Maltemal deaths associated with postpartum vulvar edema THOMAS
L.
EWING,
LEROY
E. SMALE,
FRA:VK
A.
ELLIOTT,
Baker$eld
and
Los Angeles,
M.D. M.D. M.D. California
Reported are three maternal deaths in four patients who presented with a similar syndrome following a normal antepartum course and normal labor and delivery managed by regional or local anesthesia and midline or proctoepisiotomy. Beginning about the second postpartum day, the patients developed unilateral perineal edema and induration which progressed to generalized vulvar, vaginal, perineal, and gluteal edema and induration. These patients developed marked leukocytosis, fever, and ultimately vascular collapse; three of them died. The one patient who survived had a similar course except for vascular collapse. Unilateral vulvar induration and edema associated with fever and marked leukocytosis are ominous signs. Aggressive treatment should include the use of multipka antibiotic, crystaloid, colloid, and steroid drugs and appropriate monitoring. By this report we hope to bring attention to this rare but lethal syndrome. (AM. J. OBSTET. GYNECOL. 134:173, 1979.)
REPORT concerns itself with the description of a syndrome resulting in three maternal deaths and one survival in whrch all patients had normal antenatal and intrapartum courses. Two or three days later the patients developed a progressive illness characterized first by the development of unilateral vulvar and perineal edema and induration. This extended next to the contralateral side and to the inner pelvis. All patients developed fever and marked leukocytosis. Ultimately all but one patient developed vascular collapse and died. Because of the striking similarity of the manifestations in each of the patients and the scarcity of reports concerning this syndrome, a report describing the findings in these patients seems warranted. THIS
Material and results During the period (1969 to 1976 inclusive) these patients were observed and treated, eight additional maternal deaths in 49,007 live births occurred in the County of Kern, California. Table I lists the causes of From Kern Medical Center and the DeQartment of Obstetrics and Gynecology, University of California, Los Angeles. of the Pacific Gleneden
Reprint requests: Dr. Leroy E. Smale, Bakersfield, California 93301.
G St.,
000%9378/79/100173+07$00.70/0
0
1979 The
C. V. Mosby
deaths:
Kern
County
1969 1970
2 3
3.3 4.7
6,107 6,339
1971
1
Sepsis t Pulmonary embolism Angioneurotic edema?
1.6
6,147
1972 1973
0 3
Eclampsia (dead on arrival), infectious hepatitis, and pulmonary embolism
5.6
6,005 5,388
1974 1975
0 2
-
3.1
1976 Total
0 10
Fatty metamorphosis of liver, accidental death
-
-
6,109 6,374
6,538 49,007
*One of our patients died in Los Angeles County Hospital. TReported in this paper.
Presented at the Forty-jiJth Annual Meeting Coast 0b:itetrical and Gynecological Society, Beach, Oregon, September 26-30, 1978. 2501
Table I. Maternal (1969 to 1976)”
Co.
these deaths in addition to two of the three deaths comprising this report. Table II summarizes the clinical data of the four patients comprising this report. It should be noted that 173
174
Ewing, Smale, and Elliott
Table
iI$s;
II. Data on patients Hospital Identif&lion NO.
1
2
3
Duration oj 4’
(yr.)
Parity
4435541969
of ep;.\-
PT? nancy
Deliveq
Term
Low forceps
Term
Anesthesia
Caudal plus saddle-block
iotomy
Proctoepisiatom!
Onset oj 1Vhlr
Mihi7-blood rell
c’ount
on\et
oj
1‘aMlllaI tollapw
rdema
(So. lru mm)
Second postpartum dav
33,000. 57,000. left shift
Fourth post-
Second posrpartum dav
32,500, left shift
Fourth postpartum da)
partum (la\-
46790% 1971
16
395442-
24
Term
Spontaneous deliverv
Pudendal
Midline
Third postpartum da,
30,200, left shift
Fourth postpartum day
15
Term
Low forceps
Saddle-block
Proctoepisiotomy
Third postpartum day
22,300, 27,000, left shift
None
3565 16 1976
Low forceps
Epidural
Midline with thirddegree extension
1976
4
~Y:vi”~
Case No. 3 was also included and Ledger.*
in the report
by Golde
Comment Although the microbial agents responsible for this syndrome in these women have not been identified, the clinical course represents a form of bacteria1 gangrene as described by Borkowf.’ He identified two clinical courses: (1) a gas gangrene caused by Clostridia or a necrotizing fasciitis classically caused by anaerobic streptococcus, and (2) a less virulent form of progressive synergistic bacteria1 gangrene classically caused by streptococcus and pathogenic staphylococcus in combination. These are old terms and antedate the antibiotic era. Rea and Wyrick,5 writing on necrotizing fasciitis, stated that the most significant manifestation of this infection is “extensive necrosis of the superficial fascia which results in widespread undermining of the surrounding tissue and extreme systemic toxicity regarding less of the specific bacterial etiology.” The initiating injury leading to these infections was minor trauma in 80% of their cases. In many cases the trauma was located on the abdomen or extremities. None occurred in the perineal areas. Their treatment consisted of antibiotics and wide surgical debridement. Antibiot-
its alone did not appear to lower the mortality rate. Their mortality rate was 30% and this was mostly in those patients who had diabetes or arterioscIerosis. The average time from the onset of the disease to the initiation of therapy was 4 days among the survivors. Prominent features of each of the bacteria1 gangrenes are skin manifestations, usually some evidence of necrosis and often a foul “fish-water” discharge. Stone and Martin6 described a synergistic necrotizing cellulitis. Some cases started in the perineum. These patients were almost entirely diabetic and debilitated and beyond 50 years of age. They reported a high mortality rate from these lesions because of inaccessibility and inability to debride. Our cases are characterized by massive edema (Fig. 1) and compromised blood supply as evidence by the paucity of bleeding from cut surfaces. With the exception of the patient sent to the University of Southern California Medical Center, skin necrosis was not evident and a foul-smelling “fish-water” discharge was not present. The common histologic change found in all biopsy and autopsy sections is one of subcutaneous and soft tissue edema and inflammation beginning unilaterally in the area of the episiotomies. Edema was present not only in the vulvar tissues but also at autopsy in
Volume Number
134 2
Maternal deaths associated with postpartum vulvar edema
Fever “c
