Segmental Rectum Resection for Deep Endometriosis and Excision Similarly Improve Sexual Function and Pain
Round 1
Reviewer 1 Report
The article reports on the outcomes of women treated for endometriosis. 33 had low bowel resection and 23 had conservative excision. Although the abstract (line 26) suggested a comparison between the two groups, the extent of treatment was decided based on need at the time of surgery. Both cohorts were reported to have improvement in pain and in sexual function. Detail of the extent of endometriosis or of whether women underwent hysterectomy or oophorectomy is not provided.
The results are from one centre in Brazil and the operations were done by one of two operators.
There is no indication as to the duration of follow-up.
The main points that need to be addressed are as follows:
1- Please provide details of the extent of surgery needed for this cohort including any complications.
2- 132 women were invited to take part, but only 61 accepted: what were the characteristics of those declined to take part. The rationale given for the low uptake of a before and after surgery questionnaire is that the topic of sexuality is sensitive in Brazil. Please explain in what way this is different than elsewhere in the world.
3- Line 20 Prof em Department?
4- Line 61 live long urinary? You mean life?
5- Line 71 or e sexual quality?
6- Line 136 Endo GIATM please correct
7- Line 153: After surgery desire… please revise
8- Table 1: Para? Deep endo (ml) please correct abbreviations. Pain: is that preoperative pain score?
9- Table 2: what is ED30? What is the score provided?
10- The authors suggest that both treatment approaches resulted in improvement, nevertheless despite some improvement, the total FSFI scores after surgery remained below 26 which suggest persistent sexual dysfunction. The study does not provide evidence as to what the improvement in scores translated to.
11- Table 2: CPP please explain the abbreviations and all headings provided.
12- The use of Sum of all pain is meaningless and should be removed. Data summation requires full validation before it could be used in this manner.
13- Line 177-180 should be removed. Dyspareunia is an important measure: what does the P<0.0001 refer to??
14- The data for the EHP30 scores are incompletely/inaccurately presented.
15- Line 201: The finding that restricted resection and linear excision have comparable effect on sexual function (and show some improvement) does not eliminate the concerns that arise from the impact of bowel resection on sexual function. The differences may arise from the different starting points and can serve as a warning as to the limits of possible improvement. This is particularly the case as the data presented, although showing some improvement, does not demonstrate a return to normality.
16- Line 212: please justify the statement or remove it.
17- Line 216: It is recognised that a large study is needed, but the need probably justifies the size and expense. The results of this study does not provide the required clarity of the likelihood of success or of the return to normality or of the likely long term outcomes. All of this suggest the need and that such study is ethically justified.
see above
Author Response
Thank you for this detailes and excellent review. All answers were addressed and a spell checker was run to detect eventual remaining typing errors.
1- Please provide details of the extent of surgery needed for this cohort including any complications.
Technique and results were described extensively in ref 44. Recent publications were added as res 45 and 46. The principles of surgery are described in l123 ‘Excision was visually complete without safety margins’.
L100 it was added that postoperative complications were an exclusion criteria.
2- 132 women were invited to take part, but only 61 accepted: what were the characteristics of those declined to take part. The rationale given for the low uptake of a before and after surgery questionnaire is that the topic of sexuality is sensitive in Brazil. Please explain in what way this is different than elsewhere in the world.
This has been revised as follows ‘Only 61 women accepted, probably because women were reluctant to come to the hospital during the COVID pandemic. Also, discussing sexuality might be more sensitive in Brazil than in other parts of the world.’
3- Line 20 Prof em Department?
‘Prof em Department of Obstetrics and Gynaecology
4- Line 61 live long urinary? You mean life?
Thank you yes
5- Line 71 or e sexual quality?
Thank you, the e is obviously a mistake that escaped revision
6- Line 136 Endo GIATM please correct
Thank you. After checking this indeed should be Endo Gia tri stapler
7- Line 153: After surgery desire… please revise
‘After surgery, all parameters such as desire, arousal…..’
8- Table 1: Para? Deep endo (ml) please correct abbreviations. Pain: is that preoperative pain score?
Update as follows para (deliveries) deep endo volume (ml) pain score. The legend indicates it is before surgery
9- Table 2: what is ED30? What is the score provided?
Thank you for noticing this mistake: it has been changed to EHP-30
10- The authors suggest that both treatment approaches resulted in improvement, nevertheless, despite some improvement, the total FSFI scores after surgery remained below 26 which suggests persistent sexual dysfunction. The study does not provide evidence as to what the improvement in scores translated to.
Thank you for this question. It is an ongoing debate whether bowel resections for endometriosis should be avoided to prevent sexual dysfunction, which translates into the discussion of conservative excision, discoid excision or bowel resections. As surgeons, sexuality not being our main expertise, we did not focus on sexual function after surgery. This was added in discussion l254 ‘which might explain persisting sexual dysfunction after surgery besides pain’.
11- Table 2: CPP please explain the abbreviations and all headings provided.
Thank you for pointing out that CPP was not defined when mentioned first l 144. This was corrected. We also deleted chronic pelvic pain, leaving the abbreviation l 177.
12- The use of Sum of all pain is meaningless and should be removed. Data summation requires full validation before it could be used in this manner.
Thank you for the comment. You are fully right that strictly speaking, data summation needs validation. However, we suggest leaving the sum of pain.
