Ruptured adnexal abscess with generalized peritonitis

RUPTURED

PAUL

ADNEXAL ABSCESS WITH PERITONITIS* PEDOW~Z,

M.D. AND

GENERALIZED

LAURENCE B. FELMUS, M.D.

Brooklyn, New York UPTURE of an adnexal abscess, either ovarian, pyosalpinx or tubo-ovarian, resulting in generalized peritonitis, is one of the gravest conditions encountered by the gynecologist. Despite this serious complication of peIvic inflammatory disease, little has been pubIished regarding its frequency, diagnosis and treatment.‘-9 UntiI recentIy the results in the treatment of patients with ruptured adnexal abscesses were very poor and discouraging. Marked improvement in the surviva1 rate of these patients in the last few years has emphasized the value of early diagnosis and treatment. Since the inception of the University Division at Kings County HospitaI in 1933, there have been thirty-five cases of generalized peritonitis secondary to ruptured adnexal abscesses diagnosed and treated. Rupture of an adnexa1 abscess is one of the less common complications of peIvic inffammatory disease. The majority of standard textbooks of gynecoIogy make no mention of it. The remote possibility of such an occurrence is conceded by Curtis1o who stated that he had never seen a case. Bauer” and Lenormant and Kaufmann12 both reported that ruptured pyosalpinges were found in 8 per cent of al1 patients operated upon for generaIized peritonitis. The ratter authors aIso gave a I per cent incidence of ruptured pyosalpinx in a11 patients operated upon for sarpingitis. From I 935 to I 95 I there were I 6,043 patients admitted to the gynecologic service, of which 39.9 per cent had a diagnosis of inflammatory adnexaI disease, pelvic peritonitis or generaIized peritonitis. In this group there were thirtyfive cases with ruptured adnexa1 abscesses, an incidence of I :4$8 or 0.22 per cent. The youngest patient in the series was thirteen years of age and the oldest forty-seven. Twenty-nine patients were Negro and six were white. In our institution, as in other large

R

county and municipal hospitals, the incidence of pelvic inflammatory disease is more common in the Negro race than in the white. Ruptured adnexal abscesses occurred more than four times as frequentIy among the Negroes, despite the fact that they comprised only 56 per cent of all patients admitted to the ward. Thirteen of the patients had never been pregnant and twenty-two had had one or more pregnancies. Abscesses of the adnexa usuaIly are the sequel of three types of infection, namely, gonorrheaI, pyogenic and tubercuIous. The gonorrhea1 type was the most frequent, comprising 85.7 per cent of the cases in this series. TubercuIous adnexitis was not present in any of the patients. Etiology. The gonococci reach the tubes by way of the mucosa and produce an endosalpingitis. Soon the waI1 of the tube is involved foIlowed with occlusion of the fimbriated end. Exacerbations of the disease and secondary bacteria1 invaders Iead to partial or complete sealing of the isthmus, with subsequent retention of pus within the tubal Iumen foIlowed with varying degrees of distention. Leakage of pus from a partiahy occluded ostium infects the contiguous ovary, the porta of entry being provided by the periodic rupture of the folIicIes. The invoIved tube and ovary then may fuse to form a conglomerate mass, a tuboovarian abscess. (Figs. I and 2.) In the pyogenic or puerperal type of infection the tube and ovary are involved usuaIIy via the lymphatics and bIood vessels. Commonly, it is secondary to an infected abortion which primarily involves the cehular tissues of the parametrium and broad Iigament. In such cases the tuba1 Iumen usuahy is spared and the end resuIt is a perisalpingitis. It is the ovary that most often bears the brunt of the infection as is ilIustrated by the following case. An eighteen year old Negro nullipara had

* From the Department of Obstetrics and Gynecology, of Medicine,

April,

1952

State University of New York at New York City, College and Kings County Hospital, ArookIyn, N. Y.

