Results
Demographic characteristics in this study
Following the inclusion and exclusion criteria, 727 women were included
in this study, and 85 women were excluded as shown in Fig. S1. The
prevalence of niche at CS scar was about 36.2% in our hospital during
2012.01- 2017.06 by TVS evaluation. In the analysis of clinical
symptoms, prolonged postmenstrual spotting was the most common symptom
(29.6%, 215/727), and the incidence was the highest (75.5%, 123/163)
in the large niche group as shown in Table S3. The age of all
participants ranged from 22 to 44 years, with a mean age of 29.66 ± 3.84
years, and gestational age ranged from 30 to 42 weeks at CS, with a mean
value of 38.77 ± 1.86 weeks. Of these women, 35 had once previous
vaginal delivery, 117 had undergone once CS history, and 301 had
abortion history before the last CS (ranged from 1 to 9 times).
Large niche definition
Table 1 showed the average menstruation days, endometrium thickness, and
niche parameters including depth, length, width, RMT, AMT, and depth/AMT
in two groups. The mean menstruation days of asymptomatic women with
niche was about 6.09 ± 1.10, and 9.51 ± 2.11 for the symptomatic women
with niche (p < 0.05). There was no significant
difference between the two groups in endometrium thickness on the
examination day. The mean values of depth, length, and width of niches
were significantly greater in the
symptomatic group than those of the asymptomatic group (p< 0.05). The mean values of RMT and AMT of the symptomatic
group were significantly lower than those of the asymptomatic group
(p < 0.001). The ROC curves of these variables were
shown in Fig.2 and Table S2, and the largest AUC was considered to be
the best predictor of a large niche. According to the AUC statistics,
the cut-off for depth was 0.50 cm (AUC: 0.731, 95% CI: 0.661-0.790),
RMT was 0.21 cm (AUC: 0.683, 95% CI: 0.614-0.747), and Depth/AMT ratio
was 0.56 (AUC: 0.798, 95% CI: 0.725-0.804). Therefore, we define the
large niche more than 0.50 cm in depth, or less than 0.21 cm in RMT, or
more than 0.56 in depth/AMT. This definition had 61.17% (95% CI:
0.52-0.70) specificity, 76.87% (95% CI: 0.70-0.84) sensitivity and
70.72% (95% CI: 0.65-0.76) accuracy.
Risk factors related to a large niche formation
According to the definition for the large niche mentioned above, 163
women were classified in a large niche group, 100 women were classified
in a small niche group, and 464 women were classified in the control
group. Table S3 showed the candidate predictive variables compared among
the three groups by univariate analyses. There was no significant
difference in gestational age at delivery, times of abortions, and
vaginal delivery history among different groups. The differences of age
at delivery, one CS history, bilateral tubal ligation, B-Lynch suture,
operating duration, emergent CS, MSAF, PROM, cervical dilatation more
than 4 cm, retroflexed uterine, breech, oxytocin augmentation, presence
or duration of labor at CS, anemia, and postpartum hemorrhage were
significant between control and large niche group (p <
0.05 ). However, we didn’t find any relevance between large niche and
obstetric complications, such as pre-eclampsia, diabetes, ICP, and
placenta previa. Moreover, our results showed that twin pregnancy,
macrosomia, surgeon experience, and ART didn’t influence the risk of
large niche development.
Multivariate logistic model and
assigned scores
The model predicting large niche development included the following
eight variables: age at delivery (0 for < 35 years old; 1.0
for ≥ 35 years old), history of once CS ( 1.0 for yes), operation
duration (0 for < 120 min; 1.0 for ≥ 120 min), B-Lynch suture
( 2.0 for yes), MSAF (2.0 for yes), PROM (1.0 for yes), cervical
dilatation more than 4 cm (2.0 for yes), and retroflexed uterine ( 4.0
for yes) (Table 2).
Effectiveness of the score-based prediction model of a large niche
Table 3 showed the discriminative performances of each score as the
cut-off value in identifying individuals at high-risk of large niche
formation in our study. As the cut-off value increased, the risk of
large niche formation increased. We comprehensively estimated some
predominant indices for each score cut-off value in the score-based
model, including the number and proportion of high-risk individuals,
sensitivity, specificity, Youden’s index (sensitivity + specificity –
1). The candidate cut-off value of 5 from the score-based model was
selected as the criteria for identifying high-risk individuals for large
niche formation after CS. If a cut-off value of 5 was applied to select
individuals in this study, 166 individuals were selected as the
high-risk populations. In this study, 110 individuals have been
identified to have large niches by TVS, and 107 individuals accompanied
by dot bleeding symptoms. The sensitivity and specificity were 67.48%
and 90.07%, respectively. The AUC
of this score-based model was
0.875 (95% CI: 0.848–0.902; Fig. 3).
