Pregnancy-associated diseases

Physiologic changes associated with pregnancy ( table 32.1 )

Nonobstetric surgery during pregnancy

This section of the chapter fits most closely within the title of this book, in that pregnancy is truly the “coexisting disease” in a woman who needs surgery and anesthesia while she is pregnant. Of all pregnant women in the United States, 1% to 2% will undergo surgery unrelated to their pregnancy (∼75,000 procedures requiring anesthesia per year). The most frequent nonobstetric procedures are appendectomy, cholecystectomy, breast biopsy, and surgery required because of trauma.

Generally, elective surgery should be delayed until the patient is no longer pregnant and has returned to the nonpregnant physiologic state. Depending on how elective the procedure is, surgery is sometimes delayed until after the woman has stopped breastfeeding, but this is usually not necessary (see Anesthesia in the Breastfeeding Woman, later). Procedures that can be scheduled with some flexibility but that cannot be delayed until after delivery are best performed in the second trimester. In theory, this lessens any risk of teratogenicity with first-trimester medication administration and preterm labor, which is thought to be a greater risk in the third trimester, although both of these assumptions and recommendations have come under some scrutiny and criticism recently, and many argue that delaying indicated surgery might be the larger risk in many clinical circumstances ( Fig 32.1 ).

The pregnancy-specific issues involved in nonobstetric surgery during pregnancy involve taking into account pregnancy-induced anatomic, physiologic, pharmacodynamic, and pharmacokinetic alterations, minimizing fetal drug exposure when possible and monitoring for and prevention of preterm labor, a particular risk with intrabdominal surgery.

Little outcome data exist to guide anesthetic choice, but when possible most anesthesiologists (and often patients) prefer a regional anesthetic (neuraxial or peripheral nerve block) with minimal sedation, as this would appear to be the best way to minimize fetal drug exposure. While logic and reasonable caution would suggest minimizing exposure to drugs having significant effects on the human brain, no commonly used anesthetic agents have been shown to cause teratogenicity or other significant effect on the fetus. Most anesthetic decisions should be based on the usual anesthetic and surgical considerations as well as maternal status and safety, not on the presence of the fetus. For the most commonly performed operative procedures during pregnancy (appendectomy and cholecystectomy), general anesthesia is required, particularly since these operations are now almost always performed laparoscopically. A decade ago there was significant controversy about whether laparoscopic surgery was safe during pregnancy. There were concerns with issues of the effect of intraperitoneal pressure and potentially increased Pa _CO2 , but clinical experience and some animal studies have mostly put these concerns to rest, and laparoscopic appendectomies and cholecystectomies have become the standard approach, even for pregnant women.

Airway management during surgery depends on gestational age. By late second trimester, most anesthesiologists would opt for rapid-sequence induction and intubation to reduce the risk of aspiration, as is standard for cesarean delivery under general anesthesia. Earlier in gestation, before ∼20 weeks, controversy exists as to the level and nature of this risk, and the need for rapid-sequence induction or intubation will depend on the specific nature of the patient and operative procedure.

Teratogenicity may occur at any stage of gestation. However, most organogenesis occurs in the first trimester. Although many commonly used anesthetics are teratogenic at high dosages in animals, no studies or data support teratogenic effects of anesthetic or sedative medications in humans at the dosages and time periods used for anesthesia care. There was literature in the 1960s suggesting a link between chronic benzodiazepine and other sedative use and fetal/neonatal cleft lip and palate, but even this issue of chronic exposure has been mostly refuted. There is some evidence of a link between maternal high-dose diazepam treatment and intrauterine growth restriction, but single or limited doses of benzodiazepines are almost certainly safe when used in the usual manner for perioperative anxiety. That being said, it is common to attempt to avoid these medications in pregnancy due to concerns about effects on the fetal brain and the history of controversy about their use.

Nitrous oxide has been suggested to be teratogenic in animals when administered for prolonged periods (1–2 days); the concern is biologically plausible given its effect on DNA synthesis through inhibition of methionine synthetase. As with other purported effects of anesthetics, however, it has been difficult to document clinical effects in humans. Teratogenesis has been seen in animals only under conditions (long, repeated exposure) that are not likely to be reproduced in clinical care or relevant to single exposure perioperatively. Nitrous oxide is of course widely used for labor analgesia, so fetal exposure near and at term is common without generating too much concern from either clinicians or patients, although actual data on fetal/neonatal safety are surprisingly limited. Recent studies suggest that volatile anesthetics stimulate neuronal apoptosis in rats, but it is not obvious whether these data can be extrapolated to humans. Widespread neuronal apoptosis is associated with memory and learning deficits in laboratory animals, but again this has not been shown in humans.