507

Pedowitz,

FeImus-Ruptured

Adnexal

Abscess

with

Peritonitis

FIG. 1. Microscopic section of small ovarian abscess showing surrounding normal ovarian parenchymn, X 4.0. FIG. 2. hlicroscopic section of tubo-ovarian abscess; note fusion of pyosalpins and ovarian abscess, X 15. FIG. 3. Microscopic section of pyosatpinx demonstrating marked thickness of tubal wall, X 4.0. FIG. 4. Microscopic section of wall of Iarge ovarian abscess with thin shell of ovarian parenchyma, X 80. Compare with Figure 3.

several exacerbations of pelvic peritonitis foIlowing an induced abortion. She was admitted in extremis with generalized peritonitis. She died thirty-eight hours after the acute onset of her last attack. Autopsy revealed a ruptured left ovarian abscess, Ieft saIpingitis, generatized peritonitis and a normal right tube and ovary. (Fig. 3.) An enlarging ovarian abscess practicahy destroys a11 the dorma ovarian parenchyma. Eventually the organ is converted into a mass many times the original size of the ovary. Further increase in the size of the abscess causes progressive thinning of its fibrous wall which readiIy expIains the ease with which rupture occurs. (Fig. 4.) Distention of a pyosalpinx from repeated exacerbations of an infection may lead to an enormous mass since the tuba1 muscuIature allows for considerable stretching. The size the tube must uItimately attain prior to rupture

is variabIe. Rupture of the tube depends upon the amount of focal infIammatory invoIvement of the wall, resuIting in subsequent ischemia and Ieading to necrosis and ultimate perforation. Occasionally, with repeated infections the abscess may become adherent to an adjacent hollow viscus, such as the bladder, rectosigmoid or vagina1 vauIt, and perforation of the abscess with evacuation of its contents into that organ may occur. This frequently leads to sudden relief of symptoms and eventua1 recovery. 13-r5 Almost all the reported cases of ruptured adnexaI abscesses with generalized peritonitis were tubal. Rupture of ovarian abscesses has received little attention. However, in our series ovarian abscess rupture occurred in eight cases, whereas tuba1 rupture occurred in five. In seventeen patients with tubo-ovarian abscesses the exact site of rupture couId not accurately be determined. It is our belief that

American

Journal

oj Surgery

Pedowitz,

Felmus-Ruptured

the ovary is the more Iikely site for perforation in most tubo-ovarian abscesses since this organ lacks the relatively thick distensible muscutar coat of the tube. In four other cases the diagnosis was made chnicahy but the pathoIogy and site of rupture were not determined by laparotomy in one case or autopsy in three. (Fig. 3.) An adnexa1 abscess may rupture spontaneousIy or from trauma. Fourteen cases (40 per cent) in this series ruptured prior to admission and in twenty-one (60 per cent) the rupture took place whiIe the patients were being treated in the hospital. One abscess perforated twentyseven days after conservative therapy. Repeated ruptures of adnexal abscesses occurred in six patients. In onIy three instances did rupture occur traumatically in patients under observation; one occurred during the spontaneous expulsion of an early uterine gestation and two whiIe straining at stool. In the latter cases the site of rupture was on the Ieft side, in proximity to the rectosigmoid. The injudicious use of purgatives may lead to rupture16 as may coitus or a too vigorous bimanua1 examination” or uterine curettage. We have observed a death folIowing rupture of an adnexal abscess during the course of a barium enema. Bacteriology. AIthough the etioIogic agent in most cases of peIvic inflammatory disease is the gonococcus, yet this organism was not found in any cuIture taken at operation. Petroff l* and Bricknerlg reported positive cuItures in Iess than IO per cent of patients with generaIized peritonitis. Secondary bacteria1 invaders are probabIy responsible for continued growth of the abscesses as we11 as the frequentIy fata peritonitis. The most common organisms recovered from the peritonea1 fluid and the abscess cavities have been non-hemoIytic streptococci and Escherichia coli. No growth was obtained in five of our cases, but this may have been the resuIt of effective preoperative antibiotic therapy. CIinicalIy, the symptoms and signs accompanying generalized peritonitis secondary to rupture of an adnexa1 abscess vary with the extent of spillage. If the perforation is small, adhesions may form sealing off the opening. There is IittIe doubt that this occurs often for the signs and symptoms are indistinguishable from an acute exacerbation of pelvic inhammatory disease. The condition shouId be April,

19~2

AdnexaI

Abscess

with

Peritonitis

FIG. 5. Microscopic section of ovary revealing perforation of abscess, X 4.0.