Discussion
The main results of this retrospective study proposed a large niche
definition based on the largest sample size so far, and it showed good
sensitivity and specificity. Besides, a score-based individual risk
prediction model was developed according to the multivariate analysis
results of twelve risk factors and was used to quantitatively evaluate
the risk of a large niche formation for women undergoing CS. This model
included eight variables and showed the good discriminative ability to
predict high-risk individuals. More importantly, this model provided
detailed information for the doctors to develop preventive strategies
for high-risk individuals during or after CS.
So far, only three studies focused on the large niche, and one of them
gave the definition of a large niche, and others proposed the risk
factors related to the large niche formation24-26. Our
definition was similar to the previous one16, however,
we proposed more detailed values based on a large sample size. Large
niches were uncommon with a reported varying incidence of
11-45%27 depending on the definition used. The
prevalence of large niche in our hospital during 2012-2017 was about
22.4% using our diagnostic criteria. It appeared that there was a
correlation between the size of the defect and postmenstrual
spotting4,15,27. Studies demonstrated that when using
TVS in a group of women with gynecological symptoms, half of them had a
large niche, involving more than 50% of the myometrial
thickness7. Clinically, we often found abundant newly
formed fragile vessels in the large niches under hysteroscopy as show in
Fig.1D. Therefore, postmenstrual spotting was the most common symptom in
large niches4,27 and was used as a term to distinguish
the large niches from small ones in our study.
It should be emphasized that TVS examination was done during the
mid-follicular phase of the examinees, owing to a clear visualization of
the niches in case of the fluid in the uteri
cavity28,29. The size of the niche was affected by the
thickness of endometrium30, however, we didn’t find
significant difference among the participants. As we know, depth, RMT,
and AMT values can be easily obtained from sagittal
plane31, however, the values of length and width often
varying with the pressure of the uterine cavity in case of intracavitary
fluid, which was not included in our definition. Besides, the AUC
statistics of the depth, RMT, and depth/AMT were the top three among all
the parameters of the niche. Therefore, the definition of a large niche
was the parameters of a niche meet anyone of the three items.
In this study, several well-known risk
factors5,15,23,32 were included in the final model,
including age at delivery, duration of CS, cervical dilation more than 4
cm, CS history, and uterine retroflection. The other potential risk
factors of niche6, emergent CS, the presence of labor,
anemia, oxytocin augmentation during labor, breech excluded from
multivariable logistic analysis (Table S4). These findings might suggest
that the associations between these factors and the risk of a large
niche might be weaker than the association with other robust risk
factors in our study. Some researchers reported that niche may almost
happen in the women with more than three times
CS11,18, and most women in our country are more likely
to have twice CS during their lifetime. So, we only include once history
of CS in the model. The surgical technique of uterine incision closure
is the most important determinant of the CS defect
formation33,34. However, the method of wound closure
could not be analyzed because the standard way of uterine wound closure
at our hospital was continuously locked sutures during the 2012-2017
year. Moreover, double-layer uterine closure was recommended by several
important studies33,35,36 and was accepted by the most
obstetricians in China. So, this factor was not recruited in the model.
Interestingly, B-Lynch suture and MSAF were the first time verified to
associate with a large niche. Notably, it is not possible to evaluate
the position of uterus immediately during or after the CS, but some
studies considered that the formation of niche is one year after the
operation. Therefore, scoring the position of uterus three months after
delivery does not affect the prediction of a large niche.
Compared to the conclusions of previous studies or the meta-analyses,
our model might have some merits. First, from the aspect of the study
design and data resources used in the model development, our study was
the first large size sample-based cohort study to develop a prediction
model of a large niche, thus rendering the model more convincible.
Second, from the aspect of selection of predictors, we chose all the
well-known predictors according to high-quality literature reviews,
meta-analyses, and the latest Chinese expert consensus. Besides, some
unwell-known variables such as B-Lynch suture and MSAF based on our
clinic experience were also included in the model. Therefore,
we estimated the model accuracy
and practice value as a prediction method in the large niche formation.
