Intracervical Laminaria Research Articles

Cervical ripening with laminaria tents prior to second trimester induction of labor

Background Induction abortion in the second trimester may be favored in some instances, such as in women with compounding medical conditions or when skilled providers are not available. Various methods of pre-induction cervical preparation have been used to shorten the length of induction and decrease the risk of complications. The benefits of cervical preparation with laminaria before D&E have been well studied, but the benefits of laminaria before medical induction are less clear. Objective To determine if overnight cervical preparation with laminaria tents shortens delivery interval in women undergoing 2nd trimester induction of labor (IOL) with misoprostol. Study Design This was a retrospective cohort study comparing overnight intracervical laminaria placement followed by misoprostol to misoprostol alone for 2nd trimester IOL between 1/2000 and 12/2010. Women were excluded if the reason for IOL was preterm labor or preterm premature rupture of membranes or if misoprostol was not used as the primary induction agent. The primary outcome was time from misoprostol administration to delivery. Results 126 women were analyzed including 36 (29%) who received laminaria + misoprostol and 90 (71%) who received misoprostol alone. Women in the laminaria + misoprostol group were significantly older (30 yrs [14–44] vs. 27 yrs [17-43], p = .029). Induction for fetal anomaly (92% vs. 34%, p ≤ .001) and the use of feticide (56% vs. 13%, p ≤ .001) were more common in the laminaria + misoprostol group. The mean time to delivery in the laminaria + misoprostol group was 6 h longer compared to the misoprostol only group; 19 ± 8 h compared to 13 ± 12hrs (p = .007). There was no difference in fetal to placental delivery time (p = .329), total misoprostol dose (p = .182), or length of hospitalization (p = .144) however, significantly more women completed abortion at 24 hrs in the misoprostol alone group (90% vs. 61%, p ≤ .001). Conclusions The use of laminaria tents for overnight cervical preparation does not expedite delivery times in patients undergoing 2nd trimester IOL with misoprostol.

Comparing the effects of vaginal misoprostol, laminaria, and extra amniotic saline infusion on cervical ripening and induction of labor.

ObjectiveThis study aimed to compare the effects of vaginal misoprostol, laminaria, and extra-amniotic saline infusion (EASI) on cervical ripening.MethodsThis randomized controlled trial was conducted on 195 women with singleton pregnancies and unripe cervices. Participants were randomly allocated to 3, 65-person groups: a misoprostol, a laminaria, and an EASI group. The interventions in the misoprostol, laminaria, and EASI groups included a single 25-µg vaginal misoprostol suppository, an intracervical laminaria, and a transcervical Foley catheter, respectively. The groups were compared with each other regarding time intervals from labor induction to labor active phase and delivery, cervical dilation, Bishop scores 6 hours after induction, delivery type, length of hospital stay, and complications.ResultsThere were no significant differences among the groups regarding maternal ages, gestational ages, body mass indices, baseline cervical dilations, and Bishop scores (P>0.05). Six hours after induction, the Bishop score and cervical dilation were significantly greater in the EASI group than in the other 2 groups (P<0.001). Moreover, time intervals from labor induction to labor active phase and delivery in the EASI group were significantly short (P<0.001). The rates of cesarean section, fetal distress, placental abruption, and meconium staining in the misoprostol group were significantly high (P<0.05), and the length of hospital stay in the EASI group was significantly short (P<0.001).ConclusionEASI is a safer and more effective method for cervical ripening. Considering its inexpensiveness, easy accessibility, and greater effectiveness, EASI is recommended for cervical ripening.Trial RegistrationIranian Center for Clinical Trials Identifier: IRCT20170513033941N39

Comparison of Effectiveness of Laminaria versus Vaginal Misoprostol for Cervical Preparation Before Operative Hysteroscopy in Women of Reproductive Age: A Prospective Randomized Trial