Intravenous pyelogram, sonography showed rightsided pelvic mass
Operative
intervention
Cultures
Treatment
Incision and drainage of ischiorectal fossa. No abscess found. Minimal bleeding
Pleural effusion, obliteration of psoas shadows 38.5”
Tracheotomy
Urine and blood cultures-negative. Lochiaklebsiella, aerobatter, bacteroides
38.2”
1. Episiotomy opened and incision and drainage of vagina and gluteal areas. No bleeding. 2. Colostomy. bebridement of perineum. Subtotal hysterectomy Episiotomy opened. No bleeding
Perineum and vaginal tissues-streptococci, Chtrid-
37.2”
175
ium sordellii,
Ampicillin, kanamycin, corticosteroids, chloramphenicol, crystalloids, colloids and heparin Local heat, Carbenicillin, kanamytin, gentamicin, cortisone, crystalloids Cephalothin, chloramphenicol, gentamicin, corticosteroids, crystalloids
staphylococci (coagulase negative) Urine and blood-all ative
the retroperitoneal tissues and the mesentery. No definite etiology has been determined nor has any commonalty in medications, type of suture materials, or infectious agents emerged. Special histologic examinations for both bacteria and fungi were negative. Search for possible sources of contamination in the patients comprising this report has been nonproductive. No reports of infections from contaminated suture material have appeared in the past 10 years (Med-Line search), although many decades ago catgut clostridial infections sometimes arose from viable spores remaining in the raw materials. Such infections would have identifiable epidemiologic characteristics. A possible source of infection is the patient herself. Since each of these patients had a midline episiotomy or an extension into the anal and/or rectal area by deliberate incision (proctoepisiotomy) or spontaneously, contamination from the fecal stream is a possibility and also from anaerobic microbial organisms normally present in the vagina. Except for Case No. 3 no positive blood or tissue cultures were obtained in any of the patients. The positive cultures from the lochia have no probable significance since similar organisms can be grown from asymptomatic postpartum patients. Also organisms isolated in Case No. 3 (Clos-
neg-
Penicillin, gentamitin, ampicillin, corticosteroids, crystalloids, colloids
Outcome and comment Died sixth postpartum day. Autopsy denied
Died fourth postpartum day. Autopsy findings, limited to perineal vaginal tissues, showed marked hemorrhage, inflammation, and edema Died on fourth postpartum day. Histology showed subepithelial necrosis, inflammation, and necrosis of perineum 1
Alive. Discharged on twenty-third postpartum day
tridium sordellii and coagulase-negative staphylococcus) probably are not the pathogens causing the patients’ state.4 Our patients were all healthy pregnant women with no known complicating disease. Hibbard and associate? reported unusual postpartum infections in women who had normal antenatal courses and normal labor and deliveries. Six of them had either subgluteal or retropsoas abscesses. These infections followed paracervical or pudendal blocks and often required surgical drainage and debridement. Likewise our patients had normal pregnancies, labors, and deliveries; the “infections” began in the episiotomies and were not purulent but were characterized by massive edema and induration of the vulvar, gluteal, and perineal areas and extended to one or both pelvic side walls (Figs, 1 to 3, Case No. 4). Ultimately intravascular volume could not be maintained and the patients died. Golde and Ledge? reported three postpartum patients with vulvar edema and one patient who had a necrotizing fasciitis in an incision for cesarean section. One of their patients (Case No. 3) had features similar to those of the cases reported here. Only one of their unique patients (Case No. 1) presented with features similar to those of our cases: vaginal delivery, fourth-
176
Fig.
Ewing, Smale, and Elliott
1. Unilateral
labial
edema on fourth
postpartum
day
(Case No. 4).
Fig. 3. Intravenous pyelogram showing mass shown in Fig. 3 (Case No. 4).
Fig. (+%4)
2. Ultrasonogram (Case
No.
showing unilateral
abdominal
mass
4).
degree perineal extension, and the subsequent development of a similar clinical course. Their other unique patient differed from our patients’ courses in that she developed multiple abscesses of the perivaginal and retroperitoneal tissues. From our experience we recommend early assessment of vulvar edema, particularly when it is asymmetric and &curs approximately within 48 hours after delivery. Asymmetry makes the edema distinguishable from more common causes of vulvar edema secondary to prolonged labor, etc. Temperature elevations are often of a low grade and the patient initially is reasonably asymptomatic aside from vulvar pain. The white blood cell count will most likely be greater than 20,OOOicu mm with a left shift. Once the diagnosis is suspected we recommend high doses of penicillin, gen-
deviation
OI ureter
h!
tamicin, chloramphenicol, and steroids. A tentral \enous catheter should be placed at the tirsr sign of intravascular volume depletion and perhaps intraarterial and puimonar); wedge pressures ShcJUkl be monitored to guide crystalloid and c,olloid therapy. The anatomic location ot this disease and ils rapid spread up the pelvic side wall makes pelvic, perineal debridement difficult and unsatisfactory. Out. one surviving
patient
lulitis which given. We
believe
recognition
did
not
have
tissue
responded
slowly
this
to be pertinent
seems
report critical
for
net rosis
but
to the antihioric
prevent&
because of
ccl-
drugs earl)
maternal
death. Aside from the report in 1977. bv Goidc and Ledger,’ WC have been unable to find any reports concerning vulvar edema secondary to episiotomy infection. Since only one of our patients survived, we believe also that histologic and bacteriologic stud) of subsequent surviving patients may shed information about an inherent idiosyncratic or immunologic deficiencythat may contribute to this fatal disease.
We wish to thank Dr. Elizabeth Admiral for ~1st’ of her case (Case No. 3) and Roger Terry, M.D., surgical pathologist, Los Angeles County-University of Southern California Medical Center, for his review of the tissue
sections
in Case
No.
3.
Volume
134
Number
2
Maternal deaths associated with postpartum vulvar edema
REFERENCES
1. Borkowf, 2. Golde,
S.,
H. I.: Clin. Obstet. Gynecol. l&40, 1972. and Ledger, W. J.: Obstet. Gynecol. 50:670,
1977. 3. Hibbard, L. T., Synder, E. N., and McVann, R. M.: Obstet. Gynecol. 39: 137, 1972.