Propofol and ketamine are all safe intravenous (IV) induction agents. Induction doses for these medications are unchanged in pregnancy. The choice of induction agent is usually based on provider preference and the clinical status of the patient (e.g., presence of dehydration, valvular heart disease, dysrhythmia, hypertension, or preeclampsia). None of these agents has been clearly shown to be teratogenic or have adverse effects on human brain development. Thiopental is no longer available in the United States but is also an accepted induction agent during pregnancy.

In the past decade or so there has been increasing concern in the pediatric anesthesia arena about the possible negative effects of anesthetics on behavior and learning in children, especially upon repeat or long exposure. These concerns could also be relevant to in utero exposure. While studies in young animals show a variety of behavioral and cellular neurotoxic effects of a variety of drugs and of anesthetic exposure in general, the results from human clinical studies and reviews of large databases of pediatric anesthetics have not generally shown much effect. A detailed discussion of this complex issue that is under intense investigation around the world is beyond the scope of this chapter, but the available evidence would not support any recommendation to withhold anesthesia from a pregnant woman who needs it on the basis of concern about drug effects on the fetus.

Pregnant patients are more sensitive to the action of nondepolarizing muscle relaxants but may have increased clearance of both these medications, so dosing of these medications may need to be more frequent and based on neuromuscular monitoring. Succinylcholine dose is unchanged in pregnancy; its volume of distribution is increased, but systemic pseudocholinesterase activity is decreased, resulting in increased variability in duration of action, but this is unlikely to be of any clinical significance. Muscle relaxant reversal in pregnancy involves a few unique issues. The use of neostigmine is unproblematic, but the vagolytic medication given to prevent bradycardia and other effects requires some thought. Glycopyrrolate does not cross the placenta, but neostigmine, although also a quaternary amine, does appear to cause bradycardia in some fetuses; thus atropine, which does cross the placenta, is the recommended vagolytic adjuvant in pregnancy. Suggamadex has not been approved for use in pregnancy, as very few drugs have been, but with suggamadex there is reason to consider avoiding it due to its known effect of binding steroid hormones, including progestins, which could have unknown effects on pregnancy. However, no deleterious effects or decreases in progesterone levels were reported after high doses in early pregnancy in rats.

After almost any surgical procedure during pregnancy, fetal heart rate should be monitored in the recovery room, intermittently for previable fetuses and continuously for the viable fetus. Uterine activity should also be monitored because contractions are most likely to occur proximate to the procedure and as the tocolytic effect of volatile anesthetics wears off. The use of tocolytic drugs (e.g., magnesium, indomethacin) has not been studied perioperatively, but they are commonly used. Opioids can be used as needed to control postoperative pain, as a short course of opioids is unlikely to have any detrimental effect on the fetus, but consideration should be given to peripheral nerve blocks or epidural analgesia when possible. Nonsteroidal antiinflammatory drugs (NSAIDs) are often avoided due to concerns about an increased miscarriage rate and premature closure of the ductus arteriosus seen in some studies of chronic use, although short-term perioperative use is probably safe particularly before the third trimester.

The statement of the American College of Obstetricians and Gynecologists (ACOG) Committee on Obstetric Practice titled “Nonobstetric Surgery During Pregnancy” recommends that an obstetric consult be obtained before surgery and that use of fetal monitoring be individualized. The only absolute recommendation in our institution’s policy on surgery during pregnancy is that every pregnant woman undergoing surgery must have an obstetrician on staff who is aware of her condition and able to consult as to the decisions regarding issues of fetal and labor monitoring, timing of surgery, and other issues that may arise perioperatively, including availability for possible emergent cesarean delivery. In consultation with the surgeon and anesthesiologist, the obstetrician can also make decisions about administration of corticosteroids to increase fetal lung maturity if preterm labor seems likely and the fetus is viable, and assess venous thromboembolism risk and recommend thromboprophylaxis, which is often indicated in pregnant women who are hospitalized.

