The American Fertility Society classifications of adnexal adhesions, distal tubal occlusion, tubal occlusion secondary to tubal ligation, tubal pregnancies, Mullerian anomalies and intrauterine adhesions

The American Fertility Society*

Birmingham, Alabama

The need for standard classification schemes for mechanical problems associated with infertility and fetal wastage is well recognized. In an effort to satisfy this need, The American Fertility Society (AFS) formed a classification committee chaired by Veasy C. Buttram, Jr., M.D. Subcommittees were created to distribute the responsibility for development of the various schemes desired. Committee members were chosen according to their knowledge and desire to contribute. Approval was obtained from the Board of Directors of the AFS and the Executive Committee of The Society of Reproductive Surgeons. Hopefully, editors of medical journals will not accept manuscripts for publication unless they conform to these guidelines. The classification schemes submitted (with a corresponding editorial) are relatively simple to follow. Relevant information can be incorporated into one sheet with appropriate illustrations. In several of the schemes, a scoring system is included to help the surgeon not only document the severity of disease, but to formulate a prognosis. For all schemes a space is provided for the surgeon to note his/her prognosis based upon the extent of pathology and the surgery performed. No classification is perfect, and modification is likely in the future. Use, as well as critique, of these AFS classification is encouraged. All forms, in addition to The Revised-AFS Classification of Endometriosis, will be provided by the AFS administrative office.

Received March 10, 1988. * Drafted by the following committees: Adnexal Adhesions-Editorial: Veasy C. Buttram, Jr., M.D., Committee: Veasy C. Buttram, Jr., M.D., James Daniell, M.D., Richard Dickey, M.D., Victor Gomel, M.D., Jerome Hoffman, M.D., Jaroslav Hulka, M.D., Luigi Mastroianni, Jr., M.D. and Thomas Vaughn, M.D. Distal Tubal Occlusion-Editorial: Victor Gomel, M.D., Committee: Veasy C. Buttram, Jr., M.D., Victor Gomel, M.D. and Carl Levinson, M.D. Tubal Occlusion Secondary to Tubal Ligation-Editorial: Alvin Siegler, M.D., Committee: Veasy C. Buttram, Jr., M.D., Jaroslav Hulka, M.D., Carl Levinson, M.D. and Alvin Siegler, M.D. Tubal Pregnancies-Editorial: Alan DeCherney, M.D., Committee: Veasy C. Buttram, Jr., M.D., Alan DeCherney, M.D. and John Rock, M.D. Mullerian Anomalies-Editorial: William Gibbons, M.D., Committee: Veasy C. Buttram, Jr., M.D., S. Jan Behrman, M.D., Williams Gibbons, M.D., Howard Jones, M.D. and John Rock, M.D. Intrauterine Adhesions-Editorial: Charles March, M.D., Committee: Veasy C. Buttram, Jr., M.D., Alan DeCherney, M.D., Charles March, M.D., Robert Neuwirth, M.D. and Thomas Vaughn, M.D. Reprint requests: The American Fertility Society, 2140 11th Avenue South, Suite 200, Birmingham, Alabama 35205.

ADNEXAL ADHESIONS

This classification (Table 1) is a modification of the scheme for adnexal adhesions provided in the Revised AFS Classification for Endometriosis.l Although the scores applied to differentiate between minimal, mild, moderate, and severe adhesions are arbitrary, they are considered appropriate until prospective studies are performed that dictate the need for change. Important features of this scheme are that (1) the prognosis for conception is based on the score applied for the adnexa with the least amount of pathology, (2) the importance of the fimbriated end

of the fallopian tube is emphasized, and (3) a differentiation is made between filmy and dense adhesions. The latter subjective observation may create conflict among gynecologists. However, it is the responsibility of the physician to make judgment not only of the extent of the disease, but a distinction between what are thought to be filmy and dense adhesions. Generally, filmy adhesions can be lysed by scissors, electrocautery, or laser without undue risk of bleeding or injury to adjacent organs.