Discussion DR. WILLIAM C. Scorn, ‘Tucson, Arizona. Acute vulvar edema is not strange to the gynecologist, occurring not uncommonly in association with trauma, infection, primary irritant dermatitis, allergic contact dermatitis, and Crohn’s disease and has even been reported following paracentesis. In the vast majority of instances, however, its alppearance is far from indicating a lifethreatening situation. The acute hemolytic streptococcus gangrene described by Meleney,’ in 1929, started from a superficial break in the skin and spread rapidly and alarmingly over the next 24 to 72 hours with profound prostration. The area became red, swollen, and painful with a poorly demarcated margin fading off gradually into relatively normal skin. If the process went on long enough the pathognomonic signs of dusky hew and blistering in patches developed, ending ultimately in frank gangrene. It was his belief that the rapidity and the profundity of the clinical infection suggested that the disease might be due to the entrance of organisms into the body at a time when it was in a hypersensitive state. Following Meleney’s original work the surgical literature was replete with articles describing similar syndromes later named “necrotizing fasciitis”2 and from which a variety of organisms in addition to the hemolytic streptococcus were cultured. In addition, a necrotizing lesion of a more indolent course called “progressive bacterial synergistic gangrene” was identified.3 Subsequent authors have reported postsurgical cases of progressive gangrene resistant to antibiotic therapy, for which they prescribed wide surgical excision, and a confusion ha:; existed between these two diagnoses, with more recent authors feeling that they represent different facets of the same disease.4 The first involvement of gynecology in the case reports was that of Sandusky and associates,’ in 1942,5 who reported 170 cases of anaerobic nonhemolytic streptococcal infection including three Bartholin’s gland abscesses and one case of postsurgical salpingitis. Thirty years later Roberts and Hester’ reported necrotizing fasciitis in six vulvar and Bartholin abscess cases. Keettel, in his discussion of their paper reported an additional Bartholin’s gland abscess and two post-cesarean section cases of uterine necrosis with abdominal gangrene. Hibbard and associates,’ in 1972, brought to our attention eight patients having uncomplicated labor and delivery, all of whom had either pericervical or pudendal block and who developed fascial and vulvar infection. One of their patients developed the typical labial edema, cellulitis, and course of necrotizing fasciitis but survived with an-
177
4. Johnson, R.: Personal communication. 5. Rea, W. J., and Wyrick, W. J., Jr.: Ann. Surg. 172:957, 1970. 6. Stone, H. H., and Martin, J. D., Jr.: Ann. Surg. 175:702, 1972.
tibiotic therapy. Bahary and associates,4 in 1977, and Daly and associates,s in 1978, reported gynecologic infections with a similar course. Golde and Ledger9 reported four cases, one of which is included in the present report, and now the present authors have added three additional cases to bring attention to this extremely serious postpartum pelvic infection with its fulminating course and high mortality rate. Golde and Ledger, in reporting their four cases occurring during a period when there were 109,531 deliveries at Los Angeles County-University of Southern California Medical Center, emphasized by their figures the fact that this was indeed a rare condition. I would like to add to this present series a case of my own as well as to report one from Denver because I believe that despite the paucity of reports in the literature the condition is one about which every obstetrician should be concerned. To be immediately alarmed by the presence of postpartum labial edema, especially unilateral, is to be forewarned of the necessity to promptly begin antibiotic therapy with high doses of intravenous penicillin and intramuscular gentamicin and Cleocin. A 23-year-old woman, gravida 1, para 0, with RH-positive, Type 0 blood, following a normal, uncomplicated pregnancy, was admitted at term with membranes intact and was delivered of a 6 pound, 5 ounce infant by spontaneous vaginal delivery following episiotomy and pudendal block on July 1, 1970. Twelve hours post partum the patient had a temperature of 99.2” F and pale vulvar edema which at that time was thought to be due to allergy to pHisoHex. The white count was 2 1,800icu mm. On the second post partum day the vulvar edema was unchanged, the temperature and pulse were normal, and she was discharged. Twelve hours later she was readmitted with acute vulvar pain. On the third postpartum day the temperature was normal and the edema had spread to involve the lower abdomen and upper thigh. On the morning of the fourth postpartum day the patient had a purulent vaginal discharge from which clostridia was later cultured. Suddenly at 1200 on the fourth postpartum day the patient went into endotoxic shock. Intravenous antibiotic drugs and Solu-
Cortef were started and with increasing amounts of intravenous fluids the urinary output was maintained. Despite medical management the patient went into irreversible shock and died on the fifth postpartum day. Autopsy findings revealed extreme brawny vulvar and lower abdominal edema with some degree of redness and poorly defined margins. There was retroperitoneal edema with massive amounts of fluid as well as free fluid within the abdominal and thoracic cavities.
The tissue in the area of the episiotomy and ischiorectal fascia showed necrosis. The blood culture of tissues at autopsy was negative.
In retrospect, the prompt of large doses of antibiotic
was negative
and culture
and early administration drugs at the onset of the
178
Ewing, Smale, and Elliott
disease on the first postpartum day might well have prevented the ultimate death of the patient. It certainly underscores Keettel’s warning, “The postpartum patient who becomes febrile and has hip or buttock pain should be suspect.” An almost identical case has been reported by personal communication from Dr. Watson Bowes, occurring in St. Joseph’s Hospital, Denver, Colorado. The death certificate stated “acute angioneurotic edema.” I wish to compliment the authors for bringing to our attention and through the literature to the attention of all obstetricians a rare but extremely serious complication of pregnancy. As they suggest, only prompt recognition and intensive antibiotic therapy seem at the present time to provide the decreasing mortality rates. Drs. Ewing, Smale, and Elliott have recommended in addition high doses of steroids as soon as the diagnosis is established. As steroids have accepted value only in the presence of ligneous cellulitis and endotoxic shock and presumably could interfere with leukocytic response to this overwhelming mixed infection, I seriously question their usage in the initial phases of treatment. In their cases as well as mine antibiotic therapy was too little and too late. Five of the six cases reviewed here were in primiparous patients, thus setting the stage for trauma in an area potentially colonized by clostridia and streptococci as well as anaerobic organisms. In the authors’ cases were x-ray films obtained for gas in the pelvic tissues? Were hematomas present? And since they suggest an immunologic deficiency, what were their patients’ histories of resistance to prior infection? Finally why is this complication so rare?
REFERENCES 1. 2. 3.
4. 5.
6.