$09

site of

considered if there is exacerbation of lower marked weakness, rectal abdomina1 pain, tenesmus and diarrhea. Examination reveaIs moderate pelvic peritonitis with direct and rebound abdomina1 tenderness. MiId shock may be present. On bimanual examination either a collection of pus in the cuI-de-sac or a mass in one or both fornices may be noted. Th e puIse usuahy is rapid, but the temperature is variabIe. On the other hand, massive perforation of an adnexa1 abscess gives rise to a characteristic cIinica1 picture. Almost always there is an abrupt onset of severe Iower abdomina1 pain referred to the side of the rupture. In a short time the entire abdomen is involved with a severe generaIized peritonitis and the patient is in a state of tota coIIapse. As in ruptured ectopic pregnancy, recta1 tenesmus and shoulder pain may be present. Nausea, vomiting and coId clammy perspiration, typica of shock, appear. Paralytic iIeus secondary to generalized peritonitis soon follows, and the patient appears gravely ill. The temperature immediateIy foIIowing rupture may be normal, elevated or subnormal. However, within a short time it rises rapidly and terminahy reaches 108’~. The blood pressure drops immediately to shock IeveIs an d usually remains Iow, despite the administration of blood, plasma or intravenous fluids. The rate and quality of the pulse indicate that a major calamity has occurred, for it is usuaIIy rapid and thready and out of proportion to the temperature, varying from r 10 to 170 per minute. OccasionaIJy the puJse is imperceptible.

Pedowitz,

Fehnus-Ruptured

Adnexal

An increasing p&e rate, despite a rise in bIood pressure, indicates that the peritonitis is overwhelming and treatment is ineffectua1. A previousIy paIpabIe peIvic mass may no Ionger be feIt. Abdominal tenderness and resistance become marked, with severe rebound tenderness, and the abdominal wall becomes board-like in consistency, The findings on pelvic examination depend upon the time that had eIapsed since rupture. Soon after the catastrophe remnants of a mass in one of the fornices may be dehneated, with perhaps an intact mass on the opposite side. Later, because of increasing abdomina1 distention and resistance, onIy a vague mass or fuhness is noted. The cervix is exquisitely sensitive to motion and the uterus becomes difhcult, if not impossibIe, to outIine. The Iaboratory is of IittIe aid in the diagnosis. The white bIood count ranges from 10,000 to 30,000 ceIIs, with 80 to go per cent poIymorphonucIear Ieukocytes. However, the overwhelming infection 0ccasionaIIy causes an actua1 depression of the tota count, and Iow vaIues of 6,000 to 8,000 ceIIs occurred in five of the patients in this series. The red cell sedimentation rate is rapid and of no particular diagnostic significance. DIFFERENTIAL

DIAGNOSIS

Generalized peritonitis secondary to ruptured adnexa1 abscess presents a definite cIinica1 picture and the diagnosis shouId offer IittIe difficulty. If adnexal masses had been feIt previously, an accurate diagnosis can more readiIy be made, for sudden disappearance suggests rupture. A history of either a gonorrhea1 infection of the upper genita1 tract or a recent infected abortion, together with the abrupt onset of generaIized peritonitis, a puIse out of proportion to the temperature, severe shock and the disappearance of a mass previousIy noted in the peIvis al1 point to the diagnosis. However, this condition must be differentiated from ectopic pregnancy, acute exacerbation of chronic peIvic inflammatory disease, septic abortion with uterine perforation, torsion or rupture of an ovarian cyst and rupture of an acute appendicitis. TREATMENT

