According to an American College of Obstetricians and Gynecologists (ACOG) committee opinion in 2009, all health care providers must provide patients with accurate and unbiased information and refer patients to other providers in a timely manner if they are unable to themselves provide the requested services.⁸ Similarly, women presenting with an unintended pregnancy must be presented with accurate and unbiased information regarding her options. Table 13.1 suggests the components to be discussed with the patient.⁹

Equally important in preprocedural counseling is an assessment of the mental health of the patient. As former Surgeon General C. Everett Koop testified before Congress, the development of significant psychological
problems related to abortion is “miniscule from a public health perspective.”¹⁰ Although poor psychological outcomes postabortion are uncommon, especially in those obtaining first-trimester abortions, women who may be at increased risk of such outcomes should be identified during preprocedural counseling and should be followed appropriately both during and after the procedure.¹⁰ This includes women who lack social support for their abortion, who have more conflicting feelings regarding their decision, those terminating a highly desired pregnancy, and those with pre-existing psychological conditions, which may make it difficult for them to cope with life’s stresses. The counselor should attempt to assess the patient’s current social, financial, and cultural situation; the patient’s understanding of her alternatives; any pressure she perceives on her decision-making process; her expectations of her feelings of her decision in the future; any history impacting her current decision; the circumstances surrounding the pregnancy; and her understanding of the impact her decision will have on her future and current situation.¹¹ A patient who remains undecided at the completion of her counseling session can be referred to other services (e.g., high-risk obstetrics, social worker, clergyman) if appropriate and offered an additional counseling session if she remains ambivalent.

A complete gynecologic and obstetric history should be elicited including a history of cervical conization, cesarean sections, uterine duplication anomalies, a history of bleeding during the current pregnancy, or a history of ectopic pregnancy. Any coexisting medical conditions should also be screened for and treated appropriately. The remainder of the preoperative evaluation should consist of a hematocrit or hemoglobin and testing for Rh status.

The majority of surgical abortions occur in free-standing clinics. There are very few contraindications to a clinic or office procedure. Serious pre-existing medical conditions such as severe cardiovascular or pulmonary disease are best managed in a hospital setting. Patients with suspected placenta accreta, second-trimester molar pregnancies, and cervical pregnancy are at high risk for hemorrhage. Patients with poorly controlled seizure disorders, poorly controlled hypertension, severe asthma, and coagulation disorders are also best managed in the hospital where this is possible. A conversation with the patient’s primary care physician may allow for advance treatment that will allow safe abortion to be provided in the office setting with follow-up by her primary physician. Unfortunately, in some communities, hospitals and physicians refuse to accept patients needing abortion regardless of their medical condition.

In the United States, ultrasound has become an increasingly important part of abortion care. An accurate preprocedure assessment of gestational age is important. Although the 2002 NAF survey found that 91% of abortion providers confirm pregnancy with an ultrasound, the remainder of providers perform a pelvic examination for dating, using ultrasound only when a discrepancy arises.⁵ Other clinical indications for an ultrasound include symptoms or a history of ectopic pregnancy, inability to perform a pelvic exam, and pregnancies suspected to be less than 6 weeks. In a study of patients in a South African clinic, women’s estimates of pregnancy duration averaged 19 days less than the gestational ages estimated by ultrasound.¹² The providers’ clinical estimates were somewhat more accurate. Half were within 1 week of the ultrasound estimate and three-quarters were within 2 weeks.

Basic ultrasound machines are usually sufficient for pregnancy dating. A vaginal probe 5 to 10 MHz allows for visualization of early pregnancies, assessment of the adnexa in cases of suspected ectopic, and evaluation of the endometrium if retained tissue or clot is suspected. Abdominal probes of 3 to 5 MHz are sufficient to confirm intrauterine pregnancies in many women over 7 weeks.

Typically, a gestational sac is the first ultrasonographic sign of pregnancy. A normal sac will be symmetrical, eccentrically located in the uterus, and have the appearance of an echogenic ring with a sonolucent center. The discriminatory human chorionic gonadotropin (hCG) level at which a sac can be seen transabdominally is approximately 3600 mIU/mL, and transvaginally, that level is approximately 1500 to 2000 mIU/mL, although these numbers can vary in the presence of fibroids, multiple gestations, maternal obesity, prior uterine surgery, or uterine position.^13,14 Fossum et al.¹⁵ reported the presence of a gestational sac at 34.8 ± 2.2 days from last menstrual period or an hCG of 1398 ± 155 using transvaginal ultrasound. A fetal pole was detected at 40.3 ± 3.4 days or an hCG 5113 ± 298. To distinguish a pseudogestational sac from a true intrauterine gestational sac, a yolk
sac should be present. Typically, a yolk sac should be seen with a transvaginal probe if the mean gestational sac diameter has reached 13 mm.

