Diagnosis and treatment of intrauterine adhesions by microhysteroscopy

gery ~O

FERTILITY AND STERILITY Copyright c 1983 The American Fertility Society

Vol. 39, No.3, March 1983 Printed in U.SA.

con16-

; in j

Diagnosis and treatment of intrauterine adhesions by microhysteroscopy

Am

ny.

; of Br

ec-

Jacques Hamou, M.D.*t Jacques Salat-Baroux, M.D. * Alvin M. Siegler, M.D., D.Sc.*

gic

lin

ge Ib-

Universite de Paris, Faculte de Medecine Pierre et Marie Curie, Paris, France, and State University of New York, Downstate Medical Center, Brooklyn, New York

In this report of 69 patients, a new type of hysteroscope was used to evaluate the extent and character of intrauterine adhesions, to perform lysis of them, and to monitor the effects of therapy. Additionally, prospective studies with regard to pathogenesis and endometrial regeneration can be achieved in vivo. In 59 patients the procedures were performed in an office setting using a CO2 hysteroscopic technique without the need for local anesthesia or cervical dilatation. Of 30 infertile patients, 38% subsequently had uncomplicated deliveries. The severe forms of this disease still remain very difficult to treat effectively. When the adhesions were severe or the procedure painful, the operation was scheduled under general anesthesia (ten cases). A sequential hysteroscopy with good patient acceptance affords additional opportunity for removing residual adhesions and intrauterine devices, and serves as a basis for ending treatment with steroids. Fertil SteriI39:321, 1983

Intrauterine adhesions were initially recognized by Fritsch! in 1894, and their hysterographic configurations were described by Hald 2 in 1949. Asherman 3 ,4 assessed the clinical characteristics and therapeutic possibilities and made conjectures about their cause and pathogenesis. Limited methods for simple observation of the uterine cavity precluded reliable documentation of stages of this disease, its severity, or sequential follow-up examinations. History of a curettage of a recently pregnant uterus associated with subsequent amenorrhea or hypomenorrhea should alert the physician to the

Received May 17, 1982; revised and accepted November 23, 1982. *Universite de Paris, Faculte de Medecine Pierre et Marie Curie, Paris, France. tReprint requests: Jacques Hamou, M.D., 2 Chaussee de la Muette, Paris, France. *State University of New York, Downstate Medical Center, Brooklyn, New York. Vol. 39, No.3, March 1983

possible existence of intrauterine adhesions. A hysterogram showing persistent uterine defects with sharply delineated borders increases the suspicion. However, the most precise diagnosis of the presence and extent of these adhesions can be made only by direct observation of the uterine cavity. Hysterographic findings may be equivocal or misinterpreted, as noted by Sweeney5 and Foix et al. 6 Surgical management reported by Musset and Netter7 has limitations. Conventional hysteroscopy has been a significant addition to diagnostic and therapeutic management by substituting direct observations of the lesions for hysterography, and lysis of adhesions under visual control for curettage. Currently, hysteroscopy has required general anesthesia, cervical dilatation, and ancillary surgical instruments. In this presentation of 69 patients, a new type of hysteroscope (made by Karl Storz GmbH Company, Tuttlingen, West Germany; U. S. distribution by Karl Storz Endoscopy-America, Inc.,

Hamou et al. Intrauterine adhesions and microhysteroscopy

321

Table 1. Menstrual Patterns Correlated with the Cause of Synechiae in 69 Women Menstrual pattern Cause

Curettage Abortion Induced Spontaneous Postpartum Diagnostic Other Pelvic tuberculosis Cesarean section Myomectomy Unknown Total

Amenorrhea

12 4 1 2 2 21 (30.4%)

Hypomenorrhea

9 8 2 2

8 8 1 2

1 1

2 1 1

2 25 (36.2%)

Culver City, CA) enabled the physician to search for intrauterine adhesions in clinically suspicious cases, and operative correction was possible in selected instances in an office setting. Information was obtained about pathogenesis, and monthly sequential hysteroscopic examinations were made in accordance with the clinical condition.

Normal

23 (33.3%)

Dysmenorrhea

0 3 0 0 1 1 2 7(10.1%)

Two women had had postpartum bleeding, and another had had a cesarean section. Although 48 patients had a presumptive diagnosis based on hysterograms, 16 others were referred because of menstrual disorders or infertility. They had hysteroscopy as the initial diagnostic procedure. Five others were from a previous series; they had had monthly hysteroscopic exam-

