MECHANISMS OF REPRODUCTIVE EFFECTS — Cervical surgery involves removal or destruction of tissue, which results in scarring. Theoretically, the integrity of the cervix is better preserved following ablation compared with excision. The mechanisms by which these changes affect reproductive function have not been thoroughly investigated. Available evidence and hypotheses regarding pathophysiology are presented in this section.

Tissue removal or destruction impacts both the cervical glands and stroma. Removal of cervical glands may adversely affect fertility by altering the cervical mucus that is necessary for normal sperm migration and viability. In addition, removal or destruction of a large portion of the collagen matrix that constitutes the cervical stroma may decrease tensile strength, thereby permitting the cervix to dilate prematurely during pregnancy.

Both removal of tissue and loss of cervical glands potentially increase the risk of ascending infection. A shorter cervix may predispose to migration of bacteria by increasing access from the vagina to the uterine cavity. Additionally, cervical mucus functions as a potential barrier to ascending infection. Some data suggest that women who have undergone CIN treatment are at an increased risk of microbial invasion of the amniotic fluid, which is a risk factor for preterm delivery and maternal and neonatal sepsis [2-4]. Some studies have found that conization is followed by alterations in the normal vaginal flora, specifically a decrease in the amount of lactobacilli [5]. There is no evidence that nonpregnant women with prior CIN treatment are at an increased risk of pelvic inflammatory disease.

Scarring presents as cervical stenosis or difficult cervical dilation during labor. Cervical scarring after treatment may also lead to a loss of cervical plasticity. This may render the membranes more vulnerable to shearing forces and potentially contribute to preterm premature rupture of membranes.

CERVICAL STENOSIS — Scarring from cervical surgery may result in cervical stenosis. Cervical stenosis has several potential adverse effects. A stenotic cervix impairs the clinician's ability to examine the transformation zone and endocervical canal and increases the difficulty of office procedures that typically would not require mechanical dilation (eg, endometrial biopsy). Stenosis partially or completely occludes access to the uterine cavity and may interfere with conception or impede menstrual flow; in severe cases, hematometra or pyometra can occur. In pregnant women with cervical stenosis, the cervix may fail to dilate normally during labor.

Cervical stenosis following excisional treatment for CIN has been reported in up to 8 percent of patients. Risk factors for cervical stenosis in women undergoing excisional treatment include the amount of tissue removed (eg, incision depth of ≥1 to 2 cm was associated with cervical stenosis) and postmenopausal status.

Cervical stenosis appears to be rare (1 percent or less) after cryotherapy or laser ablation. The risk of cervical stenosis following cervical ablation is increased in women with a history of in-utero exposure to diethylstilbestrol.

INFERTILITY — Treatment of CIN does not appear to impair fertility [6-9]. This was illustrated in a systematic review of four retrospective studies that found that women who had undergone cervical conization or ablation had no impairment in time to conception or total number of pregnancies [6]. Possible exceptions to this are women with severe cervical stenosis

In addition, a study based on a registry of over 800,000 deliveries (including 8295 deliveries in women with prior CIN treatment) found that the rate of pregnancies conceived through in vitro fertilization (IVF) was similar in women with and without prior CIN treatment (1.6 versus 1.5 percent) [7]. Drawing a conclusion from this study regarding the impact of CIN treatment on fertility is limited by the absence of data on non-IVF conceptions and by the lack of data on patients who underwent one or more cycles of IVF but did not ultimately have a delivery.

SECOND TRIMESTER PREGNANCY LOSS — Previous cervical conization is associated with an increased risk of second trimester pregnancy loss [10,11]. The best available data are from a population-based study which compared 15,108 births in women who had previously undergone cervical conization (cold knife, laser, or LEEP) with two other data sets: 2,164,006 births in women who had no CIN treatment and 57,136 births in women who underwent cervical conization after their index pregnancy [11]. The risk of delivery at <24 weeks of gestation was significantly higher in women with prior conization compared with either those who subsequently underwent conization or had no CIN treatment (1.5 versus 0.4 percent). A limitation of this study was that the results for each type of conization were not reported separately.

There are no data regarding second trimester pregnancy loss in women treated for CIN with ablative methods. The availability of these data are limited, since second trimester pregnancy losses are rare and birth registries do not usually collect information on second trimester miscarriages or preterm deliveries beyond 20 or 22 weeks, which further makes studies more difficult.

