In recent years, the rate of deliveries by cesarean has increased all over the world.1–4 Cesarean delivery, in particular when repeated, is associated with an increased risk of obstetric complications in subsequent pregnancies.5–7 One of the most serious complications after cesarean delivery is uterine rupture. This complication is very rare but associated with high rates of perinatal morbidity and mortality.8–11 Another consequence of cesarean delivery is uterine scar dehiscence, which is often detected incidentally at a repeat cesarean delivery. Uterine scar dehiscence may be a risk factor for uterine rupture. It would be clinically important to identify risk factors for uterine dehiscence and rupture to facilitate early diagnosis and prevention of this complication.
In several ultrasound studies, defects in the hysterotomy scar after cesarean delivery have been shown to be common in nonpregnant women.12–15 The clinical importance of large scar defects diagnosed by ultrasonography in nonpregnant women is not yet known, but large scar defects are likely to reflect incomplete healing of the scar.
The objective of our study was to estimate the association between the appearance of cesarean hysterotomy scars at transvaginal ultrasound examination of nonpregnant women and the outcome of subsequent pregnancies and deliveries. Our null hypothesis is that there is no association between large cesarean hysterotomy scar defects detected by transvaginal ultrasonography in nonpregnant women and uterine dehiscence or rupture in subsequent pregnancies.
MATERIALS AND METHODS
In 2005 and 2006 consecutive women who had been delivered by cesarean in our institution were invited to undergo an ultrasound examination of their hysterotomy scar with the aim of estimating the prevalence and appearance of cesarean hysterotomy scar defects.15,16 The volunteers of the current study are all the 162 women who accepted our invitation and fulfilled our inclusion criteria.15,16 All 162 women had undergone one or more cesarean deliveries, and their hysterotomy scars had been examined with transvaginal ultrasonography in 2005 or 2006 within the frame of our research project.15,16 The examination technique has been described in detail in our previous publications15,16 and is therefore only briefly outlined below. The 162 women were followed until January 2010 with regard to the outcome of subsequent pregnancies by scrutiny of their medical records.
A single examiner (O.V.) scanned all 162 women with transvaginal ultrasonography 6 to 9 months after the cesarean delivery and also interpreted the images: 54 women were examined only with unenhanced ultrasonography and 108 were also examined with saline contrast sonohysterography (instillation of saline into the uterine cavity during scanning), 46 having saline contrast sonohysterography performed on the same day as unenhanced ultrasonography and 62 having saline contrast sonohysterography performed on a later date.16 The ultrasound examiner was blinded to the obstetric history until all scans had been evaluated.
The uterus was scanned in a longitudinal plane to identify the hysterotomy scar and any defect in the scar. A hypoechoic or hyperechoic line in the anterior wall of the uterus was defined as a hysterotomy scar resulting from cesarean delivery. Any indentation in the scar—however small—was classified as a scar defect. The size of the hysterotomy scar defects was measured objectively. Measurements were made offline on a frozen ultrasound image immediately after the ultrasound examination, using calipers. On a longitudinal section through the uterus, measurements were taken of the thickness of the remaining myometrium over the defect, the measurements being taken on that section where the scar defect appeared to be at its largest. (Fig. 1).
When studying the association between the ultrasound appearance of the hysterotomy scar and the outcome of subsequent pregnancies, we used the definition of a large scar defect that we had determined on the basis of the results of our previous studies15,16: a scar defect was defined as large at saline contrast sonohysterography if the thickness of the remaining myometrium over the defect was no larger than 2.5 mm in women who had undergone one cesarean delivery, and no larger than 2.3 mm in women who had undergone two or more cesarean deliveries.16 Findings at unenhanced ultrasound examination were used only in those women who had not undergone saline contrast sonohysterography: a scar defect was defined as large at unenhanced ultrasound examination if the thickness of the remaining myometrium over the defect was no larger than 2.2 mm in women who had undergone one cesarean delivery, and no larger than 1.9 mm in women who had undergone two or more cesarean deliveries.15
The medical staff involved in the clinical management of the women in their subsequent pregnancies had no knowledge of the results of the ultrasound examinations of the cesarean scars. The volunteers were not told about the ultrasound measurements. They were clearly informed that the clinical importance of our ultrasound findings was unknown and would not be made available to any clinical staff.
