Best practices in surgical abortion

Best practices in surgical abortion

CLINICAL OPINION Best practices in surgical abortion Lisa M. Keder, MD Columbus, Ohio Surgical abortion in the first trimester comprises the majority ...

92KB Sizes 5 Downloads 104 Views


Best practices in surgical abortion Lisa M. Keder, MD Columbus, Ohio Surgical abortion in the first trimester comprises the majority of voluntary pregnancy interruptions performed in the United States. The majority of these procedures are done in outpatient settings with the patient under local anesthesia. Appropriate volume of and deep injection of local anesthetic can reduce pain associated with the procedure. Waiting between administration of the paracervical block and initiating the procedure does not affect pain. Intravenous administration of sedation and analgesia improves patient satisfaction but does not significantly affect pain ratings. Antibiotic prophylaxis is warranted. Vasopressin is useful for prevention of hematometra and hemorrhage. Less evidence supports the routine use of ergots. Preoperative cervical priming reduces the risk of cervical injury and uterine perforation. Attention to operative technique can reduce the risk of incomplete abortion. Routine postoperative care at 2 or 3 weeks is timed to identify complications and to reinforce pregnancy and sexually transmitted disease prevention. (Am J Obstet Gynecol 2003;189:418-22.)

Key words: Induced abortion, paracervical block, abortion complications

The recent literature regarding voluntary pregnancy interruption is dominated with articles concerning medical abortion. This represents an appropriate focus since the US introduction of mifepristone (RU 486) and its approval by the Food and Drug Administration (FDA). However, the majority of abortions performed in the United States are done by vacuum aspiration. Abortion by suction curettage remains one of the most frequently performed surgical procedures in the United States. Approximately 1.3 million abortions occur per year in this country. Eighty-eight percent of these are done in the first 12 weeks of pregnancy and 50% before 8 weeks. Although no firm data exist with regard to the percentage of early abortions that are currently performed medically versus surgically, personal experience suggests that the majority are surgical. Research from other countries indicates that, when medical abortion is available, approximately half of eligible women will choose it. However, mifepristone is currently FDA approved for abortion up to only 49 days’ gestation. Therefore, the majority of abortions, whether by reasons of personal

From the Department of Obstetrics and Gynecology, Ohio State University. Received for publication October 25, 2002; revised December 17, 2002; accepted January 9, 2003. Reprints not available from the author. Ó 2003, Mosby, Inc. All rights reserved. 0002-9378/2003 $30.00 + 0 doi:10.1067/S0002-9378(03)00107-8


choice or gestational age, will continue to be performed by vacuum aspiration. Moreover, suction curettage is also a frequently performed procedure for management of blighted ova or incomplete abortions. Information learned with regard to elective abortion may also be applicable to the procedure performed for these indications. Therefore, I will be reviewing the techniques of first-trimester surgical aspiration with regard to pain control, prevention of complications, and postabortion care to recommend ‘‘best practices’’ on the basis of the available literature. Medical versus surgical abortion A woman’s choice between surgical and medical abortions should be made with thorough counseling regarding the risks and benefits of the two procedures. In some cases, the choice will be limited by availability of the options or by feasibility of follow-up for women living remote from abortion providers. In some centers surgical abortion is restricted to gestations >7 weeks, making medical abortion the only option for very early gestations. Other centers offer early surgical abortion by vacuum aspiration or by manual vacuum aspiration. In early research addressing failure of surgical abortion, the Centers for Disease Control and Prevention (CDC) found that abortion performed at less than or equal to 6 weeks of gestation was associated with an ongoing pregnancy rate of 5.6 per 1000, 2.9 times higher than for abortion performed at greater than 6 weeks.1 Subsequently, a series

