Enhanced Recovery in Gynecologic Surgery

Diet

“NPO at midnight” was an arbitrary mantra adopted in a well-meaning, but misguided effort to prevent aspiration at induction of anesthesia. However, prolonged fasting results in patient dissatisfaction, dehydration, insulin resistance, depletion of liver-glycogen stores, and affects perioperative outcomes adversely. A recent Cochrane review showed no change in the risk of aspiration when patients were allowed to drink clear liquids preoperatively. Recognizing these findings, the American Society of Anesthesiologists recommends limiting fasting periods to 6 hours for solids and 2 hours for liquids prior to elective procedures requiring general or regional anesthesia. ⁸ In addition to limiting periods of fasting, the use of 50 g of oral carbohydrates 2 hours before induction of anesthesia has been shown to reduce insulin resistance, attenuate surgical stress, and improve healing.

Avoidance of mechanical bowel preparation

The term bowel preparation may include both mechanical and antimicrobial components in an attempt to improve healing of anastomoses and reduce infection. However, bowel preparation—particularly mechanical preparation—results in dehydration and the potential for electrolyte imbalances, and causes great patient dissatisfaction, often leading to noncompliance. Bowel preparation also requires prolonged fasting, which should be avoided for the reasons described above. No benefit was found for the use of mechanical bowel preparation in a Cochrane review of 18 randomized controlled trials (RCTs) or in a meta-analysis of 5 RCTs focused on gynecologic surgery.^9,10 It has been suggested that mechanical preparations improve visualization and bowel handling in MIS, but this has not been supported in objective investigations. ² In contrast, to mechanical preparation, oral antibiotic bowel preparation has been shown to reduce infection rates in colorectal, but not in gynecologic, surgery. An analysis of more than 200,000 hysterectomies in the OptumLabs database failed to show improvements in infectious morbidity or anastomotic leak with mechanical bowel preparation or oral antibiotic use. ¹¹ While rectal enemas are not generally considered bowel preparation per se and do not result in dehydration or electrolyte disturbances, enemas are also not associated with improvements in outcomes.

Preemptive analgesia

The concept of preemptive analgesia is based on blocking pain receptors prior to activation to improve pain control and reduce opioid requirements. Gabapentin, cyclo-oxygenase–2 inhibitors (COX-2; celecoxib), and oral acetaminophen are typically used orally prior to induction and are associated with decreased opioid requirements. ² Gabapentin has been associated with sedation, respiratory depression, and a higher rate of rapid-response calls in the recovery room, and should be used with caution in elderly patients. Many other pharmacologic agents, including intravenous (i.v.) lidocaine, magnesium, and ß-blockers, have either been investigated poorly or shown to have minimal efficacy when investigated properly.

Anesthesia

Reevaluation and standardization of anesthetic protocols should be performed at every institution with an eye toward improved pain control and rapid recovery, particularly given the growth of same-day discharge for procedures such as hysterectomy that previously were thought to require overnight admission. Together with regional anesthetic, when appropriate (discussed in further detail, below), Short-acting agents are the lynchpin of these revised anesthetic protocols. The use of total i.v. anesthesia can result in reduced side-effects, but the higher costs associated with agents such as propofol should be weighed carefully against proven benefits in future trials.

Maintaining normothermia

Hypothermia is generally defined as an intraoperative body core temperature <36°C. Hypothermia increases the risk of coagulopathy and bleeding, slows metabolism of drugs, and increases the risk of infections and cardiac morbidity through impaired oxygen transportation. To avoid these problems, active warming should begin in the preoperative area, continue intraoperatively, and extend through the recovery period. Simple and inexpensive interventions include use of standardized room thermostat settings and continuous core body–temperature monitoring. Forced-air blankets were once thought to increase the risk of infection but this has since been disproven. Other interventions that have proven efficacious alone or in combination include i.v. fluid warming, heated mattress pads, and circulating-water garments. ¹²

