Do current antiemetic practices result in positive patient outcomes? Results of a new study

As the most common of postsurgical complications, postoperative nausea and vomiting (PONV) and postdischarge nausea and vomiting (PDNV) affect approximately 25 million patients worldwide yearly,¹ with an estimated annual financial impact of several million dollars.² PONV occurs in approximately one-third of patients who require surgery and anesthesia,^3,4 and can result in prolonged hospital stays.² Although the incidence of PDNV is likely underreported due to lack of surveillance following hospital discharge,⁴ the most current available randomized controlled trial evidence suggests that approximately 30% of ambulatory surgery patients experience PDNV,⁵ confirming earlier out-patient incidence reports.⁶ As 65% of surgeries occur in the ambulatory setting, PDNV affects a significant portion of surgical patients and is a substantial problem. Emesis increases the risk of aspiration and has been associated with cases of suture dehiscence, esophageal rupture, subcutaneous emphysema, and bilateral pneumothoraces.^1,7,8

Among patients undergoing elective surgeries, PONV and PDNV is the most feared complication,⁹ and most patients prioritize vomiting as the top adverse effect of anesthesia to avoid, ahead of gagging on the tracheal tube or incisional pain.¹⁰ Many patients find nausea and vomiting even more distressing than postoperative pain.¹¹ Moreover, patients are willing to pay out-of-pocket to avoid nausea and vomiting, particularly when having recently experienced the adverse effects postoperation.¹² Having recently endured PONV while in the post-anesthesia care unit (PACU), nauseous patients were willing to pay $73 and those who had vomited nearly $100, while patients without PONV were willing to pay $40 to avoid the symptoms.¹² Similar results were obtained by surveys conducted in Germany and Turkey—patients considered the avoidance of nausea and vomiting valuable enough to pay out-of-pocket (€65–€68), especially when having recently experienced PONV (€96–€99).¹³

Specific patient characteristics associated with the proclivity to develop nausea and vomiting have been identified. Risk factors found to be predictive of PONV are female gender, a history of motion sickness or PONV, and nonsmoking status.^14,15 For patients with these risk factors, scopolamine patches have a prophylactic benefit when applied a minimum of three hours prior to surgery.¹⁶ Treatment-related risk factors include previous chemotherapy¹⁷ or surgery, certain surgical procedures, the use of volatile anesthetics or possibly nitrous oxide, and usage of postoperative opioid analgesia (Table 1).^14,15 In particular, during laparoscopic surgery, gas insufflated into the abdomen or pelvis markedly increases the likelihood of PONV. According to risk model analyses, the presence of one or two risk factors is associated with 20–40% of surgical patients developing PONV, while three to four risk factors increase the number of patients who have PONV to 60–80%.¹⁴

Patients with a 20% or greater risk of developing PONV or PDNV should be considered high risk and treated as such with appropriate prophylaxis. Using risk scores as a guide to determine which patients would benefit from prophylaxis has been shown to reduce the incidence of nausea and vomiting in surgical patients, compared to the incidence predicted without risk stratification or a nonselective approach to providing prophylaxis.^18,19 A study by Pierre and colleagues demonstrated a reduction in the incidence of PONV from 49.5% to 14.3% (p < 0.001) when a risk-score based antiemetic approach was implemented.¹⁹

Nausea, retching, and vomiting represent different stages of one process. Nausea is the urge to vomit, which can lead to retching—the rhythmic contractions of the diaphragm, abdominal wall, and chest wall muscles (otherwise known as “dry heaves”).³ Vomiting can be the final outcome of nausea and is a reflexive, rapid, and forceful oral expulsion of the upper gastrointestinal (GI) tract contents due to powerful and sustained contractions in the abdominal and thoracic musculature. There are a number of different pathways, each with unique stimuli, which can additively create nausea or vomiting. Furthermore, several areas of the brain and GI tract are involved in the pathophysiology of vomiting, including the emetic center located in the lateral reticular formation of the medulla, which is not exposed to blood and cerebrospinal fluid, in contrast to the chemoreceptor trigger zone (CTZ). The CTZ is the major protective organ against toxic substances within the brain and is involved in the development of nausea and vomiting (Figure 1).^3,20 The vestibular labyrinthian system of the middle ear can be involved in the development of nausea and vomiting when stimulated with motion or vestibular toxicity, as can the visceral afferents from the GI tract when obstructed, targeted by certain drugs, or afflicted with gastroparesis. In addition, certain emotions, sights, smells, and tastes can induce nausea and vomiting when mediated through the cerebral cortex.