The treatment of ruptured ovarian abscess or tubo-ovarian

pyosalpinx, abscess is

Abscess

with

Peritonitis

primariIy surgical, and the urgency and extent of the procedure depends upon the size of the perforation. If the peritonitis from a smaI1 perforation of an adnexa1 abscess is being controlled satisfactorily by the patient, conservative treatment may be continued. This consists of whole blood transfusions, Wangensteen suction, supportive intravenous fluids, suIfa drugs, penicillin, streptomycin and aureomycin. Since the advent of the Iatter two antibiotics there have been survivals without surgical treatment, even in the face of catastrophic perforations, which formerIy were always fata1. When a peIvic abscess forms, drainage through a posterior coIpotomy shouId be instituted. FoIIowing recovery from the acute phase, and with improvement in the general condition, hysterectomy and bilateral salpingo-oophorectomy shouId be performed before the patient is permitted to Ieave the hospita1. Two patients in this series with smaI1 perforations were treated in this manner and have made satisfactory recoveries. Failure to operate at this time is an invitation to subsequent rupture of an abscess because the patient invariabIy wiI1 be exposed to repeated infections. We have used three methods in treating Iarge perforations, namely, conservative measures, as used for smaI1 ruptures, the addition of peritonea1 cavity drainage, either abdomina1 or vaginaI, and more recentIy, extirpation of the uterus and adnexa. RESULTS

OF

TREATMENT

Conservative Treatment. Fifteen patients were treated with medica means onIy; eleven died and four recovered, a mortality rate of 73.3 per cent. The deaths occurred before 1947 prior to the advent of streptomycin and aureomycin. The four patients who survived had a stormy course and a proIonged hospital stay. They were readmitted after a short interval with either the same symptoms or for persistent abdomina1 pain. The first case was that of a twenty-nine year old coIored para 2-0-0-2 who was readmitted to the hospita1 one month folIowing discharge after apparentIy recovering from a ruptured adnexal abscess which had been treated conservativeIy. An extraperitonea1 drainage for a peIvic abscess was performed but the patient American

Journal

of Surgeq

Pedowitz,

FeImus-Ruptured

AdnexaI

faiIed to improve. Six weeks later a posterior coIpotomy was done. Three weeks Iater the abdomina1 wound was reopened and drained once more. The patient was finaIIy discharged from the hospita1 three months after the initial operative procedure, stil1 draining abdominahy and vaginahy. When examined in the outpatient chnic six months Iater the abdomina1 wound was stiI1 draining and Iarge pelvic masses were stil1 present. Two years after the initia1 operation the patient was readmitted because of persistent abdominal pain and a draining abdomina1 sinus. CuIture of the pus from the sinus revealed non-hemoIytic streptococci. On opening the abdomen a supravaginal hysterectomy and biIatera1 saIpingo-oophorectomy was the only operation feasibIe because of the extensive adhesions that were present. Accidenta severance of the left ureter occurred and was repaired following which the patient recovered UneventfuIly. The second case was that of a white twentytwo year old nulhgravida who was readmitted two months after discharge complaining of Iower abdomina1 pain. PreviousIy she had been treated conservativeIy for generalized peritonitis foIIowing rupture of an adnexal abscess. Four days later she suddenly coIIapsed into deep shock. A diagnosis of ruptured abscess was made. When she faiIed to raIIy from shock after seven hours of vigorous supportive measures, a Iaparotomy was performed and 1,500 cc. of fouI-smelling pus was evacuated from the peritonea1 cavity. A 2 cm. rent was found in a left pyosalpinx. The entire uterus and the adnexa were removed, and the abdomen cIosed without drains. She recovered without incident. The third case was that of a twenty-four year oId coIored para I-O-O-I who was admitted to the ward with severe Iower abdomina1 pains. She had been treated in another city six months before with abdomina1 incision and drainage for a ruptured adnexal abscess. Admission diagnosis was Ieft tubo-ovarian abscess. The folIowing day the patient suddenly went into shock and it was feIt that the abscess had ruptured. However, she improved rapidly with conservative treatment and ten days later was taken to the operating room in good condition for Iaparotomy. As she was being transferred to the operating tabIe her pulse became thready and extremeIy rapid; the blood pressure feI1 to 60/40. Despite the April,

