CONCERNING PROMPT SURGICAL FOR INTESTINAL PERFORATION IN TYPHOID FEVER WITH THE RELATION OF A CASE.
Only ten years ago Professor Eulenburg I wrote in an article on "Abdominal-typhus" relative to the question of operative intervention in cases of intestinal perforation, "Die Operation ist in diesen Fallen fast durchweg gut und schnell gegliickt; man hat nach der Laparotomie das Loch gefunden und schliessen konnen ; trotzdem sind alle Patienten rasch zu Grunde gegangen, und wir haben uns die Frage vorgelegt. oh iiberhaupt em n Typhuskranker im Stande sei, den Eingriff einer Laparotomie, nachdem eine Perforation erfolgt, zu ver tragen." At the time of this expression of uncertainty, only one authentic that of Wagner, was known to have re covered after laparotomy, and up to the present day a cor responding sentiment prevails, chiefly in France and Germany, among physicians, who are naturally reluctant to surrender with any degree of promptitude these cases to surgical meas ures which have seemed to be little more than the occasion of a hurried exitus. Since the papers published in 1897 by Monod and Vanwerts 1 in France and by Geselevich and Vanakh 2 in Russia, practically no contributions have been made to the subject in continental literature, and at that time the cases which had been reported were few and the results far from encouraging.
In marked contrast is the feeling which prevails to-day in America, where, according to the statistical tables, more than two-thirds of all the recorded operations seem to have been performed, and the rapid accumulation with each year in the number of reported cases gives evidence of the increasing degree of confidence with which physicians, in this country at least, have come to regard operative intervention in this serious complication. .
It is a natural sequence that there should have been a corresponding diminution in the mortality rate due to a greater familiarity with the surgical problems offered. In 1898, Dr. Westcott, for that book 3 which has done so much to stimulate interest in the surgical aspects of the complications of typhoid fever, collected eighty-three cases, covering a period of more than ten years, which gave a recovery rate of 19.3 per cent. By the end of the following year. alone Dr. Keen 4 had cogni zance of seventy-five additional cases, showing a considerable improvement in the percentage of recoveries (28 per cent.), and there seems to be little doubt but that this number will again be doubled during the present year. It must, however, be confessed that statistics, for the compilation of which one must depend chiefly upon the bibliography of the subject, will not always represent the actual results, and in all probability the true mortality rate is in excess of that which these figures would indicate, since it is presumable that individual instances of success after such an operation will with greater likelihood find their way into the literature than the isolated cases of fail ure. Furthermore, so many divers conditions must necessarily be represented in such a statistical résumé that the true relations of the question are difficult to estimate ; conditions varying all the way from the late evacuation of a circumscribed col lection of pus, possibly of appendicular origin and associated with no demonstrable intestinal lesion, to the hopeless case of generalized peritonitis in which the question of the location of the perforation or its association with_ typhoid plays little part. Be this as it may, however, to have cognizance of thirty-seven actual cases of recovery after perforative lesions of any sort in typhoid fever is sufficient cause for congratula tion, no matter what the number of failures may have been, since the complication left to run its own course is almost inevitably and promptly fatal. It is to be expected that the local mortality percentage after this operation will be lower even than that shown in the statistical tables in the favorable surroundings of individual hospitals in which risks attending operative procedures are reduced to a minimum, and in which the sentiment on the part of the medical staff is such that prompt surgical intervention in acute abdominal complications of a doubtful nature is regarded as the therapeutic measure worthy of greatest consideration. At the time of this writing, of the twelve cases of perforation which have occurred in Dr. Osler's clinic at the Johns Hopkins Hospital during the past few years and which have been operated upon, five have recovered, a percentage of 41.6. Furthermore, there is little doubt, in the light of our present comparative familiarity with the operative possibilities in perforation, but that some of our seven fatal cases would have been saved to-day, when inter vention is undertaken at a much earlier period than formerly, without waiting for an absolute demonstration of the fact of perforation by the presence of a generalized peritonitis, which of itself, in the majority of cases, renders futile all surgical procedures. It has been emphasized heretofore 1 that a dis tinction should be made in the statistical tables between opera tion for intestinal perforation in typhoid fever and operation for acute diffuse peritonitis consequent upon such a lesion; the latter condition being the one which up to recent times the surgeon has usually had to confront, the diagnostic signs of which, such as collapse, the presence of free fluid and gas, with the obliteration of liver dulness, the cessation of intestinal peristalsis, etc., occupy the attention of most writers on the subject. Perforations in individuals suffering from pro found typhoidal infections may doubtless occur, as Dr. Osler states, and, unattended by any recognizable symptoms, be only brought to light post-mortem ; but I am confident that to-day, when no abdominal pain, tenderness, or rigidity of the pa rietes whatsoever in typhoid cases is passed by without com merit, that such a calamity is much more unlikely to occur than formerly, and that in the future the diagnosis will be made in these obscure cases in the operating-room rather than on the autopsy table. For example, one of our seven fatal cases operated upon three years ago comes in this category, practically an ante-mortem exploration having been made, dis closing three large perforations and a peritonitis evidently of long standing. It is apparent from the hospital record of this however, that symptoms had been present for sev eral days which to-day would be regarded as sufficient to justify operation, symptoms to which no particular importance was attached, and which, as not uncommonly is the be came less apparent with the intoxication of a generalized peri tonitis. Similarly, the analysis by Shattuck, Warren, and Cobb 1 of the twenty-four cases which have occurred in the Boston hospitals during the past five years and have been treated surgically, has shown that abdominal symptoms ante dating the so-called diagnostic symptoms of perforation, which are usually those of general peritonitis, have been present, according to the records, in the majority of cases.