There are several structures that can be targeted in order to control PONV and PDNV, as these contain the major receptor systems involved in the etiology of PONV— dopaminergic, cholinergic (muscarinic), histaminergic (H1), serotonergic (5-hydroxytryptamine₃, 5-HT₃), and neurokinin (NK-1).²¹ Therapies that target these receptors have formed the mainstay of PONV management. While surgery directly impacts the GI tract, anesthesia can provide relief by targeting the CTZ. Antagonism of the binding of the neurotransmitters serotonin, substance P, or dopamine, respectively, to the 5-HT₃ receptor, NK-1 receptor, or dopamine receptor can relieve nausea and vomiting by affecting the emetic center and CTZ (Table 2).^22,23,24 In addition, 5-HT₃ receptors located in the GI tract and benzodiazepine receptors present in higher cortical structures can also be targeted to relieve nausea and vomiting.

The difference between PONV and PDNV is the setting and timing of the side effect. PONV is nausea and vomiting that occurs within the first 24-hour period after surgery—whether in-hospital or postdischarge—as well as anytime while an inpatient. PDNV is nausea and vomiting that occurs after discharge, usually from a health care facility following outpatient surgery (Figure 2).² Delayed PDNV or PONV is nausea and vomiting that occurs >24 hours following surgery.

Prospective observational study of treatments, outcomes and patterns of care (POSTOP) for PONV

It is clear that nausea and vomiting remain common, ongoing problems following surgery. The leading adverse effects causing the hospital admission of outpatients are nausea and vomiting;²⁵ these adverse effects can also delay hospital discharge. Although patients readily accept the likelihood of pain following surgery, they do not accept the potential for nausea and vomiting. Clinical practice guidelines for managing PONV and PDNV have been developed and published recently by the American Society of Anesthesiologists (ASA)²⁶ and the American Society of PeriAnesthesia Nurses (ASPAN).² Yet these have not led to a well-accepted consensus for managing PONV and PDNV, and patients still commonly experience nausea and vomiting for up to three days following surgery.

Until now, no study has specifically analyzed the compliance of practitioners with practice guidelines for PONV and PDNV in their routine practice, nor the impact of the recommendations for prophylaxis and treatment on emetic outcomes after general anesthesia in a large population-based study.²⁷ Moreover, antiemetic agents have been used for the management of PONV and PDNV, but the optimal antiemetic agent(s) and timing of administration for effective prevention and/or treatment of patients at risk remains variable and unclear.¹ For example, although two studies have demonstrated efficacious timing for the administration of ondansetron, the package insert fails to recommend any details on this aspect of protocol. For prolonged surgical procedures, significantly improved results (p < 0.05) were obtained when the antiemetic was administered 30 minutes prior to the end of surgery instead of prior to the induction of anesthesia.^28,29 Therefore, the POSTOP study was designed to assess the incidence and time course of PONV and PDNV, as well as the impact of emetic sequelae on the patients’ functional recovery, and the current clinical treatment patterns associated with the management of high-risk patients. Secondary to these primary objectives, the study endeavored to characterize the patterns of antiemetic use for prophylaxis and rescue of PONV before, during, and after surgery, and assess patient-reported outcomes, including interference with functioning.²⁷

The key inclusion criteria for the POSTOP trial were males or females with an ASA status of 1 to 3 who were ≥18 years old and had undergone a surgical procedure lasting a minimum of one hour with general anesthesia. Included were patients who underwent laparoscopic abdominal, gynecological surgery, or major plastic surgery and had ≥2 risk factors (such as being female, having a history of PONV/motion sickness, or a nonsmoking status) and were expected to be discharged within 24 hours following surgery. Although the inclusion criteria allowed both genders, the participants tended to be female; only 9% of the patients were male. Table 3 lists other demographic data of the POSTOP study participants, including the risk factors that enhanced their likelihood of experiencing PONV or PDNV.²⁷ The exclusionary criteria for the POSTOP study were cancer patients treated with chemotherapy or radiation therapy within 30 days prior to surgery, use of any drug that may have an antiemetic effect within 24 hours prior to the start of surgery, any prior use of palonosetron, and use of remifentanil or atropine during surgery.