Study ObjectiveTo compare the effectiveness and safety of intracervical laminaria dilator versus intravaginal misoprostol administered before surgery to facilitate cervical dilation before operative hysteroscopy. DesignA prospective randomized study (Canadian Task Force classification 1). SettingA university hospital. PatientsA total of 150 women were assigned at random to the following groups: laminaria dilation (n = 50), misoprostol dilation (n = 50), and mechanical dilation (n = 50). InterventionsHysteroscopic surgery of intrauterine lesions. Measurements and Main ResultsIn this study, 150 women were assigned at random to receive cervical priming with an intracervical laminaria dilator, 200 μg of intravaginal misoprostol, or a mechanical dilator before operative hysteroscopy. Cervical response, surgical outcome, and complications of operative hysteroscopy were assessed. Visual analog scale (VAS) pain scores were recorded in the misoprostol and laminaria dilation groups. Demographic variables of the study groups were comparable (p = .278–.988). The duration of cervical pretreatment was similar with the intracervical laminaria dilator and intravaginal misoprostol (p = .803); however, intravaginal misoprostol was associated with more adverse effects (p = .031). Compared with the misoprostol dilation group, in which all patients required additional cervical dilation, notably fewer patients in the laminaria dilation group required additional cervical dilation after cervical preparation (p = .001). VAS pain scores were significantly higher in the laminaria dilation group, however (p = .001). ConclusionCervical priming with an intracervical laminaria dilator before operative hysteroscopy reduces the need for cervical dilation and better facilitates hysteroscopic surgery compared with intravaginal misoprostol. Oral analgesic use may be required before the use of this device.

Comparison of intracervical laminaria plus vaginal misoprostol and vaginal misoprostol only with variable dosages for medical abortion within 12 hours

Copyright © 2012. Korean Society of Obstetrics and Gynecology Curettage has been the method of choice for treatment of spontaneous abortion from the past. However, such a surgical treatment method can induce abortion-related complications such as bleeding, infection, and uterine damage and perforation, and increase maternal morbidity, and even in some cases it has been reported to be related to maternal death [1,2]. Medical therapy is applied in order to solve these shortcomings of the surgical method, and among drugs used in medical treatment, prostaglandin is effective for uterine contraction and cervical ripening, and especially prostaglandin E2 (PGE2) shows an outstanding effect on medical aborORIGINAL ARTICLE

Laminaria tent versus Misoprostol for cervical ripening before surgical process in missed abortion.

To compare the efficacy of Laminaria tents with Misoprostol for cervical ripening before surgical process in missed abortion. In a prospective study, 70 women with missed abortion were assigned to have either insertion of a 3 mm intracervical Laminaria tent (n = 35) or vaginal Misoprostol 400 μg (n = 35) on the day prior to suction dilation and curettage (D/C). The women were interviewed just prior to the D/C with regard to pain, vaginal bleeding, and cervical dilator preference. Cervical dilation was greater in the Laminaria group but not significantly different from that in the Misoprostol group. However, additional cervical dilation before D/C was required in more patients in the Misoprostol group (45.7 vs 14.3%, P = 0.001). Women who received Laminaria reported significantly more pain at the time of insertion (62.8% in Laminaria group vs 22.8% in Misoprostol group) compared with women who received Misoprostol. Conversely, Misoprostol was associated with more nausea, vomiting, diarrhea and vaginal bleeding. Laminaria tents are more effective cervical dilators than vaginal Misoprostol when inserted the day prior to suction D and C. Vaginal Misoprostol insertion is more comfortable, although it is associated with an increased risk of vaginal bleeding.

Septic Shock after Intracervical Laminaria Insertion

Laminaria placement is seldom thought to be associated with postabortal sepsis. A nulliparous woman presented with high fever, low blood pressure, and signs of infection during artificial legal abortion with laminaria placement for cervical dilatation. Broad-spectrum antibiotics were given. Cultures of blood, placenta, and arterial line all yielded Enterobacter cloacae. The patient responded to antibiotics and supportive care. The use of laminaria still places patients at risk for infection because there is a certain risk of ascending colonization with potentially pathogenic microorganisms from the vaginal and cervical microflora, as in our patient. Surgical disinfection, prophylactic antibiotics, and shortened duration of laminaria placement are helpful to prevent infectious insult. Once signs of infection are noted, physicians should take action as soon as possible, such as initiating broad-spectrum antibiotics and intensive care.