PRETERM PREMATURE RUPTURE OF MEMBRANES — The risk of preterm premature rupture of membranes (PPROM) is increased in women with a history of some, but not all CIN treatment procedures. This was demonstrated in a meta-analysis of 27 retrospective studies of women who had been treated for CIN [6]. The risk of PPROM was significantly increased in women who had previously undergone LEEP compared with those who had no CIN treatment (5 versus 2 percent). Laser conization and ablation were not associated with a significant increase in PPROM frequency. There were no data regarding cold knife conization or cryotherapy.

PRETERM DELIVERY AND PERINATAL MORTALITY — The risk of preterm delivery or perinatal mortality varies by type of cervical procedure and, for excisional procedures, and also possibly by the amount of tissue removed. Preterm delivery and perinatal mortality are related outcomes; most cases of perinatal mortality in this setting are secondary to complications of prematurity.

Risks of individual treatment methods — Cold knife conization is the primary method for CIN treatment that is associated with a subsequent risk of preterm delivery and perinatal mortality. The risk of preterm delivery following LEEP is controversial. Laser conization and cervical ablation are not associated with preterm labor or perinatal mortality.

The best available data regarding the effect of cervical ablation or excision for CIN on preterm delivery and perinatal mortality were provided by a systematic review of 20 mostly retrospective studies (one prospective cohort study was included) with over 12,000 deliveries [12]. The major findings were:

Cold knife conization increased the risk of perinatal mortality and preterm delivery:

- Perinatal mortality (4.3 versus 0.5 percent, RR 2.9, 95% CI 1.4-5.8)

- Preterm delivery at <32 to 34 weeks (4.6 versus 1.6 percent, RR 2.8, 95% CI 1.7-4.5)

- Preterm delivery at <28 to 30 weeks (4.2 versus 0.8 percent, RR 5.3, 95% CI 1.6-17.4)

LEEP did not increase the risk of perinatal mortality and preterm delivery:

- Perinatal mortality (0.6 versus 0.5 percent, RR 1.2 95% CI 0.7-1.9)

- Preterm delivery at <32 to 34 weeks (2.0 versus 1.4 percent, RR 1.2, 95% CI 0.5-2.9)

- Preterm delivery at <28 to 30 weeks (0.5 versus 0.8 percent, RR 0.4, 95% CI 0.2-0.8)

This meta-analysis was designed to study very serious outcomes: perinatal mortality and very preterm (28 to 32 weeks) and extremely (under 28 weeks) preterm births, not late preterm births (from >34 to <37 weeks). While late preterm infants are less likely than those delivered earlier to have long-term sequelae of prematurity, late preterm delivery is associated with increased morbidity and mortality compared with term infants.

In contrast, several subsequent studies have found an association between LEEP and preterm delivery [13-15]. The largest of these was a Danish population-based study that reported a significantly increased risk of spontaneous preterm delivery at all gestational ages in 8180 women who had undergone LEEP compared with other women: 21 to 27 weeks (0.6 versus 0.2 percent); 28 to 31 weeks (0.8 versus 0.3 percent); and 32 to 36 weeks (5.0 versus 2.9 percent) [13]. The inclusion of these data in a meta-analysis are needed to assess whether this would change the conclusion of the meta-analysis of 20 studies [12].

Laser conization was not associated with an increased risk of perinatal mortality or preterm delivery. However, LEEP has largely replaced laser conization and therefore there are few data regarding this method and further study is needed.

Ablative methods did not increase the risk of perinatal mortality and preterm delivery. For women who underwent cryotherapy, the rate of perinatal mortality was 0.2 percent and the rates of preterm delivery were 0.6 at <32 to 34 weeks and 2.2 percent at <28 to 30 weeks. Only a small number of studies investigated perinatal risks following laser ablation; further study of this method is needed.

A limitation of most studies in this meta-analysis is that women with CIN treatment were compared to healthy obstetric controls rather than to women with CIN who did not undergo treatment. Some, but not all, studies have reported that CIN, regardless of treatment, is associated with an increased risk of preterm birth [14,16-19]. This may be due to characteristics that are found commonly in women with CIN as well as those who have a preterm delivery (eg, smoking, bacterial vaginosis, low socioeconomic status) [20-22]. As an example, a Scottish national registry study found that women with CIN3 (n = 3113) were significantly more likely to have spontaneous preterm delivery (11 versus 6 percent) and PPROM (8 versus 6 percent) than other women [18]. Of note, among women with CIN, the risks of these outcomes were increased regardless of whether the CIN was treated or not.

Thus, use of a control group without CIN may result in an overestimate of the preterm birth rate in women treated for CIN. This assertion is supported by several studies that used women with CIN as a control group and have reported a lower relative risk of preterm delivery than the meta-analysis of 20 studies [11,16,23,24].