The score of each predictor in this model represented different degrees
of impact on large niche formation, which provided important information
for the doctors to develop individualized prevention strategies. In the
future, this scoring model for large niche formation will allow a large
size of prospective studies in the clinical practice.
Strength and Limitations
This study has several strengths. First, the sample size and full-scaled
risk factors were relatively large and detailed to date. Second, the
measurement of niche and the time of examination was consistent with the
guideline in practice in Europe. Third, the definition of a large niche
was practicable and low-cost. Finally, the scoring model for the
prediction of niche formation was designed for the individual after CS.
Our study has some limitations. First, we lost a large number of women
who refused to interview the uterine scar measurement. The number of the
participants was still insufficient, so the results for design the
scoring model was limited. Second, the niche parameters in this study
are not full-scale, like the distance between the niche and the
vesico-vaginal fold, and distance between the niche and the external os
were excluded in the data collection. Third, the loss of the follow-up
of the different treatment protocols for the women with large niche and
dot bleeding symptoms, and the relationship of niche parameters with the
next pregnant result. Finally, the best suture method recommended at
present is the first-layer suture avoiding the decidua followed by a
second layer for the approximation of the myometrium. However, the
suture method of CS in our hospital was continuous single-layer uterine
closure using locking sutures, and closure of the peritoneum during
2012-2017. At present, the proposed cut-off value in the large niche
scoring model is still arbitrary, and it should be verified furtherly in
the clinic to make sure its meaning of guidance for treatment opinions.
Because the uterus position can’t be determined during or after CS, and
the uterus position sometimes may change after CS. Therefore, the uterus
position score may be determined by TVS during the first three months of
CS.
Conclusion
Further studies should focus on investigating the definition of large
niche in relation to severe long-term complications after CS, such as
obvious dot bleeding, scar pregnancy and late pregnancy uterine rupture.
In addition, the score-based prediction system of large niche should be
validated in prospective cohort study.
Acknowledgments
We thank the women that participated in the study, and our graduate
student Qian Zhang, Jialiu He for their help with statistical analysis.
Disclosure of interest
There are no conflicts of interest to declare.
Contribution to authorship
Jing Wang planned this study and wrote the manuscript. Qiushi Pang and
Shijie Yan took an active role in collecting medical history and
contacting participants. Most of the transvaginal ultrasound
examinations were done by Wenwen Wei. Mingjun Hu and Fen Huang analyzed
all the data. Linghui Cheng, Yunxia Cao and Zhaolian Wei
planned, and modified the
manuscript.
Details of ethics approval
The study was approved by the Ethics Committee of the First Affiliated
Hospital of Anhui Medical University (PJ-2019-03-12).
Funding
This study is funded by Yunxia Cao, National Key Research and
Development Program. (project number 2016YFC1000204)
The funding sources had no involvement in the study design; in the
collection, analysis and interpretation of data; in the writing of the
report; or in the decision to submit the article for publication.
Reference
1. Bij de Vaate AJ, van der Voet LF, Naji O, Witmer M, Veersema S,
Brölmann HA, et al. Prevalence, potential risk factors for development
and symptoms related to the presence of uterine niches following
Cesarean section: systematic review. Ultrasound Obstet Gynecol.
2014;43(4):372-382.
2. Betran AP, Torloni MR, Zhang JJ, Gulmezoglu AM, Section WHOWGoC. WHO
Statement on Caesarean Section Rates. BJOG. 2016;123(5):667-670.
3. Kang L, Ye S, Jing K, Fan Y, Chen Q, Zhang N, et al. A Segmented
Logistic Regression Approach to Evaluating Change in Caesarean Section
Rate with Reform of Birth Planning Policy in Two Regions in China from
2012 to 2016. Risk Manag Healthc Policy. 2020;13:245-253.
4. van der Voet LF, Bij de Vaate AM, Veersema S, Brolmann HA, Huirne JA.
Long-term complications of caesarean section. The niche in the scar: a
prospective cohort study on niche prevalence and its relation to
abnormal uterine bleeding. BJOG. 2014;121(2):236-244.
5. Kamel R, Kaelin Agten A, Noel L, Eissa T, Sharaf M, Negm S, et al.
Position and integrity of the uterine scar is determined by cervical
dilation at the time of Caesarean section. Ultrasound Obstet Gynecol.