Volume 189, Number 2 Am J Obstet Gynecol

evaluating manual vacuum aspiration (with routine postoperative ultrasonographic and tissue examination) found a failure rate of 1.3 per 1000 procedures.2 Although surgical abortion, especially in early gestation, has a failure rate, the failure rate of medical abortion is generally felt to be higher. However, failure of medical abortion is generally defined differently than for surgical abortions: any need for surgical intervention, not just surgical aspiration for ongoing pregnancy, is defined as failure. Currently, many centers are using a combination of 200 mg of oral mifepristone in combination with 800 lg of misoprostol administered vaginally. This regimen has an ongoing pregnancy rate of only 0.2% to 0.3% and a surgical intervention rate of 1.5% to 4% for 49 or fewer days of gestation. The efficacy of this regimen decreases with advancing gestational age.3 The acceptability of medical abortion has been evaluated in many international trials. In these studies, 84% to 97% of women would choose medical abortion again.3 A recent trial evaluated experience in the United States. It found that both medical and surgical abortion were highly acceptable, with 97% of those having had a medical abortion and 93% of those having a surgical abortion stating they would recommend that technique to a friend. In this study more women who had medical abortion stated they would select it in the future.4 Studies examining acceptability of medical abortion have found that factors associated with dissatisfaction include the duration of bleeding, pain, number of clinic visits, and uncertainty regarding the success of the process. Patients preferring surgical abortion often cite the ability to complete the abortion in 1 day and the certainty of completion of the abortion as reasons for choosing that option. Additionally, women choosing surgical abortion have less stated anxiety regarding the surgical procedure. Pain control The majority of abortions in the United States (approximately 70%) are performed in abortion clinics. Fewer than 10% occur in hospitals. Physicians’ offices and other clinics represent the other 20%. Most elective abortions outside the hospital setting are performed under local anesthesia with or without the use of intravenous or oral sedation. In fact, general anesthesia for induced abortion has been associated with an increased risk of death.5 The experience of pain during an elective abortion cannot be separated from the anxiety and trepidation that the woman feels before the procedure. Adequate preoperative counseling regarding expectations during and after the abortion is helpful to alleviate anxiety. The availability of trained counselors or volunteers to accompany women through the experience as well as a safe, comfortable physical environment are also important. In a study examining women’s self-reports of pain during

Keder 419

first-trimester abortion under local anesthesia, 46% of women described moderate pain, 32% severe pain, and 2% very severe pain. Younger women, mid first-trimester gestational age, and preprocedure level of fearfulness were all predictive of the degree of pain. Pain was also associated with the physician performing the abortion.6 One study examined resident versus attending physicians and found that more experienced physicians were associated with less pain during cervical dilation, but not during aspiration.7 Anecdotally, recovery room personnel note marked differences between providers both in intraoperative and postoperative pain. The specific technique of the administration of anesthetic, the skill of the operator, and the duration of the procedure all affect pain. Pain associated with the procedure occurs with injection of the paracervical block, cervical dilatation, suction aspiration, and sharp curettage. Paracervical block is effective in alleviating pain associated with application of a tenaculum to the cervix, and with cervical dilatation, but offers limited relief for the pain associated with suction and curettage. Lidocaine is the most commonly used anesthetic. It is both quick acting and inexpensive. The maximum recommended dose is 4.5 mg/kg of body weight. Twenty milliliters of a 1% solution provides 200 mg of lidocaine, close to the maximum dose for a 50-kg woman. The choice of local anesthetic for the paracervical block does not change effectiveness.8 However, the volume instilled does seem to matter: a larger volume results in better pain relief.9 One study found that bacteriostatic lidocaine was no better than bacteriostatic saline solution.10 Subsequent research using chloroprocaine has failed to confirm this.11 However, it is possible that some of the anesthetic effect of the paracervical block is related to tissue distention rather than to nerve block. A number of different injection techniques have been described for paracervical block. Deep injection appears to provide better anesthesia than does superficial injection. However, injection techniques vary markedly among research protocols. Glick first described the technique of deep injection that involves seven points of injection. This technique was found to be more effective than superficial injection by Wiebe.12 Recently, a published study failed to find any difference using two versus four injection sites.10 A slower rate of injection has been demonstrated to decrease pain associated with administration of the block.13 However, waiting between injection and dilatation does not seem to improve efficacy of the paracervical block.12,14 In my experience, a more generous administration of local anesthetic to the anterior cervical lip before placement of the tenaculum decreases the pain associated with the block itself. Some authors advocate addition of sodium bicarbonate to buffer the lidocaine solution. Others add vasopressin to the solution. This decreases the amount of bleeding during evacuation