The results of the ultrasound examinations were compared with the outcome of the pregnancies conceived after the ultrasound examination. By manual scrutiny of medical records the following complications were specifically sought: scar pregnancy (as determined by transvaginal ultrasonography before 14 completed gestational weeks), abnormal placental implantation (placenta previa; placenta accreta, increta, percreta), extremely thin myometrium, uterine scar dehiscence, and uterine rupture. The diagnoses of extremely thin myometrium, uterine scar dehiscence, and uterine rupture were established at cesarean delivery in subsequent pregnancies. Women who gave birth vaginally were not routinely examined by exploration of the uterine cavity to detect uterine scar dehiscence or rupture. In the group of women with extremely thin myometrium at cesarean delivery, we included cases where the myometrium was described as “thin as a leaf,” “membranous,” or “transparent.” Uterine scar dehiscence was defined as a subperitoneal separation of the uterine scar in the lower uterine segment with the chorionamniotic membrane visible through the peritoneum. Uterine rupture was defined as a complete separation of the uterine scar with communication between the uterine and abdominal cavities. If no comment was made in the operation report on the appearance of the lower uterine segment we assumed that there was neither an extremely thin myometrium nor dehiscence nor rupture. In cases where the operation report was unclear or difficult to interpret, the obstetrician who had performed the cesarean delivery was personally contacted and asked about details. The person retrieving information from the records and from the obstetricians performing the cesarean deliveries was unaware of the results of the ultrasound examinations.
In addition to monitoring the 162 women included in our study, we collected information by manual scrutiny of the patient records of all women with repeat cesarean delivery at our institution between 2005 and 2009 on the rate of extremely thin myometrium, uterine rupture, uterine dehiscence, placenta previa, and placenta accreta.
All clinical information was retrieved from our digitalized patient record system. Statistical calculations were performed using SPSS 12.2 or 16.0 or StatXact 9. The Fisher exact test was used to calculate the statistical significance of differences in proportions and to determine the statistical significance of a possible relationship between a large scar defect and uterine rupture or dehiscence. Odds ratios were calculated using exact univariable logistic regression.17 In these analyses only the first delivery after the ultrasound examination of the hysterotomy scar was included. A two-tailed P<.05 was considered statistically significant.
The study protocol was approved by the ethics committee of the Medical Faculty of Lund University, Sweden (Dnr 859/2004, approved December 22, 2004; Dnr 198/2006, approved May 8, 2006), and informed consent was obtained from all participants after the nature of the procedures had been fully explained.
The follow-up time for the 162 women varied between 3 years 1 month and 4 years 6 months. Of the 162 women, 6 were lost to follow-up because they moved from the area and we found no means of contacting them. Sixty-nine (44%) of the remaining 156 women became pregnant after the ultrasound examination and constitute our current study population. The characteristics of the women who did and did not become pregnant are shown in Table 1: the women who became pregnant were younger and of lower parity, and a higher proportion (93% compared with 48%) had undergone only one cesarean delivery.
Among the 69 women who became pregnant after the ultrasound examination, there were 99 pregnancies and 65 deliveries: 53 women gave birth once, and six women gave birth twice, whereas 10 women experienced miscarriage, termination of pregnancy, or ectopic pregnancy (the number of “failed pregnancies” per each of the 10 women varied between one and five). Transvaginal ultrasonography was carried out before 14 completed gestational weeks in 80 (81%) of the 99 pregnancies. No scar pregnancy was found. The clinical outcome of the 19 pregnancies not examined with ultrasonography before 14 weeks suggested that there were also no scar pregnancies among them. There were no placental complications (placenta previa; placenta accreta, increta, percreta) in any of the pregnancies.
Clinical information and the outcome of the first delivery after the ultrasound examination of the hysterotomy scar for the women who gave birth are shown in Table 2. Fifty-six (95%) women were delivered at term (37 or more completed gestation weeks) and three (5%) preterm (from 35 4/7 weeks to 36 5/7 weeks). The 5-minute Apgar score was 7 or higher in all newborns. Only one newborn had a low umbilical cord blood pH (arterial pH=6.91) but the Apgar score of this newborn was 9–10–10 at 1, 5, and 10 minutes.