1942

Abscess

with

Peritonitis

$11

patient’s desperate condition a subtota1 hysterectomy and biIateraI saIpingo-oophorectomy was performed for a ruptured Ieft tuboovarian abscess. Her postoperative course was satisfactory unti1 the third day when she went into shock during the administration of a recta1 irrigation. Perforation of the bowel was suspected. At operation a rent in the rectosigmoid was found and repaired and a coIostomy was performed. Thereafter, her course was uneventful and she was discharged on the twenty-fourth day after the hysterectomy. The fourth case was that of a thirty-seven year old para 3-0-0-3 in whom a subdiaphragmatic abscess deveIoped while the patient was being treated conservativeIy for generalized peritonitis foIlowing rupture of an adnexal abscess. The subdiaphragmatic abscess was incised and drained and the patient discharged after apparent recovery. Four months later she was readmitted for severe lower abdominal Posterior colpotomy was performed pains. twice in one month. Nine months have elapsed since rupture of the abscess and the patient is still draining vaginalIy. A Iarge adnexa1 mass is stiI1 present. She has Iost more than 50 pounds since the onset of her iIIness. At present she is being physicahy rehabihtated preparatory to major surgery. Drainage Treatment. Six patients were treated with abdomina1 or vagina1 drainage, in addition to medica therapy. Five or 83.4 per cent of these patients died. The mortality rate in simiIar cases reported by Soimaru20 and Petroff’* was 60 and 66 per cent, respectively. The deaths occurred before the introduction of streptomycin and aureomycin. The patient who survived in our series had been treated with streptomycin. A forty-seven year old white nulligravida was admitted complaining of Iower abdomina1 pain of one year’s duration and fever for one month. Examination revealed a large right adnexa1 abscess. Eight days Iater the patient suddenly went into shock and the adnexal mass was no longer paIpabIe. At Iaparotomy a ruptured right tubo-ovarian abscess was noted. Drains were pIaced into the abscess cavity and the abdomina1 wall cIosed. Non-hemoIytic streptococci were obtained on cuIture. Streptomycin, penicillin and sulfa drugs were used liberally. The patient recovered and was discharged on the twentieth postoperative day. Two and a half months Iater she was read-

512

Pedowitz,

FeImus-Ruptured

AdnexaI

mitted complaining of progressive weakness and Iower abdomina1 pain. Pus was still draining through the abdomina1 wound. Large, biIatera1, firmIy adherent adnexa1 masses rising above the umbilicus were noted on peIvic examination. One week later a large amount of pus suddenly drained through the vagina. The symptoms dramaticaIly disappeared and recovery was uneventfu1. However, several months later biIatera1 adnexa1 masses were stiI1 present. MaJ’or Surgery. Since 1947 tweIve patients have been subjected to major surgery, without a fataIity; eIeven patients had hysterectomy with biIatera1 saIpingo-oophorectomy and one uniIatera1 saIpingo-oophorectomy. An important factor in the prognosis previousIy was the interval between rupture of the adnexa1 abscess and operative treatment. Brickner,lg PetrofV and Lubke21 have shown that the lowest mortality from ruptured pyosaIpinx occurred in patients operated upon within the first twelve hours. Each additiona hour’s deIay caused an increase in the mortaIity incidence from IO to 27 per cent for the first tweIve hours to 33 per cent in the twentyfour- to forty-eight-hour period. Petroff reported a mortaIity rate of 73 per cent after forty-eight hours. In our series four patients were operated upon within the first twelve hours folIowing rupture, three at twenty-four hours, three at seventy-eight hours and the remaining two were operated upon ten and eIeven days after rupture. The satisfactory outcome in this late group probably can be attributed to the newer antibiotics. The extent of the surgical procedure depends on the genera1 condition of the patient during the operation. As a ruIe, these patients are in shock not from bIood Ioss but from the overwheIming toxemia and the perforation of the viscus. The continued seeding of the peritonea cavity with infected materia1 from the ruptured abscess is responsibIe for the maintenance of shock. AccordingIy, these patients do not respond to the usual remedial measures. OnIy when the source of contamination is removed may recovery be expected. This can be accomplished only by excision of the perforated organ. Since unilatera1 saIpingectomy or saIpingo-oophorectomy yieIds the desired result with minimum surgical manipuIation and trauma, it would appear to be the pro-