Maintenance of euvolemia

Maintaining euvolemia, as opposed to hyper- or hypovolemia is a central tenet of ERAS pathways. Repeated fluid boluses and volume excess often result in peripheral and small-bowel edema, leading to ileus, a common cause of prolonged LOS. Hypervolemia can also lead to pulmonary congestion and resulting morbidity. Conversely, fluid restriction and resulting hypovolemia can lead to impaired oxygen delivery, resulting in end-organ damage. A large trial demonstrated increased morbidity and renal dysfunction in patients randomized to restricted-fluid therapy. ¹¹ However, close analysis of the care pathway utilized in that trial suggested that that cohort could have been categorized more accurately as hypovolemic. ¹³ Compared to liberal fluid management, goal-directed therapy expedites return of bowel function, decreases nausea, reduces intensive-care unit admissions, and shortens hospital LOS. Perioperative euvolemia is not always straightforward to achieve, and can be informed by hemodynamic parameters, such as cardiac output and stroke volume, and also by preoperative fluid status (omission of mechanical bowel preparation and fasting) and insensible losses intraoperatively (MIS versus laparotomy).

Prevention of postoperative nausea and vomiting

Female gender, pelvic surgery, and MIS all increase the risk of postoperative nausea and vomiting (PONV). At least 2 of these factors are present in all patients undergoing gynecologic surgery, with nausea rates as high as 80%. Prevention of PONV is particularly important, recognizing that same-day surgery and discharge is occurring with increasing frequency. Toward this aιm, ERAS pathways include a multimodal and preemptive approach toward treatment of PONV, including use of at least 2 different classes of antiemetics. These classes include 5–hydroxytryptamine type 3 antagonists, neurokinin-1 antagonists, corticosteroids, antihistamines, anticholinergics, butyrophenones, and phenothiazines. Total i.v. anesthesia, avoidance of nitrous oxide, and minimizing opioid use reduces PONV further. Transdermal scopolamine applied 2 hours prior to anesthesia may be considered in patients with histories of PONV but should be used with caution in elderly patients, given the risk of neurologic side-effects.

Avoidance of drains and tubes

Within ERAS frameworks, using indwelling drains and tubes is discouraged. An objective review of the literature shows that most drains and tubes are unnecessary and should be eliminated. For example, a Cochrane review of 33 RCTs concluded that, compared to routine nasogastric tube (NGT) use, selective use or omission of the NGT was associated with reductions in pulmonary complications, improved bowel recovery with shorter LOS, and stable rates of anastomotic breakdowns and other complications. ¹⁴ Similarly, use of prophylactic drains was not associated with earlier detection of bleeding or anastomotic breakdown, compared to patients without drains, and should not be used routinely. ¹⁵ Indications for peritoneal drains are rare, and should be considered only in patients with a very low rectal resection in the absence of bowel diversion, or in patients with extraordinary concerns for fluid collection or bleeding. When used, peritoneal drains should be removed as early as possible.

Fluid management, postoperative feeding, and mobilization

Postoperatively, every effort should be made to maintain euvolemia, just as in the pre- and intraoperative periods. Upon arrival to the postanesthesia care unit, patients are permitted to drink and i.v. fluids are discontinued the morning following surgery or once oral intake exceeds 500 mL. Continuing i.v. fluids beyond 12–24 hours postoperatively is unusual, even following complex oncologic resections. Maintenance fluids should be given at a rate <1.2 mL/kg, and fluid boluses should be used cautiously. Providers should recall that oliguria is a normal response to surgical stress and urine output as low as 20 mL/hour does not require intervention. Crystalloids, such as Ringer's lactate, are preferred to 0.9% normal saline to avoid hyperchloremic acidosis.

Resumption of a general diet should be encouraged (not required or forced) upon recovering from anesthesia. Randomized trials in gynecologic surgery show that early feeding results in earlier return of bowel function, shorter hospital LOS, and high patient satisfaction with no change in postoperative complications (such as pulmonary complications, anastomotic leak, and poor wound healing). Notably, early feeding might result in some degree of postoperative nausea, but this is not accompanied by a higher risk of vomiting, abdominal distention, or need for NGT placement. Early mobilization prevents venous thromboembolic events and, together with early feeding, reduces deconditioning and muscle loss, and is associated with trends of shorter hospital LOS.