A Randomized Trial of Laminaria Tents Versus Vaginal Misoprostol for Cervical Ripening in First Trimester Surgical Abortion

Objective: To compare the use of laminaria tents with misoprostol for cervical ripening prior to first trimester surgical abortion.Method: In a prospective, open-label, randomized trial, 70 women were assigned to have either insertion of a 3 mm intracervical laminaria tent or vaginal misoprostol 200 µg on the day prior to suction dilatation and curettage (D&C). Cervical dilatation and operating time were measured at the time of D&C; the surgeon subjectively rated the ease of dilatation. The women were interviewed just prior to the D&C with regard to pain, vaginal bleeding, and dilator preference.Results: Laminaria produced significantly (P < 0.001) greater pre-abortion dilatation of the cervix (34.8 Pratt ± 6.2) than did misoprostol (28.4 Pratt ± 5.8). There was no demonstrable difference in ease of dilatation or operating time. Women randomized to use of laminaria reported significantly more pain on insertion than did those who received misoprostol (P < 0.001). Conversely, misoprostol was associated with more vaginal bleeding (P < 0.01). Pain following insertion was similar in each group. Two patients aborted completely after misoprostol alone. Overall, the stated patient preference for cervical dilator was more likely to be misoprostol (P < 0.01).Conclusions: Laminaria tents are more effective cervical dilators than vaginal misoprostol when inserted the day prior to suction D&C. Vaginal misoprostol insertion is more comfortable, although it is associated with an increased risk of vaginal bleeding and may abort the pregnancy. Patient preference favours misoprostol. In patient-centred care, both laminaria and misoprostol are suitable options for cervical preparation prior to suction D&C.

Second trimester abortion by laminaria followed by vaginal misoprostol or intrauterine prostaglandin F2α: a randomized trial

Intra-amniotic injection, as well as intravaginal application of prostaglandins, have been used to terminate second trimester pregnancies. There is as yet no consensus as to the most efficient protocol of such late abortions. Our goal was to compare the efficacy of intra-amniotic injection of prostaglandin F2 alpha (PGF2α) and intravaginal application of misoprostol in terminating second trimester pregnancies after pretreatment with intracervical laminaria. Women with live fetuses and requesting second trimester abortions were randomized into two groups. Eighteen hours following the insertion of intracervical laminaria, women were treated with either intra-amniotic injection of 40 mg PGF2α, or 12 hourly doses (to a maximum of 4 doses) of 200 mcg misoprostol. Fifty women were randomly assigned to each group. Failure to abort within 24 h of initiation of treatment occurred in 6 patients (12%) in the misoprostol group and 14 (28%) of the PGF2α group (p = 0.04). Mean time of induction of pharmacologic treatment to abortion was 13.6 h in the misoprostol group and 10.7 h in the PGF2α group (p = 0.03). The mean number of analgesic injections given were 0.8 in the misoprostol group and 1.6 in the PGF2α group (p = 0.0001). Only the method of abortion was predictive of abortion success and not other variables such as patient age, gestational age, gravidity, or parity. Following intracervical laminaria, vaginal misoprostol has been found to be more effective and less painful, compared with intra-amniotic PGF2α, for the termination of second trimester pregnancies with live fetuses.

Intraamniotic 15(s)-15-methyl prostaglandin F 2α and termination of middle and late second-trimester pregnancy for genetic indications: A contemporary approach

Our purpose was to determine whether intraamniotic 15(s)-15-methyl prostaglandin F2 alpha could serve effectively and safely as a surrogate for natural prostaglandin F2 alpha when used to effect middle and late second-trimester pregnancy termination for genetic indications. Further, we sought to compare efficacy and side effects with those found when intravaginal prostaglandin E2 is used. A total of 62 patients received intraamniotic 15(s)-15-methyl prostaglandin F2 alpha and urea, and intracervical laminaria. Case patients at < or = 20 weeks' gestation (n = 32) were compared with case patients at > 20 weeks (n = 30) and with a matched group (n = 64; 1:2 ratio) of control patients in whom cervical laminaria and intraamniotic urea were used with prostaglandin E2 vaginal suppositories. The mean induction-to-abortion interval among the case patients (gestational age 16 to 27 weeks) was 13 hours 11 minutes; 60 of 62 (97%) were delivered within 24 hours. There was a statistically significant negative correlation between the induction-to-abortion interval and gestational age (p = 0.04). When patients at < or = 20 weeks and those at > 20 weeks were compared, few differences were noted. The mean induction-to-abortion interval for case patients at < or = 20 weeks was 13 hours 54 minutes versus 19 hours 34 minutes for control patients (p = 0.001). One of 32 (3%) case patients remained undelivered beyond 24 hours compared with 17 of 64 (27%) control patients (p < 0.01). Immediate and delayed complications were uncommon in either group. Our study demonstrates that 15(s)-15-methyl prostaglandin F2 alpha can serve safely as a surrogate for prostaglandin F2 alpha when used in combination with urea and laminaria for termination of pregnancy. This technique appears safe for use through 27 weeks' gestation; further investigation is encouraged.

