[ Editorials | Letters | Selected Articles | Signature Series | Patient Handouts | Index ]

2002 Selected Articles

Prophylactic Benefits of Elective Cesarean Delivery

Peter K. Sand, MD

Pregnancy and vaginal delivery have long been associated with an increased incidence of urinary incontinence and anal incontinence and genital prolapse. Studies of these effects of pregnancy and delivery were reported as early as the nineteenth century.sup>1Recently, OB/GYNs have begun to question whether they can influence the development of urogenital tract disorders during pregnancy and parturition. Through careful neurologic, clinical, and epidemiologic investigation, more of the putative injuries and risk factors for prolapse and incontinence have been discovered in the past 30 years.

PHYSIOLOGY

The development of genuine stress incontinence (GSI) appears to involve a genetic predisposition toward myofascial weakness following pelvic floor injury. Investigations have shown that women who develop urogenital prolapse and incontinence heal with a higher proportion of weaker type 3 collagen than type 1 collagen.sup>2This predisposes them to weakening of pelvic-floor support. These genetic changes are probably multifactorial, and are still poorly understood.

Better understood is the association between GSI and anal incontinence with injury to the pudendal and perineal nerves. Several investigators have shown delayed pudendal nerve conduction in women with urogenital prolapse, anal incontinence, and GSI.sup>3-6Delayed conduction decreases the strength, speed, and duration of the reflex contraction of the levator ani muscles, which act as a posterior support for the bladder, urethra, vagina, and rectum. Normally, the levator ani muscles contract within 1/20 of a second following the onset of an increase in intraabdominal pressure.

This rapid contraction extends the levator plate anteriorly to close the genital hiatus, causing a cephalad displacement or flattening of the levator ani muscles to elevate the urethra, bladder, vagina, and rectum even before the increase in intraabdominal pressure starts to displace them caudally. Diminution and slowing of this reflex contraction after pudendal nerve injury allows for posterior displacement of the urogenital tract and stress on its connective tissue supports, which may eventually result in urethral hypermobility and genital prolapse. Myoneural damage to the levator ani muscles leads to widening of the genital hiatus and the further potential for downward displacement of the urogenital structures. This delayed pudendal and perineal nerve conduction also has a direct adverse effect on anal sphincter function, resulting in decreased anal resting and squeeze pressures.

NEUROLOGIC EFFECTS OF VAGINAL DELIVERY

In 1990, Snooks et alsup>5reported on the 5-year follow-up of a subpopulation of their original 1984 studysup>7in which they described the neurologic changes in the innervation of the pelvic floor musculature in 122 women who were studied prospectively through pregnancy and delivery. In the original study group, 80% of the women sustained occult but reversible pudendal nerve damage and had reduced anal pressures compared with controls. Sixty percent of these women recovered by 2 months postpartum. Parity, forceps delivery, and increased length of the second stage of labor were associated with more severe injury. The follow-up study investigated 14 of the original 24 women who had spontaneous vaginal deliveries. Five (36%) of these 14 women had persistent increased anal sphincter fiber density and increased pudendal nerve terminal motor latency measurements 5 years after vaginal delivery compared with their antepartum measurements. All five of these women developed symptomatic GSI, and three developed flatal incontinence during the 5-year follow-up period.

Allen et alsup>4examined the neurologic effects of delivery in a prospective study of 96 nulliparous woman recruited at 36 weeks’ gestation who were evaluated with concentric-needle electromyography (EMG) and pelvic-floor strength testing before labor, 2 to 5 days postpartum, and 2 months postpartum. The results showed persistent decreased pelvic-floor strength 2 months postpartum (p = 0.0006) and persistent EMG and pudendal conduction abnormalities in 80% of women after their first vaginal delivery. None of these changes occurred following elective cesarean delivery.

ANAL EFFECTS OF VAGINAL DELIVERY

Sultan et alsup>6prospectively studied 202 consecutive women 6 weeks prior to delivery. Of this group, 150 were studies 6 weeks postpartum, and 32 of them with abnormal findings 6 months after delivery to assess the effect of vaginal delivery on the anal sphincter. On anal ultrasonography, 28 of 79 primiparous women (35%) developed anal sphincter tears after vaginal delivery, which persisted in all 22 women studied at 6 months postpartum. Forty percent of the 48 multiparous women had an anal sphincter defect before the incident delivery, and 44% at 6 weeks postpartum. Sphincter defects were found in 8 of the 10 women who underwent forceps delivery. None of the 23 women who underwent cesarean deliveries were found to have sphincter defects. No women developed anal incontinence after cesarean delivery, whereas 5% of primiparas and 4% of multiparas developed anal incontinence after vaginal birth. Three percent of the primiparas and 21% of the multiparas also had symptomatic GSI 6 weeks following vaginal delivery. There were also significant decreases in anal resting and squeeze pressures on anal manometry in primiparous women (p < 0.001) and multiparous women (p < 0.004), as well as prolonged pudendal motor nerve latencies in primiparas (p <0.001) and multiparas (p <0.002) 6 weeks after vaginal delivery. None of these changes were noted following cesarean delivery.