Dependent and independent variables are important issues in statistics about independent and dependent variables. If 2 variables are strongly (++++) associated they generally carry the same information and in multivariate exploratory models either one or the other but not both variables will reach independently a p.value less than 0.05. Considering the inherent variability of variables the sum of both is probably a better estimate than each variable separately. We, therefore, added in statistics ‘Data summation, such as sumpain, needs validation However if 2 variables are very (+++) strongly associated they generally carry the same information and in multivariate exploratory models either one or the other but not both variables will reach independently a p.value less than 0.05. Considering the inherent variability of variables the sum of these strongly associated variables is probably a better estimate than each variable separately.
We also realise the use of p-values and ‘significance’. Without entering the discussion of the p-value fallacy and the ASA statement, we added to ‘statistics ’ However, we kept the word ‘significant’, despite the definition of <0.05 being arbitrary and despite the p-value fallacy with frequent erroneous conclusions in medicine since traditional frequentist statistics can only refute but cannot confirm a hypothesis.
13- Line 177-180 should be removed. Dyspareunia is an important measure: what does the P<0.0001 refer to??
This has been answered in point 12. We fully recognise the difficulty of explaining p-values and their fallacy, dependent and independent variables and model building. However, a fundamental discussion is beyond this manuscript and probably not useful for the median reader.
14- The data for the EHP30 scores are incompletely/inaccurately presented.
Paragraph 192-202 has been extensively revised as follows ‘Considering the strong correlation of FSFI (p=0.0166) and EHP30 (P=0.0009) before and after surgery, a specific effect of bowel resections on sexuality after surgery needed to be corrected by FSFI and EHP-30 before surgery. In addition, EHP-30, Sumpain and dyschezia after surgery, and other variables such as age and duration of surgery and volume of nodules had to be used as co-variables. By univariate analysis, the FSFI after surgery correlated with FSFI before surgery (P=0.0131), EHP-30 after surgery (P<0.0001) and improvement of EFP-30 (P=0.0002), with the duration of surgery (P=0.0235) and the presence of a second nodule (P<0.0001), and negatively with a hysterectomy (P=0.0056) and age (P=0.0448). By multivariate analysis (proc logistics), the only predictor of FSFI after surgery was the EHP30 after surgery (P<0.0001) or the FSFI before surgery (P<0.0001), without a significant additive effect of the type of surgery of any other variable.
’
15- Line 201: The finding that restricted resection and linear excision have comparable effect on sexual function (and show some improvement) does not eliminate the concerns that arise from the impact of bowel resection on sexual function. The differences may arise from the different starting points and can serve as a warning as to the limits of possible improvement. This is particularly the case as the data presented, although showing some improvement, does not demonstrate a return to normality.
@ questions 15 and 17
Thank you. This has been revised extensively into ‘Pain is a strong factor in decreasing sexual function. Therefore the FSFI evaluates dyspareunia and the EHP 30 the effect of pain. In gynaecology, pain is estimated by the severity and radiation of symptoms such as dyspareunia, dysmenorrhoea, CPP and dyschezia, which are intercorrelated because of the underlying pathology. Sexual functioning is influenced besides pain by psychological factors such as fear of pain decreasing arousal and lubrication and vaginal entry restriction. [1-3]. However, in women with endometriosis, it is today not clear what the relative importance is of pain and of other factors as evaluated in FSFI or EHP-30.’
16- Line 212: please justify the statement or remove it.
References added
17- Line 216: It is recognised that a large study is needed, but the need probably justifies the size and expense. The results of this study does not provide the required clarity of the likelihood of success or of the return to normality or of the likely long term outcomes. All of this suggest the need and that such study is ethically justified.
This is a difficult question about EBM and RCTs and marginally about frequentist statistics. We have discussed this in detail in a recent publication (Koninckx, P.R.; Ussia, A.; Alsuwaidi, S.; Amro, B.; Keckstein, J.; Adamyan, L.; Donnez, J.; Dan, M.C.; Wattiez, A. Reconsidering evidence-based management of endometriosis. Facts, Views and Vision in ObGyn 2022, 14, 225-233)
We fully agree that this paper does not provide a final answer. Therefore l367 was revised as follows ‘Although the data suggest that segmental rectum resection for endometriosis does not impair sexual function more than linear excision, the surgical wisdom of not doing excessive surgery remains fully valid.’
Reviewer 2 Report
This is an interesting paper on the never ending story on the association between bowel endometriosis and patient 's symptoms . Althought not giving a definitive answer it's another piece of information that less is enough (and probably better). Besides minor mistyping which I am sure will be corrected on the revision stage, I think that the research approach is correct for a surgical paper. Unfortunately double blind approaches are impossible in surgery. 120 cases of bowel endometriosis in a 2 year time interval is a correct number of a cases for a good center. The FSFI and the EFI are standardized and universally recognized scoring systems as well as VAS scores. So I cannot see which macroscopic research errors were performed. It is a decent paper that can contribute to the process of better understanding this enigmatic disease and provide surgeons and clinician more information in their consultation with the patients affected by this pathology
Author Response
Thank you very much for your comments, which I do appreciate. I also considered the Sao Paulo series a rather unique opportunity, and this initiated this study performed by the first author, Although not providing the final answer, it adds that a small bowel resection is not as detrimental as many of us tfeared and that in gynaecology the FSFI-6 will be probably as good as the full FSFI scale supplemented with a more detailed registration of pain. This is the updated message in the conclusions