Late midtrimester medical pregnancy terminations: Three different procedurew with prostaglandin F 2α and laminaria tents

One hundred twenty eight women underwent midtrimester induced abortion with: 1)combined regimen of intramniotic prostaglandin (PG) F2a injection and intracervical laminaria tents (group A, 50 women), 2)intramniotic PGF2a injection only(group B, 51 women) and 3) laminaria tents followed by intracervical PGF2a tablets insertion (group C, 27 women). The mean induction-abortion time (±SE) was 24.9±1.7 hours for group A, 28.2±2.2 hours for group B (p > 0.05) and 42.1±3.4 hours for group C, significantly longer than group A and B (p<0.001 and p<0.01 respectively). In 48 hours 98% of the patients of group A, 90% of group B (p<0.05) and 59% of group C (p<0.001) completed the abortion procedure. Parous women of group A and B presented similar induction - abortion time, while nulliparous the use of laminaria shorted the abortion procedure significantly (p<0.05). The complications rate was low. In conclusion, the intracervical PGF2a insertion is a simple but very slow abortion procedure with high failure rates. The intramniotic PGF2a injection is a successfull method for late midtrimester medical pregnancy termination and the concurrent use of laminaria tents shortens the abortion procedure, particularly in nulliparous, reduces the number of prostaglandins' reinjections and increases the incidence of successfull abortion within 48 hours.

Midtrimester pregnancy termination by intravaginal administration of prostaglandin E 2

The intravaginal administration of prostaglandin E 2 (PGE 2) in repeated doses of 20 mg was evaluated for efficacy and safety in 400 midtrimester pregnancy terminations. Patients were assigned to three treatment groups. Group A (232 patients) received PGE 2 alone, Group B (106 patients) had prior intracervical laminaria insertion, Group C (62 patients) received both laminaria and continuous intravenous oxytocin infusion. Complete abortion was achieved in 328 cases (82 per cent) and there were 13 failures (3.3 per cent). Mean instillation to abortion times was 13.0 ± 0.4 SEM hours. Differences among the treatment groups were not statistically significant but outcome was more favorable in pregnancies of 16 to 19 weeks' gestation as compared with those between 13 and 15 weeks. Gastrointestinal symptoms and temperature elevation constituted the most frequent side effects. Excessive blood loss occurred in 11 cases (2.8 per cent) and live births were encountered in 12 cases (3.0 per cent). Vaginal PGE 2 is felt to be particularly acceptable for the termination of pregnancies of 16 to 19 weeks' gestation.

Synthetic laminaria for cervical dilatation prior to vacuum aspiration in midtrimester pregnancy

Potential benefits of a synthetic tent compared to the natural laminaria japonica would be: 1) uniformly sized and shaped, 2) uniformity of dilatation characteristics, 3) assurance of sterility, 4) less expense, 5) ability to control rate and amount of dilatation, and 6) ability to control shape. 20 healthy women, from 13-17 weeks gestation desiring elective abortion had 1 intracervical laminaria tent placed in the cervix after cervical dilatation was determined. The following day the tent was removed and the amount of dilatation again determined. Abortion was accomplished with dilatation and evacuation; all received oral tetracycline, 250 mg 4 times/day for 4 days. The laminaria tents were highly effective: mean cervical dilatation was 1.4 mm before insertion and 12.8 mm after the tents were removed; with no apparent differences in dilatation among the 8 nulliparous patients compared to the 12 multiparous patients. Further dilatation in 11 women was accomplished with minimal pressure. Several had mild discomfort during laminaria insertion but all attempts were successful without anesthesia. The tents came out of the cervix spontaneously in only 1 patient. There were no cervical tears, false passages, uterine infections, or retained products of conception. Use of synthetic laminaria tents appears practicable for the following reasons: 1) clinically significant dilatation and softening occurred with overnight use, 2) side effects were minimal, 3) no visible tissue reaction occurred, 4) relatively short use (less than 24 hours) lessens possibilities for infection, and 5) the small dose of magnesium sulfate appears safe. Further studies are necessary to determine the incidence of complications, to research use in patients needing cervical dilatation for other purposes, to evaluate laminaria tents with other concentrations of magnesium sulfate and different amounts of compression.

