Journal Pre-proof Change of the residual myometrial thickness during pregnancy after laparoscopic niche resection in comparison to controls without niche surgery; a prospective comparative cohort study

Background: Reduced residual myometrial thickness before and during pregnancy is 52 associated with uterine rupture or dehiscence after vaginal birth after caesarean section. 53 Laparoscopic niche resection performed in case of gynecological symptoms has shown to 54 increase residual myometrial thickness six months after surgery. 55 Objective: To evaluate change in residual myometrial thickness from baseline value before 56 pregnancy to third trimester in women with and without laparoscopic niche resection. 57 Secondary, to evaluate niche presence, niche size during pregnancy compared to before 58 pregnancy, and obstetric outcomes including uterine rupture and dehiscence in both study 59 groups. 60 Study design: Prospective cohort study conducted in an academic medical center. Two 61 groups of pregnant women with a previously diagnosed niche were included: 1) women with 62 a large symptomatic niche (residual myometrial thickness <3mm) followed by laparoscopic 63 niche resection (Lapniche group); 2) women with a niche without niche resection because of 64 minimal symptoms or a residual myometrial thickness ≥3 mm diagnosed prior to current 65 pregnancy (expectant group). Participants underwent a transvaginal ultrasound at 12, 20 66 and 30 weeks of gestation. Changes in residual myometrial thickness, as well as changes of 67 niche measurements over time were analyzed with linear mixed models. 68 Results: 100 women were included, 61 in Lapniche group and 39 in expectant group. In the 69 Lapniche group change in residual myometrial thickness from baseline value before niche 70 resection to third trimester was +2,0mm compared to -1.6mm in the expectant group 71 (p<0.001). Residual myometrial thickness decreased from first trimester onwards in both 72 groups. Although residual myometrial thickness was thinner at baseline in the Lapniche group it remained thicker than in the expectant group during all trimesters; 3.2mm thicker 74 (p<0.001) in the first, 2.5mm (p<0.001) in the second and 1.8mm (p=0.001) in the third 75 trimester. 76 Uterine dehiscence was reported in 1/50 (2%) in the Lapniche group and 7/36 (19%) in 77 expectant group (p=0.007) and was related to the depth-of-niche/residual-myometrial- 78 thickness-ratio pre-pregnancy (after niche resection) and residual myometrial thickness in 79 second trimester. No uterine rupture was reported. The majority of the patients received a 80 scheduled caesarean section in both groups. There was more blood loss during subsequent 81 caesarean section in the Lapniche group. 82 Conclusion: Laparoscopic niche resection results in an increased residual myometrial thickness 83 during subsequent pregnancy. Also, a lower number of dehiscence was found in the Lapniche 84 group than in the expectant group without niche surgery. Per-sectio blood loss was higher in 85 the Lapniche group. In general, laparoscopic niche resection is performed to improve 86 gynecological symptoms. Currently, there is no evidence to support a laparoscopic niche 87 resection to improve obstetric outcomes, but the trend towards more uterine dehiscence 88 encourage further research. 89