A subsequent multicenter prospective trial by Chaliha et alsup>8in 286 nulliparous women evaluated anal symptoms, sensation, and manometric function in the third trimester and 3 months postpartum. As with Sultan et al,sup>6this study found that 38% of women developed anal sphincter tears after vaginal delivery versus 3% after cesarean delivery. Vaginal delivery led to significant decreases in anal resting and squeeze pressures, while there were no changes after cesarean delivery. Vaginal delivery was strongly associated with anal sphincter disruption (p < 0.0001). Xanos et alsup>9have shown that such defects are strongly associated with the subsequent development of anal incontinence, even when no symptoms are present in the postpartum period. Furthermore, 62% of women with sphincter disruptions complained of anal incontinence 8 years later, compared with 25% of women delivering vaginally who did not have anal sphincter disruptions.

URINARY EFFECTS OF VAGINAL DELIVERY

In the first prospective, longitudinal, cohort analysis of the effects of pregnancy and delivery on urinary function, Francissup>10studied 400 women attending a prenatal clinic in the first trimester. The women were carefully questioned about urinary symptoms before pregnancy, during each trimester, and postpartum. Two hundred and twenty-two of these women were primiparas, and 178 were multiparas. Antepartum GSI was found in 53% of the primiparous women and 85% of the multiparous women during gestation. This was confirmed by cystometry in all 268 subjects. Following vaginal delivery, mild GSI persisted in 38% and severe GSI in 9% of these women, but in none of the 20 women with antepartum incontinence who delivered by cesarean delivery. Stanton et alsup>11prospectively studied 181 women in the third trimester and through the puerperium. Of the 83 primiparas, 38% had GSI in the third trimester and 6% had persistent postpartum GSI. Of the 98 multiparas, 10% had a history of GSI prior to pregnancy, 42% had GSI in the third trimester, and 11% had persistent postpartum GSI.

Looking at the physiologic changes of pregnancy and delivery mode, Van Geelen et alsup>12performed urodynamic evaluations on pregnant women before and after delivery. They found a significant decrease in urethral closure pressure and functional length. No change was found in the women who underwent cesarean delivery. Meyer et alsup>13found similar changes in functional urethral length as well as intravaginal and intraanal pressure 9 weeks after vaginal delivery compared with antepartum values in a prospective study of 149 women. None of these changes were found in the 33 women who had cesarean delivery. Consistent with these differences in physiologic parameters, only one of the 33 women delivered by cesarean delivery had persistence of GSI postpartum, compared with 36% of the 25 women delivered by forceps and 21% of the 91 women who delivered spontaneously.

Peschers et alsup>14studied the anatomic effects of vaginal delivery, and found that bladder neck support was significantly worse after vaginal delivery than following cesarean delivery (p < 0.001) or in a group of 25 nulliparous controls (p < 0.001). They also found that bladder neck descent during Valsalva maneuver was significantly increased after vaginal delivery compared with cesarean delivery in both primiparous and multiparous women (p < 0.001).

Viktrup et alsup>15performed a prospective study questioning 305 primiparous women about urinary incontinence symptoms before, during, and after pregnancy. Their multivariate analysis found that length of the second stage, head circumference, episiotomy, and birth weight were associated with postpartum GSI, whereas cesarean delivery was protective against GSI. In women with no GSI during pregnancy, 21 of 167 (13%) developed the disorder postpartum compared with none of the 35 delivered by cesarean delivery (p < 0.05). At 3 months postpartum, only 4% of these women had persistent GSI, and at 1 year only 3% still had GSI. In a 5-year follow-up study, however, Vikrup and Losesup>16questioned 278 of the 305 women (91%) in their original study, and found a 30% prevalence of GSI. Nineteen percent of the women who were not incontinent in the original trial developed GSI in the ensuing 5 years. The investigators attributed this to worsening pudendal neuropathy following the initial vaginal delivery. Again, cesarean delivery was found to significantly decrease the risk of GSI.