2020.
6. Vervoort AJ, Uittenbogaard LB, Hehenkamp WJ, Brolmann HA, Mol BW,
Huirne JA. Why do niches develop in Caesarean uterine scars? Hypotheses
on the aetiology of niche development. Hum Reprod.
2015;30(12):2695-2702.
7. Tulandi T, Cohen A. Emerging Manifestations of Cesarean Scar Defect
in Reproductive-aged Women. J Minim Invasive Gynecol.
2016;23(6):893-902.
8. Rasheedy R, Sammour H, Elkholy A, Fadel E. Agreement between
transvaginal ultrasound and saline contrast sonohysterography in
evaluation of cesarean scar defect. J Gynecol Obstet Hum Reprod.
2019;48(10):827-831.
9. Du Q, Liu G, Zhao W. A novel method for typing of cesarean scar
pregnancy based on size of cesarean scar diverticulum and its
significance in clinical decision-making. J Obstet Gynaecol Res.
2020;46(5):707-714.
10. Sholapurkar SL. Etiology of Cesarean Uterine Scar Defect (Niche):
Detailed Critical Analysis of Hypotheses and Prevention Strategies and
Peritoneal Closure Debate. J Clin Med Res. 2018;10(3):166-173.
11. Ofili-Yebovi D, Ben-Nagi J, Sawyer E, Yazbek J, Lee C, Gonzalez J,
et al. Deficient lower-segment Cesarean section scars: prevalence and
risk factors. Ultrasound Obstet Gynecol. 2008;31(1):72-77.
12. Osser OV, Jokubkiene L, Valentin L. Cesarean section scar defects:
agreement between transvaginal sonographic findings with and without
saline contrast enhancement. Ultrasound Obstet Gynecol.
2010;35(1):75-83.
13. Regnard C, Nosbusch M, Fellemans C, Benali N, van Rysselberghe M,
Barlow P, et al. Cesarean section scar evaluation by saline contrast
sonohysterography. Ultrasound Obstet Gynecol. 2004;23(3):289-292.
14. Pomorski M, Fuchs T, Rosner-Tenerowicz A, Zimmer M. Sonographic
evaluation of surgical repair of uterine cesarean scar defects. J Clin
Ultrasound. 2017;45(8):455-460.
15. Antila-Langsjo RM, Maenpaa JU, Huhtala HS, Tomas EI, Staff SM.
Cesarean scar defect: a prospective study on risk factors. Am J Obstet
Gynecol. 2018;219(5):458 e1-458 e8.
16. Vikhareva Osser O, Valentin L. Risk factors for incomplete healing
of the uterine incision after caesarean section. BJOG.
2010;117(9):1119-1126.
17. Stegwee SI, Jordans I, van der Voet LF, van de Ven PM, Ket J,
Lambalk CB, et al. Uterine caesarean closure techniques affect
ultrasound findings and maternal outcomes: a systematic review and
meta-analysis. BJOG. 2018;125(9):1097-1108.
18. Osser OV, Jokubkiene L, Valentin L. High prevalence of defects in
Cesarean section scars at transvaginal ultrasound examination.
Ultrasound Obstet Gynecol. 2009;34(1):90-97.
19. Wang CB, Chiu WW, Lee CY, Sun YL, Lin YH, Tseng CJ. Cesarean scar
defect: correlation between Cesarean section number, defect size,
clinical symptoms and uterine position. Ultrasound Obstet Gynecol.
2009;34(1):85-89.
20. van der Voet LF, Jordans IPM, Brolmann HAM, Veersema S, Huirne JAF.
Changes in the Uterine Scar during the First Year after a Caesarean
Section: A Prospective Longitudinal Study. Gynecol Obstet Invest.
2018;83(2):164-170.
21. Vervoort A, Vissers J, Hehenkamp W, Brolmann H, Huirne J. The effect
of laparoscopic resection of large niches in the uterine caesarean scar
on symptoms, ultrasound findings and quality of life: a prospective
cohort study. BJOG. 2018;125(3):317-325.
22. Naji O, Abdallah Y, Bij De Vaate AJ, Smith A, Pexsters A, Stalder C,
et al. Standardized approach for imaging and measuring Cesarean section
scars using ultrasonography. Ultrasound Obstet Gynecol.