420 Keder

and may decrease the need for reaspiration.15 I have found that it decreases atony and hematometra in pregnancy termination done after 9 weeks, especially in women with retroverted uteri. Combining oral or intravenous conscious sedation with paracervical block offers additional options for women in the outpatient setting. Oral diazepam is commonly offered preoperatively. However, there is little evidence that it provides analgesic effects. In their study evaluating the technique of paracervical block, Phair et al14 found that patients who elected to receive diazepam did not report less pain associated with dilatation, aspiration, or immediate postoperative recovery. However, in this study patients were not randomized to use of diazepam. It is possible that patients who chose diazepam would have had greater anxiety and been more likely to report higher levels of pain. Three studies have examined the use of parenteral medication. Jakobbson et al16 used morphine with scopolamine, meperidine, or midazolam intramuscularly before vacuum curettage with paracervical block. They found no significant differences between groups in their self-reported pain or in recovery room personnel observation of pain. Rawling and Wiebe17 evaluated the effect of combining fentanyl with oral lorazepam and cervical block. They found that administration of 50 to 100 mcg of fentanyl lowered pain scores by 1 point on an 11-point scale. Although this was statistically significant, they question its clinical significance. Recently a randomized, double-blind, controlled trial examined the use of conscious sedation with paracervical block.18 Two milligrams of midazolam and 25 lg of fentanyl were given intravenously, followed in 2 minutes by the paracervical block. Patients rated their pain on a 10-point scale. Interestingly, their mean pain scores did not differ significantly for pain during the procedure or at 1 hour after the procedure. However, those receiving active medication were much more satisfied with their experience. It should be noted that all patients in this study received preoperative misoprostol for cervical ripening and that the dose of fentanyl administered is lower than that used in many centers. I have found that a combination of 3 or 4 mg of midazolam with 50 to 100 lg of fentanyl provides good analgesia if administered immediately before initiation of the procedure. This is especially useful in patients who have high levels of anxiety and it often permits successful outpatient surgery for a patient who could not tolerate the procedure under local anesthetic. Propofol in low doses is also an effective sedating agent for outpatient surgery. However, it has the potential for significant respiratory sedation and is expensive. Prevention of complications Antibiotic prophylaxis for elective abortion was reviewed by Sawaya et al in 1996.19 They performed a metaanalysis of published trials and concluded that routine use

August 2003 Am J Obstet Gynecol

of antibiotics could prevent up to one half of postabortion infections. However, this study included many different antibiotics given in differing doses and for varying durations. Thus, it was impossible for the authors to offer a firm conclusion on the optimal regimen. However, they suggest that the tetracyclines and the nitroimidazoles (including metronidazole) appear to offer comparable efficacy. Some authors have argued that antibiotic prophylaxis should be administered on the basis of risk profile.20 Age less than 20 years, nulligravidity, nulliparity, multiple sexual partners within the year, previous pelvic inflammatory disease or gonorrhea, currently untreated gonorrhea, chlamydia, or bacterial vaginosis are recognized risk factors for infection.19 In practice, preoperative screening for cervical infection is not always feasible. It therefore seems reasonable to adopt a strategy of universal prophylaxis. If bacterial vaginosis or a sexually transmitted disease is identified preoperatively, treatment should be administered according to CDC sexually transmitted disease treatment guidelines. Bleeding complicating first-trimester abortion is estimated to occur in 1% to 2% of cases. Risk factors for hemorrhage include advancing gestational age (>10 weeks), advanced maternal age and parity, general anesthesia, and uterine fibroids.21 Bleeding most frequently occurs as a result of atony, but it can also be caused by cervical laceration, uterine perforation, incomplete abortion, placental implantation abnormalities, or coagulopathy. Although ergot alkaloids are commonly administered to decrease bleeding in early suction curettage, there is little evidence to support their use in early first-trimester abortion. Woodard22 studied the use of methylergonovine maleate in combination with bupivicaine with epinephrine for patients undergoing midtrimester abortion and found that it made no difference in blood loss for patients at less than 17 weeks’ gestation. Schulz et al15 found that administration of dilute vasopressin in the paracervical block is associated with decreased blood loss in pregnancy termination beyond 15 weeks. I have found that intracervical vasopressin is useful in preventing atony in abortions performed at more than 10 weeks’ gestation but that methylergonovine maleate given on a prophylactic basis offers little. Patients given methylergonovine maleate for a few days postoperatively often will complain of leg cramps and will have a temporary increase in bleeding at the time of its discontinuation. Moreover, there are reported cases of myocardial infarction associated with postabortion use of ergot agents.23 Without firm evidence of efficacy for prophylactic use, it seems prudent to limit the use of ergots to those cases at higher risk of atony or to the treatment of atony with attention to any preexisting medical conditions that might contraindicate their use. Intramuscular postoperative ergonovine has been shown in one observational study to decrease the risk of