Abscess

with

Peritonitis

cedure of choice. However, in Petroff’s series of forty-six cases bilatera1 saIpingectomy was associated with a Iower mortality rate (I 1.8 per cent) than uniIatera1 salpingectomy (27.5 per cent). l8 If the patient’s condition is satisfactory during the operation, hysterectomy and bilateral salpingo-oophorectomy are preferred, provided the technical d&uIties are not too great. Both supravagina1 and tota hysterectomies have been performed in this group depending on the patient’s condition and the findings at operation; Postoperative vaginal drainage appears to make IittIe difference in the outcome but is preferabIe to abdominal drainage. A saIpingo-oophorectomy was performed on a fifteen year oId coIored gir1 who had severa previous exacerbations of peIvic inflammatory disease. Six days after admission to the hospital a right tubo-ovarian abscess ruptured, but recovery was fairIy prompt with medical treatment aIone. On the ninth hospita1 day a massive rupture occurred, and the patient went into profound shock. Her puIse had become imperceptible. At operation a right tubo-ovarian abscess was found ruptured, and despite the presence of a Ieft salpingo-oophoritis, onIy the right tube and ovary were removed. The usua1 policy of compIete operation was not foIlowed in this case. Her recovery was uneventfu1 and she was discharged on the ninth postoperative day. No bacteria1 growth was obtained from the puruIent fluid in the peritoneal cavity. Usually the uterus can be removed without too much diffrcuIty after the adnexa1 masses have been mobiIized. CompIete removal of a11 sources of infection permits better drainage and, in addition, a useIess infected organ is not Ieft behind. In onIy two of the thirty-one cases in which the pathoIogic changes had been directIy observed was there uniIatera1 adnexa1 disease. Because of this we prefer generaIIy to remove both ovaries and tubes. The remaining twentynine had biIatera1 involvement, seventeen Iarge tubo-ovarian abscesses, twelve pyosalpinges and two ovarian abscesses with tubal disease. Therefore, since the unruptured side was aIso Involved in almost a11 cases and since rupture of an abscess cannot be prognosticated with certainty, removal of both adnexa would appear to be justified. Failure to remove both sides may subject the patient to a long period American

Journal

ofSurgery

Pedowitz,

FeImus-Ruptured

of invalidism since ovarian abscesses aImost always fail to resolve. Eight of the eleven patients who had a hysterectomy and bilateral salpingo-oophorectomy recovered uneventfully. AI1 were discharged from the hospita1 by the twelfth postoperative day including three patients whose interval was longest before coming to surgery folIowing rupture of their abscesses. There were three postoperative complications. In two patients superficial wound infections developed. The third had a bowel perforation during a rectal irrigation on the third postoperative day. This probably was due to a weakened area in the wall of the bowel, a result either from the contiguous infection or from the trauma incurred during operation. Here prompt recognition and repair of the rent in the rectosigmoid was followed with an uneventfu1 recovery. To prevent bowe1 injuries during surgical procedures, no attempt is made to remove the waII of the abscess which is adherent to the intestine. Fragments left behind act as bIowout patches. Scrupulous attempts at removing the entire wal1 of the abscess are not only time consuming but aIso aImost certain to* injure the adherent organ. Postoperative perforation may occur if an additional strain is placed on this tissue, as happened in one patient. Since that accident we withhold a11 enemas and rectal irrigations postoperativeIy. The immediate postoperative care includes such supportive measures as intranasal oxygen, blood transfusions, Wangensteen drainage and intravenous fluids. Streptomycin and aureomycin offer effective means for combatting the infecting organisms, usually non-hemoIytic streptococci and Esch. coli. COMMENTS

MiIIer,22 in analyzing the deaths on the gynecoIogic service at the Charity Hospital in New OrIeans over a five-year period, found fifty-five deaths resuIting from pelvic i&ammatory disease. It is interesting to note that twenty-one cases (38.2 per cent) resuIted from rupture of an adnexal abscess. He concluded that earIy surgical intervention might have salvaged some of these patients. Treatment with the recently developed antibiotics has altered the clinical course of patients with generalized peritonitis resulting from ruptured adnexal abscesses. Before they