2012;39(3):252-259.
23. Pomorski M, Fuchs T, Rosner-Tenerowicz A, Zimmer M. Standardized
ultrasonographic approach for the assessment of risk factors of
incomplete healing of the cesarean section scar in the uterus. Eur J
Obstet Gynecol Reprod Biol. 2016;205:141-145.
24. Vervoort A, van der Voet LF, Hehenkamp W, Thurkow AL, van Kesteren
P, Quartero H, et al. Hysteroscopic resection of a uterine caesarean
scar defect (niche) in women with postmenstrual spotting: a randomised
controlled trial. BJOG. 2018;125(3):326-334.
25. Vikhareva O, Rickle GS, Lavesson T, Nedopekina E, Brandell K,
Salvesen KA. Hysterotomy level at Cesarean section and occurrence of
large scar defects: a randomized single-blind trial. Ultrasound Obstet
Gynecol. 2019;53(4):438-442.
26. Baranov A, Salvesen KA, Vikhareva O. Assessment of Cesarean
hysterotomy scar before pregnancy and at 11-14 weeks of gestation: a
prospective cohort study. Ultrasound Obstet Gynecol. 2017;50(1):105-109.
27. Bij de Vaate AJ, Brolmann HA, van der Voet LF, van der Slikke JW,
Veersema S, Huirne JA. Ultrasound evaluation of the Cesarean scar:
relation between a niche and postmenstrual spotting. Ultrasound Obstet
Gynecol. 2011;37(1):93-99.
28. Jordans IPM, de Leeuw RA, Stegwee SI, Amso NN, Barri Soldevila PN,
van den Bosch T, et al. Sonographic examination of uterine niche in
non-pregnant women: a modified Delphi procedure. Ultrasound Obstet
Gynecol. 2019;53(1):107-115.
29. Jordans IPM, de Leeuw RL, Stegwee SI, Amso NN, Barri Soldevila PN,
van den Bosch T, et al. Niche definition and guidance for detailed niche
evaluation. Acta Obstet Gynecol Scand. 2019;98(10):1351-1352.
30. Seshadri S, El-Toukhy T, Douiri A, Jayaprakasan K, Khalaf Y.
Diagnostic accuracy of saline infusion sonography in the evaluation of
uterine cavity abnormalities prior to assisted reproductive techniques:
a systematic review and meta-analyses. Hum Reprod Update.
2015;21(2):262-274.
31. Marjolein Bij de Vaate AJ, Linskens IH, van der Voet LF, Twisk JW,
Brolmann HA, Huirne JA. Reproducibility of three-dimensional ultrasound
for the measurement of a niche in a caesarean scar and assessment of its
shape. Eur J Obstet Gynecol Reprod Biol. 2015;188:39-44.
32. Pomorski M, Fuchs T, Rosner-Tenerowicz A, Zimmer M. Morphology of
the cesarean section scar in the non-pregnant uterus after one elective
cesarean section. Ginekol Pol. 2017;88(4):174-179.
33. Roberge S, Demers S, Girard M, Vikhareva O, Markey S, Chaillet N, et
al. Impact of uterine closure on residual myometrial thickness after
cesarean: a randomized controlled trial. Am J Obstet Gynecol.
2016;214(4):507 e1-507 e6.
34. Stegwee SI, Jordans IPM, van der Voet LF, Bongers MY, de Groot C,
Lambalk CB, et al. Single- versus double-layer closure of the caesarean
(uterine) scar in the prevention of gynaecological symptoms in relation
to niche development - the 2Close study: a multicentre randomised
controlled trial. BMC Pregnancy Childbirth. 2019;19(1):85.
35. Di Spiezio Sardo A, Saccone G, McCurdy R, Bujold E, Bifulco G,
Berghella V. Risk of Cesarean scar defect following single- vs
double-layer uterine closure: systematic review and meta-analysis of
randomized controlled trials. Ultrasound Obstet Gynecol.
2017;50(5):578-583.
36. Vachon-Marceau C, Demers S, Bujold E, Roberge S, Gauthier RJ,
Pasquier JC, et al. Single versus double-layer uterine closure at
cesarean: impact on lower uterine segment thickness at next pregnancy.
Am J Obstet Gynecol. 2017;217(1):65 e61-5 e5.
Table 1 Comparison of niche parameters