Meleney, F. L.: Hemolytic streptococcus gangrene, J. A. M. A. 92:2009, 1929. Wilson, B.: Necrotizing fasciitis, Am. Surg. l&416, 1952. Meleney, F. L.: A differential diagnosis between certain types of infectious gangrene of the skin, Surg. Gynecol. Obstet. 56~847, 1933. Necrotizing fasciitis, Bahary, G. M., Joel-Cohen, S. J., and Neri, A.: Obstet. Gynecol. 50~633, 1977. Sandusky, W. R., Pulaski, E. J., Johnson, B. A., and Meleney, F. L.: The anaerobic nonhemolytic streptococci in surgical infections on a general surgical service, Surg. Gynecol. Obstet. 75: 145, 1942. Roberts, D. B., and Hester, L. L.: Progressive synergistic bacterial gangrene arising from abscesses of the vulva and Bartholin’s gland duct, AM. J. OBSTET. GYNECOL. 114~285, 1972. Hibbard, L. T., Snyder, E. N., and McVann, R. M.: Subgluteal and retropsoal infection in obstetric practice, Obstet. Gynecol. 39: 137, 1972, Daly, J. W., Lokowski, M. J., and Monif, G. R. G.: Simultaneous occurrence of progressive synergistic bacterial gangrene on the abdominal wall, AM. J. OBSTET. GYNECOL. 131:624, 1978. Golde, S., and Ledger, W. J.: Necrotizing fasciitis in postpartum patients, Obstet. Gynecol. 50670, 1977.
DR. LESTER T. HIBBARD, Los Angeles, California. I would like to add two cases to Dr. Smale’s collection, both taken from the Maternal Mortality Study in California.
A 30-year-old woman, para 1, was delivered uneventfully except for a fourth-degree extension of‘s midline episiotomy. Thirty-six hours later, brawny edema was noted in the left labia majora and spread rapidly down both legs and up into the abdominal wall. A large loss of plasma into subcutaneous tissues was suspected on the basis of a 20 point rise in hematocrit. During the next 10 days, edema spread to involve the entire body, despite the use of steroids and muliple antibiotic drugs. Debridement was considered but rejected because of the extensiveness of the process. All cultures were negative and there was no crepitus or abscess formation. Subsequently the patient died. The autopsy diagnosis was toxic cellulitis of uncertain origin. A 19-year-old woman, para 0, was delivered uneventfully with the aid of a left mediolateral episiotomy. On the second postoperative day, a swelling of the left labia and left buttock was noted along with chills and fever of 101” F. Tripleantibiotic therapy was begun, but the swelling increased dramatically and the patient lapsed into shock with evidence of a massive loss of plasma. The episiotumy was reopened to
reveal nothing
more
than
cellulitis
from
which E cofi,
nonhemolytic streptococci, and Clostridiurn perfringem were recovered. However, there wlas no clinical evidence of- a clostridial infection. Two treatments in a hyperbaric chamber brought temporary improvement, but the patient died before a third treatment could be started. An autopsy was inconclusive except for the demonstration of cellulitis.
Both of these patients had the benefit of‘ multiple consultations and discussion. DR. JOHN J. MOLITOR, Fullerton, California. The majority of patients in this series had midline episiotomies with extensions into the rectum. Most of the residents today disdain the mediolateral episiotomy and prefer the midline incision. 1 still think there is a place for the mediolateral incision and I use it if I think the rectum is going to be entered, because it lessens the contamination of the perineum. I have noted no difference in postoperative pain or swelling with the use of polyglycolic sutures. Dr. Smale mentioned the elevated white blood cell counts. Several years ago, one of my colleagues was sued by a patient who developed sepsis several days after she went home following delivery. Many of our patients now go home in 24 hours. During the last 2 years, we have obtained complete blood counts on the patients the first postpartum day. We have been amazed to see that white blood cell counts of 2O,OOO/cu mm are common. Counts of twenty-three and 23,000/ cu mm and 24,OOOicu mm with a marked shift to the left are frequent. So, I doaot think that the white blood cell count particularly helps in the diagnosis. On the other hand, if these patients, on the first or second postpartum day, have an elevated white blood cell count it does not necessarily mean that they have infec-
Volume Number2
134
Maternal deaths associated with postpartum vulvar edema
tion, because most all of these patients I mentioned had no postpartum difficulties after they went home. In my experience, unilateral edema of the vulva following delivery is usually due to hematoma in the pelvis. Frequently this has occurred where the fetal head came into close approximation with the ischial spine and a vessel was ruptured submucosally. A hematoma develops and the patients experience a great deal of pain and swelling. When the hematoma is evacuated the symptoms are relieved and the patients recover. DR. KEITH P. RUSSELL, Los Angeles, California. I believe that not enough attention has been paid to another factor which may enter into this condition, and that is the use of local anesthesia injected in the area. We often take too much for granted that the injection of medication for pudendal or perineal block is without hazard. Some of these cases correlate with the type of reaction seen with local anesthesia or a contaminated injectable agent. It reminds me of the problem a number of years ago of infections after spinal anesthesia, which was being used frequently. A number of these cases were traced to cracks in the drug containers, with contamination of the material which was injected. I think that until this factor is at least considered we must look to this possible etiology and not assume that this is some rare clinical occurrence. We must review the material that is being injected into these tissues, which then may become severely edematous and/or infected.
DR. ROBERT H. GREGG, Covina, California. In the evolution of this most serious problem, I am interested in knowing the status of the vascular tree, both venous and arterial, as well as the status of the hemodynamics and intravascular clotting mechanism. DR. SMALE (Closing). In most of our patients we started steroids after the patient began to lose intravascular volume, which was early in the course of the disease. We felt that these patients were developing disseminated intravascular coagulation and we were using the steroids as one of the things to hopefully “plug up” the holes in the vessels. We did obtain x-ray pictures and there was no visualized gas, and there were no hematomas. We certainly considered hematoma as one of the first things for which to search. These were all young ladies and all were in good health. We have no idea why it is so rare, but we are glad it is such a rare disease because it is really a very lethal one in our experience. I appreciate Dr. Hibbard’s contribution of the two additional case reports and his attitude, because the more information in the literature, the more important it is going to become to all of us as obstetricians and gynecologists. As far as anesthesia is concerned, only one of the four patients had any local anesthesia; the rest had epidural or saddle-block anesthesia and one received a caudal anesthetic drug.
Copyright information The appearance of a code at the bottom of the first page of an original article in this indicates the copyright owner’s consent that copies of the article may be made for personal or internal use, or for the personal or internal use of specific clients. This consent is given on the condition, however, that the copier pay the stated per copy fee through the Copyright Clearance Center, Inc., P.O. Box 765, Schenectady, N. Y. 12301, (518) 374-4430, for copying beyond that permitted by Section 107 or 108 of the LJ. S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale.