Three of our other early fatal perforations probably would have been saved had an immediate exploration under local ananthesia been considered possible, as is the case to-day. As it was, they were given favorable prognosis, were explored at what seemed to us at the time an early hour, and survived the operation and closure of the perforation for a day or two, to ultimately succumb to an infection which we were a little too late to effectually ward off. As will be emphasized later, there are certain bacterial forms of peritonitis (chiefly strep tococcal) which it is absolutely essential to check during the first three or four hours of their progress, and it is impossible to tell, until the abdomen has been opened and a microscopical examination made of the extravasated material, what cases are destined to suffer from an infection with virulent micro organisms ; hence the need of prompt intervention in all. It is a not uncommon experience for us to keep our operating room ready for immediate use when there is a suspicious "typhoid abdomen" in the medical wards. At the present time, bearing in mind the possibility of recognition of pre perforative symptoms, the most faithful watch by both house officers and nurses is kept over all typhoid patients whose abdominal parietes present any peculiarities whatsoever sug gestive of peritoneal lesion ; and it is hardly possible that a perforation or a threatened perforation can escape notice even in patients in profound typhoidal state. I am fully convinced that the adoption of precautions such as we have taken will in the future, in similar surroundings, be the occasion of saving so or 6o per cent. of all cases of typhoid perforation.
Owing to the extraordinary difficulties of differentiation at an early hour between perforation and some other com plications of the fever, to obtain such a recovery rate will probably mean the occasional exploration of an unperforated case; but it is to be hoped that in a previous communication 1 it has been demonstrated that this should be unattended by untoward complications. In two instances of recent cocaine exploration an acutely inflamed appendix was encountered and appendectomy performed, the operation being indicated whether in association with typhoid or not. On another oc casion the abdominal symptoms were due to the involvement of the diaphragmatic pleura by a pneumonia of which no physi cal signs were apparent at the time of operation. Again, an exploration was made for acute symptoms simulating perfora tion associated with a double iliac thrombosis. Such errors, however, should become less and less frequent, and, under the operative methods which have been described, an explora tion without general narcosis should have no influence one way or another upon the progress of the actual complication ; it merely eliminates the presence of a possible perforation, to await absolute demonstration of which greatly jeopardizes the patient's chances of recovery.
Of the twelve cases above mentioned all have been re corded by Dr. Finney or the writer, with the exception of the three most recent, one of which is made the occasion of this communication. The other cases will be published subse quently by Dr. Mitchell.
Hospital No. 30,671. Typhoid fever with abdominal symptoms from onset. Local peritonitis with pre-extravasation symptoms supposed to be due to a small hemorrhage. Subsequent laparotomy (forty-five hours later) under local anesthesia for symptoms of collapse. Closure of perforation. Drainage. Sec ond perforation into wound. Temporary fistula. Recovery.
The patient, a German boy, twenty years of age, who had been in the country only two weeks, entered Dr. Osler's service on the 29th of May, 19oo, complaining of "diarrhea and vomit ing." His mother and one sister have recently died of typhoid fever. His personal history was without note ; he had always been well.