DISTAL TUBAL OCCLUSION

Salpingostomy (salpingoneostomy) is the creation of a tubal ostium in a tube with a totally occluded fimbriated end as is found in a hydrosalpinx or sactosalpinx.² The Ad Hoc Committee of the International Federation of Fertility and Sterility (IFFS) classified salpingostomy (salpingoneostomy) as (a) terminal, (b) ampullary, and (c) isthmic.^3,4 A terminal salpingostomy is preferable since the tube is conserved in its entirely and normal tubo-ovarian relationships are maintained. Specific circumstances such as the presence of extensive intratubal adhesions in the terminal portion of the tube may oblige the surgeon to excise the affected portion of the oviduct and resort to an ampullary salpingostomy. Reversal of a prior fimbriectomy (Koerner's sterilization) also requires an ampullary salpingostomy; this is outside the scope of this editorial, which is limited to distal occlusion caused by pelvic inflammatory disease. Isthmic salpingostomies have an exceedingly poor yield and should be abandoned. The utilization of microsurgical techniques in reconstructive infertility surgery has yielded significant improvements in the results of these procedures. However, improvements in outcome have not been as evident in salpingostomy for hydrosalpinx despite the high postoperative tubal patency rate obtained with microsurgery. The relative improvement in the rate of intrauterine pregnancy noted after such procedures may be due to the increase in the rate of postoperative patency, which in turn is also responsible for the significant rise in the rate of tubal pregnancy. The surgery-pregnancy time interval after salpingostomy is fairly long; more than 60% of the patients who achieved an intrauterine pregnancy did so after the first postoperative year.^5-9 Tubal function, hence the pregnancy rate subsequent to a salpingostomy for hydrosalpinx, is largely dependent on the status of the fallopian tube at the time of the surgical intervention. Individually and collectively, the following factors affect the outcome: condition of the endosalpinx, extent of ampullary dilatation, thickness and rigidity of the tubal wall, extent and nature of pelvic and periadnexal adhesions.¹⁰ The presence of well-defined regal markings within the oviduct, especially when not greatly dilated indicates a good prognosis¹¹ Conversely, intratubal adhesions, especially when they involve a large part of the oviduct, represent a contraindication to salpingostomy.¹² The prognosis is affected adversely by the extent of ampullary dilatation, especially if this is more than 3 cmY-15 A hydrosalpinx with rigid, thickened tubal wall carries an exceedingly poor prognosis. These tubes show little evidence of dilatation.^12,16,17 In addition, the prognosis is inversely proportional to the extent ofthe periadnexal and pelvic adhesive process.^12-14,16
The tools currently available to investigate the oviducts are hysterosalpingography and laparoscopy. They are complementary procedures, and they are more effective in identifying cases at the extremes of the spectrum. At one extreme are the favorable cases, with tubes that show little dilatation, exhibit rugal markings, are supple without any suggestion of tubal wall thickening, and without significant pelvic or periadnexal adhesions. At the other extreme, the findings are sufficiently severe to conclude that surgery is contraindicated; these include the presence of extensive, dense pelvic adhesions; extensive intratubal adhesions; extreme dilatation of the tubes; or changes consistent with or evidence of prior tuberculous salgingitis. The intermediate cases may be more difficult to assess. In certain instances periadnexal adhesions prevent proper visualization and assessment of the oviduct. Frequently, it is difficult to assess accurately the thickness and rigidity of the tubal wall until the time of the actual surgery. The value of salpingoscopy (performed during the diagnostic laparoscopy) in predicting postoperative prognosis is currently under investigation.
The reported viable pregnancy rates subsequent to microsurgical salpingostomy vary between 19% and 35%, and the ectopic pregnancy rate ranges between 5% and 18%. Both intrauterine and ectopic pregnancy rates increase with longer followup periods. With favorable cases the outcome is better; intrauterine pregnancy rates varying between 50% and 80% have been reported.^13,16-18 For this reason, several authors have proposed a scoring system to identify prognostically favorable cases. Owing to the progress in the field of in vitro fertilization, this therapeutic modality has become a credible alternative to surgery in cases of hydrosalpinx. This development argues in favor of a more rigorous selection of patients for microsurgical treatment. Furthermore, in selected cases laparoscopic salpingostomy provides another alternative to microsurgery; this has the advantage of being performed during the initial diagnostic laparoscopy without recourse to a laparotomy.^19-21 A better understanding of the prognostic factors will permit the selection of the most appropriate treatment modality; this individualization of therapy should enable us to achieve better overall results.
The proposed scoring system (Table 2) has been devised to predict the postoperative prognosis of salpingostomies whether performed by laparotomy or laparoscopy. It has been designed for use at the time of the actual reconstructive surgery but may also be employed at the completion of the preoperative investigation to determine the accuracy of the information gained. Various factors known to affect the prognosis have been assigned values. Although the assignment of specific numbers may appear arbitrary, they have been selected after taking into account observations and conclusions of the previously cited reports. The factors to be scored include extent and nature (type) of adhesions, thickness and rigidity of the tubal wall, the distal ampullary diameter, and the extent of preservation of mucosal folds in the neostomy site. We recommend the use of the IFFS classification to quantify the extent of adhesions, since one already exists. Minimal/mild: 1 cm of tube and/or ovary involved. Moderate: adhesions partially surround tube and/or ovary. Extensive: peritubal and periovarian adhesions totally encapsulate the tube and/or ovary. Along with other investigators,22 we recommend intraoperative salpingoscopy to visualize the whole length of the ampullary lumen. Whereas the status of the ampullary endosalpinx is also an important prognostic parameter, we elected not to include it in the scoring system at this time, since salpingoscopy is not being practiced universally.
The proposed classification and the various values (numbers) in the scoring system will need to be modified as a result of prospective studies. This task was undertaken without any illusion that it would yield the perfect classification; however, a beginning was necessary.