JOURNAL
179
L.
EWING,
LEROY
E. SMALE,
FRA:VK
A.
ELLIOTT,
Baker$eld
and
Los Angeles,
M.D. M.D. M.D. California
Reported are three maternal deaths in four patients who presented with a similar syndrome following a normal antepartum course and normal labor and delivery managed by regional or local anesthesia and midline or proctoepisiotomy. Beginning about the second postpartum day, the patients developed unilateral perineal edema and induration which progressed to generalized vulvar, vaginal, perineal, and gluteal edema and induration. These patients developed marked leukocytosis, fever, and ultimately vascular collapse; three of them died. The one patient who survived had a similar course except for vascular collapse. Unilateral vulvar induration and edema associated with fever and marked leukocytosis are ominous signs. Aggressive treatment should include the use of multipka antibiotic, crystaloid, colloid, and steroid drugs and appropriate monitoring. By this report we hope to bring attention to this rare but lethal syndrome. (AM. J. OBSTET. GYNECOL. 134:173, 1979.)
REPORT concerns itself with the description of a syndrome resulting in three maternal deaths and one survival in whrch all patients had normal antenatal and intrapartum courses. Two or three days later the patients developed a progressive illness characterized first by the development of unilateral vulvar and perineal edema and induration. This extended next to the contralateral side and to the inner pelvis. All patients developed fever and marked leukocytosis. Ultimately all but one patient developed vascular collapse and died. Because of the striking similarity of the manifestations in each of the patients and the scarcity of reports concerning this syndrome, a report describing the findings in these patients seems warranted. THIS
Material and results During the period (1969 to 1976 inclusive) these patients were observed and treated, eight additional maternal deaths in 49,007 live births occurred in the County of Kern, California. Table I lists the causes of From Kern Medical Center and the DeQartment of Obstetrics and Gynecology, University of California, Los Angeles. of the Pacific Gleneden
Reprint requests: Dr. Leroy E. Smale, Bakersfield, California 93301.
G St.,
000%9378/79/100173+07$00.70/0
0
1979 The
C. V. Mosby
deaths:
Kern
County
1969 1970
2 3
3.3 4.7
6,107 6,339
1971
1
Sepsis t Pulmonary embolism Angioneurotic edema?
1.6
6,147
1972 1973
0 3
Eclampsia (dead on arrival), infectious hepatitis, and pulmonary embolism
5.6
6,005 5,388
1974 1975
0 2
-
3.1
1976 Total
0 10
Fatty metamorphosis of liver, accidental death
-
-
6,109 6,374
6,538 49,007
*One of our patients died in Los Angeles County Hospital. TReported in this paper.
Presented at the Forty-jiJth Annual Meeting Coast 0b:itetrical and Gynecological Society, Beach, Oregon, September 26-30, 1978. 2501
Table I. Maternal (1969 to 1976)”
Co.
these deaths in addition to two of the three deaths comprising this report. Table II summarizes the clinical data of the four patients comprising this report. It should be noted that 173
174
Ewing, Smale, and Elliott
Table
iI$s;
II. Data on patients Hospital Identif&lion NO.
1
2
3
Duration oj 4’
(yr.)
Parity
4435541969
of ep;.\-
PT? nancy
Deliveq
Term
Low forceps
Term
Anesthesia
Caudal plus saddle-block
iotomy
Proctoepisiatom!
Onset oj 1Vhlr
Mihi7-blood rell
c’ount
on\et
oj
1‘aMlllaI tollapw
rdema
(So. lru mm)
Second postpartum dav
33,000. 57,000. left shift
Fourth post-
Second posrpartum dav
32,500, left shift
Fourth postpartum da)
partum (la\-
46790% 1971
16
395442-
24
Term
Spontaneous deliverv
Pudendal
Midline
Third postpartum da,
30,200, left shift
Fourth postpartum day
15
Term
Low forceps
Saddle-block
Proctoepisiotomy
Third postpartum day
22,300, 27,000, left shift
None
3565 16 1976
Low forceps
Epidural
Midline with thirddegree extension
1976
4
~Y:vi”~
Case No. 3 was also included and Ledger.*
in the report
by Golde
Comment Although the microbial agents responsible for this syndrome in these women have not been identified, the clinical course represents a form of bacteria1 gangrene as described by Borkowf.’ He identified two clinical courses: (1) a gas gangrene caused by Clostridia or a necrotizing fasciitis classically caused by anaerobic streptococcus, and (2) a less virulent form of progressive synergistic bacteria1 gangrene classically caused by streptococcus and pathogenic staphylococcus in combination. These are old terms and antedate the antibiotic era. Rea and Wyrick,5 writing on necrotizing fasciitis, stated that the most significant manifestation of this infection is “extensive necrosis of the superficial fascia which results in widespread undermining of the surrounding tissue and extreme systemic toxicity regarding less of the specific bacterial etiology.” The initiating injury leading to these infections was minor trauma in 80% of their cases. In many cases the trauma was located on the abdomen or extremities. None occurred in the perineal areas. Their treatment consisted of antibiotics and wide surgical debridement. Antibiot-
its alone did not appear to lower the mortality rate. Their mortality rate was 30% and this was mostly in those patients who had diabetes or arterioscIerosis. The average time from the onset of the disease to the initiation of therapy was 4 days among the survivors. Prominent features of each of the bacteria1 gangrenes are skin manifestations, usually some evidence of necrosis and often a foul “fish-water” discharge. Stone and Martin6 described a synergistic necrotizing cellulitis. Some cases started in the perineum. These patients were almost entirely diabetic and debilitated and beyond 50 years of age. They reported a high mortality rate from these lesions because of inaccessibility and inability to debride. Our cases are characterized by massive edema (Fig. 1) and compromised blood supply as evidence by the paucity of bleeding from cut surfaces. With the exception of the patient sent to the University of Southern California Medical Center, skin necrosis was not evident and a foul-smelling “fish-water” discharge was not present. The common histologic change found in all biopsy and autopsy sections is one of subcutaneous and soft tissue edema and inflammation beginning unilaterally in the area of the episiotomies. Edema was present not only in the vulvar tissues but also at autopsy in
Volume Number
134 2
Maternal deaths associated with postpartum vulvar edema
Fever “c
Intravenous pyelogram, sonography showed rightsided pelvic mass