MATERIALS AND METHODS

From March 1979 through January 1980, 69 patients who had a diagnosis of intrauterine adhesions were evaluated and managed by hysteroscopy. The patients were 20 to 43 years of age, with a mean of29 years. In 52 instances (82.6%) a history of curettage of a recently pregnant uterus was obtained, and 5 patients had diagnostic curettage. The other 12 women had had cesarean sections in 3 instances, pelvic tuberculosis in 3 others, a myomectomy in 1, and no history of any pelvic disorder or surgery in 5. Menstrual disorders were present in 46 patients (66.6%), with amenorrhea in 21 (30.4%) and hypomenorrhea in 25 (36.2%); an associated dysmenorrhea was present in 7 women (10.1%) (Table 1). The remaining 23 women had normal cyclic menses. Of the 39 women who wished to conceive, 33 had had previous pregnancies. Before the hysteroscopy these patients had the appropriate infertility studies, which included basal body temperature, hysterogram, and an endometrial biopsy whenever possible. Of the 33 patients who were previously pregnant, uncomplicated spontaneous abortions had occurred in 8, postabortal complications were reported to have occurred in 6, 16 women had had induced abortions, and 5 of these had been complicated by infectious morbidity. 322

Figure 1 The Hamou microcolpohysteroscope with its outer sheath is 5.2 mm in diameter. Two oculars permit observation at x 1, x 20, x 60, and x 150 magnification. (B), The sharp distal angled edge of the hysteroscope is used for the dissection of adhesions.

(A),

Hamou et aI. Intrauterine adhesions and microhysteroscopy

Fertility and Sterility

Table 2. Histologic Features of Adhesions and Their Size, Location, and Duration Histologic features Endometrial

Fibrous

Myometrial

12 5

26 11

3 12

0 15 2

14 9 14

4 3 8

13 2

5 28

1 12

Size Under 1 sq cm Over 1 sq cm Location Isthmic Central Marginal Duration (61 cases)a Under 1 year Over 1 year

aIt was not possible to estimate the duration in eight patients.

inations following first-trimester induced abortions. The authors observed the development of intrauterine adhesions. 8 A pelvic examination was performed. Hysteroscopy initially involved a panoramic view without magnification to assess the location (isthmic, marginal, central) of the adhesions. Their sizes were measured by contact with the endoscope. At x 20 magnification, the histologic nature of the adhesions was discerned, and the thickness, extension, and glandular structures of the remaining endometrium were seen. In ten patients during the initial hysteroscopic examination, lysis of intrauterine adhesions failed or was painful. Their operations were performed subsequently under general anesthesia with associated laparoscopy. Postoperatively, all patients received 100 mg ethinyl estradiol from day 1 to day 21 and 5 mg norethindrone from day 16 to day 21. Therapy was continued for two cycles. Twenty-eight women who had small adhesions of less than 1 year's duration did not receive an intrauterine device (IUD). Lippes IUDs (Ortho Pharmaceutical Corporation, Raritan, NJ) were

inserted in 41 instances. The selection of these patients was based on several factors. Eleven women had severe adhesions, 12 had adhesions for at least 1 year, 9 had fibrous adhesions, 6 others had central adhesions, and 3 women showed a reappearance of adhesions after initial control hysteroscopy. The conventional classification into mild, moderate, and severe adhesions was not deemed appropriate, because other new factors appeared significant, such as the location of the adhesions, their size, and their histologic appearance. Follow-up hysteroscopy was performed more than once in 57 women; only 15 required additional adhesiotomy. We made the repeated examinations to observe the response of the cavity to therapy and to search for recurrences. Only those patients who did not return for follow-up examination or who became pregnant did not have at least one repeat hysteroscopic examination. The hysteroscope used in this study was described in previous publications. 9 - 11 It is a microcolpohysteroscope 25 cm in length with a 4-mm diameter and a 30-degree foroblique lens. The diagnostic sheath has a 5.2-mm outer diameter. It is a multipurpose endoscope allowing panoramic . and contact vision, including x 1, x 20, x 60, and x 150 magnification (Fig. 1A). Panoramic vision without magnification can assess topography and extension of adhesions. After supravital staining by methylene blue solution, cells within glandular structures are discerned at x 60 and x 150 magnifications. Adhesiotomy was performed with the leading edge of the microhysteroscope, with the contact, magnified view to control the lysis (Fig. 1B). Carbon dioxide was used to distend the uterine cavity at a maximal flow rate of 30 mIl minute. The duration of the procedure varied, depending upon the severity of the disease. The average time for the office surgical procedure was 10 minutes.