The use of prophylactic antibiotics is routine with surgical abortion in the United States and is recommended by the ACOG.¹⁸ An extensive meta-analysis of randomized controlled trials in 1996 concluded that antibiotics were beneficial even in women in low-risk categories. An overall 42% decrease in the risk of infection was demonstrated.¹⁹ Generally, doxycycline is recommended because of its broad spectrum of antimicrobial coverage, oral absorption, and low cost.¹⁸ A common regimen is to give 200 mg of doxycycline by mouth with food immediately before or after the procedure.

Whether all or some patients requesting abortion should be screened for chlamydia and gonorrhea is a separate issue. The incidence of chlamydia in an abortion population is high, ranging from 7 to 20%.⁵ A review published in the British Journal of Obstetrics and Gynaecology concluded that the more cost-effective approach to preventing postabortal infection is universal antibiotic prophylaxis but that this approach would not prevent the long-term morbidity associated with chlamydia.²⁰ The benefits of routine testing are multiple: (a) to provide complete treatment with doxycycline for patients testing positive for chlamydia; (b) to identify patients with gonorrhea, an organism frequently resistant to tetracyclines; (c) to treat or allow identification of partners at risk. The disadvantage of screening is added cost. The appropriate strategy may depend on the local rate of gonorrhea and chlamydia and the availability of resources.

Both medical and aspiration abortion are highly effective and safe. Women express different reasons for preferring one procedure over. Some women feel that a medical abortion is less invasive, provides them with more control, offers more privacy, and allows them to avoid a surgical procedure. Other women choose a surgical procedure because they want a quick, more predictable procedure and may want anesthesia. In a study published in the British Medical Journal, 363 women less than 9 weeks were assigned to either randomization or their choice of abortion method. Twenty percent of the women presenting for an abortion had a preference for medical abortion, 26% preferred a vacuum aspiration, and the remainder were willing to undergo either method. Of women given their choice of procedure, only 4% of women would choose the other method next time. However, in those women randomly assigned to either medical or surgical abortion, 2% of the surgical abortion patients would choose medical next time, whereas 22% of the medical patients would prefer a surgical procedure with their next abortion.²¹ This suggests that choice is an important factor in a woman’s satisfaction but that having experienced medical abortion, the surgical approach seems preferable to some.

Naproxen sodium 550 mg taken 1 to 2 hours prior to the procedure has been found to reduce peak pain scores during abortion as well as to decrease postabortion pain.²³ This or other nonsteroidal anti-inflammatory drugs (NSAIDs) given preoperatively have become part of routine practice.²⁴

There is great variability in the type of pain relief offered with abortion services, but paracervical block with local anesthetics is the mainstay. In the latest NAF survey from 2002, 46% of first-trimester providers use local cervical anesthesia with or without oral premedication, one-third use local anesthesia combined with intravenous (IV) sedation, and 21% use deep sedation or general anesthesia.⁵

Despite paracervical block, many women report significant pain with their procedure. In a large study, 34% of patients undergoing first-trimester uterine aspiration reported pain that was “severe” or “very severe.”²⁵ After decades of reliance on paracervical block, there is conflicting evidence as to whether it works at all. A 2006 randomized trial of lidocaine versus bacteriostatic saline in paracervical block found no difference in pain relief.²⁶ A recent systematic review found two additional randomized trials comparing paracervical block to no treatment or to bacteriostatic saline.²⁷ The first study found no difference in pain scores with dilatation, aspiration, or postoperatively; however, both groups also received IV conscious sedation.²⁸ In the second study, the group receiving the active agent (1% chloroprocaine) did have better pain control, but the mean improvement in pain scores was only 1 point on an 11-point scale, probably too little to be clinically significant.²⁹ One group of investigators has reported that adding 30 mg of ketorolac to 18 mL of 1% lidocaine provided greater pain relief than oral ibuprofen and a plain lidocaine paracervical block.³⁰ In a new approach, an intrauterine infusion of 5 mL of 4% lidocaine via a 3 mm Novak cannula 3 minutes prior to vacuum abortion reduced procedural pain significantly when compared to paracervical block alone. Lidocaine serum levels were measured and were well below the toxic range.³¹ This is a promising technique, which should be studied further.