Operative
intervention
Cultures
Treatment
Incision and drainage of ischiorectal fossa. No abscess found. Minimal bleeding
Pleural effusion, obliteration of psoas shadows 38.5”
Tracheotomy
Urine and blood cultures-negative. Lochiaklebsiella, aerobatter, bacteroides
38.2”
1. Episiotomy opened and incision and drainage of vagina and gluteal areas. No bleeding. 2. Colostomy. bebridement of perineum. Subtotal hysterectomy Episiotomy opened. No bleeding
Perineum and vaginal tissues-streptococci, Chtrid-
37.2”
175
ium sordellii,
Ampicillin, kanamycin, corticosteroids, chloramphenicol, crystalloids, colloids and heparin Local heat, Carbenicillin, kanamytin, gentamicin, cortisone, crystalloids Cephalothin, chloramphenicol, gentamicin, corticosteroids, crystalloids
staphylococci (coagulase negative) Urine and blood-all ative
the retroperitoneal tissues and the mesentery. No definite etiology has been determined nor has any commonalty in medications, type of suture materials, or infectious agents emerged. Special histologic examinations for both bacteria and fungi were negative. Search for possible sources of contamination in the patients comprising this report has been nonproductive. No reports of infections from contaminated suture material have appeared in the past 10 years (Med-Line search), although many decades ago catgut clostridial infections sometimes arose from viable spores remaining in the raw materials. Such infections would have identifiable epidemiologic characteristics. A possible source of infection is the patient herself. Since each of these patients had a midline episiotomy or an extension into the anal and/or rectal area by deliberate incision (proctoepisiotomy) or spontaneously, contamination from the fecal stream is a possibility and also from anaerobic microbial organisms normally present in the vagina. Except for Case No. 3 no positive blood or tissue cultures were obtained in any of the patients. The positive cultures from the lochia have no probable significance since similar organisms can be grown from asymptomatic postpartum patients. Also organisms isolated in Case No. 3 (Clos-
neg-
Penicillin, gentamitin, ampicillin, corticosteroids, crystalloids, colloids
Outcome and comment Died sixth postpartum day. Autopsy denied
Died fourth postpartum day. Autopsy findings, limited to perineal vaginal tissues, showed marked hemorrhage, inflammation, and edema Died on fourth postpartum day. Histology showed subepithelial necrosis, inflammation, and necrosis of perineum 1
Alive. Discharged on twenty-third postpartum day
tridium sordellii and coagulase-negative staphylococcus) probably are not the pathogens causing the patients’ state.4 Our patients were all healthy pregnant women with no known complicating disease. Hibbard and associate? reported unusual postpartum infections in women who had normal antenatal courses and normal labor and deliveries. Six of them had either subgluteal or retropsoas abscesses. These infections followed paracervical or pudendal blocks and often required surgical drainage and debridement. Likewise our patients had normal pregnancies, labors, and deliveries; the “infections” began in the episiotomies and were not purulent but were characterized by massive edema and induration of the vulvar, gluteal, and perineal areas and extended to one or both pelvic side walls (Figs, 1 to 3, Case No. 4). Ultimately intravascular volume could not be maintained and the patients died. Golde and Ledge? reported three postpartum patients with vulvar edema and one patient who had a necrotizing fasciitis in an incision for cesarean section. One of their patients (Case No. 3) had features similar to those of the cases reported here. Only one of their unique patients (Case No. 1) presented with features similar to those of our cases: vaginal delivery, fourth-
176
Fig.
Ewing, Smale, and Elliott
1. Unilateral
labial
edema on fourth
postpartum
day
(Case No. 4).
Fig. 3. Intravenous pyelogram showing mass shown in Fig. 3 (Case No. 4).
Fig. (+%4)
2. Ultrasonogram (Case
No.
showing unilateral
abdominal
mass
4).
degree perineal extension, and the subsequent development of a similar clinical course. Their other unique patient differed from our patients’ courses in that she developed multiple abscesses of the perivaginal and retroperitoneal tissues. From our experience we recommend early assessment of vulvar edema, particularly when it is asymmetric and &curs approximately within 48 hours after delivery. Asymmetry makes the edema distinguishable from more common causes of vulvar edema secondary to prolonged labor, etc. Temperature elevations are often of a low grade and the patient initially is reasonably asymptomatic aside from vulvar pain. The white blood cell count will most likely be greater than 20,OOOicu mm with a left shift. Once the diagnosis is suspected we recommend high doses of penicillin, gen-
deviation
OI ureter
h!
tamicin, chloramphenicol, and steroids. A tentral \enous catheter should be placed at the tirsr sign of intravascular volume depletion and perhaps intraarterial and puimonar); wedge pressures ShcJUkl be monitored to guide crystalloid and c,olloid therapy. The anatomic location ot this disease and ils rapid spread up the pelvic side wall makes pelvic, perineal debridement difficult and unsatisfactory. Out. one surviving
patient
lulitis which given. We
believe
recognition
did
not
have
tissue
responded
slowly
this
to be pertinent
seems
report critical
for
net rosis
but
to the antihioric
prevent&
because of
ccl-
drugs earl)
maternal
death. Aside from the report in 1977. bv Goidc and Ledger,’ WC have been unable to find any reports concerning vulvar edema secondary to episiotomy infection. Since only one of our patients survived, we believe also that histologic and bacteriologic stud) of subsequent surviving patients may shed information about an inherent idiosyncratic or immunologic deficiencythat may contribute to this fatal disease.
We wish to thank Dr. Elizabeth Admiral for ~1st’ of her case (Case No. 3) and Roger Terry, M.D., surgical pathologist, Los Angeles County-University of Southern California Medical Center, for his review of the tissue
sections
in Case
No.
3.
Volume
134
Number
2
Maternal deaths associated with postpartum vulvar edema
REFERENCES
1. Borkowf, 2. Golde,
S.,
H. I.: Clin. Obstet. Gynecol. l&40, 1972. and Ledger, W. J.: Obstet. Gynecol. 50:670,
1977. 3. Hibbard, L. T., Synder, E. N., and McVann, R. M.: Obstet. Gynecol. 39: 137, 1972.