April,

1952

AdnexaI

Abscess

with

Peritonitis

513

were available our mortality with conservative therapy was IOO per cent but since their advent most of these patients have survived. However, the results have not been entirely satisfactory because recovery was rareIy complete. Two of the four patients so treated had a recurrent rupture within a relatively short time following their apparent recovery. The other two patients have remained chronicahy ill and subject to numerous operative procedures necessitated by the continued presence of the original infection. One of these was operated upon two years later; and because of the difficulties encountered, a ureter was accidentally severed and repaired (Case I). The subsequent course of another patient, who recovered after an abdominal incision and drainage, has not been satisfactory. She has continued to have pain and has become progressiveIy weaker, requiring hospitalization two months later. This patient still has biIateral adnexal masses which will have to be removed at a more opportune time. Results accomplished by removal of the adnexa, prior to the advent of antibiotics, have been far superior either to conservative therapy or to incision and drainage. Although excision of the perforated adnexa is the simplest procedure for removal of the source of the peritonitis, it usuahy leaves behind an acutely infected organ on the opposite side. Often this infection remains active and subject to acute exacerbations despite medical therapy. Removal of both tubes and ovaries has been shown to be associated with a lower mortality ratels than when uniIatera1 excision is done. The eventual removal of the opposite side, if not done at the time of perforation, will prove to be more diffrcuIt at a later date because of adhesions resuIting from the generalized peritonitis and the Iaparotomy. Although bilateral adnexectomy has proven to be effective in the treatment of ruptured adnexal abscesses, the added removal of the uterus at the same time offers definite advantages. It eliminates the final remaining source of infection within the peritoneal cavity, more adequate drainage per vaginum can be obtained and a functionless organ is removed as a possible source for future disease. Since the most difficult part of the surgical procedure is mobihzation of the adnexal masses preparatory to their removal, the uterus usually can be extirpated with relative ease without

Pedowitz,

574 csccssivc

prolongation

the I;tst analysis dcpcnds the

upon

iindings

the

FeImus-Ruptured

oi‘ the operation. I II extent ol’ the surgery

the condition at

AdnexaI

of the

patient

antI

faparotomy.

Obviously, if the patient remains moribund, simple saIpingooophorectomy is a11 that should be attempted. If, however, the patient improves after the operation has begun, as have a11 the patients in this series, then biIatera1 saIpingo-oophorectomy and hysterectomy should be carried out. The success of apparent radica1 surgery has been made possible by IiberaI use of bIood transfusions, advances in anesthesiology and surgical technics and development of new antibiotics. However, while the ability to carry out extensive removal of diseased tissue in seriousIy il1 patients without either immediate or deIayed mortaIity is a tribute to modern surgica1 technics and whiIe it would appear that such radicalism offers, at present, the only hope of permanent cure, castration of women during their sexualfy active years is to be deplored. The results must be continually and critically weighed in the Iight of newer antibiotics and other more conservative measures. SUMMARY I.

AND

CONCLUSIONS

Thirty-five consecutive cases of ruptured adnexal abscesses with generalized peritonitis are presented. 2. Twenty-one cases (60 per cent) ruptured in the hospita1 during the course of medical treatment. 3. There is a higher incidence of ruptured adnexa1 abscesses in the Negro race. 4. Cultures from the abscess have been consistentIy negative for the gonococcus. Non-hemoIytic streptococci and Esch. coli have been recovered. 3. The more common site of rupture, contrary to previously published reports, is the ovarian abscess. The reIative thinness and Iack of distensibility of the waII of an ovarian abscess predispose to perforation more readiIy than a pyosaIpinx. 6. A peIvic abscess frequently deveIops after the leakage of puruIent fluid from a small perforation of an” adnexal abscess. 7. Prior to 1947 treatment consisted of medical measures, and occasionally incision and drainage, with a mortality rate of IOO per cent. 8. Since 1947 nineteen patients have been treated without a fatality.

Abscess

with

Peritonitis

0. Although rccovmy ,has followed umservative therap) and incrslon and drainage, the results have not been satislactory since further surgical treatment usually is necessary. IO. The procedure of choice is extirpation of the uterus and adnexa wherever possibfe. If not feasible, then remova of the invoIved adnexa is the minimum amount of surgery that should be done. REFERENCES