Several authors have studied variations in technique for paracervical anesthesia for abortion.³² Injecting 10 to 20 mL of 1% lidocaine deep (1 to 1.5″) into the cervical stroma is more effective than superficial injections beneath the cervical mucosa.³³ Deep injections at 12, 4, and 8 avoid the lateral cervical vessels while providing as adequate analgesia as multiple injections at 12, 3, 4, 8, and 9.²⁹ In comparison with plain lidocaine, buffering lidocaine to a neutral pH reduces pain from the injection itself and also has been reported to reduce the pain with dilatation and the pain perceived at the end of procedure.³³ Buffering is accomplished by adding 5 mL of 8.4% sodium bicarbonate solution to 45 mL of 1% lidocaine solution.³⁴ Furthermore, the addition of 2 to 4 units of vasopressin has been reported to reduce blood loss from the procedure and may help prevent postabortal atony.³⁵

Increasingly, practitioners offer IV conscious sedation. Allen and colleagues³⁶ compared oral oxycodone and sublingual lorazepam to IV sedation with 100 mcg of fentanyl and 2 mg of midazolam in a randomized controlled trial. Pain relief was much better for the group who received the IV sedation. Older trials using lower doses of fentanyl 50 to 100 mcg found statistically significant decreases in pain when compared to placebo, but the women still reported less than adequate pain relief with their procedure.³⁷ One milligram of oral lorazepam by itself is not better than placebo in relief of pain.³⁸ IV sedation with narcotics and anxiolytic agents adds risk for respiratory arrest and for aspiration of gastric content. Using IV sedation in an office setting requires limiting the dosages to avoid deep sedation, monitoring with continuous pulse oximetry, the availability of oxygen supplementation if needed, and the presence of a trained assistant to observe the patient in addition to the surgeon. Wilson and colleagues³⁹ report a series of 1433 patients to 18 weeks’ gestation sedated for office abortion under conditions as described earlier with 50 to 100 mcg of fentanyl combined with 1 to 2 mg midazolam. There were no adverse events related to the sedation.

Nonpharmacologic methods such as continuous lower abdominal heat, counseling and skilled conversation, positive suggestion, guided imagery, and other relaxations techniques during the procedure may be helpful.³² Ultimately, the choice of anesthesia should take into account patient and provider preference, available resources, the clinical setting, and the patient’s comorbidities.⁴⁰

Typically, abortion providers cleanse the vagina and cervix before a procedure with either povidone-iodine or chlorhexidine, although some have called into question this routine practice. One group found no difference in the rate of endometritis after vacuum aspiration when using 0.05% chlorhexidine solution versus 0.9% saline solution.⁴¹ Varli et al.⁴² looked at vaginal and vulvar cleansing with chlorhexidine versus vulvar cleansing only and found no significant difference between methods. They hypothesize if vaginal cleansing may in fact promote the growth of more pathogenic bacteria by changing the normal flora.⁴² Likely more important in reducing the risk of infection is adherence to meticulous surgical technique including a “no-touch technique” where clinicians wearing either sterile or nonsterile gloves avoid touching those portions of instruments entering the uterus, establishment of a sterile field, and avoidance of tissue trauma.

Surgical abortion is provided in a variety of clinical settings including operating rooms, outpatient surgical centers, clinics with procedure rooms, and office examination rooms. The typical room should include an exam table with stirrups or knee-holders and adequate lighting. A source of backlighting such as a photographer’s light box and a clear dish to view specimens should also be available.⁴³ Ultrasound is helpful in cases of difficult dilatation, retroverted uteri, suspected retained tissue, the training of new providers, and second-trimester procedures but is not essential.

A sterile instrument tray should be set up in the room with easy access by the provider. A typical tray
should include a speculum, tenaculum, ring forceps, a set of graduated tapered cervical dilators (Pratt or Denniston), a small metal curette, gauze, a 21- to 25-gauge spinal needle, a 10-mL syringe, and antiseptic solution. Suppliers of instruments useful in abortion care are listed in Table 13.2.⁴⁴