His illness began four days before his admission to the hos pital, at which time he had been obliged to give up work (he was a baker by trade) and go to bed on account of weakness. He had had some nausea and vomiting on the following day with fever and considerable diarrhcea ; five or six movements, he believed.
Physical examination on admission showed a fairly well nourished young man with a natural appearance which could hardly be called typhoidal. His temperature was iol ° F.; his pulse not accelerated and of good quality. The only complaint which he made at the time was of slight abdominal pain.
The medical note on the abdomen made at the time of ad mission reads as follows : " Wall is held rather tense, but there is no spasm ; slight distention with a general tympanitic note. The chief pain is referred to the left hypochondriac and lumbar regions. The spleen is readily palpable one and a half centimetres below the costal margin ; its edge is firm. There are no rose spots." The leucocytes at this time were 8800.
May 30. Leucocytes, 5000.
From this date the patient continued in a febrile state suffi ciently akin to typhoid to justify the usual precautions with regular administration of baths, etc., though a positive Widal reaction was not obtained for several days. Various notes which were added to the history during the ensuing four days evidence the fact that some abdominal pain and rigidity were always pres ent, unassociated, however, with any apparent tenderness on pal pation. The boy invariably "took his tubs" badly ; often com plained of abdominal pain during the immersion, and was usually somewhat cyanosed after it ; once apparently, from the nurse's description, quite collapsed. On in occasion similar to this, though with somewhat more marked symptoms, a surgical consultation was held for the first time, when the following note was dictated.
June 4 (12.30 "The patient is somewhat cyanosed ; his extremities are cold ; he is groaning with abdominal pain and complaining of the tub, from which he has just been removed. He is not sweating. His abdomen is flat, held uniformly rigid, and some slight muscle spasm may be elicited by pressure in the right iliac fossa, where is located his chief complaint of pain and tenderness. There is no dulness in the flanks. The liver dulness does not reach the costal margin by two fingers' breadths. Rectal examination is negative, though disclosing a full bladder. The patient in the next bed states that the boy has hiccoughed once or twice, and a wet stain on the pillow from a mouthful of fluid which he spat up is very acid in reaction." The leucocytes at this time were io,000, the last count (May 30) having been only 5000. There was nothing peculiar in the patient's attitude. His pulse was at the time 110, of rather poor quality, and his tempera ture, which two hours before had been 103° F., had fallen four degrees.
Laboring under the disadvantage of not having seen and ex amined the patient before, and thus being unacquainted with the usual state of the parietes, which under pre-existing circumstances were said to have been uniformly rigid, in spite of the encourage ment of the medical attendants, I refrained from an immediate exploration, and apparently was justified in this decision, since during the next few hours the disturbing symptoms practically cleared up. During the night a bloody stool (200 cubic centi metres) was passed, which seemed to satisfactorily explain his upset, and on the following day his condition differed in no re spect from that of the preceding.
This upset, which, in the light of subsequent events, un doubtedly was occasioned by a pre-extravasation stage of peri toneal involvement, represents the most interesting period of the patient's illness, and, at the risk of interrupting the thread of his history, may bear analysis here. Why, when confronted by such a characteristic symptom-complex of perforation (as this appears to be on paper, at all events), intervention was withheld, may be explained by the mildness of the individual symptoms, and by the unjustified belief, on the part of the surgeon, in his ability to make what is at this period, Without opening the abdomen, an almost impossible diagnosis.
The following notes were added to the history during the few hours succeeding, and reading between the lines it is easily seen why operation was refrained from.