The multicenter trial had a prospective design; patient outcomes were observed as they occurred in usual practice at 11 U. S. sites.²⁷ Initial assessments for this observational study of PONV began with completion of pretreatment demographic questionnaires and other instruments by the patients. Then PONV-related outcome data were captured via interactive voice recording over a 72-hour postoperative observation period. In addition, PONV data were collected at 2, 6, 24, 48, and 72 hours after patients awoke from anesthesia and were able to respond to verbal commands.

A complete response (CR) was defined as no rescue medications given and no emetic episodes observed, while complete control (CC) was defined as a CR with no more than mild nausea.²⁷ The occurrence of any emetic episode, the number of emetic episodes, and the occurrence of nausea were also documented. Nausea severity was rated on a four-point Likert scale, as either none, mild, moderate, or severe. The use of rescue medications was noted, while the interference of PONV with daily living across 5 domains (appetite, sleep, physical activities, social life, and enjoyment of life) was documented using the modified Osoba module. The type and dosage of prophylactic and rescue medications administered before, during, and after surgeries while patients were in the PACU were recorded.

This observational study noted the patterns of behavior of patients at high risk for developing PONV. Despite the enhanced risk of PONV within the population studied, 8% of the patients did not receive any antiemetic prophylaxis.²⁷ Thirty-four percent received one antiemetic prophylactic agent while 35% were administered two and 23% were given three or more (Table 4); the number of antiemetic agents administered affected the outcomes (Figure 3). Overall, there was a wide variation in the number of antiemetic agents given to high-risk patients, indicating a lack of standardization for managing PONV and PDNV in clinical practice. A consideration of the pattern of usage with a single agent revealed that the 5-HT₃ class of medications is most frequently administered (Table 4), while metoclopramide, dexamethasone, and droperidol are also (infrequently) used. Typically, a 5-HT₃ inhibitor plus dexamethasone is prescribed to manage PONV; the administration of this combination has been supported by studies that have demonstrated the improved efficacy of dexamethasone combination treatments versus single agents.

Still, there are a variety of different combination regimens given for PONV, once again demonstrating the variance in current clinical practice. Seemingly, this lack of standardization is detrimental to patient outcomes. Thirty-three percent of patients received rescue medication in the first six hours following surgery, while another 28% were given a rescue medication 72 hours postsurgery. In summary, 45% of the participants required nausea and vomiting rescue medications in the first 72 hours²⁷ (a more complete data set describing the incidence and control of PONV and PDNV is listed in Table 5.)³⁰ Certainly, the quality of the anesthetic technique used affects this outcome, as do the overuse and high concentrations of inhalational anesthetics and the propensity for prescribing high-dose opioids. The more prevalent use of brain monitoring has made evident that in the U.S., in particular, patients are often treated with higher doses of anesthetics than are necessary.^31,–34 The most common medications used for rescue were the 5-HT₃ antagonists and promethazine. In the postdischarge period of 6 to 72 hours, promethazine was the most commonly prescribed antiemetic.

Despite the frequent administration of multiple antiemetic agents, 51% of patients did not experience a CR and 61% did not have complete control during the 72-hour study interval.²⁷ The lower CC rates demonstrate the impact of nausea intensity on patient outcomes. Certainly the interference of PONV/PDNV with daily life functioning data further supports this idea. The percentage of patients that experienced functional interference due to PONV or PDNV during the 72-hour study period included 65% with appetite reduction, 54% with disrupted sleep, 55% with reduced physical function, 49% with disrupted social interactions, and 59% with reduced overall enjoyment (Figure 4).³⁰