Most providers use single-toothed tenaculum, an Allis or a Bierer vulsellum tenaculum. Tears may occur less often with these latter designs, but there may also be greater slippage with strong traction. Some providers prefer ring forceps especially at later gestational ages or with more bulbous cervices. A ring forceps is also helpful for possible tamponade at a bleeding cervical laceration site and as an adjunct to suction in late first-trimester procedures.⁴³ Preferred dilators have a gradual taper, such as Pratt dilators. Denniston dilators are a plastic version of the Pratt dilator. They are much less expensive than the metal Pratt dilators and can be steam sterilized. Routine use of a metal curette is becoming increasingly less common after first-trimester abortion practices. However, a metal curette can be helpful to explore the uterine cavity when retained tissue is suspected. They increase in size starting at 1 mm in width. Cannulae come in both rigid and flexible forms. Rigid forms range from 6 to 16 mm in size. Flexible cannulae are sized from 4 to 12 mm.⁴³

The choice of manual versus electric suction is dependent on available resources, gestational age of the patient, and provider experience. In the most recent NAF survey, 49% of providers selectively use manual vacuum aspiration (MVA), an increase from 18% in 1997.⁵ MVA instruments are simple, easily carried, and require no electricity. There is no electric pump to purchase, maintain, or clean between cases. Goldberg et al.⁴⁵ in a retrospective cohort study of 1726 patients concluded that MVA is as effective as aspiration with an electric pump to 10 weeks’ gestation with no difference in complication rates. Similarly, a systematic review of 10 trials comparing manual and electric aspiration in the first trimester found evidence that MVA is as effective as electric suction.⁴⁶ At gestational ages less than 50 days, there was less blood loss and less pain associated with MVA. However, electric suction resulted in shorter procedure times. After approximately 10 weeks’ gestation, the MVA syringe may need to be emptied once or twice to complete the procedure, so many providers will switch to electric suction at this point in gestation. Many providers also prefer MVA because it is quieter than the electric suction machine. In addition to its use in uterine aspiration, MVA is valuable in treating incomplete abortion, confirming an intrauterine pregnancy in cases of suspected ectopic and in managing acute uterine hemorrhage either in an office or emergency room setting.

There are two basic types of manual aspirators. Both are 60 mL plastic syringes modified so that the plunger can be locked in the open position. One type of aspirator (Ipas MVA Plus) has a valve assembly that allows withdrawing the plunger of the syringe to establish a vacuum before inserting the attached cannula into the uterine cavity. The valves are then opened to aspirate. The other (Milex Handyvac Locking Syringe) has no valve, and the plunger is not withdrawn to create a vacuum until after the cannula is attached to the syringe and inserted into the uterus. With both instruments, the plunger locks into place to prevent loss of vacuum during the procedure.^43,44 The Ipas MVA Plus can be steam sterilized. The Milex Handyvac is meant to be disposable. Both accept plastic vacuum cannulas with a 12-mm diameter base.

An electric suction machine consists of an electric vacuum pump connected through plastic hoses to glass or plastic reservoir bottles. A plastic hose connects to a bottle on the machine while the other end attaches to a plastic control handle that receives the vacuum cannula. There are both reusable and single-use hoses.⁴³

A first-trimester abortion procedure typically begins with a patient in lithotomy position. The provider will perform a basic bimanual exam assessing for uterine size and position and the presence of any adnexal masses.
A sterile speculum is inserted into the patient’s vagina. A no-touch technique should be employed throughout the procedure. The operator takes care to avoid touching that portion of an instrument that will go inside the cervical canal. Most providers use povidone-iodine or chlorhexidine to cleanse the vagina and cervix. Approximately 2 to 4 mL of buffered local anesthetic with vasopressin is injected at 12 o’clock. A tenaculum is then placed vertically on the cervix, with one branch extending into the cervical canal. Alternatively, the tenaculum can be placed horizontally. It is important to grasp a sufficient amount of cervical stroma to avoid the tenaculum tearing through the cervix when traction is applied. At this point, the remainder of the local anesthetic is administered. The maximum advised lidocaine dose is 4.5 mg/kg or 20 mL of 1% lidocaine for a 50-kg patient. Generally, 10 to 20 mL of 1% lidocaine is used at two sites, 4 o’clock and 10 o’clock, or at multiple sites around the cervix.²² Cervical dilatation is then performed being careful to maintain no-touch technique and avoid grasping the dilator by its end. Once sufficient dilatation is achieved, the provider can insert an appropriately sized cannula gently to the uterine fundus. Many providers choose a cannula size equal to the gestational age of the pregnancy (i.e., an 8-mm cannula for an 8-week gestation). Pratt dilators are measured in the French system like urinary catheters. The number on the dilator is the circumference in millimeter. Hence, the number of the largest Pratt dilator used should be approximately three times the diameter of the planned cannula size.