Midtrimester abortion patients

Linda Dickey White, RN Paul F White, MD roviding nursing care for patients undergoing midtrimester P abortion by dilatation and evacuation (D and E) can be a difficult decision for many nurses. Over 50,000 abortions are performed each year by the D and E meth0d.l At the University of California, San Francisco, Medical Center six to ten midtrimester abortions are performed by D and E each week on an outpatient basis. An important aspect in the nursing care of these patients is the attitude of the staff toward abortion and the patient population seeking midtrimester abortions. Practicing perioperative care has been helpful in establishing rapport with these patients. To gain a better understanding of this patient population, we conducted a study to measure their anxiety levels using the Institute for Personality and Ability Testing (IPAT) trait anxiety scale and to measure mood states with the Profile of Mood States (POMS). The purpose of the study was to identify common characteristics among the patients and acquaint the staff with emotional needs of this patient population. Midtrimester abortion by D and E offers women an alternative to prostaglandin (amnio) procedures. Amnio abortions consist of placement of intracervical laminaria followed by an intraamniotic prostaglandin E injection. Laminaria are thin pieces of desiccated seaweed that cause cervical dilatation as they absorb moisture from the body and expand. The prostaglandin stimulates uterine contractions, which induce labor and fetal expulsion. The D and E method requires insertion of laminaria 12 to 24 hours prior to surgery in the physician’s office or clinic. After induction of anesthesia, the patient is placed in the lithotomy position. The physician performs a bimanual exam and removes the laminaria. After sterile preparation of the patient, the

Intra-Amniotic Prostaglandin F2α and Urea for Midtrimester Abortion

Intra-amniotic prostaglandin F2alpha, in doses of 2.5 to 20 mg, combined with 80 gm of urea, was an effective, safe, simple, and economical midtrimester abortifacient in 115 patients. The mean abortal time, 15.8 hours, was significantly less than that in prior series in which intra-amniotic hypertonic saline or urea was administered together with constant, intravenous oxytocin infusion. The use of intracervical laminaria tents did not shorten the abortal time. Only eight patients had not aborted within 30 hours; only two had not aborted within 36 hours; and only six received a second intra-amniotic injection. Operative removal of the placenta, when necessary, was accomplished under intravenous sedation in a treatment room. The incidence of infection, nausea, vomiting, and other complications was low.

Induction of Midtrimester Abortion by the Combined Method of Continuous Extra Vovular Infusion of Prostaglandin F2α and Intracervical Laminaria Tents

Midtrimester abortion was successfully induced in 55 of 60 patients with continuous extraovular infusion of prostaglandin F2alpha (PGF2alpha) following the insertion of intracervical laminaria tents. Intravenous oxytocin was also used in 38 (63%) of the 60 patients. The mean induction-abortion time (IAT) was 11.72 hours +/- 1.06 SD). Abortion was completed in 40% within 8 hours, 80% within 16 hours, and 93% within 24 hours. The mean total dose of PGF2alpha was 41.9 mg. There was no significant difference in IAT between the parous patients (13.40 hours +/- 1.90 SD) and the nulliparous patients (10.41 hours +/- 1.13 SD). There was no apparent correlation between IAT and the stages of gestation (12 to 22 weeks). The five patients who failed to abort within 24 to 36 hours underwent uterine evacuation, which was easily accomplished because there was a marked degree of cervical dilatation. Side effects and complications of the technique were few. Endometritis occurred in three patients, two of whom had had intrauterine devices in situ until just prior to the procedure. It appears that this method has a high success rate, an acceptable safety factor, good patient tolerance, and relatively few side effects.

