARSON, E. E., AND PULFORD, D. S.: Gas Gangrene of the Extremities. With Especial Reference to Trivalent Anaerobic Serotherapy. J. Am. Med. Assn., XCIV, 612, VOL XVII, NO. 4. OCTOBER 1935