Treatment recommendations from the report by the ASA Task Force in 2002 on the use of antiemetic agents for the prevention and treatment of nausea and vomiting were available during the time when the POSTOP study was conducted.²⁶ Despite the availability of guidelines, a large number of high-risk patients suffered from PONV and PDNV during the POSTOP study. Based on the cumulative data obtained during the 72-hour study period, only 61% of patients were given prophylaxis and rescue medication(s) in accordance with the 2002 ASA guidelines and 52% according to ASPAN guidelines.^2,26,27 When clinical practice failed to comply with the ASA guidelines, 17% of patients had a CR, and 67% experienced moderate-to-severe nausea in the 72 hours following surgery (Figure 5). Further consideration of the data revealed that even with PONV treatment that complied with ASA guidelines, only 69% of patients experienced a CR during the 72-hour study period. In part, the lack of efficacy, even with guideline-compliant treatment, may be due to the use of short-acting antiemetic agents along with long-acting opioids. Antiemetic agents with a longer duration of action could ameliorate this problem. Still, it is important to note that, overall, adherence with the 2002 ASA guidelines led to a reduction in moderate-to-severe nausea during the study period.

Although the observation period was longer than average, the results of the POSTOP study were still consistent with the data from other available studies. Corresponding results were found in a PONV risk-analysis study, which reviewed a database of an oncology surgical population including 8,892 adult patients, of which 4,682 were considered at high risk for PONV.³⁵ The study found that 25.8% of patients required antiemetic rescue therapy within the PACU. These single-center data parallel findings from the multicenter observational POSTOP study, which noted that 33% of patients required rescue within the six-hour period following surgery. Also, both studies further underscore the unmet need for better nausea and vomiting control—specifically in the surgical oncology patient population.

Another multicenter, prospective, observational study with 332 high-risk patients recruited from 11 freestanding surgical centers and 7 hospitals examined the burden of PONV and PDNV after outpatient ambulatory, but highly emetogenic, surgeries.³⁶ Most commonly the procedures performed were plastic (45%), laparoscopic (29%), and orthopedic (22%); similar to the POSTOP study, 90% of the participants were female. Although 92% of patients received prophylactic antiemetic agents, 20% still required inhospital rescue therapy and 43% reported being nauseous. Even though 32% of the postsurgical patients were given antiemetic treatment at discharge for home use, nearly half of the study participants (45%) who had not reported in-facility PONV experienced nausea and/or vomiting in the following one to three days after surgery. In fact, the peak incidence of nausea and vomiting occurred the first day after discharge. Seemingly, clinicians are missing opportunities to prophylactically manage PDNV. In addition, PONV was associated with a longer duration of in-facility recovery, indicating that more effective preventive measures for PONV and PDNV are needed for high-risk patients.³⁶

Summary

It is well accepted that nausea and vomiting can interfere with patient function and outcomes; therefore, controlling these common postoperative adverse effects is of considerable importance. In particular, PDNV is a challenge to control, in part because it has been poorly studied to date compared to PONV. In addition, incidences of PDNV are often underreported by patients, since the symptoms primarily occur without medical supervision present. Potential options for managing PDNV include ondansetron oral disintegrating tablets, the scopolamine patch, and stimulation of acupressure points. It is possible that PDNV may be reduced by the introduction of new prophylactic modalities that outlast the prophylactic ranges of traditional antiemetic agents. Also, longer-acting antiemetic agents, such as palonosetron, may obviate the need for patient-administered rescue treatment, simply by providing antiemetic relief further into the postdischarge period.

The POSTOP study endeavored to characterize the practice patterns associated with the management of PONV and PDNV. Although the observational study found that 92% of high-risk patients received antiemetic prophylaxis, 62% of patients experienced nausea and 30% of participants vomited during the 72-hour period directly following surgery. Moreover, nearly half of the patients in the study required rescue medication. A consideration of the practice patterns indicated that, currently, there is no standard of care for managing PONV or PDNV. Potentially, anesthesia departments could play a key role in promoting adherence to guidelines, in part by monitoring the incidences of perioperative complications, as well as by instituting policies regarding the management of PONV in line with the current recommendations.

Even with ASA guideline compliance, the incidences of PONV and PDNV remain unacceptably high—particularly with respect to nausea. Although patients experienced less PONV and PDNV when they received guideline-based antiemetic care, these results indicate that a significant proportion of patients still experience symptoms of PONV for up to three days after surgery.

Conclusion

New antiemetic agents with greater efficacies and longer durations of action may offer extended protection from nausea and vomiting throughout the 72-hour postsurgical time period, and thereby reduce the incidences of PONV and PDNV in patients undergoing surgical procedures, as will the rational use of prophylaxis based on patient stratification.

References

Author notes