Keder 421

Volume 189, Number 2 Am J Obstet Gynecol

hematometra.21 Vasopressin has been used for the same purpose. In my experience, vasopressin given as part of the cervical block is more effective with fewer side effects than methylergonovine maleate in preventing hematometra. Prevention of hemorrhage related to cervical laceration, uterine perforation, and incomplete abortion is largely a matter of operative technique. In the first trimester, injury to the cervix is most likely to occur from the tenaculum. Osmotic dilators, prostaglandins, or mifepristone are all effective for cervical priming. Often the ability to use these agents will depend on logistics for the individual woman in making return visits. Intraoperatively, cervical injury can be minimized by including an adequate amount of cervical tissue within the tenaculum’s grasp, using firm but gentle pressure while dilating, and not skipping dilators if resistance is encountered. The risk of uterine perforation is related to operator experience and may be increased by general anesthesia. It can be decreased by preoperative use of cervical dilators. Continuous traction of the tenaculum during dilation may help to straighten the angle between the cervix and uterine cavity and reduce the risk of anterior or posterior perforation in a markedly anteflexed or retroflexed uterus. Uterine sounding is not advocated as part of the abortion technique. Additionally, ultrasound guidance of dilation and aspiration should be performed if any difficulty is encountered during the procedure, for women with uterine anomalies, or when the uterus is distorted by fibroids or previous surgery. Many of the same factors that predispose to perforation also make incomplete evacuation more likely: uterine anteflexion or retroflexion, abnormal cavity, or advancing gestational age. Therefore, prevention of incomplete aspiration is also dependent on operator experience and technique. Selection of an appropriately sized cannula is important. Generally, a cannula that has a diameter equal to the gestational age in weeks will be adequate for aspiration in the first trimester. Depending on the ease of dilation, the operator may choose a slightly larger or smaller cannula. However, if a smaller cannula is used, the procedure may take longer, and it may be difficult to aspirate the placental tissue without removing the cannula from the uterus to clear the tubing or extracting tissue form the cervix. At the beginning of aspiration, the cannula should be inserted to the fundus and then withdrawn slightly so that it sits in the midfundus. A gentle rotary motion with an in-and-out motion should be used. The operator should try to minimize the number of times the cannula is removed from the cervix, to minimize the potential for introducing bacteria into the uterine cavity. Experienced physicians should be able to identify the components of the products of conception as they pass through the tubing. Suction should be continued until no

further tissue is returned. Many operators then explore the uterine cavity with a sharp curette and complete the procedure with a final pass of the suction cannula to remove dislodged tissue and blood. Inspection of the uterine aspirate is essential in recognizing incomplete or failed abortion. In early pregnancies, the gestational sac should be identified. In later gestation, the aspirate should be inspected for both fetal tissue and the gestational sac. If there is a question of incomplete aspiration, immediate ultrasound examination should be performed and reaspiration if necessary. Postoperative care Immediate postoperative care is dependent on the type of anesthetic given. Patients undergoing the procedure under local anesthetic alone generally require only a short period of observation. Those having intravenous sedation or general anesthetic need to be observed as dictated by their level of sedation. Most patients will have a short period of intense cramping immediately after evacuation that improves within 20 minutes. Continuing severe pain after this time warrants further evaluation for hematometra or uterine perforation. For patients who are Rh negative, anti-D immunoglobulin prophylaxis should be administered. Nonsteroidal anti-inflammatory agents (NSAIDs) provide adequate pain relief for most patients. Alternatively, if NSAIDs are contraindicated, acetaminophen alone or in combination with a narcotic can be used. Patients should be counseled preoperatively regarding contraceptive methods and instructed in appropriate initiation of their chosen method. Most clinicians have patients return for a visit 2 to 3 weeks postoperatively. At this visit any problems encountered since the procedure are discussed, the contraceptive choice reviewed, and bimanual examination of the uterus performed. Most complications associated with the procedure will present within the first 2 weeks; however, the clinician should be alert to signs of infection, abnormal postoperative bleeding, symptoms of continuing pregnancy, and evidence of emotional distress. Comment Suction curettage as performed for elective abortion or management of a nonviable pregnancy is a safe surgical procedure. A ‘‘best practices’’ approach will minimize the likelihood of complication. In the case of surgical abortion, routine antibiotic prophylaxis is warranted. Vasopressin may decrease the chance of hematometra and blood loss associated with later abortion. Routine use of ergots may decrease the risk of hematometra in at-risk patients but is not routinely useful for hemorrhage prevention and has a risk of serious side effects. Careful attention to surgical technique can minimize the risk of cervical injury and uterine perforation. Preoperative identification of uterine abnormalities with intraoperative use