TUBAL OCCLUSION SECONDARY TO
TUBAL LIGATION

The purpose of this operative classification (Table 3) is to be able to dentify similar procedures so that reasonable comparisons can be made among different studies, prognoses can be given to patients who then make informed choices, and physicians can respond to their queries with increased knowledge.
The type of sterilization can influence the technique of repair and prognosis. The influence of age on success is likely. Older women usually have longer intervals since sterilization, increased ovulatory disturbances, and older partners. Thus, not only age but time interval from sterilization to reversal is important to document. Electron microscopic studies have shown relative atrophy of the tubal epithelium in the proximal stumps of women sterilized more than 5 years prior to reversals and increased abnormalities between the sterilization and sterilization reversal (SR). These include flattening of mucosal folds, deciliation, adenomyosis, and polyposis. Disorders of the tubal isthmus occur with chronic (long-standing) occlusion. It is important to perform a hysterosalpingogram (HSG) prior to surgery to determine the status of the proximal portion of the fallopian tubes. Although laparoscopy is not essential when one is confident regarding the status of the distal portion of the fallopian tubes, it is generally advisable.
The lengths of the tubes after completion of the SR operation should be recorded. The types of anastomoses are intramural-isthmic, intramuralampullary, isthmic-isthmic, isthmic-ampullary, and ampullary-ampullary. Equal-sized lumens represent either isthmic-isthmic or ampullaryampullary anastomoses. It is uncommon to encounter significant periadnexal adhesions in these women during the SR, and it is not necessary to add a point scoring system for adhesions. The performance of other tubal or pelvic procedures and associated abnormalities should be listed.
Few series of more than 50 SRs have been reported; postoperatively the live births vary from 48% to 78% and the tubal pregnancies from 1.7% to 6.5%. Since there are an estimated 5,000 reversal procedures performed annually and less than 1,000 have been reported, it is obvious that most SRs are being done by the "Silent Majority."
Operations for reversals, other than anastomosis, include salpingoneostomy and utero-tubal implantation, the latter involving either the isthmus or ampulla. The length of the extrauterine portion