Discussion DR. WILLIAM C. Scorn, ‘Tucson, Arizona. Acute vulvar edema is not strange to the gynecologist, occurring not uncommonly in association with trauma, infection, primary irritant dermatitis, allergic contact dermatitis, and Crohn’s disease and has even been reported following paracentesis. In the vast majority of instances, however, its alppearance is far from indicating a lifethreatening situation. The acute hemolytic streptococcus gangrene described by Meleney,’ in 1929, started from a superficial break in the skin and spread rapidly and alarmingly over the next 24 to 72 hours with profound prostration. The area became red, swollen, and painful with a poorly demarcated margin fading off gradually into relatively normal skin. If the process went on long enough the pathognomonic signs of dusky hew and blistering in patches developed, ending ultimately in frank gangrene. It was his belief that the rapidity and the profundity of the clinical infection suggested that the disease might be due to the entrance of organisms into the body at a time when it was in a hypersensitive state. Following Meleney’s original work the surgical literature was replete with articles describing similar syndromes later named “necrotizing fasciitis”2 and from which a variety of organisms in addition to the hemolytic streptococcus were cultured. In addition, a necrotizing lesion of a more indolent course called “progressive bacterial synergistic gangrene” was identified.3 Subsequent authors have reported postsurgical cases of progressive gangrene resistant to antibiotic therapy, for which they prescribed wide surgical excision, and a confusion ha:; existed between these two diagnoses, with more recent authors feeling that they represent different facets of the same disease.4 The first involvement of gynecology in the case reports was that of Sandusky and associates,’ in 1942,5 who reported 170 cases of anaerobic nonhemolytic streptococcal infection including three Bartholin’s gland abscesses and one case of postsurgical salpingitis. Thirty years later Roberts and Hester’ reported necrotizing fasciitis in six vulvar and Bartholin abscess cases. Keettel, in his discussion of their paper reported an additional Bartholin’s gland abscess and two post-cesarean section cases of uterine necrosis with abdominal gangrene. Hibbard and associates,’ in 1972, brought to our attention eight patients having uncomplicated labor and delivery, all of whom had either pericervical or pudendal block and who developed fascial and vulvar infection. One of their patients developed the typical labial edema, cellulitis, and course of necrotizing fasciitis but survived with an-
177
4. Johnson, R.: Personal communication. 5. Rea, W. J., and Wyrick, W. J., Jr.: Ann. Surg. 172:957, 1970. 6. Stone, H. H., and Martin, J. D., Jr.: Ann. Surg. 175:702, 1972.
tibiotic therapy. Bahary and associates,4 in 1977, and Daly and associates,s in 1978, reported gynecologic infections with a similar course. Golde and Ledger9 reported four cases, one of which is included in the present report, and now the present authors have added three additional cases to bring attention to this extremely serious postpartum pelvic infection with its fulminating course and high mortality rate. Golde and Ledger, in reporting their four cases occurring during a period when there were 109,531 deliveries at Los Angeles County-University of Southern California Medical Center, emphasized by their figures the fact that this was indeed a rare condition. I would like to add to this present series a case of my own as well as to report one from Denver because I believe that despite the paucity of reports in the literature the condition is one about which every obstetrician should be concerned. To be immediately alarmed by the presence of postpartum labial edema, especially unilateral, is to be forewarned of the necessity to promptly begin antibiotic therapy with high doses of intravenous penicillin and intramuscular gentamicin and Cleocin. A 23-year-old woman, gravida 1, para 0, with RH-positive, Type 0 blood, following a normal, uncomplicated pregnancy, was admitted at term with membranes intact and was delivered of a 6 pound, 5 ounce infant by spontaneous vaginal delivery following episiotomy and pudendal block on July 1, 1970. Twelve hours post partum the patient had a temperature of 99.2” F and pale vulvar edema which at that time was thought to be due to allergy to pHisoHex. The white count was 2 1,800icu mm. On the second post partum day the vulvar edema was unchanged, the temperature and pulse were normal, and she was discharged. Twelve hours later she was readmitted with acute vulvar pain. On the third postpartum day the temperature was normal and the edema had spread to involve the lower abdomen and upper thigh. On the morning of the fourth postpartum day the patient had a purulent vaginal discharge from which clostridia was later cultured. Suddenly at 1200 on the fourth postpartum day the patient went into endotoxic shock. Intravenous antibiotic drugs and Solu-
Cortef were started and with increasing amounts of intravenous fluids the urinary output was maintained. Despite medical management the patient went into irreversible shock and died on the fifth postpartum day. Autopsy findings revealed extreme brawny vulvar and lower abdominal edema with some degree of redness and poorly defined margins. There was retroperitoneal edema with massive amounts of fluid as well as free fluid within the abdominal and thoracic cavities.
The tissue in the area of the episiotomy and ischiorectal fascia showed necrosis. The blood culture of tissues at autopsy was negative.
In retrospect, the prompt of large doses of antibiotic
was negative
and culture
and early administration drugs at the onset of the
178
Ewing, Smale, and Elliott
disease on the first postpartum day might well have prevented the ultimate death of the patient. It certainly underscores Keettel’s warning, “The postpartum patient who becomes febrile and has hip or buttock pain should be suspect.” An almost identical case has been reported by personal communication from Dr. Watson Bowes, occurring in St. Joseph’s Hospital, Denver, Colorado. The death certificate stated “acute angioneurotic edema.” I wish to compliment the authors for bringing to our attention and through the literature to the attention of all obstetricians a rare but extremely serious complication of pregnancy. As they suggest, only prompt recognition and intensive antibiotic therapy seem at the present time to provide the decreasing mortality rates. Drs. Ewing, Smale, and Elliott have recommended in addition high doses of steroids as soon as the diagnosis is established. As steroids have accepted value only in the presence of ligneous cellulitis and endotoxic shock and presumably could interfere with leukocytic response to this overwhelming mixed infection, I seriously question their usage in the initial phases of treatment. In their cases as well as mine antibiotic therapy was too little and too late. Five of the six cases reviewed here were in primiparous patients, thus setting the stage for trauma in an area potentially colonized by clostridia and streptococci as well as anaerobic organisms. In the authors’ cases were x-ray films obtained for gas in the pelvic tissues? Were hematomas present? And since they suggest an immunologic deficiency, what were their patients’ histories of resistance to prior infection? Finally why is this complication so rare?
REFERENCES 1. 2. 3.
4. 5.
6.