Modifying factors

Depth of excision — Greater depth of excision (cone depth of ≥10 mm up the cervical canal) appears to be an independent risk factor for preterm birth and PPROM in subsequent pregnancies, based on three observational studies of LEEP or laser conization [6,15,17]. As an example, a Danish population-based study evaluated over 8000 deliveries in women who had undergone LEEP conization. The risk for preterm birth increased by 6 percent for each additional excised millimeter beyond 12 mm.

Number of procedures — The risk of preterm deliveries may increase in women who undergo more than one cervical conization [19]. Some, but not all, observational studies have reported that the risk of preterm delivery in women who had two or more compared with one conization procedures increased two- to fivefold (5 versus 24 percent in one study [15]) [17,19,25].

There are no data regarding the effect of repeat ablation procedures on obstetric outcomes.

Short interval from treatment to pregnancy — The effect on preterm delivery of a short interval between the excisional procedure and delivery is controversial. The working hypothesis is that the risk of preterm birth may be increased if the cervical epithelium has not had adequate time to heal (short procedure-to-conception period). In a population-based retrospective cohort study of over 8000 women who had undergone LEEP, those who delivered within <1 year compared with ≥1 year post-LEEP had no significant difference in the rate of preterm delivery (6.6 versus 5.9 to 7.1 percent) [17]. In contrast, a nested case-control study reported that women who had a preterm delivery had a shorter conization (either cold knife or LEEP) to conception interval than women who delivered at term (2.5 versus 10.5 months) [26]. In contrast, short interval between cervical treatment and delivery did not increase the risk for preterm birth in all studies [17,27]. The measure used in the second study, treatment to conception interval, is preferable to treatment to delivery interval. This is because treatment to delivery interval may be altered by differences in pregnancy duration (ie, preterm pregnancies are shorter than term pregnancies).

Multiple gestation — Some data suggest that a history of LEEP may further increase the risk of preterm delivery in women with multiple gestation. As an example, a retrospective study of over 9000 women with multiple gestation, 166 of whom had undergone a LEEP, reported a higher rate of preterm delivery in the women with a history of LEEP (43 versus 33 percent) [28]. Further study of this issue is needed.

Summary — The main priorities for women with CIN are to diagnose or prevent cervical cancer. As noted above, the gold standard for CIN treatment is excisional treatment, but less invasive methods appear to result in fewer reproductive risks.

Ablative methods (cryotherapy, laser ablation) are not associated with increased risks of perinatal mortality or preterm delivery, although there are few data regarding laser ablation. For women who are planning future pregnancy and are candidates for all methods of CIN treatment, we suggest ablation in selected cases. Ablative methods can be used for women with squamous disease that is low-grade or selected women with small high-grade lesions.

Women who are at a higher risk of invasive cervical cancer should be treated with excisional therapy. LEEP does not appear to increase the risk of perinatal mortality or of preterm delivery at ≤34 weeks. The effect of LEEP on preterm delivery requires further study. Data conflict regarding the association between LEEP and delivery at ≤34 weeks; the highest quality data, a meta-analysis of 20 observational studies, found no increased risk. On the other hand, it appears that there is an increased risk of late preterm delivery in women who have undergone LEEP. The depth of excision appears to influence the risk of preterm delivery.

Available evidence shows no effect of laser conization on these perinatal outcomes. In contrast, cold knife conization is associated with a threefold or higher increased risk of perinatal mortality and preterm delivery.

Given this difference between cold knife conization compared with other excisional treatments, we recommend not using this method in women who are planning pregnancy, whenever possible. Decisions regarding the need for cold knife conization are made by most surgeons for each patient based upon characteristics of the patient (cervical cancer risk factors including results of HPV serotype testing) or lesion (grade, size, histology, recurrence).

Women who desire future childbearing and in whom cold knife conization is planned should be counseled about subsequent reproductive risks. Shared decision-making between the clinician and such patients is necessary regarding the trade-off between the risk of cervical cancer and potential reproductive risks.

PERFORMANCE OF PROCEDURES DURING PREGNANCY — Cervical ablation or excision performed during pregnancy is associated with an increased risk of adverse pregnancy outcome, including heavy vaginal bleeding and spontaneous abortion. As such, every effort should be made to avoid such procedures during pregnancy, although they may be needed in select cases to exclude invasive disease. In these cases, conization in the operating room, with removal of a shallow cone to rule out invasive disease, is usually advised.