( I) "The patient is in some degree collapsed; is cyanosed and with cold extremities. He, however, has just been removed from the bath, from which he was on a previous occasion taken in somewhat the same condition. There is no sweating, such as is often seen in the collapse of perforation. (2) There has been a drop in temperature of several degrees (to 99° F.). A similar drop has occurred after a 'tub' on a former occasion. The temperature, as at that time, has rapidly returned to its usual figure, 104°. (3) Vomiting. The patient was not seen to vomit. His cheek and a spot upon the pillow were damp with a fluid which was found to be acid in reaction, suggesting that it came from the stomach. (4) A board-like abdominal rigidity is pres ent. This, however, is said to have been the usual condition after the bath. There was no especial muscle spasm. Abdominal move ments were present with respiration. (5) Pain. Not especially characteristic, and no more complaint than on previous occasions. (6) Leucocytosis of io,000 counted immediately after his first symptoms. This number has fallen rapidly in the course of a few hours to 4000, which probably represents his usual 'typhoid num ber. (The temporary relative leucocytosis and the history of previous complaints of abdominal pain were the two most sus picious symptoms.) (7) Attitude not peculiar. Knees not drawn up. No restlessness. Patient is quiet between our exami nations. (8) There is considerable tenderness on pressure, espe cially in the right iliac fossa. No dulness on percussion, and several other typhoids in the ward seem equally tender. (9) Rectal examination. Tenderness was considerable in the recto vesical region. The bladder was quite full, and the patient was catheterized (500 cubic centimetres of high-colored urine). Evacuation apparently caused no pain, and after it much deeper abdominal palpation was allowed than before." June 4 (continued). During the night the leucocytes fell as follows. When first counted after the onset of the symptoms at midnight, they were to,000, as stated ; at 1 A.M., 9900 ; at 2 A.M., 5400; at 3 A.M., 5200, and at 4 A.M., 4200. By morning the patient was so much better, and so free from abdominal symp toms in any degree more marked than during his first few days in the hospital, that he was demonstrated to the clinic as a case of simulative perforation. At 3 P.M. he passed a few blood-clots, and at 5 P.M. 200 cubic centimetres of blood clotted and liquid. This was believed to be the "tell-tale" of the disturbance the night before.
June 5. Patient passed a fairly comfortable night. His condition apparently was unaltered. During this day he seemed dull and drowsy, though with no appreciable change in abdominal symptoms. Late in the afternoon, however, his general condition became somewhat worse. At 5.30 P.M., the leucocytes, which had not been counted previously during the day, were found to be 17,000, and he vomited a small amount after taking nourishment.
At 7 P.M. the leucocytes were 14,500; at 8 P.M., 14,000. A positive Widal reaction had been obtained for the first time on this date. He rapidly grew much weaker ; he was very dull ; his pulse very feeble, 140. Evidently, some critical change for the worse had taken place, and, though there was apparently less indication of a peritoneal lesion than there had been forty-five hours before at the time of the sudden upset above described, an exploratory laparotomy was with reluctance determined upon.
June 5, 9 P.M. Operation under local anwsthesia after one eighth grain of morphia.
Exploratory laparotomy. Perforation in ileum with recent extravasation into general cavity. Closure of perforation. Irri gation. Drainage. No shock.
With Schleich's infiltration method ( 1 to moo cocaine), an oblique incision was made over the situation of the appendix. On opening the general cavity, evidences of a local peritoneal reaction of many hours' duration were encountered. The cmcum, lower part of the ileum, and the parietal peritoneum over the right iliac fossa were covered with fibrin, the oldest and most marked degree of peritonitis being in this situation. The ap pendix, which was involved in this process, showed no external evidences of ulceration. Several ounces of pus were present be tween the coils of bowel in this corner of the abdomen. After carefully wiping out this exudate and taking cultures the ileum was then examined, and almost the first coil which was drawn out into the wound showed a perforation about one-half by one centi metre in its diameter. This was situated possibly eight centi metres from the cmcum. It was immediately closed by a double row of sutures. Beyond this perforation, the ileum, which was deeply injected, showed evidences for several centimetres of un derlying deep ulcerations, many very thin areas corresponding to Peyer's patches threatening immediate perforation. It would have been impossible to have inverted all of these areas. The pelvis contained several ounces of turbid seropurulent fluid, which a microscopical examination showed to contain only leucocytes and numerous bacilli ; no streptococci. The general cavity was irrigated for half an hour with Ringer's solution by means of large tubes introduced among the coils in various directions. Closure. Owing to the threatening appearance of the lower part of the ileum, the omentum, which fortunately was long, was wrapped about the most suspicious looking coil, ten or twelve centimetres in length, and caught in place by one or two fine in testinal sutures. This portion of bowel was then placed parallel with the abdominal wound and held loosely in position by means of gauze, with the view of having it under control should it be desired subsequently to establish an artificial anus in case of in testinal paralysis from peritonitis, and also to enable a subsequent perforation, should one occur, to have exit by way of the wound. The two ends of the incision were closed. During the entire operation, which was performed leisurely and lasted over an hour, the patient made absolutely no complaint of pain, although greatly frightened when taken to the operating-room and object ing to the preliminary hypodermic of morphia and the first inser tion of the infiltration needle. His condition, especially during the long irrigation, greatly improved, and he left the operating table with a much better pulse and general appearance than he had brought to it.