Seventy-five percent (9/12, 95% confidence interval [CI] 43–95%) of the women with an intact cesarean scar gave birth vaginally compared with 50% (17/34, 95% CI 32–68%) of those with a small scar defect and 46% (6/13, 95% CI 19–75%) of those with a large scar defect (P=.29). Elective cesarean delivery (“maternal request” or “previous cesarean delivery”) was performed in 17% (2/12, 95% CI 2–48%), 24% (8/34, 95% CI 11–41%), and 23% (3/13, 95% CI 5–54%) of the women, respectively (P=1.00), and emergency cesarean delivery in 8% (1/12, 95% CI 0.2 −38%), 24% (8/34, 95% CI 11–41%), and 31% (4/13, 95% CI 9–61%), respectively (P=.43). Ten women (38%, 10/26) were in active labor at the cesarean delivery.
Uterine dehiscence (n=2) and uterine rupture (n=2) were diagnosed in 6.8% (4/59, 95% CI 2–16%) of the women: in 2.2% (1/46, 95% CI 0.1–12%) of the women with an intact scar or a small scar defect compared with in 23.1% (3/13, 95% CI 5–54%) of the women with a large scar defect (P = .03). This corresponds to an odds ratio of 12.7 (95% CI 0.9–724) for uterine rupture or dehiscence in women with a large scar defect. There were no clinical signs of uterine rupture in the women who gave birth vaginally.
Among the women delivered by cesarean, 5.3% (1/19, 95% CI 0.1–26%) of those with an intact scar or a small scar defect were found to have uterine dehiscence or rupture at the cesarean delivery compared with 42.9% (3/7, 95% CI 10–82%) of the women with a large defect (P=.047). This corresponds to an odds ratio of 11.8 (95% CI 0.7–746) for uterine rupture or dehiscence in women with a large scar defect. Three of the four women with uterine dehiscence or rupture had undergone only one previous cesarean delivery, all three having a large scar defect; one had undergone two previous cesarean deliveries and had a small scar defect.
One of the two women with uterine rupture underwent an elective cesarean delivery because of two previous cesarean deliveries. At saline contrast sonohysterography she had been judged to have a small defect in her uterine scar, the thickness of the remaining myometrium over the scar defect being 5.2 mm. At the cesarean delivery the whole lower uterine segment was described as “thin as a leaf” with a 10-cm complete separation of the scar with communication between the uterine and abdominal cavities. The amniotic sack was bulging into the abdominal cavity with no visceral peritoneum covering the defect. The other woman with uterine rupture underwent an emergency cesarean delivery at 8 cm cervical dilatation because of fetal distress. She had been judged to have a large defect in her uterine scar at saline contrast sonohysterography, the thickness of the remaining myometrium over the scar defect being 0.8 mm. At the cesarean delivery the lower uterine segment was extremely thin and contained a total defect connecting the uterine and abdominal cavities. There was a fresh hematoma at the edge of the defect.
Both women with uterine dehiscence underwent emergency cesarean delivery because of failure to progress. The first woman with uterine dehiscence was delivered at 5 cm cervical dilatation. She had been judged to have a large defect in her uterine scar at saline contrast sonohysterography, the thickness of the remaining myometrium over the scar defect being 2.0 mm. At the cesarean delivery, the left side of the isthmus was very thin, and, in one area, the uterine and abdominal cavities were separated only by the peritoneum. The second woman with uterine dehiscence was delivered when the cervix was fully dilated. She had been judged to have a total scar defect at saline contrast sonohysterography with no remaining myometrium over the defect. At the cesarean delivery, the lower uterine segment was extremely thin, with a dehiscence constituting a large part of it.
Six women gave birth twice after the ultrasound examination of their cesarean scars. Four underwent uncomplicated vaginal delivery twice, one of them having an intact scar, two a small scar defect, and one a large scar defect. The other two women underwent elective cesarean delivery twice, one of them having an intact scar and the other a large scar defect. Uterine scar dehiscence or rupture was not described at any of the deliveries.