OBSTETRIC MANAGEMENT IN SUBSEQUENT PREGNANCIES — At a preconceptional or first obstetric visit, women should be asked regarding prior treatment for CIN. Women with this history should be counseled about the risks of preterm delivery and associated complications (eg, preterm premature rupture of membranes, perinatal mortality).

Prospective series have found that a short cervix on second trimester transvaginal ultrasound examination was predictive of an increased risk of preterm birth in women who have had previous conization, similar to other women [29,30]. Given the risk of preterm delivery following cold knife conization, we recommend surveillance with serial transvaginal sonographic cervical length measurement. Ultrasound examinations are typically performed every other week from 16 to 32 weeks. Cervicovaginal fetal fibronectin (fFN) testing can be added as clinically indicated after 22 weeks gestation to further refine the risk for preterm birth.

The role of prophylactic cerclage in women who have undergone cold knife conization has not been well studied. In one retrospective study of 108 women with prior conization, the rate of preterm delivery was similar in those with and without a prophylactic cerclage (23 versus 21 percent) [31]. In general, the effectiveness of prophylactic cerclage is controversial and its use is associated with perinatal risks. We suggest not performing prophylactic cerclage in women based solely on a history of cold knife conization.

For women who have had a LEEP, the risk of preterm delivery, if any, is after 34 weeks. This is beyond the usual period of cervical length surveillance (up to 32 weeks). Thus, we suggest not performing routine cervical length surveillance in women who have undergone LEEP. Surveillance may be reasonable in women who have had a very deep LEEP conization (>1 to 2 cm) or two or more LEEP procedures.

There are no data on the use of progesterone supplementation to prevent preterm birth in women at high risk by virtue of a prior cone biopsy alone.

SUMMARY AND RECOMMENDATIONS

Cervical intraepithelial neoplasia (CIN) is a precursor to cervical cancer that is managed with either surveillance or cervical treatment procedures. CIN is treated with either cervical excision (ie, conization) or ablation. (See 'Excision versus ablation' above.)

Cervical stenosis occurs more frequently following excisional than ablative procedures. (See 'Cervical stenosis' above.)

Treatment of CIN does not appear to impair fertility in most women. (See 'Infertility' above.)

Excisional procedures are associated with an increased risk of second trimester pregnancy loss. (See 'Second trimester pregnancy loss' above.)

The risk of preterm premature rupture of membranes is increased for women following some (eg, LEEP), but not all, CIN treatment procedures (laser conization or laser ablation). There are no data regarding cold knife conization and the risk of membrane rupture. (See 'Preterm premature rupture of membranes' above.)

Cold knife conization increases the risk of extremely preterm and very preterm delivery as well as perinatal mortality. It appears that LEEP is not associated these extremely severe obstetric outcomes, but this issue requires further study. LEEP probably increases the risk for late preterm birth (from 34 to <37 weeks of gestation). Other CIN treatment methods do not increase the risk of perinatal mortality or extremely preterm and very preterm delivery. (See 'Risks of individual treatment methods' above.)

Women who are planning future pregnancy and have indications for CIN treatment should be treated with the method that best diagnoses or prevent cervical cancer and also incurs the lowest risk of reproductive effects.

- For women who are planning future pregnancy and are candidates for all methods of treatment, we suggest ablation rather than excision (Grade 2B). This includes women with squamous disease that is either low-grade or selected women with small high-grade lesions.

- For women who are planning future pregnancy, who require excision and are candidates for all excisional methods, we recommend not using cold knife conization (Grade 1B). Decisions regarding the need for cold knife conization are generally made individually based upon characteristics of the patient (cervical cancer risk factors) and lesion (grade, size, histology, recurrence).

(See 'Summary' above.)

Pregnant women who have a history of CIN treatment should be counseled about the risks of preterm delivery and options for obstetric management.

- For pregnant women who have had a cold knife conization, we recommend serial sonographic cervical length measurements.

- We suggest not performing prophylactic cervical cerclage in pregnant women based solely upon a history of cold knife conization (Grade 2C).

- For women who have undergone LEEP, we suggest not performing serial sonographic cervical length measurements. Cervical length surveillance may be reasonable in women who have had very deep LEEP excisions (>1 to 2 cm) or two or more LEEP procedures.

注：：A Grade 1B recommendation is a strong recommendation, and applies to most patients. Clinicians should follow a strong recommendation unless a clear and compelling rationale for an alternative approach is present.

   A Grade 2B recommendation is a weak recommendation; alternative approaches may be better for some patients under some circumstances.