The postoperative history may be dismissed with a few words. The day after the operation there was considerable dis tention, which was relieved by turpentine enemata. His post operative leucocytosis persisted for a couple of days, apparently due to an infection of that part of the parietes which had been closed. As is often observed, an abdominal incision which has been soiled during such an operation takes care of micro-organ isms less well than the underlying peritoneum. The wound had to be laid open its full length, disclosing an infection of the muscle and panniculus. The postoperative observations on the leuco cytes are as follows : June 5, 8 P.M., 14,000 (before operation).
June 6, II A.M., 17,400; 10.30 P.M., 15,600.
June 7, II A.M., 14,000; 9 P.M., 9000.
June 8, 10 A.M., 7000; 6 P.M., 4000.
June 9, 6 P.M., 4500.
June io, 5 P.M., 3000.
June II, 6 P.M., 3000.
Two days after the laparotomy the patient's abdomen was soft and free from rigidity for the first time since his admission to the hospital. On June 14, eight days after the operation, at a point of the bowel exposed in the lower angle of the wound just proximal to the coil which had been covered by the omental graft, a thin area gave way, and a fmcal fistula developed which persisted for some time. The abdominal wound, owing to the necessity of abandoning the sutures at the angles, had gaped considerably, and it was possible to keep under observation the underlying loop of ileum. The omental graft played its part well and possibly saved that part of the bowel from other perforations ; an imita tion of nature's method of protecting such surfaces. The wound had closed, and the patient was discharged from the hospital early in August.
The necessity of acknowledgment in this particular case of an error of judgment on the part of the operator, and his failure to take advantage of that very stage which has been heretofore described as the elective one for intervention, per haps emphasizes the more what may be given as the rule to follow in these cases, namely, "when reasonably in doubt, explore." By good fortune alone was delay in this instance not followed by disastrous results. The character of the pen tonitis encountered at the operation leaves it beyond doubt that a peritoneal lesion had existed, possibly for some days, and had been protected by neighboring adhesions. At the time of the exacerbation of the abdominal symptoms there probably had been some increase consequent upon the bath in this local reaction which again subsided. An analogous condition is often seen in the temporary spread of a circum scribed appendicular peritonitis. The occasion of final giving way of adhesions with extravasation of pus and intestinal contents into the free cavity was unassociated with pronounced abdominal symptoms, only with general symptoms of collapse. It probably occurred not many hours before our exploration. Just such cases as this were the ones which formerly were brought to light at the autopsy table, comparatively mild and neglected preliminary abdominal symptoms merging inappre ciably into those of an unrecognized general peritonitis.
The frequent association of haemorrhage from the bowel in cases of perforation has been brought out by the careful analysis of Shattuck, Warren, and Cobb, it having occurred in more than one-third of their series. It is possible, therefore, that this symptom should have been regarded by us rather as favoring the presence of a perforation than taken in itself as the occasion of the symptoms.
The case furthermore may be offered in illustration of that fact which the statistical study of the successful cases has brought to light, namely, that the greatest percentage of re coveries has occurred among those cases operated upon during the period from eight to twenty-four hours after evidence of perforation. A natural explanation is that patients who have for that length of time survived the perforation, and still are regarded as fit subjects for surgical intervention, either have suffered from a mild bacterial infection or else the process has been one slowly progressive and well combated by peritoneal reaction. It is interesting in this connection to learn from the statistics the number of recoveries which have attended opera tions performed at a comparatively late hour (all unfortunately without bacteriological observations), and to see of this number in what a large percentage the operation consisted in exposing a circumscribed abscess, and in how many instances, therefore, failure to find the perforation is recorded. This naturally means a very mild bacterial process occasioning a localized peritonitis of the adhesive variety. To await the reaction which may follow the collapse of perforation and extravasation is another way of distinguishing between the favorable and un favorable cases, viz., those which improve sufficiently to be considered fit subjects for operation after several hours and those which do not. The latter are presumably rarely operated upon, since the condition, for want of a better term called shock, merges imperceptibly into that of a profound septicxmia, which is even more unfavorable from an operative stand-point.
Fortunately, for this particular patient, the actual perfo ration was unassociated with the extravasation of organisms other than of bacillary types ( B. coli communis was alone isolated). Had the existing flora of the intestine been rich in streptococcal forms, an operation only at the time of the pre extravasation symptoms could with any degree of probability have warded off a fatal issue.