Table 3. Results from Adhesiotomy on Menses and Fertility Method of treatment in 60 women Location

Isthmic (20) Marginal (23) Central (15) Severe (l1)a

One complete treatment 14 14 13 3 63.8%

Repeated treatment

Results in 39 infertile women

Under general anesthesia and laparoscopy

5 4 2 4 21.7%

1 5 0 4 14.5%

Normal menses

Infertility

15 (75%) 23 (100%) 15 (100%) 6 (54%)

11 12 8 8

59 (85.5%)

Pregnancy

5 7 4 4 20 (51.3%)

Term (uncomplicated) 3 7 3 2 15 (38.4%)

aDefined as more than 2 sq em on hysteroscopy. Vol. 39, No.3, March 1983

Hamou et al. Intrauterine adhesions and microhysteroscopy

323

Table 4. A Comparison Between Results of Hysterograms and Hysteroscopic Examinations in 55 Patients Hysteroscopy Hysterogram

Isthmic

Marginal

Severe

Normal

11

Isthmic Marginal Central Severe Normal Polyp

12

2

10 1

5

Polyp or malformation

2 2 1

4

1

2

2

RESULTS Three types of intrauterine adhesions were identified at x 20 magnification: (1) Endometrial adhesions appeared white, with some glandular and vascular patterns similar to those in the surrounding endometrium. They were easily dissected (17 cases). (2) Synechiae composed mostly of fibrous or connective tissue appeared transparent, thin, bridge-like, and poorly vascularized and formed stumps after lysis; usually they were central or isthmic (37 cases). (3) Myometrial adhesions, limiting uterine distension, were highly vascular, extensive, and required general anesthesia for their lysis (seven cases). The fibrous synechiae of the second type were more common in moderate adhesions (less than 1 sq cm) and in 14 of 18 women with isthmic adhesions of more than 1 year's duration. Adhesions predominantly of endometrial tissue were mostly central and under 1 year's duration. Myometrial tissue was found in extensive and old adhesions (Table 2). Synechiolysis was attempted in all cases during the first hysteroscopic examination and was successful in 44 patients (63.8%). Partial recurrence or persistence required reoperation 1 month later in eight patients, and in seven others lysis was completed at a third hysteroscopic examination. The ten patients in whom the operation was not completed because of pain or discomfort had severe, old, or marginal adhesions that were lysed under general anesthesia, two with a tenotome (Medimex Inc., Paris, France) (Table 3). Uterine perforation occurred in two patients, without.abnormal bleeding or apparent sequelae; one had had a previous uterine perforation during an earlier attempt at lysis by tenotome. Sequential hysteroscopic operations were done at monthly intervals until a satisfactory uterine cavity and normal endometrium were found. Complete recovery occurred within 1 month if endometrial adhesions were less than 1 sq cm. When synechiae were extensive, fibrous, or muscular, healing of the en324

Central

dometriurn began within 2 months. At this time a thin irregular endometrium was seen with few glandular openings at x 150 magnification. Of the 28 patients who did not have IUDs inserted postoperatively, 3 developed recurrent intrauterine adhesions, and 2 others with severe synechiae had recurrences. In contrast, of 41 women with IUDs inserted during 2 postoperative months, only 3 with large central synechiae (over 2 sq cm) had partial recurrences. Of 21 patients with amenorrhea, 6 underwent hysteroscopy at least 4 months after postabortal curettage. Isthmic synechiae were seen with a

Figure 2 (A), Capillaries are seen in the adhesive band (x 20). (B), Months later only a few adhesions are present in the fundus.

Hamou et al. Intrauterine adhesions and microhysteroscopy

Fertility and Sterility

Figure 3 (A), Central, fibrous adhesions separate the uterine cavity into two compartments. (B), Dissection with the distal end of the hysteroscope begins. (C), Small adhesions still persist. (D), The fundal endometrium comes into view, and both cornua are identified.

normal cavity above, and the endometrium was less than 4 mm thick. Four other women had multiple, extensive, and diffuse adhesions in a small uterine cavity. In the remaining two, from 8 to 9 weeks after an induced abortion isthmic, synechiae associated with hematometria were found. In these 12 women, abortion had been by aspiration alone in 7, by aspiration and curettage in 4, and by curettage twice in 1. All 25 women with hypomenorrhea had intrauterine adhesions, significant in 20 and minor in 5 (less than 1 sq cm); their endometrium was thin, pale, and had few glandular openings. Only mild synechiae were seen in 23 women who had normal menses, and their endometria seemed to be of normal thickness. Reestablishment of normal menses without dysmenorrhea occurred in 59 (85.5%) patients; 3 had persistent amenorrhea, 2 of whom had a history of pelvic tuberculosis; and 5 hypomenorrheic patients reported no improvement. Of the 39 women who wanted to conceive, 20 did s0-14 within 3 months after the operation. Fifteen had uncomplicated deliveries at term, and 5 had spontaneous early abortions; seven had othVol. 39, No.3, March 1983

er factors associated with infertility. The direct view of the intrauterine adhesions is more reliable than interpretations of hysterograms, as established by comparisons in 55 patients (Table 4). In 37 patients (67.2%), the diagnosis by the two methods agreed, but the two methods disagreed in 12 (17.4%) and were not wholly in accord in another 6 women (10.9%). In five women, isthmic adhesions prevented the contrast medium from entering the uterine cavity on hysterographic examination. In one case, a single central adhesion was seen, but the hysterogram showed total obliteration of the cavity. In four others, filling defects were interpreted as polyps, but proved to be marginal or isthmic adhesions. One patient with a normal hysterogram had a central adhesion seen in hysteroscopy. Hysterographic defects interpreted as representing synechiae in seven women actually were polyps, malformations, or artifacts. DISCUSSION