of tubal segment should be measured as well as its new location in the uterine wall. Salpingoneostomy is done for reversal of a previous fimbriectomy, and the length of the extrauterine tubal segment should be noted. Laparoscopy is essential in these patients before SR is attempted.
A separate classification is needed for anastomosis operations performed because of pathologic rather than iatrogenic occlusions. These procedures have a different prognosis from those done for SR. Invariably, these women will need a screening HSG and laparoscopy; however, the categories for anastomosis are quite similar to those for SR. We would suggest including a category entitled "Surgical Pathology Report." The listings would be luminal fibrosis, salpingitis isthmica nodosa, endometriosis, tuberculosis, and other. Of concern is that in approximately 10% of specimens for histologic examination, no pathologic lesions are found.

TUBAL PREGNANCIES

Approximately 60,000 ectopic pregnancies occur in this country in a single chronologic year, making surgery for an ectopic pregnancy a frequently performed procedure in reproductive surgery.
New technology has changed ectopic pregnancy as an entity, both in diagnosis and in management. The combination of serial beta pregnancy testing and ultrasound, both abdominal and vaginal, has increased our diagnostic acuity so that ectopic pregnancies are now diagnosed between 6 and 8 weeks gestation, instead of being diagnosed at 10 weeks or later as in the past. This change has ushered in the age of conservative surgery in the management of ectopic pregnancy, in contradistinction to the ablative surgery that was done a decade ago. At that time, ectopic pregnancies were diagnosed after rupture has occurred, and very little in the way of alternative remained.
Classification for ectopic pregnancy should be comprehensive, and not overwhelming. It should be helpful and informative. Completion of the forms should be a welcome addition to management, rather than a chore. This classification of ectopic pregnancy (Table 4) includes some historic facts indigenous to the patient that indicate a high risk for ectopic pregnancy, i.e., previous tubal surgery, previous ectopic pregnancy.
The anatomic location of the ectopic pregnancy is critical. Ampullary ectopic pregnancies usually grow outside of the lumen, and therefore postoperative patency is to be expected. Linear salpingostomy seems to be the treatment of choice in these cases. Conversely, isthmic ectopic pregnancies grow within the lumen and destroy the tubal mucosa; in these cases, segmental resection seems to be the treatment of choice.²³ Interstitial pregnancies in time will most likely be treated by chemotherapy, not surgery; therefore, this is also important to note. Ruptured or unruptured should be noted, and site of rupture, or imminent rupture are also critical issues since ectopics that rupture into the broad ligament, rather than dorsally, represent an entirely distinct entity requiring more extensive surgery.
It is essential to record the mode of treatment. Many ectopic pregnancies are now removed by a laparoscopic approach as opposed to laparoscopic diagnosis with treatment at laparotomy. The method of surgical repair should also be included: magnification vs. no magnification. The method of surgical approach needs to be elucidated (no surgery, expression [milking the tube], linear salpingostomy, or segmental resection with primary or secondary anastomosis or no anastomosis planned). Also to be considered in the classification are other anatomic observations including adhesions, status of the contralateral tube, and corpus luteum. And last, adjunctive therapy and postoperative course are helpful.
In any classification, foresight is the key. What will you want to know in the future about this particular patient? This information is critical in counseling patients about future success rates, especially now that patients are having multiple ectopic pregnancies, rather than only two as in the past.²⁴