Evaluation of Intramuscular 15(S)-15-Methyl Prostaglandin F2α Tromethamine Salt for Induction of Abortion, Medications to Attenuate side effects, and Intracervical Laminaria Tents

Sixty gravidas 8 to 20 menstrual weeks' gestation were studied to evaluate (1) the efficacy of intramuscularly administered 15(S)-15-methyl prostaglandin F2alpha tromethamine (15(S)-Me-PGF2alpha) as an abortifacient; (2) the effectiveness of prochlorperazine and Lomotil for attenuation of vomiting and diarrhea; and (3) the practicability of augmenting this prostaglandin dose schedule with intracervical laminaria tents. Group I subjects received 250 mug of 15(S)-Me-PGF2alpha intramuscularly every 2 hours for the initial 24 hours and 500 mug for the next 24-hour period. Group II received the same dose schedule of prostaglandin and prearranged doses of prochlorperazine and Lomotil. Group III received the same dose schedule of prostaglandin after intracervical laminaria tents had been inserted, and prochlorperazine and Lomotil were administered by the prearranged dose schedule. It appears that (1) the 15(S)-Me-PGF2alpha was effective in inducing abortion; (2) a significant decrease in body temperature occurred; (3) the abortifacient effectiveness of this prostaglandin dose schedule was not altered by the regimen of prochlorperazine and Lomotil; (4) only diarrhea was significantly attenuated with the regimen of prochlorperazine and Lomotil; and (5) laminaria augmentation was not useful.

Combination therapy for midtrimester abortion: Laminaria and analogues of prostaglandins

Thirty-four women 13–20 weeks pregnant were aborted by IM 15(S)-15 methyl prostaglandin E 2, 10 mcg every four hours. Thirteen women had pretreatment with intracervical laminaria tents. The use of laminaria tents produced a highly significant reduction in time to abortion without an increase in complications. A reduction in the dose of methyl PGE 2 required for abortion was also noted. Other investigators have already reported similar improved results with combinations of laminaria and analogues of prostaglandin and they deserve a wider clinical trial.

First trimester abortion using intramuscular (15 S) 15 methyl prostaglandin F 2α

The efficacy of intramuscular (15 S) 15 methyl prostaglandin F 2α (15-Me-PGF 2α) as an abortifacient was evaluated in forty women between six and twelve weeks of gestation. 250 μg of 15-Me-PGF 2α was given intramuscularly at two-hour intervals to these patients. All the patients aborted within twenty-four hours and the abortion was complete in 82.5 percent of the cases. The injection-abortion interval was significantly lower in the twenty patients who had intracervical laminaria insertion prior to treatment with 15-Me-PGF 2α than in the twenty women without laminaria insertions. The frequency and type of side effects are discussed in both groups. The use of intramuscular 15-Me-PGF 2α is also compared with suction curettage for first trimester abortion. This study demonstrates the availability of the intramuscular route as a simple, safe, and reliable means of terminating first trimester pregnancies.

Prostaglandin E2 vaginal suppository as a midtrimester abortifacient

The efficacy of prostaglandin E2 (PGE2) vaginal suppositories as an abortifacient was evaluated in 40 women between 16 and 20 weeks' gestation. All pregnancies were successfully terminated with a mean time of 12.7 hours (S.D. 5.7). Eight patients (20 per cent) were treated with oxytocin augmentation after spontaneous rupture of the membranes during PGE2 vaginal suppository therapy. Gastrointestinal side effects and drug fever were frequent but self-limiting. The success of this approach eliminates amniocentesis for drug instillation as well as the supplementary use of intravenous oxytocin in high concentrations and/or intracervical laminaria.

Intra-amniotic prostaglandin PGF 2a (tham salt) and the laminarta tent in midtrimester termination of pregnancy

Twenty patients underwent intra-amniotic administration of PGF 2a for termination of pregnancies ranging from 14–70 weeks gestation. Success was achieved in 18 patients (90%) with this method. The two failures were terminated vaginally with adjunctive use of intracervical laminaria and IV pitocin induction. Ten of the patients had intracervical laminaria placed 4–24 hours prior to drug administration. This latter group was noted to have a shorter induction-abortion (I-A) interval, shorter onset of uterine contractions, and a decreased incidence of vaginal bleeding prior to abortion. No serious side effects occurred in either group of patients. Clinical toxicity in the form of gastrointestinal symptoms such as nausea and vomiting were common but could generally be controlled with antiemetics. No significant changes were recorded in blood counts, platelet counts, liver function tests, and renal function tests from pretreatment values and at 24 and 48 hours post-treatment. There were no instances of hemorrhage, fever, or infection which occur as complications of hypertonic saline abortion.