Complications from pregnancy-related curettage are the most common causes of intrauterine

Hamou et al. Intrauterine adhesions and microhysteroscopy

325

adhesions. Menstrual disorders, either hypomenorrhea or amenorrhea, are the chief complaints of the majority of women. 12, 13 Hysteroscopy as an office procedure permits sequential examinations that give a new approach to the study of the pathogenesis of intrauterine adhesions. Hysteroscopy at x 20 magnification showed capillaries converging on the adhesions (Fig. 2A). Hysteroscopy 1 month later showed further changes, and months later the synechiae had a bridge-like appearance (Fig. 2B). With microhysteroscopy, the components of the adhesions could be discerned. At x 20 magnification, fibrous and myometrial tissue appeared more frequently when the adhesions had been present over 1 year; endometrial glands and stroma were present in "younger" adhesions. This observation suggests that endometrial tissue might be transformed into fibrous tissue and in the advanced state into adhesions composed of myometrial tissue. As is shown in Table 4, interpretations of hysterograms failed completely to coincide with what was clearly seen by hysteroscopy in 12 of 55 cases (21.8%), and agreement was only partial in another 6 cases (10.9%). Evaluation of isthmic synechiae associated with amenorrhea can be difficult to study hysterographically because contrast medium may not reach the cavity. In 5 of the 21 amenorrheic patients, a single isthmic adhesion was found. Management of these synechiae by hysteroscopic dissection enables good endometrial regeneration. Conventional surgical techniques and a tenotome were used only when microhysteroscopic lysis failed (ten cases). Monthly hysteroscopy enabled the observation ofthe healing process and the decision as to when to retrieve the IUD and terminate steroid treatment. Also, additional lysis may be performed (Fig. 3A to D).

326

Repeated atraumatic microhysteroscopy is justified and ethical only if there is more benefit to the patient than contingent risks or drawbacks. Hysteroscopy as reported by other investigators generally included local anesthesia with paracervical block, cervical dilatation, the use of a 7-mm sheathed hysteroscope, and liquid media for uterine distension. 14 , 15 REFERENCES 1. Fritsch H: Ein Fall Von Volligen Schwund der Gerbarmutterhole nach Auskratzung. Zentralbl Gumsrl 18: 1337, 1894 2. Hald H: On uterine atresia consequent to curettage. Acta Obstet Gynecol Scand 28:169, 1949 3. Asherman JG: Amenorrhea traumatica (atretica). J Obstet Gynaecol Br Emp 55:23, 1948 4. Asherman JG: Traumatic intrauterine adhesions. J Obstet Gynaecol Br Emp 57:892, 1950 5. Sweeney WJ: Accuracy of preoperative hysterosalpingograms. Obstet Gynecol 11:640, 1958 6. Foix A, Bruno RO, Davidson T, Lema B: The pathology of pact curettage intrauterine adhesions. Am J Obstet Gynecol 96:1027, 1966 7. Musset R, Netter A: Synechies uterines. Encyciopedie Gynecol Paris 10:140, 1972 8. Kremer L, Hamou J, Salat-Baroux J: Aspects of hysteroscopiques dans les suites d'interruptions volontaires de grossesse. These, Faculte de Medecine de Paris, 1980, p 33 9. Hamou J: Microhysteroscopy. Acta Endosc 10:415, 1980 10. Hamou J: Microhysteroscopy: a new procedure and its application in gynecology. J Reprod Med 26:275, 1981 11. Hamou J: Hysteroscopy and microhysteroscopy. Acta Eur Fertil 12:29, 1981 12. Siegler AM, Kemmann EK: Hysteroscopy. Obstet Gynecol Surv 30:567, 1975 13. Valle RF: Hysteroscopy: diagnostic and therapeutic applications. J Reprod Med 20:115, 1978 14. Sugimoto 0: Diagnostic and therapeutic hysteroscopy for traumatic intrauterine adhesions. Am J Obstet Gynecol 131:539, 1978 15. March eM, Israel R, March AD: Hysteroscopic management of intrauterine adhesions. Am J Obstet Gynecol 130:653, 1978

Hamou et al. Intrauterine adhesions and microhysteroscopy

Fertility and Sterility