Condensation:
The residual myometrium in women with previous laparoscopic niche 27 resection remained thicker during pregnancy than in women without previous niche surgery. 28 29 Short title: Niche evaluation in pregnancy after surgical resection 30 31 AJOG at a Glance: 32 A. Why was this study conducted? 33 To evaluate the change in residual myometrial thickness during pregnancy in women 34 with and without a laparoscopic niche resection. 35 B. What are the key findings? 36 Despite larger niches and thinner residual myometrial thickness at baseline before a 37 laparoscopic niche resection, this procedure resulted in a thicker residual myometrium 38 during the pregnancy which was related to a lower number of dehiscence than in 39 women without niche resection. 40 C. What does this study add to what is already known? 41 No previous study has reported on the changes of residual myometrial thickness during 42 pregnancy and/or obstetric outcomes after previous laparoscopic niche resection. 43 Insight in its behavior during pregnancy and its relation to pre-pregnancy measures may 44 facilitate the development of prediction models on risk of uterine rupture or dehiscence 45 and contribute to future decision making concerning the mode of delivery. Background: Reduced residual myometrial thickness before and during pregnancy is 52 associated with uterine rupture or dehiscence after vaginal birth after caesarean section. 53 Laparoscopic niche resection performed in case of gynecological symptoms has shown to 54 increase residual myometrial thickness six months after surgery. 55 Objective: To evaluate change in residual myometrial thickness from baseline value before 56 pregnancy to third trimester in women with and without laparoscopic niche resection. 57 Secondary, to evaluate niche presence, niche size during pregnancy compared to before 58 pregnancy, and obstetric outcomes including uterine rupture and dehiscence in both study 59 groups. 60 Study design: Prospective cohort study conducted in an academic medical center. Two 61 groups of pregnant women with a previously diagnosed niche were included: 1) women with 62 a large symptomatic niche (residual myometrial thickness <3mm) followed by laparoscopic 63 niche resection (Lapniche group); 2) women with a niche without niche resection because of 64 minimal symptoms or a residual myometrial thickness ≥3mm diagnosed prior to current 65 pregnancy (expectant group). Participants underwent a transvaginal ultrasound at 12, 20 66 and 30 weeks of gestation. Changes in residual myometrial thickness, as well as changes of 67 niche measurements over time were analyzed with linear mixed models. 68 Results: 100 women were included, 61 in Lapniche group and 39 in expectant group. In the 69 Lapniche group change in residual myometrial thickness from baseline value before niche 70 resection to third trimester was +2,0mm compared to -1.6mm in the expectant group 71 (p<0.001). Residual myometrial thickness decreased from first trimester onwards in both 72 groups. Although residual myometrial thickness was thinner at baseline in the Lapniche 73 J o u r n a l P r e -p r o o f group it remained thicker than in the expectant group during all trimesters; 3.2mm thicker 74 (p<0.001) in the first, 2.5mm (p<0.001) in the second and 1.8mm (p=0.001) in the third 75 trimester. 76 Uterine dehiscence was reported in 1/50 (2%) in the Lapniche group and 7/36 (19%) in 77 expectant group (p=0.007) and was related to the depth-of-niche/residual-myometrial-78 thickness-ratio pre-pregnancy (after niche resection) and residual myometrial thickness in 79 second trimester. No uterine rupture was reported. The majority of the patients received a 80 scheduled caesarean section in both groups. There was more blood loss during subsequent 81 caesarean section in the Lapniche group.

82
Conclusion: Laparoscopic niche resection results in an increased residual myometrial thickness 83 during subsequent pregnancy. Also, a lower number of dehiscence was found in the Lapniche 84 group than in the expectant group without niche surgery. Per-sectio blood loss was higher in 85 the Lapniche group. In general, laparoscopic niche resection is performed to improve 86 gynecological symptoms. Currently, there is no evidence to support a laparoscopic niche 87 resection to improve obstetric outcomes, but the trend towards more uterine dehiscence 88 encourage further research.

94
A worldwide increase of the caesarean section (CS) rate has led to a rising number of 95 patients with complications after a CS. A niche, defined as a defect at the site of the uterine 96 scar with a depth of at least 2mm, is visible with sonohysterography (with gel or saline 97 infusion) in 60-70% of women after CS. 1-3 A large niche, defined as a niche with a residual 98 myometrial thickness <3mm or with a depth of >50% of the myometrial thickness is reported 99 in approximately 25% of all women after a CS. 3,4 A niche is frequently described to be 100 associated with long-term symptoms, including gynecological symptoms (abnormal uterine 101 bleeding, dysmenorrhea), fertility problems, and obstetric complications such as uterine 102 dehiscence or rupture in subsequent pregnancy. 2,5-7 The effectiveness of uterine repair by 103 performance of a laparoscopic niche resection (LNR) has been reported for gynecological 104 symptoms, but less is known about the effect of obstetric outcomes. 8-10 105 Residual myometrial thickness or thickness of the lower uterine segment (LUS) may be 106 predictive for the risk of uterine rupture. Although no exact cut-off point is determined yet, a 107 residual myometrial thickness <3mm before and during subsequent pregnancy was 108 associated with uterine rupture or dehiscence after vaginal birth after CS (VBAC). 11,12 Naji et 109 al. 13 demonstrated that residual myometrial thickness, measured transvaginally in pregnant 110 women with a niche, decreases as the pregnancy progresses and that niche width increases. 111 The influence of LNR prior to pregnancy on the thickness and changes of the residual 112 myometrium in the subsequent pregnancy and its associated risk of uterine rupture or 113 dehiscence is unknown. Therefore we conducted a prospective cohort study including 114 pregnant women with a niche diagnosed before their current pregnancy. The aim of the 115 study was to compare change in residual myometrial thickness from baseline before 116 J o u r n a l P r e -p r o o f 7 pregnancy (before LNR in the Lapniche group) to third trimester in women with and without 117 previous LNR. Also, we evaluated changes in niche size in pregnancy compared to baseline, 118 and related them to the obstetric outcomes and the occurrence of uterine dehiscence.