Meleney, F. L.: Hemolytic streptococcus gangrene, J. A. M. A. 92:2009, 1929. Wilson, B.: Necrotizing fasciitis, Am. Surg. l&416, 1952. Meleney, F. L.: A differential diagnosis between certain types of infectious gangrene of the skin, Surg. Gynecol. Obstet. 56~847, 1933. Necrotizing fasciitis, Bahary, G. M., Joel-Cohen, S. J., and Neri, A.: Obstet. Gynecol. 50~633, 1977. Sandusky, W. R., Pulaski, E. J., Johnson, B. A., and Meleney, F. L.: The anaerobic nonhemolytic streptococci in surgical infections on a general surgical service, Surg. Gynecol. Obstet. 75: 145, 1942. Roberts, D. B., and Hester, L. L.: Progressive synergistic bacterial gangrene arising from abscesses of the vulva and Bartholin’s gland duct, AM. J. OBSTET. GYNECOL. 114~285, 1972. Hibbard, L. T., Snyder, E. N., and McVann, R. M.: Subgluteal and retropsoal infection in obstetric practice, Obstet. Gynecol. 39: 137, 1972, Daly, J. W., Lokowski, M. J., and Monif, G. R. G.: Simultaneous occurrence of progressive synergistic bacterial gangrene on the abdominal wall, AM. J. OBSTET. GYNECOL. 131:624, 1978. Golde, S., and Ledger, W. J.: Necrotizing fasciitis in postpartum patients, Obstet. Gynecol. 50670, 1977.
DR. LESTER T. HIBBARD, Los Angeles, California. I would like to add two cases to Dr. Smale’s collection, both taken from the Maternal Mortality Study in California.
A 30-year-old woman, para 1, was delivered uneventfully except for a fourth-degree extension of‘s midline episiotomy. Thirty-six hours later, brawny edema was noted in the left labia majora and spread rapidly down both legs and up into the abdominal wall. A large loss of plasma into subcutaneous tissues was suspected on the basis of a 20 point rise in hematocrit. During the next 10 days, edema spread to involve the entire body, despite the use of steroids and muliple antibiotic drugs. Debridement was considered but rejected because of the extensiveness of the process. All cultures were negative and there was no crepitus or abscess formation. Subsequently the patient died. The autopsy diagnosis was toxic cellulitis of uncertain origin. A 19-year-old woman, para 0, was delivered uneventfully with the aid of a left mediolateral episiotomy. On the second postoperative day, a swelling of the left labia and left buttock was noted along with chills and fever of 101” F. Tripleantibiotic therapy was begun, but the swelling increased dramatically and the patient lapsed into shock with evidence of a massive loss of plasma. The episiotumy was reopened to
reveal nothing
more
than
cellulitis
from
which E cofi,
nonhemolytic streptococci, and Clostridiurn perfringem were recovered. However, there wlas no clinical evidence of- a clostridial infection. Two treatments in a hyperbaric chamber brought temporary improvement, but the patient died before a third treatment could be started. An autopsy was inconclusive except for the demonstration of cellulitis.
Both of these patients had the benefit of‘ multiple consultations and discussion. DR. JOHN J. MOLITOR, Fullerton, California. The majority of patients in this series had midline episiotomies with extensions into the rectum. Most of the residents today disdain the mediolateral episiotomy and prefer the midline incision. 1 still think there is a place for the mediolateral incision and I use it if I think the rectum is going to be entered, because it lessens the contamination of the perineum. I have noted no difference in postoperative pain or swelling with the use of polyglycolic sutures. Dr. Smale mentioned the elevated white blood cell counts. Several years ago, one of my colleagues was sued by a patient who developed sepsis several days after she went home following delivery. Many of our patients now go home in 24 hours. During the last 2 years, we have obtained complete blood counts on the patients the first postpartum day. We have been amazed to see that white blood cell counts of 2O,OOO/cu mm are common. Counts of twenty-three and 23,000/ cu mm and 24,OOOicu mm with a marked shift to the left are frequent. So, I doaot think that the white blood cell count particularly helps in the diagnosis. On the other hand, if these patients, on the first or second postpartum day, have an elevated white blood cell count it does not necessarily mean that they have infec-
Volume Number2
134
Maternal deaths associated with postpartum vulvar edema
tion, because most all of these patients I mentioned had no postpartum difficulties after they went home. In my experience, unilateral edema of the vulva following delivery is usually due to hematoma in the pelvis. Frequently this has occurred where the fetal head came into close approximation with the ischial spine and a vessel was ruptured submucosally. A hematoma develops and the patients experience a great deal of pain and swelling. When the hematoma is evacuated the symptoms are relieved and the patients recover. DR. KEITH P. RUSSELL, Los Angeles, California. I believe that not enough attention has been paid to another factor which may enter into this condition, and that is the use of local anesthesia injected in the area. We often take too much for granted that the injection of medication for pudendal or perineal block is without hazard. Some of these cases correlate with the type of reaction seen with local anesthesia or a contaminated injectable agent. It reminds me of the problem a number of years ago of infections after spinal anesthesia, which was being used frequently. A number of these cases were traced to cracks in the drug containers, with contamination of the material which was injected. I think that until this factor is at least considered we must look to this possible etiology and not assume that this is some rare clinical occurrence. We must review the material that is being injected into these tissues, which then may become severely edematous and/or infected.
DR. ROBERT H. GREGG, Covina, California. In the evolution of this most serious problem, I am interested in knowing the status of the vascular tree, both venous and arterial, as well as the status of the hemodynamics and intravascular clotting mechanism. DR. SMALE (Closing). In most of our patients we started steroids after the patient began to lose intravascular volume, which was early in the course of the disease. We felt that these patients were developing disseminated intravascular coagulation and we were using the steroids as one of the things to hopefully “plug up” the holes in the vessels. We did obtain x-ray pictures and there was no visualized gas, and there were no hematomas. We certainly considered hematoma as one of the first things for which to search. These were all young ladies and all were in good health. We have no idea why it is so rare, but we are glad it is such a rare disease because it is really a very lethal one in our experience. I appreciate Dr. Hibbard’s contribution of the two additional case reports and his attitude, because the more information in the literature, the more important it is going to become to all of us as obstetricians and gynecologists. As far as anesthesia is concerned, only one of the four patients had any local anesthesia; the rest had epidural or saddle-block anesthesia and one received a caudal anesthetic drug.
Copyright information The appearance of a code at the bottom of the first page of an original article in this indicates the copyright owner’s consent that copies of the article may be made for personal or internal use, or for the personal or internal use of specific clients. This consent is given on the condition, however, that the copier pay the stated per copy fee through the Copyright Clearance Center, Inc., P.O. Box 765, Schenectady, N. Y. 12301, (518) 374-4430, for copying beyond that permitted by Section 107 or 108 of the LJ. S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale.
JOURNAL
179