Cardiac surgery requiring cardiopulmonary bypass (CPB) is rarely needed in pregnant women and when possible should be delayed until after pregnancy as maternal and fetal mortality after CPB are 3% to 13% and 16% to 38%, respectively. If surgery and CPB are unavoidable (aortic dissection, mechanical valve thrombus, severe valvular lesion with maternal decompensation, etc.), then the parameters shown to improve fetal outcomes are shorter CPB and aortic cross-clamp times, normothermia, higher pump flows (>2.5 L/min/m ² ), perfusion pressure greater than 70 mm Hg, and maintaining the hematocrit above 28%. If cardiac surgery is necessary, concomitant cesarean delivery (if fetus is viable) or staged CPB may be the preferable strategy. Fetal heart monitoring may be useful intraoperatively as fetal arrhythmias or bradycardia can be treated by improving maternal hemodynamics. If cardiac surgery is performed in pregnancy, the second trimester has the best maternal and fetal outcomes. Continuous fetal monitoring and uterine contraction monitoring for 12 to 24 hours postoperatively is recommended as uterine contractions are the most important predictor for fetal demise, and contractions can occur on rewarming.

Anesthesia in the breastfeeding woman

The issue concerning anesthesia for a breastfeeding woman, and what to tell her to do after the anesthetic, has been controversial for years. The concern is mostly that anesthetic drugs, when passed into breast milk and consumed by the breastfeeding infant, could be toxic or at least cause sedation or behavioral problems. Recommendations that breastfeeding women who receive anesthesia care express milk and discard it (“pump and dump”), for some period of time, often 24 hours, have been common over the past decades. With increasing understanding of the magnitude of drug transfer (very low in most cases) and with improved pharmacokinetic profiles of many anesthetics, combined with the further establishment of the benefits of breastfeeding, recommendations have changed significantly in the past few years. A strong consensus appears to have been reached that there is less of an issue than there used to appear to be, and almost all women are now advised the resume breastfeeding after surgery and anesthesia as soon as they are awake and able to breastfeed. A recent formal statement from the American Society of Anesthesiologists (ASA) and a detailed set of guidelines and recommendations from the Association of Anaesthetists of Great Britain and Ireland (AAGBI) both state that pump and dump is unnecessary.

The AAGBI document contains some of the most detailed and evidenced-based guidelines. It makes the following points: (1) Breastfeeding is of proven benefit to women and infants, (2) most drugs are transferred in very low quantities into breastmilk, (3) most anesthetic drugs leave the woman’s system or decrease in concentration very quickly (e.g., propofol, volatile anesthetic agents), (4) many recommendations have erred on the side of caution with little evidence that the advice to discard breast milk for 12 or 24 hours is based on science or clinical experience, and (5) as the infant will be ingesting any of the drugs of concern orally, and oral availability of most anesthetic/analgesic/sedative medications is low, there is a built-in element of safety ( Table 32.2 ).

There are a few drugs that are controversial and might and usually can be avoided in breastfeeding women. Tramadol and codeine are both metabolized by cytochrome CYP2D6, and this metabolism is greatly affected by the specific genetic variant present; the US Food and Drug Administration (FDA) advises against their use in breastfeeding women, although similar agencies in other countries have not taken this position. As there are many other analgesics to choose from, it would seem that avoidance of these two should not be difficult. Benzodiazepines and other long-acting sedatives should be used in limited doses, but the usual perioperative use of these medications should be not be problematic. When necessary for postoperative analgesia, opioids can and should be used, but at the lowest effective doses, and the infant should be monitored for any signs of any sedation. Sugammadex is a large molecule probably not found in any significant concentration in breast milk, but it does interfere with pharmacologic contraception so could be avoided in women relying on that form of birth control.

Obstetric anesthesia care

Because this is a chapter about pregnancy as a “coexisting disease” when anesthesia care is required, we will not go into detail on general strategies of standard obstetric anesthesia management of labor pain or operative delivery in healthy pregnant women, a subject more properly accessed in books and review articles devoted exclusively to obstetric anesthesia and analgesia. We will focus on obstetric anesthesia in women with some of the more common coexisting diseases that occur or exist during pregnancy and that make anesthesia or analgesia care more complex for the anesthesiologist.