312
To our knowledge, no previous study has reported on uterine CS scar evaluation during 313 pregnancy and related these findings to obstetric outcomes after a LNR. Few studies 18-20 314 reported on obstetric outcomes after LNR, but no sonographic measurements were 315 performed during subsequent pregnancy. 316 There are previous studies that evaluated the relation between residual myometrial 317 thickness, thickness of the lower uterine segment (LUS) in women with a previous CS. In line 318 with our findings, a gradual decrease in residual myometrial thickness and thickness of the 319 LUS during pregnancy in women with a previous CS was reported. 13,21-24 One previous study 320 evaluated thickness of the LUS in the third trimester and found a correlation with uterine 321 rupture or dehiscence. 25 The prognostic value of D/RMT-ratio just before pregnancy on uterine dehiscence in women 323 with a previous CS was previously described by Pomorski et al. 17 They reported a significantly 324 higher mean D/RMT-ratio before pregnancy in women with uterine dehiscence (n=7) than in 325 women without dehiscence (n=34); mean D/RMT-ratio was 1.4 (SD ±0.39) and 0.36 (SD 326 ±0.07), respectively. Our results are in line with these findings but the number of women 327 with a dehiscence and registered D/RMT-ratio in our study was too small to draw any strong 328 conclusion concerning its value. Mean D/RMT-ratio increased significantly after LNR. No 329 other studies evaluated this value after LNR in comparison to controls. 330 Higher blood loss during subsequent CS after previous LNR was not earlier reported. In one 331 case in our study blood loss >1000mL was due to abnormal adhesive placentation and in 332 another case due to placenta previa. However, the relation between LNR and placental 333 problems is unknown. Currently, it is more likely that presence of a uterine cesarean scar 334 was related to the placental problems in these cases based on available literature. 26

336
Clinical implications 337 Insight in the behavior of the uterine CS scar during pregnancy and its relation to pre-338 pregnancy measures may facilitate the development of prediction models on risk of uterine 339 rupture or dehiscence and may contribute to future decision making concerning the mode of 340 delivery. In this study, we showed that residual myometrial thickness during pregnancy 341 decreases during the gestation. Cut-off values of residual myometrial thickness or thickness 342 of the LUS during pregnancy to predict a successful vaginal birth after CS or uterine 343 rupture/dehiscence still need to be determined in future research. However, despite the fact 344 that most women received a scheduled CS, we found a significantly higher prevalence of 345 uterine dehiscence during CS in women without LNR. Mean residual myometrial thickness was <3mm during entire pregnancy in women with a dehiscence during the CS. Residual 347 myometrial thickness in second trimester seemed to be the most discriminating value for a 348 dehiscence at term. 11,12 349 LNR has a positive effect on the residual myometrial thickness before and during the entire 350 pregnancy and may decrease the prevalence of uterine dehiscence, although this study was 351 not powered to evaluate the latter outcome. This lower prevalence of dehiscence must be 352 weighed against higher blood loss during repeat CS.

361
This is the first study describing the effect of LNR on residual myometrial thickness and niche 362 measurements during subsequent pregnancy. Another strength of this study is the long term 363 follow-up after LNR; besides the measurements during subsequent pregnancy these 364 outcomes were also related to pregnancy outcomes. Residual myometrial thickness 365 measurements before and during pregnancy provide a clear insight of changes of the CS scar 366 features over time. In this study we deliberately chose to evaluate niche changes during 367 pregnancy in relation to baseline before surgery and not after surgery, since our aim was to 368 study the effect of surgery on the RMT and uterine CS features during pregnancy and 369 therefore baseline data should be similar in both groups. Follow-up rates were high; US measurements were reported in 84-97% at the various moments and obstetric outcomes 371 were available in 86% of the included women. Another strength is that measurements were 372 performed in a structured way by experienced sonographers. 373 Additionally, we compared the outcomes after LNR to a control group of women with 374 expectant management. However due to the non-randomized design selection bias was    0.123 Apgar score After 1 minute 9 (8-9) 9 (9-9) 0.431 After 5 minutes 10 (9-10) 10 (9-10) 0.385 Admission to NICU 13 (26%) 9 (25%) 0.830 Data are reported as mean ± standard deviation, median (interquartile range) or n (valid %).

Group
No.

Number of previous CSs
Residual myometrial thickness prepregnancy* (in mm) Appendix S1 Median time between pre-pregnancy uterotomy* and onset of subsequent 542 pregnancy in both study groups 543