In the study presented here, we analysed the rate of recurrence of 206 patients between the ages of 17 and 59 years who underwent surgery for moderate to severe endometriosis in a period of eleven years and who completed the questionnaire sent to them. Common benign comorbidities did not exclude the patients from the analysis, which, to our mind, represents a real cross section of the normal population.
We also analysed the pregnancy rate for 120 women who wanted to become pregnant after surgery. The majority of these women was under the age of 45 (N = 117; 97.5%). Compared to other publications, this is a high number of patients. Recently, Sun et al. published a study with 59 infertile patients after laparoscopic cystectomy with six to ten years of follow-up [24].
Our long follow-up period of four to 14 years enabled the registration and evaluation of the long-term efficacy of surgical endometriosis lesions resection with regard to the development of recurrence and postoperative pregnancy. However, as a limitation of our study, the long follow-up period was problematic to the extent that a number of patients could no longer be reached per mail or telephone or through the attending physicians. After contacting 456 of the patients enrolled in the study, the return rate of the questionnaires was 45.2%. This is, however, in the higher range when compared to other questionnaire-based studies [25–27].
As Haas et al. [28] have shown in their retrospective epidemiological study, endometriosis affects young adolescents as well as perimenopausal women. Our study also included these women, so that overall, a cross section of all affected age groups was achieved. Obviously the fertility capacity decreases considerably over the age of 45 years; therefore, the low number of three women over the age of 45 years who were trying to become pregnant after surgery was to be expected. Amazingly one woman at the age of 48 gave birth to two children consecutively.
Irrespective of the high number of patients analysed in this paper, we are fully aware that the retrospective approach of the study limits the value of the results.
Stage-related surgical procedures and postoperative rate of recurrence
Surgical treatment for endometriosis is technically challenging and is sometimes similar in its complexity to the radical surgeries in gynaecological oncology. During the excision of extensive endometriosis lesions, close attention must be paid to the preferably complete removal of the endometriosis lesions and the preservation of the functional, reproductive structures in the minor pelvis. Therefore, especially in women wanting to have children, patient age, previous pharmacological or surgical therapies and the extent of the disease must therefore be considered [29, 30]. Furthermore, the benefits of excessive surgery before IVF/ICSI are still uncertain and must be weighted carefully against the risks [15].
Ford et al. [31] and Garry et al. [32] showed that radical endometriosis lesions resection can help clearly relieve symptoms and improve the quality of life. However, in the case of extensive excision, the risk of damage to the urinary and intestinal tract with respective late damage must be noted [30, 33, 34]. In our cohort, 8.3% of patients (17/206) required a procedure for intestinal or bladder surgery due to deep infiltrating endometriosis. Therefore, vaginal-rectal palpitation is obligatory before surgery.
The wish to have children can limit the extent of resection so as to preserve the function of the tubes and ovaries and to diminish the ovarian reserve as little as possible. Furthermore, some endometriosis lesions are located in places where surgical removal is technically difficult, impossible or associated with a high rate of complications affecting, for example, adenomyosis as well as pararectal lesions.
Complete surgical removal of endometriosis lesions was achieved in 90.8% (187/206) of our study population, which is – with respect to the high stages of disease – a high proportion. The most common reason for incomplete resection was the patient’s existing wish to have children with the relevant high risk of serious destruction of the ovaries and tubes.
Sometimes, the extent of endometriosis involvement is difficult to assess before surgery despite optimal screening, which was recently reported by Wattiez et al. and Espada et al. [35, 36]. The rASRM classification system does show the stage during surgery with pinpoint precision, but it does not always correlate with disease activity or the level of the state of pain. Furthermore, the rASRM classification is not indicative of the postoperative fertility potential [33, 37].
With 6.8%, the peri- and postoperative rate of rather minor complications was low despite the partly very extensive disease and is in line with or even lower than the data published in the literature [20, 30, 38, 39]. In half of these patients, the complications were associated with the immediate operative distance of the endometriosis lesions at the respective anatomical structures. The most common postoperative complication described was renal congestion [20], which developed in 1% of our study population.
The complete or partial relief of symptoms associated with endometriosis was achieved in 93.2% of our patient population, whereby complete remission of symptoms accounted for the largest percentage with 76.6%. This result is high when compared with other published data [40–42]. We were not able to establish any differences in symptom relief regarding the rASRM stage, which also corresponds to the results of Vercellini et al. [33]. Comptour et al. [43] also demonstrated that laparoscopic endometriosis lesions resection irrespective of the rASRM stage has a positive, minimum three-year effect on both symptoms (lower abdominal pain, dyspareunia) and the quality of life.