Between 2005 and 2009 there were 21,420 deliveries in our institution; the rate of cesarean delivery was 13.7% (2,928/21,420) and the rate of repeat cesarean delivery was 3.3% (709/21,420). Table 3 shows the findings at the latest cesarean delivery in women undergoing repeat cesarean delivery in our institution between 2005 and 2009. Among the patients with uterine rupture, 85% (17/20) were in active labor when the decision to deliver by cesarean was taken. In the years 2005 to 2009, no woman who delivered vaginally after having undergone cesarean delivery had total placenta previa, placenta accreta, placenta increta, or placenta percreta or a diagnosis of uterine rupture.
The results of this study suggest a likely association between large defects in the hysterotomy scar after cesarean delivery detected by transvaginal ultrasonography in nonpregnant women and uterine rupture or dehiscence in subsequent pregnancy.
The strength of our study is that it contributes new information. It has been known for some time that defects in hysterotomy scars after cesarean delivery can be detected at transvaginal ultrasound examination of nonpregnant women12–16 and that such defects are common.12–15 However, despite extensive literature search in the PubMed (search strategy using the MESH terms “cesarean section/adverse effects AND ultrasonography” plus then looking at related articles) we did not find published studies investigating a possible association between scar defects seen by ultrasonography in nonpregnant women and the outcome of subsequent pregnancies and deliveries.
Our study has limitations. First, it is small. Although our data point toward an association between large cesarean scar defects detected at ultrasound examination of nonpregnant women and uterine dehiscence or rupture or both in a subsequent pregnancy, we cannot determine with any precision the strength of this association. Moreover, our study is not large enough to explore interaction between variables, nor to determine whether there are other factors than the ultrasound appearance of the uterine scar that contribute to uterine rupture or dehiscence. A very large number of women examined with ultrasonography after cesarean delivery and with uterine rupture or dehiscence in subsequent pregnancies would be needed to determine this. On the basis of the results of our own studies15,16,18 (including the current study) and information from our clinical obstetric database, we have estimated that to identify 20 cases of uterine dehiscence or rupture at repeat cesarean delivery we would need to scan the hysterotomy scar of 800 primipara and follow them with regard to subsequent pregnancy for 4 years. A second limitation is that information on delivery outcome in subsequent pregnancy was collected retrospectively. The descriptions in the operation reports of the condition of the uterine wall were sometimes difficult to interpret. We have been conservative in our classification and used the term “dehiscence” only when this diagnosis was unequivocal. A third limitation is that women who gave birth vaginally were not routinely examined with exploration of the uterine cavity to detect uterine scar dehiscence or rupture. Undiagnosed dehiscences or ruptures might complicate future pregnancies, and so might be clinically relevant.
The rate of uterine rupture and dehiscence was higher in our study group than in our total population of women undergoing repeat cesarean delivery between 2005 and 2009 (Table 3). We cannot exclude that women with previous complicated pregnancies or complicated cesarean deliveries were more eager to participate in our study than women with less complicated pregnancies or deliveries, and that women in our study were therefore at greater risk of uterine rupture and scar dehiscence than women who declined participation. On the other hand, the combined rate of extremely thin myometrium, uterine dehiscence, and uterine rupture was similar in our study group and in all women giving birth by repeat cesarean delivery between 2005 and 2009 (15.4% compared with 13.0%). All three conditions are likely to be a spectrum of the same condition. The rates of uterine dehiscence or rupture (“uterine scar defect”) at repeat cesarean delivery in our study fall within the extremely wide range reported in studies with a design similar to ours (average 6.6%, range 1–46%).19 Research in this area is difficult because the International Classification of Diseases does not have a code for incomplete uterine rupture (dehiscence) or extremely thin myometrium. Therefore, retrospective studies relying on International Classification of Diseases codes are likely to underestimate the true rate of dehiscence or extremely thin myometrium or both.20 In our study, we retrieved information on the appearance of the uterine isthmus by manually scrutinizing the operation reports of all women who had undergone repeat cesarean delivery in our institution between 2005 and 2009.
On the basis of current knowledge, it is not possible to know whether a defect in a cesarean hysterotomy scar detected by ultrasound examination in a nonpregnant woman is more or less predictive of uterine rupture or dehiscence than the thickness of the myometrium in the uterine isthmus measured by ultrasound examination in late pregnancy.19 Our results cannot be used to provide recommendations on how to routinely use ultrasound examinations of cesarean hysterotomy scars in nonpregnant women to manage subsequent pregnancies and deliveries. However, they support that women in whom a large cesarean scar defect is detected at a gynecologic scan are likely to be a high-risk group for uterine dehiscence or rupture, and that therefore it may be justified to scan their lower uterine segment near term if they become pregnant again, or at least to be aware that these women belong to a high-risk group.