422 Keder

of ultrasound guidance for difficult procedures can decrease the risk of incomplete abortion and uterine injury. REFERENCES 1. Kaunitz A, Rovira EZ, Grimes D. Abortions that fail. Obstet Gynecol 1985;66:533-7. 2. Edwards J, Creinin MD. Surgical abortion for gestations of less than 6 weeks. Curr Probl Obstet Gynecol Fertil 1997;20:11-9. 3. Newhall EP, Winikoff F. Abortion with mifepristone and misoprostol: regimens, efficacy, acceptability and future directions. Am J Obstet Gynecol 2000;183(Suppl):S44-53. 4. Jensen JT, Harvey SM, Beckman LJ. Acceptability of suction curettage and mifepristone abortion in the United States: a prospective comparison study. Am J Obstet Gynecol 2000;182:1292-9. 5. Peterson HB, Grimes DA, Cates W Jr, Rubin GL. Comparative risk of death from induced abortion at less than or equal to 12 weeks gestation performed with local versus general anesthesia. Am J Obstet Gynecol 1981;141:763-8. 6. Smith GM, Stubblefield PG, Chirchirillo L, McCarthy MJ. Pain of first trimester abortion: its quantification and relations with other variables. Am J Obstet Gynecol 1979;133:489-98. 7. Edelman A, Nichols MD, Jensen J. Comparison of pain and time of procedures with two first-trimester abortion techniques performed by residents and faculty. Am J Obstet Gynecol 2001;184:1564-7. 8. Malzer DS, Malzer MC, Wiebe ER, Halvorson-Boyd G, Boyd C. Pain management. In: Paul M, Lichtenberg ES, Borgatta L, Grimes DA, Stubblefield PB, editors. A clinician’s guide to medical and surgical abortion. Philadelphia: Churchill Livingstone; 1999. p. 73-89. 9. Stubblefield PG. Control of pain for women undergoing abortion. Int J Gynecol Obstet 1989;S3:131-40. 10. Miller L, Jensen MP, Stenchever MA. A double-blind randomized comparison of lidocaine and saline for cervical anesthesia. Obstet Gynecol 1996;87:600-4. 11. Glantz JC, Shomento S. Comparison of paracervical block techniques during first trimester pregnancy termination. Int J Gynaecol Obstet 2001;72:171-8.

August 2003 Am J Obstet Gynecol

12. Wiebe ER. Comparison of the efficacy of different local anesthetics and techniques of local anesthesia in therapeutic abortions. Am J Obstet Gynecol 1992;167:131-4. 13. Wiebe ER, Rawling M. Pain control in abortion. Int J Gynaecol Obstet 1995;50:41-6. 14. Phair N, Jensen JT, Nichols MD. Paracervical block and elective abortion: the effect on pain of waiting between injection and procedure. Am J Obstet Gynecol 2002;186:1304-7. 15. Schulz KF, Grimes DA, Christensen DD. Vasopressin reduces blood loss from second-trimester dilatation and evacuation abortion. Lancet 1985;2:353-6. 16. Jakobbson J, Andreen M, Westgren M, Thomasson K. Discomfort after outpatient abortion using paracervical block: a comparison between two opiods and one non-opiod drug for premedication. Gynecol Obstet Invest 1990;30:71-4. 17. Rawling MJ, Wiebe ER. A randomized controlled trial of fentanyl for abortion pain. Am J Obstet Gynecol 2001;185:103-7. 18. Wong CY, Ng EH, Ngai SW, Ho PC. A randomized, double blind, placebo-controlled study to investigate the use of conscious sedation in conjunction with paracervical block for reducing pain in termination of first trimester pregnancy by suction evacuation. Human Reprod 2002;17:1222-5. 19. Sawaya G, Grady D, Kerlikowske K, Grimes DA. Antibiotics at the time of induced abortion: the case for universal prophylaxis based on a meta-analysis. Obstet Gynecol 1996;87:884-90. 20. Hemsell DL. Prophylactic antibiotics in gynecologic and obstetric surgery. Rev Infect Dis 1991;13:S821-41. 21. Lichtenberg ES, Grimes DA, Paul M. Abortion complications: prevention and management. In: Paul M, Lichtenberg ES, Borgatta L, Grimes DA, Stubblefield PB, editors. A clinician’s guide to medical and surgical abortion. Philadelphia: Churchill Livingstone; 1999. p. 197-216. 22. Woodward G. Intraoperative blood loss in midtrimester dilatation and extraction. Obstet Gynecol 1983;62:69-72. 23. Fujiwara Y, Yamanaka O, Nakamura T, Yokoi H, Yamaguchi H. Acute myocardial infarction induced by ergonovine administration for artificially induced abortion. Jpn Heart J 1993;34: 803-8.