MULLERIAN ANOMALIES

The reason for providing a standardized classification is its value to the practicing physician. It should be simple to use to promote compliance and flexible enough to provide the capacity to fit almost "every" possibility. A frequently used classification by Strassman25 grouped the anomalies into the symmetric double malformations (didelphus, bicornuate, septate) or the asymmetric ones (unicornuate with or without a rudimentary horn). This resulted in a reduction in the amount of available information that reached the literature because of arbitrary groupings particularly with regard to the bicornuate and septate uteri. Consequently, if the clinician wishes to inform, for example, a patient with a unicornuate uterus with a

rudimentary functional horn of her prognosis, the sources are limited to individual case reports or a single series containing few cases. Few reports in the literature provide fetal wastage rates for separate uterine malformations.^26-29 The conclusions reached in these individual series vary greatly because they are based on such a small number of observations. An easy-to-use flexible reporting system will allow the clinician to group cases with others so that in the future appropriate conclusions can be reached based on adequate numbers. This classification (Table 5) organizes the anomalies according to the major uterine anatomic types. It allows the user to indicate the malformation type as well as the associated variations involving the vagina, cervix, tubes, ovaries, and urologic system. Later, when data is compiled, a unique "computerizable" code can be generated for reporting purposes. The classification committee had difficulty deciding how to include the arcuate uterus. Because the arcuate uterus is externally unified, it could be classified as a form of a partial septate uterus. However, since in contrast to the other malformations, the arcuate uterus appears to behave benignly, it was thought that it should be classified separately for the present. Thus, data can be generated that can be used to determine if it should remain in a classification of abnormal uterine malformations or is a variant of normal anatomy.

INTRAUTERINE ADHESIONS

The charge to our committee was to develop a simple classification system for intrauterine adhesions (IVA). Although other schema to grade IVA have been available, the widespread use of hysteroscopy demonstrated their flaws. The first endoscopic classification was accurate but failed to consider menstrual pattern, thus the prognostic significance of endometrial sclerosis or atropy could not be considered.³⁰
The classification proposed (Table 6) provides an objective scoring system that permits both the indirect (hysterographic) and direct (hysteroscopic) grading of IVA. The location of adhesions is likely to be of prognostic importance for infertile women because most implantations occur in the top-fundal portion of the uterus and because adhesions in the cornual recesses may cause tubal obstruction. Drawings should be made to document location and extent of adhesions. Thus, these findings, as well as those of laparoscopy should be recorded in the section entitled "Additional Findings."
The number of classifications of IVA is exceeded only by the number of treatment regimens. Because suitable intrauterine contraceptive devices are no longer readily available to use as "splints" within the uterine cavity, many treatment schemes are likely to change. Therefore, these data are even more important.
Finally, it is advised that one classification form be completed immediately after surgery and another following all therapy including the follow-up hysterosalpingogram or hysteroscopy that is performed before permitting the patient to attempt pregnancy. The comparison of the two forms will provide objective data about the success of therapy and will permit the true worth of the classification system to be assessed.
As with the classification system for endometriosis, continuous evolution is expected. The comparison with an earlier successful classification reported in this journal attests to the workability of the new system. However, broader application may cause flaws to become more evident resulting in additions, deletions and/or modifications.