Our findings should not lead to any changes in clinical practice. After all, 10 (77%) of the 13 women in our study with a large scar defect underwent a delivery in which there were no signs of uterine dehiscence or rupture. We do believe, however, that our results justify the performance of an appropriately powered prospective study—prospective also with regard to evaluation of the lower uterine segment if repeat cesarean delivery is carried out—designed to determine with precision the association between large cesarean hysterotomy scar defects and uterine dehiscence or rupture in subsequent pregnancies and also the effect of other factors on this risk.
1. Blanc B, Capelle M, Bretelle F, Leclaire M, Bouvenot J. The worrying rise in the frequency of cesarean section [French]. Bull Acad Natl Med 2006;190:905–13.
2. Althabe F, Sosa C, Belizan JM, Gibbons L, Jacquerioz F, Bergel E. Cesarean section rates and maternal and neonatal mortality in low-, medium-, and high-income countries: an ecological study. Birth 2006;33:270–7.
3. Betran AP, Merialdi M, Lauer JA, Bing-Shun W, Thomas J, Van Look P, et al. Rates of caesarean section: analysis of global, regional and national estimates. Paediatr Perinat Epidemiol 2007;21:98–113.
4. Ben Nagi J, Ofili-Yebovi D, Marsh M, Jurkovic D. First-trimester cesarean scar pregnancy evolving into placenta previa/accreta at term. J Ultrasound Med 2005;24:1569–73.
5. Hamilton BE, Minino AM, Martin JA, Kochanek KD, Strobino DM, Guyer B. Annual summary of vital statistics: 2005. Pediatrics 2007;119:345–60.
6. Clark SL, Koonings PP, Phelan JP. Placenta previa/accreta and prior cesarean section. Obstet Gynecol 1985;66:89–92.
7. Grobman WA, Gersnoviez R, Landon MB, Spong CY, Leveno KJ, Rouse DJ, et al. Pregnancy outcomes for women with placenta previa in relation to the number of prior cesarean deliveries. Obstet Gynecol 2007;110:1249–55.
8. Kennare R, Tucker G, Heard A, Chan A. Risks of adverse outcomes in the next birth after a first cesarean delivery. Obstet Gynecol 2007;109(2 pt 1):270–6.
9. Diaz SD, Jones JE, Seryakov M, Mann WJ. Uterine rupture and dehiscence: ten-year review and case-control study. South Med J 2002;95:431–5.
10. Silver RM, Landon MB, Rouse DJ, Leveno KJ, Spong CY, Thom EA, et al. Maternal morbidity associated with multiple repeat cesarean deliveries. Obstet Gynecol 2006;107:1226–32.
11. Landon MB, Hauth JC, Leveno KJ, Spong CY, Leindecker S, Varner MW, et al. Maternal and perinatal outcomes associated with a trial of labor after prior cesarean delivery. N Engl J Med 2004;351:2581–9.
12. 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;23:289–92.
13. 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;31:72–7.
14. Armstrong V, Hansen WF, Van Voorhis BJ, Syrop CH. Detection of cesarean scars by transvaginal ultrasound. Obstet Gynecol 2003;101:61–5.
15. Osser OV, Jokubkiene L, Valentin L. High prevalence of defects in Cesarean section scars at transvaginal ultrasound examination. Ultrasound Obstet Gynecol 2009;34:90–7.
16. 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;35:75–83.
17. Mehta CR, Patel NR. Exact logistic regression: theory and examples. Stat Med 1995;14:2143–60.
18. Vikhareva Osser O, Valentin L. Risk factors for incomplete healing of the uterine incision after caesarean section. BJOG 2010;117:1119–26.
19. Jastrow N, Chaillet N, Roberge S, Morency AM, Lacasse Y, Bujold E. Sonographic lower uterine segment thickness and risk of uterine scar defect: a systematic review. J Obstet Gynaecol Can. 2010;32:321–7.
20. Al-Zirqi I, Stray-Pedersen B, Forsén L, Vangen S. Uterine rupture after previous caesarean section. BJOG 2010;117:809–20.