Similar conclusions were reached by Meyer et alsup>17in their prospective study of 149 primiparous women during and after pregnancy. They found that 31% of these primiparas had antepartum GSI that persisted in 21% after spontaneous vaginal delivery, but in only 3% following cesarean delivery. Anal incontinence was also noted in 5.5% of primiparas after vaginal delivery, but in none following cesarean delivery.

Population-based trials since 1998 by Hojberg et al,sup>18Persson et al,sup>19Moller et al,sup>20and Marshall et alsup>21consistent with the prospective trials show that there is a strong association between vaginal delivery and parity with GSI. Hojberg et alsup>18studied 1,781 primiparas at 16 weeks’ gestation, and showed an odds ratio (OR) of 5.7 for GSI after vaginal delivery compared with 1.3 following cesarean delivery. Persson et alsup>19studied 10,074 Swedish women undergoing surgery for GSI, and found that the OR of prior cesarean delivery versus vaginal delivery was 0.21. There was also a strong association between GSI and parity. Moller et alsup>20studied 502 women with lower urinary tract symptoms and 742 controls. There was an association between parity and GSI with an OR of 2.2 after one vaginal delivery, 3.9 after a second vaginal delivery, and 4.5 after a third delivery. Marshall et alsup>21studied 7,771 women early in the puerperium, and found a strong association between GSI and constipation with parity. Continued straining at bowel movements would theoretically further increase urethral hypermobility, with concomitant GSI and urogenital prolapse.

Iosif and Ingemarssonsup>1showed that cesarean delivery may also be related to the development of GSI, but much less commonly than vaginal birth. In 1982, they performed a retrospective trial involving 204 of 264 women who underwent elective cesarean delivery 1 to 6 years earlier. They found that 4.7% of these women had persistent GSI after primary cesarean delivery, and 4.1% after a second cesarean delivery.

These data clearly suggest a putative role of vaginal delivery in the development of GSI, anal incontinence, and neurologic injury leading to the development of these disorders and genital prolapse. These patients often require pelvic reconstructive surgery, with the inherent costs and morbidity that are usually not considered in cost analyses of delivery mode. Olsen et alsup>22performed a retrospective cohort study of 149,554 women in a managed care organization, and found an 11.1% lifetime risk for prolapse and anti-incontinence surgery. Twenty-nine percent of these operations were repeat surgeries, which are associated with potentially even more morbidity. Wagner and Husup>23reported direct costs of $26.3 billion in 1995 for urinary incontinence in patients over age 64. This averaged out to an annual cost of $3,565 for each incontinent patient. Korn and Learmansup>24reported that anti-incontinence operations increased from 78,000 in 1988 to 121,000 in 1991, such that $500 million was spent on anti-incontinence operations in 1991.

Chung et alsup>25performed a cost-effectiveness analysis of vaginal birth after cesarean delivery (VBAC), and concluded that below a 74% success rate for VBAC, repeat elective cesarean delivery is more cost-effective, with better outcomes than a trial of labor. This study examined only the costs incurred during the incident delivery, and did not consider the long-term effects of vaginal birth after cesarean delivery on the pelvic floor.

CONCLUSION

These data show that vaginal delivery is associated with injury to the pudendal nerve, which is further aggravated when forceps or an episiotomy is utilized. These changes and direct damage to the fibromuscular supports of the urogenital tract after vaginal delivery lead to persistence of GSI after vaginal delivery to a much greater extent than after cesarean delivery. Anal sphincter injury and anal incontinence are clearly linked to vaginal delivery, and are aggravated by forceps usage and episiotomy. The literature supports the avoidance of forceps and episiotomy in all patients to help prevent these complications.

The development of genital prolapse appears to be related to pudendal nerve injury and loss of levator muscle strength. These changes occur chronically, and are worsened by constipation and chronic increases in intraabdominal pressure. Vaginal delivery has been shown to predispose women to these conditions.

While more data are clearly needed before elective cesarean delivery can be recommended for all women, this strategy should be considered for certain populations at high risk for the development of urogenital prolapse and incontinence. In addition, VBAC, which may often be cost-ineffective,sup>25should be abandoned in women at risk for the development of urinary and anal incontinence, as well as in those who are at risk of failing a trial of labor. This strategy could result in significant cost savings, relief from the suffering and stigma of incontinence, and the elimination of thousands of operations for prolapse and incontinence every year.sup>22,23

Peter K. Sand, MD, is associate professor of obstetrics and gynecology, director of
the Division of Urogynecology, and director of the Evanston Continence Center for Evanston Northwestern Healthcare at Northwestern University Medical School
in Chicago.