A recurrence that led to repeated surgery developed in 21.8% of patients in our study population. This value falls into the lower range of data published in the literature, where a recurrence rate of between 6.0% and 67.0% after five years is specified [18–20, 22, 34, 44, 45]. It should be kept in mind that in some of our patients, 14 years had passed since surgery. Nirgianakis et al. evaluated the recurrence patterns after surgery depending on the different endometriosis subtypes: superficial peritoneal endometriosis, ovarian endometrioma and deep infiltrating endometriosis. Although peritoneal endometriosis and ovarian endometrioma are more likely to present with the same subtype at recurrence, the time to recurrence was independent of the lesion subtype at first surgery [22].
We were not able to establish a statistically significant association between the recurrence rate and the disease stage; there is discord in the literature in this regard. While Vercellini et al. [33] were not able to demonstrate an association between stage and risk of recurrence and attribute a rather low predictive power to the rASRM classification, Busacca et al. [46, 47] showed that both the rASRM stage as well as previous surgeries constituted influencing factors for recurrence. Tobiume et al. published data from 352 patients who underwent surgery for endometrioma and determined that only the rASRM score was correlated with a recurrence of 28.7% after five years [45].
Possible reasons for this discrepancy can be the difference in the technical performance of the surgery or the expertise of the surgeon, the definition of the measured endpoint (symptom relief or repeated surgery) as well as the length of the follow-up period and the patient population [19]. The studies of Busacca et al. [46, 47] focused on patients with stage III and IV endometriosis but defined the development of recurrence by means of postoperative sonographic findings and questionnaires without histological and/or laparoscopic confirmation. Vercellini et al. [33], on the other hand, examined patients of all rASRM stages and defined the development of recurrence by means of sonographic findings and a new surgical intervention. The findings of our study essentially correspond with the results of Vercellini et al. [33]. Our study concentrated only on patients with moderate and severe endometriosis where the development of recurrence was defined by the need for new surgery with histological confirmation.
Even if the initial stage of endometriosis had no effect on recurrence in our study, recurrence developed significantly more often in those with incomplete resection (36.8%) than in those with complete resection (20.3%). These observations correspond with other publications [19, 44, 48]. Hormonal follow-up treatments are indicated after endometriosis surgery since they verifiably decrease the recurrence rates [48–51].
Therefore, based on the recommendations of the ESHRE guidelines from 2014, it is currently advised that patients start postoperative hormone therapy until there is a wish for children. This includes administrations or even combined long-cycle oral or vaginal contraceptives seen as a ´pseudopregnancy regimen` without therapy pause and intrauterine gestagens/hormonal IUDs [16]. Intermittent use is associated with a higher rate of recurrence [50]. Most international studies used oral contraceptives due to good tolerance and a good safety profile, cost efficiency and the option of a longer intake period. Schäfer and Kiesel [52] recommend a postoperative maintenance therapy in the form of gestagen monotherapy or continuous use of a combination product (´long-cycle`)..
Thirty-five women in our population (33.0%) received postoperative therapy with GnRH agonists, often as pre- and postoperative ´sandwich therapy`, to ensure treatment success and optimize the likelihood of becoming pregnant in the long term. However, because of the hypoestrogenism, GnRH analogues are characterized by pronounced side effects, such as menopausal symptoms, mood swings and listlessness as well as iatrogenic osteoporosis [47]. Therefore, an add-back-therapy using transdermal oestrogens can be used to reduce these side effects. Nevertheless, it should be noted that, after six months of therapy, 50% of vital ectopic endometrium could still be established [53].
A total of 100 women (48.5%) in our population did not receive postoperative pharmacological therapy. In 53%, the reason for this was indicated as the active wish to have children directly after surgery. In 22.8% (47/206), adequate postoperative pharmacological therapy was not administered or taken, which should be improved.
The current gold standard for surgical endometriosis lesions resection is laparoscopy. This was performed in a total of 63.1% of interventions in our patient population. Reasons for conversion from laparoscopy to laparotomy, which took place in 15%, can be an unexpected extensive endometriosis involvement or a pronounced state of adhesions that makes a laparoscopic procedure impossible or prone to complications [54, 55]. Especially in patients with severe endometriosis with pronounced adhesions and/or deep infiltrating endometriosis, it is not uncommon to plan the surgery in the form of a laparotomy. In our study, a laparotomy was performed in 21.8% of patients. Mostly, this was intended beforehand since fertility-preserving and restorative interventions on the tubes and uterus are performed depending on the equipment of the particular hospital, as the use of a surgical microscope and microsurgical instruments is necessary.
Surgical access, on the other hand, had no effect on the rate of recurrence in our patient population. This observation corresponds to international data [56, 57].