Early urinary catheter removal

Urinary catheters should be used only when necessary and should be discontinued as soon as possible. Randomized trials and meta-analyses have demonstrated that prolonged indwelling catheter use increases the risk for infectious morbidity, while early removal increases the risk for urinary retention. One investigation in patients undergoing total abdominal hysterectomy compared urinary catheter removal in cohorts immediately after surgery, 6 hours, or 24 hours postoperatively. ¹⁶ Removal of catheters 6 hours postoperatively resulted in fewer urinary-tract infections, earlier ambulation, and shorter hospital LOS, compared to the other 2 cohorts.

Perioperative pain management

Adequate perioperative pain control is a key component of enhanced recovery. The concept of multimodal analgesia is based on synergism between 2 or more medications. Non-steroidal anti-inflammatory drugs (NSAIDs) such as COX-2 inhibitors are effective for postoperative pain control, and reducing opioid requirements and PONV; NSAID effectiveness is increased when combined with acetaminophen. Initial concerns regarding an association between NSAID use with anastomotic leaking and postoperative bleeding have been disproved following well-designed multicenter trials. Due to common side-effects, including constipation, nausea, impaired bowel function, impaired mental status leading to reduced mobilization, and pulmonary complications resulting from respiratory depression, opioids should be used sparingly.

In addition, ∼6% of patients will engage in new persistent opioid use following surgical procedures; this use is related to social and behavioral factors rather than surgical approach or complexity. Responsible postoperative opioid prescribing is important to minimize the contribution of surgical specialties to the ongoing opioid crisis in the United States. Many groups have published results demonstrating that, even after complex laparotomies, the majority of patients may be managed with 5–10 tablets of oxycodone at discharge. ¹⁷

Several regional analgesic techniques may be used with the goal of improving pain control further and reducing opioid use but should be evaluated carefully and objectively for efficacy. These techniques include thoracic epidural analgesia (TEA), transversus abdominis plane (TAP) blocks, and wound infiltration with local anesthetic.

TEA is considered standard of care in many ERAS pathways, including colorectal, hepatobiliary, urologic, thoracic, and vascular surgeries. However, supporting data in gynecologic surgery are conflicting, and many investigators have failed to prove benefit using pain control and hospital LOS as endpoints.^18,19 Furthermore, TEA is associated with failure rates as high as 30%, can result in hypervolemia if fluid boluses are needed to address hypotension secondary to sympathetic blockade, and might interfere with early mobilization and early urinary-catheter removal.

TAP blocks involve infiltration of local anesthetic into the plane between the internal oblique and transversus abdominis muscles. TAP blocks reduce postoperative pain up to 24 hours postoperatively, compared to placebo. While TAP blocks are an option for patients undergoing open surgery and MIS, ultrasound guidance is often necessary for accurate injection, and simple infiltration of the wound with local anesthetic could have practical advantages with similar or even improved efficacy.

Wound infiltration of the surgical site with bupivacaine is safe, effective, quick, and easy to perform. Elastometric pumps have been developed to address the short duration of action of bupivacaine and to provide continuous wound infiltration (CWI). Results with CWI are conflicting in both gynecologic- and nongynecologic-surgery literature, with some investigations failing to show benefit. Disadvantages of CWI include difficulty of use, the potential for catheters to become dislodged, and the potential for surgical-site infection. To achieve a longer duration of action, liposomal bupivacaine may be used in place of bupivacaine, with a 72–96-hour duration of action, compared to 8 hours. In comparison to regular bupivacaine, liposomal bupivacaine resulted in stable to improved pain scores depending on surgical complexity, an 80% reduction in use of opioids in the first 48 hours, and a sevenfold reduction in the need for patient-controlled analgesia (PCA) in patients undergoing laparotomy for ovarian cancer, despite omitting TEA. ²⁰ However, benefit from liposomal bupivacaine appears to be limited—from a cost-effectiveness standpoint—to the most complex laparotomies (i.e., 4 quadrant oncologic resections). For this reason, regular bupivacaine should be used in patients undergoing simple laparotomy or laparoscopy.