REFERENCES

1. Limbo RK, Wheeler SR. When a Baby Dies: A Handbook for Healing and Helping. LaCrosse, WI: Bereavement Services; 1998.
2. Woods JR Jr. Pregnancy-loss counseling: the challenge to the obstetrician. In: Woods JR Jr, Esposito Woods JL, eds. Loss During Pregnancy or in the Newborn Period: Principles of Care With Clinical Cases and Analyses. Pitman, NJ: Jannetti Publications; 1997:71–123.
3. Sainsbury MK. Grief in multifetal death. Acta Genet Med Gemellol (Roma). 1988;37(2):181–185.
4. Bryan E, Denton J, Hallett F. Guidelines for Professionals: Multiple Pregnancy. London, England: Multiple Births Foundation; 1997.
5. McKinney MK, Tuber SB, Downey JI. Multifetal pregnancy reduction: psychodynamic implications. Psychiatry. 1996;59 (4):393–407.
6. Ball HL, Hill CM. Reevaluating twin infanticide. Curr Anthropol. 1996;37(5):856–863.
7. Lewis E, Bryan EM. Management of perinatal loss of a twin. BMJ. 1988; 297(6659):1321–1323.
8. Netzer D, Arad I. Premature singleton versus a twin or triplet infant death: parental adjustment studied through a personal interview. Twin Res. 1999;2(4):258–263.
9. Bryan EM. The death of a twin. Palliat Med. 1995;9(3): 187–192.
10. Cuisinier M, deKleine M, Kollee L, et al. Grief following the loss of a newborn twin compared to a singleton. Acta Paediatr. 1996;85(3):339–343.
11. Multiple Birth Loss and the Hospital Caregiver. Anchorage, AK: Center for Loss in Multiple Birth; 1993:1–4.
12. Harrigan R, Naber MM, Jensen KA, et al. Perinatal grief: response to the loss of an infant. Neonatal Netw. 1993; 12(5):25–31.
13. Bryan E, Hallett F. Guidelines for Professionals: Bereavement. London, England: Multiple Births Foundation; 1997.
14. Wilson AL, Fenton LJ, Stevens DC, Soule DJ. The death of a newborn twin: an analysis of parental bereavement. Pediatrics. 1982;70(4):587–591.
15. Thorpe K, Golding J, MacGillivray I, Greenwood R. Comparison of prevalence of depression in mothers of twins and mothers of singletons. BMJ. 1991;302(6781):875–878.
16. Pector, EA, Smith-Levitin, M. Grief and psychological issues in multifetal pregnancy loss. Isr J Obstet Gynecol. 2000;11 (4):155–166.
17. Pearlman EM, Ganon JA. When a twin dies. In: Pearlman EM, Ganon JA, eds. Raising Twins. New York, NY: HarperCollins; 2000:209–225.
18. Van der Zalm JE. Accommodating a twin pregnancy: maternal processes. Acta Genet Med Gemellol (Roma). 1995;44 (2):117–133.
19. Baram DA. Termination of pregnancy for fetal abnormalities. In: Woods JR Jr, Esposito Woods JL, eds. Loss During Pregnancy or in the Newborn Period: Principles of Care with Clinical Cases and Analyses. Pitman, NJ: Jannetti Publications; 1997:307–330.
20. Sweeney MM. The value of a family-centered approach in the NICU and PICU: one family’s perspective. Pediatr Nurs. 1997;23(1):64–66.
21. Noble E. Having Twins. Boston, MA: Houghton Mifflin; 1991:307–328, 380–383.
22. O’Leary J, Parker L, Thorwick C. After Loss: Parenting in the Next Pregnancy. Minneapolis, MN: Abbott Northwestern Hospital; 1998:11–12, 84–90.
23. Hanrahan J. Grieving while lactating. Clin Issues Lactation. 1999;4(1).
24. O’Leary JM, Thorwick C. Impact of pregnancy loss on subsequent pregnancy. In: Woods JR Jr, Esposito Woods JL, eds. Loss During Pregnancy or in the Newborn Period: Principles of Care with Clinical Cases and Analyses. Pitman, NJ: Jannetti Publications; 1997:431–443.
25. Raffensperger J. A philosophical approach to conjoined twins. Pediatr Surg Int. 1997;12(4):249–255.

back to top