Laparoscopic endometriosis lesion resection is associated with advantages such as shorter hospital stays, faster recovery, less postoperative pain and, last but not least, better cosmetic results than laparotomy [16]. However, well-performed laparoscopic surgery requires more technical expertise from the surgeon than laparotomy, especially if there is extensive endometriosis involvement [58].
Stage-dependent postoperative pregnancy rate
The postoperative pregnancy rate in our study was 65.8% (79/120) if the women wanted to have children. The average number of pregnancies was 1.4 and therefore higher than the postoperative pregnancy rates described in the literature, which are between 32.4% and 65.5% after the excision of a stage III and IV endometriosis lesions [24, 59, 44, 46, 60]. The birth rate was 80.5%, and a previous pregnancy improved the chance of becoming pregnant again after surgery (76.0% versus 61.4%).
As expected, we were able to determine a statistically significant correlation between the onset of pregnancy and maternal age (p < 0.0017). A total of 76.7% of women up to the age of 34 years became pregnant, while only 48.9% of women from the age of 35 up were able to conceive. The oldest woman who became pregnant twice was 48 years old. The fertility capacity decreases as a women age, and moreover, with the resection of ovarian endometrioma, there is a risk of reduction in the ovarian reserve and thus less likelihood of conceiving [60].
However, not all women wanting to have children benefit from endometriosis lesion excision, therefore, in patients with tubal or andrological factors, assisted reproduction such as in-vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI), may also be indicated [10, 61].
In our study, we deliberately did not ask about the method of conceiving (natural or by IVF). Sterility is a multifactorial process that cannot always be treated with the endometriosis surgery performed here and sterility factors of the partner, for instance, are beyond our control. For an analysis of the chance of natural conception and the need for assisted reproduction, a distinctly larger patient group is required.
In the international literature, a higher percentage of endometriosis patients become pregnant by means of IVF. In a study by Soritsa et al. [59], the pregnancy rate after removal of stage I and II endometriosis lesions was 66.9% in 121 patients and was 65.5% after the removal of stage III and IV endometriosis lesions in 58 patients. Twenty-eight percent of all patients conceived spontaneously, and 72% conceived by IVF, mainly after pre-therapy with GnRH analogues over 4.9 months. The percentage of IVF pregnancies was surprisingly slightly higher in the group with moderate and severe endometriosis. There were no differences regarding the rate of miscarriages and births. Vercellini et al. showed similar data [33]: Even though 70.4% of women became pregnant after the resection of moderate endometriosis lesions and (only) 56.4% after the removal of severe endometriosis lesions, a statistically significant effect of disease stage on the postoperative pregnancy rate could not be established. This observation corresponds with that of the study from Guzick et al. [62]. Another meta-analysis showed that in patients with prolonged, severe endometriosis, the postoperative administration of GnRH analogues over three to six months can be advisable before the start of assisted reproduction [63].
In most studies investigating the effect of surgical therapy of endometriosis with regard to several outcome parameters such as symptom relief, recurrence rate and pregnancy rate, the surgery was performed as laparoscopy [25, 33, 40, 44, 64]. A prospective study compared the effect of the resection of colorectal endometriosis lesions with laparoscopy and laparotomy on fertility and showed that the rate of spontaneous postoperative pregnancies was distinctly higher in the group with laparoscopic access [65]. Daraï et al. also showed that laparoscopic endometriosis lesions resection is associated with higher postoperative pregnancy rates than resection via laparotomy, and has similar positive effects on symptoms and the quality of life [66].
Our results substantiate these findings with regard to postoperative pregnancy rates after laparoscopic surgery, which were statistically higher (74.3%) than after laparotomy (61.3%) as well as after conversion surgery (42.1%). The advantage of laparoscopy is likely to be found in the low formation of adhesions, since the tissue damage is lower overall [67]. In our collective patients, we had a high percentage of women with severe tubal damage who were treated by laparotomy to restore fertility, using surgical microscopy and microsurgical instruments. Possibly, the lower rate of postoperative pregnancy after laparotomy was caused by these unfavourable conditions of the fallopian tubes.
After complete resection of endometriosis lesions, the pregnancy rate was higher (68.5%) than after incomplete excision (41.7%). However, the correlation was weak. The retrospective cohort study of Soriano et al. [64] was able to demonstrate that radical surgical removal that also included a procedure for intestinal and bladder surgery does not have to involve a decrease in fertility. Surgical therapy is able to restore the pelvic anatomy that is changed by endometriosis [10]. However, it is essential to strike a balance between complete resection and the preservation of the reproductive organs and their function. In addition to the risk of diminishing the ovarian reserve, attention must be paid to the formation of new or rather further adhesions [68].
A new and future possibility to preserve fertility can be oocyte vitrification for patients not only in oncological situations but also in situations such as severe endometriosis that may induce the risk of premature ovarian failure [69–71].