I. VINCENT, G. Rupture of pyosaIpins into free peritoneum; subtotal hysterectomy with peritonization and closure without drainage. Bull. Sot. d’obst. et de gynh., 21: 747, 193% 2. ALHAIQUE, A. Surgical treatment of acute peritonitis due to rupture of pyosalpinx. Arch. ed atti d. Sot. ital. di cbir., 39: 709, 1933. 3. FOURNIER, R. Rupture of pyosaIpinx into free peritoneum; recovery after surgica1 treatment in first hours: 2 cases. Bull. Sot. d’obst. et de I g,ynec., 22: 236, 1933. 4. MASSABLJAU,G. and GIUBAL, A. Perforation of pyosalpinx into free peritoneum; subtotal hysterectomy, Mikulicz drain; cure. Bull. Sot. d’obst. et de gyh., 22: 587, 1933. 5. UBERMUTH, H. Rupture of pyosalpinx and of tuboovarian abscesses as cause of gynecoIogica1 peritonitis. Ztscbr. j. Geburtsb. u. Gynaek., I 13: 258, 1936. 6. FINCK, C. Spontaneous rupture of Iarge pyosalpinx into peritoneat cavity; case report. Rev. mex. de cir., ginec. y ccincer, 6: 23, 1938. 7. MARTZ, H. and FOOTE, R4. N. GeneraIized peritonitis secondary to ruptured pyosalpinx. Am. J. Obst. CTGynec., 36: 1009, 1938. 8. GARIBALDI, D. D. Pyosalpinx ruptured into peritoneaI cavity. Semana m6d., 2: 1305, 1940. 9. HALL, J. E. and NELMS, W. F. Hysterectomy in presence of peritonitis, salpingitis and uretera obstruction. Brooklyn Hosp. J., 5: 89, 1947. IO. CURTIS. A. H. Textbook of GvnecoIogv. 6th ed., 195o.” W. B” .Saunders P. 194. Philadelphia, & co. I I. BAUER, G. Zur Behandlung der AppendicitisPeritonitis mit Besonderer Berucksichtigung der Frage nach Primarsutur und der Behandiung von Postoperativum Ileus. Acta cbir. Scandinav., 70: 1, 1933. 12. LENORMANT, C. and KAUFMANN, R. Contribution ii I’Etude de Ia Rupture des Pyosalpinx. Presse mtd., 35: 9.45. 1927. IX. DONOHUE, P. F. Vesico-ovarian listula; ovarian abscess with rupture into bladder. J. Ural., 40: 27. 1938. 14. FALK, H. C. and HOCHMAN, S. Rupture of peIvic inflammatory masses into the urinary bIadder. Am. J. Obst. ICY Gynec., 38: 654, rg3g. 15. CASTALLO, M. A. and FETTER, T. R. Perforation of tubo-ovarian abscess into urinary bladder. Ural. I?? Cutan. Rev., 45: 425, 1941. 16. JANEWAY, E. G. Obscure case of obstipation. Neul York M. J., 32: 522, 1880.

American

Journal

of Surgery

Pedowitz, I 7.

hlaRTrN, A. Torsion

Felmus-Ruptured

du p&dicuk D’un Hydrosalpinu Droit. Rev. pmt. de g,v&c. d’ohs~. cl dc pcdinl., 19: 230, 19~10. I8. ~'WIIOPF, M. A. krforation et f
AdnexaI

Abscess

with

Peritonitis

20. SOIMARU, A. PCritonites

Gkvxtliskcs par Kupturc du Pyosalpins. G,vnkdo~ie, 34: 2 1, 1)3$. 21. LURKE, I<. A. Frequency, causes, prognosis and treatmrnt of rupture of salpinu. These de I tambourg, abstract in Zenlrulh/. Gynaek., 4~): 2423, 1925. 22. MILLER, FI. E. Deaths in gynecology. Am. J. Obst. P* Qnec., 48: 824, 1944.

ONE of the most successfu1 operations for the creation of an artificial vagina is the making of an opening through the perineum toward the uterus (if one is present) and lining it with split-skin grafts pIaced upon a moId which is held in this cavity until the grafts take. Of course, the smooth outer skin surface faces the moId and the raw, basaI part of the skin graft Iies externally on the mold. Even in the absence of the uterus such a satisfactory canal for maritaI purposes is worth while, especiaIIy since the procedure is a simple one. Simpler stiI1 is the operation in which a perineal cana is opened up and a smooth vaginal prothesis is inserted and kept in pIace unti1 the canal (between bIadder and rectum) becomes compIeteIy epithelized, starting from the norma perinea1 skin. In a few of these cases norma1 pregnancy and parturition has occurred. (Richard A. Leonardo, M.D.)

April,

1952

5’5