REFERENCES

1. The American Fertility Society: Revised classification of endometriosis. Fertil Steril 43:351, 1985
2. Gomel V: Correction of terminal occlusion ofthe oviduct. In Microsurgery in Female Infertility, Boston, Little, Brown and Company, 1983, p 163
3. Gomel V: Recent advances in surgical correction of tubal diseases producing infertility. Curr Probl Obstet Gynecol 1:10, 1978
4. Gomel V: Classification of operations for tubal and peritoneal factors causing infertility. Clin Obstet GynecoI23:1259, 1980
5. Gomel V: Salpingstomy by microsurgery. Fertil Steril 29:380, 1978
6. Gomel V, Swolin K: Salpingostomy: microsurgical technique and results. Clin Obstet Gynecol 23:1243, 1980
7. Jansen RPS: Surgery-pregnancy time intervals after salpingolysis, unilateral salpingostomy, and bilateral salpingostomy. Fertil Steril 34:222, 1980
8. Russell JB, DeCherney AH, Laufer N, Polan ML, Naftolin F: Neosalpingostomy: comparison of 24- and 72-month follow- up time shows increased pregnancy rate. Fertil Steril 45:296, 1986
9. Kitchin JD III, Nunley WC Jr, Bateman BG: Surgical management of distal tubal occlusion. Am J Obstet Gynecol 155:524, 1986
10. Gomel V: Results of reconstructive infertility surgery. In Microsurgery in Female Infertility, Boston, Little, Brown and Company, 1983, p 225
11. Ozaras H: The value of plastic operations on the fallopian tubes in the treatment of female infertility: a clinical and radiologic study. Acta Obstet Gynecol Scand 47:489,1968
12. Gomel V: Contraindications. In Microsurgery in Female Infertility, Boston, Little, Brown and Company, 1983, p 129
13. Rock JA, Katayama P, Martin EJ, Woodruff JD, Jones HW Jr: Factors influencing the success of salpingostomy techniques for distal fimbrial obstruction. Obstet Gynecol 52:591, 1978
14. Verhoeven HC, Berry H, Frantzen C, Schlosser HW: Surgical treatment for distal tubal occlusion: a review of 167 cases. J Reprod Med 28:293, 1983
15. Donnez J, Casanas-Roux F: Prognostic factors of fimbrial microsurgery. Fertil Steril 46:200, 1986
16. Boer-Meisel ME, te Velde ER, Habbema JDF, Kardaun JWPF: Predicting the pregnancy outcome in patients treated for hydrosalpinx: a prospective study. Fertil Steril 45:23,1986
17. Mage G, Pouly JL, de Joliniere JB, Chabrand S, Riouallon A, Bruhat M-A: A preoperative classification to predict the intrauterine and ectopic pregnancy rates after distal tubal microsurgery. Fertil Steril46:807, 1986
18. Frantzen G, Schlosser HW: Microsurgery and postinfectious tubal infertility. Fertil Steril 38:397, 1982
19. Gomel V: Salpingostomy by laparoscopy. J Reprod Med 18:265, 1977
20. Gomel V, Taylor PJ, Yuzpe AA, Rioux JE: Laparoscopy and Hysteroscopy in Gynecologic Practice. Chicago, Year Book Medical Publishers, Inc, 1986, p 148
21. Daniell JF, Herbert CM: Laparoscopic salpingostomy utilizing the CO2 laser. Fertil Steril 41:558, 1984
22. Henry-Suchet J, Loffredo V, Tesquier L, Pez JP: Endoscopy of the tube (tuboscopy): its prognostic value for tuboplasties. Acta Eur Fertil16:139, 1985
23. DeCherney AH, Maheuz R: Modern management of ectopic pregnancy. Curr Prob Obstet Gynecol, Chicago, Year Book Medical Publishers, 6:9, 1983
24. Silidker J, Tarlatzis BC, DeCherney AH: Fecundite apres deux grossesses ectopique. Contraception-Fertilite-Sexualite 12:701, 1984
25. Strassman P: Die operative vereinigung eines doppelten uterus. Zentralbl Gynakol 31:1322, 1907
26. Caprara BJ, Chuang JT, Randall CL: Improved fetal salvage after metroplasty. Obstet Gyneco131:97, 1968
27. Michalas S, Prebedourakis C, Lolis D, Antasalkis A: Effect of congenital uterine abnormalities on pregnancy. Int Surg 61:557, 1976
28. Behrman SJ, Musich JR: Obstetrical outcome before and after metroplasty in women with uterine anomalies. Obstet Gynecol 52:63, 1978
29. Buttram VC Jr, Gibbons WE: Mullerian anomalies: a proposed classification (an analysis of 144 cases). Fertil Steril
32:40,1979 30. March CM, Israel R, March AD: Hysteroscopic management of intrauterine adhesions. Am J Obstet Gynecol 130:653, 1978